Hybrid Communities: Biosocial Approaches to Domestication and Other Trans-Species Relationships

Hybrid Communities

Domestication challenges our understanding of human-environment relationships because it blurs the dichotomy between what is artificial and what is natural. In domestication, biological evolution, environmental change, techniques and practices, anthropological trajectories and sociocultural choices are inextricably interconnected. Domestication is essentially a hybrid phenomenon that needs to be explored with hybrid scientific approaches. Hybrid Communities: Biosocial Approaches to Domestication and Other Trans- species Relationships attempts for the first time to explore domestication viewed from across disciplines both in its origins and as an ongoing process. This edited collection proposes new biosocial approaches and concepts which integrate the methods of social sciences, archaeology and biology to shed new light on domestication in diachrony and in synchrony. This book will be of great interest to all scholars working on human-environment relationships, and should also attract readers from the fields of social anthropology, archaeology, genetics, ecology, botany, zoology, history and philosophy.

Charles Stépanoff is a social anthropologist (Laboratoire d’anthropologie sociale, École pratique des hautes études, Sorbonne, France). His research interests include human-animal relationships in hunting, herding and shamanism in North Asia.

Jean-Denis Vigne is an archaeologist (Centre national de la recherche scientifique, Muséum national d’Histoire naturelles, Sorbonne Universités, France). His research interests lie in archaeozoology, focused on interaction dynamics between animals and human societies, namely domestication, since the last hunters to the preindustrial farmer societies, mostly in the Mediterranean area, Southwest Asia and Central Asia and China. Routledge Studies in Anthropology

Counterfeit Itineraries in the Global South The Human Consequences of Piracy in China and Brazil Rosana Pinheiro-Machado

Culture as a System How We Know the Meaning and Significance of What We Do and Say David B. Kronenfeld

Distortion Social Processes Beyond the Structured and Systemic Edited by Nigel Rapport

Critical Times in Greece Anthropological Engagements with the Crisis Edited by Dimitris Dalakoglou and Georgos Agelopoulos

An Ethnography of Global Environmentalism Becoming Friends of the Earth Caroline Gatt

Linguistic and Material Intimacies of Cell Phones Edited by Joshua A. Bell and Joel C. Kuipers

Hybrid Communities Biosocial Approaches to Domestication and Other Trans-species Relationships Edited by Charles Stépanoff and Jean-Denis Vigne

Orthodox Christian Material Culture Of People and Things in the Making of Heaven Timothy Carroll

For more information about this series, please visit: www.routledge.com/Routledge- Studies-in-Anthropology/book-series/SE0724 Hybrid Communities Biosocial Approaches to Domestication and Other Trans-species Relationships

Edited by Charles Stépanoff and Jean-Denis Vigne First published 2019 by Routledge 2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN and by Routledge 711 Third Avenue, New York, NY 10017 Routledge is an imprint of the Taylor & Francis Group, an informa business © 2019 selection and editorial matter, Charles Stépanoff and Jean-Denis Vigne; individual chapters, the contributors The right of Charles Stépanoff and Jean-Denis Vigne to be identified as the authors of the editorial material, and of the authors for their individual chapters, has been asserted in accordance with sections 77 and 78 of the Copyright, Designs and Patents Act 1988. All rights reserved. No part of this book may be reprinted or reproduced or utilised in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. Trademark notice : Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation without intent to infringe. British Library Cataloguing-in-Publication Data A catalogue record for this book is available from the British Library Library of Congress Cataloging-in-Publication Data Names: Stépanoff, Charles, editor. | Vigne, Jean-Denis, editor. Title: Hybrid communities : biosocial approaches to domestication and other trans-species relationships / edited by Charles Stépanoff and Jean-Denis Vigne. Description: Abingdon, Oxon ; New York, NY : Routledge, 2018. | Series: Routledge studies in anthropology ; 46 | Includes bibliographical references and index. Identifiers: LCCN 2017058517 (print) | LCCN 2018016153 (ebook) | ISBN 9781315179988 (ebook) | ISBN 9781351717984 (web pdf) | ISBN 9781351717977 (epub) | ISBN 9781351717960 (mobi/kindle) | ISBN 9781138893993 (hardback : alk. paper) Subjects: LCSH: Domestication. | Nature—Effect of human beings on. | Human-animal relationships. | Human-plant relationships. Classification: LCC SF41 (ebook) | LCC SF41 .H94 2018 (print) | DDC 636—dc23 LC record available at https://lccn.loc.gov/2017058517 ISBN: 978-1-138-89399-3 (hbk) ISBN: 978-1-315-17998-8 (ebk)

Typeset in Times New Roman by Apex CoVantage, LLC Contents

List of figures ix List of tables xi List of contributors xiii

Introduction 1 CHARLES STÉPANOFF AND JEAN-DENIS VIGNE

PART I Liminal processes: beyond the wild and the domestic 21

1 A genetic perspective on the domestication continuum 23 LAURENT A. F. FRANTZ AND GREGER LARSON

2 Self-domestication or human control? The Upper Palaeolithic domestication of the wolf 39 MIETJE GERMONPRÉ, MARTINA LÁZNIČKOVÁ-GALETOVÁ, MIKHAIL V. SABLIN AND HERVÉ BOCHERENS

3 Beyond wild and domestic: human complex relationships with dogs, wolves, and wolf-dog hybrids 65 NICOLAS LESCUREUX

4 Wild game or farm animal? Tracking human-pig relationships in ancient times through stable isotope analysis 81 MARIE BALASSE, THOMAS CUCCHI, ALLOWEN EVIN, ADRIAN BĂLĂŞESCU, DELPHINE FRÉMONDEAU AND MARIE-PIERRE HORARD-HERBIN

5 Arable weeds as a case study in plant-human relationships beyond domestication 97 AMY BOGAARD, MOHAMMED ATER AND JOHN G. HODGSON vi Contents PART II How domestication changes humans’ bodies and sociality 113

6 From fighting against to becoming with: viruses as companion species 115 CHARLOTTE BRIVES

7 Milk as a pivotal medium in the domestication of cattle, sheep and goats 127 MÉLANIE ROFFET-SALQUE, ROSALIND E. GILLIS, RICHARD P. EVERSHED AND JEAN-DENIS VIGNE

8 Watching the horses: the impact of horses on early pastoralists’ sociality and political ethos in Inner Asia 145 GALA ARGENT

PART III Shared places, entangled lives 163

9 Growing a shared landscape: plants and humans over generations among the Duupa farmers of northern Cameroon 165 ÉRIC GARINE, ADELINE BARNAUD AND CHRISTINE RAIMOND

10 Fig and olive domestication in the Rif, northern Morocco: entangled human and tree lives and history 179 YILDIZ AUMEERUDDY-THOMAS AND YOUNES HMIMSA

11 Cooperating with the wild: past and present auxiliary animals assisting humans in their foraging activities 197 EDMOND DOUNIAS

12 Why did the Khamti not domesticate their elephants? Building a hybrid sociality with tamed elephants 221 NICOLAS LAINÉ

13 Cognition and emotions in dog domestication 235 SARAH JEANNIN Contents vii PART IV Ongoing transformations 249

14 Domestication and animal labour 251 JOCELYNE PORCHER AND SOPHIE NICOD

15 Human-dog-reindeer communities in the Siberian Arctic and Subarctic 261 KONSTANTIN KLOKOV AND VLADIMIR DAVYDOV

16 Domesticating the machine? (Re)configuring domestication practices in robotic dairy farming 275 SÉVERINE LAGNEAUX

17 From parasite to reared insect: humans and mosquitoes in Réunion Island 289 SANDRINE DUPÉ

Index 302

Figures

0.1 Hybrid communities. 12 1.1 Schematic of various models of domestication and their effect on genetic diversity. 25 2.1 Dorsal view of a Pleistocene wolf skull from the Gravettian Předmostí site. 43 2.2 Dorsal view of the Palaeolithic dog skull from the Goyet cave. 44 4.1 Results from stable carbon (δ 13 C) and nitrogen (δ 15 N) analysis of bone collagen from pigs, sheep and human remains from the mediaeval city of York. 84 4.2 Reliance of pigs on domestic farmed food (millet, animal protein scraps) in Dadiwan in ancient China, as evidenced from stable isotope composition of bone collagen. 87 4.3 Gradual changes through time in pigs’ diet in Xiawanggang, with an increasing contribution of millet and animal proteins. 89 4.4 A: Stable isotope ratios in bone collagen of the main species from the Gumelniţa culture assemblages at Borduşani-Popină, Hârşova-tell and Vităneşti-Măgurice. B : Stable isotope ratios in bone collagen from suids with small ‘domestic’, large ‘wild’ and large ‘domestic’ molars (from geometric morphometric analyses). 91 5.1 Display of gathered plants including weedy Malva spp., market, Jeblia, Rif region, Morocco. 101 5.2 Vicia sativa subsp. nigra , growing as a weed of cereals in the oasis of Imin-o-Iaouane on the southern slopes of the High Atlas, Morocco. 102 5.3 Harvested décrue barley field, showing spiny weeds (Echinops spinosus ), Guelmim province, Morocco. 103 7.1 Proportion of animal fat residues identified as milk fats, ruminant and non-ruminant adipose fats and aquatic fats in archaeological sherds from the Neolithic in Europe and the Near East. 133 7.2 Detail of a milking scene from rock art in a rockshelter at Tasigmet, Oued Djerrat (Tassili-n-Ajjer). 135 7.3 Hypothetical cattle kill-off profiles. 136 8.1 Reconstruction of the Berel 11 burial mound. 147 x Figures 8.2 Typical Pazyryk bridle structure with both the headstall (3) and throat latch (4) fastening on the horse’s left side. 153 8.3 Scythian armor and weaponry. 155 8.4 Pazyryk shields. 155 9.1 Sorghum is represented by a large diversity in the Duupa subsistence system from cultivated to wild morphotype. 171 9.2 Leaving in weeds. 173 9.3 Children sorting leaves. 174 10.1 Children in the Bni Ahmed region caprifying a fig tree, Rif, northern Morocco. 184 10.2 Large mixed fig, olive and cereal agroecosystem, Rif, northern Morocco. 185 10.3 Oleasters grafted with olive varieties within a tended orchard in Sidi Redouane, Rif, northern Morocco. 187 11.1 Domesticated and non-domesticated auxiliary animals assisting humans in their foraging activities. 208 12.1 Elephant tied to the lak chang. 223 12.2 Learning commands. 225 12.3 Ivory statue representing Utingna . 228 12.4 The relational dynamics between the Khamti and elephants. 230 15.1 Ratio of reindeer, sled, pastoral, and hunting dogs in the nomadic households of the Russian North, according to the data from the Polar Census of 1926–1927. 263 15.2 Ratio of reindeer, sled, pastoral, and hunting dogs in settled households of the Russian population of the Russian North, according to the data from the Polar Census of 1926–1927. 264 16.1 The view from Marc’s office of the entirety of the robotic stable, Belgium, August 2016. 276 16.2 Marc at his office computer, which shows a lactation curve in decline, Belgium, July 2017. 282 Tables

2.1 Comparison of canid products (cf. Sigaut, 1980) in the ethnographic and archaeological (Upper Palaeolithic) record (non-exhaustive list). 48 2.2 Comparison of possible forms of familiarization of wolves (cf. Sigaut, 1988) in the ethnographic and archaeological (Upper Palaeolithic) record (non-exhaustive list). 51 2.3 Comparison of possible forms of appropriation of captive wolves/dogs (cf. Sigaut, 1988) in the ethnographic and archaeological (Upper Palaeolithic) record (non-exhaustive list). 53 7.1 Composition of cow, sheep, goat and human milk. 128 11.1 Profiles of wild and untamed auxiliary animals assisting humans in their foraging activities. 210 15.1 Number of people, dogs, and reindeer in the households of the Russian North. Calculated on the basis of data from the 1926–1927 Polar Census. 262 15.2 Main typological groups of HDR communities. 266 17.1 The domesticity of mosquitoes, as an encompassing category, according to Mason’s categories (1985). 292 17.2 The domesticity of mosquitoes from the gardens, according to Mason’s categories (1985). 293 17.3 The domesticity of experimental mosquitoes, according to Mason’s categories (1985). 296 17.4 The domesticity of released mosquitoes, according to Mason’s categories (1985). 297 17.5 The domesticity of Réunionese Aedes albopictus , according to Mason’s categories (1985). 299

Contributors

Gala Argent is a lifelong equestrienne and an interdisciplinary scholar whose work concerns the relationships between humans and other animals. She has taught in departments of art, communication studies and anthropology, and currently teaches through the Animal Studies major within the Psychology Department at Eastern Kentucky University, USA. Her recent work focuses on theorizing human-horse interactions using models of nonverbal and interpersonal communication, and on the ways in which humans and other animals come together in corporeal, temporal and spatial ways to co-create mutually interdependent selves and societies. She shares her social world with horses, dogs, birds and other humans. Mohammed Ater is an ethnobotanist and biologist with particular interests in traditional agrosystems and crop varieties (Department of Biology, Université Abdelmalek Essaâdi, Tétouan, Morocco). He has conducted many field studies of tree and annual crops in Morocco, including in the Rif region and in southern oasis systems. Yildiz Aumeeruddy-Thomas is an ethnoecologist who studies the cultural context which guides peoples’ connectedness to nature. She is a research director at the French National Center for Scientific Research (CNRS) working at the Center for Functional and Evolutionary Ecology in Montpellier (CEFE). She analyzes agrosylvopastoral practices that transform nature and affect social identities and organizations including relationship to others, non-humans and non-tangible entities. Her work among the Jbala society in Morocco looks at ways products such as olive oil may have had a major role in pre-domesticary processes and the on-going nature of domestication. She is currently developing a comparative study in Sicily on the island of Pantelleria. Adrian Bălăşescu is a zooarchaeologist at the Vasile Pârvan Institute of Archaeol- ogy, Romanian Academy, Bucharest, Romania. He has participated in various research teams on numerous archaeological excavations (in Romania, France and Armenia), being responsible mainly for the zooarchaeological investigations. His background is in Balkan Prehistory. He also has research interests in Armenian Neolithic and Chalcolithic zooarchaeology. xiv Contributors Marie Balasse is a researcher at the French National Center for Scientific Research (CNRS), affiliated to the Muséum national d’Histoire naturelle in Paris, France. She is a zooarchaeologist specialized in stable isotope analysis of faunal remains. Her main area of expertise is husbandry practices in the Neolithic of Europe. Adeline Barnaud is a researcher at the French National Research Institute for Sustainable Development (IRD), France. She is a plant geneticist, interested in social, environmental and biological factors shaping African cereals’ genetical resources. Research unit DIAD (DIversité - Adaptation - DEveloppement des plantes), Institut de recherche pour le développement, Université de Montpel- lier, France. Hervé Bocherens is a palaeobiologist in the Department of Geosciences and Senckenberg Centre for Human Evolution and Palaeoenvironment, University of Tübingen, Germany. His main area of research is to track species habitat and trophic relationships within ancient terrestrial communities using stable isotopes in fossil remains, mostly in the Quaternary and the Neogene. Amy Bogaard is an archaeologist at the School of Archaeology, University of Oxford, UK, and external professor at the Santa Fe Institute, working on the ecology, evolution and social implications of early farming systems in Europe and western Asia, with comparative interests in other parts of the world. Key methods in her research include survey of present-day ‘traditional’ agrosystems to build heuristic models for comparison with archaeobotanical data. Charlotte Brives is a research fellow at the Centre national de la recherche scientifique, Centre Emile Durkheim, Bordeaux, France. After having worked for several years on the production of evidence and scientific knowledge in clinical trials on HIV/AIDS in sub-Saharan Africa, she is now studying human/ microbes relationships through two main fieldworks: the production of scientific knowledge on the human microbiota, and the development of phage therapy to combat multidrug-resistant bacterial infections. Thomas Cucchi is a zooarchaeologist employed by the CNRS at the National Natural History Museum of Paris, France. His research focuses on the origin, spread and evolution of commensal and domestic mammals, throughout the Holocene. He relies on geometric morphometris approaches combined with genetic and isotopic approaches when possible. He is currently leading a research project on new morphometric markers of anthropogenic changes in animals’ biomechanic environments. Vladimir Davydov is Head of the Siberian Ethnography department at Peter the Great Museum of Anthropology and Ethnography (Kunstkamera), Russian Academy of Sciences, Russia. He got his PhD at the University of Aberdeen (2012). He is a co-author and an editor of the monograph Fire, water, wind and stone in Evenki landscapes. Human-nature relations in Baikal Siberia (2016). Contributors xv Edmond Dounias works for the French National Research Institute for Sus- tainable Development (IRD), France, and is a Senior Research Associate to the Center for International Forestry Research (CIFOR). He is affiliated to the Center for Functional and Evolutionary Ecology (CEFE) at Montpellier, France. He explores the resilience of socio-ecological systems in response to external drivers of change by focusing on the biocultural interactions between forest dwellers and tropical rainforests, with a long fieldwork experience among forest dependent peoples in Congo Basin, Sumatra and Borneo. He is co-editor of several books, including Animal Symbolism: Animals, Keystone of the Relationship between Man and Nature (2007), Peoples and Natures (2012), and Climate Change: What Challenges for the South (2015). Sandrine Dupé has a PhD in anthropology, and is an Associate Research Fellow at the National Museum of Natural History in Paris, France. She is currently working on an interdisciplinary research project on coastal agriculture, and is interested in the recede of animals from Breton coastline. Richard P. Evershed FRS is a Professor of Biogeochemistry at the Organic Geo- chemistry Unit, University of Bristol, UK. His research is highly interdisciplinary, applying the principles, techniques and rigor of organic and analytical chemistry to tackle questions in the fields of archaeological chemistry, biogeochemistry and biomolecular palaeontology. Allowen Evin is a researcher at the French National Center for Scientific Research (CNRS), working at the Institut for Evolutionary Sciences in Montpellier (ISEM), France. She is mainly using multi-disciplinary approaches combining bioarchaeology and evolutionary biology to study the evolution of domestic animals. Laurent A. F. Frantz (PhD in Genetics at Wageningen University, 2015) spent two years as an European Research Council (ERC) postdoctoral researcher at the University of Oxford, UK, and one year as a Wolfson College Junior fellow, also at the University of Oxford. In 2017, he became a Lecturer in Molecular Genetics and Informatics at Queen Mary University of London, UK, where he studies evolutionary genetics and animal domestication. Delphine Frémondeau is currently a post-doctoral research assistant at the Uni- versity of Reading, UK, and a research associate at the University of Leuven, Belgium. As a Zooarchaeologist specialised in stable isotope analysis, she is investigating food supply to the Roman and Medieval town of Exeter, UK, subsistence economy and land use in Late Bronze and Iron Age SE Bulgaria, and early farming practices in the Balkan Peninsula. Éric Garine is lecturer at the department of Anthropology of Paris Nanterre Uni- versity, France. He is studying various aspects of human-environment relationships in West Africa. xvi Contributors Mietje Germonpré is a palaeontologist and archaeozoologist at the Royal Bel- gian Institute of Natural Sciences, Belgium. Her main areas of research include human-animal interactions from hunting to domestication during the Middle and Upper Palaeolithic in northern Eurasia. Rosalind E. Gillis is a postdoctoral researcher at the Christian-Albrechts- Universität in Kiel, Germany. Multi-proxy approaches to animal husbandry and foddering practices are the principal foci of her research, with an interest on how these influenced – and vice versa – the development of social structure and ideologies. Younes Hmimsa is Professor at University Abdelmalek Essaadi at the Polydisci- plinary Faculty of Larache. He is a specialist of the traditional agrobiodiversity of the Rif in northern Morocco, with a particular focus on the ethnobotany of Ficus carica. He is now involved with work in chemical ecology on relationships between humans, local fig varieties and fig pollinating insects, including changes that are occuring within the context of climate change, in collaboration with a group of researchers from the Centre for Functional Evolutionary Ecology in Montpellier. John G. Hodgson is a plant ecologist and field botanist whose research interests include arable weed flora and comparative work between modern and ancient floras. He was a member of the Unit of Comparative Plant Ecology at the Uni- versity of Sheffield, UK. Marie-Pierre Horard-Herbin is lecturer in Protohistory at the University of Tours, France. She is a zoorchaeologist specialized in Iron Age: her main areas of interests are husbandry, meat diet, and cynophagy. Her lastest book is Pour une Histoire de la viande: fabrique et représentations de l’Antiquité à nos jours (2017). Sarah Jeannin is clinical psychologist; she has a PhD in Ethology and is research associate of the laboratory “Ethology, Cognition and Development (LECD)” at the University Paris Nanterre, France. She works as trainer in Ethology and Psychology at the training center “Animal University” and is partner consultant in behavioural veterinary medicine consultation at the veterinary clinic of Dr T. Bedossa. Her current research is focused on human-pet communication and emotions. Konstantin Klokov is Professor at the Chair of Regional Politics and Political Geography at Saint Petersburg State University, Russia. Issues of reindeer herding and traditional hunting of indigenous peoples of Siberia lie in the main focus of his research. He is a co-author of the monograph The 1926/27 Soviet Polar Census Expeditions , ed. David G. Anderson (2011). Séverine Lagneaux has a PhD in anthropology and is a member of the Laap (Laboratoire d’anthropologie prospective) at the University of Louvain, Bel- gium. Her work focuses on the relationships between humans and non-humans Contributors xvii in farms with different technical systems. She is the author of a monograph entitled Eternal Provisional. Ethnography of the Romanian peasantry at the European (2016). Nicolas Lainé holds a PhD in Ethnology from Paris West University, France (2014). He is currently affiliated with Laboratoire d’anthropologie sociale in Paris, France. Specializing in human-animal relations, his research is at the crossroads of the anthropology of nature and conservation. He has published several articles on human/elephant interspecies community in South and South- east Asia, and has co-edited the collective volume Nature, Environment and Society (2012). He is currently pursuing his research interest on the issues on health and environment, and the link between biodiversity and cultural diversity. Greger Larson received his bachelor’s degree in 1996 from Claremont McKenna College, USA. After working for an environmental consultancy in Azerbaijan, Greger studied at Oxford, UK, and the University of Colorado, USA before receiving his PhD in Zoology in 2006. He then spent two years in Uppsala, Sweden, on an European Molecular Biology Organization (EMBO) postdoctoral fellowship before starting a job in the department of archaeology at Dur- ham University, UK. He has since moved to Oxford University, UK, to become the Director of the Palaeogenomics & Bio-Archaeology Research Network, where he is continuing his focus on the use of ancient DNA to study the pattern and process of domestication. Martina Lázničková-Galetová is an anthropologist and prehistorian at the Mora- vian Museum Brno; Hrdlicka Museum, Charles University, Czech Republic. She has a research focus on prehistoric art, ornaments and industry of hard animal tissues, its manufacturing technology and its roles in the life of prehistoric people of the Upper Palaeolithic, as well as on human-animal relationships during that period. Nicolas Lescureux is ethnoecologist at the French National Center for Scientific Research, France. His research interests are focused on hunters and livestock breeders’ interactions with wild animals – notably large carnivores – and how they participate in their knowledge construction in various socioecological contexts and during major social changes. Sophie Nicod worked for the Ludo Circus Forum for 15 years, and then created her own stable in the bay of Mont Saint-Michel, France. As a member of the French National Agency for Research (ANR) research program “COW” directed by Jocelyne Porcher, she studied a community of feral horses, the Garanos, in northern Portugal and participated in a comparative study of chimpanzee/horse cognition under the direction of Masaki Tomonaga, from the Primate Research Institute, University of Kyoto, Japan. She has now joined the Ecole Blondeau, France, where she manages the training of riders. Jocelyne Porcher is Research Director at the National Institute of Agronomic Research, France. Her researches focus on labor relationships between humans xviii Contributors and animals. She is the author of numerous books, including The Ethics of Ani- mal Labor: A Collaborative Utopia (2017). Christine Raimond is Research Director, CNRS (Centre national de Recherche Scientifique) UMR Prodig (Pôle de recherche et de diffusion de l’information géographique), Paris, France. She is studying the dynamics of rural territories in West Africa. She has published books and journal’s special issues on agrobiodiversity (Frontières des hommes, frontières des plantes cultivées ) and food systems (Sharing food, Ed Univ Guadalajarà ). She is currently working on international programs about the impact of global policies on crop’s genetic resources. Mélanie Roffet-Salque is a postdoctoral Research Associate in the School of Chemistry, University of Bristol, UK. Her research is focussed on studying lipids preserved in archaeological artefacts, mainly pottery, for reconstructing past exploitation of natural resources and human diet. Mikhail V. Sablin is a Senior Scientist at the Zoological institute Russian Acad- emy of Sciences of Saint-Petersburg, Russia. He has specialized in palaeontology, archaeozoology, craniometry and osteometry of mammals. His PhD thesis was about the Ice Age carnivores and herbivores of the central Russian Plain (1998). His recent projects concern domestication, ancient DNA and stable isotopes of the Pleistocene/Holocene mammals. Charles Stépanoff is an ethnologist member of the Laboratory of Social Anthro- pology, holding the chair “Religions of Northern Asia and Arctic” at the École pratique des hautes études at the Sorbonne in Paris, France. His main research subject is human-nonhuman relationships in shamanism, hunting and herding practices among indigenous peoples in Siberia. He coordinated the Interna- tional Research Group GDRI “Nomadism, society and environment” (France, Russia, Kirghizstan) and co-edited the book Nomadismes d’Asie centrale et septentrionale (2013). Jean-Denis Vigne is Research Director and Director of Research, Expertise, Val- orization and Teaching at the Muséum National d’Histoire Naturelle, France. He devotes his archaeozoological researches to the dynamics of the interactions between animals and human societies, namely domestication, since the last hunters to the preindustrial farmer societies, mostly in the Mediterranean area, Southwest Asia and Central Asia and China. He is the author of several books, among them Les débuts de l’élevage (2012), and he coedited First steps of animal domestication: new archaeozoological approaches (2005). Introduction

Charles Stépanoff and Jean-Denis Vigne

The end of domestication? In the past few decades, we have seen crisis after crisis in the world of agriculture and animal farming, leading us more than ever to question the nature and meaning of the links that unite us with domestic plants and animals. Mad-cow disease, bird flu, scandals around the maltreatment of animals in industrial farming and slaughterhouses, widespread worries concerning pesticides, endocrine disruptors, GMOs: the very conditions in which we produce our food and manage live raise questions and doubts. With each new epizootic episode, the public witnesses the destruction of animal populations suspected of carrying diseases caused by the very conditions in which they are bred. In 2001, during the mad-cow-disease crisis, Claude Lévi-Strauss noticed a “diffuse feeling that our species is paying the price for having contravened the natural order”. He saw modern breeding as having reached an impasse and imagined a future where it would be abandoned: “Our former herds will be set free and become just like any other game, in a countryside returned to the wild” ( Lévi-Strauss 2001 , 13). This feeling of crisis has intensified ever since with the fast disappearance of wild biodiversity and a growing awareness of the global environmental changes which the concept of the Anthropocene refers to: the Earth System itself seems now threatened by human activities which would have become comparable to a geological force, bringing down the walls between nature and society (Hamilton et al. 2015; Latour 2017 ). One key question is that of the origins of this chain of changes that humanity seems to have unleashed around it and has lost control of: when did the anthropocene begin? Some date the dividing point to 1945 and the first atomic bomb, others to the invention of the steam machine. For others still, it is with the first instances of plant and animal domestication that humans started to set in motion radical changes within the ecosystems which ultimately affected the climate (Smith and Zeder 2013). Neolithic scholar Jacques Cauvin already maintained it: “in our domination of the Earth, the decisive step was taken in the Neolithic Age and we are its heirs and direct product.” ( Cauvin 1994 , 15). For Cauvin, the rise of this domination resulted from a mental and environmental process of separation between the wild world and the domestic world: “Little by little, wild nature was confronted by an initially limited portion of land that was 2 Charles Stépanoff and Jean-Denis Vigne constantly expanded, worked, ‘civilized’ and reshaped by human activities and exploitation: a process of conquest was started” (Cauvin ibid ). This vision of a war waged by the domestic against the wild is not different from the environmentalist Paul Shepard’s statement that “in the ideology of farming, wild things are enemies of the tame; the wild Other is not the context but the opponent of ‘my’ domain” ( Shepard 1996 , 178). With the growing awareness of the environmental crisis, the once triumpha- list narrative of this struggle has taken on ever darker nuances. The nostalgia for pre-agricultural and pre-state societies is not new, going back as it does to at least Jean-Jacques Rousseau (18th century). However, for some people, in recent years it has taken on a new dimension in the guise of a phobia of domestication, the latter being perceived as an original sin containing the germ of all subsequent disasters. A turning point has undoubtedly been paleopathology highlighting the Neolithic’s negative impact on human health (Cohen et al. 1984), which, together with the spread of inequality and violence, has led Jared Diamond to denounce the agricultural revolution as “the worst mistake in the history of human race” (Diamond 1987). A growing number of authors see domestication as a radical change of attitude toward the environment, a taking of ownership of living beings the result of which are domination, slavery, and the current global ecological crisis (Diamond 1998; Harari 2014; Lestel 2015; Oelschlaeger 1991; Scott 2017; Shepard 1996 ). For the philosopher Dominique Lestel, “we need to either radically change our culture or disappear; nothing can be salvaged. [. . .] The West truly has a major problem with nature and its collapse was programmed from its origins” (Lestel 2015). The use and abuse of domestic animals have become the archetypal images of an irredeemable environmental and moral failure. Taking note of the negative effects of domestication, increasingly less marginal social movements but also influential thinkers (Nibert 2013; Francione 2009; Regan 2005; Singer 1976) have been promoting global veganism, “cruelty-free consumption”, and “abolitionism” – that is, the abolition of the right to treat animals as property and resources. Is not this, in fact, a way to absolve ourselves of a relationship with animals whose either roots and causes, or consequences and goals, we no longer understand? In not such a distant, more optimistic past, the great march of domestication was supposed to lead to “civilization” and the “liberation of humanity from the constraints of nature”. Today, for a growing fringe of Western societies it leads to a deadly procession of disasters. These apparently contradictory positions paradoxically share the same determinist view of domestication as the already mapped-out route of a divorce between nature and society, for better or for worse. Both points of view ignore the animals’ slow familiarization with the humans, well before the advent of Neolithic agriculture and breeding, and the bio-cultural process spanning thousands of years which saw domestic animals and plants adapting to humans just as much as human societies were shaped on the basis of the animals and plants they incorporated. In this work, we explore views of domestication which differ from that of a great programmed march leading to a sole destination. Bringing together Introduction 3 archaeobotany, archaeozoology, genetics, ethology, ethnoecology, ethnobotany, and social anthropology, our work draws on both the variety of the present situations observed as well as the long period during which they emerged and highlights the multiple ways in which a domus , a home, accommodating several species can be built. “Domestication” refers to a becoming, a process set in time which causes living beings to go through different states, between the outside and the domus , between far and near, between predation and familiarity. Domestication does not simply cross these boundaries; it blurs them and questions categories, whilst at the same time adding its own apparent order by describing a succession of changes as oriented stages. Talking about domestication means giving narrative shape to the events’ diversity and taking the risk of finalism by bringing ex post meaning to that which had none at the beginning.

The tale of the nature-culture divorce On this point, the great narrative dominating in the West is that of the Neolithic Revolution. Deplored or hailed, this change of trajectory split human history in two: an age of foraging was followed by an age of production to which we currently belong. It was the Marxist archaeologist V. Gordon Childe who coined the phrase “Neolithic Revolution” and outlined its scenario: “The escape from the impasse of savagery was an economic and scientific revolution that made the participants active partners with nature instead of parasites on nature” (Childe 1942, 55). According to this author, this “new aggressive attitude” (ibid.) toward nature was supposedly triggered by the climatic change at the end of the Pleisto- cene and the drying up of the Near East, spurring human societies to invent new sources of subsistence. It is striking to note that Childe’s narrative is based on a division between a natural world governed by causal mechanisms and a human world marked by intentionality, innovation, and agency. It is in this sense that domestication is often (and of course wrongly) seen as the entry of humanity into the world of culture, which is subjected to human will and lies outside the laws of nature. Archaeologists have long since repudiated Childe’s scenario which, nevertheless, retains its strong influence as a founding myth. First, the period when early domestication occurred in the Near East covers a vast region comprising different environments and spans thousands of years between the last Ice Age and the first bending of the Holocene climatic optimum. We can no longer link early domestication with a unique climatic or environmental situation characterized by a simple event such as a great drought or, conversely, a period of natural abundance. Besides, the issue is no longer so much to find a cause as to identify the system of interaction between multiple environmental and human factors which, here and there, went out of control and thus triggered the upheaval of domestication (Vigne 2011, 2015 ). Whilst this transition may have spanned tens of human generations, “whatever motivations might have been there at the beginnings of domestication they may no longer have played a role later in the process or at the end of the 4 Charles Stépanoff and Jean-Denis Vigne domestication process” (Fuller 2010, 9). The role of human control in this lengthy emergence has been downplayed: domestication is now described as a “process of unintentional entanglement” with plants and animals, people becoming trapped in feedback loops of increasing labour, domestic plants and animals evolving in response to human innovations ( Larson and Fuller 2014 , 120). The idea of a gulf between nature and culture opened up by domestication paradoxically comes up against the evidence of hybridizations which it has in fact helped proliferate. The life and evolution of domesticated living beings inextricably link biological processes with cultural choices. Agricultural, health, and ecological crises continually remind us that domestication breaks barriers. As Budiansky (1999, X) noted: “Where domestication is concerned, the terms artificial and natural lose any sharp meaning.” In this domestication paradox we recognize the moderns’ fundamental contradiction revealed by Bruno Latour (1993 ): we postulate a metaphysical split between nature and culture whilst surrounding ourselves with hybrids dissolving the nature-culture divide. Postulating a unilineal pathway leading from domestication to the modern Western way of managing and dominating nature means ignoring the variety of the centers of domestication which the world has known during the Holocene: the Near East 11,500 years ago, China 10,000 years ago, South America 10,000 years ago in the case of plants, New Guinea probably 7,000 years ago, India 5,000 years ago, and East Africa 5,000 years ago (Price and Bar-Yosef 2011). Many instances of domestication, including that of the horse, the dromedary, and, indeed, the rabbit, even started outside these great centers, in extremely diverse biogeographical and cultural contexts: Southwestern Europe in the Central Middle Ages in the case of the lagomorph (Callou 2003), Central Asia and Arabia in the fifth or fourth millennium in the first two cases (Outram et al. 2009; Almathen et al. 2016). In each of these regions, people have developed original ways of building long- standing hybrid communities with plants and animals. What avenues should now be adopted in order to conceptualize domestication outside the Eurocentric and contradictory myth of the Neolithic Revolution?

Other narratives Western archaeologists are of course not the only ones to have taken an interest in the question of the origin of cultivated plants and domestic animals. In the various centers of domestication throughout the world, people know of narratives explaining the close relationship formed with certain species. In Amazonia, the Kayapo mythology tells that Indians used to eat rotten wood until a small rat came and told a woman about the existence of maize, which humans could then adopt as their food (Lévi-Strauss 1964, 175–176). For the Achuar, the humans got the manioc from Nunkui, the mother of cultivated plants and protective spirit of gardens, who, in the narrative, takes the shape of a sloth (Descola 1986, 239–244). In Niger, the cattle-breeding Fula people say that the first domestic bovines came out of a river, attracted by human campfire. Ever since then, the Fula have made fires every night so cows may gather around them to avoid mosquitos and predators (Dupire 1962). Introduction 5 In Siberia too, narratives often invoke the animals’ curiosity or attraction to human camps. A Nenets myth recounts that the reindeer used to live separately from humans. Because they were harassed by wolves, some of them decided to go to the humans in search of safety. The other reindeer refused and remained wild. The humans harnessed the reindeer that had come to them and they started to live together ( Golovnev 1995 , 220). It is striking to note that, in these mythologies, we never encounter the trium- phalist image of humans transforming wild plants and animals into submissive domesticated beings as in the narrative of the Neolithic Revolution. Rather, they are multi-agent narratives in which, alongside human agency, the agency of cos- mic or landscape entities and that of the plants and animals themselves play a part. Often, a kind of agreement is made as humans and non-humans find it in their interest to live side by side. For example, both in Africa and Siberia, livestock find safety and comfort next to humans. These multi-agent conceptualizations espoused by cultivators and herders should make archaeologists and anthropologists wonder. We are probably partly trapped by our academic language. The verb “to domesticate” imposes the syn- tactic construction “[a subject] domesticates [an object]” which grants agency to humans and turns plants and animals into passive objects, artefacts even. How- ever, this paradigm is very recent since the Romans themselves did not have the verb domesticare , which is a Medieval Latin invention. In the past, people used other models, different from the subject-object schema, in order to theorize the becoming of their relations with their plants and their animals.

How do we recognize domestication? All theories constructed on the nature-culture divide tend to present domestication as a radical turning point and consider the criteria for differentiating between wild and domestic a crucial issue. Since the middle of the nineteenth century, zoology, botany, archaeology, and anthropology have sought to find, looking through the lens of a rather essentialist perspective, the fundamental identity of domestic beings. We shall see that each of these criteria sheds a light on important aspects of the domesticatory relationship, but neither of them is able to embrace its diversity on its own.

Reproduction control “Domestication necessarily presupposes reproduction under man’s control”: this famous definition by the zoologist Isidore Geoffroy Saint-Hilaire still carries authority with many biologists and anthropologists. In fact, in his work Accli- matation et domestication des animaux utiles ( 1861 , 157), Geoffroy Saint-Hilaire explains that he simply borrowed this definition from the rules of a competition organized by the Société impériale d’acclimatation (Imperial Society of Accli- matization). What did this competition consist of? The one in 1857, for example, offered a 1,000-franc medal to someone able to “domesticate and multiply a great 6 Charles Stépanoff and Jean-Denis Vigne Kangaroo species”, with the condition that two generations be obtained in captivity (Duméril, 1860, II–III). People taking part in this competition were wealthy amateurs, princes and members of the bourgeoisie who enjoyed the use of private zoos where they could conduct their experiments on exotic animals got from colonies at great expense. In hindsight, it is doubtful that these practices marked by nineteenth-century colonialism and scientism could serve as a model for understanding the experiences of hunter-gatherers involved in the Neolithic transition. The control of reproduction and the isolation of domestic subpopulations has long been the acid test for distinguishing the artificial selection through which humans produce domestic plants and animals from the natural selection through which wild species are derived. True, humans often intervene in the reproduction of domestic plants and animals by sowing cereals, planting tubers, castrating, or selecting breeders. Nevertheless, there can be very different degrees of intervention and these do not necessarily seek control. In Europe, today, to what extent is the reproduction of cats controlled by humans? When domestic and wild populations of the same species live in proximity, crossbreeding is frequent and can even happen by design. In Siberia, Chukchi herders let wild male reindeer mate with domestic female reindeer, which is seen as a way of strengthening a herd’s blood ( Bogoras 1904–1909 ). In Papua New Guinea, many groups breed only sows, leaving the task of inseminating them to wild boars ( Dwyer 1996 ). These practices, which appear exotic to us, nevertheless existed in Europe: in Bourgogne, in the eighteenth century, sows were allowed to mate with boars so as to improve the robustness of their lineage (Poplin 1976). Recent palaeogenomic data have confirmed that neither reproductive isolation nor intentional breeding was as significant as previously thought ( Marshall et al. 2014 , cf. Frantz and Lar- son, this volume).

The domestication syndrome If the criterion of reproduction control still holds sway, this is because it seems to be the best explanation for the many morphological and genetic changes observed in domestic populations. Often grouped under the concept of “domestication syndrome”, these traits are frequently seen as defining by biologists. In the case of cereals, the appearance of non-shattering rachis requiring human intervention for spreading the seeds has been considered a marker of the beginning of domestication; however, we now know that cereals were for a long time cultivated in a morphologically wild form (pre-domestic cultivation, Willcox and Stordeur 2012 ). Plants and animals can undergo a process of “domestication” over a long period of time without it being possible to qualify them as “domestic” in a biological sense ( Vigne 2015 ). Conversely, morphologically domestic animals can be taken out of the domestication process if they are feralized, as was the case with the emblematic Corsican mouflon as early as the Neolithic age ( Poplin 1979 ; Vigne 1988 ). The breaking of ties with the humans on a social level never leads to a return to a wild morphology, even after many generations ( Zeder 2015 ). Introduction 7 The theory of neoteny seeks to give a general explanation of the domestication syndrome noticed in domestic animals. Shortened snout, small size, slender skeleton, dependent behaviour: all these changes supposedly show an immaturity in domestic animals which, throughout their lives, retain juvenile morphological and behavioural traits ( Coppinger and Smith 1989 ; Budiansky 199 9). This theory ties up with the idea of the captive animals’ reduced environmental skills (Hemmer 1990). The neotenic scenario is in fact an extension of the old idea held by certain naturalists that domestic forms degenerated in relation to the pure and complete forms represented by the wild species (cf. Lescureux, this volume). Recently, new morphological and ethological studies have damaged the cred- ibility of the neotenic theory. Compared ontogeny assisted by three-dimensional geometric morphometric analysis has shown that dogs’ skulls are not comparable to those of juvenile wolves, nor pigs’ skulls to those of juvenile boars (Drake 2011 ; Owen et al. 2014 ). Furthermore, the fantastic recent developments in the cognitive ethology of domestic animals have led us to question the disparaging condescendence with which we have viewed them. Dogs, goats, horses, and even pigs have proven capable of cognitive operations not seen in chimpanzees, such as shared attention and the comprehension of social cues (Kaminski et al. 2005; Nawroth et al. 2014; Proops et al. 2010). As shown by Sarah Jeannin in this volume, dogs show social and communication skills which are completely unknown in the world of wolves. These hybridized socio-cognitive abilities acquired by domestic mammals in contact with humans show that domestication cannot be reduced to cognitive decline anymore than to neoteny.

Human domination Domestication as domination is inscribed in the origins of naturalism. As Buffon ( 1753 , 169) put it:

Man changes the natural state of animals by forcing them to obey and serve him for his use: a domestic animal is a slave of which one has fun, which one uses, abuses, alters, disorients and denatures, while wild animal, obeying only to Nature, knows no other laws than those of need and freedom.

Following Buffon, Geoffroy Saint-Hilaire saw the control of the reproduction of domestic animals as a manifestation of “man’s domination over the rest of Cre- ation” (Geoffroy Saint-Hilaire 1861, 155). The domination paradigm has long had a strong influence on the works of anthropologists and archaeologists. For Digard, “humans domesticated animals, and still do today, first and foremost in order to satisfy their intellectual need to know and their megalomaniac com- pulsion to dominate and appropriate the world and the living beings” (Digard 1999, 109). As an illustration of this “domesticating zeal”, he cites the famous zebra carriage which Lord Rothschild drove in the nineteenth century at Buck- ingham Palace, showing off his ability to subdue this willful equid (Digard 1990, 214–215). It is doubtful that such an example can be generalized, all the more 8 Charles Stépanoff and Jean-Denis Vigne since these experiments never led to the zebra’s domestication; nevertheless, the Neolithic expert Jacques Cauvin borrowed the concept of “domesticating zeal” from Digard, using it in his theory of “mental revolution” which seeks to explain early domestication in the Near East by the emergence of a domination ideology (Cauvin 1994, 175). And yet, Lévi-Strauss had already highlighted in his work The Savage Mind (1962 ) what he termed the “Neolithic paradox”: the inventions of the Neolithic age were accomplished by people who could not have shared the world-objectifying concepts of modern science. This paradox must not be forgotten. Domination is only one avenue among many: ethnography gives many examples of farming and herding societies which, drawing on the autonomy and agency of plants and animals, have preferred to build their relationships on a model of “respectful friendship” rather than control and objectification (Haudri- court 1962 ; Ingold 1996 ).

Property To the morphological criterion put forward by biologists, social scientists have long preferred that of property. In this view, domestication exists when living animals are objects of ownership and, consequently, exchange or trade (Clutton- Brock 1989, 7; Ducos 1978, 54; Ingold 1980 ; Russell 2007). Marx himself saw domestic animals as instruments of labour devoid of will, the appropriation of which represented a primary source of inequality between humans. However, the property criterion has the disadvantage of including the tamed wild animals liven- ing up the homes of many indigenous peoples ( Erikson 1987 ) or the game living in seigniorial and royal forests which belonged only to the landowners ( Sigaut 1988 ). Conversely, this criterion rejects outside domestication feral and stray animal populations which can very well keep living dependently on humans even though they no longer belong to anyone. Have the thousands of cats and dogs ending up in rescue centers become wild simply because their owners no longer want them?

Intimacy Recently, archaeologists and anthropologists unsatisfied by the limitations of the domination paradigm have highlighted the emergence within domestication of a daily hybrid intimacy and sociality impossible to find in hunting and gathering (Anderson et al. 2017; Armstrong Oma 2010; Knight 2012, 2005 ; Fijn 2011; Sté- panoff 2017 ). The approaches centered on the techno-economic uses of animals ignore what makes these possible: daily interaction, reciprocal adaptation, joint commitment. And yet, archaeology provides striking evidence of these close relations accompanying domestication from its very beginnings: they are the seeds, dogs, cats, lambs, or horses that humans have placed in the tombs of their dead (e.g. Davis and Valla 1978 ; Vigne et al. 2004 ). Once again though, on its own, this criterion would lead to qualifying the Amerindians’ wild pets as domestic and excluding all escaped animals as well Introduction 9 as millions of battery animals that no longer have any close relationship with the humans.

Symbiosis and co-evolution In his 1868 work The Variation of Animals and Plants under Domestication , Dar- win set himself apart from Geoffroy Saint-Hilaire and other zoologists of his time by attributing limited importance to human control and the wild-domestic divide. Darwin suggested that domestication should be approached as “an experiment on a gigantic scale” enabling the modeling of natural selection (Darwin 1868, 3). Darwin distinguished two forms of selection in domestication: the “methodical selection” which was intentional and the “unconscious selection” which produced slight unforeseen changes accumulated over generations. This unconscious selection introduced natural selection mechanisms inside domestication processes. Recent research on the concept of unconscious selection has shown the major role it has played in the evolution of domestic species. The cases of the house mouse ( Mus musculus) and the house sparrow ( Passer domesticus) show that animals can go through morphological changes by adapting to anthropized environments without undergoing any form of human-controlled selection, even several millennia before the dawn of the Neolithic ( Leach 2007 ; Wiessbrod et al. 2017). Even the enlarging of crop seeds, previously thought to have been caused by deliberate selection, seems to be a non-intentional general effect of domestication; indeed, this has been noticed in the case of the potato, cassava, and sweet potato, where people do not plant seeds, let alone harvest them (Kluyver et al. 2017). In the Near East, we now know that many morphological traits seen as markers of early domestic animals appeared not through selection but simply as a quasi-automatic biological consequence of being made captive (Zohary et al. 1998; Vigne 2008). Many morphological changes in domestic plants and animals could thus be the effect of adapting to anthropized environments and a relaxation of natural selective pressures, rather than human decisions (Franz & Larson, this volume), domestication being seen as an environmental change and therefore a change in selection pressures ( Vigne 2011 , 2015 ). Protected, cared for, and spread throughout the world by humans, domestic populations have benefited from obvious advantages compared to wild populations. The domestic horse or cattle conquered the world while the wild one disappeared. From this point of view, some authors describe domestication as a natural phenomenon of co-evolution comparable to the symbiosis or domestication of aphids and fungi by ants (Rindos 1984; Budiansky 1999 ; Zeder 2015 ). However, symbiosis implies an obligatory state of mutual dependence which does not in any way correspond to the autonomy of domestic plants and animals, nor to the forms of intermittent co-existence preferred by various cultivating and herding societies (e.g. Stépanoff et al. 2017 ). Generally speaking, co-evolution phenomena are undoubtedly involved in domestication; however, the concept of co-evolution is clearly defined as a strictly biological interaction between two or several lineages, whilst domestication is a socio-environmental phenomenon 10 Charles Stépanoff and Jean-Denis Vigne characterized by constant historical and technological changes (Kallis 2007; Vigne 2011). For the sake of accuracy, we should speak of socio-environmental co-evolution . As we can see, the various characterizations of domestication as reproduction control, domination, property, intimacy, and co-evolution reveal real and important aspects of the processes involved, but neither of them is entirely satisfactory on its own, making clear the heterogeneous nature of domestication. Insufficient by themselves, these criteria and approaches benefit from being used in a complementary fashion: for example, the morphological approach shows the long-lasting marks left on bodies by multigenerational relations, whilst the social approach centered on intimacy describes these relations as they are experienced by the living beings.

An ongoing transformation The authors of this volume belong to different disciplinary fields, each chapter reflecting their personal points of view, but at the same time their works reveal converging aspects of domestication which we shall try to outline here. This picture of domestication which emerges from this collective work is not a homogeneous reality but a variety of ongoing interconnected biological and social transformations which extend through a continuum of interactions between control and autonomy . Domestication is an ongoing process. From a biological point of view, it is revealed by morphological changes – and these changes do not stop at the moment the biological product of the process has become phenotypically (or even genetically) different from its wild ancestor. These changes are continuous, producing ever new lineages, varieties, or breeds whose standards evolve as well. Unlike those models that view domestication as a defined state and speak of “semi-”, “proto-”, and “full” domestication (the full version given by animals bred in the modern West, of course!), the domestication we are looking at here is never- ending. Domestication is one of the ways in which the living world evolves, where the main selection pressure, whether intentional or not, is the anthropic milieu with its socio-cultural complexity and its fluctuations across time (e.g. Vigne 2011). Domestication is polymorphous : from a genetic point of view, Laurent A. F. Frantz and Greger Larson note in their chapter that “domestication is highly species-specific”. Genomes are affected in dramatically different ways reflecting specific “human-animal relationships during domestication”. Even at a species level, ethnology and, more recently, archaeology ( Vigne et al. 2011 ) have shown that different types of relationship alternate, causing plants and animals to go through the different states of autonomous agent, partner, demanding tyrant, slave, food, and raw material. Various contributions from ethnologists and archaeologists highlight modes of domestication based not on the separation of the domestic from the wild but on their complementarity and continuity . In their chapter, Marie Balasse and her colleagues have managed to describe, thanks to stable-isotope data continuously Introduction 11 stored in tooth enamel throughout the life, the types of interactions that pigs had with humans during the Neolithic or the Middle Ages. In various archaeological assemblages, they identified the concomitant presence of wild, feral, and domestic pigs, showing that, for a long time, the pig evolved between commensalism and feralization, as was also the case in many parts of the world (Albarella et al. 2007). In the case of plants, the discontinuity between crops and weeds, which seems obvious in a world where modern agriculture sees all spontaneous intrusion onto a field as a risk of yield loss, is not, however, a universal model. As Amy Bogaard and her colleagues show, Neolithic farmers in the Near East and Europe gathered, stored, and sometimes cultivated “opportunistic” weed plants adapted to soil disturbed by agriculture. The ethnography of Éric Garine and his colleagues describes a similar situation with the Duupa of Cameroon: they practice shifting cultivation whereby domestic and wild plants – two porous categories – are used complementarily. In the Upper Palaeolithic, the appearance of the first biologically defined dogs was preceded, and made possible, by a long history of ritualized socialization between humans and wolves, as suggested by the archaeological discoveries of Mietje Germonpré and her colleagues. Nicolas Lescureux’s chapter explores the multiple interactions, complementarities and hybridizations between dogs and wolves, all the way to the present day, which are not easily accepted by conserva- tionist policies based on a domestic-wild divide. Nicolas Lainé’s ethnography shows that the relationship between the Khamti and their elephants is based on the circulation between tamed individuals used for work and the forest universe: the tamed are captured from among the wild, and some tamed males are then led to inseminate wild females. Unpredictable and open, the domestication link is fragile and can break in a very short period of time, making it radically different from the biotic interactions described by ecology. Since the Neolithic Age, technological innovation has continuously led humans, animals, and plants to modify their mutual adaptations. Sandrine Lagneaux tells us that the relationship between farmers and their cows has been unpredictably and drastically changed by the introduction of robots on farms: it can either intensify or disappear and cause disasters. This fragility of the co-operative link in labour with animals is highlighted by the study authored by Jocelyne Porcher and Sophie Nicod. In Portugal, mechanization and the economic crisis have led owners to abandon their horses in wild areas, where they often become the prey of wolves. The end of labour and the breaking of the domestic link are nothing like the “liberation” imagined by the antispeciesists.

An entanglement of human and nonhuman agencies It is the great lesson of the myths of horticulturists, farmers, and herders about their relationship with their plants and animals: humans are not enough to create and maintain a domesticatory link. Non-human agencies, those of the environment as well as the plants and animals themselves, are also needed. They may be the 12 Charles Stépanoff and Jean-Denis Vigne animals’ ancestors or owners, as shown in the chapters by Éric Garine and Nicolas Lainé. This view stems from the repeated experience that human will is not enough for a species to be successfully tamed. The Egyptians and many other peoples have, to this date, tried and always failed to domesticate many species, including the gazelle. Advanced archaeological knowledge has shown that the domestication of several important species such as the cat, the pig, and maybe even the dog has been the result not so much of human intentionality as of those species’ attraction to anthropized environments, at least in the initial stages of the process. Biologists have indicated that certain dispositions are necessary on the part of the plants and animals for them to become domestic. As early as 1865, the anthropologist Francis Galton reckoned that, in order to be domesticated, animals “should have an inborn liking for man” (Galton 1865, 137; see Zeuner 1963 ). Jeannin’s chapter discusses what cognitive ethology can tell us today about the basis for this “fondness for man” in the case of dogs. To what extent is the notion of fondness applicable to plants? Weeds are plants adapted to soil disturbed by agriculture; some are “opportunistic”, while others, in the words of Amy Bogaard and colleagues, “mimic” crops. Species such as rye and oats were initially commensal before being, probably inadvertently, domesticated. Yildiz Aumeeruddy-Thomas’s chapter shows that, in Morocco, fig trees can spontaneously grow at the foot of a wall, in which case they are seen as a kind of gift that needs to be protected. People see trees as “powerful agents that mediate between humans, as well as between humans and supranatural forces, thus ensuring reciprocal exchanges” (Aumeeruddy-Thomas, this volume). The diagram in Figure 0.1 sums up this interpenetration between human and nonhuman initiatives and dispositions in hybrid communities within which domestication may (or may not) crystallize.

Figure 0.1 Hybrid communities. Introduction 13 Recently, in social anthropology, the anthropo- and male-centric model of “man domesticating nature” has strongly been challenged under the effect of eco- feminist approaches which work with concepts such as “companion species”, “bound”, and “entanglement” ( Haraway 2003 ), and claim that “human nature is an interspecies relationship” (Tsing 2012, 141). In this volume, Charlotte Brives has taken her inspiration from the notion of “companion species” in order to describe the “domestication of viruses”, taking the concept to its farthest limits. She shows how viruses, pathogenic or otherwise, have for millions of years made their way into our DNA and our history. These approaches raise the question of the reciprocity of the transformations within domestication. In the classic narrative of the Neolithic Revolution, it is in their role as property and accumulation objects that crops and livestock make their effects felt on human societies, triggering the rise of inequalities. The relational approach developed by Gala Argent in her chapter is completely different: she examines the effect of the horse’s central presence among Altai early nomads not as an object but as a living being endowed with a particular behaviour. She suggests that the meaning of the synchrony within the movement present in equine sociality has been able to influence the way people perceived and organized their relationship with their environment. Domestication has a social and relational, but also technological and somatic impact on humans, which is often neglected but is, in fact, symptomatic of the fact that humans are just as much subject to the domesticatory interaction as animals and plants are. The beginnings of milk exploitation, which we now know to have been, in the West, contemporaneous with the early domestication of ungulates 10,000 years ago (Debono Spiteri et al. 2016; Gillis et al. 2017), have had a strong impact on food practices, leading to the development of new techniques (cheesemaking) and, in several parts of the world, the rapid selection of genes that enable lactose digestion at an adult age (Mélanie Roffet-Salque et al. this volume). Generally speaking, different traits of the domestication syndrome, such as skeleton gracilization and brain size shrinking, are partly found in humans, and are interpreted as a process of “self-domestication” ( Leach 2003 ; Theofanopoulou et al. 2017). Humans are not only agents but also objects of the domestication transformations.

Domus and the common ground of domestication We often forget that domestication implies three terms: the human, the non- human, and the domus where their entangled lives take root in the long term. That is why, beyond the classic domesticator-domesticated dyad, we develop a triadic perspective on the interactional dynamics which transform the human, the non-human, and their shared habitat. This shared habitat may be the human body itself (microbes, parasites, etc.), the human home (pets, commensals, etc.), or the whole anthroposystem (crop plants, weeds, livestock, etc.). The anthroposystem or socioecosystem is forged not only by humans but also by the domesticated species which modify the environment and have to adapt one to another, thus making 14 Charles Stépanoff and Jean-Denis Vigne it necessary to use global comprehensive approaches (Vigne 2015). For example, as illustrated by the chapter authored by Konstantin Klokov and Vladimir Davy- dov, in Siberia, the domus that reindeer are a part of also includes dogs, and vice versa: whereas their wild ancestors used to be predators and prey, they now have to submit to “reciprocal learning”. This co-presence implies both constraints and opportunities, and Klokov and Davydov draw a vast overview of the extraordi- narily diverse “human-dog-reindeer communities” that indigenous people of the Far North have derived from it. Beyond the domus lies the issue of the common ground of an association between different species. The search for favourable soil and the clearing of forests are means of literally producing common ground, attractive to both cultivated plants and their cultivators. Communication with animals relies on mediators such as nomadic routes (Stépanoff et al. 2017); work tools, including sleds (Klokov and Davydov, this volume), saddles, or plows; and chants and pet-directed talk (Lainé, Jeannin, this volume). The common ground can also consist of an activity such as milk exchange, as shown in the chapter by Mélanie Roffet-Salque and her colleagues. In various societies, humans and animals exchange milk, forming a relationship based on intimacy and mutual trust. Whilst the theory of the “Secondary Products Revolution” presumed that the relationship with domestic ungulates must have, at first, been centered on meat production, having then, over several thousands of years, become more refined and turned to the exploitation of living animals for milk, wool, and labour (Sherratt 1983), it seems that milk may have been a catalyst for the initial domestication of cattle, sheep, and goats. The consequences of this change in perspective are anthropologically important: the starting point of domestication was probably not the treatment of animals as matter but as living animals playing the role of co-operating partners (milking, traction, riding, prestige).

Hybrid communities In ecology, the notions of biocenosis or of community refer to a set of species interacting in a generally natural environment or space. Throughout the twentieth century and still often to day, the human species was mostly excluded. On the contrary, when social scientists speak of “community”, they mean a group of humans which excludes any other species. It is obvious that we need more inclusive concepts and it is precisely among the farming and herding populations that we can find them. For example, the Tuva nomadic herders of southern Siberia use the phrase aal-kodan to refer to their living place. This lexical pair includes the encampment area and the grazing land around it, the families living in it, as well as their herds. All the elements of this community are interdependent: human misbehaviour toward spirits can lead to illness within the herd whilst, conversely, the sanctified horse ydyk is charged with watching over the prosperity and health of the whole aal-kodan . Similar notions of inclusive communities can be found throughout the world. The Andean ayllu community refers to the relationships between humans and Introduction 15 non-humans (animals, plants, mountains, etc.) living on a given territory (De La Cadena 2010 ). The notion of “hybrid communities”, introduced by Lestel with regard to the interactions between apes and primatologists (Lestel 2004), can be extended to refer to what is described by the Tuva notion of aal-kodan and the Quechua notion of ayllu : forms of long-standing multispecies associations between humans, plants, and animals around a shared habitat . Domestication is often preceded and made possible by various kinds of hybrid communities that are not in themselves tantamount to domestication. That is why we have adopted an open approach, both ecological and cultural, which takes into account the porousness of different types of interaction, such as: parasitism (an association profitable to the parasite and harmful to the host), commensalism (an association not harmful to the host), mutualism (an interaction profitable to both parties), symbiosis (an interaction indispensable to both parties), and taming (an individual relationship without the notion of intergenerational continuity). We include in “hybrid communities” these various forms of association in the vicinity of domestication. The archaeology of the island of Cyprus has revealed the intensity of the relations that may precede animal domestication. The pre-Neolithic hunter-gatherers who colonized the island brought on their boats wild boars which they released to the wild on the island and were then able to use as game, 1,000 years before domestic pigs made their appearance ( Vigne et al. 2009 ). Carrying boars by sea implies a high degree of management and familiarity. Without the humans, the boars alone would not have come to the island of Cyprus but, conversely, without the boars, people would have found it hard to feed themselves and settle on the island in the long run. Hunters and boars formed a particular type of hybrid community characterized by intermittent links. Humans and animals can be involved in different kinds of partnership which are not domestication but are worthy to compare with domestication. In his chapter, Edmond Dounias draws an overview of the amazing cases where wild untamed animals assist humans in their foraging tasks on the basis of a mutualistic relationship. May such relationships have evolved in the past into domestication processes? It is possible, but not necessary: in the cases presented by Dounias, animal auxiliaries, like dolphin and honeybirds, have remained wild and do not share habitat with humans. Biotechnologies open up new gaps in categorial boundaries, as shown by the contribution of Sandrine Dupé in her chapter “From parasite to reared insect: humans and mosquitoes in Réunion”. Mosquitos have a close relationship with humans: they share our dwellings and partly feed on our blood, which makes them ectoparasites. Recently though, some mosquito populations have displayed certain domestication traits: humans ensure they reproduce and feed them because their multiplication has become in their interest. Even though these insects belong to a species parasitic on humans, have these particular mosquito populations become domestic? But how can we ecologically contemplate a population whose only mission is to destroy one’s species! This case multiplies the paradoxes around the boundaries of domestication, whilst at the same time it strongly emphasizes the 16 Charles Stépanoff and Jean-Denis Vigne originality behind the domesticating and, more generally, anthropic phenomenon within a biosphere which is more than 3 billion years old. By exploring hybrid communities, our work calls for the complementarity of biological and social approaches, the studies included here being representative of new forms of research which join up concepts, questions, and methods from different disciplines. The hybridized realities of domestication require innovative hybrid concepts and approaches.

References Albarella, U., Dobney, K., Ervynck, A., and Rowley-Conwy, P., eds., 2007. Pigs and Humans, 10,000 Years of Interaction . Oxford: Oxford Univ. press. Almathen, F., Charruau, P., Mohandesan, E., Mwacharo, J. M., . . . and Burger, P. A., 2016. Ancient and Modern DNA Reveal Dynamics of Domestication and Cross-continental Dispersal of the Dromedary. Proceedings of the National Academy of Sciences , 113(24), 6707–6712. Anderson, D. G., Loovers, J. P. L., Schroer, S. A., and Wishart, R. P., 2017. Architectures of Domestication: On Emplacing Human-animal Relations in the North. Journal of the Royal Anthropological Institute , 23(2), 398–416. Armstrong Oma, K., 2010. Between Trust and Domination: Social Contracts Between Humans and Animals. World Archaeology , 42(2), 175–187. Bogoras, W., 1904–1909. The Chukchee. The Jesup North-Pacific Expedition 7. Leiden, NY: Brill. Budiansky, S., 1999. The Covenant of the Wild: Why Animals Chose Domestication . New Haven and London: Yale University Press. Buffon, G.-L. L., 1753. Histoire naturelle, générale et particulière , vol. 4. Paris: Imprim- erie Royale. Callou, C., 2003. De la garenne au clapier: Étude archéozoologique du Lapin en Europe occidentale . Paris: Muséum national d’Histoire naturelle. Cauvin, J., 1994. Naissance des divinités, naissance de l’agriculture: La revolution des symboles au Néolithique . Paris: CNRS. Childe, V. G., 1942. What Happened in History . Harmondsworth and New York, NY: Penguin. Clutton-Brock, J., 1989. Introduction to domestication. In: Clutton-Brock, J. ed. The Walk- ing Klarder. Patterns of Domestication, Pastoralism, and Predation . London: Unwin Hyman, 7–9. Cohen, M. N. and Armelagos, G. J., 1984. Paleopathology at the Origins of Agriculture . New York, NY: Academic Press. Coppinger, R. P. and Smith, C. K., 1989. A Model for Understanding the Evolution of Mammalian Behavior. Current Mammalogy , 2, 335–374. Darwin, C., 1868. The Variation of Animals and Plants Under Domestication . London: Murray. Davis, S. J. R. and Valla F. R., 1978. Evidence for Domestication of the Dog 12,000 Years Ago in the Natufian of Israel. Nature , 276, 608–610. Debono Spiteri, C., Gillis, R. E., Roffet-Salque, M., Castells Navarro, L., Guilaine, J., Manen, C., . . . Evershed, R. P., 2016. Regional Asynchronicity in Dairy Production and Processing in Early Farming Communities of the Northern Mediterranean, Proceedings of the National Academy of Science USA , 113(48), 13594–13599. De La Cadena, M., 2010. Indigenous Cosmopolitics in the Andes: Conceptual Reflections Beyond ‘Politics’. Cultural Anthropology , 25(2), 334–370. Introduction 17 Descola, P., 1986. La nature domestique: symbolisme et praxis dans l’écologie des Achuar . Paris: Maison des sciences de l’homme. Diamond, J., 1987. The Worst Mistake in the History of the Human Race. Discover (May), 64–66. Diamond, J., 1998. Guns, Germs, and Steel: A Short History of Everybody for the Last 13,000 Years . London: Vintage. Digard, J.-P., 1990. L’homme et les animaux domestiques: Anthropologie d’une passion . Paris: Fayard. ———, 1999. Les Français et leurs animaux . Paris: Fayard.Drake, A. G., 2011. Dispelling Dog Dogma: An Investigation of Heterochrony in Dogs Using 3D Geometric Morpho- metric Analysis of Skull Shape. Evolution & Development , 13(2), 204–213. Ducos, P., 1978. ‘Domestication’ Defined and Methodological Approaches to Its Recogni- tion in Faunal Assemblages. In: Meadow, R. H. and Zeder, M. A., eds. Approaches to Faunal Analysis in the Middle East . Harvard: Peabody Museum Bulletin 2, 53–56. Duméril, A., 1860. Procès verbal de la séance publique annuelle de la Société impériale zoologique d’acclimatation. Bulletin de la Société impériale zoologique d’acclimatation , 7, I–VI. Dupire, M., 1962. Peuls nomades: étude descriptive des Wodaabe du Sahel nigérien . Paris: Karthala. Dwyer, P. D., 1996. Boars, Barrows, and Breeders: The Reproductive Status of Domestic Pig Populations in Mainland New Guinea. Journal of Anthropological Research , 52(4), 481–500. Erikson, P., 1987. De l’apprivoisement à l’approvisionnement: chasse, alliance et familiarisation en Amazonie amérindienne. Techniques & Culture , 9, 105–140. Fijn, N., 2011. Living With Herds: Human-Animal Coexistence in Mongolia . Cambridge: Cambridge University Press. Francione, G. L., 2009. Animals as Persons: Essays on the Abolition of Animal Exploita- tion . New York, NY: Columbia University Press. Fuller, D. Q., 2010. An Emerging Paradigm Shift in the Origins of Agriculture. General Anthropology , 17(2), 1–12. Galton, F., 1865. The first steps towards the domestication of animals. Transactions of the Ethnological Society of London , 3, 122–138. Geoffroy Saint-Hilaire, I., 1861. Acclimatation et domestication des animaux utiles . Paris: Librairie agricole de la maison rustique. Gillis, R. E., Kovacikova, L., Bréhard, S., Guthmann, E., Vostrvska, I., Nohalova, H., . . . Vigne, J.-D., 2017. The evolution of dual meat and milk cattle husbandry in Linearband- keramik societies. Proceedings of the Royal Society, B , 284, 20170905. Golovnev, A. V., 1995. Govoriashchie kul’tury: Traditsii samodiicev i ugrov . Ekaterinburg: RAN. Hamilton, C., Gemenne, F., and Bonneuil, C., eds., 2015. The Anthropocene and the Global Environmental Crisis: Rethinking Modernity in a New Epoch . London: Routledge. Harari, Y. N., 2014. Sapiens: A Brief History of Humankind . London: Vintage. Haraway, D. J., 2003. The Companion Species Manifesto: Dogs, People, and Significant Otherness . Chicago, IL: Prickly Paradigm Press. Haudricourt, A., 1962. Domestication des animaux, culture des plantes et traitement d’autrui. L’Homme , 2(1), 40–50. Hemmer, H., 1990. Domestication: The Decline of Environmental Appreciation . Cam- bridge: Cambridge University Press. Ingold, T., 1980. Hunters, Pastoralists, and Ranchers: Reindeer Economies and Their Transformations . Cambridge: Cambridge University Press. 18 Charles Stépanoff and Jean-Denis Vigne ———., 1996. Growing Plants and Raising Animals: An Anthropological Perspective on Domestication. In: Harris, D.R., ed. The Origins and Spread of Agriculture and Pasto- ralism in Eurasia . London: UCL, 12–24. Kallis, G., 2007 When Is It Coevolution? Ecological Economics , 62(1), 1–6. Kaminski, J., Riedel, J., Call, J., and Tomasello, M., 2005. Domestic Goats, Capra Hircus, Follow Gaze Direction and Use Social Cues in an Object Choice Task. Animal Behav- iour , 69(1), 11–18. Kluyver, T. A., Jones, G., Pujol, B., Bennett, C., Mockford, E. J., Charles, M., Rees, M., and Osborne, C. P., 2017. Unconscious Selection Drove Seed Enlargement in Vegetable Crops. Evolution Letters , 1(2), 64–72. Knight, J., ed., 2005. Animals in Person: Cultural Perspectives on Human-Animal Intima- cies . Oxford: Berg. ———., 2012. The Anonymity of the Hunt. Current Anthropology , 53(3), 334–355. Larson, G. and Fuller, D. Q., 2014. The Evolution of Animal Domestication. Annual Review of Ecology, Evolution, and Systematics , 45(1), 115–136. Latour, B., 1993. We Have Never Been Modern . New York, NY: Harvester Wheatsheaf. ———., 2017. Facing Gaia: Eight Lectures on the New Climatic Regime .Cambridge: Pol- ity Press. Leach, H., 2003. Human Domestication Reconsidered. Current Anthropology , 44(3), 349–368. ———., 2007. Selection and the Unforeseen Consequences of Domestication. In: Cassidy, R. and Mullin, M., eds. Where the Wild Things Are Now, Domestication Reconsidered . Oxford: Berg, 71–100. Lestel, D., 2004. L’animal singulier . Paris: Seuil. ———., 2015. A quoi sert l’homme ? Paris: Fayard. Lévi-Strauss, C., 1962. La Pensée Sauvage . Paris: Plon. ———., 1964. Mythologiques. 1, Le cru et le cuit . Paris: Plon. ———., 2001. La leçon de sagesse des vaches folles. Études rurales , 157–158, 9–14. Marshall, F. B., Dobney, K., Denham, T., and Capriles, J. M., 2014. Evaluating the Roles of Directed Breeding and Gene Flow in Animal Domestication. Proceedings of the National Academy of Sciences , 111(17), 6153–6158. Nawroth, C., Ebersbach, M., and von Borell, E., 2014. Juvenile Domestic Pigs (Sus Scrofa Domestica) Use Human-given Cues in an Object Choice Task. Animal Cognition , 17(3), 701–713. Nibert, D., 2013. Animal Oppression and Human Violence: Domesecration, Capitalism, and Global Conflict . New York, NY: Columbia University Press. Oelschlaeger, M., 1991. The Idea of Wilderness: From Prehistory to the Age of Ecology . New Haven, CT and London: Yale University Press. Outram, A. K., Stear, N. A., Bendrey, R., . . . and Evershed, R. P., 2009. The Earliest Horse Harnessing and Milking. Science , 323(5919), 1332–1335. Owen, J., Dobney, K., Evin, A., Cucchi, T., Larson, G., and Strand Vidarsdottir, U., 2014. The Zooarchaeological Application of Quantifying Cranial Shape Differences in Wild Boar and Domestic Pigs (Sus Scrofa) Using 3D Geometric Morphometrics. Journal of Archaeological Science , 43, 159–167. Poplin, F., 1976. Origine du porc. Ethnozootechnie , 16, 6–13. ———., 1979. Origine du mouflon de Corse dans une nouvelle perspective paleon- tologique: par marronnage. Annales de Génétique et Sélection animale , 11, 133–143. Price, T. D. and Bar-Yosef, O., 2011. The Origins of Agriculture: New Data, New Ideas: An Introduction to Supplement 4. Current Anthropology , 52(S4), S163–S174. Introduction 19 Proops, L., Walton, M., and McComb, K., 2010. The Use of Human-given Cues by Domes- tic Horses, Equus caballus , During an Object Choice Task. Animal Behaviour , 79(6), 1205–1209. Regan, T., 2005. Empty Cages: Facing the Challenge of Animal Rights . Lanham, MD: Rowman & Littlefield. Rindos, D., 1984. The Origins of Agriculture: An Evolutionary Perspective . New York: Academic Press. Russel, N., 2007. The Domestication of Anthropology. In : Cassidy, R. and Mullin, M. eds., Where the Wild Things Are Now . Domestication Reconsidered . Oxford, New York: Berg, 27–48. Scott, J., 2017. Against the Grain: A Deep History of the Earliest States . New Haven, CT: Yale University Press. Shepard, P., 1996. The Only World We’ve Got: A Paul Shepard Reader. San Francisco: Sierra Club Books. Sherratt, A., 1983. The Secondary Exploitation of Animals in the Old World. World Archaeology , 15(1), 90–104. Sigaut, F., 1988. Critique de la notion de domestication. L’Homme , 28(108), 59–71. Singer, P., 1976. Animal Liberation: A New Ethics for Our Treatment of Animals . London: Jonathan Cape. Smith, B. D. and Zeder, M. A., 2013. The Onset of the Anthropocene. Anthropocene , 4, 8–13. Stépanoff, C., 2017. The Rise of Reindeer Pastoralism in Northern Eurasia: Human and Animal Motivations Entangled. Journal of the Royal Anthropological Institute , 23(2), 376–396. Stépanoff, C., Marchina, C., Fossier, C., and Bureau, N., 2017. Animal Autonomy and Intermittent Coexistences: North Asian Modes of Herding. Current Anthropology , 58(1), 57–81. Theofanopoulou, C., Gastaldon, S., O’Rourke, T., Samuels, B. D., Messner, A., Martins, P. T., s . . . Boeckx, C., 2017. Self-domestication in Homo Sapiens: Insights From Com- parative Genomics. PLoS ONE , 12(10), e0185306. Tsing, A., 2012. Unruly Edges: Mushrooms as Companion Species. Environmental Humanities , 1(1), 141–154. Vigne, J.-D., 1988. Les Mammifères post-glaciaires de Corse , étude Archéozoologique . Paris: CNRS. ———., 2008. Zooarchaeological Aspects of the Neolithic Diet Transition in the Near East and Europe, and Their Putative Relationships With the Neolithic Demographic Transi- tion. In: Bocquet-Appel, J.-P. and Bar-Yosef, O., eds. The Neolithic Demographic Tran- sition and Its Consequences . New York, NY: Springer, 179–205. ———., 2011. The Origins of Animal Domestication and Husbandry: A Major Change in the History of Humanity and the Biosphere. Comptes rendus biologies, 334(3), 171–181. ———., 2015. Early Domestication and Farming: What Should We Know or Do for a Bet- ter Understanding? Anthropozoologica , 50(2), 123–150. Vigne, J.-D., Carrère, I., Briois, F., and Guilaine, J., 2011. The Early Process of the Mam- mal Domestication in the Near East: New Evidence From the Pre-Neolithic and Pre- Pottery Neolithic in Cyprus. Current Anthropology , S52(4), S255–S271. Vigne, J.-D., Evin, A., Cucchi, T., Dai, L., Yu, C., Hu, S., . . . Yuan, J., 2016. Earliest ‘Domestic’ Cats in China Identified as Leopard Cat (Prionailurus bengalensis ). PLoS ONE , 11(1), e0147295. Vigne, J.-D., Guilaine, J., Debue, K., Haye, L., and Gérard, P., 2004. Early Taming of the Cat in Cyprus. Science , 304, 259. 20 Charles Stépanoff and Jean-Denis Vigne Vigne, J.-D., Zazzo, A., Saliège, J.-F., Poplin, F., Guilaine, J., and Simmons, A., 2009. Pre- Neolithic Wild Boar Management and Introduction to Cyprus More than 11,400 Years Ago. Proceedings of the National Academy of Sciences , 106(38), 16135–16138. Weissbrod, L., Marshall F., Valla, F., Khalaily, H., Bar-Oz, G., Auffray, J.C., . . . Cucchi, T., 2017. The origins of house mice in ecological niches created by settled hunter-gatherers in the Levant 15,000 y ago. Proceedings of the National Academy of Sciences , 114 (16), 4099–4104. Willcox, G. and Stordeur, D., 2012. Large-scale Cereal Processing Before Domestication During the Tenth Millennium Cal BC in Northern Syria. Antiquity , 86, 99–114. Zeder, M. A., 2015. Core Questions in Domestication Research. Proceedings of the National Academy of Sciences , 112(11), 3191–3198. Zeuner, F. E., 1963. A History of Domesticated Animals . New York, NY: Harper & Row. Zohary, D., Tchernov, E., and Kolska Horwitz, L., 1998. The Role of Unconscious Selec- tion in the Domestication of Sheep and Goat. Journal of Zoology London, 245, 129–135. Part I Liminal processes Beyond the wild and the domestic

1 A genetic perspective on the domestication continuum

Laurent A. F. Frantz and Greger Larson

Introduction Beginning with dogs over 15,000 years ago, the domestication of plants and animals has played a key role in the development of modern societies. The importance of domestication for archeology, biology and humanities means that it has been studied extensively. Over the past decades, new conceptual models of animal domestication have emerged from the archeological literature that is trying to break away from anthropocentric view of domestication. These models describe domestication as a gradual process, but also question the ubiquity of human intent during the early stages of the process (Vigne 2011; Zeder 2011; Ervynck et al. 2001 ). Under this view, domestication is thought as a continuous process by which a species alters, voluntarily or involuntarily, the phenotype of another and assumes a significant degree of influence over its care and reproduction (Zeder 2015). These models stand in stark contrast with the traditional views that involve intent and strong directed breeding from humans ( Larson and Fuller 2014 ; Clutton-Brock 1992 ; Bökönyi 1989 ). This view of domestication also refrains from generalising a process that might be highly species specific (Vigne 2011). This is done by broadly defining three pathways to domestication: commensal, prey and directed pathway, each with different varying degree of intent (Zeder 1982). Under the commensal pathway, for example, domestication is most diffuse and involves little intent, while under the directed domestication is fully intentional with human directed breeding starting during early phases of domestication. In the latter model, humans apply a strong dichotomy between what is considered wild and domestic. The striking morphological and behavioural changes associated with the process of domestication also make it an excellent model to study evolution. Under- standing the broad mechanisms that allow for fast evolution in domestic taxa is a central question in evolutionary biology. In contrast with species specific pathways to domestication advocated by some archeologists and biologists, however, are often more interested in drawing conclusions that are globally applicable to the process than species specific. For example, multiple studies have focused on understanding common biological features (e.g. developmental processes) underlying a general “domestication syndrome” in both plants ( Sakuma et al. 2011 ; 24 Laurent A. F. Frantz and Greger Larson Allaby 2014 ) and animals ( Wilkins et al. 2014 ; Sánchez-Villagra et al. 2016 ). This is an interesting contrast between “archeological” and “biological” views of domestication. Genetics have been instrumental in studying biological aspects of domestication. For simplicity, however, genetic studies generally model domestication as a process involving strong bottlenecks (also called founder events), reproductive isolation between wild and domestic, and strong artificial selection. These models provide a great framework for geneticists to address questions such as geographic origin, strength of selection and demography of domestication (Frantz et al. 2015). This view of domestication, however, implies strong control of the breeding of domestic species as well as strong artificial selection that also stands in contrast with the archaeological view of domestication as a more fluid process. Armed with new DNA sequencing technologies, geneticists are now able to sequence and analyse large scale genetic information from both modern and archeological samples. These genomes provide an ever greater resolution to address fundamental questions about domestication, and in particular to test the veracity of recent theories. Here we first discuss how genetics can provide key information to test various models of domestication, and then review the degree to which genetic evidence supports these models.

Models of domestication and their expectations Conceptual models of domestication based on the idea of intent involve a combination of key characteristics: 1) a founder effect (also called a bottleneck; see Figure 1.1a ) as only a few individuals are domesticated ( Larson and Burger 2013 ; Eyre-Walker et al. 1998 ; Doebley et al. 2006 ); 2) strong artificial selection for traits that define domestic species (e.g. behaviour; Figure 1.1b) (Trut et al. 2009; Trut 1999); 3) reproductive isolation enforced between wild and domestic population to facilitate trait differentiation (e.g. keep animals tame; Figure 1.1c) (Mar- shall et al. 2014); and 4) simpler genetic architecture of traits (Andersson-Eklund 2013). While these characteristics provide a great framework for geneticists to study domestication, as key events (e.g. geographical origin and timeframe of domestication), they do occasionally contradict the expectations associated with the more diffuse models of domestication proffered by archeologists (Marshall et al. 2014; Vigne 2011; Zeder 2011). The first and maybe most obvious difference between these models of domestication is the expected strength of bottlenecks. Figure 1.1a is a schematic of a model in which a wild population undergoes a “domestication” bottleneck as a result of a human-directed domestication. In this example, the wild population prior to domestication has two alleles (black and white). After domestication, however, only the black remains, as the result of a founder effect (since the founding population possessed only the black allele). While drastic, this example highlights one major expectation of a directed model with human intent: a severe bottleneck resulting in a significant loss of genetic diversity. Bottlenecks can, as we will see later in this chapter, also result from directed breeding (for example, A genetic perspective on domestication 25 a) b) Ne Ne

Time Time

Domestication Domestication

Breed formation Bottleneck and/or Strong selection

Wild Domestic Wild Domestic Population Population Population Population c) Ne

Time

Domestication

Wild Domestic Population Population

Figure 1.1 Schematic of various models of domestication and their effect on genetic diversity. Black and white dots represent alleles at a hypothetical locus in the genome of a domestic species. Ne is the effective population size: the number of breeding individuals in an idealised population (different from the census population size). Ne is directly related to genetic diversity. In these schematics, population size increases and decreases due to various events (e.g. bottlenecks) over time. This has an effect of overall genetic diversity. Figure 1.1a basic model involving a founder effect and/or strong selection; Figure 1.1b a model involving a mild domestication bottleneck and a strong bottleneck during breed formation – this illustrates that the genetic diversity of the modern population under this model and a model in Figure 1.1a are the same, yet the models are different; Figure 1.1c this last panel represents the effect of gene flow from a wild to a domestic population following a bottleneck. We can see in this schematic how genetic diversity can increase in the domestic population as a result of gene flow. during the formation of a specific breed) as highlighted in Figure 1.1b . In a more diffuse model, however, in which domestication proceeds unintentionally, the expectation is that a domestication bottleneck will be less severe (or even absent). This is because the process will happen somehow more “naturally” and thus more individuals will be involved in the founding of the domestic population and will have more chance to capture an equal number of black and white alleles. Conscious selection by humans for specific traits is also expected to reduce genetic diversity. In our example in Figure 1.1a , the white allele could be represented by aggressive individuals. Strong “artificial” selection would be applied against these individuals in a domestic population. On the other hand, if domestication 26 Laurent A. F. Frantz and Greger Larson proceeds unintentionally, selection is not expected to be as strong simply because individuals that are, for example, aggressive (or simply less tame) will not be directly selected against. The expectations of a directed model of domestication, as opposed to a more diffuse model with no intent, are rather similar to the one of severe/mild bottlenecks: differential level of genetic diversity (Figures 1.1a, 1.1b ). Another expectation of directed models of domestication is the maintenance of reproductive isolation (prevent breeding between wild and domestic population). As we show in Figure 1.1c, gene flow (exchange of genetic material) between wild and domestic populations can increase genetic diversity. If conscious selection is applied on a domestic population, people might have imposed strict reproductive isolation from wild populations. Preventing gene flow from wild animals would provide a framework for faster selection of domestic traits ( Frantz et al. 2015 ). On the other hand, without conscious selection, domestication proceeds similarly to parapatric speciation models which can (and often do) involve gene flow ( Larson and Fuller 2014; Frantz et al. 2015; Vigne 2011). By taking advantage of modern DNA sequencing, geneticists are now able to generate the necessary data, including genomes from modern and archeological samples, to test these key expectations underlying intentional models of domestication.

Domestication, genetic diversity, bottlenecks and breed formation Genetic studies of domestication often assume a domestication bottleneck in both plants ( Xu et al. 2011 ; Zhu et al. 2007 ) and animals such as dogs ( Freedman and Wayne 2017; Frantz et al. 2016; Lindblad-Toh et al. 2005), cattle (Bollongino et al. 2012; Scheu et al. 2015; Beja-Pereira et al. 2006; MacLeod et al. 2013), goats (Gerbault et al. 2012) and rabbits (Carneiro et al. 2014). The idea of a domestication bottleneck is so embedded in the genetic literature that it is often taken as prior knowledge (Bollongino et al. 2012; Scheu et al. 2015; Gerbault et al. 2012). For at least some species, genetic data has supported the idea of a lower genetic diversity in domestic populations, especially dogs (Marsden et al. 2016), cattle (Bollongino et al. 2012; MacLeod et al. 2013) and rabbits (Carneiro et al. 2014). Dogs are the most drastic example, especially as breed dogs are significantly more inbred than wolves ( Frantz et al. 2016 ; Marsden et al. 2016 ). Interestingly, in dogs, most genetic diversity is thought to have been lost during severe bottlenecks linked to breed formation ( Wang et al. 2013 ; Wang et al. 2014 ) and therefore may not reflect processes that took place during domestication itself. Domestication may have induced a reduction of genetic variation of less than 20% (Wang et al. 2014). Recent breeding efforts in purebred dogs have thus resulted in more severe bottlenecks than domestication itself, as indicated by the higher level of genetic diversity observed in free living dogs (village dogs; see Marsden et al. 2016). This more recent and drastic bottleneck, however, does not explain all differences between wolves and dogs, suggesting that a milder bottleneck took place during domestication ( Marsden et al. 2016 ). The exact timing of this earlier bottleneck and therefore its association with domestication has yet to be tested with ancient DNA (as discussed later). A genetic perspective on domestication 27 While reduced genetic diversity is a clear pattern common to multiple domestic species, it is not generally applicable to all domestic animals. Indeed, recent studies of pigs ( Frantz et al. 2015 ; Bosse et al. 2014 ), bees ( Wallberg et al. 2014 ), chickens (Rubin et al. 2010) and yaks (Qiu et al. 2015) found no reduction of genetic diversity in domestic populations, suggesting that no domestication bottlenecks have taken place in these species. Pigs and chickens are thought to have been domesticated via a commensal pathway (Larson and Fuller 2014; Zeder 2011 ). Under this pathway, early stages of domestication are thought to be fully unintentional. The lack of a bottleneck in these species might reflect the specific- ity of this pathway. Multiple factors, however, complicate the interpretation the genetic diversity level in pigs, chickens and bees, especially their propensity to interbreed with wild stock ( Eriksson et al. 2008 ; Frantz et al. 2015 ; Wallberg et al. 2014). Outbreeding with wild populations could indeed inflate genetic diversity as pictured in Figure 1.1c . Another factor that complicates interpretation of these findings is the genetic diversity in wild populations. Indeed, most analyses inferring bottlenecks often rely on comparing genetic diversity with wild counterparts. This, however, can be problematic if wild populations have themselves undergone recent bottlenecks, or in more dramatic cases, have gone extinct as it is the case for aurochs. This is an issue for many species such as wolves ( Fan et al. 2015 ; Freedman et al. 2014 ), wild boars (Bosse et al. 2012; Frantz et al. 2015) and wild horses (Schubert et al. 2014; Jónsson et al. 2014). Understanding the dynamics in wild populations is therefore key to characterise the severity of bottlenecks during domestication.

Reproductive isolation Archeologists have suggested that interbreeding between wild and domestic forms may be an important feature of most domestication processes (Marshall et al. 2014). The consequences of interbreeding are numerous. First, as mentioned previously (and as shown in Figure 1.1c ), gene-flow can increase the level of genetic diversity in a population. Indeed, if gene flow from a wild population into a domestic population takes place after a (domestication) bottleneck, genetic diversity will be restored, to some extent (depending on the number of wild individuals that contribute to the domestic population). Gene flow from wild stock can also affect the phenotypic evolution of a domestic species. This is because gene flow can have homogenising effect – in other words, gene flow could prevent the domestic population from sustaining divergent phenotypes. By reducing gene flow, it is therefore possible to accelerate the process of phenotypic differentiation (e.g. behavioural) between wild and domestic. Similarly, the establishment of barrier to gene flow is thought to be an important process during speciation, especially when differentiated, yet closely related, species/subspecies occur in parapatric conditions ( Feder et al. 2012 ). The processes that reduce gene flow between differentiated populations also enforce reproductive isolation. The characteristics and types of reproductive isolation have been studied extensively in wild populations ( Seehausen et al. 2014 ). 28 Laurent A. F. Frantz and Greger Larson Reproductive isolation can be classified roughly into three categories: extrinsic, postzygotic/prezygotic and intrinsic (Seehausen et al. 2014). Two species are intrinsically isolated if they cannot reproduce due to intrinsic biological specific- ity or genetic incompatibility (e.g. difference in ploidy). On the other hand, two species might be partially isolated due to, for example, behavioural differences (extrinsic prezygotic isolation) or due the low fertility of their offspring (extrinsic postzygotic isolation). In general, while in many cases permeable, these barriers to gene flow are a key aspect of speciation as they are necessary for two species to maintain their different phenotype. Just as in speciation, the maintenance of reproductive isolation is key for domestication (Price 2002). This is particularly true under a model in which domestication is intentional (Marshall et al. 2014). Under this scenario, reproductive isolation in domestic animals can be imposed by humans (in domestic populations, though this is not always the case) and by natural selection (in wild populations). These barriers to gene flow can be seen as extrinsic prezygotic (domestic) or postzygotic (wild), and will assist in the maintenance of domestic traits. In domestic populations, humans can act as an extrinsic prezygotic isolation mechanism by preventing mating with wild individuals. In wild populations, however, mating is not necessarily prevented, but the fitness of a cross between wild and domestic offspring might be lower (due to relaxed selection in domestic population). This can be seen as postzygotic isolation. Maintaining reproductive isolation between wild and domestic populations, however, varies dramatically among species. Many plant species have and continue to interbreed with their wild relatives, including banana and plantain (Andersson and de Vicente 2010), barley (Baum et al. 1992), cassava (Duputié et al. 2007), maize (da Fonseca et al. 2015), pearl millet (Miura and Terauchi 2005), potato (Scurrah et al. 2008), sorgho (see Garine, this volume) and apples ( Cornille et al. 2012 ). Levels of outcrossing are, however, highly variable among species (Andersson and de Vicente 2010). Many species have also formed partial intrinsic reproductive barriers with their wild progenitors, often as a result of changes in level of ploidy such as in cotton and sweet potato ( Andersson and de Vicente 2010 ). A recent review, however, suggested that only 38% of all cultivated plants are partially or fully reproductively isolated from their wild ancestor ( Dempewolf et al. 2012 ). Outcrossing between wild and domestic also seems common in domestic animals, including cattle (Park et al. 2015), pigs (Frantz et al. 2015), dromedary camels (Almathen et al. 2016), cats (Ottoni et al. 2017) and horses ( Lippold et al. 2011 ). Armed with whole genome sequences, geneticists are now able to evaluate the degree of gene flow between wild and domestic populations in greater detail. Taken together, these studies have demonstrated how the rate of genetic exchange varies among species. In pigs and dogs, for example, levels of genetic exchange are dramatically different. Besides widespread gene flow from dogs into wolf populations (see Lescureux, this volume; (Fan et al. 2015) and a few examples of deliberate crossing between wolves and dogs, such as the Sarloos breed ( Morris 2008), global dog populations form a genetically homogeneous cluster with little A genetic perspective on domestication 29 evidence for outbreeding with wolves (Freedman et al. 2014; Frantz et al. 2016; Fan et al. 2015; Wang et al. 2016). The lack of outbreeding with wolves is not solely a recent phenomenon as demonstrated by ancient DNA studies demonstrating that Neolithic dogs from Europe (>5,000 years old) also show little evidence of interbreeding with wild canids (Frantz et al. 2016; Thalmann et al. 2013; Bot- igué et al. 2017 ). Pigs, on the other hand, show a drastically different pattern, with high level of outcrossing between wild and domestic populations (Frantz et al. 2015). While it is clear that outcrossing has taken place within the last few hundred years (since formation of some breeds), this phenomenon is likely to have taken place for thousands of years as suggested by ancient DNA analysis (Larson et al. 2007). Interestingly, the direction of interbreeding between wild and domestic forms seems to be reversed in pigs and canids. Interbreeding takes place most often in canids from domestic dogs into wolf populations (see Lescureux, this volume), while in suids, it often takes place from wild boars into domestic pigs ( Frantz et al. 2015 ). The ubiquity of outcrossing raises the question of how the highly derived phenotypic characteristics associated with domestication are maintained in the face of gene flow. One possibility is that recurrent strong artificial selection for domestic traits (e.g. behaviour) reduced the effects of gene flow which would otherwise prevent the domestic traits from increasing in frequency ( Larson and Fuller 2014 ; MacHugh et al. 2017 ). Speciation theory suggests that this should leave footprints of selection in the region of the genome that controls for the phenotypic differences between two species ( Turner et al. 2005 ). These regions have coined “islands of speciation”. In pigs, while the domestic/wild status of an individual is virtually impossible to determine within most of the genome, some regions show a clear pattern of genetic homogeneity among domestic pigs ( Frantz et al. 2015). One of these regions, for example, is linked to the higher number of vertebrae found in domestic pigs. This provides an interesting similarity between the processes of domestication and speciation, suggesting the potential existence of “islands of domestication” in the genomes of domestic species (Frantz et al. 2015 ). It is thus clear that outcrossing must have pervasive effects on phenotypes that differentiate domestic and wild populations (e.g. behaviour). It is interesting to note that such mechanisms may be responsible for the limited amount of gene flow from wolf to dogs suggested previously. Gene flow, however, may also be beneficial in some cases; for example, for traits that are highly variable among breeds (e.g. coat colour). As such, gene flow may provide novel genetic variation in domestic populations and enhance the potential for breeders to diversify traits. In addition, gene flow from wild populations may also provide the means for domestic populations to adapt to novel environments, such as in high altitude dogs ( vonHoldt et al. 2017 ; Miao et al. 2017 ) or by augmenting immune capabil- ity (Vilà et al. 2005) or by providing novel coat colours (Linderholm et al. 2016; Fang et al. 2009 ). While our understanding of outcrossing between wild and domestic populations is still rudimentary, its clear association with domestication suggests some 30 Laurent A. F. Frantz and Greger Larson interesting parallels with the process of speciation. As for speciation, gene flow between domestic and wild population appears to be variable among species, suggesting that domestication is highly species specific. Gene flow is also suggestive that the modern dichotomy between wild and domestic populations may be something relatively recent, suggesting that intent might not be an important feature of domestication ( Larson and Fuller 2014 ).

Artificial and relaxed selection As for bottlenecks and reproductive isolation, strong and deliberate artificial selection is also often associated with intentional domestication as a means to select for traits that differentiate domestic populations from their wild progenitors. Strong artificial selection can affect overall genetic diversity during bottlenecks (Figure 1.1a) and leaves specific footprints in the genome (localised reduction of genetic diversity, also known as a selective sweep). These can be detected and used to identify genes under selection ( Vitti et al. 2013 ). The application of these methods, combined with recent sequencing technologies, have revealed many regions under selection in the genome of domestic plants including rice, tomato and maize (Lin et al. 2014; Xu et al. 2011; Lai et al. 2010), and animals including chickens, rabbits, dogs and horses ( Rubin et al. 2010 ; Rubin et al. 2012 ; Carneiro et al. 2014; Frantz et al. 2015; Axelsson et al. 2013; Librado et al. 2017). Genetic mapping (also known as linkage mapping) has also been employed to characterise the genetic architecture of traits that differentiate within domestic populations (Andersson-Eklund 2013; Schoenebeck et al. 2012)). Although less common, this method has also been applied to compare wild and domestic populations ( Andersson-Eklund et al. 1998 ). Scans for the footprints of selection are often conducted by comparing genomes derived from modern populations ( Girdland Flink et al. 2014 ). While selection can often be reasonably well identified, these methods possess little power to infer the time at which selection occurred (Peter et al. 2012). This is problematic because comparing modern genomes does not provide the means to disentangle selection that happened recently (e.g. breed improvement) from selection that took place during the early phases of domestication ( Larson and Fuller 2014 ), especially given that selection during early phases are not necessarily expected to lead to strong morphological differentiation ( Zohary et al. 1998 ; Vigne 2008 ). The recent generation of ancient DNA (aDNA) sequences is starting to provide clues regarding the importance of artificial selection in the past, including the type of traits that were and were not selected early on. In chickens, for example, two recent aDNA studies (Girdland Flink et al. 2014; Loog et al. 2017) demonstrated that the selection for an allele involved in seasonal reproduction that differenti- ates modern chickens and Red Jungle fowls (Rubin et al. 2010) only began driving the allele to fixation about 1,000 years ago. Similarly, alleles linked to better starch digestion in dogs, a phenotype that strongly differentiate them from wolves ( Axelsson et al. 2013 ) only arouse >7,000 years after they were domesticated ( Ollivier et al. 2016 ). A genetic perspective on domestication 31 Artificial selection is not the only type of selection expected during domestication. By moving plants and animals from natural to different environments (e.g. introduction of wild boars to Cyprus), especially to anthropogenic niches (Zeder 2016), humans may have relaxed natural selection constraints on domestic species. Relaxed selection is an interesting proxy to study domestication from a genetic perspective due to some relatively well understood and expected outcomes (MacHugh et al. 2017). Relaxation of selective constraints predicts that more deleterious mutations (mutations that have a negative impact on an individual) would segregate within domestic than wild populations. These deleterious mutations are also expected to be more common in populations that have undergone recent bottlenecks. Studying relaxed selection therefore provides an ideal proxy to test the importance that human directly played during early stages of domestication, because deleterious alleles would be expected to increase gradually as animals or plants are moved from natural to anthropogenic ecosystems and/or because of bottlenecks. Studies of horses and dogs have shown that these species carry more deleterious mutations than expected under the relaxed selection hypothesis (Librado et al. 2017; Marsden et al. 2016). This, however, does not generally apply to all domesticated species (Moray et al. 2014). In addition, issues such already low level of deleterious mutations in wild progenitors (e.g. Przewalski’s horses) can also complicate inference of relaxed selection. Nevertheless, a recent study of ancient horse genomes demonstrated clear patterns that can be associated with relaxed selection by showing that modern horses have accumulated more deleterious mutations in their genome than early domestic populations ( Librado et al. 2017 ). Relaxed selection, as measured via deleterious load, can also be difficult to disentangle from other mechanisms. For example, the frequency of deleterious mutations can increase together with other other alleles that are under positive selection (increasing in frequency) – a process called genetic hitchhiking. Lastly, populations that are not at an equilibrium (e.g. that underwent recent expansion or bottleneck) will have skewed frequency of neutral variants ( Brandvain and Wright 2016 ), which can affect methods that compare the frequency of neutral and deleterious variation in a population. Future research is needed to investigate how to overcome these challenges.

The rise of ancient genomes Using modern genetic data alone, domestication can appear to be an abrupt process during which a species has been drastically altered relative to its wild pro- genitor. Modern data, however, only provides a snapshot of what archaeologists showed to be a long and continuous process. In many cases, modern genomes can only inform us about the recent past (e.g. selection scan and genes linked to improvement). Recent technological advances in DNA sequencing and methods to recover DNA from bones provide the means to solve these issues. Through sequencing of ancient genomes of domestic and wild species, it is possible to obtain direct measurements of genetic diversity, relaxed/artificial selection and 32 Laurent A. F. Frantz and Greger Larson gene flow in ancient populations. By comparing these measurements through time, it is possible to assess how gradual the process of domestication has been. Besides a few examples, ancient DNA studies have yet to tackle fundamental questions about the process of domestication. This can be mostly attributed to the prohibitive cost of associated with generating a high quality nuclear genome from ancient DNA. Genome sequencing effort of ancient horses (Librado et al. 2017; Schubert et al. 2014), dogs (Frantz et al. 2016) and maize (da Fonseca et al. 2015) have, nevertheless, generated sufficient information to draw some interesting conclusions about the process of domestication in these species by demonstrating, for example, a sharp reduction of genetic diversity in both modern dogs and horses and a sharp increase of genetic load in the latter.

Conclusions Even the most recent studies have only begun to scratch the surface of the enor- mous potential of genomes as a window into domestication. An interesting trend is nevertheless emerging: domestication is highly species specific. Bottlenecks, reproductive isolation and strong selection are by no means ubiquitous features, but rather are linked to species (and even region) specific pathways of domestication. The process of domestication in animals such as dogs, pigs, chickens, bees and many other plant and animal species affected their genomes in dramatically different ways. This probably reflects specific human-animal/plant relationships during domestication, as has been suggested by numerous archaeologists. Like speciation, domestication comes in many forms, parapatric or allopatric, with intrinsic and/or extrinsic reproductive isolations, strong or diffuse selection. Inter- estingly, this highly taxon-specific depiction of domestication that is emerging from field such as genetics and archeology provide a scientific basis for the soci- ologists that question the usefulness of imposing a general wild/domestic dichotomy in wildlife conservation (see Lescureux, this volume). Genome-wide information, especially ancient genomes, offers an incredible tool to study domestication. For example, studying relaxed selection through time, but especially during early stages, will provide critical evidence for the importance that selection and intent played in specific domestication processes but also for our understanding of fundamental evolutionary questions (Brandvain and Wright 2016). Understanding past genetic diversity in domestic taxon will not only allow us to understand the importance that genetic variability played during domestication, but also to understand the future and potential sustainability of our farming systems.

References Allaby, R.G., 2014. Domestication syndrome in Plants. In : C. Smith, ed. Encyclopedia of Global Archaeology . New York, NY: Springer, 2182–2184. Almathen, F., Charruau, P., Mohandesan, E., . . . Burger, P.A., 2016. Ancient and modern DNA reveal dynamics of domestication and cross-continental dispersal of the dromedary. Proceedings of the National Academy of Sciences of the United States of America , 113 (24), 6707–6712. A genetic perspective on domestication 33 Andersson-Eklund, L., 2013. Molecular consequences of animal breeding. Current Opin- ion in Genetics & Development , 23 (3), 295–301. Andersson-Eklund, L., Marklund, L., Lundström, K., Haley, C.S., Andersson, K., Hans- son, I., . . . Andersson, L., 1998. Mapping quantitative trait loci for carcass and meat quality traits in a wild boar x Large White intercross. Journal of Animal Science , 76 (3), 694–700. Andersson, M.S. and de Vicente, M.C., 2010. Gene Flow Between Crops and Their Wild Relatives . Baltimore, USA: JHU Press. Axelsson, E., Ratnakumar, A., Arendt, M.-L., . . . Lindblad-Toh, K., 2013. The genomic signature of dog domestication reveals adaptation to a starch-rich diet. Nature , 495 (7441), 360–364. Baum, M., Lagudah, E.S., and Appels, R., 1992. Wide crosses in cereals. Annual Review of Plant Physiology and Plant Molecular Biology , 43 (1), 117–143. Beja-Pereira, A., Caramelli, D., Lalueza-Fox, C., . . . Bertorelle, G., 2006. The origin of European cattle: Evidence from modern and ancient DNA. Proceedings of the National Academy of Sciences of the United States of America , 103 (21), 8113–8118. Bökönyi, S., 1989. Definitions of animal domestication. The Walking Larder: Patterns of Domestication, Pastoralism, and Predation , 24–27. Bollongino, R., Burger, J., Powell, A., Mashkour, M., Vigne, J.-D., and Thomas, M.G., 2012. Modern taurine cattle descended from small number of near-eastern founders. Molecular Biology and Evolution , 29 (9), 2101–2104. Bosse, M., Megens, H.-J., Madsen, O., Frantz, L.A.F., Paudel, Y., Crooijmans, R.P.M.A., and Groenen, M.A.M., 2014. Untangling the hybrid nature of modern pig genomes: A mosaic derived from biogeographically distinct and highly divergent Sus scrofa populations. Molecular Ecology , 23 (16), 4089–4102. Bosse, M., Megens, H.-J., Madsen, O., . . . Groenen, M.A.M., 2012. Regions of homozygosity in the porcine genome: Consequence of demography and the recombination landscape. PLoS Genetics , 8 (11), e1003100. Botigué, L.R., Song, S., Scheu, A., . . . Veeramah, K.R., 2017. Ancient European dog genomes reveal continuity since the Early Neolithic. Nature Communications , 8, ncomms16082. Brandvain, Y., and Wright, S.I., 2016. The limits of natural selection in a nonequilibrium world. Trends in Genetics: TIG , 32 (4), 201–210. Carneiro, M., Rubin, C.-J., Di Palma, F., Albert, F.W., . . . and Andersson, L., 2014. Rabbit genome analysis reveals a polygenic basis for phenotypic change during domestication. Science , 345 (6200), 1074–1079. Clutton-Brock, J., 1992. The process of domestication. Mammal Review , 22 (2), 79–85. Cornille, A., Gladieux, P., Smulders, M.J.M., . . . Giraud, T., 2012. New insight into the history of domesticated apple: Secondary contribution of the European wild apple to the genome of cultivated varieties. PLoS Genetics , 8 (5), e1002703. da Fonseca, R.R., Smith, B.D., Wales, N., . . . Gilbert, M.T.P., 2015. The origin and evolution of maize in the Southwestern United States. Nature Plants , 1, 14003. Dempewolf, H., Hodgins, K.A., Rummell, S.E., Ellstrand, N.C., and Rieseberg, L.H., 2012. Reproductive isolation during domestication. The Plant Cell , 24 (7), 2710–2717. Doebley, J.F., Gaut, B.S., and Smith, B.D., 2006. The molecular genetics of crop domestication. Cell , 127 (7), 1309–1321. Duputié, A., David, P., Debain, C., and McKey, D., 2007. Natural hybridization between a clonally propagated crop, cassava (Manihot esculenta Crantz) and a wild relative in French Guiana. Molecular Ecology , 16 (14), 3025–3038. Eriksson, J., Larson, G., Gunnarsson, U., . . . Andersson, L., 2008. Identification of the yellow skin gene reveals a hybrid origin of the domestic chicken. PLoS Genetics , 4 (2), e1000010. 34 Laurent A. F. Frantz and Greger Larson Ervynck, A., Hongo, H., Dobney, K., and Meadow, R., 2001. Born free ? New evidence for the status of Sus scrofa at Neolithic Çayönü Tepesi (Southeastern Anatolia, Turkey). Paléorient , 27 (2), 47–73. Eyre-Walker, A., Gaut, R.L., Hilton, H., Feldman, D.L., and Gaut, B.S., 1998. Investiga- tion of the bottleneck leading to the domestication of maize. Proceedings of the National Academy of Sciences of the United States of America , 95 (8), 4441–4446. Fan, Z., Silva, P., Gronau, I., . . . Wayne, R.K., 2015. Worldwide patterns of genomic variation and admixture in gray wolves. Genome Research , 26 (2), 163–173. doi:10.1101/ gr.197517.115. Fang, M., Larson, G., Ribeiro, H.S., Li, N., and Andersson, L., 2009. Contrasting mode of evolution at a coat color locus in wild and domestic pigs. PLoS Genetics, 5 (1), e1000341. Feder, J.L., Egan, S.P., and Nosil, P., 2012. The genomics of speciation-with-gene-flow. Trends in Genetics: TIG , 28 (7), 342–350. Frantz, L.A.F., Mullin, V.E., Pionnier-Capitan, M., . . . and Larson, G., 2016. Genomic and archaeological evidence suggest a dual origin of domestic dogs. Science , 352 (6290), 1228–1231. Frantz, L.A.F., Schraiber, J.G., Madsen, O., . . . Groenen, M.A.M., 2015. Evidence of long- term gene flow and selection during domestication from analyses of Eurasian wild and domestic pig genomes. Nature Genetics , 47 (10), 1141–1148. Freedman, A.H., Gronau, I., Schweizer, R.M., . . . Novembre, J., 2014. Genome sequencing highlights the dynamic early history of dogs. PLoS Genetics , 10 (1), e1004016. Freedman, A.H. and Wayne, R.K., 2017. Deciphering the origin of dogs: From fossils to genomes. Annual Review of Animal Biosciences , 5, 281–307. Gerbault, P., Powell, A., and Thomas, M.G., 2012. Evaluating demographic models for goat domestication using mtDNA sequences. Anthropozoologica , 47 (2), 64–76. Girdland Flink, L., Allen, R., Barnett, R., . . . Larson, G., 2014. Establishing the validity of domestication genes using DNA from ancient chickens. Proceedings of the National Academy of Sciences of the United States of America , 111 (17), 6184–6189. Jónsson, H., Schubert, M., Seguin-Orlando, A., . . . Orlando, L., 2014. Speciation with gene flow in equids despite extensive chromosomal plasticity. Proceedings of the National Academy of Sciences of the United States of America , 111 (52), 18655–18660. Lai, J., Li, R., Xu, X., . . . Wang, J., 2010. Genome-wide patterns of genetic variation among elite maize inbred lines. Nature Genetics , 42 (11), 1027–1030. Larson, G., Albarella, U., Dobney, K., . . . Cooper, A., 2007. Ancient DNA, pig domestication, and the spread of the Neolithic into Europe. Proceedings of the National Academy of Sciences of the United States of America , 104 (39), 15276–15281. Larson, G. and Burger, J., 2013. A population genetics view of animal domestication. Trends in Genetics: TIG , 29 (4), 197–205. Larson, G. and Fuller, D.Q., 2014. The evolution of animal domestication. Annual Review of Ecology, Evolution, and Systematics , 45 (1), 115–136. Librado, P., Gamba, C., Gaunitz, C., . . . and Orlando, L., 2017. Ancient genomic changes associated with domestication of the horse. Science , 356 (6336), 442–445. Lin, T., Zhu, G., Zhang, J., . . . Zhang, Z., 2014. Genomic analyses provide insights into the history of tomato breeding. Nature Genetics , 46 (11), 1220–1226. Lindblad-Toh, K., Wade, C.M., Mikkelsen, T.S., . . . and Lander, E.S., 2005. Genome sequence, comparative analysis and haplotype structure of the domestic dog. Nature , 438 (7069), 803–819. Linderholm, A., Spencer, D., Battista, V., . . . Larson, G., 2016. A novel MC1R allele for black coat colour reveals the Polynesian ancestry and hybridization patterns of Hawaiian feral pigs. Royal Society Open Science , 3 (9). A genetic perspective on domestication 35 Lippold, S., Knapp, M., Kuznetsova, T., . . . Hofreiter, M., 2011. Discovery of lost diversity of paternal horse lineages using ancient DNA. Nature Communications , 2, 450. Loog, L., Thomas, M.G., Barnett, R., . . . Eriksson, A., 2017. Inferring allele frequency trajectories from ancient DNA indicates that selection on a chicken gene coincided with changes in medieval husbandry practices. Molecular Biology and Evolution, 34 (8), 1981–1990. MacHugh, D.E., Larson, G., and Orlando, L., 2017. Taming the past: Ancient DNA and the study of animal domestication. Annual Review of Animal Biosciences , 5, 329–351. MacLeod, I.M., Larkin, D.M., Lewin, H.A., Hayes, B.J., and Goddard, M.E., 2013. Infer- ring demography from runs of homozygosity in whole-genome sequence, with correc- tion for sequence errors. Molecular Biology and Evolution , 30 (9), 2209–2223. Marsden, C.D., Ortega-Del Vecchyo, D., O’Brien, D.P., . . . Lohmueller, K.E., 2016. Bot- tlenecks and selective sweeps during domestication have increased deleterious genetic variation in dogs. Proceedings of the National Academy of Sciences , 113 (1), 152–157. Marshall, F.B., Dobney, K., Denham, T., and Capriles, J.M., 2014. Evaluating the roles of directed breeding and gene flow in animal domestication. Proceedings of the National Academy of Sciences of the United States of America , 111 (17), 6153–6158. Miao, B., Wang, Z., and Li, Y., 2017. Genomic analysis reveals hypoxia adaptation in the Tibetan Mastiff by introgression of the gray wolf from the Tibetan plateau. Molecular Biology and Evolution , 34 (3), 734–743. Miura, R. and Terauchi, R., 2005. Genetic control of weediness traits and the maintenance of sympatric crop-weed polymorphism in pearl millet (Pennisetum glaucum). Molecular ecology , 14 (4), 1251–1261. Moray, C., Lanfear, R., and Bromham, L., 2014. Domestication and the mitochondrial genome: Comparing patterns and rates of molecular evolution in domesticated mammals and birds and their wild relatives. Genome Biology and Evolution , 6 (1), 161–169. Morris, D., 2008. Dogs: The Ultimate Dictionary of Over 1,000 Dog Breeds. London, UK: Trafalgar Square. Ollivier, M., Tresset, A., Bastian, F., . . . Hänni, C., 2016. Amy2B copy number variation reveals starch diet adaptations in ancient European dogs. Royal Society Open Science , 3 (11), 160449. Ottoni, C., Van Neer, W., De Cupere, B., . . . and Geigl, E.-M., 2017. The palaeogenetics of cat dispersal in the ancient world. Nature Ecology & Evolution , 1 (7). doi:10.1038/ s41559-017-0139. Park, S.D.E., Magee, D.A., McGettigan, P.A., . . . MacHugh, D.E., 2015. Genome sequencing of the extinct Eurasian wild aurochs, Bos primigenius, illuminates the phylogeogra- phy and evolution of cattle. Genome Biology , 16 (1), 234. Peter, B.M., Huerta-Sanchez, E., and Nielsen, R., 2012. Distinguishing between selective sweeps from standing variation and from a de novo mutation. PLoS Genetics, 8 (10), e1003011. Price, E.O., 2002. Animal Domestication and Behavior . New York, NY: CABI Publishing. Qiu, Q., Wang, L., Wang, K., . . . Liu, J., 2015. Yak whole-genome resequencing reveals domestication signatures and prehistoric population expansions. Nature Communica- tions , 6, 10283. Rubin, C.-J., Megens, H.-J., Martinez Barrio, A., . . . Andersson, L., 2012. Strong signatures of selection in the domestic pig genome. Proceedings of the National Academy of Sciences of the United States of America , 109 (48), 19529–19536. Rubin, C.-J., Zody, M.C., Eriksson, J., . . . Andersson, L., 2010. Whole-genome resequencing reveals loci under selection during chicken domestication. Nature , 464 (7288), 587–591. 36 Laurent A. F. Frantz and Greger Larson Sakuma, S., Salomon, B., and Komatsuda, T., 2011. The domestication syndrome genes responsible for the major changes in plant form in the Triticeae crops. Plant & Cell Physiology , 52 (5), 738–749. Sánchez-Villagra, M.R., Geiger, M., and Schneider, R.A., 2016. The taming of the neural crest: A developmental perspective on the origins of morphological covariation in domesticated mammals. Royal Society Open Science , 3 (6), 160107. Scheu, A., Powell, A., Bollongino, R., . . . Burger, J., 2015. The genetic prehistory of domesticated cattle from their origin to the spread across Europe. BMC Genetics, 16, 54. Schoenebeck, J.J., Hutchinson, S.A., Byers, A., . . . and Ostrander, E.A., 2012. Variation of BMP3 contributes to dog breed skull diversity. PLoS Genetics , 8 (8), e1002849. Schubert, M., Jónsson, H., Chang, D., . . . and Orlando, L., 2014. Prehistoric genomes reveal the genetic foundation and cost of horse domestication. Proceedings of the National Academy of Sciences , 111 (52), E5661–E5669. Scurrah, M., Celis-Gamboa, C., Chumbiauca, S., Salas, A., and Visser, R.G.F., 2008. Hybridization between wild and cultivated potato species in the Peruvian Andes and biosafety implications for deployment of GM potatoes. Euphytica/ Netherlands Journal of Plant Breeding , 164 (3), 881–892. Seehausen, O., Butlin, R.K., Keller, I., . . . Widmer, A., 2014. Genomics and the origin of species. Nature Reviews. Genetics , 15 (3), 176–192. Thalmann, O., Shapiro, B., Cui, P., . . . Wayne, R.K., 2013. Complete mitochondrial genomes of ancient canids suggest a European origin of domestic dogs. Science , 342 (6160), 871–874. Trut, L., 1999. Early Canid Domestication: The Farm-Fox Experiment Foxes bred for tamability in a 40-year experiment exhibit remarkable transformations that suggest an interplay between behavioral genetics and development. American Scientist, 87 (2), 160–169. Trut, L., Oskina, I., and Kharlamova, A., 2009. Animal evolution during domestication: The domesticated fox as a model. BioEssays: News and Reviews in Molecular, Cellular and Developmental Biology , 31 (3), 349–360. Turner, T.L., Hahn, M.W., and Nuzhdin, S.V., 2005. Genomic Islands of speciation in anopheles gambiae. PLoS Biology , 3 (9), e285. Vigne, J.-D., 2008. Zooarchaeological aspects of the Neolithic diet transition in the near East and Europe, and their putative relationships with the Neolithic demographic transition. In : The Neolithic Demographic Transition and Its Consequences . Dordrecht: Springer, 179–205. Vigne, J.-D., 2011. The origins of animal domestication and husbandry: A major change in the history of humanity and the biosphere. Comptes Rendus Biologies, 334 (3), 171–181. Vilà, C., Seddon, J., and Ellegren, H., 2005. Genes of domestic mammals augmented by backcrossing with wild ancestors. Trends in Genetics: TIG , 21 (4), 214–218. Vitti, J.J., Grossman, S.R., and Sabeti, P.C., 2013. Detecting natural selection in genomic data. Annual Review of Genetics , 47, 97–120. vonHoldt, B., Fan, Z., Ortega-Del Vecchyo, D., and Wayne, R.K., 2017. EPAS1 variants in high altitude Tibetan wolves were selectively introgressed into highland dogs. PeerJ , 5, e3522. Wallberg, A., Han, F., Wellhagen, G., . . . Webster, M.T., 2014. A worldwide survey of genome sequence variation provides insight into the evolutionary history of the honeybee Apis mellifera. Nature Genetics , 46 (10), 1081–1088. Wang, G.-D., Xie, H.-B., Peng, M.-S., Irwin, D., and Zhang, Y.-P., 2014. Domestication genomics: Evidence from animals. Annual Review of Animal Biosciences , 2, 65–84. A genetic perspective on domestication 37 Wang, G.-D., Zhai, W., Yang, H.-C., . . . Zhang, Y.-P., 2013. The genomics of selection in dogs and the parallel evolution between dogs and humans. Nature Communications , 4, 1860. Wang, G.-D., Zhai, W., Yang, H.-C., . . . Zhang, Y.-P., 2016. Out of southern East Asia: The natural history of domestic dogs across the world. Cell Research , 26, 21. Wilkins, A.S., Wrangham, R.W., and Tecumseh Fitch, W., 2014. The ‘domestication syndrome’ in mammals: A unified explanation based on neural crest cell behavior and genetics. Genetics , 197 (3), 795–808. Xu, X., Liu, X., Ge, S., . . . Wang, W., 2011. Resequencing 50 accessions of cultivated and wild rice yields markers for identifying agronomically important genes. Nature Biotech- nology , 30 (1), 105–111. Zeder, M.A., 1982. The domestication of animals. Reviews in Anthropology , 9 (4), 321–327. Zeder, M.A., 2011. Pathways to animal domestication. In: A. Damania and P. Gepts, eds. Harlan II: Biodiversity in Agriculture: Domestication, Evolution and Sustainability . Davis, CA: University California Press, 227–229. Zeder, M.A., 2015. Core questions in domestication research. Proceedings of the National Academy of Sciences of the United States of America , 112 (11), 3191–3198. Zeder, M.A., 2016. Domestication as a model system for niche construction theory. Evolu- tionary Ecology , 30 (2), 325–348. Zhu, Q., Zheng, X., Luo, J., Gaut, B.S., and Ge, S., 2007. Multilocus analysis of nucleotide variation of Oryza sativa and its wild relatives: Severe bottleneck during domestication of rice. Molecular Biology and Evolution , 24 (3), 875–888. Zohary, D., Tchernov, E., and Kolska Horwitz, L., 1998. The role of unconscious selection in the domestication of sheep and goats. Journal of Zoology , 245 (2), 129–135.

2 Self-domestication or human control? The Upper Palaeolithic domestication of the wolf

Mietje Germonpré, Martina Lázničková-Galetová, Mikhail V. Sablin and Hervé Bocherens

1 Introduction The dog is the only species that was domesticated before the origin of agriculture. Today the global dog population is estimated to number nearly one billion individuals (Gompper, 2014a). Dogs are mainly dependent on human-derived foods for their resources ( Gompper, 2014a , 2014 b ; Coppinger and Coppinger, 2016 ). Their influence and burden on human societies and wildlife communities can thus not be underestimated. Different categories of dogs have been proposed such as owned dogs, free-ranging dogs and feral dogs; the boundaries between these categories are fluid and dogs can in the course of their life span move from one category to another (Vanak and Gompper, 2009; Coppinger and Coppinger, 2016; Boitoni et al., 2017 ). Owned dogs can assist their owners in various ways. They can provide company, function as watchdog, guard livestock, facilitate transport as beasts of burden or aid in hunting. In some cultures, they are consumed; their skin or hair can be used for the tailoring of cloth, or they can play a ritual role. The origin of the dog is currently a very active field of research (Grimm, 2015; Morell, 2015; Tarlach, 2016). During the past 20 years, a multitude of papers has been published on this topic (e.g. Vilà et al., 1997 ; Larson et al., 2012 ; Frantz et al., 2016). Recent genetic analyses have shown that the ancestors of modern dogs are fossil wolves that lived in Eurasia during the Pleistocene (Thalmann et al., 2013; Freedman et al., 2014; Skoglund et al., 2015; Frantz et al., 2016) and that genetic modifications of the coat colour and of the capabilities of digesting starch due to the domestication process were already ongoing before the Neolithic transition ( Ollivier et al., 2013 , 2016 ). Although the dog’s ancestor is now known, many questions remain, such as when, where, how and why the wolf was domesticated. One thing is certain: the dog is the oldest domesticated animal. Most researchers accept that the beginnings of this domestication took place towards the end of the Pleistocene, when human populations were still living in small groups as hunter-gatherers and agriculture was not yet practiced (e.g. Thalmann et al., 2013; Morey and Jeger, 2015). The Upper Palaeolithic, which lasted from approximately 45,000 years ago to about 11,700 years ago, is associated with Anatomically Modern Humans (AMH). This period was preceded by the Middle Palaeolithic associated with the Neanderthals. The beginnings of the Upper Palaeolithic are characterized among 40 Mietje Germonpré et al. other features by the repetitive and systematic appearance of art, such as cave paintings and mobile art that represent almost exclusively large and/or dangerous mammals (White, 2003, 2007 ; Floss and Hussain, 2015; Porr and de Maria, 2015 ). In addition, Upper Palaeolithic sites yield – in contrast to Middle Palaeo- lithic sites – important quantities of animal-based products: personal ornaments made from mammoth ivory and animal teeth, in the early Upper Palaeolithic especially from foxes, wolves and bears, and tools fabricated in ivory, antler or bone ( Vanhaeren and d’Errico, 2006 ; Pacher, 2005 ; Klein, 2009 ). The societies of the Upper Palaeolithic hunter-gatherers have been proposed to be largely egalitarian, with at their core the principle of general reciprocity, e.g. sharing of killed game ( Yamada, 2001 ). Furthermore, the human-animal relationships deciphered from the archaeological record of the Early Upper Palaeolithic from southwestern Germany have been shown to be essentially animistic (Hussain and Floss, 2015; Porr and de Maria, 2015). In addition, Peoples et al. (2016) have proposed, on the basis of a suite of phylogenetic comparative methods, that animism is the oldest trait of religious behaviour since it is shared by the most recent common ancestor of present-day hunter-gatherers. It is in the framework of an animated worldview of Upper Palaeolithic peoples, drawn upon a range of archaeozoological, archaeological and ethnographic data (cf. Germonpré and Hämäläinen, 2007 ; Hussain and Floss, 2015 ) that we intend to situate the early beginnings of the domestication of the wolf. Although the biodiversity of the Pleistocene fauna was very different from that of the extant North- ern fauna – animals like the woolly mammoth and cave lion are now extinct, and domestic ungulates did not yet exist – a range of herbivores, such as reindeer and elk, and of carnivores, from wolverine to wolf and brown bear, still survive in the north (cf. Germonpré and Hämäläinen, 2007). The ethnography of contemporary Northern peoples (e.g. Hamayon, 1990 ; Ingold, 1986 ; Willerslev, 2007 ; Willerslev et al., 2015) can help to understand how animals were perceived during the Upper Palaeolithic (Robert-Lamblin, 2001; Germonpré and Hämäläinen, 2007). In this contribution, we first review some discussions of the concept domestication. Then we detail two large canid morphotypes from the Upper Palaeolithic. Next, we discuss the two main hypotheses on the domestication of the wolf, focusing on ethnographic sources for canid-related practices among Northern hunting peoples of the modern era. Subsequently, we consider some stepping stones on the path of the domestication of the wolf. The discussion on the utility of the Palaeolithic dogs will be detailed in a forthcoming paper ( Germonpré et al., in preparation ).

2 Domestication, the utility of domestic mammals and the domestication syndrome Domestication of animals and plants is an ongoing process connecting humans with certain non-human species. The beginnings of this process are deemed as gradual, dynamic, mutual and unintentional (e.g. Bökönyi, 1989; Lien, 2015; Sau- tchak, 2016 ), and consist of a strengthening of the relationships between animals and humans (Vigne, 2015). For some, the concept of domestication implies a Self-domestication or human control? 41 protective domination (e.g. Ingold, 1986 ; Hamayon, 1990 ; Clutton-Brock, 1995 ), for others a social and ecological two-way relationship (e.g. O’Connor, 1997 ; Morey and Jeger, 2015 ) (cf. Russell, 2002 , 2012 ; Anderson, 2017 ). According to Sautchak (2016, p. 46), the scheme provided by Sigaut’s (1980) approach to domestication permits us to explore the diversity of the relationship between humans and animals. According to Sigaut (1988), human-animal relations can be situated on at least three levels, which can be mutually exclusive: utilization, familiarization and appropriation. A logical set of configurations can thus be developed which is much more varied than that which is ordered according to the polarity of wild/domesticated. Concerning utilization, Sigaut (1980 ) distinguishes four main types of products that a wild or domestic animal can provide: corporal products, energy, behavioural products and signs. All these product types should be considered on the same level in the domestication debate (Sigaut, 1980). Sautchak (2016) recommends that these products should not be seen as defining features, but rather as innovations integrated in a set of practices in specific ethnographic situations. A set of phenotypic alterations of the morphology, behaviour, physiology and reproduction can be observed across a range of domesticated species that are absent in their wild ancestors: the so-called domestication syndrome. It has been proposed that domesticated phenotypes could evolve because of correlated responses when animals were selected for tameness, reduced fear of humans and hypersociability towards humans ( Belyaev, 1979 ; Trut et al., 2009 ; Wilkins et al., 2014 ; Bélteky et al., 2016; Nelson et al., 2017; Sànchez-Villagra et al., 2016; Pendleton et al., 2017; von- Holdt et al., 2017 ). This could be due to pleiotropic effects, correlated mechanisms at the genomic level or structural variants in specific genes, or be driven by deficits in neural crest cells during embryonic development ( Wilkins et al., 2014 ; Wright, 2015; Bélteky et al., 2016; Sànchez-Villagra et al., 2016; vonHoldt et al., 2017). The modern dog is the domestic mammal that shows most of the features of this domestication syndrome, such as a reduced general body size and snout length, an increase in the snout width, an increased variability in the number of vertebrae and an increased tameness compared to its wild ancestor (Benecke, 1987; Pionnier et al., 2011; Wilkins et al., 2014; Sànchez-Villagra et al., 2016). Aberrant neural crest cells migrations and a decrease in the number of these cells have been correlated with a decrease in jaw size ( Wilkins et al., 2014 ). Furthermore, a recent study has shown that during early dog domestication, initial selection targeted genes essential to the initiation and regulation of the neural crest. This selection altered the proper development of neural crest cells-derived tissues ( Pendleton et al., 2017 ). In addition, directional selection likely targeted a unique set of linked behavioural genes that contribute to extreme sociability, permitting behavioural deviation between early dogs and ancestral wolves ( vonHoldt et al., 2017 ).

3 Two canid morphotypes in the Upper Palaeolithic We have shown, on the basis of multivariate analyses of traditional morphometrics of the skull and mandible, that in several Upper Palaeolithic sites of Europe, 42 Mietje Germonpré et al. two morphotypes of fossil large canids can be distinguished. A first morphotype is quite similar in size and shape to the recent wild wolves from Northern Eurasia, although the snout of this fossil morphotype is on average longer and wider than the muzzle of the recent Northern wolves (Germonpré et al., 2017). This morphotype has been named by us “Pleistocene wolf ” ( Figure 2.1 ). Skulls and mandibles of the second fossil morphotype, dating from the early and middle Upper Palaeolithic, have been found in European sites located above 45° North latitude: Goyet (50°N) (Figure 2.2), Hohlefels (48°N), Předmostí (49°N), Kostenki-8 (51°N) (Germonpré et al., 2009, 2012, 2015b; Camarós et al., 2016). The Palaeolithic dog morphotype has a smaller skull size and a shortened snout with a proportionally wide palate and a shorter lower jaw compared to the wild type (Germonpré et al., 2009, 2012, 2015a, 2017a; Galeta et al., submitted), features related to the domestication syndrome. Based on other fossil and modern samples of wolves and on different morphometric techniques, several authors opposed the assertion that the Palaeolithic dog morphotype falls outside the morphological variability of wolves (Boudadi-Maligne and Escarguel, 2014; Drake et al., 2015); others consider that it could represent a different ecotype of fossil wild wolves (Perri, 2016). We responded to these objections by showing that the Palaeolithic dog morphotype has a unique morphology that falls outside the size and shape variability of our sample of Pleistocene and recent Northern wolves and recent Northern indigenous dogs, implying that Palaeolithic dogs likely are not a part of the natural variation in Pleistocene wolves although they still differ from recent Northern dogs (Germonpré et al., 2015a, 2017).

4 Two hypotheses on the Upper Palaeolithic domestication of the wolf Although some Palaeolithic dogs may be, but are not necessarily, the direct ancestors of modern dogs, an unswerving relationship between Upper Palaeoli- thic humans and these animals existed as shown by the manipulated canid skulls from the Russian Eliseevichi site (Sablin and Khlopachev, 2002) and the Czech Předmostí site (Germonpré et al., 2012), two sites that are characterized by huge amounts of mammoth remains. Can this relationship be connected to a domestication process? And how was the domestication of the wolf initiated? Two main scenarios have been proposed to explain the early beginnings of this process (e.g. Russell, 2002 , 2012 ). They can be summarized as follows.

4.1 Self-domestication of wolves This scenario is based on the natural consequence of an ecological association between people and canids (Russell, 2012; Morey and Jeger, 2015). A first and obvious ecological association between wolves and humans is the competition for the same prey ( Vigne, 2015 ). Typical prey species of Upper Palaeolithic humans were the horse, the reindeer, the red deer and large bovids (e.g. Germonpré, 1996 ; Stewart, 2004 ; Jimenez et al., 2016 ). Pleniglacial wolves in Europe were Figure 2.1 Dorsal view of a Pleistocene wolf skull (Total Skull Length: 276.54 mm) from the Gravettian Předmostí site; the braincase shows a perforation on its left side. Photo credit: Oldřich Kroupa. 44 Mietje Germonpré et al.

Figure 2.2 Dorsal view of the Palaeolithic dog skull (estimated Total Skull Length: 227 mm) from the Goyet cave; this incipient dog has a calibrated age of about 36,000 years BP. Photo credit: Royal Belgian Instit.ute Natural Sciences. outcompeted by cave hyenas and fed especially upon red deer, giant deer and smaller bovids ( Bocherens, 2015 ). After the extinction of the cave hyena, Euro- pean wolves consumed primarily horses and large bovids (e.g. Germonpré et al., 2009; Bocherens, 2015). Also, fossil wolves in Eastern Beringia – a region never invaded by the cave hyena – preyed mainly on horses and bison (Leonard et al., 2007). The self-domestication model puts forward that some fossil wolves were attracted to prehistoric garbage dumps. These wolves scavenged on the remains of prey animals left by prehistoric people at human settlements. Those wolves that were less anxious and aggressive thrived; they became habituated to the human presence, adapted to the human niche and colonized the human dominated Self-domestication or human control? 45 environments, resulting in a commensal relationship. Gradually, generation after generation, the first primitive dogs emerged from this group (O’Connor, 1997; Coppinger and Coppinger, 2001 ; Zeder, 2012 ; Larson and Burger, 2013 ).

4.2 Human control of wolves by Upper Palaeolithic people Pet keeping is very well recorded in the ethnographic literature, especially among societies that lack domestic farm animals (Russell, 2012). This model proposes that Upper Palaeolithic peoples over large parts of Europe and Asia adopted wolf pups and this for several reasons. These pups likely were less than two weeks old when captured. At this young age, wolf pups are still deaf and blind. At two weeks old, wolf pups start to explore their environment using olfaction. Their critical period of socialization ends when they are about six weeks old. Exposure to other species during this period can permit social attachments ( Lord, 2013 ; Hall et al., 2015 ) to, in this case, humans. The Upper Palaeolithic hunter-gatherers probably let only the most docile wolf pups live into adulthood; the more aggressive ones were eliminated. Occasionally, the Upper Palaeolithic peoples probably allowed that the most gentile wolves in their care could mate and their offspring could live. After several generations of selection for behaviour, the first dogs emerged (Clutton-Brock, 1995; Morey, 2010). This means that in this model, in order to have a sufficient number of founders, the capture and rearing of young animals must have been a rather widespread practice during the Upper Palaeolithic. People during that time must have placed particular value on the personal interaction with animals (cf. Porr and de Maria, 2015 ), as is also shown that by the fact that, at the end of the Upper Palaeolithic, canids were buried together with humans in Europe and Southwest Asia ( Davis and Valla, 1978 ; Nobis, 1979 ; Maher et al., 2011 ).

4.3 Common core of both hypotheses Both hypotheses have in common the idea that the wolves on the way toward becoming domesticated displayed reduced fear and aggressiveness and had a reduced flight response towards humans compared to the wild type (Morey and Jeger, 2015 ). Furthermore, in both scenarios, the food these wolves were feeding upon can be supposed to be derived from prey animals killed by the Upper Pal- aeolithic hunter-gatherers. In the self-domestication hypothesis, the canids were eating food stored by the prehistoric people and/or refusal/waste. In the human control hypothesis, the canids were being fed by people.

4.4 Critique on the self-domestication hypothesis Palaeolithic hunter-gatherers probably had little surplus or refuse that could have provided food for the invasive wolves. In addition, they likely stowed away food in secure places in order to avoid uninvited guests. Northern peoples construct elevated racks and above-ground caches for food-storage or use tree caches to obstruct predators; they use also underground caches (e.g. Fair, 1997 ). After the 46 Mietje Germonpré et al. consumption of the meat, they often deposit heads and bones of prey animals on platforms or in trees to keep these body parts out of the reach of dogs and other carnivores (e.g. Tanner, 1979; Lavrillier, 2011). Furthermore, the ethnographic record shows that small communities in general do not have enough refuse to sustain many dogs. The dogs that subsist on refuse are in poor health and have a reduced life span dying at ages of not more than three years ( Lupo, 2017 and references herein). According to the Denésuliné hunters of northern Canada, wolves are shy and are reluctant to approach things that smell of humans ( Sharp and Sharp, 2015 , pp. 37–38). After game has been killed, Denésuliné hunters keep scavengers off by covering and concealing the meat that they cannot immediately transport back to the camp. This meat pile is placed at some distance of the gut pile that tends to attract scavengers and provide in this way some protection for the stacked meat ( Sharp and Sharp, 2015 , p. 9). Tanner (1979 , pp. 157–158) observed that the Cree buried the meat of the slain prey animal in the snow or covered it with spruce boughs to protect it from scavengers until they come back with enough people to transport the meat from the kill site to the camp. Nevertheless, predators regularly scavenge the remains at the kill sites ( Sharp and Sharp, 2015 ). In addition to the limited and protected food surplus/waste at Upper Palaeoli- thic campsites, another essential critique on the self-domestication model is that wolves habituated to associate humans with food are dangerous: they can attack and kill people (Linnell et al., 2002, 2003 ). It has been shown that wolves prey on humans, mostly unattended children, near or at settlements, especially when improper waste disposal methods are used or when the wolves live in the vicinity (Rajpurohit, 1999; McNay and Mooney, 2005; Behdarvand and Kaboli, 2015). People tend to react to this behaviour by exterminating adult wolves and pups (Behdarvand et al., 2014). We believe that Upper Palaeolithic people did everything possible to avoid large predators lurking around or raiding their campsites; most likely they stored food and waste in a safe place and killed young carnivores at the den sites in the neighbourhood of their camps. Occasionally, they probably spared some young and brought home very small pups and cubs (see 4.2). Surprisingly, little research has been done on the diet of these wolves that would have lived near Upper Palaeolithic hunter-gatherers. A recent study of the collagen content of the lower jaws of some large canids from Předmostí, a Gra- vettian site dated at c. 28,000 years ago, revealed that the two canid morphotypes had a separate diet. The Pleistocene wolves fed essentially on horses and mammoths; they shared the access to these carcasses, presumably at the kill sites, with smaller carnivores such as foxes and wolverines. The Palaeolithic dogs were predominantly consuming reindeer meat, in contrast with the Gravettian human who consumed mainly mammoth meat ( Bocherens et al., 2015 ). They likely were fed by the Předmostí inhabitants, were probably tethered and apparently did not have access to the mammoth bones ( Bocherens et al., 2015 ). Ethnographic comparisons of the feeding of arctic dogs are given in Bocherens et al. (2015 ). One example will suffice here: Inuit dogs were fed worn-out clothing made from the skin of the prey hunted by the Inuit (Laugrand and Oosten, 2015 p. 71). Self-domestication or human control? 47 4.5 Incentives for the adoption of young carnivores Evidence of the keeping of young carnivores, such as wolf and fox pups and bear cubs, can be found in the ethnographic literature on Northern peoples (e.g. Drucker, 1951; Prokof’yeva, 1964; Sokolova, 2000; Hamayon, 2012). Not only carnivore young were raised, but also other animals like owls, ducks, geese and eagles were reared ( Prokof’yeva, 1964 ; Sokolowa, 1982; Nelson, 1983 ). Further- more, wild animals were treated in a helpful way: Evenki put recipients filled with water near their dwellings so that migratory birds could drink when having a rest during their journey ( Lavrillier, 2013 ). The indigenous cosmologies of these peoples show several parallels. All these people were very concerned about the ritual treatment of the bones of the dead, humans and animals, so as to ensure the continuation of their rebirths ( Willerslev et al., 2015 ). They believed that the soul of the hunted animal would return to its congeners to be recycled into a new body of the same species by the master spirit “giver of game”. The hunters were very anxious to ensure reincarnation by the treatment of the carcass according to strict rules, in particular the skull and bones that, according to them, contained the soul of the animal and its potential for rebirth. These skeletal elements were reverently deposited on platforms or in trees. So, the killing of an animal is both an act of destruction and a rite of renewal since the stock of souls is limited. It is a process of endless reciprocity and exchange in which the animals willingly take part ( Ingold, 1986 ; Hamayon, 1990 , 2012 ; Brightman, 2002 ; Jordan, 2003 ; Yamada, 2001 ; Stépanoff, 2013 ; Willerslev et al., 2015 ). Hunting adult wild wolves is not easy. An efficient and safe way to obtain a wolf is to take it as a very young pup from its den. These pups can be killed the following winter to acquire their winter fur (Lescureux, 2007). John Murdoch (1892) observed in Alaska an Iñupiat family taking two wolf pups. The pups were tethered outside the village and fed until winter when their fur had grown long enough for use in trimming hoods (Murdoch, 1892, p. 263). Similarly, Siberian peoples like the Khanty and Mansi ( Prokof’yeva et al., 1964 ) and the Sel’kups (Prokof’yeva, 1964) held fox pups in captivity, fed them fish and killed them in autumn for their fur. Sometimes young carnivores, like foxes, wolves and bears, were kept in captivity by Siberian peoples to be fed as “ongon” (Zélénine, 1952) in order to appease the spirits who were waiting for the death of the hunters as compensation for the prey killed by these hunters (Hamayon, 1990, 2012 ). Young carnivores were sometimes breast-fed by the wife of the hunter ( Hamayon, 1990 ). Northern peoples carried out a ceremony when a bear was hunted by treating respectfully the skeleton, especially the skull of the slain bear, so that the soul of the animal would not avenge itself on the hunter and to ensure the availability of animals as prey in the future as the bear can be reborn and killed again. The Khanty perforated the braincase of the killed bear with a special ceremony and ate the brain in communion (Jordan, 2003). Dogs could not gnaw and defile the skull and bones of the bear. After the feast, the skeletal remains of the hunted bear must be deposited respectfully in a specific location (e.g. Hallowell, 1926; Paulson, 1965). The origin 48 Mietje Germonpré et al. of these bear-related rituals may be located in the Upper Palaeolithic (see Germon- pré and Hämäläinen, 2007 and references cited). Some peoples of East Asia such as the Nivkh, Nanay and Ainu raised bear cubs, captured after their mother was killed, in order to have a bear feast when it had grown up (e.g. Kitagawa, 1961 ; Yamada, 2001 ; Willerslev et al., 2015 ). The Ainu did not consider that they imprisoned the cub but rather that they took care of it as an invited guest (Kimura, 1999). After a few years the animal was killed during a sending-away ceremony. The bear offered its fur and meat to its hosts and the bear’s spirit returned to the sacred mountain, accompanied by prayers and gifts. The bear informed the other bear spirits about his trip so that his relatives too would visit the human world to be hunted, to bring hide and meat as a present and to receive many gifts (Kitagawa, 1961; Kimura, 1999; Yamada, 2001). The Ainu believed that the spirit of an animal lodged in its skull; during the sending-away ceremony, they sometimes made a hole in the bear’s braincase to release its spirit (Walker, 2005). The Ainu used not only bears for their ceremonies, but sometimes also owls (Yamada, 2001), foxes and raccoon dogs ( Akino, 1999 ), and wolves ( Walker, 2005 ). The Ainu likely raised canid pups for this ceremony; canid skulls were sometimes manipulated during the ceremony so that the spirit of the wolf was set free from its earthly body by perforating the braincase (Walker, 2005 pp. 92–94).

5 The early beginnings of the wolf domestication Sigaut (1980) made a very practical table with the utility of four main types of animal products that we follow and adapt here by comparing animal products that could have been of possible use in Upper Palaeolithic societies. These products from wild and captive wolves are listed in Table 2.1. They are based on data from the ethnographic literature and confronted, for certain of these products, with the possible registration of uses in the archaeological record ( Table 2.1 ). The first type of canid products corresponds to bodily products (Table 2.1). Sigaut (1980 ) argues that to obtain bodily products, it makes no difference whether the animal was killed as a wild animal or as a captive animal. We propose that Upper Palaeolithic peoples captured canid pups and raised them until

Table 2.1 Comparison of canid products (cf. Sigaut, 1980 ) in the ethnographic and archaeological (Upper Palaeolithic) record (non-exhaustive list).

Comparison of canid products Wild/captive wolf (non-exhaustive) Ethnographic Archaeological (UP) record record

Body fur (d) + + Murdoch (1892 ); Wojtal et al. (2012 ); Bogoras (1904– Wilczyński et al. 1909 ); Stefansson (2015 ) and Wissler (1919) Comparison of canid products Wild/captive wolf (non-exhaustive) Ethnographic Archaeological (UP) record record meat and brain (d) + + Charlier (2013 , Wojtal et al. (2012 ); 2015 ) Wilczyński et al. (2015 ); Germonpré et al. (2012 ); Germonpré et al. (2017b) long bones (d) + + Drucker (1951 ) Wojtal et al. (2012 ); Wilczyński et al. (2015 ); Germonpré et al. (2017b) Sign pendant/amulet (dentition/ + + metapodials) (d) Issenman (1997 ) Germonpré et al. (2012 ); Lázničková- Galetová (2016 ) ritual (skull manipulation) (d) + + Walker (2005 ) Germonpré et al. (2012 ) ritual (skull deposition in hearth) (d) − + Pitulko et al. (2012 ) “ongon” (l) + ? Zélénine (1952 ); Hamayon (2012 ) Behaviour (i) food-related cleaning waste/ + (captive wolf) ? being fed (l) Murdoch (1892 ) (ii) territorial instincts guarding/ ? ? sentinel (l) (iii) social aspects company/pet (l) + (captive wolf) (+) Drucker (1951 ); Lescureux (2007 ) invited guest + (+) Walker (2005 )

UP: Upper Palaeolithic l: living d: dead +: clear evidence −: absence of evidence (+): likely ?: possible 50 Mietje Germonpré et al. the following autumn or winter, when they killed the grown canids to harvest the fur. In this way, the Upper Palaeolithic hunter-gatherers would have guaranteed access to a sufficient number of skins to tailor cold weather clothing. The interest of Upper Palaeolithic people in wolf skins can be deduced from wolf bones bearing cut marks related to skinning discovered at Gravettian sites in Central Europe ( Wojtal et al., 2012 ; Wilczyński et al., 2015 ). In addition, Collard et al. (2016) have shown that canid bones are significantly more present in early Upper Palaeolithic sites than in Middle Palaeolithic sites. Also, Patou-Mathis (2012 ) has demonstrated that in Eastern Europe, canid remains are much rarer in Middle Pal- aeolithic sites than in Upper Palaeolithic sites. This discrepancy probably reflects the use of specialized cold weather clothing such as garments with fur trim during the Upper Palaeolithic (Collard et al., 2016). Indeed, the ethnographic record shows that canid pups are regularly taken captive and raised for their fur (see 4.5). The killed canid bodies could at the same time be a source of meat and long bones (Table 2.1). Cut marks on long bones of Pleistocene wolves from Gravettian sites in Central Europe indicate that their meat and brains were consumed and their bones used for the manufacture of tools ( Wojtal et al., 2012 ; Wilczyński et al., 2015 ; Germonpré et al., 2017b). A second type of canid products is related to signs ( Table 2.1 ). However, in contrast with Sigaut (1980 ), we propose to include here also skeletal parts. Therefore, some functions grouped under this type can overlap with bodily functions. These products, like teeth and skulls, can be obtained from wild or captive canids. Per- forated canid teeth are present in several European Upper Palaeolithic sites, especially those dating from the early and middle Upper Palaeolithic and from higher latitudes ( Vanhaeren and d’Errico, 2006 ; Lázničková-Galetová, 2016 ; Lázničková- Galetová et al., 2016). These modified teeth have been interpreted as signalling an ethnic dimension (Vanhaeren and d’Errico, 2006) or as amulets (Germonpré et al., 2012). At the middle Upper Palaeolithic site of Yana, a site in northern Siberia with direct evidence of mammoth hunting ( Nikolskij and Pitulko, 2013 ), a wolf skull together with its lower jaw was deposited in a hearth, probably as part of a ritual handling (Pitulko et al., 2012). At the Gravettian Předmostí site, a number of canid remains bearing sign-related modifications has been unearthed (Germonpré et al., 2012). Several perforated canine teeth were identified as coming from Pleisto- cene wolves ( Germonpré et al., 2012 ). A Pleistocene wolf skull was at the time of its death manipulated and modified by prehistoric humans (Figure 2.1): its braincase was perforated on its left side ( Germonpré et al., 2012 ), in a way akin to the perforations executed during bear and wolf sending-away ceremonies of the Ainu (Walker, 2005) or during Khanty bear rituals (Jordan, 2003). The ethnographic literature of Northern peoples shows that such treatments of carnivore skulls can have a ritual meaning associated with the sending of souls, and that in some societies the carnivores were captured young and raised in captivity until adult (see 4.5 and third type of canid products [Table 2.1]: “invited guest”). A third type of canid products is related to the behaviour of the canids ( Table 2.1 ). Several subtypes are proposed by Sigaut (1980 ) – (i) food-related: Wild canids can scavenge on the waste of abandoned campsites and kill sites. However, it Self-domestication or human control? 51 is probable that Upper Palaeolithic people managed the storage of food and the disposal of waste in such a way that large predators were not attracted to enter inhabited campsites to feed there. On the other hand, it is likely that Pleistocene wolves and other predators scavenged the kill sites (cf. Bocherens et al., 2015) (see also 4.4). At campsites, captive canids could have been fed or could eventually have had access to prey remains and offal; (ii) territorial instincts: Possibly, captive wolves could notify through vocalizations their caretakers of approaching carnivores; (iii) social aspects: Recent experiments have shown that, thanks to intensive human socialization, young non-domesticated wolves can form attachments to their human caretakers (e.g. Hall et al., 2015 ). In the ethnographic literature, ample evidence exists that canid pups can function as a companion animal (see references in Table 2.1 ), sometimes in anticipation of the full development of the fur (e.g. Sokolowa, 1982 p. 167). Not only canids, but also other carnivores were taken care of. Inuit sometimes adopted bear cubs. The cub was well treated, the children played with it, it could sleep inside the igloo and it was fed. Once the cub grew too big, it would be released (Laugrand and Oosten, 2015 p. 182–183). Among the Siberian Ket, families without children raised bear cubs, which were captured when their mother was killed, as their son or daughter, and released the bears when they were grown up (Alexejenko, 1963). The Ainu took care of bear cubs as invited guests. Once the cubs were full grown, they were sent away by a ceremonial killing. Also, canids were raised as invited guests, their skulls were sometimes perforated to free the spirit of the animal ( Table 2.1 ). In a more recent paper, Sigaut (1988) reflects on two other aspects of early domestication: familiarization and appropriation. According to Sigaut (1988 ),

Table 2.2 Comparison of possible forms of familiarization of wolves (cf. Sigaut, 1988) in the ethnographic and archaeological (Upper Palaeolithic) record (non- exhaustive list).

Possible forms of familiarization of wolves Wild/captive wolf (non-exhaustive) Ethnographic/ Archaeological modern record (UP) record

Self-domestication hypothesis feeding at kill-sites (no or limited + + habituation?) Tanner (1979 ); Bocherens et al. Sharp and Sharp (2015 ) (2015 ) feeding at campsites/villages + − (habituation) storage/waste Linnell et al. (2002 ) living at the outskirts of campsites (habituation) + (+)

(Continued) 52 Mietje Germonpré et al.

Possible forms of familiarization of wolves Wild/captive wolf (non-exhaustive) Ethnographic/ Archaeological modern record (UP) record preying on children Rajpurohit (1999 ); McNay and Mooney (2005 ) Human-control hypothesis capturing wolf pups + (+) Drucker (1951 ); Lescureux (2007 ) constructing food-storage caches to obstruct predators + + Fair (1997 ) Vereshchagin and Kuz’mina (1977 ) waste management to obstruct predators + (+) Tanner (1979 ); Lavrillier (2011 ) feeding wolf pups + (+) Murdoch (1892 ); Zélénine (1952 ); Hamayon (1990 ) familiarization by human contact + (+) Lord (2013 ); Hall et al. (2015 ); Jeannin (this volume) selection on behaviour + (+) vonHoldt et al. (2017 )

UP: Upper Palaeolithic +: clear evidence − : absence of evidence (+): likely the animal must develop another relation with humans than that of fear, flight or hostility – without, however, becoming too familiarized, a characteristic which could hinder its utilization. The association between animal and human is built up mutually by their respective behaviours towards each other. This familiarization expands when young pups are being taken care of and are being fed. In the self- domestication hypothesis, the adult wolves claim access to the food and waste available at the campsites (see references in Tables 2.1 and 2.2 ). Ample evidence exists in the ethnographic record that access to food and waste at campsites was obstructed to large predators, although they could scavenge on the remains left at the kill-sites (see references in Table 2.2 ). In addition, habituation to humans Self-domestication or human control? 53 Table 2.3 Comparison of possible forms of appropriation of captive wolves/dogs (cf. Sigaut, 1988) in the ethnographic and archaeological (Upper Palaeolithic) record (non-exhaustive list).

Possible forms of appropriation of captive wolves/ Wild wolf/captive wolf/dog dogs (non-exhaustive) Ethnographic/ Archaeological modern record (UP) record

Human-control hypothesis property (dog/wolf ): owned by + − an individual or a family Ellickson (2013); Wilson (1924) structured relationship: human-dog + Street et al. (2015) exchanging dog puppies + (+) Wilson (1924)

UP: Upper Palaeolithic +: clear evidence − : absence of evidence (+): likely is not the same as familiarization; habituated wolves can live in the vicinity of human settlements, associate humans with food, behave aggressively towards humans and undertake predatory attacks on children (see references in Table 2.2). This aggressive behaviour can result in hunting campaigns against the predatory wolves (e.g. Behdarvand et al., 2014). In fact, it has been proposed that during the Upper Palaeolithic, people intervened with large carnivores and caused a decline in their populations (Fan et al., 2016). In the human-control model, the captive carnivores are familiarized through human contact; they do not have free access to food but are being fed by humans (see references in Table 2.2). In the initial phases of the domestication process the breeding of captive wolves was likely controlled by selection on behaviour (vonHoldt et al., 2017) that probably installed a mutual cycle in which humans were tended to look after the cute pups of the captive canids (see discussion regarding oxytocin in Jeannin [this volume]). Another aspect of a captive wild animal is its appropriation which forms an interplay between social and ecological relations. According to Sigaut (1988 ), to be appropriated it is not necessary for the animal to be familiarized with humans, or even to be utilized; it is enough that the animal can be transferred from one human to another. Street et al. (2015) deem that the concept of ownership is not entirely fitting in the context of the egalitarian Upper Palaeolithic hunter- gatherers, and prefer to circumscribe that dogs had a structured relationship to people. According to Ellickson (2013), indirect evidence suggests that dogs, and their lupine ancestors on the way of being domesticated, had to be owned pri- vately by an individual or a family. He proposes that visual signalling to others provided clues about this private property. Wilson (1924) describes that in the 54 Mietje Germonpré et al. Hidatsa bands dogs belonged to the women of a family. The women exchanged pups – only those that were independent of their mother – for gifts, but never sold trained dogs ( Table 2.3 ).

Discussion We propose that the capturing and raising of wolf pups to obtain access to their products could have been a first stepping stone on the path to the domestication of the wolf (Table 2.1). One motivation for the keeping of captive wolves could be related to the animistic cosmology of some Upper Palaeolithic societies. In that sense, the captive canids could have been considered as invited guests that were upon adulthood ceremonially killed. According to Girard (1987), the use of animals for sacrifice must have preceded domestication, a process that he considers to be a by-product of the ritual practice. Also, Russell (2012) considers the need for sacrificial animals as a motivation for domestication. Another motive may lie in the guaranteed access to wolf skins to tailor cold-adapted clothing (cf. Collard et al., 2016). The captive canids would then possibly have been regarded as property (Table 2.3). Further drives, which would not leave traces in the archaeological record, though, could be the use of young wolves at the campsites as pets and sentinels. Also, the fact that wolf keeping could possibly enhance status could be considered. All these incentives likely were intertwined. We propose that the initial beginning of the wolf domestication process was closely linked with the cultural traditions of Upper Palaeolithic societies. Behavioural and social parallels between wild wolves and human hunters – both living in reproductive units (families) – likely helped the integration of selected wolves in the life of Upper Palaeolithic peoples (cf. Morey and Jeger, 2015). Other canids, like foxes, can be tamed and through behavioural selection become domesticated (Nelson et al., 2017). We think that in the Upper Palaeolithic, foxes – since they are not so dangerous – were permitted to visit campsites to look for food and were tamed as pets and providers of fur. They were not domesticated probably because they were rather ubiquitous present and could be, in contrast with wolves, more easily captured when needed. In wolves, the basic social unit is the breeding pair, with as the natural extension its offspring that can remain with the parents for up to four years. Wolves rely on pack members for foraging and pup-rearing, and are tolerant and cooperative towards pack members (Mech and Boitani, 2003). They can to a certain extent behave similarly to humans ( Virányi and Range, 2014 ). The wolf-wolf cooperation, characterized by emotional and cognitive skills, could therefore have formed the base of the dog-human cooperation. Recent studies have shown that hand-reared wolves, removed from their mother when about 10 days old, can become attached to their human caregivers (e.g. Lord, 2013; Hall et al., 2015 ) ( Table 2.2 ). However, human-socialized adult wolves can display offensive aggression towards humans as they sometimes fight for their dominance (Klinghammer and Goodmann, 1987). The predatory impulses of wolves stirred up by small children (e.g. Linnell et al., 2003) likely necessitated that captive wolves were tethered or confined at the Palaeolithic living sites. The offsprings of those Self-domestication or human control? 55 wolves that had a tendency to be attracted and sociable to humans would then, through directional selection, lose their fear of humans and be able to expand their social skills to interrelate with humans (Virànyi and Range, 2014; Marshall-Pescini et al., 2017 ; vonHoldt et al., 2017 ). A review of the emotional connections and communications between humans and dogs is given in Jeannin (this volume). Probably only a small number of the wolf pups that were brought to the Palaeo- lithic campsites could survive into adulthood and could breed. In order to have a sufficient number of founders, a large number of pups must have been brought to the camps where they were familiarized and fed ( Table 2.2 ). Upper Palaeoli- thic peoples probably let live at their camps those individual wolves with a less aggressive and gentler demeanour or wolves presenting some special feature like an unusual coat colour, provoking the early emergence of new coat colours (Olliv- ier et al., 2013). The former is in accordance with the hyper-sociability hypothesis that poses that selective breeding acted on the hyper-social response of individual wolves towards people so that not only juvenile animals but also adult canids could maintain social engagements with humans, permitting their coexistence ( vonHoldt et al., 2017 ). These sociable individuals must have survived until they delivered at least one litter. In that sense, the prehistoric hunter-gatherers must have screened a very large gene pool of Pleistocene wolves, and they must have permitted or accepted that some docile wolves became sexually mature. Such a screening was probably done not only by one group at one time, but could have been repeated by a large number of families/bands/societies over long periods of time over vast areas (Europe and/or Asia). Whenever different groups of hunter- gatherers would reunite, it is possible that the canids in their care were meeting too and that pups could have been exchanged (Table 2.3). The offspring of these “gentle” wolves would then have gone through the same type of selection process during which certain individuals were favoured and the unwanted ones culled, leading gradually and ultimately to primitive dogs. Genetic data from extant dogs confirm that at least 500 founding wolves were involved in the domestication process of the modern dog (Pang et al., 2009; Niskanen et al., 2013). Probably some of the early domesticated wolf lineages became extinct, as could have been the case with the Goyet Palaeolithic dog ( Figure 2.2 ), and/or were replaced by incoming dogs that accompanied migrant peoples during and after the Upper Pal- aeolithic (e.g. Frantz et al., 2016 ).

Conclusion The ancestor of the first domesticated animal in human history is a redoubtable carnivore: the wolf. The first steps of this domestication can be placed in the early Upper Palaeolithic. We think that a self-domestication of the wolf is unlikely because access to surplus food or waste at the Palaeolithic campsites would probably have been obstructed to large carnivores in order to avoid predatory attacks from habituated wolves with a food-conditioned behaviour. We do not believe that the beginning of the domestication of the wolf was a goal-oriented process. Rather, we believe that this early domestication was an unanticipated side effect of 56 Mietje Germonpré et al. the capturing and raising of canid pups by Upper Palaeolithic hunter-gatherers, a practice that likely formed part of the cultural tradition of certain Upper Palaeolithic societies. Several intertwined motives imply the complex interaction between man and carnivore during this process. The wolf/dog domestication is just one example of the diverse paths in the range of domestication processes. The early beginnings of wolf/dog domestication included different forms of association between humans and animals such as familiarization and mutualism and led to a hybrid community in which both species shared a habitat. The domesticated wolves started to transform the way of life of the Upper Palaeolithic people with whom they resided. Later on, the domestication process of the wolf probably did not serve as a template for the domestications of other animal species (cf. Uerpmann and Uerpmann, 2017 ).

Acknowledgments The authors would like to express their gratitude to Jean-Denis Vigne and Charles Stépanoff for inviting them to contribute to this volume on domestication and hybrid communities and for their helpful and constructive comments on an earlier draft of the text. This study is supported by a grant from the Czech Science Foun- dation GAČR 15-06446S “The relationships between humans and large canids – the dogs and wolves of the Gravettian Předmostí site (Moravia)”. The participation of ZIN RAS (state assignment № АААА-А17–117022810195–3) to this research is acknowledged.

References Alexejenko, J.A., 1963. Der Bärenkult der Keten (Jenissei-Ostjaken). In : V. Dioszegi, ed. Glaubenswelt und Folklore der sibirischen Völker. Budapest: Akademiai Kiado, 191–208. Akino, S., 1999. Spirit-sending ceremonies. In : W.W. Fitzhugh and C.O. Dubreuil, eds. Ainu, Spirit of a Northern People. Washington: Arctic Studies Center, National Museum of Natural History, Smithsonian Institution, 248–255. Anderson, D.G., 2017. Comments. In: C. Stépanoff, C. Marchina, C. Fossier and N. Bureau, eds. Animal autonomy and intermittent coexistences. North Asian modes of herding. Current Anthropology , 58, 70–71. Behdarvand, N. and Kaboli, M., 2015. Characteristics of gray wolf attacks on humans in an altered landscape in the west of Iran. Human Dimensions of Wildlife , 20, 112–122. Behdarvand, N., Kaboli, M., Ahmadi, M., Nourani, E., Mahini, A.S. and Aghbolaghi, M.A., 2014. Spatial risk model and mitigation implications for wolf – Human conflict in a highly modified agroecosystem in western Iran. Biological Conservation , 177, 156–164. Bélteky, J., Agnvall, B., Johnsson, M., Wright, D. and Jensen, P., 2016. Domestication and tameness: Brain gene expression in red junglefowl selected for less fear of humans suggests effects on reproduction and immunology. Royal Society Open Science , 3, 160033. Belyaev, D.K., 1979. Destabilizing selection as a factor in domestication. Journal of Heredity , 70, 301–308. Benecke, N., 1987. Studies on early dog remains from Northern Europe. Journal of Archaeological Science , 14, 31–49. Bocherens, H., 2015. Isotopic tracking of large carnivore palaeoecology in the mammoth steppe. Quaternary Science Reviews , 17, 42–71. Self-domestication or human control? 57 Bocherens, H., Drucker, D.G., Germonpré, M., Lázničková-Galetová, M., Naito, Y.I., Wissing, C., Brůžek, J. and Oliva, M., 2015. Reconstruction of the Gravettian food-web at Předmostí I using multi-isotopic tracking (13 C, 15 N, 34S) of bone collagen. Quaternary International , 359–360, 211–228. Bogoras, W., 1904–1909. Part I. The chukchee – Material culture. Memoirs of the Ameri- can Museum of Natural History 11. Publications of the Jesup North Pacific Expedition 7. Leiden: Brill. Boitoni, L., Francisci, F., Ciucci, P. and Andreoli, G., 2017. The ecology and behavior of feral dogs: A case study from central Italy. In : J. Serpell, ed. The Domestic Dog: Its Evolution, Behaviour and Interactions With People. Cambridge: Cambridge University Press, 342–368. Bökönyi, S., 1989. Definitions of animal domestication. In : J. Clutton-Brock, ed. The Walking Larder: Patterns of Domestication, Pastoralism, and Predation . London: Unwin Hyman, 22–27. Boudadi-Maligne, M. and Escarguel, G., 2014. A biometric re-evaluation of recent claims for Early Upper Palaeolithic wolf domestication in Eurasia. Journal of Archaeological Science , 45, 80–89. Brightman, R., 2002. Grateful Prey: Rock Cree Human-animal Relationships . Berkeley: University of California Press . Camarós, E., Münzel, S.C., Cueto, M., Rivals, F. and Conard, N.J., 2016. The evolution of Paleolithic hominin – Carnivore interaction written in teeth: Stories from the Swabian Jura (Germany). Journal of Archaeological Science: Reports , 6, 798–809. Charlier, B., 2013. Le loup mongol: pratiques et représentations. In : C. Stepanoff, C. Fer- ret, G. Lacaze and J. Thorez, eds. Nomadisme d’Asie centrale et septentrionale . Paris: Armand Colin, 221–223. Charlier, B., 2015. Faces of the Wolf: Managing the Human, Non-human Boundary in Mongolia . Leiden: Brill. Clutton-Brock, J., 1995. Origins of the dog: Domestication and early history. In : J. Serpell, ed. The Domestic Dog: Its Evolution, Behaviour and Interactions With People . Cam- bridge: Cambridge University Press, 7–20. Collard, M., Tarle, L., Sandgathe, D. and Allan, A., 2016. Faunal evidence for a difference in clothing use between Neanderthals and early modern humans in Europe. Journal of Anthropological Archaeology , 44 (Part B), 235–246. Coppinger, R. and Coppinger, L., 2001. Dogs: A Startling New Understanding of Canine Origin, Behavior, and Evolution . New York: Scribner. Coppinger, R. and Coppinger, L., 2016. What Is a Dog? Chicago: University of Chicago Press. Davis, S.J.M. and Valla, F.R., 1978. Evidence for domestication of the dog 12,000 years ago in the Natufian of Israel. Nature , 276, 608–610. Drake, A.G., Coquerelle, M. and Colombeau, G., 2015. 3D morphometric analysis of fossil canid skulls contradicts the suggested domestication of dogs during the late Paleolithic. Scientific Reports , 5, 1–8. Drucker, P., 1951. The Northern and Central Nootkan tribes. Bulletin Smithsonian Institute, Bureau American Ethnology , 144. Ellickson, R.C., 2013. Stone-age property in domestic animals. Yale Law & Economics Research Paper , 468. http://dx.doi.org/10.2139/ssrn.2241675 . Fair, S.W., 1997. Story, storage, and symbol: Functional cache architecture, cache narratives, and roadside attractions. Perspectives in Vernacular Architecture , 7, 167–182. Fan, Z., Silva, P., Gronau, I., Wang, S., Serres Armero, A., Schweizer, R.M., Ramirez, O., Pollinger, J., Galaverni, M., Ortega Del-Vecchyo, D., Du, L., Zhang, W., Zhang, 58 Mietje Germonpré et al. Z., Xing, J., Vila, C., Marques-Boner, T., Godinho, R., Yue, B. and Wayne, R.K., 2016. Worldwide atterns of genomic variation and admixture in gray wolves. Genome Research , 26, 163–173. Floss, H. and Hussain, S., 2015. Substantiating the saltationist view of Aurignacian emergence in Central and Western Europe: A reassessment of qualitative and quantitative arguments. In : Abstracts Annual 57th meeting of Hugo Obermaier Society , 7–11 April 2015 Heidenheim, 33. Frantz, L.A.F., Mullin, V.E., Pionnier-Capitan, M., Lebrasseur, O., Ollivier, M., Perri, A., Linderholm, A., Mattiangeli, V., Teasdale, M.D., Dimopoulos, E.A., Tresset, A., Duf- fraisse, M., McCormick, F., Bartosiewicz, L., Gál, E., Nyerges, E.A., Sablin, M.V., Bréhard, S., Mashkour, M., Bălășescu, A., Gillet, B., Hughes, S., Chassaing, O., Hitte, C., Vigne, J.-D., Dobney, K., Hänni, C., Bradley, D.G. and Larson, G., 2016. Genomic and archaeological evidence suggest a dual origin of domestic dogs. Science , 352, 1228–1231. Freedman, A.H., Gronau, I., Schweizer, R.M., Ortega-Del Vecchyo, D., Han, E., Silva, P.M., Galaverni, M., Fan, Z., Marx, P., Lorente-Galdos, B., Beale, H., Ramirez, O., Hor- mozdiari, F., Alkan, C., Vilà, C., Squire, K., Geffen, E., Kusak, J., Boyko, A.R., Parker, H.G., Lee, C., Tadigotla, V., Siepel, A., Bustamante, C.D., Harkins, T.T., Nelson, S.F., Ostrander, E.A., Marques-Bonet, T., Wayne, R.K. and Novembre, J., 2014. Genome sequencing highlights the dynamic early history of dogs. PLoS Genetics, 10, e1004016. Galeta, P., Lázničková-Galetová, M., Sablin, M. and Germonpré, M., submitted. Morpho- logical evidence for early dog domestication in the European Upper Palaeolithic: The randomisation approach. Germonpré, M., 1996. Preliminary results on the mammals of the Magdalenian upper horizon of Goyet (Belgium). Notae Praehistoricae , 16, 75–85. Germonpré, M., Fedorov, S., Danilov, M., Galeta, P., Jimenez, E.L., Sablin, M.V. and Losey, R.J., 2017a. Palaeolithic and prehistoric dogs and Pleistocene wolves from Yaku- tia: Identification of isolated skulls. Journal of Archaeological Science , 78, 1–19. Germonpré, M. and Hämäläinen, R., 2007. Fossil bear bones in the Belgian Upper Pal- aeolithic: The possibility of a proto-bear ceremonialism. Arctic Anthropology, 44, 1–30. Germonpré, M., Lázničková-Galetová, M., Jimenez, E.L., Losey, R.J., Sablin, M.V., Bocherens, H. and Van den Broeck, M., 2017b. Consumption of canid meat at the Gra- vettian Předmostí site, Czech Republic. Fossil Imprint , 73, 360–382. Germonpré, M., Lázničková-Galetová, M., Losey, R.J., Räikkönen, J. and Sablin, M.V., 2015a. Large canids at the Gravettian Předmostí site, the Czech Republic: The mandible. Quaternary International , 359–360, 261–279. Germonpré, M., Lázničková-Galetová, M. and Sablin, M., 2012. Palaeolithic dog skulls at the Gravettian Předmostí site, the Czech Republic. Journal of Archaeological Science , 39, 184–202. Germonpré, M. Lázničková-Galetová, M., Sablin, M.V. and Bocherens, H., in preparation. Could incipient dogs have enhanced differential access to resources among Upper Pal- aeolithic hunter-gatherers in Europe? Germonpré, M., Sablin, M.V., Lázničková-Galetová, M., Després, V., Stevens, R.E., Stiller, M. and Hofreiter, M., 2015b. Palaeolithic dogs and Pleistocene wolves revisited: A reply to Morey (2014). Journal of Archaeological Science , 54, 210–216. Germonpré, M., Sablin, M.V., Stevens, R.E., Hedges, R.E.M., Hofreiter, M., Stiller, M. and Després, V.R., 2009. Fossil dogs and wolves from Palaeolithic sites in Belgium, the Ukraine and Russia: Osteometry, ancient DNA and stable isotopes. Journal of Archaeo- logical Science , 36, 473–490. Self-domestication or human control? 59 Girard, R., 1987. Things Hidden Since the Foundation of the World . Stanford: Stanford University Press. Gompper, M.E., 2014a. The dog-human-wildlife interface: Assessing the scope of the problem. In : M.E. Gompper, ed. Free-ranging Dogs and Wildlife Conservation . Oxford: Oxford University Press, 9–54. Gompper, M.E., 2014b. Introduction: Outlining the ecological influences of a subsidized, domesticated predator. In : M.E. Gompper, ed. Free-ranging Dogs and Wildlife Conser- vation . Oxford: Oxford University Press, 1–8. Grimm, D., 2015. Dawn of the dog. Science , 348, 274–279. Hall, N.J., Lord, K., Arnold, A.-M.K., Wynne, C.D.L. and Udell, M.A.R., 2015. Assess- ment of attachment behaviour to human caregivers in wolf pups (Canis lupus lupus ). Behavourial Processes , 110, 15–21. Hallowell, A.I., 1926. Bear ceremonialism in the Northern Hemisphere. American Anthro- pologist , NS ., 28, 1–175. Hamayon, R.N., 1990. La chasse à l’âme. Esquisse d’une théorie du chamanisme sibérien . Nanterre: Société d’Ethnologie. Hamayon, R.N., 2012. The three duties of good fortune: ‘Luck’ as a relational process among hunting peoples of the siberian forest in Pre-Soviet Times. Social Analysis , 56, 99–116. Hussain, S.T. and Floss, H., 2015. Sharing the world with mammoths, cave lions and other beings: Linking animal-human interactions and the Aurignacian ‘belief world’. Quartär , 62, 85–120. Ingold, T., 1986. The Appropriation of Nature: Essays on Human Ecology and Social Rela- tions . Manchester: Manchester University Press. Issenman, B.K., 1997. Sinews of Survival: The Living Legacy of Inuit Clothing . Vancouver: University of British Columbia Press. Jimenez, E.-L., Smolderen, A., Jadin, I. and Germonpré, M., 2016. Exhumation de la collection faunique d’Edouard Dupont provenant du Trou Magrite (Pont-à-Less, Belgique). Quelles données et quelles perspectives pour une collection du XIX siècle? Notae Prae- historicae , 36, 167–190. Jordan, P., 2003. Material Culture and Sacred Landscape: The Anthropology of the Sibe- rian Khanty . Oxford: Altamira Press. Kimura, T., 1999. Bearing the ‘bare facts’ of ritual: A critique of Jonathan Z. Smith’s study of the bear ceremony based on a study of the Ainu iyomante. Numen , 46, 88–114. Kitagawa, J.M., 1961. Ainu bear festival (iyomante). History of Religions , 1, 95–151. Klein, R.G., 2009. The Human Career: Human Biological and Cultural Origins . Chicago: University of Chicago Press. Klinghammer, E. and Goodman, P., 1987. Socialization and management of wolves in captivity. In : H. Frank, ed. Man and Wolf: Advances, Issues, and Problems in Captive Wolf Research . Dordrecht: Dr W. Junk Publishers, 31–61. Larson, G. and Burger, J., 2013. A population genetics view of animal domestication. Trends in Genetics , 29, 197–205. Larson, G., Karlsson, E.K., Perri, A., Webster, M.T., Ho, S.Y., Peters, J., Stahl, P.W., Piper, P.J., Lingaas, F., Fredholm, M., Comstock, K.E., Modiano, J.F., Schelling, C., Agoulnik, A.I., Leegwater, P.A., Dobney, K., Vigne, J.D., Vilà, C., Andersson, L. and Lindblad-Toh, K., 2012. Rethinking dog domestication by integrating genetics, archeology, and biogeography. Proceedings of the National Academy of Sciences , 109, 8878–8883. 60 Mietje Germonpré et al. Laugrand, F. and Oosten, J., 2015. Hunters, Predators and Prey: Inuit Perceptions of Ani- mals . New York: Berghahn, 408 p. Lavrillier, A., 2011. The creation and persistence of cultural landscapes among the Siberian Evenkis: Two conceptions of ‘sacred’ space. In : P. Jordan, ed. Landscape and Culture in Northern Eurasia . Walnut Creek: Left Coast Press, 215–231. Lavrillier, A., 2013. La taiga sibérienne, forêt boréale. In : C. Stepanoff, C. Ferret, G. Lacaze and J. Thorez, eds. Nomadisme d’Asie centrale et septentrionale . Paris: Armand Colin, 28–30. Lázničková-Galetová, M., 2016. Perforated animal teeth. In : J. Svoboda, ed. Dolní Věstonice II: Chronostratigraphy, Paleoethnology, Paleoanthropology . Brno: Academy of Sciences of the Czech Republic, Institute of Archeology, The Dolní Věstonice Stud- ies, 21, 313–322. Lázničková-Galetová, M., Balzeau, A., Zikmund, T. and Kaiser, J., 2016. CT scanning of a crust with perforated animal teeth belonging to individual DV 13. In : J. Svoboda, ed. Dolní Věstonice II: Chronostratigraphy, Paleoethnology, Paleoanthropology . Brno, Academy of Sciences of the Czech Republic, Institute of Archeology, The Dolní Věstonice Studies 21, 372–376. Leonard, J.A., Vila, C., Fox-Dobbs, K., Koch, P.L., Wayne, R.K. and Van Valkenburgh, B., 2007. Megafaunal extinctions and the disappearance of a specialized wolf ecomorph. Current Biology 17, 1146–1150. Lescureux, N., 2007. Maintenir la réciprocité pour mieux coexister? Ethnographie du récit kirghiz des relations dynamiques entre les hommes et les loups. Thesis (PhD). Muséum national d’Histoire Naturelle, ED 227, Sciences de la Nature et de l’Homme. Lien, M., 2015. Becoming Salmon: Aquaculture and the Domestication of a Fish . Oakland: University of California Press. Linnell, J.D.C., Andersen, R., Andersone, Z., Balciauskas, L., Blanco, J.C., Boitani, L., Breitenmoser, U., Kojola, I., Liberg, O., Løe, J., Okarma, H., Pedersen, H.C., Prom- berger, C., Sand, H., Solberg, E.J., Valdmann, H. and Wabakken, P., 2002. The fear of wolves: A review of wolf attacks on humans. Oppdragsmelding, Trondheim: Norwegian Institute of Nature Research , 731, 1–65. Linnell, J.D.C., Solberg, E.J., Brainerd, S., Liberg, O., Sand, H., Wabakken, P. and Kojola, I., 2003. Is the fear of wolves justified? A Fennoscandian perspective. Acta Zoologica Lituanica , 13, 27–33. Lord, K., 2013. A comparison of the sensory development of wolves (Canis lupus lupus ) and dogs ( Canis lupus familiaris ). Ethology , 119, 110–120. Lupo, K.D., 2017. When and where do dogs improve hunting productivity? The empirical record and some implications for early Upper Paleolithic prey acquisition. Journal of Anthropological Archaeology , 47, 139–151. Maher, L.A., Stock, J.T., Finney, S., Heywood, J.J.N., Miracle, P.T. and Banning, E.B., 2011. A unique human-fox burial from a Pre-Natufian cemetery in the Levant (Jordan). PLoS ONE , 6, e15815. Marshall-Pescini, S., Cafazzo, S., Virányi, Z. and Range, F., 2017. Integrating social ecology in explanations of wolf – Dog behavioral differences. Current Opinion in Behav- ioral Sciences , 16, 80–86. McNay, M.E. and Mooney, P.W., 2005. Attempted predation of a child by a Gray Wolf, Canis lupus , near Icy Bay, Alaska. Canadian Field-Naturalist , 119, 197–201. Mech, L.D. and Boitani, L., 2003. Wolf social ecology. In : L.D. Mech and L. Boitani, eds. Wolves: Behavior, Ecology, and Conservation . Chicago: University of Chicago Press, 1–35. Morell, V., 2015. How wolf became dog. Scientific American , 313 (1), 48–55. Self-domestication or human control? 61 Morey, D.F., 2010. Dogs: Domestication and the Development of a Social Bond . Cam- bridge: Cambridge University Press. Morey, D.F. and Jeger, R., 2015. Paleolithic dogs: Why sustained domestication then? Journal of Archaeological Science: Reports , 3, 420–428. Murdoch, J., 1892. Ethnological results of the Point Barrow expedition. In: Ninth Annual Report of the Bureau of Ethnology to the Secretary of the Smithsonian Institution 1887–’88, John Wesley Power, General Editor . Washington: Government Printing Office, 19–441. Nelson, R.K., 1983. Make Prayers to the Raven: A Koyukon View of the Northern Forest . Chicago: University of Chicago Press. Nelson, R.M., Temnykh, S.V., Johnson, J.L., Kharlamova, A.V., Vladimirova, A.V., Gulev- ich, R.G., Shepeleva, D.V., Oskina, I.N., Acland, G.M., Rönnegård, L., Trut, L.N., Carl- borg, O. and Kukekova, A.V., 2017. Genetics of interactive behavior in silver foxes ( Vulpes vulpes ). Behavior Genetics , 47, 88–101. Nikolskij, P. and Pitulko, V., 2013. Evidence from the Yana Palaeolithic site, Arctic Siberia, yields clues to the riddle of mammoth hunting. Journal of Archaeological Science , 40, 4189–4197. Niskanen, A.K., Hagström, E., Lohi, H., Ruokonen, M., Esparza-Salas, R., Aspi, J. and Savolainen, P., 2013. MHC variability supports dog domestication from a large number of wolves: High diversity in Asia. Heredity , 110, 80–85. Nobis, G., 1979. Der älteste Haushund lebte vor 14000 Jahren. Umschau Wissenschaft und Technik , 79, 610. O’Connor, T.P., 1997. Working at relationships: Another look at animal domestication. Antiquity , 71, 149–156. Ollivier, M., Tresset, A., Bastian, F., Lagoutte, L., Axelsson, E., Arendt, M.-L., Bălăşescu, A., Marshour, M., Sablin, M.V., Salanova, L., Vigne, J.-D., Hitte, C. and Hänni, C., 2016. Amy2B copy number variation reveals starch diet adaptations in ancient European dogs. Royal Society Open Science , 3, 160449. Ollivier, M., Tresset, A., Hitte, C., Petit, C., Hughes, S., Gillet, B., Duffraisse, M., Pionnier- Capitan, M., Lagoutte, L., Arbogast, R.-M., Balasescu, A., Boroneant, A., Mashkour, M., Vigne, J.-D. and Hänni, C., 2013. Evidence of coat color variation sheds new light on ancient canids. PLoS ONE , 8 (10): e75110. Pacher, M., 2005. Die Verwendung von Bärenzähnen als Schmuck im Paläolithikum. In : D. Nagel, ed. Festschrift für Prof. Gernot Rabeder . Vienna: Mitteilungen der Kommis- sion für Quartärforschung, 14, 135–151. Pang, J.-F., Kluetsch, C., Zou, X.-J., Zhang, A.-B., Luo, L.-Y., Angleby, H., Ardalan, A., Ekström, C., Sköllermo, A., Lundeberg, J., Matsumura, S., Leitner, T., Zhang, Y.-P. and Savolainen, P., 2009. mtDNA data indicate a single origin for dogs south of Yangtze river, less than 16,300 years ago, from numerous wolves. Molecular Biology and Evolu- tion , 26, 2849–2864. Paulson, I., 1965. Die rituelle Erhebung des Bärenschädels bei arktischen und subark- tischen Völkern. Temenos , 1, 150–173. Patou-Mathis, M., 2012. Interactions between Neanderthals and carnivores in Eastern Europe. Journal of Taphonomy , 10, 277–290. Pendleton, A.L., Shen, F., Taravella, A.M., Emery, S., Veeramah, K.R., Boyko, A.R. and Kidd, J.M., 2017. Selective sweep analysis using village dogs highlights the pivotal role of the neural crest in dog domestication. bioRxiv . Available from: www.biorxiv.org/ content/early/2017/03/21/118794 . Peoples, H.C., Pavel Duda, P. and Marlowe, F.W., 2016. Hunter-Gatherers and the origins of religion. Human Nature , 27, 261–282. 62 Mietje Germonpré et al. Perri, A.R., 2016. A wolf in dog’s clothing: Initial dog domestication and Pleistocene wolf variation. Journal of Archaeological Science , 68, 1–4. Pionnier-Capitan, M., Bemilli, C., Bodu, P., Célérier, G., Ferrié, J.G., Fosse, P., Garcià, M., Vigne, J.D., 2011. New evidence for Upper Palaeolithic small domestic dogs in South- Western Europe. Journal of Archaeological Science , 38, 2123–2140. Pitulko, V.V., Pavlova, E. Yu., Nikolsky, P.A. and Ivanova, V.V., 2012. Material culture and symbolic behavior of the Upper Paleolithic settlers of Arctic Siberia (with particular reference to the Yana site). Russian Archaeological Yearbook , 2, 33–102. Porr, M. and de Maria, K., 2015. Perceiving animals, perceiving humans. Animism and the Aurignacian mobiliary art of Southwest Germany. In : S. Sázelová, M. Novák and A. Mizerová, eds. Forgotten Times and Spaces: New Perspectives in Paleoanthropological, Paleoetnological and Archeological Studies . 1st Edition. Brno: Institute of Archeology of the Czech Academy of Sciences, Masaryk University, 293–302. Prokof’yeva, E.D., 1964. The Sel’kups. In: M.G. Levin and L.P. Potapov, eds. The Peoples of Siberia . Chicago: University of Chicago Press, 587–606. Prokof’yeva, E.D., Chernetsov, N. and Prtkova, N.F., 1964. The Khants and Mansi. In : M.G. Levin and L.P. Potapov, eds. The Peoples of Siberia . Chicago: University of Chi- cago Press, 511–546. Rajpurohit, K.S., 1999. Child lifting: Wolves in Hazaribagh, India. Ambio , 28, 162–166. Robert-Lamblin, J., 2001. Un regard anthropologique. In : J. Clottes, ed. La grotte Chauvet. L’art des origines . Paris: Seuil, 200–208. Russell, N., 2002. The wild side of animal domestication. Society and Animals , 10, 285–302. Russell, N., 2012. Social Zooarchaeology: Humans and Animals in Prehistory . Cambridge: Cambridge University Press. Sablin, M.V. and Khlopachev, G.A., 2002. The earliest Ice Age dogs: Evidence from Eli- seevichi. Current Anthropology , 43, 795–799. Sànchez-Villagra, M.R., Geiger, M. and Schneider, R.A., 2016. The taming of the neural crest: A developmental perspective on the origins of morphological covariation in domesticated mammals. Royal Society Open Science , 3. Sautchak, C.E., 2016. Eating (with) piranhas: Untamed approaches to domestication. Vibrant , 13, 38–57. Sharp, H.S. and Sharp, K., 2015. Hunting Caribou: Subsistence Hunting Along the North- ern Edge of the Boreal Forest . Lincoln: University of Nebraska Press. Sigaut, F., 1980. Un tableau des produits animaux et deux hypothèses qui en découlent. Production Pastorale et Société , 7, 20–36. ———., 1988. Critique de la notion de domestication. L’Homme , 28 (108), 59–71. Skoglund, P., Ersmark, E., Palkopoulou, E. and Dalén, L., 2015. Ancient wolf genome reveals an early divergence of domestic dog ancestors and admixture into high-latitude breeds. Current Biology , 25, 1–5. Sokolova, Z.P., 2000. The bear cult. Archaeology, Ethnology, Anthropology of Eurasia , 2, 121–130. Sokolowa, S., 1982. Das Land Jugorien . Leipzig: VEB F.A. Brockhaus Verlag. Stefansson, V. and Wissler, C., 1919. Stefansson-Anderson arctic expedition. Anthropo- logical Papers American Museum Natural History , 14. Stépanoff, C., 2013. L’ours en Sibérie. In : C. Stepanoff, G. Ferret, G. Lacaze and J. Thorez, eds. Nomadisme d’Asie centrale et septentrionale . Paris: Armand Colin, 220–221. Stewart, J.R., 2004. Neanderthal-modern human competition? A Comparison between the mammals associated with Middle and Upper Palaeolithic industries in Europe during OIS 3. International Journal of Osteoarchaeology , 14, 178–189. Self-domestication or human control? 63 Street, M., Napierala, H. and Janssens, L., 2015. The late palaeolithic dog from Bonn- Oberkassel in context. In: L. Giemsch, R.W. Schmitz, eds. The Late Glacial Burial from Oberkassel Revisited . Bonn: Verlag Philipp von Zabern in Wissenschaftliche Buchge- sellschaft, Reinische Ausgrabungen 72, 253–273. Tanner, A., 1979. Bringing Home Animals: Religious Ideology and Mode of Production of the Mistassini Cree Hunters . London: C. Hurst and Company. Tarlach, G., 2016. The origins of dogs. Discover Magazine , December, 32–39. Thalmann, O., Shapiro, B., Cui, P., Schuenemann, V.J., Sawyer, D.K., Greenfield, D.L., Germonpré, M.B., Sablin, M.V., López-Giráldez, F., Domingo-Roura, X., Napierala, H., Uerpmann, H.-P., Loponte, D.M., Acosta, A.A., Giemsch, L., Schmitz, R.W., Worthing- ton, B., Buikstra, J.E., Druzhkova, A., Graphodatsky, A.S., Ovodov, N.D., Wahlberg, N., Freedman, A.H., Schweizer, R.M., Koepfli, K.-P., Leonard, J.A., Meyer, M., Krause, J., Pääbo, S., Green, R.E. and Wayne, R.K., 2013. Complete mitochondrial genomes of ancient canids suggest a European origin of domestic dogs. Science , 342, 871–874. Trut, L., Oskina, I., Kharlamova, A., 2009. Animal evolution during domestication: The domesticated fox as a model. Bioessays , 31, 349–360. Uerpmann, H.-P. and Uerpmann, M., 2017. Chapter 7: The ‘commodification’ of animals. In : A. Tsuneki, S. Yamada and K.-I. Hisada, eds. Ancient West Asian Civilization . Media Singapore: Springer Science+Business, 99–113. Vanak, A.T. and Gompper, M.E., 2009. Dogs Canis familiaris as carnivores: Their role and function in intraguild competition. Mammal Review , 39, 265–283. Vanhaeren, M. and d’Errico, F., 2006. Aurignacian ethno-linguistic geography of Europe revealed by personal ornaments. Journal of Archaeological Science , 33, 1105–1128. Vereshchagin, N.K. and Kuz’mina, I.E., 1977. Remains of mammals from Palaeolithic sites on the Don and Desna Rivers. Proceedings of the Zoological Institute, Academy of Sciences of the USSR , 72, 77–110. Vigne, J.-D., 2015. Early domestication and farming: What should we know or do for a better understanding? Anthropozoologica , 50, 123–150. Vilà, C., Savolainen, P., Maldonado, J.E., Amorim, I.R., Rice, J.E., Honeycutt, R.L., Cran- dall, K.A., Lundeberg, J. and Wayne, R.K., 1997. Multiple and ancient origins of the domestic dog. Science , 279, 1687–1689. Virányi, S. and Range, F., 2014. Chapter 2 – On the way to a better understanding of dog domestication: Aggression and cooperativeness in dogs and wolves. In : J. Kaminski and S. Marshall-Pescini, eds. The Social Dog, Behavior and Cognition . Amsterdam: Else- vier, Academic Press, 35–62 . vonHoldt, B.M., Shuldiner, E., Koch, I.J., Kartzinel, R.Y., Hogan, A., Brubaker, L., Wanser, S., Stahler, D., Wynne, C.D.L., Ostrander, A.E., Sinsheimer, J.S. and Udell, M.A.R., 2017. Structural variants in genes associated with human Williams-Beuren syndrome underlie stereotypical hypersociability in domestic dogs. Science Advances , 3, e1700398. Available from: http://advances.sciencemag.org/content/3/7/e1700398 . Walker, B.L., 2005. The Lost Wolves of Japan . Seattle: University of Washington Press. Wilczyński, J., Wojtal, P., Roblíčková, M. and Oliva, M., 2015. Dolní Vĕstonice I (Pavlov- ian, the Czech Republic) – Results of zooarchaeological studies of the animal remains discovered on the campsite (excavation 1924–52). Quaternary International, 379, 58–70. Wilkins, A.S., Wrangham, R.W. and Tecumseh Fitch, W., 2014. The ‘domestication syndrome’ in mammals: A unified explanation based on neural crest cell behavior and genetics. Genetics , 197, 795–808. Willerslev, R., 2007. Soul Hunters: Hunting, Animism, and Personhood Among the Sibe- rian Yukaghirs . Berkeley: University of California Press. 64 Mietje Germonpré et al. Willerslev, R., Vitebsky, P. and Alekseyev, A., 2015. Sacrifice as the ideal hunt: A cosmological explanation for the origin of reindeer domestication. Journal of the Royal Anthropological Institute (N.S.) , 21, 1–23. Wilson, G.L., 1924. The horse and the dog in Hidatsa culture. Anthropological Papers of the American Museum of Natural History 15, 125–311. White, R., 2003. Prehistoric Art: The Symbolic Journey of Humankind. New York: H.N. Abrams Inc Publishers. White, R., 2007. Systems of personal ornamentation in the Early Upper Palaeolithic: Methodological challenges and new observations. In : P. Mellars, K. Boyle, O. Bar- Yosef and C. Stringer, eds. Rethinking the Human Revolution: New Behavioural and Biological Perspectives on the Origin and Dispersal of Modern Humans . Cambridge: McDonald Institute Monographs, McDonald Institute for Archaeological Research, 287–302. Wojtal, P., Wilczyński, J., Zbigniew, M., Bocheński, Z.M. and Svoboda, J.A., 2012. The scene of spectacular feasts: Animals remains from Pavlov I south-east, the Czech Repub- lic. Quaternary International , 252, 122–141. Wright, D., 2015. The genetic architecture of domestication in animals. Bioinformatics and Biology Insights , 9(S4), 11–20. Yamada, T., 2001. The World View of the Ainu: Nature and Cosmos Reading From Lan- guage . Oxford: Routledge. Zeder, M.A., 2012. Pathways to animal domestication. In : P. Gepts, T.R. Famula, R.L. Bettinger, S.B. Brush, A.B. Damania, P.E. McGuire and C.O. Qualset, eds. Biodiversity in Agriculture: Domestication, Evolution, and Sustainability. Cambridge: Cambridge University Press, 227–259. Zélénine, D., 1952. Le culte des idoles en Sibérie . Paris: Payot. 3 Beyond wild and domestic Human complex relationships with dogs, wolves, and wolf-dog hybrids

Nicolas Lescureux

Introduction Milton (2000) suggested that conservationists invoke three culturally defined boundaries: 1) the interspecies boundary, 2) the boundary between natives and aliens, and 3) the boundary between human and non-human processes. The last two categories clearly reflect the dichotomy between nature and culture on which modern western science is largely based ( Latour 1993 ; Descola 2013 ). The same is true of the boundary between wild and domestic, which is a consequence of the distinction between human and non-human processes, and appears to not be universal (Descola 2004 ). Interestingly, the complex relationship between humans, dogs, and wolves – and notably the hybridization between the latter – questions both the interspecies boundary and the boundary between wild and domestic. Indeed, dogs descend from wolves and although they have gone through different historical processes, wolves and dogs still interact across a large part of the Northern Hemisphere. Question- ing these boundaries also implies reflecting on conservation discourses and actions about wolf-dog hybrids and hybrids in general, both conceptually and practically. After clarifying current thinking on the ancestry relationships between wolves and dogs, we will quickly draw the almost antithetical historical trajectories of human-dog and human-wolf relationships from the time dogs first became domesticated. Then we will discuss the question of wolf-dog hybridization, its definition, its level, and how it can be perceived as a double pollution from the nature conservation perspective. This will lead us to question the relevance of interspecies and wild-domestic boundaries used in science in general and in the conservation sciences in particular in the case of wolf-dog hybrids. Beyond its biological aspects, hybridization thus appears as a social problem and managing hybrids certainly requires taking account of the existence of culturally constructed conceptual boundaries that humans have established between wolves and dogs that parallel those between wild and domestic, or nature and culture.

Wolves and dogs: ancestors and descendants According to archaeological data, hominids and grey wolves have had a relatively close relationship for at least 300,000 years (Olsen 1985). This close relationship probably led to the progressive emergence of commensal wolves partly living on 66 Nicolas Lescureux human refuse, thus entering into a domestication process eventually leading to the emergence of domestic dogs ( Morey 1994 ; Clutton-Brock 1995 ; Larson and Fuller 2014 ). However, the debate remains open about when, where, and how the process of domestication happened. Some fossils dated from ca. 30,000 years BP or even earlier have been morphologically identified as dog remains ( Germonpré et al . 2009 ; Ovodov et al . 2011 ; Germonpré et al . 2012 ; Germonpré et al . 2013 ; see also Germonpré et al. in this volume), but this proposal has been seriously questioned (Pionnier-Capitan et al. 2011; Crockford and Kuzmin 2012; Boudadi- Maligne and Escarguel 2014 ; Perri 2016 ). The combination of archaeological and genetic data has begun to provide a better understanding of the dog domestication process. According to Larson et al . (2012 ), the first undisputed domestic dog remains date back to ca. 15,000 years BP in Europe and ca. 12,000 years BP in several places including Syria, Cyprus, Iraq, northern China, and the Russian far east. Palaeogenetic analysis has revealed the existence of at least two centres of dog domestication ( Frantz et al . 2016 ). It also appears that interbreeding (backcrossing) between dogs and local wolf populations often occurred during the early stages of the domestication process (Vilà et al. 1997; Ardalan et al. 2011; vonHoldt et al . 2011 ; Wayne and vonHoldt 2012 ). This would reinforce the suggestion that dog domestication in its early stages was probably more a stochastic evolutionary process than one guided by human design, implying no intention of domestication but more a specialization of some wolves to a new niche offered by humans (Morey 1994; Coppinger and Coppinger 2001; Galibert et al. 2011; Larson et al. 2012). The hypothesis of self-domestication is contested, notably by Germonpré et al . (in this volume), who suggest the widespread existence of wolf pup rearing in Upper Palaeolithic human groups. None- theless, wolf pup rearing did not imply any human intention to domesticate. This is all the more probable as humans had no prior experience with domestication. According to the most recent studies combining archaeology and genetic data, earliest livestock domestication events for other species (cattle, sheep, goats, pigs) all occurred in the Near East around the same period between 8,500 and 11,000 years BP ( Vigne 2011 ; Zeder 2011 ). At the time and place these domestication events occurred, dogs were already domesticated in Eurasia and certainly present in the Near East (Dayan 1994; vonHoldt et al . 2010 ; Larson et al. 2012). It could well be that the definitive separation between dogs and wolves is relatively recent and would have followed Neolithic animal domestication, which implies a physical separation between dogs and wolves as a consequence of the incompatibility of wolf presence around human settlements where livestock were kept (Clutton-Brock 1995; Sablin and Khlopachev 2002 ; Verginelli et al . 2005 ), even though events of hybridization have been detected later during the neolithization process ( Ollivier et al . 2013 ).

Dogs’ and wolves’ relationships with humans: antithetical trajectories? Whatever the place(s) of origin, the nature of the process or the exact time of dog domestication, accompanying humans somehow appears as a good strategy for Beyond wild and domestic 67 dogs from an evolutionary point of view. In a few thousand years, they spread across all the continents and in almost all islands in the world in the wake of human migrations. Dogs became capable of digesting starch as early as 7,000 years BP in Southeastern Europe, suggesting a biocultural coevolution of humans and dogs, allowing the latter to live on a rich starch diet paralleling the development of farming societies ( Ollivier et al . 2016 ). Despite low genetic diversity mainly due to recent efforts to create purebred dogs (Frantz & Larson, this volume), dogs have reached a very high phenotypic diversity with around 400 different breeds ( Galibert et al. 2011) adapted to very different purposes, from providing meat to herding flocks to various other uses such as hunting, following scents, guiding blind people, carrying burdens, protecting livestock against wolves, or just to be pets and run after balls and sticks. As a consequence the largest dog is more than 100 times heavier than the smallest one ( Galibert et al . 2011 ). In addition, beyond these 400 breeds, there are a lot of crossbreeds as well as village dogs unrelated to any breeds ( Boyko et al . 2009 ). Dogs have become one of the most ubiquitous domestic species and the most common carnivore. Their worldwide population is estimated to be close to 900 million (60% of them in rural areas) and certainly growing ( Gompper 2014 ), while their ancestor population is estimated at around 400,000 individuals ( Mech and Boitani 2003 ). Indeed, from the time dogs and later ungulates were domesticated, the history of humans and wolves has been rather complex and often conflictual. Humans and grey wolves ( Canis lupus) have been sharing the same landscape, the same habitats and even some similarities in their hunter’s way of life for a long time ( Olsen 1985 ; Clutton-Brock 1995 ). However, it is quite probable that the domestication of ungulate species (cattle, sheep, and goats) which started around 11,500 BP and spread across Eurasia through the neolithization process ( Tresset and Vigne 2011 ) led to keeping wolves at a distance from human settlements ( Clutton-Brock 1995 ; Sablin and Khlopachev 2002 ; Verginelli et al . 2005 ). Indeed, from this period, wolves became a potential threat for livestock, and they still appear as one of the most conflictual species wherever their presence overlaps with herding activities. Conflicts between wolves and livestock breeding have probably been responsible for motivating the past reduction in the number and distribution of large carnivores on a worldwide level ( Mech 2017 ). In the western world, animals used to be categorized either as useful or as harmful for human activities. As they killed livestock and even occasionally humans ( Linnell et al . 2003 ; Moriceau 2007 ), there is no doubt that wolves were on the harmful side. However, despite the worldwide expansion of agriculture and livestock breeding, the wolf’s adaptability combined with low human density allowed them to survive in most of their historical range until a relatively recent past. Except for the British Isles (Hickey 2011), only from the middle of the 19th century did the increase in human population, deforestation, and progress in hunting equipment and organization start to dramatically impact wolf populations, leading to the worldwide reduction of the wolf’s historical range by a third in the middle of the 20th century ( Mech 1995 ; Boitani 2003 ). 68 Nicolas Lescureux However, the sociocultural context in North America and Western Europe also dramatically changed during this period. Faced with industrialization and the concentration of human populations in large cities, a romantic vision of nature and an attraction towards the wilderness developed in North America and Western Europe at the end of the 19th century (Descola 2013). The development of ecology also changed the perception of animals, displaying their ecological role beyond their usefulness or harmfulness for human activities. The combination of attraction towards nature and the wilderness and concerns about the ecological role of animal species drove some people to be concerned about endangered species. In the USA, this concern was translated into the Endangered Species Act (1966) and even extended to wolves, which were protected in 1973. In the same year was created the wolf specialist group in the International Union for Conservation of Nature (IUCN). Information about wolves in the wild was not extensive at that time and essentially based on studies made in Alaska and Minnesota in the 1940s and the 1970s ( Mech 1970 ; Mech and Frenzel 1971 ; Murie 1985 [1944 ]). Studies on wolf ecology and behaviour then became more numerous. Their circulation in popularized versions contributed to positively change the wolf’s image, but also propagated incomplete information or even misconceptions towards mainly urban populations without experience-based knowledge about wolves ( Mech 1995, 2017 ). Thus, wolves started to be perceived as a symbol of the wilderness, a keystone species able to self-regulate, and which only attack livestock in the absence of wild prey. These campaigns also led to the idea that wolves could not stand intense human activity and avoided settling in such anthropized areas. This idealized vision of the wolf contributed to the multiplication of environmentalist associations defending and protecting wolves, in the USA and then in Europe, and permitted a kind of wolf rehabilitation campaign, as well as its protection in several European countries. Nowadays, in many countries, land abandonment, drastic changes in rural land occupation, and/or conservation legislation are leading to the recovery of wolf populations in multiple-use landscapes (Linnell et al . 2001 ; Falcucci et al . 2007 ; Chapron et al . 2014 ). Accordingly, many conflicts are currently appearing or increasing in several countries (Skogen et al . 2008 ; Dressel et al . 2015 ; Garde 2015 ; Mech 2017 ).

The construction of the wolf-dog hybrids problem As a consequence of wolves and dogs being closely related, they can interbreed and produce fertile offspring. For centuries humans have been deliberately crossbreeding wolves and dogs in order to obtain wolf-dog hybrids. The first written record of this practice comes from Aristotle (ca. 350 BC ) and Pliny (AD 77), who reported that people from Gaul tied their bitches to trees so they could mate with wolves and produce hybrids (Iljin 1941). Deliberate wolf-dog crossbreeding in order to improve dog breeds was apparently widespread in the 17th and 18th centuries, even if only occasionally practiced, and has been reported for Indian dogs, Inuit dogs, Hungar- ian dogs, etc. ( Iljin 1941 ). Nowadays, several wolf-dog breeds exist.1 Beyond wild and domestic 69 Evidence of uncontrolled wolf-dog hybridizations While humans have been, and still are, crossbreeding wolves and dogs, hybridization can also occur in uncontrolled situations. Thus, hybridization was certainly a reoccurring part of the early dog domestication process and was possible because humans and wolves were living in close contact and early dogs and wolves were certainly morphologically similar. Nowadays, uncontrolled hybridization between dogs and wolves still occurs. Hybridization has to go through two steps in order to have an impact on wolf populations. Firstly, crossbreeding between wolves and dogs must generate hybrids (generation F1). These hybrids can reproduce among themselves, but can also backcross with wolves. Ultimately, an introgression of dog genes can occur into wolf populations ( Randi 2011 ). Analyses based on mtDNA have not detected much introgression of dog mtDNA into wolf populations ( Muñoz-Fuentes et al . 2010 ). Therefore, wolf-dog hybridization has long been considered a very rare event, since it was considered that crossing between male dogs and female wolves was unlikely to be successful because male dogs do not assist females in pup rearing and care (Vilà and Wayne 1999 ). However, more recent genetic studies that include Y-chromosome analysis tend to show that crosses between male dogs and female wolves not only occur but are primarily responsible for the hybridization process (Vilà et al . 2003 ; Iacolina et al . 2010 ; Godinho et al . 2011 ). Crosses between female dogs and male wolves can occur, but remain rare (Hindrikson et al. 2012). Evidence for modern day hybridization in the wild has been detected in numerous places including Bulgaria, Canada, Italy, Latvia, Spain, Portugal, and Scandinavia (Caniglia et al . 2013 ; Lescureux and Linnell 2014 ; Torres et al . 2017 ). It is often assumed that the risks of hybridization are higher in areas where wolves are either rare, highly per- turbed, or in contact with a large population of free-ranging dogs ( Vilà and Wayne 1999 ; Randi et al . 2000 ; Hailer and Leonard 2008 ). However, it appears that hybridization between wild canids and domestic dogs can also occur even when the wild canid population is relatively abundant ( Adams et al . 2003 ). Nonetheless, wolf-dog hybridization remains relatively rare compared to wild/domestic species couples like wild boars (Sus scrofa) and pigs (Sus scrofa domesticus) (Frantz & Larson, this volume).

Wolf-dog hybridization: double pollution In some parts of conservation scientific literature, while natural hybridization (between wild living forms) is seen as having a role in speciation, anthropogenic hybridization (between wild and domestic forms) is seen as a pollution potentially compromising the genetic integrity of existing taxa (cf. Lorenzini et al . 2014). This concern, which is expressed in several papers about the introgression of dog genes into wolf populations, directly raises the questions of boundaries , purity, and pollution which are well-developed concepts in disciplines such as the anthropology of nature and the sociology of sciences (see e.g. Douglas 1966; Latour 1993 ; Knight 2000 ; Forsyth 2003 ). Purity is associated with the respect 70 Nicolas Lescureux for boundaries, whereas pollution occurs when boundaries are crossed, i.e. when the social understanding of the contextualized environmental order is disturbed ( Knight 2000 ). In addition, wolf – dog hybridization and dog genetic introgression into wolf populations are perceived as a double pollution. Indeed, on the one side they cross the boundary between two species (Canis lupus and Canis familiaris ) and on the other side they cross the boundary between wild and domestic, which reflects the boundary between natural processes and human processes as proposed by Milton (2000). The boundary between wild and domestic is well established in Western societies and in Western science (Descola 2004, 2013 ). In the view of some conservationists, dogs – as domestic animals – are already “polluted” by human processes. They no longer belong to nature. This could probably explain the previous lack of interest among biologists in studying dog ecology (cf. Vanak and Gompper 2009 ). Therefore, when dogs become feral, they fall between two categories; they are no longer domestic without, however, belonging to wildlife. They are not in the “right” place and they “pollute” nature (Knight 2000). As a consequence, they are perceived as a conservation problem and an anthropogenic threat to nature. When these feral dogs interbreed with wolf populations, they cross the interspecies boundary as well as the domestic/wild one, generating disorder in conservationists’ social understanding of the environment. What is the result of this hybridization? Is it wild, or is it domestic? Is it Canis lupus, or Canis familiaris ? These hybrids fall between categories and, as such, generate pollution. But one can question the relevance of these categories. Indeed, the boundaries generating this sense of pollution are not rigidly established once and for all.

Domestic, wild, Canis lupus, Canis familiaris: relevant categories? The socially constructed concepts of nature as separated from humans, and wild separated from domestic, are increasingly being questioned in the anthropology of nature (Descola and Pálsson 1996; Ellen and Katsuyoshi 1996; Ingold 2000; Descola 2004 , 2013 ) and the sociology of science ( Latour 1993 ), notably because they do not appear to be universal and also because they generate numerous conceptual “hybrids”, which are neither pure natural objects nor pure social subjects, as in the case of wolf-dog hybrids, which are neither wild animals nor owned domestic animals. In addition, these conceptual boundaries are quite frequently transgressed in conservation actions (Linnell et al. 2015) since reintroductions, translocations, and other forms of wildlife management are above all human processes, and domestic animals are even sometimes used to maintain “natural” landscapes. Finally, processes like climate change also question the mere existence of natural processes outside of any human influence. The interspecies border between domestic dogs and wolves can also be questioned, and it has been particularly obvious in the uncertain taxonomic status of the dingo in Australia, which has variously been viewed as either a wolf subspecies, Canis lupus dingo, or a feral dog, Canis familiaris dingo ( Newsome et al . 1980 ; Beyond wild and domestic 71 Newsome and Corbett 1982, 1985 ; Corbett 1995), even if it now clearly appears as Canis familiaris ( Larson et al. 2012). The situation is made even more complex because of documented hybridization between domestic dogs of European origin and dingoes (Daniels and Corbett 2003). Similar taxonomic uncertainty also exists for many other wild canids, such as the status of the eastern wolf in North America (Canis lupus vs Canis lycaon ), the role of hybridization between coyotes and wolves in the origins of the red wolf (Canis rufus ) ( Nowak 1992 ; Nowak and Federoff 1998 ; Wayne et al . 1998 ; Wilson et al . 2000 ; Grewal et al . 2004 ; Mech 2010 ; Benson et al. 2013), the adaptive introgression of wolf and dog genes into coyote populations (Monzón et al. 2014), and the identity of the Great Lakes wolves (Leonard and Wayne 2008, 2009; Cronin and Mech 2009; Mech 2009). It has also been claimed that North African jackals could potentially be considered as African wolves Canis lupus lupaster ( Rueness et al . 2011 ; Gaubert et al . 2012 ). It now appears that hybridization exists between golden jackals (Canis aureus ) and dogs (Galov et al. 2015). This pattern of cryptic relationships is beginning to emerge as a recurrent theme among the larger canid species and may reflect more fluid species borders than many biologists, and almost all legislation, are used to dealing with.

Hybridization as a social problem Conservation biologists and wildlife managers are concerned with hybridization as a potential threat to small wolf populations in close contact with free-ranging and feral dogs ( Randi 2008 ; Iacolina et al . 2010 ). Indeed, hybridization could drive species or populations to lose specific adaptations and even cause their extinction as a distinct taxon (Gottelli et al . 1994 ; Simberloff 1996 ; Randi 2008 ; Muñoz-Fuentes et al . 2010 ; Allendorf et al . 2013 ), notably if hybrids reveal themselves as competitors to wolf populations ( Bassi et al . 2017 ). On the other hand, it has been shown that introgression can also be adaptive (Castric et al . 2008 ; Hed- rick 2013) and it is possible that hybridization with dogs could sometimes provide advantages for their descendants. For instance, under certain circumstances, black wolves 2 may have a better life expectancy, especially in the face of environmental changes (Anderson et al . 2009 ; Hedrick 2009 ; Coulson et al. 2011), even if its rar- ity in wolf populations and its early appearance in dogs would suggest this mutation was strongly counter-selected in strictly wild contexts ( Ollivier et al . 2013 ). Behind conservationists’ concerns about hybridization and introgression seen as pollution, there is no apparent attested impact of wolf-dog hybridization and dog genes introgression on wolf behavioural patterns or wolf conservation status. It appears that the main concerns result from culturally defined boundaries which are not always biologically relevant (the one between canid species) and not universal (the one between wild and domestic). Conservation biologists are probably facing contradictions between their concerns – as conservationists – about human negative impacts on the planet (anthropophobia), and their concerns – as scientists – about the influence of their own human normative values on the way they do science (autophobia). Indeed, the influence of normative values appears 72 Nicolas Lescureux rather obvious on hybridization problems, and the contradictions resulting from the combination of anthropophobia and autophobia will probably generate what Robbins and Moore call “ecological anxiety disorder” ( 2013 ). Our anthropological view on wolf-dog relationships does not de-legitimize conservationists’ concerns for hybridization as a potential threat for wolf populations. However, it shows that in addition to legal and technical aspects, their concern is based on values linked with culturally defined (socially constructed) boundaries. Therefore, the management of feral dogs and/or of wolf-dog hybridization goes beyond being a biological problem and requires consideration of the perceptions of society.

Hybrids as unmanageable animals? Whatever the consequences, hybridization between dogs and wolves emerges as a major challenge for wildlife managers and conservation biologists for a number of reasons. First, identification of wolf-dog hybrids remains complex even with the latest advances in genetic techniques (cf. notably Lorenzini et al . 2014 ). Secondly, the legal status of these hybrids is very difficult to assess. If the wolf is protected, what is the status of a wolf-dog hybrid? The only international legislation that specifically addresses the issue is the Convention on International Trade in Endangered Species which offers hybrids the same protection as the wild species (CITES, Conf. 10.17, Rev. Cop14). Faced with the revelation of the relatively high percentage of hybridization in wolf populations in Europe, a similar recommendation to protect hybrids was made by the standing committee of the Convention on the Conservation of European Wildlife and Natural Habitats of the Council of Europe in 2014 ( Trouwborst 2014 ). These issues raise a number of difficult questions that conservationists need to address about the management of hybrids. The question has been raised if all black wolves or wolves with dewclaws (cf. Ciucci et al. 2003) should be removed from the wild in Italy because there was an introgression of dog genes in their karyotype at some point in their history. Intro- gression of dog genes into wolf populations can be the result of ancient hybridizations. Robbins and Moore (2013) suggested that conservation sciences are edenic sciences in the sense they always refer to a previous point in history as a reference point, an a priori baseline when things were better or more natural. In that case, would an ancient hybridization be more acceptable than a more recent one? Were dogs more natural 300 or 3,000 years ago than they are now? Were wolves “pure” at that time? This also leads to further technical questions about what is an acceptable level of introgression. What percentage of dog genes introgression would make the wolf a dog? How can this percentage be quantified objectively? What would be an acceptable level of “pollution”? Finally, a wide range of ethical questions about the acceptability of various management interventions are apparent. Blurry boundaries and the impact of human society on wolf-dog relationships are notably reflected in the varying legal status of wolf-dog hybrids and the legal status of free-ranging dogs. The latter is in general much more diverse than the Beyond wild and domestic 73 legislation governing wolves and shows considerable variation between countries, from the shoot-on-sight policy for loose dogs in the Baltics to the full protection provided for them in Italy, from the use of lethal control of hybrids (e.g. Norway in 2004) to live capture and placement in a captive setting (e.g. Latvia in 2000, Germany in 2003, and Italy in 2013). Virtually all discussions about hybrid management take place in highly emotional debates, and there are no widely accepted best practice management protocols for their management. The desirability of wolf-dog hybrids being ecological surrogates for wolves has never been raised in the mainstream conservation literature, although there has been some local public debate about it. One of the main problems is that there is little data on the behaviour and ecology of wolf-dog hybrids under free-ranging conditions.

Conclusion Human relationships with wolves and dogs definitely disrupt established categories and boundaries. In a part of conservation literature, intro -gression is perceived as a trans -gression, the pollution of the wild by the domestic world, associated with degeneration. Such a view of domestication as degeneration is also present in the theory of neoteny. Even if now contested ( Drake and Klingenberg 2010 ), this theory postulates that domestic animal morphology and behaviour can be associated with different developmental stages of their wild counterparts, only the wild animal reaching the last stage ( Coppinger and Coppinger 2001 ). The huge variability in human relationships with both wolves and dogs brings into question the relevance of interspecific and wild-domestic boundaries, which are rather status and static boundaries. A more relational approach could appear more relevant, considering that a reasonable proportion of dogs are free-ranging and do not have close relationships with humans, while some wolves are closely tracked with GPS, trapped, captured, identified, and released, and others are selected according to their behaviour when dangerous or harmful individuals are shot. Some wolves also feed on refuse in cities, while free-ranging dogs hunt roe deer or wild boars in forests. We could almost say that there are now wild dogs and domestic wolves. Both archaeology and genetics define differences between wild and domestic animals based on morphology, morphometry, phenotype, and genotype. However, the domestication process could rather be seen as an intensification of human- animal relationships ( Vigne 2011 , 2015 ) and appears to be very species specific (Frantz & Larson, this volume). It should also be noted that intense relationships can exist without domestication, notably in the case of pet keeping in hunter societies (see e.g. Erikson 2000 ). Nonetheless, numerous debates have arisen about measures and variability in order to determine if a particular skull is closer to the dog or wolf type. Of course, such differences can probably tell us if at one point in the ancestry of the individual there was enough human influence to modify its genotype and/or phenotype. However, it doesn’t tell us if this animal was domestic or wild in the sense it had a particular relationship with humans. The dingo is not a wolf, but is it a domestic animal? Beyond the selection that operates at the 74 Nicolas Lescureux species level and on an intergenerational scale, it appears that domestication is also an action that has to be repeated, a relation that has to be (re)constructed each generation in order to rebuild the special bond that exists between humans and animals in a domestic context.

Notes 1 E.g. the Saarloos wolf dog, the Czechoslovakian wolf dog, the Lupo Italiano, the Kun- ming wolf dog. 2 The black coat colour in wolves is probably the result of the introgression of a mutation resulting from hybridization with dogs (see Ollivier et al . 2013 ).

References Adams, J. R., Leonard, J. A. and Waits, L. P. 2003. Widespread occurrence of a domestic dog mitochondrial DNA haplotype in southeastern US coyotes. Molecular Ecology , 12, 541–546. Allendorf, F. W., Luikart, G. and Aitken, S. N. 2013. Conservation and the genetics of populations . Oxford, UK: Wiley-Blackwell. Anderson, T. M., et al . 2009. Molecular and evolutionary history of melanism in North American Gray Wolves. Science , 6(323), 1339–1343. Ardalan, A., et al . 2011. Comprehensive study of mtDNA among Southwest Asian dogs contradicts independent domestication of wolf, but implies dog-wolf hybridization. Ecology and Evolution , 1(3), 373–385. Bassi, E., et al . 2017. Trophic overlap between wolves and free-ranging wolf × dog hybrids in the Apennine Mountains, Italy. Global Ecology and Conservation , 9, 39–49. Benson, J. F., et al . 2013. Genetic and environmental influences on pup mortality risk for wolves and coyotes within a Canis hybrid zone. Biological Conservation , 166, 133–141. Boitani, L. 2003. Wolf conservation and recovery. In: Mech, L. D. and Boitani, L. eds. Wolves: Behavior, ecology, and conservation. Chicago, IL: The University of Chicago Press, 317–344. Boudadi-Maligne, M. and Escarguel, G. 2014. A biometric re-evaluation of recent claims for Early Upper Palaeolithic wolf domestication in Eurasia. Journal of Archaeological Science , 45, 80–89. Boyko, A. R., et al. 2009. Complex population structure in African village dogs and its implications for inferring dog domestication history. Proceedings of the National Acad- emy of Science , 106(33), 13093–13908. Caniglia, R., et al. 2013. Black coats in an admixed wolf x dog pack is melanism an indicator of hybridization in wolves? European Journal of Wildlife Research , 59, 543–555. Castric, V., et al . 2008. Repeated adaptative introgression at a gene under multiallelic balancing selection. PLoS Genetics , 4(8), e1000168. doi:1000110.1001371/journal.pgen.1000168. Chapron, G., et al . 2014. Recovery of large carnivores in Europe’s modern human-dominated landscapes. Science , 346(6216), 1517–1519. Ciucci, P., et al. 2003. Dewclaws in wolves as evidence of admixed ancestry with dogs. Canadian Journal of Zoology , 81, 2077–2081. Clutton-Brock, J. 1995. Origins of the dog: Domestication and early history. In: Serpell, J. ed. The domestic dog: Its evolution, behaviour and interactions with people . James Serpell ed. Cambridge: Cambridge University Press, 7–20. Beyond wild and domestic 75 Coppinger, R. and Coppinger, L. 2001. Dogs: A startling new understanding of canine origin, behavior and evolution . New York, NY: Scribner. Corbett, L. K. 1995. The Dingo in Australia and Asia. Ithaca, NY: Cornell University Press. Coulson, T., et al . 2011. Modeling effects of environmental change on wolf population dynamics, trait evolution, and life history. Science , 334, 1275–1278. Crockford, S. J. and Kuzmin, Y. V. 2012. Comments on Germonpré et al., Journal of Archaeological Science 36, 2009 ‘Fossil dogs and wolves from Palaeolithic sites in Bel- gium, the Ukraine and Russia: Osteometry, ancient DNA and stable isotopes’, and Ger- monpré, Lázničková-Galetová, and Sablin, Journal of Archaeological Science 39, 2012 ‘Palaeolithic dog skulls at the Gravettian Předmostí site, the Czech Republic’. Journal of Archaeological Science , 39, 2797–2801. Cronin, M. A. and Mech, L. D. 2009. Problems with the claim of ecotype and taxon status of the wolf in the Great Lakes region. Molecular Ecology , 18, 4991–4993. Daniels, M. J. and Corbett, L. K. 2003. Redefining introgressed protected mammals: When is a wildcat a wild cat and a dingo a wild dog? Wildlife Research , 30(3), 213–218. Dayan, T. 1994. Early domesticated dogs of the near East. Journal of Archaeological Sci- ence , 21, 633–640. Descola, P. 2004. Le sauvage et le domestique. Communications , 76(1), 17–39. Descola, P. 2013. Beyond nature and culture . Chicago, IL: University of Chicago Press. Descola, P. and Pálsson, G. 1996. Nature and society: Anthropological perspectives . London, UK: Routledge. Douglas, M. 1966. Purity and danger: An analysis of the concepts of pollution and taboo . London and New York, NY: Ark Paperbacks. Drake, A. G. and Klingenberg, C. P. 2010. Large-scale diversification of skull shape in domestic dogs: Disparity and modularity. The American Naturalist , 175(3), 289–301. Dressel, S., Sandström, C. and Ericsson, G. 2015. A meta-analysis of studies on attitudes toward bears and wolves across Europe 1976–2012. Conservation Biology , 29(2), 565–574. Ellen, R. and Katsuyoshi, F. 1996. Redefining nature: Ecology, culture and domestication . Oxford, UK: Berg. Erikson, P. 2000. The social significance of pet-keeping among Amazonian Indians. In: Pod- berscek, A. L., Paul, E. S. and Serpell, J. A. eds. Companion animals and us: Exploring the relationships between people and pets . Cambridge: Cambridge University Press, 7–26. Falcucci, A., Maiorano, L. and Boitani, L. 2007. Changes in land-use/land-cover patterns in Italy and their implication for biodiversity conservation. Landscape Ecology , 22, 617–631. Forsyth, T. 2003. Critical politcal ecology: The politics of environmental science . London and New York, NY: Routledge. Frantz, L. A. F., et al. 2016. Genomic and archaeological evidence suggest a dual origin of domestic dogs. Science , 352(6290), 1228–1231. Galibert, F., et al. 2011. Toward understanding dog evolutionary and domestication history. Comptes Rendus Biologies , 334, 190–196. Galov, A., et al . 2015. First evidence of hybridization between golden jackal (Canis aureus ) and domestic dog (Canis familiaris ) as revealed by genetic markers. Royal Society Open Science , 2(12). Garde, L. 2015. Sheep farming in France: Facing the return of the wolf. Carnivore Damage Prevention News , 11, 17–27. Gaubert, P., et al . 2012. Reviving the African Wolf Canis lupus lupaster in North and West Africa: A mitochondrial lineage ranging more than 6,000 km wide. PLoS ONE , 7(8), e42740. 76 Nicolas Lescureux Germonpré, M., et al. 2009. Fossil dogs and wolves from Palaeolithic sites in Belgium, the Ukraine and Russia: Osteometry, ancient DNA and stable isotopes. Journal of Archaeo- logical Science , 36(2), 473–490. Germonpré, M., et al . 2013. Palaeolithic dogs and the early domestication of the wolf: A reply to the comments of Crockford and Kuzmin (2012). Journal of Archaeological Sci- ence , 40, 786–792. Germonpré, M., Lázničková-Galetová, M. and Sablin, M. V. 2012. Palaeolithic dog skulls at the Gravettian Předmostí site, the Czech Republic. Journal of Archaeological Science , 39(2012), 184–202. Godinho, R., et al. 2011. Genetic evidence for multiple events of hybridization between wolves and domestic dogs in the Iberian Peninsula. Molecular Ecology, 20, 5154–5166. Gompper, M. E. 2014. The dog-human-wildlife interface: Assessing the scope of the problem. In: Gompper, M. E. ed. Free-ranging dogs and wildlife conservation. Oxford, UK: Oxford University Press, 9–54. Gottelli, D., et al. 1994. Molecular genetics of the most endangered canid: The Ethiopian wolf Canis Simensis . Molecular Ecology , 3, 301–312. Grewal, S. K., et al. 2004. A genetic assessment of the Eastern Wolf (Canis lycaon) in Algonquin Provincial Park. Journal of Mammalogy , 85(4), 625–632. Hailer, F. and Leonard, J. A. 2008. Hybridization among three native North American Canis species in a region of natural sympatry. PLoS ONE , 3(10), e3333. Hedrick, P. W. 2009. Wolf of a different colour. Heredity , 103, 435–436. Hedrick, P. W. 2013. Adaptive introgression in animals: Examples and comparison to new mutation and standing variation as sources of adaptive variation. Molecular Ecology , 22(18), 4606–4618. Hickey, K. 2011. Wolves in Ireland . Dublin: Open Air. Hindrikson, M., et al. 2012. Bucking the trend in wolf-dog hybridization: First evidence from Europe of hybridization between female dogs and male wolves. PLoS ONE , 7(10), e46465. Iacolina, L., et al . 2010. Y-chromosome microsatellite variation in Italian wolves: A contribution to the study of wolf-do hybridization patterns. Mammalian Biology, 75, 341–347. Iljin, N. A. 1941. Wolf-dog genetics. Journal of Genetics , 42(3), 359–414. Ingold, T. 2000. The perception of the environment: Essays in Livelihood, dwelling and skill . London, UK: Routledge. Knight, J. 2000. Introduction. In: Knight, J. ed. Natural enemies: People-wildlife conflicts in anthropological perspective . London, UK: Routledge, 1–36. Larson, G. and Fuller, D. Q. 2014. The evolution of animal domestication. Annual Review of Ecology, Evolution, and Systematics , 45(1), 115–136. Larson, G., et al . 2012. Rethinking dog domestication by integrating genetics, archaeology, and biogeography. Proceedings of the National Academy of Science, 109(23), 8878–8883. Latour, B. 1993. We have never been modern . Cambridge, MA: Harvard University Press. Leonard, J. A. and Wayne, R. K. 2008. Native Great Lakes wolves were not restored. Biol- ogy Letters , 4, 95–98. Leonard, J. A. and Wayne, R. K. 2009. Wishful thinking: Imagining that the current Great Lakes wolf is the same entity that existed historically (invited reply). Biology Letters , 5, 67–68. Lescureux, N. and Linnell, J. D. C. 2014. Warring brothers: The complex interactions between wolves (Canis lupus) and dogs (Canis familiaris) in a conservation context. Biological Conservation , 171, 232–245. Linnell, J. D. C., et al . 2003. Is the fear of wolves justified? A Fennoscandian perspective. Acta Zoologica Lituanica , 13, 34–40. Beyond wild and domestic 77 Linnell, J. D. C., et al. 2015. Framing the relationship between people and nature in the context of European conservation. Conservation Biology , 29, 978–985. Linnell, J. D. C., Swenson, J. E. and Andersen, R. 2001. Predators and people: Conserva- tion of large carnivores is possible at high human densities if management policy is favourable. Animal Conservation , 4, 345–349. Lorenzini, R., et al. 2014. Wolf-dog crossbreeding: ‘Smelling’ a hybrid may not be easy. Mammalian Biology , 79, 149–156. Mech, L. D. 1970. The wolf the ecology and behavior of an endangered species . Minne- apolis, MN: University of Minnesota Press. Mech, L. D. 1995. The challenge and opportunity of recovering wolf populations. Conser- vation Biology , 9, 270–278. Mech, L. D. 2009. Crying wolf: Concluding that wolves were not restored (comment). Biology Letters , 5, 65–66. Mech, L. D. 2010. What is the taxonomic identity of Minnesota wolves? Canadian Journal of Zoology , 88, 129–138. Mech, L. D. 2017. Where can wolves live and how can we live with them? Biological Conservation , 210, Part A, 310–317. Mech, L. D. and Boitani, L. 2003. Wolves: Behavior, ecology, and conservation . Chicago, IL: The University of Chicago Press. Mech, L. D. and Frenzel, L. D. 1971. Ecological studies of the timber wolf in Northeastern Minnesota . St Paul, MN: U.S. North Central Forest Experiment Station, Research Paper NC-52. Milton, K. 2000. Ducks out of water: Nature conservation as boundary maintenance. In: Knight, J. ed. Natural enemies . New York, NY: Routledge, 229–248. Monzón, J., Kays, R. and Dykhuizen, D. E. 2014. Assessment of coyote-wolf-dog admixture using ancestry-informative diagnostic SNPs. Molecular Ecology , 23(1), 182–197. Morey, D. F. 1994. The early evolution of the domestic dog. American Scientist , 82(4), 336–347. Moriceau, J.-M. 2007. Histoire du méchant loup. 3 000 attaques sur l’homme en France . Paris: Fayard. Muñoz-Fuentes, V., et al . 2010. The genetic legacy of extirpation and re-colonization in Vancouver Island wolves. Conservation Genetics , 11, 547–556. Murie, A. 1985 (1944). The wolves of Mount McKinley. Seattle, WA: University of Wash- ington Press. Newsome, A. E. and Corbett, L. K. 1982. The identity of the Dingo II. Hybridization with domestic dogs in captivity and in the wild. Australian Journal of Zoology , 30(2), 365–374. Newsome, A. E. and Corbett, L. K. 1985. The identity of the Dingo III. The incidence of dingoes, dogs and hybrids and their coat colours in remote and settled regions of Austra- lia. Australian Journal of Zoology , 33(3), 363–375. Newsome, A. E., Corbett, L. K. and Carpenter, S. M. 1980. The identity of the Dingo I. Morphological discriminants of dingo and dog skulls. Australian Journal of Zoology , 28(4), 615–625. Nowak, R. M. 1992. The red wolf is not a hybrid. Conservation Biology , 6(4), 593–595. Nowak, R. M. and Federoff, N. E. 1998. Validity of the red wolf: Response to Roy et al. Conservation Biology , 12(3), 722–725. Ollivier, M., et al . 2013. Evidence of coat color variation sheds new light on ancient canids. PLoS ONE , 8(10), e75110. Ollivier, M., et al . 2016. Amy2B copy number variation reveals starch diet adaptations in ancient European dogs. Royal Society Open Science , 3(11). 78 Nicolas Lescureux Olsen, S. J. 1985. Origins of the domestic dog: The fossil record . Tucson, AZ: The Univer- sity of Arizona Press. Ovodov, N. D., et al. 2011. A 33,000-year-old incipient dog from the Altai Mountains of Siberia: Evidence of the earliest domestication disrupted by the last Glacian Maximum. PLoS ONE , 6(7), e22821. Perri, A. 2016. A wolf in dog’s clothing: Initial dog domestication and Pleistocene wolf variation. Journal of Archaeological Science , 68, 1–4. Pionnier-Capitan, M., et al . 2011. New evidence for Upper Palaeolithic small domestic dogs in South-Western Europe. Journal of Archaeological Science , 38, 2123–2140. Randi, E. 2008. Detecting hybridization between wild species and their domesticated relatives. Molecular Ecology , 17, 285–293. Randi, E. 2011. Genetics and conservation of wolves Canis lupus in Europe. Mammal Review , 41(2), 99–111. Randi, E., et al. 2000. Mitochondrial DNA Variability in Italian and East European Wolves: Detecting the consequences of small population size and hybridization. Conservation Biology , 14(2), 464–473. Robbins, P. and Moore, S. A. 2013. Ecological anxiety disorder: Diagnosing the politics of the Anthropocene. Cultural Geographies , 20(1), 3–19. Rueness, E. K., et al. 2011. The cryptic African wolf: Canis aureus lupaster is not a golden jackal and is not endemic to Egypt. PLoS ONE , 6(1), e16385. Sablin, M. V. and Khlopachev, G. A. 2002. The earliest ice age dogs: Evidence from Eli- seevichi 1. Current Anthropology , 43, 795–799. Simberloff, D. 1996. Hybridization between native and introduced wildlife species: Impor- tance for conservation. Wildlife Biology , 2, 143–150. Skogen, K., Mauz, I. and Krange, O. 2008. Cry wolf!: Narratives of wolf recovery in France and Norway. Rural Sociology , 73, 105–133. Torres, R. T., et al. 2017. Hybridization between wolf and domestic dog: First evidence from an endangered population in central Portugal. Mammalian Biology – Zeitschrift für Säugetierkunde , 86, 70–74. Tresset, A. and Vigne, J.-D. 2011. Last hunter-gatherers and first farmers of Europe. Comptes Rendus Biologies , 334(3), 182–189. Trouwborst, A. 2014. Applying the Bern Convention on the Conservation of European Wildlife and Natural Habitats to the Problem of Hybridisation Between Wolves (Canis Lupus) and Domestic Dogs – Document T-PVS/Inf (2014) 15 . Council of Europe – Standing Committee of Convention on the Conservation of European Wildlife and Natu- ral Habitats of the Council of Europe. Vanak, A. T. and Gompper, M. E. 2009. Dogs Canis familiaris as carnivores: Their role and function in intraguild competition. Mammal Review , 39(4), 265–283. Verginelli, F., et al. 2005. Mitochondrial DNA from prehistoric canid highlights relationships between dogs and South-East european wolves. Molecular Biology and Evolution , 22(12), 2541–2551. Vigne, J.-D. 2011. The origins of animal domestication and husbandry: A major change in the history of humanity and the biosphere. Comptes Rendus Biologies, 334(3), 171–181. Vigne, J.-D. 2015. Early domestication and farming: What should we know or do for a better understanding? Anthropozoologica , 50(2), 123–150. Vilà, C., et al . 1997. Multiple and ancient origins of the domestic dogs. Science , 276(5319), 1687–1689. Vilà, C., et al . 2003. Combined use of maternal, paternal and bi-parental genetic markers for the identification of wolf-dog hybrids. Heredity , 90, 17–24. Beyond wild and domestic 79 Vilà, C. and Wayne, R. K. 1999. Hybridization between wolves and dogs. Conservation Biology , 13(1), 195–198. vonHoldt, B. M., et al . 2010. Genowe-wide SNP and haplotype analyses reveal a rich history underlying dog domestication. Nature , 464, 898–902. vonHoldt, B. M., et al. 2011. A genome-wide perspective on the evolutionary history of enigmatic wolf-like canids. Genome Research , 21, 1294–1305. Wayne, R. K., Roy, M. S. and Gittleman, J. L. 1998. Origin of the red wolf: Response to nowak and federoff and gardener. Conservation Biology , 12(3), 726–729. Wayne, R. K. and vonHoldt, B. M. 2012. Evolutionary genomics of dog domestication. Mammalian Genome , 23, 3–18. Wilson, P. J., et al. 2000. DNA profiles of the eastern Canadian wolf and the red wolf provide evidence for a common evolutionary history independent of the gray wolf. Cana- dian Journal of Zoology , 78, 2156–2166. Zeder, M. A. 2011. The origins of agriculture in the Near East. Current Anthropology , 52(4), S221–S235.

4 Wild game or farm animal? Tracking human-pig relationships in ancient times through stable isotope analysis

Marie Balasse, Thomas Cucchi, Allowen Evin, Adrian Bălăşescu, Delphine Frémondeau and Marie-Pierre Horard-Herbin

Introduction The beginnings of animal domestication involved a gradual intensification of the long-standing relationship between humans and animals (Vigne et al . 2009 ; Vigne 2011). Different pathways to domestication have been proposed, resulting from associations of different kinds ( Zeder 2012 ; Larson and Burger 2013 ): the commensal pathway, where animals taking advantage of the resources in human habitats progressed from anthropophily to commensalism, to mutualistic relationships as a foundation for domestication; the prey pathway, where game management strategies designed to organize prey availability eventually turned into herd management; and the directed pathway, where animals were brought directly and deliberately under human control to satisfy other purposes such as, for example, transportation (Zeder 2012; Larson and Burger 2013; Larson and Fuller 2014). In this scheme, the intention-driven directed pathway probably emerged once humans had become familiar with domesticated animals ( Zeder 2012 ). Although the increasing entanglement of humans with the controlled species may have occurred unintentionally in the very first steps of both the commensal and prey pathways ( Larson and Fuller 2014 ), human intent must be considered as a key element in reaching the final destination of the process (Zeder 2009, 2012 ; Vigne 2011). One question is whether human intent may be used to define the start of the domestication process per se, or if the process of changing human-animal relationships should be considered as a continuum ( Ervynck et al . 2002 ; Vigne 2011 , 2012 ; Marshall et al. 2014). The recent recognition of long-term gene flows between wild and domestic stocks to a larger extent than previously thought, underlines the complexity of the trajectories and the difficulty to establish boundaries, raising questions regarding interpretation of the archaeological record ( Marshall et al. 2014). One important challenge is to be able to identify biological markers of (an ancestry of) domestication, but key to the debate is also a better characterization of the human-animal relationships (Marshall et al . 2014 ). This means documenting, beyond the goals, the terms of the relationship and, most importantly, reaching the herd/individual scale rather than the lineage scale. 82 Marie Balasse et al. Zooarchaeology deals with dead animals whose remains are found in association with a human presence at the archaeological sites. These remains are the testimony of the interactions between people and animals, but the intentionality and the behaviour associated with this relationship are irremediably lost and can only be inferred indirectly. Despite this, the skeleton retains a record of events that occurred during the lifetime of the animal, which is not genetically inherited, providing invaluable information on the status of the individual in the human-animal relationship. In combination with osteology, the stable isotope analysis of zooarchaeological remains has developed during the last 15 years specifically towards achieving this goal, through the reconstruction of husbandry practices (Balasse 2014). The stable isotope composition of elements in the skeleton informs about the animal’s diet, from which they are assimilated, and indirectly about the environmental setting of the husbandry system. This approach helps to capture the practices that were implemented to organize cohabitation between humans and animals; this includes feeding strategies and the place allocated to animals within the human domestic sphere. Any effort in this direction would help to define an eco-sociological approach ( Vigne 2015 ) and to reconcile archaeology with anthropology on the subject of animal domestication ( Digard 1988 ). In this chapter, we present how the stable isotope analysis of animal remains provides new avenues to explore ancient human-animal relationships, focusing on pig husbandry. Pigs have a long-standing history of domestication. Zooarchaeol- ogy indicates they were independently domesticated in at least two distant regions of the world, in eastern Anatolia around 9000 cal BP ( Ervynck et al . 2002 ; Larson et al . 2007a ) and in central China around 8600 cal BP ( Cucchi et al . 2011 ). In eastern Anatolia, pig domestication involved a long stage of wild boar management (Ervynck et al . 2002 ; Vigne et al. 2009). The prey and commensal pathways may both be adduced to define the mechanisms behind the intensification of the relationship between pigs and humans (Ervynck et al . 2002 ; Larson and Füller 2014; Marshall et al. 2014). Subsequent diffusion of Near Eastern and Chi- nese domestic pigs into Europe and Southeast Asia involved complex scenarios of dispersal and introgressions between introduced domestic pigs and local wild boars (Larson et al . 2007b ; Ottoni et al . 2013 ; Marshall et al . 2014 ; Larson and Burger 2013; Evin et al. 2015b). The mechanisms behind the extent of these gene flows can be better understood in the light of husbandry practices, which partly determine occasions for local wild and introduced domestic populations to meet and admix. Pigs are adaptable to a variety of husbandry systems. The ethnology of traditional husbandry documents a large panel of cohabitation types between pigs and humans, from the closest to more distant interactions: backyard pigs/ stalled pigs/extensively herded pigs/free-range pigs ( Dwyer 1996 ; Albarella et al . 2007 , 2011 ; Halstead and Isaakidou 2011 ; Hadjikoumis 2012 ; Silitoe 2007 ). The proximity of pig-human cohabitation may define the intensity of the relationship. In their ethnological survey of traditional pig husbandry in Greece, Halstead and Isaakidou (2011 , p.164) mention that stall-fed pigs could become very tame, “rolling on (their) back to be petted before feeding”. This nearly pet status would bring pigs to the most intense degree of the animal-human relationship (Vigne Wild game or farm animal? 83 2011). While traditional genetic and morphological attributes do not provide any hint in this matter, stable isotope biochemistry may contribute greatly to documenting this range of relationships. Among the four studies chosen to illustrate our approach, the first one is taken from a mediaeval Western European archaeological context where the subsistence economy was largely dominated by husbandry with hunting playing a very minor role. This study was chosen as a starting point to illustrate how stable isotope analyses may be used to investigate the scale of pig husbandry in a case where the domestic status of the pigs was certain. The following examples, in Neolithic China and Chalcolithic Romania, illustrate how the level of complexity rises when the challenge is also to distinguish between the wild and the domestic components among archaeological suids, which stable isotopes help to define. Complex scenarios sometimes lead to the necessity of a new reading embracing the varying degrees in the intensity of the animal-human relationship rather than adopting the often over-simplistic dichotomic categorization of wild versus domestic. Last, stable isotopes can provide insights into the feral component among archaeological suids. This feral component – an animal with a domestic ancestry but a free direct status – is particularly illuminating when we seek to capture a contrast between a biological definition of domesticity as a genetically inheritable character, by opposition to an anthropological definition of it, which considers the direct relationship to humans and the place where the animal lives.

Pig diet and scale of herding Stable isotope analysis is often considered the most straightforward tool to address the scale of pig herding, through the characterization of pigs’ diet. The opportunistic feeding behaviour of pigs makes them adaptable to a variety of feeding systems. In particular, compared to extensively herded pigs, household pigs may become more dependent on domestic sources of animal protein (leftovers). The consecutive ascension in the food chain may be traced through the analysis of bone stable nitrogen isotope ratio (δ 15 N), which significantly increases with each trophic step in a food web (Ervynck et al. 2007). Illustrative of this approach is the study of pig husbandry in the mediaeval city of York. York has been the site of numerous archaeological excavations, covering the period from its foundation at the end of the 1st century AD to the modern town. The zooarchaeological remains testify to the presence of pigs from the Roman to post-mediaeval times. Pigs provided significant amounts of meat, although they were never the principal source. Cereals seem to have been the main staple food, while beef provided most of the red meat and other food sources included chickens as a source of eggs, and fish. Hunting seems to have supplied very little meat (O’Connor 2000). Cattle and sheep were slaughtered beyond the optimal age for meat production, suggesting that their meat could have been a by-product of countryside husbandry systems focused on secondary productions (milk, wool, dung, traction): most of York’s meat supply would then have derived from what the agrarian economy of the surroundings could spare ( O’Connor 2000 ). Pigs, in contrast, were slaughtered as 84 Marie Balasse et al. sub-adults, at their optimal age for meat production. From this observation and from the findings of a few neonatal pig bones implying the presence of breeding sows in the city, it was suggested that pigs could have been raised in the backyards with the chickens and could have had a role in refuse disposal ( O’Connor 2000 ). Against this well-argued depiction of backyard pig husbandry in mediaeval York, the hypothesis of pigs feeding on human refuse was not supported by consecutive stable isotope analyses of pig bones ( Figure 4.1 ). Pig remains from the late 12th to 13th centuries yielded δ15 N values similar to those measured in sheep recovered from the same refuse dumps, suggesting that pigs were not holding a higher position in the food chain (Hammond and O’Connor 2013). Pigs also yielded stable carbon isotope values (δ 13C), reflecting a predominantly terrestrial diet. Those were by no means comparable to the stable isotope signatures in the human populations recovered from York cemeteries from the 11th century

Figure 4.1 Results from stable carbon (δ 13 C) and nitrogen (δ 15 N) analysis of bone collagen from pigs, sheep and human remains from the mediaeval city of York. The pigs grouped in the convex hull yielded values similar to those measured in sheep, suggesting countryside herding rather than household keeping; one pig from an early 13th century context yielded values within the range of those measured in the human population, i.e . the expected value for a refuse consumer. Source : Data from Müldner and Richards (2007 ); Hammond and O’Connor (2013 ). Wild game or farm animal? 85 on, marked by higher δ13 C values reflecting the growing importance of marine fish in diet (Müldner and Richards 2007). In fact, the great similarity between stable isotope values measured in pigs and sheep, and the significant discrepancy between values measured in pigs and humans both suggest that pigs were not fed from human refuse. Although it may not be ruled out that pigs were actually kept in backyards and predominantly fed vegetable/cereal by-products (Halstead and Isaakidou 2011 ), they could also have been raised in countryside farms. Only one pig from the early 13th century context yielded δ 15 N and δ 13C values within the range of variation of those measured in the human population, i.e . the expected value for a refuse consumer ( Figure 4.1 ; Hammond and O’Connor 2013 ). In any case, in light of these results, the previous picture of town pig husbandry is greatly nuanced. A similar scenario occurred when stable isotope analyses were implemented to investigate pig husbandry practices in the Celtic village of Levroux Les Arènes (2nd century BC , centre of France). In a context of demographic management dedicated to standardized meat production, and against the well-argued hypothesis of a constrained system where pigs would have been penned within the village and fed from consumption refuse and agricultural by-products, stable isotope analyses of pig remains highlighted a predominantly herbivorous diet on a yearly scale ( Frémondeau et al . 2013 ) and autumn foraging in the forest ( Frémon- deau et al. 2017). These two examples emphasize the multiplicity and complexity of pig husbandry systems, and the need to describe independently each of the different elements of these systems in the manner of anthropological surveys.

Interconnections between pig husbandry and millet agriculture in Ancient China Beyond household husbandry, and by opposition to extensive herding where natural resources are exploited, the animal-man relationship may include cultivated lands and their productions as a third party. Defining interactions between animal husbandry and agriculture is highly relevant to characterize the intensity of the relationship between animals and humans. From the earliest steps of agriculture, the attractiveness of cultivated fields could have played a role in the commensal pathway of pigs’ domestication. In all instances, those connections may help to reveal investment in stock feeding and assess the place of the animal within the human environment (animals in the fields/fed agricultural by-products/herded further away to prevent damage to the fields). Ethnographical studies of traditional pig husbandry in the Mediterranean area report both the necessity to keep the pigs away from growing crops and their feeding on cereal stubble after harvesting as elements of the pastoral cycle ( Albarella et al . 2011 ; Halstead and Isaakidou 2011 ). In this perspective, connections between millet cultivation and pig husbandry in ancient China have been the subject of many studies. Millet agriculture dates back to the Early Neolithic in northeastern China ( Lu et al . 2009 ). In the Yellow River basin, millet contributed significantly to human diet from the earliest Yang- shao times (6600–5400 cal BP) at a period when rice is rare in the archaeobotanical record (Pechenkina et al . 2005 ; Barton et al. 2009). Pigs are the main domestic 86 Marie Balasse et al. animal in the Early and Middle Neolithic farming economy. Domesticated in central China around 8600 cal BP ( Cucchi et al . 2011 ), they remained central to husbandry until the Longshan period (4600–3900 cal BP) at a time when cattle and sheep were introduced and started playing an increasingly important role in daily life and ritual activities (Flad et al. 2007). Interconnections between millet cultivation and husbandry in the Yellow River basin during Neolithic times have been highlighted repeatedly ( Pechenkina et al . 2005 ; Barton et al . 2009 ; Zhang et al . 2011 ; Chen et al . 2016a , 2016b ; Cucchi et al . 2016 ; Dai et al . 2016 , in press). Millet is a grass using the C 4 photosynthetic process, as opposed to plants using the C3 photosynthetic process, which predominate in the surrounding environment of those studies; C3 and C4 plants may be identified owing to a marked difference in their δ 13 C values. Elevated δ13 C values in bone collagen reflect significant contributions of millet to the diet of humans as well as animals (dogs, pigs and later on cattle and sheep). Pechenkina et al . (2005 ) have suggested that the interconnection of millet cultivation and pig raising helped to ensure the stability of Middle Neolithic farming economies in the Yellow River basin, by offering flexibility through variable redistribution of agricultural products, depending on annual fluctuations in agricultural outputs and the availability of wild resources. Later on, during the Late Neolithic in China, the different extent to which cattle, sheep and pigs benefitted from this resource has also been linked to their role in social or ritual-related activities ( Dai et al . in press ). As a consequence of this interconnection, δ13 C ratios heavily marked by a 15 C 4 plant (millet) signature, in combination with elevated δ N values reflecting ascension in the food chain, have been designated as “the isotopic identity of domestication” at the Dadiwan site in the Yellow River basin (Barton et al . 2009 , p. 5525). Dadiwan was first occupied intermittently from 7900 to 7200 cal BP (pre-Yangshao phase), and again more intensively from 6500 to 4900 cal BP by farmers carrying the Yangshao cultural package. Between both phases, a gradient in the relationship between people, millet, dogs and pigs has been observed. In this study, Sus species remains were analyzed isotopically with no reference to any osteomorphometrical criterion regarding their identification as wild boar or domestic pig. In Phase 1, while broomcorn millet was represented in the archaeobotanical record by a few seeds, some dogs yielded a signal of significant consumption of millet, much likely provisioned by humans, while pigs and all large wild mammals belong to a C3 ecosystem ( Figure 4.2 ). This phase was described as a “low-intensity domestic relationship” between hunter-gatherers, plants and animals. By contrast, in the Yangshao phase, where both broomcorn and foxtail millets are attested, a more intensive farming system was set up. Most pigs – and all dogs – display high δ 13 C and δ 15 N values clearly reflecting their entrance into the domestic farming sphere ( Figure 4.2 ). A gradient can still be observed, between a majority of pigs maintained in very close proximity to humans (fed millet and animal food scraps; these pigs yielded the highest δ 13 C and δ15 N values); a group of pigs with intermediate δ13 C and δ 15 N values representing either wild animals that raided millet fields (and entered in a commensal niche) or free- ranging pigs occasionally fed with millet; and finally very few specimens with Wild game or farm animal? 87

Figure 4.2 Reliance of pigs on domestic farmed food (millet, animal protein scraps) in Dadiwan in ancient China, as evidenced from stable isotope composition of bone collagen. Source : Modified after Barton et al . (2009 ). low δ13 C and δ 15N values, which foraged and were presumably hunted wild boars or pigs foraging in the wild ( Barton et al . 2009 ). In the study at Dadiwan, the “isotopic identity of domestication” reflects a strong intensification of the relationship between pigs and humans, from a state where pigs do not belong to the farming sphere (while some dogs do) to a state where most of them rely to a very high extent on a millet food chain. It is conceivable that stronger control may have been applied in the very first steps of intentional control of humans on animals, which could then be released once the relationship was strongly imprinted in the animals’ behaviour. A reversal on this scale of proximity may occur afterwards. In the large rural village of Xiawang- gang (Henan Province, northeastern China), a diachronic study spanning the occupation from the Yangshao period to the Han Dynasty (200 BC to AD 200) made it possible to highlight changes through time in the intensity of the linkage between millet cultivation and a husbandry system dominated by pigs during the whole period ( Cucchi et al . 2016 ). Bone collagen stable isotope analysis on Sus scrofa 88 Marie Balasse et al. remains from all phases demonstrated that millet contributed to pigs’ diet from the earliest phase of occupation (Middle Yangshao, 6600–5400 cal BP), although to varying amounts. This contribution gradually intensified to reach its highest level in the Longshan period (4600–3900 cal BP) ( Figure 4.3 ). Interestingly, geometric morphometric analyses conducted on pig molars demonstrated a rupture at the Longshan period from the population history of Xiawanggang pig herds from the Yangshao period to Han dynasty ( Cucchi et al . 2016 ). The Longshan pigs were idiosyncratic and very small, suggesting either an allochthonous origin or a drastic change in husbandry practice that impacted their growth and development. This change would have been a greater confinement through strict penning as suggested by the intensification of millet foddering, i.e . tighter control of diet. Alongside these confined and millet-fed small pigs occurred individuals with predominantly C3 diet (foraging in the wild) and displaying larger-size third molars, possibly reflecting the presence of wild boars in this assemblage ( Figure 4.3 ). During the following Erlitou phase (Early Bronze Age), a lower contribution of millet in the pigs’ diet was observed, potentially reflecting a return to a system mixing extensive foraging and millet foddering, while the high trophic level of some of the pigs still clearly suggested an elevated position within the domestic food chain ( Figure 4.3 ). In this context, however, the interpretation of intermedi- 13 ate δ C values is less straightforward as they could also reflect contribution of C3 domestic crops, including rice, present from the Yangshao period, but also wheat, whose cultivation widely developed during the Han dynasty ( Cucchi et al . 2016 ).

When morphological, genetic and isotopic categorizations do not match: the identification of feral pigs From the different studies described previously, it appears that small (domestic- sized) suids can be found on a gradient from feeding in the natural environment to relying heavily on farming food, reflecting a panel of practices from extensive herding to penning, including mixed practices. By contrast, the large (wild- sized) suids are not found in the human household (no case of heavy reliance on produced food), but could partly be feeding in cultivated lands (commensalism). This would suggest that the diet criterion globally supports the size criterion in assigning small suids to a domestic status. Identification of wild and domestic pigs in the archaeological record remains challenging since they have very close skeleton morphology, especially in the earliest stage of the domestication process; furthermore, the two forms can hybridize, rendering the identification even more problematic. The size criterion has been widely used because domestication led to a decrease in body size in many species ( Boessneck and von den Driesch 1978 ; Meadow 1989 ; Vigne et al. 2005). However, the use of a size criterion alone to distinguish between the wild and domestic forms is now debated in pigs ( Cucchi et al . 2011 ; Evin et al . 2013 , 2015a ). With the advance of geometric morphometrics (GMM) in the field of zooarchaeology ( Cucchi et al . 2015 ), some limitations of classical morphometrics have been alleviated. This morphometric approach, which relies on two- or three-dimensional coordinates instead of Figure 4.3 Gradual changes through time in pigs’ diet in Xiawanggang, with an increasing contribution of millet and animal proteins. Specimens from the Longshan period (black diamonds) with lower stable isotope values (dotted outline) have third molars within the range of wild boars. Source : Data from Cucchi et al . (2016 ). 90 Marie Balasse et al. linear measurements, provides a better capture of the complexity of the biological forms. The possibility to align the points’ configuration by a standardization procedure called Procrustes superimposition also provides the possibility to compare the size and shape components of the specimens more precisely. In pigs, this approach has been extensively applied to explore the dental size and shape variation of wild and domestic pigs over Eurasia and has provided clear evidence for significant differences between modern wild, domestic and hybrid forms ( Larson et al . 2007b ; Cucchi et al . 2009 ; Evin et al . 2015a; Cucchi et al . 2016 ). In particular, when applied to assemblages dating from the Neolithic to the Iron Age in Romania, the approach has revealed, beside the two contrasted groups of specimens with small size and ‘domestic’ shaped molars, on the one hand, and large and ‘wild’ shaped molars, on the other hand, the presence of a third group, consisting of suids with large size and ‘domestic’ shaped molars (Evin et al . 2015b). Those specimens were found to be on average smaller than those with a wild shape, in agreement with the size reduction observed during domestication. Ancient DNA revealed that a great majority of the Neolithic and Chalcolithic suids with small size ‘domestic’ shaped molars carried the Near-Eastern Y1 mitochondrial haplotype establishing a link with Anatolia from where this domestic lineage came to Europe along the Danubian route ( Larson et al . 2007a ). Among the specimens with large and ‘domestic’ shaped molars that were genotyped, the earliest one, from the Middle Neolithic (second half of the sixth millennium BC ) carried the European haplotype E1-C (possibly reflecting the introduction of local wild boar into domestic stock); later on, in the Late Chalcolithic (second half of the fifth millennium BC ) while small domestic pigs carrying the Near-Eastern Y1 haplotypes were still present, this haplotype was also present in large specimens displaying both a ‘wild’ or ‘domestic’ shape, suggesting in all instances gene flows between introduced domestic pigs and local wild boar at least at this period of time ( Evin et al . 2015b ). This brief description reveals a very complex situation where groups potentially described as ‘domestic’ by contrast to ‘wild’, either from size, shape or genetic data, may not always correspond at the margins, because they actually cover different biological or ecological realities. Currently, the relationship to humans of these suids with large ‘domestic’ shaped molars remains to be defined since these specimens can be free from human control over their lifetime but carry an inherited domestic imprint in their morphological and/or a genetic make-up. Stable isotopes were introduced with the purpose of determining the degree of these pigs’ food dependence on humans as an additional element to disentangle their relationship with Romanian Chalcolithic societies. The study focused on three contemporaneous assemblages in Borduşani-Popină, Hârşova-tell and Vităneşti-Măgurice in southern eastern Romania. All belong to the Gumelnița culture (Late Chalcolithic, 4600–3900 cal BC ), at a time period when the pig significantly increased in importance after one and a half millennia of sparse representation in early Neolithic to early Chalcolithic assemblages ( Bălăşescu 2014). Results from δ 15 N and δ 13C analyses of bone collagen from a spectrum of wild species provided baseline isotopic signatures for the surrounding landscape ( Figure 4.4A ). The incorporation of these pigs in the farming food Figure 4.4 A : Stable isotope ratios in bone collagen of the main species from the Gumelniţa culture assemblages at Borduşani-Popină, Hârşova-tell and Vităneşti-Măgurice. Domestic herbivores = cattle, sheep; wild terrestrial herbivores = red deer, roe deer, aurochs; aquatic piscivores = sander, Northern pike, catfish; aquatic omnivores = carp. A high trophic level for pigs (higher δ 15 N values) suggests household husbandry. Wild and domestic pigs were identified based on traditional size measurements. B : Stable isotope ratios in bone collagen from suids with small ‘domestic’, large ‘wild’ and large ‘domestic’ molars (from geometric morphometric analyses). The large suids with domestic shaped molars show lower δ 15 N values than the small domestic pigs. Source : Modified after Balasse et al . (2016 ). 92 Marie Balasse et al. chain was then evaluated from comparisons with dogs as a reference for a human leftover disposer, on the one hand, and wild boar (identified from a size criterion) as a reference for an extensive free-ranging animal on the other hand. At all three sites, high δ 15 N values for pigs (sometimes comparable to those measured in dogs) reflected diets consisting of significant amounts of animal protein, likely from various human wastes. Mid-range δ13 C values suggested the contribution of remnants of hunted game, and/or dairy products counterbalanced on the δ 13 C values scale by aquatic resources. Pigs could benefit from all these resources from deliberate provisioning with leftovers or by free ranging in and around the settlement, adopting the role of refuse disposers ( Balasse et al . 2016 , 2017 ). At Vităneşti-Măgurice, stable isotope analyses targeted the mandibular bone of suids that had been subjected to a geometric morphometric analysis. Suids with large ‘wild’ shaped molars (wild boar) delivered δ 13 C values similar to those measured in large herbivores from open environments. By comparison, suids with small ‘domestic’ shaped molars (domestic pigs) hold in majority a higher trophic level than wild boars (higher δ 15 N values; Balasse et al . 2016 ; Figure 4.4B ). The unresolved third group of suids with large ‘domestic’ shaped molars yielded δ15 N values setting them apart from the small domestic pigs based on their diet (Fig- ure 4.4B). The hypothesis of extensively herded domestic stock could not be ruled out; this system could have favoured hybridization with local wild boar, which would explain their larger size. However, it was considered more likely that the large domestic-shaped suids could have been feral pigs that escaped domestic herding and hybridized with wild boar. In this hypothesis, the feral pigs cannot be considered biologically to be wild boars because they have kept an imprint of their domestic ancestry; they also cannot be considered functionally domestic pigs, since they had probably lost association with people and were acquired through hunting. From an anthropological point of view, they should therefore be considered to belong to the wild category ( Balasse et al . 2016 ). Of interest is the significance of the presence of these feral animals – what does it tell us about husbandry practices? how much of it resulted from a deliberate mixing of wild and domestic populations – as well as their relationship to people – were feral pigs identified as such, i.e ., did the perception of these communities correspond to our modern categorization between these three groups: domestic, wild and feral pigs? Unfortunately, these questions might be extremely difficult to address in archaeology.

Conclusion The question regarding feral pigs in the last case study clearly illustrates the difference between the biological and anthropological definitions of the domestic animal. The latter takes a socio-ecological view, where domestic animals are those care for by people, while wild animals have no (or have lost all) association with people. Even in this definition, the categories are not clear-cut, depending on the perception that the herders have of their own investment in pig management ( Dwyer 1996 ), and this makes categorization even more difficult in the Wild game or farm animal? 93 archaeological record where this last dimension is lost. The biological and anthropological definitions of the domestic animal are not to be opposed. They fundamentally refer to different scales despite sharing the same terminology: while biology traces plurigenerational relationships (using inheritable markers), the social approach describes individual trajectories. Typically, the feral animals, with a domestic ancestry but wild ecological behaviour, have to do with both definitions used in conjunction. The demonstration of their presence in the archaeological record is not straightforward. It must combine evidence from inherited characters (which is the case of all genetic, and part of the morphometric, characters) and from the relationship that they actually share with people. We have seen that combining geometric morphometrics and stable isotope analyses could help progress in this matter, although even in this case, alternative hypotheses could still be proposed between feral pigs or extensively herded larger size domestic pigs. Efforts must be pursued to describe the terms of human-animal relationships in the past. Stable isotope analyses allow us to address key parameters of these relationships: management of animal feeding and indirectly the place allocated to the animal within the domestic sphere; they help to provide nuance in the larger picture, but do not provide definite answers. The answers are likely to come from the confrontation of multiple criteria.

Acknowledgments We would like to thank Charles Stépanoff and Jean-Denis Vigne for inviting us to be part of the “Domestication et communautés hybrides” conference held in April 2016 in Paris, and for their valuable comments to this chapter.

References Albarella, U., Manconi, F. and Trentacoste, A., 2011. A week on the plateau: Pig husbandry, mobility and resources exploitation in Central Sardinia. In : U. Albarella, ed. Ethnoar- chaeology: The Present Past of Human: Animal Relationships. Oxford: Oxbow, 143–159. Albarella, U., Manconi, F., Vigne, J.-D. and Rowley-Conwy, P., 2007. Ethnoarchaeology of pig husbandry in Sardinia and Corsica. In : U. Albarella, K. Dobney, A. Ervynck and P. Rowley-Conwy, eds. Pigs and Humans: 10,000 Years of Interactions. Oxford: Oxford University Press, 285–307. B ălăşescu, A., 2014. Arheozoologia neo-eneoliticului de pe Valea Teleormanului . Muzeul Național de Istorie a României, Colecția Muzeului Național, Seria Cercetări Puridisci- plinare XIV. Cluj Napoca: Editura Mega. Balasse, M., 2014. Isotopic studies of husbandry. In : C. Smith, ed. Encyclopedia of Global Archaeology . New York: Springer, 4120–4128. Balasse, M., Bălăşescu, A., Tornero, C., Frémondeau, D., Hovsepyan, R., Gillis, R. and Popovici, D., 2017. Investigating the scale of herding in Chalcolithic pastoral communities settled by the Danube river in the 5th millennium BC : A case study at Borduşani- Popină and Hârşova-tell (Romania). Quaternary International , 436, Part B, 29–40. Balasse, M., Evin, A., Tornero, C., Radu, V., Fiorillo, D., Popovici, D., Andreescu, R., Dob- ney, K.M., Cucchi, T. and Bălăşescu, A., 2016. Wild, domestic and feral? Investigating 94 Marie Balasse et al.

the status of suids in the Romanian Gumelniţa (5th mil. cal BC ) with biogeochemistry and geometric morphometrics. Journal of Anthropological Archaeology , 42, 27–36. Barton, L., Newsome, S.D., Chen, F.-H., Wang, H., Guilderson, T.P. and Bettinger, R.L., 2009. Agricultural origins and the isotopic identity of domestication in northern China. Proceedings of the National Academy of Science , 106, 5523–5528. Boessneck, J. and von den Driesch, A., 1978. The significance of measuring animal bones from archaeological sites. In : R.H. Meadow and M.A. Zeder, eds. Approaches to Faunal Analysis in the Middle East . Bulletin n 2. Cambridge: Peabody Museum, 25–40. Chen, X.-L., Fang, Y.-M., Hu, Y.-W., Hou, Y.-F., Lü, P., Yuan, J., Song, G.-D., Füller, B.T. and Richards, M.P., 2016a. Isotopic reconstruction of the late longshan period (ca. 4200–3900 BP) dietary complexity before the onset of state-level societies at the wadian site in the Ying River Valley, Central Plains, China. International Journal of Osteoar- chaeology , 26, 808–817. Chen, X.-L., Hu, S.-M., Hu, Y.-W., Wang, W.-L., Ma, Y.-Y., Lü, P. and Wang, C.-S., 2016b. Raising practices of neolithic livestock evidenced by Stable Isotope analysis in the Wei River Valley, North China. International Journal of Osteoarchaeology , 26, 42–52. Cucchi, T., Baylac, M., Evin, A., Bignon-Lau, O. and Vigne, J.-D., 2015. Morphométrie géométrique et archéozoologie: Concepts, méthodes et applications. In : M. Balasse, J.-P. Brugal, Y. Dauphin, E.-M. Geigl, C. Oberlin and I. Reiche, eds. Messages d’os. Archéo- métrie du squelette animal et humain. Paris: Editions des Archives Contemporaines, 197–216. Cucchi, T., Dai, L., Balasse, M., Zhao, C., Gao, J., Hu, Y., Yuan, J. and Vigne, J.-D., 2016. Social complexification and pig (Sus scrofa) husbandry in ancient China: A combined geometric morphometric and isotopic approach. PLoS ONE . doi:10.1371/journal. pone.0158523. Cucchi, T., Fujita, M. and Dobney, K., 2009. New insights into pig taxonomy, domestication and human dispersal in Island South East Asia: Molar shape analysis of Sus remains from Niah Caves, Sarawak. International Journal of Osteoarchaeology , 19, 508–530. Cucchi, T., Hulme-Beaman, A., Yuan, J. and Dobney, K., 2011. Early Neolithic pig domestication at Jiahu, Henan Province, China: Clues from molar shape analyses using geometric morphometric approaches. Journal of Archaeological Science , 38, 11–22. Dai, L., Balasse, M., Yuan, J., Zhao, C., Hu, Y. and Vigne, J.-D., in press. Cattle and sheep raising and millet growing in the Longshan Age in central China: Stable isotope investigation at the Xinzhai site. Quaternary International . doi:10.1016/j.quaint.2016.02.035 Dai, L.L., Li, Z.P., Zhao, C.Q., Yuan, J., Hou, L.L., Wang, C.S., Füller, B.T. and Hu, Y.W., 2016. An isotopic perspective on animal husbandry at the Xinzhai site during the initial stage of the legendary Xia Dynasty (2070–1600 BC ). International Journal of Osteoar- chaeology , 26, 885–896. Digard, J.-P., 1988. Jalons pour une anthropologie de la domestication animale. L’Homme , 108 (4), 27–88. Dwyer, P.D., 1996. Boars, barrows, and breeders: The reproductive status of domestic pig populations in Mainland New Guinea. Journal of Anthropoplogical Research , 52, 481–500. Ervynck, A., Dobney, K., Hongo, H. and Meadow, R., 2002. Born free? New evidence for the status of Sus scrofa at Neolithic Çayönu Tepesi (Southeastern Anatolia, Turkey). Paléorient , 27 (2), 47–73. Ervynck, A., Lentacker, A., Müldner, G., Richards, M. and Dobney, K., 2007. An investigation into the transition from forest dwelling pigs to farm animals in medieval Flanders, Belgium. In : U. Albarella, K. Dobney, A. Ervynck and P. Rowley-Conwy, eds. Pigs and Humans: 10,000 Years of Interaction . Oxford: Oxford University Press, 170–193. Wild game or farm animal? 95 Evin, A., Cucchi, T., Cardini, A., Strand Vidarsdottir, U., Larson, G. and Dobney, K., 2013. The long and winding road: Identifying pig domestication through molar size and shape. Journal of Archaeological Science , 40, 735–743. Evin, A., Dobney, K., Schafberg, R., Owen, J., Strand Vidarsdottir, U., Larson, G. and Cuc- chi, T., 2015a. Phenotype and animal domestication: A study of dental variation between domestic, wild, captive, hybrid and insular Sus scrofa . BMC Evolutionary Biology , 15 (6). doi:10.1186/s12862-014-0269-x. Evin, A., Girdland Flink, L., Bălăşescu, A., Popovici, D., Andreescu, R., Bailey, D., Mirea, P., Lazăr, C., Boroneanţ, A., Bonsall, C., Strand Vidarsdottir, U., Bréhard, S., Tresset, A., Cucchi, T., Larson, G. and Dobney, K., 2015b. Unravelling the complexity of domestication: A case study using morphometrics and ancient DNA analyses of archaeological pigs from Romania. Philosophical Transactions of The Royal Society B, 370, 20130616. doi:10.1098/rstb.2013.0616. Flad, R.K., Yuan, J. and Shuicheng, L., 2007. Zooarcheological evidence for animal domestication in northwest China. In : D.B. Madsen, C. Fa-Hu and G. Xing, eds. Late Quaternary Climate Change and Human Adaptation in Arid China. Developments in Quaternary Sciences 9. Amsterdam: Elsevier, 167–203. Frémondeau, D., Horard-Herbin, M.-P., Ughetto-Monfrin, J. and Balasse, M., 2013. L’alimentation des troupeaux porcins et la production de viande à Levroux Les Arènes (Indre): une analyse isotopique. In : A. Colin and F. Verdin, eds. L’âge du Fer en Aqui- taine et sur ses marges. Mobilité des hommes, diffusion des idées, circulation des biens dans l’espace européen à l’âge du Fer. Actes du 35e colloque international de l’AFEAF (Bordeaux, 2–5 juin 2011). Bordeaux: Aquitania, 747–752. Frémondeau, D., Nuviala, P. and Duval, C., 2017. Pigs and cattle in Gaul: The role of Gal- lic societies in the evolution of husbandry practices. European Journal of Archaeology , 20, 494–509. Hadjikoumis, A., 2012. Traditional pig herding practices in southwest Iberia: Questions of scale and zooarchaeological implications. Journal of Anthropological Archaeology , 31, 353–364. Halstead, P. and Isaakidou, V., 2011. A pig fed by hand in worth two in the bush: Eth- noarchaeology of if husbandry in Greece and its archaeological implications. In : U. Albarella, ed. Ethnoarchaeology: The Present Past of Human: Animal Relationships . Oxford: Oxbow, 160–174. Hammond, C. and O’Connor, T., 2013. Pig diet in medieval York: Carbon and nitrogen stable isotopes. Archaeological and Anthropological Science , 5, 123–127. Larson, G., Albarella, U., Dobney, K., Rowley-Conwy, P., Schibler, J., Tresset, A., Vigne, J.-D., Edwards, C.J., Schlumbaum, A., Dinu, A., Balaçsescu, A., Dolman, G., Taglia- cozzo, A., Manaseryan, N., Miracle, P., Van Wijngaarden-Bakker, L., Masseti, M., Brad- ley, D.G. and Cooper, A., 2007a. Ancient DNA, pig domestication, and the spread of the Neolithic into Europe. Proceedings of the National Academy of Science , 104 (39), 15276–15281. Larson, G. and Burger, J., 2013. A population genetics view of animal domestication. Trends in Genetics , 29, 197–205. Larson, G., Cucchi, T., Fujita, M., Matisoo-Smith, E., Robins, J., Anderson, A., Rolet, B., Spriggs, M., Dolman, G., Kim, T.H., Randi, E., Doherty, D., Awe Due, R., Bolt, R., Djiubiantono, T., Griffin, B., Intoh, M., Keane, E., Kirch, P., Li, K.-T., Morwood, M., Pedrina, L.M., Piper, P.J., Rabett, R.J., Shooter, P., van den Bergh, G., West, E., Wickler, S., Yuan, J., Cooper, A. and Dobney, K., 2007b. Phylogeny and ancient DNA of Sus provides new insights into Neolithic expansion in Island Southeast Asia and Oceania. Proceeding of the National Academy of Science , 104, 4834–4839. 96 Marie Balasse et al. Larson, G. and Füller, D.Q., 2014. The evolution of animal domestication. Annual Review of Ecology, Evolution and Systematic , 45, 115–136. Lu, H., Zhang, J., Liu, K., Wu, N., Li, Y., Zhou, K., Ye, M. and Zhang, T., 2009. Earliest domestication of common millet (Panicum miliaceum) in East Asia extended to 10,000 years ago. Proceedings of the National Academy of Science , 106 (18), 7367–7372. Marshall, F.B., Dobney, K., Denham, T. and Capriles, J.M., 2014. Evaluating the roles of directed breeding and gene flow in animal domestication. Proceeding of the National Academy of Science , 111, 6153–6158. Meadow, R.H., 1989. Osteological evidence for the process of animal domestication. In : J. Clutton-Brock, ed. The Walking Larder: Patterns of Domestication, Pastoralism and Predation . London: Unwin Hyman, 80–90. Müldner, G. and Richards, M.P., 2007. Stable isotope evidence for 1500 years of human diet at the city of York, UK. American Journal of Physical Anthropology, 133, 682–697. O’Connor, T., 2000. Bones as evidence of meat production and distribution in York. In : E. White, ed. Feeding a City: York. The Provision of Food From Roman Times to the Begin- ning of the Twentieth Century . Devon: Prospect Books, 43–60. Ottoni, C., Girdland Flink, L., Evin, A., Geörg, C., De Cupere, B., Van Neer, W., Bar- tosiewicz, L., Linderholm, A., Barnett, R., Peters, J., Decorte, R., Waelkens, M., Van- derheyden, N., Ricaut, F.-X., Hoelzel, A.R., Mashkour, M., Fatemeh, M.K., Sheikhi, A., Seno, S., Daujat, J., Brock, F., Pinhasi, R., Hongo, H., Perez-Enciso, M., Rasmus- sen, M., Frantz, L., Megens, H.J., Crooijmans, R., Groenen, M., Arbuckle, B., Benecke, N., Strand Vidarsdottir, U., Burger, J., Cucchi, T., Dobney, K. and Larson, G., 2013. Pig domestication and human-mediated dispersal in western Eurasia revealed through ancient DNA and geometric morphometrics. Molecular Biology and Evolution , 30, 824–832. Pechenkina, E.A., Ambrose, S.H., Xiaolin, M. and Benfer, R.A. Jr., 2005. Reconstructing Northern Chinese subsistence practices by isotopic analysis. Journal of Archaeological Science , 32, 1176–1189. Silitoe, P., 2007. Pigs in the New Guinea Highlands: An ethnographic example. In : U. Albarella, K. Dobney, A. Ervynck and P. Rowley Conwy, eds. Pigs and Humans: 10,000 Years of Interaction . Oxford: Oxford University Press, 330–358. Vigne, J.-D., 2011. The origins of animal domestication and husbandry: A major change in the history of humanity and the biosphere. Comptes Rendus Biologies , 3334, 171–181. Vigne, J.-D., 2012. Domestication (origines de la). In : J.-P. Poulain, ed. Dictionnaire des cultures alimentaires . Paris: Presses Universitaires de France, 390–398. Vigne, J.-D., 2015. Early domestication and farming: What should we know or do for a better understanding? Anthropozoologica , 50, 123–150. Vigne, J.-D., Peters, J. and Helmer, D. (Eds.), 2005. The First Steps of Animal Domestica- tion: New Archaeological Approaches . Proceedings of the 9th ICAZ Conference, Dur- ham 2002. Oxford: Owbow Books. Vigne, J.-D., Zazzo, A., Saliège, J.-F., Poplin, F., Guilaine, J. and Simmons, A., 2009. Pre- Neolithic wild boar management and introduction to Cyprus more than 11,400 years ago. Proceedings of the National Academy of Science , 106, 16135–16138. Zeder, M.A., 2009. The Neolithic macro(r)evolution: Macroevolution theory and the study of culture change. Journal of Archaeological Research , 17, 1–63. Zeder, M.A., 2012. The domestication of animals. Journal of Anthropological Research , 68, 161–190. Zhang, X., Zhao, X. and Chen, L., 2011. Human diets of Yangshao Culture in the Central Plains. Chinese Archaeology , 11, 188–196. 5 Arable weeds as a case study in plant-human relationships beyond domestication

Amy Bogaard, Mohammed Ater and John G. Hodgson

Introduction Arable weeds are plants that invade habitats created by people for the cultivation of other species. Though they are not the target of human cultivation, their growth in arable habitats means that they, like crops, are under human selection. Genetic studies of weedy crop relatives have documented traits (shattering/dehiscence, asynchronous flowering, seed dormancy, etc.) that allow weeds to escape detection and eradication by farmers, and give them a competitive advantage over crops ( Thurber et al . 2010 , 2011 ; Qi et al . 2015 ). Selection of these traits in weedy crop relatives thus constitutes a (partial) reversal of the domestication syndrome (Hammer 1984), sometimes called ‘de-domestication’ (Ellstrand et al . 2010 ). But weeds can also adapt by taking on domestic traits (Hammer 1984); Harlan (1992: 66, 94) reports non-shattering populations of the weed Bromus secalinus and of weedy oats, for example. Moreover, genetic study of wild crop relatives and their domesticated counterparts has confirmed the importance of allele introgression from proximate wild populations (Song et al . 2014 ; Gutaker et al . 2017 ). The history of crops and weeds is thus deeply intertwined, the selection of traits in one shaping (directly and indirectly) the evolution of the other, in a particularly clear instance of mutual evolution through niche construction. Another aspect of this complex relationship is summarised by Vavilov’s distinction between ‘primary’ and ‘secondary’ domesticates, the latter thought to originate as arable weeds of the former ( Vavilov 1992 : 20). Hammer (1984 ) links secondary domesticates with the unconscious and gradual domestication of seed crops, in contrast to the conscious selection of traits in garden crops cultivated for their vegetative parts. Human selection in seed crop cultivation tends to operate at the level of the stand/field rather than the individual plant, thus encompassing both crops and weeds and bringing both under potential selection for domestic traits. Widely cited examples of secondary domesticates are rye (Secale cereale L.) and oat (Avena sativa L.). It is now known that both wild rye and oat were likely under cultivation by the 10th millennium BC (Pre-Pottery Neolithic A) in the middle Euphrates and Jordan Valley regions of Southwest Asia, respectively ( Weiss et al . 2006 ; Willcox et al. 2008, 2009), alongside the wild progenitors of other 98 Amy Bogaard, Mohammed Ater, John G. Hodgson cereals (barley, einkorn and emmer wheat). Unlike the latter, however, neither rye nor oat was part of the early domesticated package of crops that emerged in Southwest Asia, though domestic (non-shattering) rye was reported from the established Neolithic site of Can Hasan III, central Anatolia (Hillman 1978). Spo- radic archaeobotanical finds of rye in Southwest Asia and Europe suggest that it mostly contaminated other cereal crops before emerging as a staple in temperate Europe from the Iron Age, expanding significantly during the early mediaeval period (Behre 1992; Zohary et al. 2012: 62–66). Oat possesses remarkable phenotypic plasticity as a weed, and can be a tolerated weedy supplement of other cereals ( Harlan 1992 : 66, 94). Archaeobotanical evidence suggests that it infested other cereals for millennia before emerging as a domesticated crop in central and western Europe, from the Roman period onward ( Zohary et al . 2012 : 69). Weedy origins have also been suggested for pulse crops such as common vetch (Vicia sativa L.) (Zohary et al. 2012: 95) and for other annual seed crops including opium poppy ( Hammer 1984 ). It is thus apparent that weeds are under genetic selection for invasiveness, and that their successful persistence in arable fields simultaneously brings them to the attention of farmers, who may choose to cultivate and (inadvertently) domesticate them. The weedy ecological niche thus both impinges negatively on crop growth, and offers an opportunity for human exploitation. The practices of traditional subsistence farmers demonstrate a keen awareness of this paradox; though they may well regard weeds as harmful competitors, they are often knowledgeable of their potential uses, and open to opportunities to make use of the weeds they encounter in arable fields, in accordance with their needs. In this chapter, we seek to disentangle crop husbandry and weed use further by distinguishing relevant weed ecological strategies and identifying different forms of human exploitation shaped by different agrosystems. We then use case studies to illustrate the potential of archaeobotanical remains of prehistoric weed flora associated with crops for reconstructing complex relationships among people, crops and weeds in the deep past. Our present-day and archaeological focus is Western Asia, North Africa and Europe, and we have furthermore targeted annual cereal and pulse (seed) crops and their weeds.

Ethnobotanical and ecological perspectives on arable weeds There is ample ethnobotanical support for the claim that arable weeds can form a valued part of the subsistence economy, including the study in northern Cameroon by Garine et al. (this volume). Palmer (1998 ) showed that hand-weeding in Jordan is associated with provision of fodder in small-scale herding, for example, while Ertuğ (2000 : Table 1) has recorded the use of dozens of invasive species in central Anatolia, mainly as fodder and/or food. Here we distinguish two broad categories of weed use: 1. intentional collection of species with distinctively different growth habits and/or seasonality to the crop; and 2. collection (intentional or not) of species that closely mimic the crop in morphology and seasonality, making it difficult to avoid harvesting Arable weeds as a case study 99 both. For simplicity, we refer to the first group as ‘opportunists’ and the second group as ‘mimics’, though we recognise that there is a spectrum and thus overlap between these strategies. While both categories include noxious weeds that are inedible and even harmful to human or livestock consumers, both include species that can make a significant contribution to subsistence, as explored further later. The ‘opportunist’ category is not only readily distinguished from the crop in its morphology, but may also set seed well before and/or after crop harvest time, relying on a persistent bank of seeds in the soil from which to regenerate, and/or on vegetative regeneration from fragments of root/rhizome/stolon. Weeds of the ‘mimic’ category, by contrast, are difficult to distinguish visually from the crop, often set seed around harvest time, hold their seeds at a similar height to the crop and possess seed morphology that imitates the size/shape of crop seed, enhanc- ing their likelihood of being re-sown with the seed corn. These species may also have some seed dormancy, allowing them to persist in the soil to an extent, though mimics of large-seeded crops are less likely to form large persistent seed banks in the soil than small-seeded species (Thompson et al . 1993 ; Saatkamp et al . 2009 , 2011 ). The ‘mimic’ category includes weedy crop relatives such as the wild millets discussed by Garine et al. (this volume). In cereal and pulse fields, the distinction between opportunists and mimics parallels that between the broad phytosociological classes of Chenopodietea and Secalietea ( Ellenberg et al . 1992 : 72): root-/row-crop weeds that may grow opportunistically between row-sown or dibbled crops, versus mimicking weeds that thrive in dense crop stands. In a general sense, the Chenopodietea and Seca- lietea characterise and help to discriminate between crop fields that are managed with relatively high and low labour intensity, respectively ( Jones et al . 1999 ). To summarise thus far, ‘opportunist’ weeds include those that exploit fertile and highly disturbed habitats, while ‘mimic’ weeds follow the strategy of the crop as closely as possible. Cereals and pulses tend to be relatively competitive (tall/wide, fast growing), and to varying degrees demand fertile conditions, but require a certain period without mechanical soil disturbance in order to reach maturity. Thus, crop mimics will tend to be less disturbance-tolerant than opportunists, at least under highly fertile conditions. We would therefore expect ‘opportunist’ weeds to be more closely associated with labour-intensive maintenance of highly disturbed conditions (i.e. thoroughly tilled and hand-weeded), whereas crop mimics would be more prominent in less disturbed conditions (less thorough tillage and limited/ no hand-weeding). These broadly distinguished weed strategies create different opportunities for farmers. Opportunist weeds offer a potential source of human food and/or animal fodder over an extended time period, or at least for a period offset from that of the crop harvest, whereas mimics may usefully stretch/bulk out the main crop harvest itself (e.g. Garine et al., this volume). Crop mimics, however, compete most directly with the crop, and pose an additional threat if they are noxious species that are difficult to separate from the crop and eradicate (e.g. Agrostemma githago ; Lolium temulentum infected with ergot fungus – Freeman 1904 ). 100 Amy Bogaard, Mohammed Ater, John G. Hodgson Hand-weeding in labour-intensive agrosystems in Morocco, including study sites in a recent weed ecological survey involving the present authors, offers a relevant case study ( Bogaard et al . 2016a ). Weeds in these systems tended to be viewed by farmers as a resource rather than merely as plants lowering crop yield, and were often used as fodder or food. We use functional ecological traits as a partial explanation of the evident value of weeds in this kind of system. Our observations suggest that hand-weeding is motivated not only by the removal of plants that compete with the crop, but also by the palatability of weed species that grow in highly fertile, disturbed and often shaded conditions. Studies of herbivory and defensive characteristics (including chemical defense) in a range of plant species, from trees to ruderals, suggest that palatability is negatively correlated with growth rate and leaf life-span ( Coley 1983 ; Southwood et al . 1986 ; Grime et al. 1996; Wright and Westoby 2002). Leaf traits have been shown to conform to a broad spectrum of variation irrespective of life form ( Wright et al . 2004 ). We hypothesise that growth rate in arable weeds, as gauged through traits such as specific leaf area (SLA, the ratio of leaf area to leaf dry weight), is positively correlated with palatability, such that the weeds that thrive in very fertile and disturbed conditions also tend to be the most attractive for human and/ or animal consumption. Intensive hand-weeding of annual crops that form the basal layer of the oasis polyculture system is mentioned by various authors (e.g. Despois 1964 ; Baudoux et al. 2004; Ministry of Agriculture and Maritime Fishing 2009). As in the small Amtoudi oasis (Guelmim province, Morocco), where we conducted quadrat weed- survey of 16 cereal fields (six-row hulled barley, Hordeum vulgare L.) ( Bogaard et al. 2016a), this is a task carried out mainly by women, who are therefore the experts in distinguishing useful and noxious species. We also surveyed a small number of cereal and pulse fields in the coastal part of Chefchaouen, northern Morocco, where small plots were also very intensively hand-weeded. A study of wild plant usage in the Rif region, including the prominent role played by arable weeds, is in progress ( Bensbih and Ater 2017 ). Despite clear climatic and biogeographical differences in the vegetation of northern and southern Morocco, the species composition of arable weed floras of intensively managed plots in both regions is very similar ( Bogaard et al . 2016a : Fig. 2a). Labour-intensive management of these plots entailed manuring and (where needed) irrigation, as well as thorough tillage and hand-weeding ( Bogaard et al. 2016a). Particularly characteristic of these plots were species such as Stel- laria media (L.) Vill., Euphorbia peplus L. and Fumaria officinalis L., all with relatively thin leaves and high SLA values. A combination of thin leaves and high SLA values in annual plants predicts rapid growth through production of leaves with low construction costs, and an ability to make efficient use of phytosynthate in shaded conditions (Hodgson et al. 2011). In a UK study, Stellaria media was shown to score particularly highly for herbivore palatability, coinciding with a short leaf life expectancy of around two weeks (Southwood et al . 1986 ). These are species of modest height and typically with smaller seeds than the crop; they fit within an opportunist rather than crop mimic category. The weed flora in these Arable weeds as a case study 101 fields also contained other species with lower SLA values, including several relatively large-seeded grasses (Avena sterilis L. , Bromus madritensis L. , B. rubens L. , Phalaris minor Retz.,) and other large-seeded species that tend to infest seed corn (e.g. Convolvulus arvensis L.), but classic crop mimics did not dominate the weed flora. Phalaris minor is considered noxious to livestock in Morocco (Lam- naouer and Abdennebi 1994 ). Weeds in southern Moroccan oases are used particularly as animal fodder, in conjunction with small-scale and intensive livestock husbandry (e.g. house goats) (personal observations; cf. Palmer 1998). In the Rif region, by contrast, herding is large-scale and extensive, and an ongoing study by Bensbih and Ater (2017 ) reveals that arable weeds instead have a particular role as cooked greens in the human diet. The taxa recorded in the latter study include Papaver rhoeas L., a species of moderately high SLA also documented in intensively managed crop fields in coastal Chefchaouen (Bogaard et al. 2016a). Other species collected for human consumption in the Rif region and recorded as arable weeds in intensively managed fields include Malva spp. All of these taxa are collected (and sold) as ‘bouquets’ and also consumed as mixtures of species (Bensbih and Ater 2017) ( Figure 5.1 ).

Figure 5.1 Display of gathered plants including weedy Malva spp., market, Jeblia, Rif region, Morocco. Photo credit: Mohammed Ater & Hasnae Bensbih. 102 Amy Bogaard, Mohammed Ater, John G. Hodgson Traditional Moroccan agrosystems also harbour cases of wild crop relatives that are considered useful plants in their own right. Vicia sativa subspecies nigra (L.) Ehrh. is an example (Figure 5.2). This wild vetch grows in oasis barley fields as an opportunist, and displays distinctively ‘wild’ characteristics (asynchronous flowering, shattering) relative to cultivated forms of common vetch. It has been observed collected in bunches and drying on roofs as a fodder plant in the oasis of Imin-o-Iaouane, a relatively high-altitude oasis (with almond orchards rather than date palms) on the southern slopes of the High Atlas, in the hinterland of Skoura (Ater, personal observations). Elsewhere in Morocco, and in the wider Mediterranean, it is possible to observe more extensive traditional production systems with fewer labour inputs per unit area, including little to no hand-weeding (Bogaard et al . 2016a , 2016b). These

Figure 5.2 Vicia sativa subsp. nigra , growing as a weed of cereals in the oasis of Imin-o- Iaouane on the southern slopes of the High Atlas, Morocco. Photo credit: Mohammed Ater. Arable weeds as a case study 103 weed floras variously reflect less fertile, disturbed and shaded conditions, and are characterised by weed species of varying palatability for livestock and/or human consumers. In the Sault region and adjacent areas of Haute Provence, a quadrat-weed survey of 60 fields has recently documented the weed flora associated with extensive organic cultivation of einkorn and other protected cereal landraces (Bogaard et al. 2016b). This weed flora is characterised by classic cereal mimics such as Agrostemma githago, the seeds of which are moderately toxic, and must be carefully removed to avoid spoiling flour and causing ‘githagism’ in consumers (www.toxiplante.fr/monographies/nielle.html). Other cereal mimics in this weed flora, such as Avena sterilis, have edible seeds and can also be used as fodder. A range of opportunistic weeds also occur in this weed flora. An ongoing collaborative project involving two of the present authors (Bogaard, Hodgson) and Yildiz Aumeeruddy-Thomas aims to clarify/communicate the potential uses (as human food, medicine or animal fodder) of the weed species in these organic cereal fields. Finally, a recent study of the arable weed flora of décrue (flood recession) fields in the Guelmim province of southern Morocco offers a case study in weed use in the context of low-input cultivation (Ater and colleagues, personal observations). Here, women practice the oasis-based tradition of weeding cereal fields for livestock, but must contend with a high proportion of non-palatable taxa (e.g. sclerophyllous and/or spiny/thorny) (Figure 5.3). Their strategy is to weed fields

Figure 5.3 Harvested décrue barley field, showing spiny weeds (Echinops spinosus ), Guelmim province, Morocco. Photo credit: Mohammed Ater and Mohamed El Mahroussi. 104 Amy Bogaard, Mohammed Ater, John G. Hodgson and then separate the weedy harvest into two groups: fodder (for their sheep, goats and cattle) and non-fodder. Spiny/thorny taxa are discarded on the margins of fields, and may be consumed by the camels of passing nomads. The farmers prefer hand-weeding to mechanised weed collection so that they can avoid incorporating spiny taxa in fodder.

Archaeobotanical evidence for the use of arable weeds in intensive and extensive agrosystems Three case studies serve to illustrate the potential of archaeobotanical remains of ancient arable weed flora for understanding complex ecological relationships among people, crops and weeds in the deep past, extending ethnobotanical insights from present-day studies. Here we briefly sketch the context and evidence available for agroecology and weed use at Neolithic Çatalhöyük, south-central Turkey (late 8th to late 7th millennium BC ), the Neolithic Linearbandkeramik (Linear Pottery Culture) complex of central and western Europe (later 6th millennium BC ) and the northern Mesopotamian Late Chalcolithic-Early Bronze Age (4th–3rd millennia BC ). While the two Neolithic case studies shed light on intensive use of ‘opportunist’ weeds in labour-intensive agrosystems, the northern Mesopotamian case presents a contrasting scenario of high infestation of crop stores by (edible) weedy crop mimics that evidently made a significant contribution to the economy. A notable constraint of the charred (carbonised) archaeobotanical evidence we consider here is that we are limited in our statements about past weed use to the collection/processing/storage of dense plant parts, especially seeds. Ethnobotani- cally, however, we know that farmers often make use of multiple plant parts, and some of the taxa considered below have edible vegetative parts as well as seeds.

Çatalhöyük The archaeobotanical assemblage from the 1,000-year Neolithic sequence at Çat- alhöyük (c. 7100–5950 BC – Bayliss et al . 2015 ; Marciniak et al. 2015) attests to the collection and storage of a diverse range of wild species, including likely arable weeds (Fairbairn et al . 2007 ; Bogaard et al. 2013a, 2017 ). Potential weed taxa found in ‘storage’ concentrations include members of the Cruciferae/Brassicaceae family (especially Descurainia sophia [L.] Webb ex Prantl) and several grasses ( Taeniatherum caput-medusae Nevski, Eremopyrum sp. [Ledeb.] Jaub. & Spach). None of these species are close crop mimics. We focus here on the wild crucifer Descurainia sophia since it is attested very consistently throughout the sequence, and in great abundance in some ‘storage’ concentrations (e.g. 30 litres of cleaned seed in the storage bin of one burned building, Building 52) (Bogaard et al . 2013a , 2017 ). Wild mustard was likely valued for its oil-rich seeds; linseed/flax is very rare in the assemblage, and wild mustard offered a source of plant oil, alongside almonds and other nuts. A seed of this taxon was also recently observed embedded in a fragment of charred ‘bread’ ( Gonzalez Carretero et al . 2017 ), adding support to the suggestion that it was a condiment alongside staple cereals and pulses Arable weeds as a case study 105 ( Bogaard et al . 2013a : 128). The tiny seeds of this taxon are much smaller than those of crops at Çatalhöyük; while is it thus not associated with relatively ‘clean’ (processed grain/seed) crops in archaeological storage contexts, its ubiquity and association in some deposits with crop processing by-products, together with its invasiveness in arable and ruderal habitats in the Konya plain today (Bogaard, personal observations), suggest that it was also an arable weed. Descurainia sophia thus appears to offer a clear example of an invasive weedy species that was valued in its own right as a source of human nutrition. Similar examples of the collection of weedy members of the Cruciferae/ Brassicaceae family as human food are also known elsewhere in western Asia (e.g. 10th millennium BC Jerf el Ahmar – Willcox 2002 ) and Europe ( Schlichthe- rle 1981 ). A particularly clear example including the same species (Descurainia sophia) is offered by a burned village-wide destruction layer at late Neolithic (early 4th millennium BC ) Hornstaad-Hörnle IA, on the southwestern shore of Lake Constance/Bodensee. Here charred clusters of the oil-rich seeds of D. sophia , other Cruciferae and some Labiatae/Lamiaceae (Galeopsis spp.) attest to deliberate/ separate storage, while the dispersal of low levels of these same taxa through stored cereals shows that they were invasive in crop fields. As at Çatalhöyük, a case can be made that these species may have been the target of cultivation (Maier 2011), but their archaeological ubiquity and ecological invasiveness suggest that it would be reasonable to assume that they were (also) weeds of other crops. The absence of domesticated cattle through much of the sequence at Çatal- höyük, and their low-level introduction in later phases, indicates that the agrosystem could only be based on manual cultivation. Detailed ecological work on the agroecology of Çatalhöyük is still underway, but the available evidence suggests that productive conditions were probably maintained to varying degrees by thorough tillage, hand-weeding and manuring (with sheep dung) and/or middening, a key constraint on the labour-intensity of management being the varying distance of fields across the surrounding alluvial plain from the densely nucleated settlement (including penning areas) ( Bogaard et al . 2013a ). It is apparent, therefore, that Çatalhöyük represents an established early farming community that practised relatively intensive arable land management and took advantage of the nutritional opportunities offered by certain ‘opportunist’ arable weed species. Occa- sional finds of the seeds of taxa that persist as arable weeds in cereal stubble after the harvest (Chenopodium L., Polygonum aviculare L.) in the matrix of charred sheep/goat dung pellets at Çatalhöyük suggests that certain weeds also played a role in the livestock diet ( Bogaard et al . 2013a : 103).

Linearbandkeramik (LBK – Linear Pottery Culture) complex It has long been appreciated by archaeobotanists working on assemblages from this widespread Neolithic horizon in lowland Eastern-Central-Western Europe that for centuries LBK farmers made use of certain species that were almost certainly also arable weeds. The clearest case concerns fat hen/goosefoot (Che- nopodium album L.), discussed extensively by various authors ( Knörzer 1967 , 106 Amy Bogaard, Mohammed Ater, John G. Hodgson 1973 , 1988 , 1997 ; Bakels 1979 , 1983/4 , 1991 ; Willerding 1980 ; Lüning 2000 : 92; Bogaard 2004 , 2011 ; Kreuz 2007 ). The potential productivity and nutritional properties of C. album, a small-seeded relative of quinoa, have been discussed by Stokes and Rowley-Conwy (2002 ). As a potential cereal weed, it is an opportunist species of relatively fertile/disturbed habitats, and its association with glume/ hulled (einkorn and emmer) processing by-products on many LBK sites confirms that it was a regular constituent of cereal fields. The use of its seeds as a likely human food is directly suggested by the finding of ‘pure’ concentrations in pit fills on a range of sites, including the earliest LBK site of Niederhummel, Bavaria (Bogaard unpublished), as well as numerous later LBK settlements. Detailed agroecological characterisation of the LBK Neolithic through weed ecological analysis and more recently stable isotope analysis of associated crops ( Bogaard 2004 , 2011 ; Bogaard et al . 2013b ) indicates that arable fields were managed to varying degrees with labour-intensive inputs, including likely hand- weeding and manuring/middening; the traditional ‘slash and burn’ model of Neo- lithic European farming simply does not fit the evidence. As at Çatalhöyük, and indeed some later Neolithic complexes in central Europe (lakeshore sites in the Alpine foreland – Jacomet et al . 2016 ), the LBK complex presents compelling evidence for a relatively small-scale intensive agrosystem that included use of species alongside crops that were likely arable weeds. Moreover, the key species appear to be opportunists that did not closely mimic the crop and could be readily separated/collected/stored in their own right.

Northern Mesopotamia, Late Chalcolithic-Early Bronze Age The third case study presents a deliberate contrast to the others. The dry-farming belt of northern Mesopotamia was a cradle of early urbanism, parallel but agro- ecologically distinct from that which developed in the irrigated southern alluvium ( Oates et al . 2007 ; Ur 2010 ). While early towns like Tell Zeidan on the Euphrates reached 12 ha by the 5th millennium BC , by the later 4th millennium BC (Late Chalcolithic 3), sites like Tell Brak exceeded 100 ha. This remarkable sequence of early urbanisation, involving radically larger populations living in what appear to have been densely occupied cities, including temples and other institutional build- ings, entailed equally dramatic agroecological change. A recent study by Styring et al . (2017 ) has shown that this agricultural process is best understood as extensification: the expansion of low-input management of cereals like two-row hulled barley and emmer wheat outwards from the urban core, facilitated by animal traction, and inter alia accelerating land-based inequality ( Bogaard et al . 2018 ). Large-scale storage of cereals is particularly well evidenced at EBA Tell Brak (TC complex – Emberling and MacDonald 2001 , 2003 ; Hald and Charles 2008 ) and Tell Leilan (Acropolis storage complex – Weiss et al. 2002). Both of these complexes feature storage not only of domesticated cereals (e.g. hulled barley, emmer wheat), but also very abundant seeds of a cereal-mimicking weed, Aegil- ops spp. This genus is closely related to wheats and barleys, falling within the same tribe (Triticeae) of the grass family. A. tauschii Coss. was the donor of the Arable weeds as a case study 107 D chromosomes in hexaploid bread wheat (Triticum aestivum L.), which arose through alloploidisation of A. tauschii with domesticated emmer (T. dicoccum Schrank) ( Kilian et al . 2010 ). The large seeds of Aegilops closely mimic wheats and barleys, as do the plants in the field in seasonality and growth habit. Though we cannot be sure that Aegilops was an explicitly valued contribution to the economy, its abundance in stores and edibility indicate that, in effect, it was a significant component of the cereal harvest. The large scale of the TC complex at Brak, and apparently institutional nature of the Acropolis store at Leilan, suggest that it was above all extensive cultivation aimed at fulfilling institutional/ taxation requirements, and forming an important basis of elite land-based power, that encouraged expansive of low-input cultivation favouring crop mimics such as Aegilops .

Discussion and conclusions Ecologically, the use/consumption (deliberate or not) of an edible crop mimic such as Aegilops can be contrasted with that of weedy species in the Neolithic cases reviewed in this chapter. While Aegilops broadened the subsistence base to an extent, it also competed directly with the desired crop, and would be subject to similar hazards of rainfall/weather since the ecological niches of crop and weed mimic would have been very similar. There is a hint from stable carbon isotope analysis of cereal grains and associated Aegilops grains in the Tell Leilan Acropo- lis store ( Styring et al . 2017 : Fig. 5a) that Aegilops completed photosynthesis even earlier than barley, suggesting that there were physiological differences that favoured earlier ripening of weed than crop. In the cases of Çatalhöyük and the LBK complex, however, the exploited ‘opportunist’ weeds contrasted much more clearly with their associated crops, in ecology and nutritional value. The agrobiodiversity of these small-scale intensive systems, therefore, was significantly enhanced through exploitation of the weed flora, and more so than in the (inadvertent?) use of crop mimics in the extensive production systems of Early Bronze Age northern Mesopotamia. More- over, had the crop mimic been a toxic species such as Agrostemma githago , the consequences for the value of the crop itself could be devastating. Potentially catastrophic effects of toxic crop mimics on crop stores are indeed known from Roman and mediaeval contexts, for example (e.g. Wiethold 1996 ). In conclusion, it is clear that an appreciation of the potential uses of weeds – borne largely of necessity in subsistence farmer contexts – has been a valuable component of the economy for thousands of years. Consideration of weed ecology and the nature of the wider agrosystem upholds our hypothesis that labour- intensive subsistence systems tend to foster more valuable and diversified weed use than low-input extensive systems, such as those encouraged by the formation of dense urban centres with wide agricultural hinterlands and marked social differentiation. The resilience of long-lived Neolithic communities such as that at Neolithic-early Chalcolithic Çatalhöyük, which persisted in total from the late 8th millennium to the mid-6th millennium BC , relied not only on established staple 108 Amy Bogaard, Mohammed Ater, John G. Hodgson crops but also systematic use of species that were invasive in arable habitats and presented new opportunities for diversification of the subsistence base ( Bogaard et al . 2017 ). A simplistic ‘domestic/wild’ dichotomy fails to capture the subtlety and mechanics of crop/weed communities and their exploitation by humans.

References Bakels, C.C. 1979. Linearbandkeramische Früchte und Samen aus den Niederlanden. Archaeo-Physika 8, 1–10. Bakels, C.C. 1983/4. Pflanzenreste aus Niederbayern – Beobachtungen in rezenten Aus- grabungen. Bericht der Bayerischen Bodendenkmalpflege 24/25, 157–166. Bakels, C.C. 1991. Tracing crop processing in the Bandkeramik culture. In J. Renfrew (ed.) New Light on Early Farming , pp. 281–288. Edinburgh, Edinburgh University Press. Baudoux, L., Nieullet, E., Druguet, S., Jouve, P., Seuge, C., Ouabou, R., Mouradi, H. and Loussert, R. 2004. Femmes et développement oasien, situation actuelle, initiatives et perspectives. Etudes de cas dans la région de Tata. Etudes thématiques en vue du dével- oppement des oasis de la région de Tata (Maroc) effectuées par des étudiants du CNE- ARC, Etude No. 2 . Montpellier, CNERAC. Bayliss, A., Brock, F., Farid, S., Hodder, I., Southon, J. and Taylor, R.E. 2015. Getting to the bottom of it all: A Bayesian approach to dating the start of Çatalhöyük. Journal of World Prehistory 28, 1–26. Behre, K.-E. 1992. The history of rye cultivation in Europe. Vegetation History and Archaeobotany 1, 141–156. Bensbih H. and Ater M. 2017. Gathered Food Plants in Northern Morocco . Unpublished Poster. Bogaard, A. 2004. Neolithic Farming in Central Europe . London, Routledge. Bogaard, A. 2011. Plant Use and Crop Husbandry in an Early Neolithic Village: Vai- hingen an der Enz, Baden-Württemberg. Frankfurter Archäologische Schriften. Bonn, Habelt-Verlag. Bogaard, A. The Archaeobotany of LBK Niederhummel, Bavaria . Unpublished report. Bogaard, A., Charles, M., Livarda, A., Ergun, M., Filipović, D. and Jones, G. 2013a. The archaeobotany of mid-later Neolithic occupation levels at Çatalhöyük. In I. Hodder (ed.) Humans and Landscapes of Çatalhöyük: Reports from the 2000–2008 Seasons , pp. 93–128. Los Angeles, CA, Monographs of the Cotsen Institute of Archaeology, Univer- sity of California at Los Angeles. Bogaard, A., Filipović, D., Fairbairn, A., Green, L., Stroud, E., Fuller, D. and Charles, M. 2017. Agricultural innovation and resilience in a long-lived early farming community: The 1500-year sequence at Neolithic-early Chalcolithic Çatalhöyük, central Anatolia. Anatolian Studies , 67, 1–28. Bogaard, A., Fraser, R.A., Heaton, T.H.E., Wallace, M., Vaiglova, P., Charles, M., Jones, G., Evershed, R.P., Styring, A.K., Andersen, N.H., Arbogast, R.-M., Bartosiewicz, L., Gardeisen, A., Kanstrup, M., Maier, U., Marinova, E., Ninov, L., Schäfer, M. and Stephan, E. 2013b. Crop manuring and intensive land management by Europe’s first farmers. Proceedings of the National Academy of Sciences 110, 12589–12594. Bogaard, A., Styring, A., Ater, M., Hmimsa, Y., Green, L., Stroud, E., Whitlam, J., Diffey, C., Nitsch, E., Charles, M., Jones, G. and Hodgson, J. 2016a. From traditional farming in Morocco to early urban agroecology in northern Mesopotamia: Combining present-day Arable weeds as a case study 109 arable weed surveys and crop isotope analysis to reconstruct past agrosystems in semi- arid regions. Environmental Archaeology . doi:10.1080/14614103.2016.1261217. Bogaard, A., Hodgson, J., Nitsch, E., Jones, G., Styring, A., Diffey, C., Pouncett, J., Her- big, C., Charles, M., Ertuğ, F., Tugay, O., Filipović, D. and Fraser, R. 2016b. Combining functional weed ecology and crop stable isotope ratios to identify cultivation intensity: a comparison of cereal production regimes in Haute Provence, France and Asturias, Spain. Vegetation History and Archaeobotany 25: 57–73. Bogaard, A., Styring, A., Whitlam, J., Fochesato, M. and Bowles, S. 2018. Farming, inequality and urbanization: A comparative analysis of late prehistoric northern Meso- potamia and south-west Germany. In T.A. Kohler and M.E. Smith (ed.) Ten Thousand Years of Inequality: The Archaeology of Wealth Differences , pp. 201–229. Tucson, AZ, University of Arizona Press, Amerind Seminar Series. Coley, P.D. 1983. Herbivory and defensive characteristics of tree species in a lowland tropical forest. Ecological Monographs 53, 209–234. Despois, J. 1964. Les paysages agraires traditionnels du Maghreb et du Sahara septentri- onal. Annales de Géographie 73, 129–171. Ellenberg, H., Weber, H.E., Düll, R., Wirth, V., Werner, W. and Paulissen, D. 1992. Zeiger- werte von Pflanzen in Mitteleuropa. Scripta Geobotanica 18. Ellstrand, N.C., Heredia, S.M., Leak-Garcia, J.A., Heraty, J.M., Burger, J.C., Yao, L., Nohzadeh-Malakshah, S. and Ridley, C.E. 2010. Crops gone wild: Evolution of weeds and invasives from domesticated ancestors. Evolutionary Applications 3, 494–504. Emberling, G. and MacDonald, H. 2001. Excavations at Tell Brak 2000: Preliminary report. Iraq 63, 21–54. Emberling, G. and MacDonald, H. 2003. Excavations at Tell Brak 2001–2002: Preliminary report. Iraq 65, 1–75. Ertuğ, F. 2000. An ethnobotanical study in central Anatolia (Turkey). Economic Botany 54, 155–182. Fairbairn, A.S., Martinoli, D., Butler, A. and Hillman, G. 2007. Wild plant seed storage at Neolithic Çatalhöyük East, Turkey. Vegetation History and Archaeobotany 16, 467–479. Freeman, E.M. 1904. The seed-fungus of Lolium temulentum, L., the Darnel. Philosophi- cal Transactions of the Royal Society of London , B 196, 1–27. Gonzalez Carretero, L., Wollstonecroft, M. and Fuller, D.Q. 2017. A methodological approach to the study of archaeological cereal meals: A case study at Çatalhöyük East (Turkey). Vegetation History and Archaeobotany . Grime, J.P., Cornelissen, J.H.C., Thompson, K. and Hodgson, J.G. 1996. Evidence of a causal connection between anti-herbivore defence and the decomposition rate of leaves. Oikos 77, 489–494. Gutaker, R.M., Zaidem, M., Fu, Y.-B., Diederichsen, A., Smith, O., Ware, R. and Allaby, R.G. 2017. Adaptation to European latitudes through assimilation of wild diversity at the LuTFL1 locus altered architecture and promoted fiber production in flax. bioRxiv . doi: 10.1101/178772. Hald, M.M. and Charles, M. 2008. Storage of crops during the fourth and third millennia BC at the settlement mound of Tell Brak, north-east Syria. Vegetation History and Archaeo- botany 17(Suppl 1), 35–41. Hammer, K. 1984. Das Domestikationssyndrom. Kulturpflanze 32, 11–34. Harlan, J.R. 1992. Crops and Man . Second ed. Madison, WI, American Society of Agronomy. Hillman, G. 1978. On the origins of domestic rye – Secale cereale: The finds from Ace- ramic can Hasan III in Turkey. Anatolian Studies 28, 157–174. 110 Amy Bogaard, Mohammed Ater, John G. Hodgson Hodgson, J.G., Montserrat-Martí, G., Charles, M., Jones, G., Wilson, P., Shipley, B., Sharafi, M., Cerabolini, B.E.L., Cornelissen, J.H.C., Band, S.R., Bogaard, A., Castro- Díez, P., Guerrero-Campo, J., Palmer, C., Pérez-Rontomé, M.C., Carter, G., Hynd, A., Romo-Díez, A., de Torres Espuny, L. and Royo Pla, F. 2011. Is leaf dry matter content a better predictor of soil fertility than specific leaf area? Annals of Botany 108, 1337–1345. Jacomet, S., Ebersbach, R., Akeret, Ö., Antolín, F., Baum, T., Bogaard, A., Brombacher, C., Bleicher, N.K., Heitz-Weniger, A., Hüster-Plogmann, H., Gross, E., Kühn, M., Rent- zel, P., Steiner, B.L., Wick, L. and Schibler, J.M. 2016. On-site data cast doubts on the hypothesis of shifting cultivation in the late Neolithic (c. 4300–2400 cal BC ): Landscape management as an alternative paradigm. The Holocene 26, 1858–1874. Jones, G., Bogaard, A., Halstead, P., Charles, M. and Smith, H. 1999. Identifying the intensity of crop husbandry practices on the basis of weed floras. Annual of the British School at Athens 94, 167–189. Kilian, B., Mammen, K., Millet, E., Sharma, R., Graner, A., Salamini, F., Hammer, K. and Özkan, H. 2010. Aegilops. In C. Kole (ed.) Wild Crop Relatives: Geonomic and Breed- ing Resources, Cereals , pp. 1–76. Heidelberg, Springer-Verlag. Knörzer, K.-H. 1967. Subfossile Pflanzenreste von bandkeramischen Fundstellen im Rhe- inland. Archaeo-Physika 2, 3–29. Knörzer, K.-H. 1973. Pflanzliche Großreste. Rheinische Ausgrabungen 13, 139–152. Knörzer, K.-H. 1988. Untersuchungen der Früchte und Samen. In U. Boelicke, D. von Brandt, J. Lüning, P. Stehli and A. Zimmerman (ed.) Der bandkeramische Siedlung- splatz Langweiler 8, Gemeinde Aldenhoven, Kreis Düren , pp. 813–852. Rheinische Aus- grabungen 28. Cologne, Rheinland-Verlag GmbH. Knörzer, K.-H. 1997. Botanische Untersuchung von 16 neolithischen Siedlungsplätzen im Bereich der Aldenhovener Platte, Kr. Düren und Aachen. In J. Lüning (ed.) Studien zur neolithischen Besiedlung der Aldenhovener Platte und ihrer Umgebung, pp. 647–684. Cologne, Rheinland-Verlag GmbH. Kreuz, A. 2007. Archaeobotanical perspectives on the beginning of agriculture north of the Alps. In S. Colledge and J. Conolly (ed.) The Origins and Spread of Domestic Plant in Southwest Asia and Europe , pp. 259–294. Walnut Creek, CA, Left Coast Press. Lamnaouer, D . and Abdennebi, E.H. 1994. Intoxications d’origine végétale chez les ovins du Maroc. In B. Boulanouar and R. Paquay (ed.) L’élevage des moutons et ses systèmes de production au Maroc , pp. 335–352. Rabat, Nadacom. Lüning, J. 2000. Steinzeitliche Bauern in Deutschland – die Landwirtschaft im Neolithi- kum . Universitätsforschungen zur prähistorischen Archäologie aus dem Seminar für Vor- und Frühgeschichte der Universität Frankfurt/M. 58. Maier, U. 2011. Archäobotanische Flächenuntersuchungen in der endneolithischen Sied- lung Torwiesen II. Hemmenhofener Skripte 9. Marciniak, A., Baranski, M.Z., Bayliss, A., Czerniak, L., Goslar, T., Southon, J. and Taylor, R.E. 2015. Fragmenting times: Interpreting a Bayesian chronology for the Late Neo- lithic occupation of Çatalhöyük East, Turkey. Antiquity 89, 154–176. Ministry of Agriculture and Maritime Fishing. 2009. Elaboration de plans de développe- ment des filières de production agricoles et des produits de terroir de la région de Guelmim Es-smara, Rapport de la 2ème phase. Ministry of Agriculture and Maritime Fishing, Regional Director of Agriculture Guelmim Es-smara. Oates, J., McMahon, A., Karsgaard, P., Al Quntar, S. and Ur, J. 2007. Early Mesopotamian urbanism: A new view from the north. Antiquity 81, 585–600. Palmer, C. 1998. The role of fodder in the farming system: A case study from northern Jordan. Environmental Archaeology 1, 1–10. Arable weeds as a case study 111 Qi, X., Liu, Y., Vigueira, C.C., Young, N.D., Caicedo, A.L., Jia, L., Gealy, D.R. and Olsen, K.M. 2015. More than one way to evolve a weed: Parallel evolution of US weedy rice through genetic mechanisms. Molecular Ecology 24, 3329–3344. Saatkamp, A., Affre, L., Dutoit, T. and Poschlod, P. 2009. The seed bank longevity index revisited: Limited reliability evident from a burial experiment and database analyses. Annals of Botany 104, 715–724. Saatkamp, A., Affre, L., Dutoit, T. and Poschlod, P. 2011. Germination traits explain soil seed persistence across species: The case of Mediterranean annual plants in cereal fields. Annals of Botany 107, 415–426. Schlichtherle, H. 1981. Cruciferen als Nutzpflanzen in neolithischen Ufersiedlungen Süd- westdeutschlands und der Schweiz. Zeitschrift für Archäologie 15, 113–124. Song, B.-K., Chuah, T.-S., Tam, S.M. and Olsen, K.M. 2014. Malaysian weedy rice shows its true stripes: Wild Oryza and elite rice cultivars shape agricultural weed evolution in Southeast Asia. Molecular Ecology 23, 5003–5017. Southwood, T.R.E., Brown, V.K. and Reader, P.M. 1986. Leaf palatability, life expectancy and herbivore range. Oecologia 70, 544–548. Stokes, P. and Rowley-Conwy, P. 2002. Iron age cultigen? Experimental return rates for Fat Hen ( Chenopodium album L.). Environmental Archaeology 7, 95–99. Styring, A.K., Charles, M., Fantone, F., Hald, M.M., McMahon, A., Meadow, R.H., Nich- olls, G.K., Patel, A.K., Pitre, M.C., Smith, A., Sołtysiak, A., Stein, G., Weber, J.A., Weiss, H. and Bogaard, A. 2017. Isotope evidence for agricultural extensification reveals how the world’s first cities were fed. Nature Plants 3, 17076. doi:10.1038/nplants.2017.76. Thompson, K., Band, S.R. and Hodgson, J.G. 1993. Seed size and shape predict persistence in soil. Functional Ecology 7, 236–241. Thurber, C.S., Hepler, P.K. and Caicedo, A.L. 2011. Timing is everything: Early degradation of abcission layer is associated with increased seed shattering in U.S. weedy rice. BMC Plant Biology 11, 14, doi:10.1186/1471-2229-11-14. Thurber, C.S., Reagon, M., Gross, B.L., Olsen, K.M., Jia, Y. and Caicedo, A.L. 2010. Molecular evolution of shattering loci in U.S. weedy rice. Molecular Ecology 19, 3271–3284. Ur, J. 2010. Cycles of civilization in northern Mesopotamia, 4400–2000 BC . Journal of Archaeological Research 18, 387–431. Vavilov, N.I. 1992. Origin and Geography of Cultivated Plants. Cambridge, Cambridge University Press. Weiss, E., Kislev, M.E. and Hartmann, A. 2006. Autonomous cultivation before domestication. Science 312, 1609–1610. Weiss, H., deLillis, F., deMoulins, D., Eidem, J., Guilderson, T., Kasten, U., Larsen, T., Mori, L., Ristvet, L., Rova, E. and Wetterstrom, W. 2002. Revising the contours of history at Tell Leilan. Annales archéologiques arabes syriennes 45, 59–74. Wiethold, J. 1996. Von Dinkel, Einkorn und Kornrade: Archäobotanische Untersuchungen im Bereich der römischen Villenanlage von Borg, Kr. Merzig-Wadern. In 10 Jahre Aus- grabung Perl-Borg 1986–1996 . Merzig, Landkreis Merzig-Wadern, Arbeitsverwaltung Merzig-Wadern. Willcox, G. 2002. Charred plant remains from a 10th millennium B.P. kitchen at Jerf el Ahmar (Syria). Vegetation History and Archaeobotany 11, 55–60. Willcox, G., Buxo, R. and Herveux, L. 2009. Late Pleistocene and early Holocene climate and the beginnings of cultivation in northern Syria. The Holocene 19, 151–158. Willcox, G., Fornite, S. and Herveux, L. 2008. Early Holocene cultivation before domestication in northern Syria. Vegetation History and Archaeobotany 17, 313–325. 112 Amy Bogaard, Mohammed Ater, John G. Hodgson Willerding, U. 1980. Zum Ackerbau der Bandkeramiker. Materialhefte zur Ur- und Früh- geschichte Niedersachsens 16, 421–456. Wright, I.J., Reich, P.B., Westoby, M. et al . 2004. The worldwide leaf economics spectrum. Nature 428, 821–827. Wright, I.J. and Westoby, M. 2002. Leaves at low versus high rainfall: Coordination of structure, lifespan and physiology. The New Phytologist 155, 403–416. Zohary, D., Hopf, M. and Weiss, E. 2012. Domestication of Plants in the Old World . Fourth ed. Oxford, Oxford University Press. Part II How domestication changes humans’ bodies and sociality

6 From fighting against to becoming with Viruses as companion species

Charlotte Brives

In 1347, the Black Death struck the city of Constantinople, and began to spread across Europe. Following improvements in navigation techniques and the establishment of contact between the American and European continents, the 15th and 16th centuries saw entire populations decimated by viruses such as smallpox. In 1683, Anton van Leeuwenhoek systematized his descriptions of what he decided to call in 1677 “animalcules”, as seen through his microscope. More than a century later, in 1796, Edward Jenner developed the technique for vaccination against smallpox. In 1854, John Snow established the hydrologic cartography of the spread of cholera by careful investigation of different London neighborhoods, after having theorized the role of water pumps in the transmission of cholera in 1849. In the 1860s, Louis Pasteur developed the idea, already present in earlier writings such as those of Gianfranco Fracastore in 1546, though in a very different form, that infectious diseases are caused by living organisms, which he called “germs”. This theory was taken up and validated by the work of Robert Koch, who, in 1876, succeeded in cultivating the anthrax agent, Bacillus anthracis , against which Pas- teur developed a vaccine in 1885. In 1892, Dmitri Ivanowski, who was working on the tobacco mosaic, proved the existence of tiny pathogens, identified in 1898 by the Dutch chemist Martinus Beijerinck and later renamed as “viruses”. Between 1918 and 1920, the Spanish Flu caused more than fifty million deaths. It was not until 1937, and the invention of the electron microscope, that the first image of a virus was obtained. Although smallpox was considered to be permanently eradicated from the world in 1980, new infections were identified in 1981. Its agent was first isolated in 1983. After different controversies concerning its discovery, the term human immunodeficiency virus (HIV) was coined in 1986 to name the causal agent of acquired immune deficiency syndrome (AIDS). These dates mark the recent history of germs and humans, a history presented as a war, an arms race between species. Because of the success of germ theory, and through an obliteration of the subtler interpretations of our relationship with microbes, especially those developed by Pasteur himself, it is indeed through the metaphor of struggle that the relations between humans and microbes are most often not only presented, but also considered and analyzed. While the idea that bacteria can be classified as something other than dangerous pathogens is becoming increasingly widespread, in particular through emphasis on probiotics and the 116 Charlotte Brives development of microbiota work in recent years, viruses continue to suffer from a certain form of ignorance which limits them to the role of mortal entities. Through their destructive and disruptive potential, viruses both frighten and fascinate. This metaphor has not changed, even though its uses have a profound impact on our ways of apprehending each other, whether in the field of science itself, in the theories and hypotheses developed by researchers, as brilliantly demonstrated by Evelyn Fox Keller (2002), or in the public and media spheres (Larson et al ., 2005). The use of metaphor, the assimilation of the virus as the foe which must be destroyed, and from which we must protect ourselves, has led, as Priscilla Wald has meticulously analyzed, to the construction of a constantly repeated narrative, the “outbreak narrative”, structured around three major events: the emergence of infection, its spread via a global network of contacts and ultimately its containment (or eradication as in the case of smallpox). These stories, far from being insignificant, affect our responses to the challenge of epidemics in a globalized world, and stigmatize individuals and populations, behaviors and lifestyles ( Wald, 2008 ). Yet the vast amount of data accumulated on certain viruses by many disciplines over the decades is matched only by our profound ignorance of them. Indeed, although some infectious viruses may be fairly well characterized, and although a whole corpus exists on a handful of bacteriophage viruses that have become formidable tools for the development of molecular biology,1 the diversity of these entities, their constitution, their ecology, their hosts and vectors, their characteristics and properties, not to mention their number, make them the most adaptable and the most widespread entities on Earth. If they are the ultimate parasites (because they can rarely survive outside of their hosts) and are found everywhere, viruses are most often totally harmless to humans. We thus develop unrecognized and complex relationships with them, which we are only just beginning to try to understand from a new angle.2 My participation in a colloquium on domestication offered the opportunity to work on these relationships and to renew the way we see these entities. While the term “domestication” is generally used to describe our relationship with certain animals (for example, but not limited to, pets), and also with cultivated plant species, in this chapter I propose an interpretation of our relationship that is no longer based on the paradigm of struggle, but on one that leaves open the possibility of observing virus performance to show that, in turn, this affects the co- performances of humans and viruses. To do so, I decided not to build on the existing concepts and theories of microbial ecology, and of biology in general. As mentioned, our conceptions of viruses are strongly related to the knowledge scientists have patiently produced about/with them. Thinking with those concepts could then inevitably lead to the same metaphors and analogies.3 Instead, in order to conduct a different analysis of our deep entanglements, I propose to tell stories of humans and viruses. And to introduce this little narrative, I would like to begin with the words of a leading scientist who, at the end of a long career devoted to genetics and microorganisms, shows us how urgent it is to think differently about these invisible creatures. From fighting against to becoming with 117 Lederberg (2000 : 287) wrote in the journal Science:

Our most sophisticated leap would be to drop the Manichean view of microbes – “we good, they evil.” Microbes indeed have a knack for making us ill, killing us, and even recycling our remains to the geosphere. But in the long run microbes have a shared interest in their hosts’ survival: a dead host is a dead end for most invaders too. Domesticating the host is the better long- term strategy for pathogens.

Although Lederberg does not completely depart from the metaphor of the struggle in this excerpt – nor from the outbreak narrative described by Wald, going so far as to speak of “invaders” – domestication, presented as a quasi-military strategy, leads to a shift in the conceptualization of the relationships between pathogens and humans. The very use of the term implies for Lederberg the restitution of an agency to the microbes: they must domesticate humans, in the sense that they must learn to live with representatives of this species without destroying them totally, without annihilating them. They must adapt to them and to the resources they offer without overwhelming them. They must use them without destroying them. Three viruses or types of virus are outlined in the following sections: the smallpox virus (Variola major), endogenous retroviruses and, finally, the viral archetype of our time: HIV. These three viruses or types of virus tell three stories, which reveal a generally ignored aspect of our relationship with pathogens, and surmount Lederberg’s final obstacles, offering a history of reciprocal domestication for our consideration. These three moments of reflection will then lead us to consider, in the wake of Donna Haraway, viruses as companion species (Haraway, 2008).

Smallpox and the Spanish conquest The specific pathogenic virus/host relationship is subtle and extremely sensitive. It is ever-changing, and dependent on the performance of representatives of both species, as illustrated by the history of the Spanish conquest of the Americas. One of the most incontrovertible examples of the dangerousness and destructive and disruptive potential of viruses and pathogens in general is, without doubt, the one sadly offered by the arrival of Europeans in America, following the discovery of a new navigation route by Christopher Columbus in 1492. It is estimated that by the beginning of the 17th century, only one million of the eighteen million people who lived in Mexico before the Spanish Conquest had survived. The decimation of indigenous populations is of course explained by a complex set of factors. First, by the way they were treated by the conquistadors, but also by their coming into contact with numerous pathogens, first and foremost smallpox. Although the story of the introduction of smallpox in Central America can be told in the context of an outbreak narrative and perceived as a human and viral invasion of a continent, against which the natives battled, there is another story to tell from the encounter between natives, colonizers and viruses. This second story has a more environmental context, and takes into account the long and shared 118 Charlotte Brives history of Europeans and Variola major, and the properties of each agent. At that time, European populations had regularly been affected by smallpox epidemics over a period of several centuries. The repeated interactions of the virus with the populations had nevertheless allowed them to become progressively less suscep- tible to the virus. At an environmental level, the relationship between humans and viruses had been somewhat long established: the virus was part of the Euro- pean environment, it was part of the community in the ecological sense (community understood as all species inhabiting a given space). On the other hand, the Amerindian populations shared no common history with the smallpox virus. Their brutal encounter with the Europeans was thus amplified by an equally brutal encounter with this virus (and others), making the latter an agent of the Spanish conquest. The oft-mentioned question of the deliberate or unintentional dissemination of the virus,4 although controversial when considered within the framework of colonial relations, is interesting when thinking about the symmetry of the relationship between humans and viruses; microbes may well domesticate humans, as Leder- berg says, but it seems that the latter have fully understood how their co-history could enable them to use the former, thus committing us to a concept of human/ virus relations close to the notion of mutualism . However, for our purposes, the story of smallpox and the Spanish conquest presents one major drawback: the fact that it fails to release us from the metaphor of war. After all, it is a story of conquest, domination, struggle, enslavement and death. In order to part with this metaphor completely, I offer another example of our relationship with viruses. This second example is very different from the first, in that it concerns a theoretically non-pathogenic type of virus.5 This detour via endogenous retroviruses will subsequently allow us to address, in the final part of this article, the archetypal virus of the end of the 20th century and, undoubtedly, of part of the 21st: HIV.

Endogenous retroviruses and placentation Retroviruses are RNA (RiboNucleic Acid) viruses that have the ability to transform their RNA genes into DNA (DeoxyriboNucleic Acid) by means of an enzyme known as reverse transcriptase, so that they can be inserted into the host DNA and amplified by its cellular machinery. The result is the production of thousands of copies of the virus, or virions, released by the destruction of the infected cell.6 For reasons that may vary, but partly due to host cell defense mechanisms, some of these viruses are “trapped” in the host DNA. If this event occurs in a germline cell, it is then transmitted to subsequent generations. This concerns endogenous retroviruses, or HERVs (Human Endogenous Retroviruses), as opposed to exogenous retroviruses. Transfection of a germline cell, which may appear to be rare, has been sufficiently frequent in human history for endogenous retroviruses to account for 8% of our DNA ( Belshaw et al ., 2004 ; Griffiths, 2001 ). These DNA fragments may have no function. Some, however, play a major role in the development of mammals. From fighting against to becoming with 119 Syncytins are proteins encoded by endogenous retroviruses which play a crucial role in placental development. They enable the differentiation of some cells and their fusion to form the syncytiotrophoblast, part of the placenta responsible for feto-maternal interactions and other functions of the placenta ( Mallet et al ., 2004 ; Dupressoir et al., 2009). They also possess immunosuppressive properties, thereby playing a role in the immune protection of the fetus (Kämmerer et al ., 2011). The different origins of syncytin in mice and humans implies that the infection of mammalian cells by retroviruses encoding these proteins has occurred independently several times throughout history. In genetics, this insertion of viral DNA, which is then used by the host – a process that involves co-evolution and lasting interactions between entities – is called domestication . Viruses, far from simply being pathogenic entities, have thus played a fundamental role in the evolution of mammals, enabling them to exhibit new performances. We are because, on several occasions in history, subtle DNA interactions have occurred and become stabilized, born of repeated interactions between species that we now consider fundamentally distinct, when we do not overlook them completely. We cohabit with these viruses which are inserted deeply within our cells, into the nucleus, and which have fused their DNA with ours – DNA which, since the discovery of its structure in 1953, and at the height of the subsequent reductionist wave, symbolized and still symbolizes for many today the very essence of what makes us human. It is this point in particular, however, that must give us pause. Although the example of the endogenous retrovirus allows us to grasp the magnitude of the co-history of viruses and humans, it could nevertheless inspire, in comparison with the knowledge we have developed on smallpox, a somewhat misleading idea that humans are ultimately the passive receptors of endogenous viruses’ omnipotence. Viruses can make us suffer, and can even kill us, but, ironically for the partisans of the metaphor of war, it seems that our very ability to give life, as we know it, is a property that we owe in part to them. Power of life, power of death: could viruses be the new gods? Taking viruses seriously and recognizing their performance must not lead us to neglect the performance of humans. This can be seen when we address the question of potential culpability behind the spread of smallpox. Though of a more complex nature, it can also be seen in the immune defense mechanisms that have led to the stable insertion of viral DNA into our genomes. However, in order to restore their respective agencies to both humans and viruses, a sine qua non for domestication, I would like to address one last example: the very specific relationship between humans and HIV.

HIV and politicization HIV and its relationship to humans is extremely complex from an analytical point of view because of the amount of knowledge and experience we have accumulated in recent decades. The virus was discovered in 1983 and sequenced in 1985. Along with basic research, epidemiological data and also the establishment of 120 Charlotte Brives blood sampling banks, the study of observation groups, and the study of the consequences of the various control and prevention strategies implemented over the last thirty years, have all helped to build up a vast body of knowledge, including, but not limited to: virology, molecular biology, evolutionary biology, phylogenet- ics, medicine, pharmacology, epidemiology, human and environmental ecology, history, geography, health economics and anthropology. The euphoria that resulted from the discovery of HIV and the hope of a rapidly available treatment for patients infected at the beginning of the epidemic, have unfortunately been followed by years of doubt, uncertainty, suffering and stigmatization. New stories were created to replace the old outbreak narratives, characterized by the victory of humans and the end of the battle against a virus which had been silenced ( Wald, 2008 ). But the end of the epidemic now seems out of reach. The stories of HIV, replete with the fantasies of the Western elite, have often been brutal and extremely violent in their desire to place responsibility for an epidemic that no one could have predicted on others, from the African male and his unbridled sexuality, to the four Hs (hemophiliacs, Haitians, homo- sexual males, heroin users) not to mention a poor steward reduced to the status of patient zero, all sharing an obvious lack of morality (Treichler, 1987; Packard and Epstein, 1991 ). At the same time, the designation of HIV as the cause of AIDS quickly led to the naturalization – or rather, the biologization – of the disease, leading to the concentration of almost all action on scientific and medical research, and thus on the virus itself (to the detriment of a dynamic and ecological approach that recognizes the fundamental, and not just the contextual, role of the environment). Interna- tional programs, motivated by the same thought process, offset out to eradicate the virus, but many specialists have raised their voices to highlight the dangers and limitations of biomedicalizing the response to the epidemic (Nguyen et al ., 2011 ). Joshua Lederberg (2000: 5, my italics), again, draws up a slightly critical assessment of the efforts made:

AIDS and HIV have spurred the most concentrated program of biomedical research in history, yet they still defy our counterattacks. And our focus on extirpating the virus may have deflected less ambitious, though more prag- matic, aims, including learning to live with the virus, by nurturing in equal measure the immune system that HIV erodes. After all, natural history points to analogous infections in simians that have long since achieved a mutually tolerable state of equilibrium.

Once again, although Lederberg cannot abandon the war metaphor (no matter how the Empire strikes back, the force always seems to be with the viruses), it nevertheless leads us down another path, that of “living with the virus”, and invites us to draw links between ourselves and the natural histories of other species, as well as introducing an element of humility to our attitudes. But Lederberg does not invite everyone to the negotiating table. By bringing the subject back From fighting against to becoming with 121 to the immune system, he remains imbued with a naturalism that sends HIV back to a biological entity and reduces the human dimension to its physiological components. Just as before, however, I would like to explore Lederberg’s “learning to live with” approach and extend my reflections on the domestication of viruses, based on two sets of data: the first, from Jacques Pépin’s book, The Origins of AIDS , in order to consider our relationship with HIV from a historical point of view; the second, from my own fieldwork on several clinical trials conducted on HIV in sub- Saharan Africa, in order to describe and analyze the hybrid community formed by humans and viruses.

The Origins of AIDS In The Origins of AIDS, Jacques Pépin (2011) retraces the complex history of the emergence of HIV. HIV originates in the SIVcpz virus (genetically identical to HIV), present for several hundred years in the Pan troglodytes troglodytes chimpanzee species. While the infection of a human with the chimpanzee virus is an event that may have occurred during the hunting of this animal or its preparation as a source of food, the spread of the epidemic nevertheless required the concordance of various factors, the first of which is colonization. Phylogenetic data make it possible to estimate that the common ancestor of the current HIV strains dates back to 1921. This does not mean that chimp/human interspecies transmissions didn’t occur before this date, only that the necessary conditions for an amplification of the number of infections were not present: a hunter or his wife, contaminated at the time of hunting or cooking, would have infected each other and perished a few years later in their village without having had the opportunity to transmit the virus outside the family circle. In 1921, however, following the various waves of colonization, Africa had undergone profound social, political and medical transformations. Pépin explains that it has certainly taken two distinct transmission paths to generate the epidemic, which he carefully reconstitutes on the basis of a meticulous historical approach: a first phase, by parenteral transmission (that is, by needles and syringes which have been contaminated and poorly sterilized during vaccination or prevention campaigns, in particular against sleep- ing sickness, and/or during leprosy treatment campaigns in the same period).7 The increase in the number of individuals infected via this route in their own villages then makes sexual transmission and amplification more likely to happen. Prior to colonization, an infected man leaving his village to go to the city (presumably Leopoldville/Brazzaville) and contaminating a sex worker is a very low probability event. The changes brought about by the arrival of settlers, including the migration of men to the cities to find work and the massive development of the sex industry in the following decades, changed the situation considerably, opening up a second phase of amplification via sexual transmission. Pépin also explains the emergence of a number of epidemic outbreaks, such as in Haiti, and he traces, on the basis of the properties of the viral strains as well as the socio-cultural data available, the different trajectories of the virus, and that of humans, in the world. 122 Charlotte Brives Far from the simplistic story of a jungle-like virus that spread by taking advantage of people’s loose morals, Pepin depicts the deep entanglement of humans and nonhumans, and shows how historical events, medical innovations and social and cultural changes have allowed the spread of the virus, but also how it has adapted, mutated and differentiated itself according to the properties specific to each strain. The emergence of HIV is above all a co-history of humans and viruses.

Antiretrovirals (ARVs) and the domestication of HIV This intertwining of humans and non-humans has, since 1981 and the U.S. Cen- ters for Disease Control and Prevention (CDC) press release on the first cases of AIDS, never ceased to grow and become more complex. Nearly a century after the appearance of the founder strain, the epidemic, by virtue of its scale, its evolution and its duration, made it possible to dramatically highlight a certain perpetuation – and then institutionalization – of the links between humans and HIV, involving more and more actors.8 While the Global North-South divide in access to treatment is still very wide today, it is reasonable to speak of a chronic epidemic in which, especially in the Global North, people can live several decades with HIV without developing AIDS. We now speak of People Living with HIV (PLWHIVs). If we compare this idea to that developed by Lederberg, what it underpins far exceeds it, and is anchored in the daily lives of infected individuals. Focusing on the question of treatment based on data collected from fieldwork in Côte d’Ivoire, Burkina Faso and South Africa, I would like to show how this “living with” can be conceptualized in terms of domestication. From 2011 to 2013, I was, as a post- doctorate researcher, part of a program aimed at analyzing the scientific, clinical and individual reasoning that prevailed in the setting up and running of a ran- domized clinical trial (RCT) on the evaluation of early antiretroviral treatment of HIV-infected people.9 Through careful study of the protocol, one of our goals was to understand how investigators were conceptualizing the relationship between viruses and humans in order to act on the former for the benefit of the patient. The issue of early antiretroviral treatment can be presented as follows: there is a threshold of infection at which point it is generally agreed a patient should receive antiretroviral treatment.10 This threshold is partly determined by a benefit/ risk ratio that balances the treatment’s side effects and the risks of viral resistance to treatment, but also compliance with treatment.11 In the study, the aim was to lower the threshold for treatment, i.e. to place pients on treatment earlier on in the course of infection. Early treatment comes from a desire to eradicate the virus: asymptomatic patients are given treatment to control their infection, but also to limit the contamination of new individuals.12 During this research program, I was able to show how the trial constructed a complex device in which humans were reduced to a biological body, summarized into biochemical and physiological data and reduced to a standardized body. For the purposes of the RCT,13 the human body was in fact assimilated to the environment of the virus. Only the characteristics considered by the investigators to be relevant for analyzing the relationships between humans and viruses and From fighting against to becoming with 123 monitoring their evolution were retained. In this device, antiretroviral drugs were used to modify this environment so as to make it unfit for the reproduction of the virus. 14 By ingesting specific, targeted molecules, it is possible to “control” the virus population in the body, and in this sense to domesticate it, if we mean by domestication “the control of reproduction”, but also from a behavioral point of view “the mitigation of dangerous virus behaviors”. The RCT thus constructed a simple device: a biological body, viruses, and molecules that are injected into the body in order to control/domesticate the viral population. The ingestion of the treatment, however, is based on complex dynamics, reminding us that the patient is not only a biological body, but also a social individual with a life of their own, constrained by all kinds of needs. The treatment can therefore only be understood by placing the infected individual back into the domestic space, into their everyday interactions with their familial and professional surroundings (Bessette et al., 2001). In other words, the domestication of the virus through treatment implies both the patient’s cooperation with a certain number of rules concerning the correct way to self-administer their treatment, and considerable changes to their daily life. Domesticating the virus also involves learning that what contains it, constrains it and also constrains us. Viruses and humans form what could be termed a hybrid community, which requires precise rules, including the regular and controlled intake of antiretroviral drugs. Although this domestication can be read on an individual level, the consequences of the existence of this hybrid community and its evolution are much broader. In the fight against HIV, the involvement of infected persons has been crucial at various levels, from the demand for rights to treatment (Epstein, 1998) to the creation of support groups for patients whose status is far from being systematically shared with others around them. The need to survive the HIV epidemic led to the politicization of patients and some of their relatives, which was reinforced by the arrival of antiretroviral drugs and the demand for what they promised to achieve ( Nguyen, 2010 ). When, as part of the ANRS12242 research program, we arrived in Abidjan to observe the trial, we sought to contact support groups, which are usually frequently available, in order to discuss with patients their participation in or, conversely, their exclusion from, the trial, and to evaluate the effects. However, it seemed that such groups did not exist for this trial. In investigating this state of affairs, we came to the conclusion that because the trial guaranteed access to treatment and to high quality healthcare and, to a certain extent, protection from stigmatization, there was no need for infected patients to regroup. This partial depoliticization of individuals does not mean a return to a pre-antiretroviral period, but rather the establishment of new relationships with viruses.15 The depoliticization effect due to access to ARVs was also felt in South Africa, when we met with former contacts of Frédéric Le Marcis, who were part of the Treatment Action Campaign (TAC) in the early 2000s. They told us that they had left the movement after receiving treatment. In other words, the creation of a hybrid community thanks to ARVs led to the dislocation of the community of patients, who no longer felt so strongly the need to be together. 124 Charlotte Brives While the objective of the World Health Organization (WHO) remains an out- right eradication of HIV, as has been achieved with smallpox,16 the experience of millions of people leads us to think more in terms of domestication and coexistence.

Conclusion: reciprocal domestication and companion species In her wonderful book, Donna Haraway ( 2008 ) introduces us to our companion species, and reminds us that domestication is always a matter of reciprocity. Emi- lie Hache (2015 ) illustrates this with her use of the title of one of Haraway’s chapters, “If I have a dog, my dog has a human”, stressing the desire of the American philosopher not to place the accent on non-human animals but on the relationship between humans and non-humans. While Haraway, as an accomplished biologist, includes microbes from the very first pages of her introduction, and also cites bacteria, fungi17 and protists, she does however overlook viruses. There are perhaps many reasons for this oversight, but it is nonetheless symptomatic of the difficulty posed by these entities, situated on the fringes of our categories of thought. Yet, if dogs have helped in the making of humans, in the construction of civilizations through the different tasks they have taken on, what about viruses? They have mixed their DNA with ours, made it possible for us to conquer new territories by favoring certain populations, and contribute daily to politicization in all its forms as well as to the way we live, feel and experiment. Through the scale and duration of the epidemic, HIV allows us to appreciate the scale of this reciprocal domestication and its effects, if we leave behind the rhetoric of eradication. When discussing companion species, Haraway uses the particularly apt expression, “becoming with ” (Haraway, 2008: 3). This makes sense when we stop naturalizing our relationship with viruses, and take a serious look at what they are doing to us, at what we become with them. As for the acronym PLWHIV (People Living With HIV), which implies a certain amount of passivity, the mere observation of the omnipresence of a virus within human cells, circulating in our bloodstreams, leads me to believe that PBWHIV, People Becoming With HIV, is a better term. It has the advantage of restoring the respective agencies to the different human and non-human entities, recognizing them as more than simple puppets of a powerful Mother Nature. This is not to deny or ignore the problems faced by people with HIV. We say we take viruses seriously, but only the threat they represent, reinforced by decades of terrifying stories of contagion, is the subject of sustained attention. As well as foolish dreams of eradication, the consequences of surveillance networks and biosafety rhetoric ( Lakoff and Collier, 2008 ) go beyond the scope of discourse and impose themselves on our actions. In the name of the fight against viruses, decisions are made which are sometimes violent and often disruptive. To speak of domestication, of becoming with viruses, by no means implies neglecting or denying the sufferings that they can inflict, but rather offers a way to recall that they reveal, and not create, inequalities and violence. This shift points the way to new paths towards the development of a more ecological conception of our future, both individual and collective, and towards imagining other possible futures. From fighting against to becoming with 125 Notes 1 See the work of the Phage group, initiated by Max Delbrück (Cairnset al ., 1966) 2 For a philosophical approach to some of the questions posed by these entities, see the special issue of the journal Studies in History and Philosophy of Biological and Bio- medical Sciences , coordinated by Thomas Pradeu et al . (2016 ). 3 A systematic analysis of these concepts is necessary, however, when it comes to viruses, but this is not the point of this chapter. 4 By the use of contaminated blankets left in the vicinity of villages, for example. 5 I say “theoretically” because it would appear that some endogenous retroviruses could be involved in human pathologies (Dolei, 2006). 6 This, in simple terms, describes how a virus like HIV functions. 7 Africa was also built up as a vast laboratory, especially for Pasteurians (Latour, 1993; Tilley, 2011 ). 8 These include the foundation of UNAIDS in 1996; the Global Fund to Fight AIDS, Tuberculosis and Malaria in 2002; and PEPFAR in 2003. 9 “L’essai clinique comme espace de rencontre. Des logiques scientifiques aux expériences individuelles: construction des participants, biosocialité et expériences de subjectivation (Abidjan – Côte d’Ivoire)” (“Clinical trials as a meeting space. Scientific reasoning for personal experiences: the making of participants, bio-socialness and experiments in sub- jectification (Abidjan – Cote d’Ivoire)”), Program ANRS12242, coordinated by Frédéric Le Marcis (LADEC FRE2002) and Mariatou Koné (Cocody-Abidjan University). 10 This threshold is defined by the number of CD4 cells, which are one of the targets of the virus, and which play an important role in the immune response system. The lower the number, the weaker the immune system of the infected person. 11 Compliance is defined as patient respecting the dosages and the hours at which drugs must be taken. 12 Patients undergoing ARV treatment generally have undetectable virus levels and are non-contagious. 13 Depending on the discipline, but also according to the mechanisms in place to fight HIV, the environment of the virus will be defined differently, therefore having varied consequences on the way solutions are proposed. 14 For more detail, see Brives (2013 ), and Brives and Le Marcis (2015 ). 15 For an extensive and more complex development of this argument, see Brives and Le Marcis (2015 ). 16 Proof of the importance of understanding viruses not alone but in their complex relationships to humans. HIV and Variola major are two fundamentally different viruses, with which we have established distinct relationships. What was possible for the former was hardly conceivable for the latter. 17 See Anna Tsing’s beautiful article, “Unruly edges: Mushrooms as companion species” (Tsing 2012).

References Belshaw, R., Pereira, V., Katzourakis, A., Talbot, G., Paces, J., Burt, A. and Tristem, M., 2004. Long-term reinfection of the human genome by endogenous retroviruses, PNAS , 101(14), 4894–4899. Bessette, D., Bungener, M., Costagliola, D., Flori, Y.-A., Matheron, S., Morin, M., Setbon, M. and Souteyrand, Y. (eds.), 2001. L’observance aux traitements contre le VIH/sida. Mesure, déterminants, évolution , Paris: ANRS collection sciences sociales et sida. Brives, C., 2013. Identifying ontologies in a clinical trial, Social Studies of Science , 43(3), 397–416. 126 Charlotte Brives Brives, C. and Le Marcis, F., 2015. Réimaginer des communautés? Les conséquences du traitement précoce contre le VIH/sida en Côte d’Ivoire, Terrain , 64, 84–103. Cairns, J., Stent, G.S and Watson, J.D., 1966. Phage and the Origins of Molecular Biology . Cold Spring Harbor, New York: Cold Spring Harbor Laboratory Press. Dupressoir, A., Vernochet, C., Bawa, O., Harper, F., Pierron Gérardn Opolon, P. and Heid- mann, T., 2009. Syncitin-A knockout mice demonstrate the critical role in placentation of a fusogenic, endogenous retrovirus-derived, envelope gene, PNAS , 106, 12127–12132. Dolei, A., 2006. Endogenous retroviruses and human disease, Expert Review of Clinical Immunology , 2(1), 149–167. Epstein, S., 1998. Impure Science, Aids, Activism and the Politics of Knowledge , Berkeley, CA: University of California Press. Griffiths, D. J., 2001. Endogenous retroviruses in the human genome sequence, Genome Biology , 2(6), reviews 1017.1–1017.5. Hache, E., 2015. If I have a dog, my dog has a human, Canadian Journal of Continental Philosophy/Revue canadienne de philosophie continentale , 19(2), 7–21. Haraway, D., 2008. When Species Meet , Minneapolis, MN: University of Minnesota Press. Kämmerer, U., Germeyer, A., Stengel, S., Kapp, M. and Denner, J., 2011. Human endogenous retrovirus K (HERV-K) is expressed in villous and extravillous cytotrophoblast cells of the human placenta, Journal of Reproductive Immunology , 91, 1–8. Keller, E. F., 2002. Making Sense of Life, Explaining Biological Development With Models, Metaphors and Machines , Cambridge, MA: Harvard University Press. Lakoff, A. and Collier, S. (eds.), 2008. Biosecurity Interventions Global Health and Security in Question: Global Health and Security in Practice , New York, NY: Columbia Univer- sity Press. Larson, B. M. H., Nerlich, B. and Wallis, P., 2005. Metaphors and biorisks: The war on infectious diseases and invasive species. Science Communication , 26(3), 243–268. Latour, B., 1993. The Pasteurization of France, Cambridge, MA: Harvard University Press. Lederberg, J., 2000. Infectious history, Science , 288(5464), 287–293. Mallet, F., Bouton, O., Prudhomme, S., et al., 2004. The endogenous retroviral locus ERVWE1 is a bona fide gene involved in hominoid placental physiology, PNAS , 101(6), 1731–1736. Nguyen, V.-K., 2010. The Republic of Therapy: Triage and Sovereignty in West Africa’s Time of AIDS . Durham and London: Duke University Press. Nguyen, V.-K., Bajos, N., Dubois-Arber, F., O’Malley, J. and Pirkle, C. M., 2011. Remedi- calizing an epidemic, from HIV treatment as prevention to HIV treatment is prevention, AIDS , 25(3), 291–293. Packard Randall, M. and Epstein, P., 1991. Epidemiologists, social scientists, and the structure of medical research on AIDS in Africa, Social Science and Medicine, 33(7), 771–783. Pépin, J., 2011. The Origins of AIDS , Cambridge: Cambridge University Press. Pradeu, T., Kostyrka, G. and Dupré, J., 2016. Understanding viruses, Studies in History and Philosophy of Biological and Biomedical Sciences , 59, 57–63. Tilley, H., 2011. Africa as a Living Laboratory, Empire, Development, and the Production of Scientific Knowledge , Chicago, IL: The University of Chicago Press. Treichler, P. A., 1987. AIDS, homophobia and biomedical discourse: An epidemic of signi- fication, Cultural Studies , 1(3), 263–305. Tsing, A., 2012. Unruly edges: Mushrooms as companion species, Environmental Humani- ties , 1, 141–154. Wald, P., 2008. Contagious, Cultures, Carriers, and the Outbreak Narrative, Durham and London: Duke University Press.

7 Milk as a pivotal medium in the domestication of cattle, sheep and goats

Mélanie Roffet-Salque, Rosalind E. Gillis, Richard P. Evershed and Jean-Denis Vigne

Introduction: milking and humans’ relationship with domestic animals The domestication of animals and the development of food-producing economies facilitated the introduction of new foods into the human diet. Moreover, the introduction of animal milks completely reshaped the human culture, biology and economy. The multi-billion Euro modern dairy economy is the direct consequence of the both humans’ ability to tame ruminant species and their genetic adaptation to animal’s milk, namely lactase persistence (LP), with most people in Europe being tolerant to lactose. Milk links humanity with animals, and indeed, breastfeeding pets, such as dogs, pigs, bears and monkeys, has been reported in hunter-gatherer societies ranging from Australian aborigines to South American Indians (Milliet, 2007). This form of positive care is believed to ensure the survival of the young animal, for example for economic gain, such as for the valuable sleigh dogs among the Inuit (Gray and Young, 2011). Seasonal observations of dams suckling young would have resonated within hunter-gatherer populations familiarising them with animal milk as a food source ( Fischler, 1988 ). The capture of dams with their offspring would have been easier than a buck and likely aided the domestication process, with young being hand reared by humans, thereby reducing the flight instinct ( Hemmer, 1990 ). Once in settlements, tamed animals could have been exploited for their milk. Archaeozoological evidence shows clearly that domesticated ruminants from the Near East at some of the earliest farming sites were exploited for their milk (Helmer, 1994; Helmer et al ., 2007 ). Those captive animals would have been valued as living animals, rather than as a source of meat. In the archaeological record, direct evidence of positive care has been observed in early managed goats with broken hind limbs at the pre-pottery Neolithic B (PPNB) site of Jarmo (Central Zagros) being treated instead of, as one would expect, slaughtered for their meat (Bendrey, 2014). Studies have shown that a strong relationship of mutual trust develops between the farmer and each dairy animal ( Porcher and Schmitt, 2012 ). This attachment relates both to affection, partly because milking requires a physical proximity between the dairyper- son and the animal, and utility. In contrast, cattle raised for meat are often seen 128 Mélanie Roffet-Salque et al. as a “tool of production” and considered as part of a larger entity, the herd, rather than individual animals (Bock et al., 2007:121). This invokes a situation where domesticated animals were chosen for their capacity to produce milk and where the developing relationship between prehistoric humans and early domesticated animals was cemented by the act of milking.

Milk as a nutritious food Milk is a rich nutritional food source, made up of water, sugars (mainly lactose), proteins, lipids and minerals. At present, it is a major source of dietary energy, protein and fat, as well as containing numerous nutrients providing a significant contribution to the body’s daily requirements (McClellan et al., 2008). In 2007, 84.9 kg of milk and dairy products per capita were consumed globally ( Gerosa and Skoet, 2013 ). The composition of milk varies between species and is a reflection of the long-term nutritional needs of the offspring. Lactose is important in the intestinal absorption of calcium, magnesium and phosphate, and the utilisation of vitamin D (Wijesinha-Bettoni and Burlingame, 2013). Human milk is high in lactose but low in protein in comparison with the other species (Table 7.1). This variation in milk composition between humans and ruminants is a reflection of the semi-altricial state that babies are born in compared to cattle, sheep and goats’ offspring ( McClellan et al ., 2008 ). Calves, lambs and kids are able to stand within hours of their birth and grow quickly so require high-quality protein to support maximal growth. Milk composition also changes over the lactation period to meet the need of the infant and as a result of foddering practices (Wijesinha-Bettoni and Burlingame, 2013 ). Animal milks, pure or mixed with cereals and pulses, are often used as a weaning food for human infants. Due to nutritional differences in composition between human milk and ruminant milk, such as the lack of iron in ruminant milk, it is recommended that infants should not be given untreated non- human milk before 12 months old, particularly cow’s milk ( Leung and Sauve, 2003). However, after 12 months, the consumption of milk and dairy products has noticeable positive effects on child health and development ( Wijesinha-Bettoni and Burlingame, 2013 ).

Table 7.1 Composition of cow, sheep, goat and human milk (after Table 3.3; Wijesinha- Bettoni and Burlingame, 2013 ).

Species Water Fat Protein Lactose Ash/Minerals (g/100g) (g/100g) (g/100g) (g/100g) (g/100g) Average Range Average Range Average Range Average Range Average Range

Cow 87.8 87.3–88.1 3.3 3.1–3.3 3.3 3.2–3.4 4.7 4.5–5.1 0.7 0.7–0.7 Sheep 82.1 80.7–83.0 6.4 5.8–5.7 5.6 5.4–6.0 5.1 4.5–5.4 0.9 0.8–1.0 Goat 87.7 86.4–89.0 3.9 3.3–4.5 3.4 2.9–3.8 4.4 4.2–4.5 0.8 0.8–0.8 Human 87.5 – 4.4 – 1.0 – 6.9 – 0.2 – Milk as a pivotal medium in domestication 129 Livestock domestication and the role of milk Animal remains can provide evidence for a species being domesticated, as archaeozoological analyses allow the detection of an animal translocated outside its normal ecological range, morphological changes (e.g. decrease in bone size, changes in teeth shape) and/or changes in slaughtering profiles and/or herd composition (e.g. increase in females) usually implying an anthropogenic impact on the species in question (Vigne, 2011). Evidence for intensive cattle, sheep and goat management appeared around 10.5 kyBP in the middle and upper Euphrates valley and the Zagros Mountains within PPNB communities (Helmer et al., 2005a; Peters et al., 2005 ; Zeder, 2005 ). By 10.4–10.2 kyBP, these intensive practices appeared in the South Levant ( Horwitz et al ., 1999 ) and Cyprus ( Vigne et al ., 2011 ). This intensification appeared to have little effect on hunting practices with large game, such as gazelle and ibex, continuing to be exploited and were probably the principle source of meat (Peters et al., 2005). Indeed, in many societies worldwide, meat is laden with social meaning and is often reserved for formal occasions or special ceremonies. The symbolic value of meat from domesticates compared to wild animals varies across populations, with evidence that in some foraging communities meat makes up a small proportion of overall diet (Marciniak, 2005). Full scale widespread herding did not establish itself until 9.5 kyBP in Anato- lia and the Near East ( Vigne, 2011 ), which begs the question of the role of these animals prior to the establishment of domesticate husbandry. The hypothesis of milk as a catalyst for domestication contrasts with the long-held Secondary Prod- ucts Revolution hypothesis, where dairy husbandry evolved later perhaps during the 4th–3rd millennium BC in Mesopotamia and then spread through either transmission of ideas or by physical migration of populations into Europe ( Sherratt, 1981 ). The original hypothesis was based on three lines of evidence: (i) humans’ natural predisposition of lactose intolerance (human genetics); (ii) adoption of pottery shapes appropriate for the manipulation of liquids from the Bronze Age in Europe (material evidence); (iii) early domesticates’ let-down ability and production capacity (ruminant genetics).

Humans and lactose intolerance The latin word “lac ” for milk was chosen to name the sugar contained in milk: lactose. Indeed, most mammalian milks contain lactose, with the exceptions of the monotremes (platypus and echidnas) and the pinnipedia (seals, walruses and allies; Fuquay et al., 2011). Lactose is the primary source of energy in milk. This disaccharide sugar is broken down into glucose and galactose by the enzyme lactase present in the consumer’s small intestine. This enzyme is crucial for the young as it enables them to obtain the nutritious simple sugars from the disaccharide lactose. However, in mammals, including most humans, the enzyme activity is down-regulated after weaning, leading to lactase non-persistence or lactose intolerance . The consumption of lactose-rich foodstuffs by lactose-intolerant individuals can induce abdominal discomfort, flatulence and diarrhoea (Flatz 130 Mélanie Roffet-Salque et al. and Rotthauwe, 1977 ). More than two-thirds of the population worldwide do not express the enzyme lactase in adulthood and are thus unable to digest lactose ( Itan et al ., 2010 ). The rest of the human population is able to express the enzyme lactase in adulthood, making them lactase persistent (LP) or lactose tolerant . In Europe, the ability to digest lactose in adulthood is related to a single mutation (−13,910*T) on one gene (LCT). In contrast, five different alleles (−13,907*G, 13,910*T, −13,915*G, −14,009*G and −14,010*C), including the one observed in Europe, are detected in lactase-persistent Africans. Genetic studies have demonstrated that the European LP allele (−13,910*T) only appeared in European populations very recently in the last 2–7 to 12–20 kyBP. Current frequency of the LP allele in modern populations is extremely high ( Gerbault et al ., 2013 for a review) and genetic drift alone is unlikely to explain the rapid increase in frequency of this allele in few millennia. A high natural selection must thus be invoked with genetic studies showing that a selection strength between 0.8% and up to 19% is needed to explain the distribution of the −13,910*T allele. Such positive natural selection is one of the highest ever seen in the human genome in the last 30,000 years, and may even have fluctuated over time and space. LP is distributed unevenly worldwide ( Itan et al ., 2010 ) with populations with a history of dairying having higher frequencies of LP (Simoons, 1970; Holden and Mace, 1997). The human genome has thus been shaped by the introduction of milk and dairy foods into the diet. This genetic adaptation to a dietary shift is described as gene-culture co-evolution ( Holden and Mace, 1997 ). The consumption of lactose-rich products must have exerted a high selective advantage since the appearance of the −13,910*T allele in the human genome. Several hypotheses have been formulated to explain the evolution of lactose tolerance (McCracken, 1971 ; Holden and Mace, 1997 ), with the calcium assimilation hypothesis being often invoked for the European populations. Calcium assimilation in the gut is promoted by vitamin D sourced from the diet or synthesised in the skin from its precursor 7-dehydrocholesterol and UV-B light ( Engelsen, 2010 ). Communities in the high latitudes may have been deficient in vitamin D and therefore milk would have provided a source of calcium and vitamin D, preventing farmers from developing rickets induced by high cereal consumption (Cordain, 1999). This is evident when Mesolithic hunter-gatherer communities in high latitudes shifted from aquatic resources (particularly rich in vitamin D) to resources obtained from domesticated plants and animals in the Neolithic ( Cramp et al ., 2014 for the UK). Individuals living in farming communities with access to milk would have benefitted from its many advantages, such as its high nutritional value ( Simoons, 1970 ; McCracken, 1971 ) and non-LP individuals could have avoided the lactose intolerance symptoms by milk processing techniques. Indeed, although fresh milk from cattle, sheep and goats contains high levels of lactose (up to 5 g per 100 g; Table 7.1 ), the manufacture of dairy products can lower the lactose content, making them more digestible by lactose-intolerant individuals. Curdling the milk and straining the whey from the curds separates the lactose-rich whey (lactose is water-soluble) from the lactose-poor curds. Con- sequently, cheeses such as camembert, roquefort and cheddar contain only traces Milk as a pivotal medium in domestication 131 of lactose, and parmesan less than 1 g per 100 g (Food Standards Agency, 2002). Cream also has a lower concentration of lactose than milk (2.2 g per 100 g of cream vs. up to 5 g per 100 g of milk; Food Standards Agency, 2002 ) as it is obtained by letting milk stand for a few days to allow the fat-rich cream to float on the surface of skimmed milk containing the water-soluble lactose. The manufacture of butter from cream by churning also reduces the lactose content (0.6 g per 100 g) by removing the buttermilk (5.0 g per 100 g; Food Standards Agency, 2002). The manufacture of yoghurt from milk involves the fermentative conversion of lactose into lactic acid by bacteria, thereby reducing the lactose concentration of yoghurt compared to milk. However, yoghurt still contains relatively high concentration of lactose (less than 5 g per 100 g; Food Standards Agency, 2002 ). Milk and dairy products from sheep, goats, cattle and horses form the corner- stone of modern pastoral societies’ subsistence practices, such as those of the Eurasian steppe. Those populations practice extensive pastoral systems and raise animals primarily for their milk and meat with little or no supplementary agricultural activity. Milk is commonly processed into fermented liquids, i.e. kumys or airag (fermented mare milk), and dried products, particularly aaruul (dried milk curd). Winter and spring can be harsh on the steppe and these products, produced throughout the lactation period, are stored for the demanding coming months, extending the shelf life of milk ( Sadler et al ., 2010 ).

Detecting milk use in prehistory: organic residues in pottery vessels Ceramic containers appear in the Near East after c. 9 kyBP, in a region spanning from Central Anatolia and Upper Mesopotamia ( Le Mière and Picon, 1998 ). The introduction of ceramics allowed the newly settled agricultural communities to improve their storage facilities and provide more efficient ways of processing food ( Redman, 1978 ; Moore, 1995 ). Cooking increases the palatability and digestibility of otherwise toxic and/or inedible foodstuffs ( Dunne et al ., 2016 ). A portion of lipids (or fats) contained in foodstuffs are readily absorbed in the clay pores within the vessel walls and preserved from microbial degradation for thousands of years (Evershed, 2008b). Ceramic sherds excavated at archaeological sites and the lipids preserved therein are, thus, considered to reflect the variety of foodstuffs processed in the pots, recording the subsistence strategies as practiced by early farmers (Roffet-Salque et al., in press). Lipids trapped in the clay pores of pottery sherds are extracted using organic solvents and characterised using state-of-the-art analytical methods (Evershed et al ., 1990 ; Charters et al ., 1993; Dudd and Evershed, 1998; Correa-Ascencio and Evershed, 2014). Com- pounds are separated using chromatography, identified using mass spectrometry and further characterised using isotope ratio mass spectrometry. These analytical methods allow both culinary and technical uses of organic materials to be investigated (Roffet-Salque et al., in press). Beeswax (e.g. Roffet-Salque et al ., 2015 ), resins and tars (e.g. Regert et al., 2008), plant waxes (e.g. Dunne et al ., 2016 ) and animal fats (e.g. Dudd and Evershed, 1998) have characteristic molecular 132 Mélanie Roffet-Salque et al. composition, and thus can be identified based on the detection of “biomarkers”, biomolecular proxies, for the processing a wide range of animal and plant products (Evershed, 2008b). Animal fats are usually too degraded to identify their source using their molecular composition only, and thus compound-specific isotopic analyses are performed on fatty acids to identify whether the fats originate from non-ruminant or ruminant animals, and carcass or milk (Dudd and Evershed, 1998 ; Copley et al ., 2003 ). The advent of organic residue analyses of lipids preserved in archaeological pottery vessels and the possibility of identifying dairy fats, have provided new avenues of investigation regarding the use of early pottery vessels and the emergence of dairying in early farming communities. The extensive study of more than 2,200 pottery sherds from Southeastern Europe, the Near East and the Levant provided direct evidence for early milk use. Archaeological sites around the Sea of Marmara showed intensive dairy practices from 8.5 kyBP, with more than 70% of the residues detected in the sherds investigated in the study revealing traces of milk residues, with the faunal remains pointing to a dominance of cattle herding ( Evershed et al ., 2008 ). Investigations of pottery containers coupled with osteological analyses of domesticates demonstrated that dairying was part of the subsistence economy at early farming sites along the Northern Mediterranean coast ( Debono Spiteri et al ., 2016 ). Finally, the onset of the Neolithic in the UK was characterised by a dramatic shift in subsistence practices, with the abandonment of aquatic resources for animal products obtained from domesticated animals, particularly milk (Richards et al ., 2003 ; Cramp et al., 2014). The nutritional quality of milk clearly contributed to the success of the farming communities living in such marginal extreme environments ( Cramp et al ., 2014 ). Milk was also adopted in the Baltic region at the same time as farming practices, although fishing, hunting and gathering activities continued alongside agriculture (Craig et al ., 2011 ). In this region, milk thus complemented existing staple foods rather than replacing them. The adoption of milk in the Neolithic as seen through the lens of ceramic vessels is widespread across Europe (Figure 7.1). The detection of dairy lipids in pots is usually interpreted as evidence for milk processing, as heating up foodstuffs promotes the absorption of lipids in the clay walls of vessels (Evershed, 2008a ). However, the detection of milk residues in ceramic containers, such as pots, pans and dishes, rarely allows explicit definition of the exact processing practices, except when typological similarities between modern-day dairy utensils and archaeological vessels provide credence for the use of such ceramic vessels in milk processing. Indeed, Copper Age vessels from Hungary were suggested to be “milk jugs” although lipid residue analyses on eight such vessels failed to prove that they were used for storing, processing or serving milk (Craig et al ., 2003 ). Perforated vessels excavated in Neolithic Central Europe were hypothesised to be cheese strainers due to their similarity to modern-day and ethnographic cheese strainers (Bogucki, 1984). These bowl- and funnel-like perforated vessels are found in sites from the Neolithic Linear Pottery culture (LBK) in Central Europe. Bogucki (1984) also noticed that cattle herding strategies at these sites reflect Figure 7.1 Proportion of animal fat residues identified as milk fats (white, Δ13 C ≤ −3.1‰), ruminant and non-ruminant adipose fats (grey, Δ13 C > −3.1‰) and aquatic fats (black, aquatic biomarkers detected) in archaeological sherds from the Neolithic in Europe and the Near East. Numbers (n) represent the number of residues identified as animal fats for each group of sites. a. Copley et al. (2003); b. Craig et al. (2011); c. Cramp et al. (2014); d. Debono Spiteri et al. (2016); e. Evershed et al. (2008); f. Gregg et al . (2009 ); g. Matlova et al . (2017 ); h. Salque et al . (2013 ); i. Smyth and Evershed (2016 ); j. Šoberl et al . (2008). 134 Mélanie Roffet-Salque et al. dairy subsistence practices. Recent analyses of 50 sieve fragments from the region of Kuyavia (north-central Poland) showed that milk lipids were present in most of the sieves (Salque et al., 2013), supporting Bogucki’s hypothesis. The idea is that the sieves were used to separate the curds from the whey to make (cow’s- milk) cheese as early as 7.2 kyBP, although the geographical and chronological extent of this practice is still largely unknown. Cheese-making provides digestible dairy products for lactose-intolerant early farming communities and extends the shelf life of milk, allowing the nutritional benefits of milk to be available outside of the milking season ( Salque et al ., 2013 ).

Detecting the evolution of dairy husbandry in the faunal record Understanding the mechanisms behind milk production, different modes of dairy husbandry and the factors that influence the decisions of farmers are crucial in tracking dairy husbandry amongst early European Neolithic farmers ( Gillis, 2017 ). A longstanding method for identifying dairy husbandry in archaeozoological assemblages is by examining the animal remains (mainly dental remains) and determining the age at death for individuals. These data can be used to construct mortality profiles, which are compared with hypothetical models of production ( Payne, 1973 ; Helmer et al ., 2005b ). The production of milk is dictated by the ability of animals to let-down their milk, and this varies between species and breed types. Indeed, the ejection of milk to the teat cisterns is stimulated by the hormone oxytocin, and the production of the hormone by the sensory and tactile stimulus of the young or by artificial means. Milk is stored temporarily in the alveolar units and then actively transferred in the teat cisterns by the ejection reflex induced by oxytocin. The proportion of cisternal milk (available without ejection reflex) is higher in caprines than cattle, particularly in goats. Conse- quently, the need for the offspring to stimulate the production of oxytocin is less important for small stock (review by Balasse, 2003). Ethnographic examples have unsurprisingly shown that lambs and kids can be removed without affecting the flow of milk ( Halstead, 1998 ). The goat’s long lactation and high milk production, together with their tolerance to harsh conditions, ideal in marginal environments, make them known as the “poor man’s cow” (Merrill and Taylor 1976: 74). The age of slaughter is also related to the demands of market or family consumption, regional traditions and the replacement needs of the herd ( Hal- stead, 1998 ; Boyazoglu and Morand-Fehr, 2001 ). For example, for caprines, the young tender meat (0–3 months) is a delicacy in the modern-day Mediterranean, whereas in Middle Eastern and African societies, older heavier lambs are prized (Boyazoglu and Morand-Fehr, 2001). If sheep milk is highly valued, then the lamb will be removed early to ensure maximum production for human consumption and not shared with the infant. In contrast for cattle, the let-down of milk without the infant may inhibit production, which is particularly evident in traditional breeds of cattle (Clutton-Brock, 1981 ). This was one of the lines of evidence used to support Sherratt’s Second- ary Products Revolution hypothesis (Sherratt, 1981). However, many traditional Milk as a pivotal medium in domestication 135 societies practice milk sharing, where the infant animal is allowed to suckle prior to removing the milk for human consumption. These practices, beautifully represented on Saharan rock art iconography (Le Quellec, 2011: Figure 7.2), are applied to all dairy ruminant and non-ruminant species, including horses. Milk sharing between the young and the herder has been proposed as being distinctive for cattle dairy husbandry in archaeological contexts, where such practices would be translated into high proportion of calves being slaughtered at the end of the lactation period ( Peske, 1994 : Figure 7.3a). When milk is the primary objective, an intensive cattle dairy model is where an increased infant slaughter prior to 6 months is expected, particularly of unwanted male calves ( Legge, 1981 ; Figure 7.3b). Such a slaughter strategy may inhibit milk let-down (Balasse, 2003); however, there is extensive historical and prehistoric evidence for the stimulation of cattle milk let-down when the calf is absent (Ireland: Lucas, 1989; sub-Saharan Africa: Le Quellec, 2011 ). Therefore, it is conceivable that ingenuous tactics, such as blow- ing into cows’ vaginas and replacement of the calf with a dummy, could have been employed by early prehistoric farmers. Cows are also less likely to need to the calf to stimulate milk ejection after the first parturition. Furthermore, allomater- nal nursing, where the cow will allow foreign calves to suckle, is more common after the first parturition ( Le Neindre, 1989 ). This ability to adopt calves to free younger lactating females was recognised by Columella (1745: VI: XXIV), where he describes the small Altinian cattle from the land of Cevas (region now Switzer- land) as having great quantities of milk and had the disposition to allow foreign calves to suckle.

Figure 7.2 Detail of a milking scene from rock art in a rockshelter at Tasigmet, Oued Djerrat (Tassili-n-Ajjer). Note that the calf is present during milking and probably tied beside its dam. Source : Drawing from Balasse et al . (2000 ). 136 Mélanie Roffet-Salque et al.

Figure 7.3 Hypothetical cattle kill-off profiles showing (a) a post-lactation slaughtering and (b) an increased calf slaughter before 6 months, where milk production could have been maintained using artificial means or sharing of calves between multiple females. Source : Method based on Gerbault et al . (2016 ).

The evolution of dairying in Europe Both lipid residue analyses of ceramic containers for demonstrating milk use, and archaeozoological studies for reconstructing herding practices, have been used extensively to detect dairying practices in the past. Studies of early archaeozoological assemblages indicate that shared milking and intensive slaughtering was practiced during the Neolithic. This selection of animals for specific roles was probably based on their individual productive capacities and physiologies. The choice of dairy animal appears to be governed in part by environmental constraints. Stud- ies have shown that cattle prefer water-rich environments ( Gander et al ., 2003 ), whereas caprines can survive in areas of aridity like the Mediterranean and the Balkans ( Boyazoglu and Morand-Fehr, 2001 ). Milk exploitation was a component of early Neolithic farming across most regions of Europe ( Helmer and Vigne, 2004 ; Helmer et al ., 2007 ; Vigne and Helmer, 2007 ; Evershed et al ., 2008 ; Debono Milk as a pivotal medium in domestication 137 Spiteri et al ., 2016 ; Ethier et al ., 2017), with caprines and cattle being the main milk producers along the Mediterranean and in Central Europe, respectively. There is little evidence for dairying in areas of the initial expansion of Neolithic cultures in Greece and Southeastern Europe, although by the time Neolithic culture spread to the Cycladic islands, caprines appear to be managed for milk (Debono Spiteri et al ., 2016 ). There are differences in the way sheep and goats were managed during the Neolithic, with the latter perhaps being used for dairy with evidence of early slaughter of infants, and the former as a source of fast-growing meat in South- ern Europe (Helmer and Vigne, 2004; Greenfield and Arnold, 2015). By 7 kyBP, there is evidence from stable isotopic analysis of age-specific mandible samples for post-lactation slaughter of cattle in France and Romania (Balasse and Tresset, 2002 ; Gillis et al., 2013), which may have developed due to increasing herd sizes. Large herds of cattle are observed in Central Europe, with milk being transformed in cheese using sieves in the LBK site of Ludwinowo ( Salque et al ., 2013 ). Inten- sive slaughtering of calves has been proposed at some early Neolithic sites in Spain ( Gillis et al ., 2016 ) and Southern Europe ( Vigne and Helmer, 2007 ), which may be a deliberate strategy to increase milk supply but may also be a way to allevi- ate stocking pressures caused by extreme weather. Mediterranean livestock were introduced beyond their natural climatic range by farming groups dispersing into the interior of the Balkans by 7.6 kyBP. Climate-driven selective pressures led to a better adaptation of the livestock to their new environment (Ethier et al ., 2017). Husbandry practices had a role in shaping the breeds that exist today, through natural selection of animals adapted to feed scarcity or ability to reduce metabolism in winter (Ethier et al., 2017) and artificial selection of phenotypes with advan- tageous traits, such as milk production and ability to let-down milk with little intervention. Genetic changes linked to milk production were actively selected for during the last century. Indeed, dairy management practices and dairy breeding rapidly evolved after the Second World War to maximize dairy species’ production capacities (Labussière, 1999; Orland, 2003). Specialised sheep breeds now produce double the milk of traditional breeds, whereas dairy goat breeds can produce five times as much milk as traditional breeds ( Haenlein, 2007 ). The most dramatic change in production is seen in cattle where specialised breeds, such as Friesian, produce ten times as much as traditional breeds for twice as long (Gillis, 2017).

Final comments The exploitation of milk from domesticates played an important role in feeding entire communities across Europe, with infants and adults benefitting from the introduction of animal milk into their diet. Milk provided a new range of weaning foods, and the majority of today’s infant formulas are still based on cow’s milk. Cow’s milk was given to infants in Greco-Roman Egypt, as reported in a wet-nursing contract from the period (Fildes, 1986). Ethnographic studies suggest that weaning liquids and solids are generally introduced earlier in food- producing communities compared to hunter-gatherer societies (Sellen and Smay, 2001), although it seems that early weaning is not entirely a consequence of using 138 Mélanie Roffet-Salque et al. animal milk as a weaning food, as a range of weaning foods is also available for hunter-gatherers (Sellen and Smay, 2001). The reduction in breastfeeding duration in farming communities is suggested to be related to the working patterns of women and the constraints of carrying a toddler to distant arable fields ( Fouts et al ., 2005 ; Ghosh et al., 2006). The possible early weaning of babies at the start of the Neolithic would have shortened the period of breastfeeding, thereby reducing intervals between births. Indeed, the Neolithic is well-known to have been a period of increased population growth, attributed to an increased fertility rate followed by an increased mortality rate (Bocquet-Appel and Bar Yosef, 2008). Milk must have provided a stable seasonal resource to early farming communities and increased the quality of the overall diet, probably increasing the life expectancy of Neolithic people ( Vigne, 2008 ). The exploitation of domesticated animals for milk has had an impact on the human genome, with one-third of the world’s population being able to digest the lactose in milk in adulthood. Nowadays, cattle dominate dairy production, particularly in Northern Europe, while sheep and goats play important roles in providing milk and meat for pastoral communities across the Mediterranean basin and the Balkan region. This distribution of dairy species in modern-day Europe was initiated in the Neolithic, where caprines and cattle were already exploited in the geographical areas we see today. The advances of analytical methods are now allowing us to study at least some of the mechanisms underlying the emergence of dairying in prehistory, in particular subsistence and herding practices at the household, site and regional levels, which are shedding light on patterns we see today. Over the last 20 years, the increasing integration of multiple strands of evidence has worked towards drawing a picture of the processes at play for the domestication of ungulates, the emergence of LP in Europe and the role of milk in those processes.

Acknowledgements The European Research Council (Advanced Grant NeoMilk FP7-IDEAS-ERC/ 324202 attributed to R. P. E.) and Christian-Albrechts-Universität are thanked for funding M. R.-S. and R. E. G. We thank the Natural Environment Research Council for partial funding of the mass spectrometry facility at Bristol (R8/H10/63). Charles Stépanoff is acknowledged for comments that greatly improved the manuscript.

References Balasse, M. 2003. Keeping the young alive to stimulate the production of milk? Differ- ences between cattle and small stock. Anthropozoologica , 7, 3–10. Balasse, M. and Tresset, A. 2002. Early weaning of Neolithic domestic cattle (Bercy, France) revealed by intra-tooth variation in nitrogen isotope ratios. Journal of Archaeo- logical Science , 29, 853–859. Balasse, M., et al . 2000. Un abattage ‘post-lactation’ sur des bovins domestiques néo- lithiques. Etude isotopique des restes osseux du site de Bercy (Paris, France). Gestion Milk as a pivotal medium in domestication 139 démographique des animaux à travers le temps, Colloque internatinal de Turin, 16–18 septembre 1998, Ibex 5; Anthropozoologica 31 , 39–48. Bendrey, R. 2014. Care in the community? Interpretations of a fractured goat bone from Neolithic Jarmo, Iraq. International Journal of Paleopathology , 7, 33–37. Bock, B. B., et al . 2007. Farmer’s relationship with different animals: The importance of getting close to the animals – Case studies of French, Swedish and Dutch cattle, pig and poultry farmers. International Journal of Sociology of Agriculture and Food , 15(3), 108–125. Bocquet-Appel, J.-P. and Bar Yosef, O., eds. 2008. The Neolithic demographic transition and its consequences . New York, NY: Springer. Bogucki, P. I. 1984. Ceramic sieves of the Linear Pottery culture and their economic implications. Oxford Journal of Archaeology , 3(1), 15–30. Boyazoglu, J. and Morand-Fehr, P. 2001. Mediterranean dairy sheep and goat products and their quality: A critical review. Small Ruminant Research , 40(1), 1–11. Cordain, L. 1999. Cereal grains: humanity’s double-edged sword. In: Simopoulos, A. P. ed. Evolutionary aspects of nutrition and health. Diet, exercise, genetics and chronic disease . Basel, Karger: World Review of Nutrition and Dietetics, 84, pp 19–73. Charters, S., et al. 1993. Quantification and distribution of lipid in archaeological ceramics: Implications for sampling potsherds for organic residue analysis and the classification of vessel use. Archaeometry , 35(2), 211–223. Clutton-Brock, J. 1981. Contributions to discussion. In: Mercer, R. J. ed. Farming practice in British Prehistory . Edinburgh: Edinburgh University Press, 218–220. Columella, L. J. M. 1745. L. Junius Moderatus Columella of husbandry: In twelve books: And his book concerning trees; Translated into english, with illustrations from Pliny, Cato, Varro, Palladius and other ancient and modern authors . London: A. Millar. Copley, M. S., et al . 2003. Direct chemical evidence for widespread dairying in prehistoric Britain. Proceedings of the National Academy of Sciences of the United States of America , 100(4), 1524–1529. Correa-Ascencio, M. and Evershed, R. P. 2014. High throughput screening of organic residues in archaeological potsherds using direct methanolic acid extraction. Analytical Methods , 6, 1330–1340. Craig, O. E., et al . 2003. ‘Milk jugs’ and other myths of the Copper Age of Central Europe. European Journal of Archaeology , 6(3), 251–265. Craig, O. E., et al. 2011. Ancient lipids reveal continuity in culinary practices across the transition to agriculture in Northern Europe. Proceedings of the National Academy of Sciences of the United States of America , 108(44), 17910–17915. Cramp, L. J. E., et al . 2014. Immediate replacement of fishing with dairying by the earliest farmers of the northeast Atlantic archipelagos. Proceedings of the Royal Society B , 281, 20132372. Debono Spiteri, C., et al . 2016. Regional asynchronicity in dairy production and processing in early farming communities of the northern Mediterranean. Proceedings of the National Academy of Sciences of the United States of America , 113(48), 13594–13599. Dudd, S. N. and Evershed, R. P. 1998. Direct demonstration of milk as an element of archaeological economies. Science , 282, 1478–1481. Dunne, J., et al . 2016. Earliest direct evidence of plant processing in prehistoric Saharan pottery. Nature Plants , 3, 16194. Engelsen, O. 2010. The relationship between ultraviolet radiation exposure and vitamin D status. Nutrients , 2(5), 482–495. 140 Mélanie Roffet-Salque et al. Ethier, J., Bánffy, E., Vuković, J., Leshtakov, K., Bacvarov, K., Roffet-Salque, M., Ever- shed, R. P. and Ivanova, M. 2017. Earliest expansion of animal husbandry beyond the Mediterranean zone in the sixth millennium BC . Scientific Reports , 7 (1), 7146. Evershed, R. P. 2008a. Experimental approaches to the interpretation of absorbed organic residues in archaeological ceramics. World Archaeology , 40(1), 26–47. Evershed, R. P. 2008b. Organic residue analysis in archaeology: The archaeological bio- marker revolution. Archaeometry , 50(6), 895–924. Evershed, R. P., Heron, C. and Goad, L. J. 1990. Analysis of organic residues of archaeological origin by high-temperature gas chromatography and gas chromatography-mass spectrometry. The Analyst , 115, 1339–1342. Evershed, R. P., et al . 2008. Earliest date for milk use in the Near East and southeastern Europe linked to cattle herding. Nature , 455, 528–531. Fildes, V. A. 1986. Breasts, bottles and babies: A history of infant feeding . Edinburgh: Edinburgh University Press. Fischler, C. 1988. Food, self and identity. Social Science Information , 27(2), 275–292. Flatz, G. and Rotthauwe, H. W. 1977. The human lactase polymorphism: Physiology and genetics of lactose absorption and malabsorption. Progress in Medical Genetics , 2, 205–249. Food Standards Agency 2002. McCance and Widdowson’s the composition of foods . Cam- bridge: Royal Society of Chemistry. Fouts, H. N., Hewlett, B. S. and Lamb, M. E. 2005. Parent-offspring weaning conflicts among the Bofi farmers and foragers of Central Africa. Current Anthropology , 46(1), 29–50. Fuquay, J. W., Fox, P. F. and McSweeney, P. L. H. 2011. Encyclopedia of dairy sciences . 2nd edition. London: Elsevier Ltd. Academic Press . Gander, A., et al. 2003. Habitat use by Scottish Highland cattle in a lakeshore wetland. Bulletin of the Geobotanical Institute ETH , 69, 3–16. Gerbault, P., et al. 2013. How long have adult humans been consuming milk? IUBMB Life , 65, 983–990. Gerbault, P., et al. 2016. Statistically robust representation and comparison of mortality profiles in archaeozoology. Journal of Archaeological Science , 71, 24–32. Gerosa, S. and Skoet, J. 2013. Milk availability: Current production and demand and medium-term outlook. In: Muehlhoff, E., Bennett, A. and McMahon, D. eds. Milk and dairy products in human nutrition . Rome: Food and agriculture organization of the United Nations, 11–37. Ghosh, R., Mascie-Taylor, C. G. N. and Rosetta, L. 2006. Longitudinal study of the frequency and duration of breastfeeding in rural Bangladeshi women. American Journal of Human Biology , 18(5), 630–638. Gillis, R. E. 2017. A calf’s eye view of milk production: Tony Legge’s contribution to dairy husbandry studies. In: Rowley-Conwy, P., Serjeantson, D. and Halstead, P. eds. Eco- nomic zooarchaeozoology: Studies in hunting, herding and early agriculture . Oxford: Oxbow Books, 135–142. Gillis, R. E., et al . 2013. Sophisticated cattle dairy husbandry at Borduşani-Popină (Roma- nia, fifth millennium BC ): The evidence from complementary analysis of mortality profiles and stable isotopes. World Archaeology , 45(3), 447–472. Gillis, R. E., et al. 2016. Neonatal mortality, young calf slaughter and milk production during the early Neolithic of north western Mediterranean. International Journal of Osteo- archaeology , 26(2), 303–313. Gray, P. B. and Young, S. M. 2011. Human – Pet dynamics in cross-cultural perspective. Anthrozoös , 24(1), 17–30. Milk as a pivotal medium in domestication 141 Greenfield, H. J. and Arnold, E. R. 2015. ‘Go(a)t milk?’ New perspectives on the zooarchaeological evidence for the earliest intensification of dairying in south eastern Europe. World Archaeology , 1–27. Gregg, M. W., et al . 2009. Subsistence practices and pottery use in Neolithic Jor- dan: Molecular and isotopic evidence. Journal of Archaeological Science, 36(4), 937–946. Haenlein, G. F. W. 2007. About the evolution of goat and sheep milk production. Small Ruminant Research , 68(1–2), 3–6. Halstead, P. 1998. Mortality models and milking: Problems of uniformitarianism, optimal- ity and equifinality reconsidered. Anthropozoologica , 27, 3–20. Helmer, D. 1994. La domestication des animaux d’embouche dans le Levant Nord (Syrie et Sinjar), du milieu du 9e millénaire BP à la fin du 7e millénaire BP. Nouvelles données d’après les fouilles récentes. Anthropozoologica , 20, 41–54. Helmer, D., et al. 2005a. Identifying early domestic cattle from Pre-Pottery Neolithic sites on the Middle Euphrates using sexual dimorphism. In: Vigne, J.-D., Peters, J. and Helmer, D. eds. Proceedings of the 9th conference of the international council of archaeozoology 2002, first steps of animal domestication, new archaeozoological approaches . Oxford: Oxbow, 86–95. Helmer, D., et al . 2005b. L’élevage des caprinés néolithiques dans le sud-est de la France: saisonnalité des abattages, relations entre grottes-bergeries et sites de plein-air. Anthro- pozoologica , 40(1), 167–189. Helmer, D., Gourichon, L. and Vila, E. 2007. The development of the exploitation of products from Capra and Ovis (meat, milk and fleece) from the PPNB to the Early Bronze in the northern Near East (8700 to 2000 BC cal.). Anthropozoologica , 42(2), 41–69. Helmer, D. and Vigne, J. D. 2004. La gestion des cheptels de caprinés au néolithique dans le midi de la France. In: Française, S. P. ed. Approches fonctionnelles en Préhistoire – Actes XXV ème congrès Préhistorique de France, Nanterre, 24–26 novembre 2000 , 397–407. Hemmer, H., ed. 1990. Domestication: The decline of environmental appreciation, Cam- bridge: Cambridge University Press. Holden, C. and Mace, R. 1997. Phylogenetic analysis of the evolution of lactose digestion in adults. Human Biology , 69(5), 605–628. Horwitz, L. K., et al. 1999. Animal domestication in the Southern Levant. Paléorient , 25(2), 63–80. Itan, Y., et al. 2010. A worldwide correlation of lactase persistence phenotype and genotypes. BMC Evolutionary Biology , 10(1), 36. Labussière, J. 1999. The physiology of milk ejection: Consequences on milking techniques. In: Martinet, J., Houdebine, L.-M. and Head, H. H. eds. Biology of lactation . Paris: INRA, 307–333. Legge, A. J. 1981. Aspects of cattle husbandry. In: Mercer, R. J. ed. Farming practice in British Prehistory . Edinburgh: Edinburgh University Press, 169–181. Le Mière, M. and Picon, M. 1998. Les débuts de la céramique au Proche-Orient. Paléori- ent , 5–26. Le Neindre, P. 1989. Influence of cattle rearing conditions and breed on social relationships of mother and young. Applied Animal Behaviour Science , 23(1), 117–127. Le Quellec, J.-L. 2011. Provoking lactation by the insufflation technique as documented by the rock images of the Sahara. Anthropozoologica , 46(1), 65–125. Leung, A. K. C. and Sauve, R. S. 2003. Whole cow’s milk in infancy. Paediatrics and Child Health , 8(7), 419–421. Lucas, A. T. 1989. Cattle in ancient Ireland . Kilkenny: Boethius Press. 142 Mélanie Roffet-Salque et al. Marciniak, A. 2005. Placing animals in the Neolithic: Social zooarchaeology of Prehis- toric farming communities . London: UCL Press. Matlova, V., et al . 2017. Defining pottery use and animal management at the Neolithic site of Bylany (Czech Republic). Journal of Archaeological Science: Reports , 14, 262–274. McClellan, H. L., Miller, S. J. and Hartmann, P. E. 2008. Evolution of lactation: Nutri- tion v. protection with special reference to five mammalian species. Nutrition Research Reviews , 21, 97–116. McCracken, R. D. 1971. Lactase deficiency: An example of dietary evolution. Current Anthropology , 12(4/5), 479–517. Merrill, L. B. and Taylor, C. A., 1976. Take note of the versatile goat. Rangeman's Journal , 3(3), 74–76. Milliet, J. 2007. L’allaitement des animaux par des femmes, entre mythe et réalité. In: Dou- nias, E., Motte-Florac, E. and Dunham, M. eds. Le symbolisme des animaux. L’animal, clef de voûte de la relation entre l’homme et la nature? Paris: IRD Editions, 881–911. Moore, A. M. T. 1995. The inception of potting in western Asia and its impact on economy and society. In: Barnett, W. K. and Hoopes, J. W. eds. The emergence of pottery: Tech- nology and innovation in ancient societies . Washington: Smithsonian Institution Press, 39–54. Orland, B. 2003. Turbo-cows: Producing a competitive animal in the nineteenth and early twentieth centuries. In: Schrepfer, S. and Scranton, P. eds. Industrializing organisms: Introducing evolutionary history . New York/London: Routledge, 167–189. Payne, S. 1973. Kill-off patterns in sheep and goats: The mandibles from Aşvan Kale. Anatolian Studies , 23, 281–303. Peske, L. 1994. Příspěvek k poznání počátku dojení skotu v pravěku (Contribution to the beginning of milking in Prehistory). Archeologické Rozhledy , 46, 97–104. Peters, J., von den Driesch, A. and Helmer, D. 2005. The upper Euphrates-Tigris basin: Cradle of agro-pastoralism? In: Vigne, J. D., Peters, J. and Helmer, D. eds. Proceedings of the 9th conference of the international council of archaeozoology 2002, Durham, first steps of animal domestication, new archaeozoological approaches . Oxford: Oxbow Books, 96–124. Porcher, J. and Schmitt, T. 2012. Dairy cows: Workers in the shadows? Society and Ani- mals , 20(1), 39–60. Redman, C. L. 1978. The rise of civilization: From early farmers to urban society in the ancient Near East . San Francisco: Freeman and Co. Regert, M., et al. 2008. Reconstructing ancient Yemeni commercial routes during the Mid- dle Ages using a structural characterization of terpenoid resins. Archaeometry , 50(4), 668–695. Richards, M. P., Schulting, R. J. and Hedges, R. E. M. 2003. Archaeology: Sharp shift in diet at onset of Neolithic. Nature , 425(6956), 366. Roffet-Salque, M., et al. 2015. Widespread exploitation of the honeybee by early Neolithic farmers. Nature , 527, 226–230. Roffet-Salque, M., et al. 2017. From the inside out: Upscaling organic residue analyses of archaeological ceramics. Journal of Archaeological Science: Reports, 16, 627–640 . Sadler , K., et al . 2010. The fat and the lean: Review of production and use of milk by pastoralists. Pastoralism , 1(2), 291–324. Salque, M., et al . 2013. Earliest evidence for cheese making in the sixth millennium BC in northern Europe. Nature , 493, 522–525. Sellen, D. W. and Smay, D. B. 2001. Relationship between subsistence and age at weaning in ‘preindustrial’ societies. Human Nature , 12(1), 47–87. Milk as a pivotal medium in domestication 143 Sherratt, A. 1981. Plough and pastoralism: Aspects of the secondary products revolution. In: Hodder, I., Isaac, G. and Hammond, N. eds. Pattern of the past: Studies in honour of David Clarke . Cambridge: Cambridge University Press, 261–305. Simoons, F. 1970. Primary adult lactose intolerance and the milking habit: A problem in biologic and cultural interrelations – II. A culture historical hypothesis. Digestive Dis- eases and Sciences , 15(8), 695–710. Smyth, J. and Evershed, R. P. 2016. Milking the megafauna: Using organic residue analysis to understand early farming practice. Environmental Archaeology , 21(3), 214–229. Šoberl, L., et al. 2008. Early herding practices revealed through organic residue analysis of pottery from the early Neolithic rock shelter of Mala Triglavca, Slovenia. Documenta Praehistorica XXXV , 253–260. Vigne, J. D. 2008. Zooarchaeological aspects of the Neolithic diet transition in the Near East and Europe, and their putative relationships with the Neolithic demographic transition. In: Bocquet Appel, J.-P. and Bar-Yosef, O. eds. The Neolithic demographic transition and its consequences . New York: Springer Verlag, 179–205. Vigne, J.-D. 2011. The origins of animal domestication and husbandry: A major change in the history of humanity and the biosphere. Comptes Rendus Biologies, 334(3), 171–181. Vigne, J.-D. and Helmer, D. 2007. Was milk a ‘secondary product’ in the Old World Neo- lithisation process? Its role in the domestication of cattle, sheep and goats. Anthropo- zoologica , 42(2), 9–40. Vigne, J.-D., et al . 2011. The early process of mammal domestication in the Near East: New evidence from the pre-Neolithic and pre-Pottery Neolithic in Cyprus. Current Anthropology , 52(S4), S255–S271. Wijesinha-Bettoni, R. and Burlingame, B. 2013. Milk and dairy product composition. In: Muehlhoff, E., Bennett, A. and McMahon, D. eds. Milk and dairy products in human nutrition . Rome: Food and agriculture organization of the United Nations, 41–102. Zeder, M. A. 2005. A view from the Zagros: New perspectives on livestock domestication in the Fertile Crescent. In: Vigne, J. D., Peters, J. and Helmer, D. eds. Proceedings of the 9th conference of the international council of archaeozoology 2002, Durham, first steps of animal domestication, new archaeozoological approaches. Oxford: Oxbow Books, 125–146.

8 Watching the horses The impact of horses on early pastoralists’ sociality and political ethos in Inner Asia

Gala Argent

When I adopted Sabre, I thought he was a lot like me. He was confident, brave, charismatic and proud. Eleven years later, I looked into his eyes and realized I was wrong. Sabre wasn’t like me after all – I had always wanted to be like him. Nevyn O’Kane, speaking of his dog1

Over the past several decades, the anthropocentric lens through which domestication has been seen as entirely human-directed has shifted to a more ecocentric view “where domestication appears as a relation between two species” ( Noske 1997:10; see also Boyd 2017). As this volume brings into perspective, definitions of domestication as a process wherein human subjects act upon animal objects have broadened to allow for the notion of coevolution and reciprocal adaptations between and among humans, co-domesticates and the habitats they share. We now recognize that nonhuman animals (hereafter: animals) possess agential qualities that influence humans, and that the process of domestication for any given species can neither be considered as a static point in time, nor as a fixed set of characteristics. This significant paradigm shift allows for – indeed, necessitates – a reassessment of prior archaeological interpretations which have assumed interspecies relations as unidirectional. In what follows, I suggest that for one such human-horse configuration, the Iron Age Pazyryk archaeological culture, influence between the species was bi-directional and that horses influenced not only human social structures, but also political ethos – that humans mimicked horses’ ways of being. The earliest evidence for equine domestication currently points to Kazakhstan some 5,500 years ago, where chemical signatures in pottery suggest that horse milk played a role in the motivations for horse domestication ( Outram et al. 2009 ). Over the next several millennia, the people of Inner Asia refined the human-horse relationship in ways that fostered new social, economic and political structures. As the Bronze Age shifted to the Iron Age around 3,000 years ago, this transition is thought to have included the completion of the nascent shift from pastoral-agricultural economies to fully transhumant pastoralism, a refinement of horse riding technology and the emergence of mounted warfare, with the first development of militaristic state society (see, e.g., Anthony 2007 ; Hildinger 1997 ). 146 Gala Argent At the eastern edge of this vast sociocultural landscape – not on the steppe, but in high mountain plateaus of the Altai Mountains spanning present-day southern Siberia, Mongolia, China and Kazakhstan – lived the people of the archaeological culture we call the Pazyryk (6th–3rd centuries BC E). We know of them not through their writings – they left none – but through the accounts of Greek and Chinese historians situated on the steppe’s edges. We know perhaps more about them through their burial mounds, dotted across the landscape, which archaeologists began excavating in the late 19th century and the early 20th century (Radloff 1884; Gryaznov 1950; Rudenko 1970[1953]). The graves include implements of daily living, wooden furniture, food for the journey to the next world, and the well-preserved bodies and clothing of those buried, all preserved in permafrost. We learn from these burials that the Pazyryk people were foragers, hunters and fishers, but primarily pastoralists – herders who moved with their animals up and down the mountains sharing time between winter and summer camps where stock could be grazed. We can assume that horses were crucial to all of these activities. So important to these people were the horses that they buried them with their dead. Horses with saddles, bridles and elaborate costumes – each decorated differently – were sacrificed and interred with their humans in Pazyryk burial mounds. Even small graves held one horse, and the larger graves contained up to 22 ( Figure 8.1 ). Across the broader region, the appearance in burials of bridled horses and weaponry has been argued to indicate “the formation of a social strata of mounted warriors within society” ( Bokovenko 1995b :266), and a “warrior or aristocratic elite” ( Chang et al. 2007 :32) who subscribed to a “militaristic lifestyle” based upon a “warrior ethos” (Hanks 2002:187, iii, emphasis in original). Based upon material remains, archaeologists identified a very broad ethnocultural zone – from the Scythians in the west to the Pazyryk in the east – with assumed shared cultural traits and beliefs, carrying these traits and beliefs into interpretations of Pazyryk culture. The Pazyryk were “cruel nomads” (Van Noten and Polosmak 1995:76) and “militarists in constant combat, one group with another” (Lamberg-Karlovsky 1998 :n.p.). The studies of these “early nomads” coming to these conclusions have operated from the anthropocentric base of conventional archaeological inquiry, reflecting a sense of horses’ separation from human social and cultural life. This position is at odds with the actual types of interpersonal and social associations that can and do develop between humans and horses who live closely with one another. Further- more, it leaves no room for the fluid and reciprocal agencies that may have been at work within these hybrid communities made up of individuals of both social species working at learning how to work together. Privileging hierarchy, domination and control over care, concern and cooperation is well entrenched in academic disciplines situated within Euro-American scholar- ship. The belief that humans engage with the world in an aggressive fashion comes into play in the present discussion in (at least) two ways. First, it is a means of projecting back in time current hierarchical social structures and political concerns that may or may not have existed, where such historical anachronism can promote Watching the horses 147

Figure 8.1 Reconstruction of the Berel 11 burial mound. Source : After Samashev et al. (2000 :13). interpretive bias. Second, it upholds the anthropocentric objectification of animals as subalterns within human worlds. As I shall show in both cases, a paradigm based upon the assumption of control, aggression and domination – whether aimed at other animals or at humans – is not the only way to view these interactions. In this chapter, I argue that archaeological interpretations of the Inner Asian early nomads as “fierce warriors” are reflective of both this linear tracking of interspecies agency and influence, and of current political focuses. From my position as a lifelong equestrienne and horse breeder, I utilize an inside auto- ethnographic strategy to reevaluate archaeological data gathered in fieldwork in the Russian Federation. I do so through a query of how the smaller-scale mutual interpersonal adaptations that occur between humans and horses – and are necessitated to large measure by horses’ ways of being, as shown through ethological studies – might have influenced larger-scale social arrangements within these 148 Gala Argent societies refining equine domestication. Considering the potential of human-horse relationships as bidirectional, I apply to the Pazyryk material an ethno-ethological strategy which allows for viewing these interactions as an “interactive relational system that links humans and non-humans . . . [granting] all living beings the status of relational beings, that is, agents interacting on the phenomenon of ‘culture’ that was hitherto reserved for human beings” (Lestel et al. 2006:168). In other words, about these people who lived with and on their horses, I ask: How might horses’ ways of being have contributed to human political ethos and social structures in this time and place?

Dominance, hierarchy and other ideological frameworks Despite current awareness of the need to consider other cultures as developing and existing outside of our own constructs, the focus on hierarchy, power and domination is pervasive in Euro-American academic thought regarding both human-human and human-animal interactions – as evidenced by the previous interpretations of Pazyryk culture. The belief that human social life is structured hierarchically implies a vertical linearity, with value judgments attached. There is an up and a down, a better and a worse, a dominant and a submissive. Similarly, social evolutionary theory posits a stadial categorization of social structures and subsistence patterns through time which supposes such values (Pluciennik 2005). Under this view, prior (at least) to the agricultural revolution humans were mere hunter-gatherers, and complex societies are a product of the necessity of organization fostered by the change to an agricultural lifeway. It is perhaps natural for those situated within Euro-American traditions to consider the past as leading up to the pinnacle of the present. Western schol- arship traces many philosophical and intellectual traditions back to Axial Age Greece (c. 600–300 BC E), contemporary with the Pazyryk burials. It is during this time that Herodotus penned in 440 BC E descriptions of the Black Sea Scythians and Iron Age Eurasians farther to the east. The finding of Scythian patterns of grave goods – horses and/or horse equipment, weaponry and artifacts decorated with a particular style of animal art – within Pazyryk burials led to their being labeled Scytho-Siberian, and has been used to attribute to the Pazyryk other traits ascribed to the Scythians. These include social structure (hierarchical), economy/ subsistence (wandering nomads) and political ideology (fierce warriors). There are problems with utilizing this model, and Herodotus’s descriptions, in this way. First, Herodotus did not himself visit the Altai region, some 7,000 km to the west of the Black Sea, and the second-hand reports he relied upon thus should be considered less than reliable. More importantly, because histories are written by people with underlying biases, historical “fact” as written does not exist. Rather, interpretations have to be distilled from material which is shaded by the assumptions of historians and their cultures. As the following passage makes clear, Herodotus (1964: Histories IV:46) acknowledges his bias against Scythian customs: Watching the horses 149 The Scythians indeed have in one respect, and that the very most important of all those that fall under man’s control, shown themselves wiser than any nation upon the face of the earth. Their customs otherwise are not such as I admire. The one thing of which I speak is the contrivance whereby they make it impossible for the enemy who invades them to escape destruction, while they themselves are entirely out of his reach, unless it please them to engage with him. Having neither cities nor forts, and carrying their dwellings with them wherever they go; accustomed, moreover, one and all of them, to shoot from horseback; and living not by husbandry but on their cattle, their wagons the only houses that they possess, how can they fail of being unconquerable, and unassailable even?

The passage also illustrates Herodotus’s focus on pastoralism as a military strategy rather than a way of life ( Ascherson 1995 :54), reﬂecting the Axial Age’s concern with warfare and conﬂict, a point that carries through to the present day. Greek society of the time was indeed structured hierarchically, and conflict, warfare, control and power were a focus of social life. Indeed, warfare might well have helped to vertically structure Greek, and thus some Euro-American, social patterns. According to Turchin (2011 :5): “I know of no empirical evidence indi- cating that other selective forces could even approach the intensity of selection (for larger group size and greater social complexity) imposed by warfare.” Con- sidering Herodotus’s cultural lens, it is understandable that Greek social structure and ideology influenced descriptions of the characteristics ascribed by the Greeks to Scythians. It is also important that Herodotus mentions the Scythian tactic of retreating to their lands and refusing to engage their enemies, a point to which I return. Regarding the economy of the Scythians, Herodotus sets up their nomadism “against Greek city-state patriotism which was about settledness, continuity, love of place” (Ascherson 1995:53). Apparent here is a social evolutionary construct that judges pastoral lifeways as primitive, while considering settled ways of living as more evolved. Considering the Pazyryk economy as different rather than less evolved might point toward a more useful model with which to assess Pazyryk ideology. Archaeologist Claudia Chang (2012 :130–131, following Riches [2000 ]), notes of the pastoral way of life two driving principles that foster not a hierarchical social structure, but rather an egalitarian one:

(1) the individual herdsman or household must ‘share pasture resources’ although livestock can be accumulated, which results in both rich and poor pastoralists; and (2) the social-leveling mechanisms used in pastoral societies are necessary to preserve the ethics of open access to communal resources such as pasture and water.

The Euro-American worldview also inherited its notion of the separation of the cultured from the natured from the Greek Axial Age. From the time of Aristotle, other animals have been distinguished from “man,” deﬁned as the rational animal 150 Gala Argent (Weil 2012:xv). This paradigm carried forward into other Western institutions. Cultures subscribing to the monotheistic, Abrahamic religious traditions (Islam, Judaism and Christianity) conceive of a hierarchy extending downwards from God, to “man” (where the term is meant to imply both male and female human beings, but this too is most often ordered hierarchically), to nonhuman animals (cf. Skeen 2011). Following this path, the natural tendency is to explore human- animal interactions in terms of a sole focus on economics and social inequality, where frameworks of human control and dominance are applied. Within this hierarchically conceived world, in addition to being anthropocentrically objectiﬁed, all domestic animals are subalterns. A focus on economic and cosmological views other than Western suggests different interpretations of what the Pazyryk might have believed about horses, and about themselves. Here, a more heterarchical view of animals might be gleaned from the Pazyryk pastoral way of life because, within these economies, both “species and individual animals [are] completely integrated culturally and economically within the social fabric” (Schwabe 1994:37). From a cosmological perspective, within ancient Eurasian shamanic traditions, it is only through animals that humans may traverse between the seen and unseen worlds (Baldick 2000:89), and there is a reciprocity between human, environmental, spirit and animal worlds (Zvelebil 2003:5–7). Similarly, within North American indigenous traditions, animals are viewed as subjects having their own customs and indeed social structures, and “Native theologies are based on close observation and experience with living animals driven by a deep curiosity and dependence on animals for both sustenance and spiritual knowledge and power” ( Matamonasa-Bennett 2015 :32–34). Within these belief systems, “spiritual views of animals and nature are exactly the opposite of hierarchical, dominance view from Western cultures in that greater power lies within nature than with humans” ( Matamonasa-Bennett 2015 :34).2 Might the Pazyryk have viewed their horses similarly? Assuming a less hierarchical and more ecocentric, holistic and inter-relational interspecies framework of the Pazyryk, what might they have gleaned from watching the horses who were so important to them in every sense?

The ways of horses The assumption of hierarchy underlies the conventional view of how horses interact with each other, where the key element in their social lives is seen as a rigid, vertical chain of command. Moreover, the notion prevails in training relationships such as “natural horsemanship” promoted in the United States and Europe. Here – although it is often couched as “leadership” rather than “domination” – humans are encouraged to insert themselves above the horse in this alleged linear hierarchy in order to dominate horses. This view is inaccurate with regard to horses’ innate way of being. Equine ethologist Lucy Rees (2016 ) has spent many decades studying wild horse bands in Uruguay. She has concluded that what is often interpreted in domestic horses as “pecking order” hierarchy (e.g., dominant individuals move others away in order to eat before less-dominant ones) is, rather, a function of feeding and space Watching the horses 151 regimens that place the horse in the unnatural position of competing for a limited resource. In other words, feeding horses once or twice a day in small areas changes their natural way of interacting. In the wild, because space is not limited, food is not limited, and there is thus no need to move another away from the pile in order to have more for yourself. According to Rees (2016 :n.p.), “truly free-living horses show almost no aggression at all. They do not establish pecking orders like domestic horses do and the vast majority of their interactions are friendly.” In wild and feral bands, the key component of equine sociality is cooperation, not hierarchy. Equine sociality consists of a fluid, contextual set of statuses and roles, negotiated by and within the group, dealing with age and connections and abilities (Boyd and Keiper 2005:55–56; Fey 2005:83). Horses are cooperative in socially complex ways. Mares, for example, will rotate “sentry duty,” allowing other mares and foals in the group to rest. Within this social context, horses display empathy for one another and a desire to work things out before resorting to aggression or violence. When faced with aggression or threat, horses usually try to avoid it by moving away ( Goodwin et al. 2009 :7). However, when fleeing does not eliminate the danger, horses will fight to protect themselves and the weaker members of their communities. Stallions will defend not their territory, but their mares. If avoidance is not possible – if, for instance, there is a real threat of wolf attack – the entire band will form a circle with the foals inside and the larger horses with their hind legs turned outward to kick the wolves (Kiriushkin and Tishkin 1987). With these points in mind, it can be noted that there is indeed a hierarchy within naturally occurring equine social structures, but it is one based protection and care, not upon aggression and domination. The degree of importance of cooperation is manifested through embodied syn- chronous movement. This corporeal synchrony is visible when horses match with each other their speed, direction and even the foot-for-foot mirroring of movement, serving as an embodied metaphor for the importance of cooperation to horses (Argent 2012). Horses extend this willingness toward cooperation, as embodied through synchrony, to their dealings with humans, where they will match human movement on the ground or while the human is riding. It is important to note that such embodied mimicry, according to ethologist Frans de Waal (n.d.), is precisely the means by which empathy both develops and is expressed. Thus, within this corporeal synchrony offered by horses to humans resides the proposal for empathy and connection, affection and care. It is a potent if soft communicative query, the acceptance of which allows a moving from a dominant/submissive relation- ality to one of a partnership. However, it is not a proposal that many humans today recognize, particularly those who subscribe to the dominance paradigm of human-horse interactions. The result of this, according to Rees (2016:n.p.) is that horses “stop communicating because we are just telling them to shut up all the time, or we just don’t have the ears for it.” Learning the physical skills and social norms necessary to synchronize movement through riding facilitates, for both human and equine students, empathy for members of the other species (Argent 2013). At an individual level, the physicality of this type of movement puts horse and rider the zone of intimate 152 Gala Argent personal space between two known individuals (Argent 2012). Touching and moving together, sweat and breath are exchanged, contributing to mutual empathy and deep bonding. Patterns of feeding, grooming, tacking, riding, schooling and traveling together are shared, bodies and mindsets change, adapting to one another in a fluid and reciprocal feedback loop. For members of both species thoughts change behaviors and behaviors change thoughts. Because humans and horses both recognize, and can bond with, members of the other species, shared histories become embedded within not only the memories of the horse- human dyad, but also within the broader social structure of the community of horses and riders. In this way, engaging in joint action forms a community of practice, within which commitments, shared beliefs and memories form a collective interspecies identity. These aspects of horses I have brought forward – these characteristics that allow them to create with humans an interspecies intersociality – derive here from scholarly ethological studies and my lived experience, but they are not beyond the grasp of other astute horse people who take the time to watch horses. With this in mind, I suggest the Pazyryk people also understood these characteristics of horses, and perhaps much more about them.

Evidence of Pazyryk human-horse connections Returning to the archaeological data, what might be gleaned from the archaeological material about the nature of the human-horse relationships within Pazyryk culture? Before touching on two archaeological features, I address the manner in which it the horses were likely kept – the interface between the humans, horses and habitat – because this relates to how the Pazyryk people saw their horses interacting with one another. The Altai region’s ecological characteristics are such that great numbers of horses could exist without human assistance or intervention. Ecological conditions favored stockbreeding, where horses “could be left to look after themselves, needing only general oversight” ( Rudenko 1970 [1953]:55). Today, indigenous Altaians (and Mongolians; see Fijn 2011) keep a few horses in the settlements for riding, but most – including trained riding horses – are pastured on their own. This type of ancient horse management, utilizing year- round pasturing, is termed tabun keeping in Asia (Rolle 1989[1980]:105). These management techniques were noted by nineteenth century travelers in the region, when a tabun could consist of 1,000 horses, including multiple stallions, who were sometimes herded, but often left alone ( Rolle 1989 [1980]:105). While we cannot be certain that the Pazyryk people practiced tabun horsekeeping, we might infer some similar process because, although archaeological evidence of small corrals has been found in Eneolithic Kazakhstan ( Outram et al. 2009 ; Stiff et al. 2006), no barns or other enclosures attributed to the Iron Age in the Altai have yet been found. If the Pazyryk horses were kept in this fashion, then the people would have witnessed, as ethologist Lucy Rees has in wild herds, only a heterarchical, cooperative, caring equine social structure, not the hierarchical behavior of domesticated horses kept in small quarters. Watching the horses 153 Archaeological evidence first reveals that moving into the Pazyryk timeframe the types of bridle bits found buried with the horses moves from highly variable to fixed (see Argent 2016a ). At least 37 variants led up to the design settled upon by the Pazyryk bit-makers (Bokovenko 2000). So functional were these jointed bits, now termed “snaffle” bits, that they are still widely used today, and they are considered mild bits, among the least severe in action. This indicates that over several centuries, the people experimented to find horse bits that worked well, and that considerable time, thought and effort were spent finding the best technological methods to communicate with the horse, methods that did not consist of applying fear or pain, but instead allowed subtle communication. Second, as can be seen, Pazyryk bridles fastened on the horses’ left sides ( Fig- ure 8.2). This indicates that, as we do today, the Pazyryk horses were handled and mounted from their left. The conventional explanation for why we in contemporary Euro-American horse culture tack, handle and mount horses from the left is that this is a habit developed when mounting war horses, where long sword scab- bards on the (right-handed) riders’ left sides meant mounting from the left kept them from sitting on their swords. Because no swords have been found in Pazyryk archaeological contexts, we must look for another explanation. Horses’ brains are lateralized in how information is processed differently in the right or left hemispheres, with the left eye keying to the right hemisphere,

Figure 8.2 Typical Pazyryk bridle structure with both the headstall (3) and throat latch (4) fastening on the horse’s left side. Source : After Gryaznov (1950 , fig. 20). 154 Gala Argent and vice versa. Recent studies show that horses prefer to scan the environment, particularly where humans are present, with their left eye (Farmer et al. 2008). They exhibit the same left eye/right hemispheric bias when attempting to gauge human emotions (Smith et al. 2016) or look at novel objects that provoke negative emotional states ( DeBoyer Des Roches et al. 2008 ). Thus, in handling horses from the left, we are allowing them to use the preferred part of their brains to assess our intentions and desires. It is likely, then, that in watching and engaging with their horses, the Pazyryk tacked and mounted them from the left understanding that this enables them to be less reactive ( Farmer et al. 2008 ). From these aspects of Pazyryk horse-keeping and handling we can infer that the Pazyryk people witnessed their horses behaving cooperatively and with mutual care, and carried this understanding of their nature forward in developing and nurturing relationship of cooperation and mutual trust. We can infer that within this community the horses were given a voice in how they were treated. This is not to say there was no power differential between humans and their horses. Certainly, there was, or we would not see horses sacrificed. However, this paradox seems to point not to the lack of care or concern for the horses. Their killing rather might be seen to indicate the cosmological importance of the deep interpersonal bonds developed between the individual horses and the deceased in this world, bonds that should maintained in the afterlife (see Argent 2013, 2016b ). Clearly within their lives, these aspects of human-horse material culture connote that every effort was made to provide the type of collaborative, pleasant relationships with the horses that horses enjoy with each other.

Evidence of violence or battle I now turn to archaeological evidence of the Pazyryk as fierce warriors. First, while archaeological evidence indicates the Scythians of the Black Sea were well armored, the Pazyryk burials contained no human body armor of any kind (Fig- ure 8.3 ). Their horse bits demonstrate that metallurgy was well developed, yet no metal shields are found. A low percentage of flimsy, woven wooden shields are interred outside the burial chambers, with the horses, implying they were the horses’ grave goods (Figure 8.4). They thus seem intended for purposes other than battle, perhaps for keeping brush off rider and horse when riding. Second, Pazyryk weaponry included knives, “fighting” axes, bows, and bone, antler and bronze arrowheads ( Bokovenko 1995a :289); as noted earlier, no swords have been found. Even in the undisturbed graves, “armaments” are scant (e.g., Polosmak 1994 , 1995 ). Knives were found with women as well as men, and many of the knives were positioned with meat on the tables in the inner chambers, for their journey to the afterlife. Except for the class of weapons labeled as “battle axes” – which also could have had other purposes – all items deemed “weaponry” rather might be considered as implements necessary for daily living: the daggers as cutlery, and bows and arrows as hunting tools. Third, the causes of death of the preserved bodies were initially primarily ascribed to battle wounds, but can be interpreted differently. The “warrior” buried Figure 8.3 Scythian armor and weaponry. Source : Rolle (1989[1980 ] , fig. 42).

Figure 8.4 Pazyryk shields. Photo credit: Gala Argent. 156 Gala Argent at Verkh-Kaldzhin 2–1, for instance, died from a stomach wound that either could have occurred in battle or been caused by an animal (Bogucki 1996:150). The Pazyryk people preserved their dead for the next world by embalming the bodies, and it is likely that many of the holes in the crania were to enable the removal of the brains for this process. Additionally, recently a great deal of scholarly attention has been turned to these bodies, with conclusions that many of them had been trepanned, an advanced surgical technique used to relieve cranial pressure on an injured brain through cutting or drilling a hole in the skull (Chikisheva et al. 2014; Krivoshapkin et al. 2014; Goodrich 2015). The man’s head wounds in Berel grave 11, for instance, could be attributed to attempts to remove clots from traumatic injury through trepanning, or to posthumously remove the brains during embalming ( Samashev 2007 :40). For the Pazyryk 2 male who had similar wounds and was scalped (Rudenko 1970[1953]:221), and the Pazyryk 2 female who was trepanned, the same is possible. Even when we find bodies with traumatic injuries indicative of violence, they do not seem to support the contention of warfare. For instance, one particular Pazyryk cemetery yielded 10 skeletons (seven adult males, one adult female and two children) with a total of 14 traumatic injuries. The study concludes that 12 of those were related to interpersonal violence, yet these injuries seem “related to raids or surprise attacks” ( Jordana et al. 2009 :1319). Raids are unlikely because Pazyryk goods appear to have been obtained through trade, not looting ( Rudenko 1970 [1953]:222–223). The authors note the injuries are “not a result of routinized or ritualized violence” yet – reflecting the standard war-mongering narrative of the Pazyryk – still contend that this verifies Herodotus’s 5th century BC E “ancient histories of warfare and violence” ( Jordana et al. 2009 :1319), a statement that seems at odds to the evidence.

Social mimicry – the wisdom of the horse It might be adequate at this point to conclude that the archaeological evidence does not support the model of Pazyryk ideology as warlike and violent. I am not the first to suggest otherwise: From an art historical perspective, Jacobson (1993) concluded that the deer image present in early Siberian art represented the role of the Great Mother in the cycle of birth, death and rebirth. It has been posited that the original Pazyryk burial ground was a “corporate cemetery of high priests” and that the Altai was the “sacral center of the Scythian world” (Kuroch- kin 1993 cited in Cheremisin 2007 :91; Latham 1958 ). However, these authors came to these conclusions by assessing the Pazyryk funerary materials related to the horses as reflecting entirely human abstract concerns. I would like to take this in a different direction. Opening up archaeological interpretations of this hybrid interspecies community to allow for mutual human-horse interdependencies and the bidirectional transfer of knowledge, I would like to address what meanings and understandings might have been passed along from the horses to the Pazyryk people who watched them so closely. Although we have no proto-historical reports of the Pazyryk people’s tactics in war, we do have accounts of the battle strategy of the Scythians, which Watching the horses 157 are contemporaneous with the Pazyryk burials. As noted earlier, Herodotus’ reports (Herodotus (1964: Histories IV:118–142)) describe Scythian warriors around the Black Sea as having no qualms about retreating from a battle. In the 5th century BC , Darius I (“the Great”) of Achaemenid Persia undertook a campaign into Scythian territory. The Scythians did not stand up and fight, but fled, leading Darius’ army into the wilderness, wearying and demoralizing his soldiers.

This had gone on so long, and seemed so interminable, that Darius sent [to them] the following message: “Thou strange man, why dost thou keep on fly- ing before me, when there are two things thou mightest do so easily? If thou deemest thyself able to resist my arms, cease thy wanderings and come, let us engage in battle.”

The Scythians were viewed as disdainfully cowardly for retreating thusly, but the effectiveness of the tactic was apparent: Darius eventually left the area, unsuccessful. When asked to join the ﬁght against Darius, several Scythian leaders declined, one stating:

We, on our part, did no wrong to these men in the former war, and will not be the first to commit wrong now. If they invade our land, and begin aggressions upon us, we will not suffer them; but, till we see this come to pass, we will remain at home. Herodotus (1964: Histories IV:119)

I suggest that the Scythians watched, understood and adapted horses’ tactics to their own battle strategies, and that an ideological rhetoric embracing the wisdom of the horse might be read in these proto-historical accounts (cf. Kohanov 2001 ). In these two encounters, we see the Scythian “fierce warriors” following exactly the logic of the horses they watched so closely, perhaps mimicking an equine worldview. Their approach – like that of the horses – was to protect the people, not the land; to flee when attacked if possible; and when flight was not possible, then to gather together and present their hooves to the wolves. Thus, the Scyth- ians in these cases might be seen, like horses, as less expansionistic than defensive warriors, and an examination of Pazyryk funerary materials leads to similar, but stronger, interpretations. When taken together, the Pazyryk archaeological evidence and Scythian proto-historical accounts suggest that although Pazyryk people may have feuded with one another, there is certainly little to imply that they engaged in warfare, if we consider that to mean “organized aggression between autonomous political units” ( Thorpe 2003 :171). Here, assumptions that battle was a daily way of life might be seen to reflect the Western (over-)concern with power, warfare and domination. With this in mind, prior interpretations regarding the “fierce warriors” of the Eurasian Iron Age might be viewed as the result of the context of the interpreters, a Western bias which includes “tendencies to emphasize thought over emotion, logic over 158 Gala Argent intuition, territory over relationship, goal over process, and force over collaboration” ( Kohanov 2001 :xvii).

Conclusion and implications The implications of considering the agency of horses – embedded as active participants in a mutually impactful culture – points to a revised view of Pazyryk group ideology which I suggest draws from the horses themselves. As evidenced archaeologically, the focus upon finding the right, mild horse bits shows that the Pazyryk were astute observers of equine behavior and desired a cooperative relationship with their horses. They tacked and mounted their horses from the left, understanding that horses are lateralized to the left eye and thus less reactive when handled from the left. They understood, as today’s working riders know and recent behavioral studies show, that positive interactions with humans lead to lasting positive memories in horses (Sankey et al. 2010). The Pazyryk people listened to their horses and they heard what their horses told them; the horses taught the Pazyryk people how they wished to be treated. Through the physical and psychological aspects of moving together synchro- nously, the Pazyryk horses called on the people to act toward them with generosity and care, and the archaeological evidence supports that the people did so. Perhaps engaging with horses in a cooperative manner influenced other ways the Pazyryk thought about the world and themselves, creating beliefs which were also contingent upon the ways of the horses with whom they shared the landscape, histories and identity. Rather than operating from a militaristic ethos, these people of the Pazyryk community preferred to stay at home – not in walled citadels, but in a communal, hybrid human-horse conceptual and rhetorical mindscape focused upon collaboration, relationship, empathy and care. As they followed their horses up and down the mountains, they also perhaps mimicked their horses’ ways of being in the world. If this is so, then the abilities and behaviors of horses played a crucial part not only in how the horses were conceptualized and treated, but also in how the people conceived of themselves as fitting into the larger world. Ethologist Frans de Waal (2017 ) recently noted, “It is clear that academics love to wage war. It is much less clear that war is in human DNA.” In these musings, I have turned away from the explicit or implicit Euro-American focus applied in many archaeological studies upon hierarchy, domination and control as the sole means by which humans and animals – and humans with each other – engage meaningfully. This has allowed an appreciation of the pro-social attitudes both humans and horses bring to their interchanges, and how those interchanges powerfully impact both individuals and communities. While human beings can indeed behave malignly and manipulatively toward others, this is not how we always act. The Pazyryk archaeological materials present a rich case study showing that the consideration of interspecies intersociality – a consideration wherein animals do not merely accom- modate human ways, but also reciprocally influence those ways – can offer fresh interpretations of animals’ impacts on human cultures in the past, interpretations very different from those previously asserted. Watching the horses 159 Notes 1 Souls Awakened: The Animals Who Have Shaped Us exhibition. An online version of this exhibit is available at http://theanimalmuseum.com/exhibits/souls-awakened/ . Originally published by National Museum of Animals and Society, now The Animal Museum, in Los Angeles. 2 This is not to imply a direct association between the Pazyryk and Amerindian traditions, but rather to open our minds to other models in order to consider how animals might have been incorporated into Pazyryk social and spiritual fabric, outside of the Euro- American paradigm. 3 For other suggestions of humans mimicking animals’ ways of being, in archaeological interpretation Fowler (2004: 148) has suggested regarding later Mesolithic Scandinavia that ‘the embedded nature of human existence in the animal world meant that social activity might have been understood through reference to animal sociality’; from an ethno-ethological perspective, see Florence Brunois (2005).

References Anon. 2012. Fashion and beauty secrets of 2,500 year old Siberian ‘princess’ from her permafrost burial chamber. The Siberian Times . Available from: http://siberiantimes.com/culture/ others/features/fashion-and-beauty-secrets-of-a-2500-year-old-siberian-princess-from- her-permafrost-burial-chamber/ [accessed 25 January 2017]. Anthony, D. 2007. The Horse the Wheel and Language: How Bronze-Age Riders From the Eurasian Steppes Shaped the Modern World . Princeton, NJ: Princeton University Press. Argent, G. 2012. Toward a privileging of the nonverbal: Communication, corporeal synchrony and transcendence in humans and horses. In Experiencing Animal Minds: An Anthology of Animal-Human Encounters. Julie A. Smith and Robert W. Mitchell, eds. New York, NY: Columbia University Press, 111–128. ——— 2013. Inked: Human-horse apprenticeship, tattoos and time in the Pazyryk world. Society & Animals 21, 178–193. ——— 2016a. Horses, mourning: Interspecies embodiment, belonging and bereavement in the past and present. In Mourning Animals . Margo DeMello, ed. East Lansing, MI: Michigan State University Press, 21–30. ——— 2016b. Killing (constructed) horses – Interspecies elders, empathy and emotion, and the Pazyryk horse sacrifices. In Ethnozooarchaeology: Global Perspectives. Lee G. Broderick, ed. Oxford: Oxbow, 19–32. Ascherson, N. 1995. Black Sea . New York, NY: Hill & Wang. Baldick, J. 2000. Animal and Shaman: Ancient Religions of Central Asia . New York, NY: New York University Press. Bogucki, P. 1996. Pazyryk and the Ukok princess. In Tombs, Grave sand Mummies . Paul G. Bahn, ed. New York, NY: Barnes & Noble, 146–151. Bokovenko, N. A. 1995a. Scythian culture in the Altai Mountains. In Nomads of the Eur- asian Steppes in the Early Iron Age . Jeannine Davis-Kimball, Vladimir A. Bashilov and Leonid T. Yablonsky, eds. Berkeley, CA: Zinat Press, 285–295. ——— 1995b. Tuva during the Scythian period. In Nomads of the Eurasian Steppes in the Early Iron Age. Jeannine Davis-Kimball, Vladimir A. Bashilov and Leonid T. Yablon- sky, eds. Berkeley, CA: Zinat Press, 265–281. ——— 2000. The origins of horse riding and the development of ancient Central Asian nomadic riding harness. In Kurgans, Ritual Sites, and Settlements: Eurasian Bronze and Iron Age. Jeannine Davis-Kimball, Eileen M. Murphy, Ludmilla Koryakova and 160 Gala Argent Leonid T. Yablonsky, eds. (BAR International Series 870). Oxford: Archaeopress, 304–310. Boyd, B. 2017. Archaeology and human–animal relations: Thinking through anthropocen- trism. Annual Review of Anthropology 46, 299–316. Boyd, L. and R. Keiper. 2005. Behavioural ecology of feral horses. In The Domestic Horse: The Evolution, Development and Management of Its Behaviour . Daniel S. Mills and Sue M. McDonnell, eds. Cambridge: Cambridge University Press, 55–82. Brunois, F. 2005. Man or animal: who copies who? Interspecific empathy and imitation among the Kasua of New Guinea. In Animal Names. Alessandro Minelli, Gherardo Ortalli and Glauco Sanga, eds. Venezia: Istituto Veneto di Scienze, 369–381. Chang, C. 2012. Lines of power: Equality or hierarchy among the Iron Age pastoralists of Southeastern Kazakhstan. In The Archaeology of Power and Politics in Eurasia: Regimes and Revolutions. Charles W. Hartley, Bike Yazicioglu and Adam T. Smith, eds. Cambridge: Cambridge University Press, 123–142. Chang, C., F. P. Grigoriev and P. A. Tourtellotte. 2007. Farmers, herders, and rich folk: The Issyk burial and the Talgar Iron Age sites. In Of Gold and Grass: Nomads of Kazakhstan . Claudia Chang and Katherine S. Guroff, eds. Bethesda, MD: Foundation for Interna- tional Arts and Education, 29–34. Cheremisin, D. V. 2007. The meaning of representations in the Animal Style and their relevance for the reconstruction of Pazyryk ideology. Archaeology, Ethnology & Anthro- pology of Eurasia 3(31), 87–102. Chikisheva, T. A., A. V. Zubova, A. L. Krivoshapkin, V. P. Kurbatov, P. V. Volkov and Anatoli T. Titov. 2014. Trepanation among the Early Nomads of Gorny Altai: A multidisciplinary study. Archaeology, Ethnology and Anthropology of Eurasia 42(1), 130–141. DeBoyer Des Roches, A., M.-A. Richard-Yris, S. Henry, M. Ezzaoui and M. Hausberger. 2008. Laterality and emotions: Visual laterality in the domestic horse (Equus caballus ) differs with objects’ emotional value. Physiology and Behavior 94, 487–490. de Waal, F. n.d. Bodies in Sync: Contagious laughter, yawns, and moods offer insight into empathy’s origins. Natural History Magazine. Available from: www.naturalhistorymag. com/features/271555/bodies-in-sync [accessed 2 August 2012]. ——— 2017. Franz de Waal-Public Page. Facebook. 25 July 2017. www.facebook.com/ franspublic/ Farmer, K., K. Krueger and R. W. Byrne. 2008. Visual laterality in the domestic horses ( Equus caballus ) interacting with humans. Animal Cognition 13, 229–238. Fey, C. 2005. Relationship and communication in socially natural horse herds. In The Domestic Horse: The Evolution, Development and Management of Its Behaviour . Daniel S. Mills and Sue M. McDonnell, eds. Cambridge: Cambridge University Press, 83–93. Fijn, N. 2011. Living With Herds: Human-Animal Coexistence in Mongolia . Cambridge: Cambridge University Press. Fowler, C. 2004. The Archaeology of Personhood: An Anthropological Approach. London: Routledge. Goodrich, J. T. 2015. Early surgeons performing trepanation: An examination of Scythian trepanations in the Gorny Altai at Hippocratic times. World Neurosurgery 83(3), 305–307. Goodwin, D., P. McGreevy, N. Waran and A. McLean. 2009. How equitation science can elucidate and refine horsemanship techniques. The Veterinary Journal 181, 5–11. Gryaznov, M. P. 1950. Pervyj Pazyrykskij Kurgan . St. Petersburg: Hermitage. Hanks, B. K. 2002. The Eurasian steppe ‘nomadic world’ of the first millennium BC : Inher- ent problems within the study of Iron Age nomadic groups. In Ancient Interactions: Watching the horses 161 East and West in Eurasia. Katie Boyle, Colin Renfrew and Marsha Levine, eds. Oxford: Oxbow, 183–197. Herodotus . 1964. The Histories . G. Rawlinson, trans., H. Bowden, ed. (Everyman Library Series). London: J.M. Dent. Hildinger, E. 1997. Warriors of the Steppe: A Military History of Central Asia 500 BC to 1700 AD . Cambridge, MA: Da Capo Press. Jacobson, E. 1993. The Deer Goddess of Ancient Siberia: A Study in the Ecology of Belief . Leiden: Brill. Jordana, X., I. Galtes, T. Turbat, D. Batsukh, C. Garcia, A. Isidro, P.-H. Giscard and A. Malgosa. 2009. The warriors of the steppes: Osteological evidence for warfare and violence from Pazyryk tumuli in the Mongolian Altai. Journal of Archaeological Science 36, 1319–1327. Kiriushkin, U. F. and A. A. Tishkin. 1987. Razvedenie i Ispolzovanie Lomashnikh Zhivot- nynh v Panneskifskoe Vremya na Territorii Gornogo Altaya, in Skifskaya Epokha Gor- nogo Altaya [Breeding and use of domestic animals in the early Scythian time in Altai Mountains]. In Scythian Epoch of Altai Mountains . Barnaul: O.F., 98–104. Kohanov, L. 2001. The Tao of Equus . Novato, CA: New World Library. Krivoshapkin, A. L., T. A. Chikisheva, A. V. Zubova, V. P. Kurbatov, A. P. Titov and P. V. Volkov. 2014. Scythian trepanations in the Gorny Altai in Hippocratic times: Modern expert appraisal of ancient surgical technologies. World Neurosurgery 82(5), e649–e655. Kurochkin, G. N. 1993. Sakralnyi tsentr kochevnikov na Altae (arkeologetnograficheskaya rekonstructivnaya model). In Problemy kulturogeneza i kulturnoe nasledie: Materialy Mezhdunar. Konferentsii. Pt. 2: Arkeologiya i izuchenie kulturnykh protsessov I yavlenii . In: Problems of cultural-genesis and cultural heritage: Proceedings of the International Conference. Pt. 2: Archaeology and the Study of Cultural Processes and Phenomena ). St. Petersburg: O.F., 93–98. Lamberg-Karlovsky, C. C. 1998. Interview in: ‘Ice mummies: The Siberian Ice Maiden’ [online]. Public Broadcasting Service. Available from: www.pbs.org/wgbh/nova/transcripts/2517 siberian.html [accessed 26 December 2006]. Latham, R., trans. 1958. The Travels of Marco Polo . New York, NY: Penguin Classics. Lestel, D., F. Brunois and F. Gaunet. 2006. Etho-ethnology and Ethno-ethology. Social Science Information 45(2), 155–177. Matamonasa-Bennett, A. 2015. Putting the horse before Descartes: Native American para- digms and ethics in equine-assisted therapies. Business & Professional Ethics Journal 34(1), 23–43. Noske, B. 1997. Beyond Boundaries: Humans and Animals . Rev. ed. Montreal: Black Rose Books. Outram, A. K., N. A. Stear, R. Bendrey, S. Olsen, A. Kasparov, V. Zaibert, N. Thorpe and R. P. Evershed. 2009. The earliest horse harnessing and milking. Science 323(5919), 1332–1335. Pluciennik, M. 2005. Social Evolution . London: Duckworth. Polosmak, N. V. 1994. The Ak-Alakha frozen grave barrow. Ancient Civilizations From Scythia to Siberia 1(3), 346–364. ——— 1995. Investigations of a Pazyryk barrow at Kuturguntas. Ancient Civilizations From Scythia to Siberia 2(1), 92–111. Radloff, V. V. 1884. Aus Siberien , vol. 2. Leipzig: T.O. Weigel. Rees, L. 2016. Real Ethology With Lucy Rees: Part 5, Bonding Behaviour Largely Ignored by Scientists. Available from: https://epona.tv/bonding-behaviour-largely-ignored-by- scientists [accessed 5 May 2017]. 162 Gala Argent Riches, D. 2000. The holistic person: Or, the ideology of egalitarianism. Journal of the Royal Anthropological Institute 6(4), 669–685. Rolle, R. 1989[1980]. The World of the Scythians . F. G. Walls, trans. Berkeley, CA: Uni- versity of California Press. Rudenko, S. I. 1970[1953]. Frozen Tombs of Siberia: The Pazyryk Burials of Iron-Age Horsemen . M. W. Thompson, trans. Berkeley, CA: University of California Press. Samashev, Z. S. 2007. Culture of the nomadic elite of Kazakhstan’s Altai Region (based on materials from the Berel necropolis). In Of Gold and Grass: Nomads of Kazakhstan . Claudia Chang and Katherine S. Guroff, eds. Bethesda, MD: Foundation for Interna- tional Arts and Education, 35–44. Samashev, Z. S., G. Bazarbaeva, G. Zhumabekova and S. Sungatai. 2000. Berel . Almaty: O.F. Sankey, C., M.-A. Richard-Yris, H. Leroy, S. Henry and M. Hausberger. 2010. Posi- tive interactions lead to lasting positive memories in horses. Animal Behaviour 79(4), 869–875. Schwabe, C. W. 1994. Animals in the ancient world. In Animals and Human Society: Changing Perspectives. Aubrey Manning and James Serpell, eds. London: Routledge, 36–58. Skeen, J. 2011. Predator prey relationships: What humans can learn from horses about being whole. In The Psychology of the Human-animal Bond: A Resource for Clinicians and Researchers . Christopher Blazina, David Shen-Miller and Guler Boyraz, eds. New York, NY: Springer, 81–106. Smith, A. V., L. Proops, K. Grounds, J. Wathan and K. McComb. 2016. Functionally relevant responses to human facial expressions of emotion in the domestic horse (Equus caballus ). Biology Letters 12, 20150907. Stiff, A. R., R. C. Capo, J. B. Gardiner, S. A. Olsen and M. F. Rosenmeier. 2006. Geo- chemical evidence of possible horse domestication at the Copper Age Botai settlement of Krasnyi Yar, Kazakhstan. Geological Society of American conference, 22–25 Octo- ber, Philadelphia, PA. Thorpe, I. J. N. 2003. Death and violence – The later Mesolithic of southern Scandinavia. In Peopling the Mesolithic in a Northern Environment. Lynne Bevan and Jenny Moore, eds. (BAR International Series 936). Oxford: Archaeopress, 171–180. Turchin, P. 2011. Warfare and the evolution of social complexity: A multilevel-selection approach. Structure and Dynamics 4(3), 1–37. Van Noten, F. and N. Polosmak. 1995. The frozen tombs of the Scythians. Endeavor 19(2), 76–82. Weil, K. 2012. Thinking Animals: Why Animal Studies Now? New York, NY: Columbia University Press. Zvelebil, M. 2003. People behind the lithics. Social life and social conditions of Mesolithic communities in temperate Europe. In Peopling the Mesolithic in a Northern Environ- ment . L. Bevan and J. Moore, eds. (BAR International Series 936). Oxford: Archaeo- press, 1–26. Part III Shared places, entangled lives

9 Growing a shared landscape Plants and humans over generations among the Duupa farmers of northern Cameroon

Éric Garine, Adeline Barnaud and Christine Raimond

The idea that domestication is better conceptualized as a continuum of material relationships between humans and other species, rather than as two polar- ized states – wild and domesticated – has been proposed by archaeology-minded scholars ( Harris 1989 ; Zeder 2006 ). Several dimensions of variation are relevant in this continuum: the intensity of human energy devoted to produce plants or animals, the intentionality of human actions that influence the development and reproduction of plants or animals and the phenotypic and genetic transformations of these organisms through repeated human actions. All these dimensions are relevant to study in the “longue durée” – the evolution of human/plant relations from societies gathering plants in virtually unmanaged ecosystems to those fully dependent on intensive production of domesticated plants, which are themselves dependent for their maintenance on their symbiotic relations with humans. It should be added that this diversity of human/plant interactions is also borne out by ethnographic accounts of indigenous subsistence systems, especially in the tropics, where people simultaneously manage many species using a variety of techniques ranging from gathering to intensive cultivation. Anthropologists, along with other scholars studying the social and symbolic life of humans, have established that drawing a clear-cut dichotomy between domesticated and wild states for animal and plant species, as Western common sense does, is not only simplistic but, through the analysis of ethnographic data about indigenous cosmogonies, is also rather ethnocentric ( Ingold 2000 ; Descola 2004 ). There is also a psychological dimension in lay views about domesticated entities: they are familiar, easily witnessed and handled in well-known places. This feature is more evidently associated with relations of humans with animals rather than with plants. However, even if the idea of “taming plants” may seem awk- ward, it deserves a little thought. Emotional attachment to an individual plant of an annual species seems unlikely, as these are mostly handled in great numbers. However, perennials can be individually known, and humans can become attached to them. Many generations of humans can reproduce and live under the canopy of a single tree ( Rival 1993 ). Some plant species, both annuals and perennials, spontaneously flourish in habitats that are intensively known and used by humans. Whether we are considering 166 Éric Garine et al. villages, fields or paths in a forest, their characteristics are derived from humans living in these places and the ecological disturbances they cause. The species, rather than the individuals, may be recognized as “old friends”. Spontaneously occurring “anthrophilic” plants, such as weeds, are not domesticated in the strict genetic sense, but they grow in domestic, familiar, places. It should be added that, whatever they think or do, humans act according to social rules. This is especially the case for activities such as mating, building settlements and feasting, but it also holds for the transmission of knowledge about plants and for the tending, the harvesting and the consumption of plants as well: anthropophilic species of plants mainly grow in places where humans collectively dwell. They are a by-product of humans’ social life. A final point to be borne in mind is that domestication is not something that can be observed at a single point in time. In addition to the long-term natural and cultural evolution that eventually led to domesticated species and subsistence systems based on cultivation, temporality also needs to be examined on a finer scale: the succession of generations of humans, as well as of the plants and animals with which they interact ( Ingold 2000 ). These ideas will be at test on the ethnographic case of the Duupa people of northern Cameroon. In a mountainous savanna environment, their economy and cultural identity is largely dominated by the shifting cultivation of a large inventory of plants, of which cereals (sorghum and pearl millet1 ) are the most important in terms of their nutritional as well as their symbolic value (Garine 2001). Besides the intensive tending of domesticated crops, Duupa individuals engage daily with numerous species in various parts of the landscape and for various purposes. We will discuss three types of human/plant interactions:

• cultivation and selection of species that have evolved genetic adaptations in response to interactions with humans; their propagules are selected and they depend on humans for their reproduction. For some of these species, wild, weedy and domesticated forms can be found in sympatry in the Duupa landscape. • consumption as pot herbs of a variety of species, some of which can be selected and sown, but which also grow and reproduce without direct human intervention in various habitats, including fields, where they may be regarded as weeds. • maintenance of trees in farmed parkland through the cycles of clearing, cultivating and fallowing of fields.

These three kinds of interactions have been thoroughly investigated by scholars from various ﬁelds, but most of the time they have been treated separately. Our point is that because Duupa farmers simultaneously engage with plants in all these three ways (among others) on a daily basis, it is worth studying them together to get a comprehensive view of human/plant relationships. Growing a shared landscape 167 Selecting seeds and farming domesticates while avoiding their wild counterparts The Duupa cultivate almost 50 species of domesticated plants. These are the species they deliberately sow/plant in chosen places (fields and gardens) using selected propagules (seeds, stems, tubers). In addition to cereals, they produce root crops, pulses, vegetables, oil seeds, fibre plants and medicinal plants, for their own use or for sale locally. For some of these species, the Duupa also identify similar plants that grow spontaneously in the wild (unmanaged savanna). Within this category are some minor species (in terms of the extent of their use) of vegetables and tubers. They are named as “crop species name from the bush” (such as haaa hõozoka “Dioscorea dumetorum from the bush” discussed below in this chapter) and they are not managed in any way, even though their similarity with domesticated species is noticed. Another example of this categorical and nomenclatural association is gap golla “spirits’ melon”, Cucumis melo L. subsp. agrestis (Naudin) Pangalo, referring to the wild form of melon cultivated by the Duupa for their edible fruits and leaves. The nomenclatural rule reflects the categorical opposition between the “things of the bush”, hen hõozoka, and “things of the village”, hen kãariya, which applies to all living things such plants, animals and ancestors’ spirits. Such a categorization translates rather well the occidental dichotomy between wild and domesticated things. Linguistically, the name of the wild species is always marked by the epithet. Most of these wild counterparts of domesticated plants are botanically similar (species of the same genus or family). Indeed, some of these species are known by botanists to grow in the region as domesticated crops as well as wild forms. They include yams and pulses. Domesticated and weedy forms of sorghum are also discussed in the chapter.

Yams Dioscorea abyssinica Hochst. & Kunth, haata in duupa language, is a wild yam commonly found in the Duupa area. Its tuber is edible and most of the time casu- ally dug and consumed as a snack while travelling. Hunters commonly taste the yams they find along their path in the wild. Tubers vary in bitterness, and when an exemplar is found to be sweet, people can replant it close to houses or near yam fields. On one occasion, we heard the testimony of a farmer who told us about his Dioscorea abyssinica that he took with him and replanted near his yam field every time he had to move his swidden. This case seems rare, and besides the fact that one of the landraces of domesticated Dioscorea cayenensis-rotundata complex has leaves very similar in shape to those of D. abyssinica, there is no specific process by which the Duupa explicitly “ennoble” wild yam and encourage it to become part of their crop, as has been described in West Africa (Dumont et al. 2005 ; Scarcelli et al. 200 6 ). However, it is known that where wild Dioscorea 168 Éric Garine et al. abyssinica and domesticated Dioscorea cayenensis-rotundata occur in sympatry, genetic exchange between the two occurs (Dumont et al. 2005: 64–65), but to the Duupa, the genetic continuum between the two is not conceived. Another species of yam shows a slightly different pattern concerning its uses: Dioscorea dumetorum (Kunth) Pax. This species is common in forest galleries where it grows wild. Tubers of this form are known to be very bitter, thus toxic in Duupa’s view, and are never used as food, even in cases of food shortage. However, some other bitter types of Dioscorea dumetorum are also grown, for medicinal and magical uses, always near houses. These two bitter, wild and planted, types are called by the same name haaa bèèè “D. dumetorum bad” or haaa hõozoka “ D. dumetorum from the bush”. These medicinal D. dumetorum can be transmitted in ritual context. Last, a third type of the same species is much appreciated for food and is named with an unmarked label: haaa . It also belongs to Dioscorea dumetorum , and it is domesticated and commonly grown. Two landraces are known and regularly cultivated. The same trilogy of edible-cultivated in outfields, toxic-cultivated near houses and toxic-wild in the bush is observed for Dioscorea bulbifera L. (tooo ), another yam species.

Pulses Two species of pulses are known under various forms, each species growing both cultivated and spontaneously: Vigna unguiculata L. and Vigna subterranea (L.) Verdc. Feral as well as cultivated, and perhaps also wild, forms of Vigna unguiculata are known to be present in northern Cameroon ( Pasquet and Fotso 1994 ). The Duupa call the domesticate zëmma , and women and men commonly cultivate it for its seeds and its leaves, both of which are frequently eaten and can be sold. Five cultivated landraces of zëmma are distinguished according to the colors of the seeds and the shape of the pod. They can all be cultivated together. The Duupa also maintain ancient crops in their agrosystem. Among these, a rare subspecies, Vigna unguiculata var. textilis , is still cultivated. It was formerly widespread in northern Cameroon ( Pasquet and Fotso 1994 ) and Duupa farmers still sow this type named tekka . They use the fibres of the peduncle and also con- sume the seeds, which are used to prepare a culinary condiment after fermenta- tion. This subspecies has been described as “primitive” by scholars ( Feleke et al. 2006) because of its similarity to wild forms (small seeds, semi-dehiscent pods). The seeds of the two subspecies zëmma and tekka are managed separately. Vigna unguiculata subsp. textilis is grown in small quantities, mostly by women who are more active in fibre use and basket making. Some Duupa farmers say that another “wilder” form, with even smaller seeds, can unfortunately appear after selected seeds have been sown. Although they cannot explain the process, the Duupa have observed that some plants with highly domesticated phenotypes can produce seeds that will develop into plants of wild-type phenotype. The known mechanism of transformation goes from the domesticated form to the wild, not the reverse. Growing a shared landscape 169 The ethnobotanical picture is similar for another species of pulse, well known in Africa for its underground seeds: Vigna subterranea bëy dunna . The Duupa, mostly women, cultivate a very common form with large seeds, as elsewhere in West Africa, but they have also developed a specific variety called bëy dun tèèè . This variety, vanishing today, was formerly sown in yam mounds, a practice for which there is an underlying agronomic rationale. The variety has primitive characters: small uniformly coloured seeds that are buried deeper (because of long peduncles) than those of the common cultivated form. Digging up the seeds of this form is harder work than digging those of the common form, whose seeds are buried near the base of the stem; shorter peduncle is an evolved trait in subterranean pulses. The deep burial of seeds of this Duupa variety is not counter-selected, because the yam mounds in which it grows are dug up in any case to harvest yam tubers, an action that permits harvest of the deeply buried V. subterranea seeds with no extra effort. The Duupa also recognize, in remote places, a wild form of Vigna subtarrenea they name after their “primitive” variety bëy dun tèè hoozoka (“ V. subterranea variety from the bush”). While the continuum from the more domesticated to the wild forms, and the naming of the forms, follow the same pattern for both species of Vigna , we have never collected for Vigna subterranea the same narrative as that for V. unguiculata about a transformation from one form to another.

Sorghum The most important species in the Duupa subsistence system is sorghum, which is represented by a considerable diversity of landraces. Other cereals, Pennisetum glaucum (L.) R. Br. and Eleusine coracana (L.) Gaertn., were more important in the past; however, over the course of the 20th century, sorghum has gained its pivotal role in the agrosystem. Duupa farmers recognize and name close to 50 landraces of sorghum.We have some genetic information for 21 of these landraces ( Barnaud et al. 2006 ). It is puzzling to understand how such diversity is maintained because sorghum is partly allogamous (i.e. cross-fertilized) and the Duupa mix various landraces in the same field, thus favouring outcrossing between landraces. Among the named types of sorghum three labels denote “off types”, individual plants not conforming to one of the culturally defined and named phenotypes, these plants are perceived as going through a transformation process. Duupa say that a form, described as ugly, can sometimes appear after the sowing of some properly selected seeds (Barnaud et al. 2009). These off-type forms are called genkiya . Their seeds can be eaten and used like those of any landrace, but care is taken to avoid their inclusion among seeds selected for the next year’s crop. If seeds of genkiya are sown (or germinate spontaneously), they may grow to produce an even more degraded form called haariya – and if seeds of haariya are sown (or germinate spontaneously), they might produce an even more wild- looking form called naa baa seee “cattle cropping-negative ”. Another Duupa narrative gives a complementary model (not all farmers agree) for how this form appears in fields: its seeds are dispersed in the faeces of cattle after they have 170 Éric Garine et al. eaten sorghum. This journey into the bovine digestive system is thought to induce the transformation. These two last-named weedy forms are considered noxious and when they appear near or within fields, they are eradicated to the extent this is possible when branching reveals the degraded form before they produce panicle. Genkiya panicles do not fit into the categories of properly identified and named landraces; haariya shows what botanists would consider “wild-type” characters: small black seeds, long glumes and shattering panicles. These negatively valued characters are even more pronounced in naa baa seee (Figure 9.1). Duupa farmers consciously select the sorghum seeds they will use for next year’s crop, choosing during harvest and threshing the best panicles (biggest and beautifully developed) of the named landraces they desire. Fostering diversity, of species and landraces, is a prominent trait of Duupas’ agricultural practice. Diversity is seen as an adaptation to changing ecological as well as economic conditions; it is sought for the diversity of material uses of plant crops, and it is also desired for symbolic uses and, for some individuals, for aesthetic reasons. Part of a farmer’s seeds come from his or her own stock, but the diverse port- folios of individual farmers (from 4 to 18 landraces in a single field) are also completed through exchanges with other farmers (Barnaud et al. 2008). Each year, about one-third of the farmers mix their own seeds with those they get from others. The transmission of sorghum seeds involves no monetary exchange; seeds can be given through kinship or friendship networks, and the Duupa have an insti- tutionalized rule according to which anyone can pick publicly a few panicles during the threshing of the crop of any other farmer. The maintenance through years and generations of the diversity of domesticated sorghum is achieved through the aggregation of individual behaviours of selection and sharing seeds. Discarding degraded forms of sorghum is a matter of individual action, but the maintenance of the diversity of proper landraces is achieved through cooperation and reciprocal exchange involving individuals from different generations. Among all the plants known and managed by the Duupa, the domesticated ones are the most valued, and cereals appear at the top of this hierarchy. The well- defined and familiar landraces of cereals are sought out by farmers because of their usefulness for making the staple food, as well as the beer that is essential to all social and ritual occasions. These landraces are the most desired type of plants and their production involves the greatest expenditure of energy, as well the greatest ritual involvement, because the help of the ancestors’ spirits is needed to cultivate them, and because, like humans and cattle, they must be protected from witchcraft.

“Back to the wild”, rather than the other way around For yams, pulses or cereals, the categorization and classification of plants, as well as the “hierarchy” of desired/feared features of the various forms of domesticated species and their wild counterparts, are easily comparable between Duupa farmers and Western botanists. Small size of seeds, bitterness of tubers and shattering of panicles are negatively valued in terms of plants’ usefulness to humans. However, even though the taxonomic categories and their valuation are similar Figure 9.1 Sorghum is represented by a large diversity in the Duupa subsistence system from cultivated to wild morphotype: A : selected landraces among the fifty named landraces managed by Duupa farmers; B : off-type forms are called genkiya; C : more degraded off-type form called haariya; D : wild-looking form called naa baa seee “cattle cropping-neg.” Photo credit: A. Barnaud 172 Éric Garine et al. between Duupa farmers and Western botanists, the underlying rationale differs in two fields: the cognitive model of reproduction and the influence of human action. Whether they reproduce plants from vegetative propagules or seeds, Duupa farmers do not conceive of reproduction as a sexual process, and we never heard any narrative that would parallel the idea of gene flow between individual plants. In the Duupa vision, plants do not mate. The transformation of the plant from one generation to the following is not attributed to crossing. A degenerating individual of sorghum or of a pulse is somehow an “unlucky” disabled phenotype that appears from the development of a regular seed. The most valued panicles selected for seeds are said by the Duupa to have their “eyes wide open”, referring to the fact that the grain should stand out of the glume. Glumes are carefully removed during the preparation of seeds for planting, as they can be a cause of poor development of the plant. The mechanical pressure they are supposed to have in the ground is viewed by the Duupa as a probable cause of the improper development of “neat” landrace seeds into the weedy forms of sorghum. Somehow, the agency for transformation is from the grain itself, not the farmer. Duupa do not view themselves as “ennobling” plants. They select the best exem- plars of known landraces and try to get rid of the disabled ones, and they act to keep the various forms in spatially distinct places, moving their fields when too many weedy forms of sorghum appear in a place. The spatial organisation of human behaviour is an essential dimension of the description of human/plant interactions among the Duupa.

Leafy vegetables in the anthropogenic landscape Domesticated species, subjected to conscious selection, such as the cereals, pulses and yams discussed in the previous section, are definitely of primary importance in terms of the extent of their use as well as their symbolic meaning. However, Duupa farmers also rely on minor cultivated crops and on gathered plants from the different biota of the patchy landscape they help create with shifting cultivation: from farms and compounds, to fallows and the wild. To grasp some of this diversity of human management of plants, we will focus on one use – relishes – for which many different species are suitable. The structure of Duupa meals follows a classic dual pattern of West African food systems: the staple is a thick porridge made out of flour of cereals (most often sorghum), which is served with a tasty side dish. The composition of the staple is monotonous, but the relishes vary. Fish and meat can enter into the composition of meals on rare occasions, but in most cases, plants are used, both as vegetables and seasoning. Almost 50 species of leafy vegetables are known in Duupa gastronomy ( Garine 1996 ). Most of these species are denoted by binomial names which finish with the word ha77a meaning “soup”. Some of these species are domesticated and deliberately sown and tended (for example, Amaranthus hybridus L., Vigna unguiculata [L.] Walp., Hibiscus sabdariffa L. and Cucurbita maxima L.), some grow spontaneously and are gathered in the wild (Hibiscus rostellatus Guill. & Perr., Psophocarpus palustris Desv.). However, most of these species are found in Growing a shared landscape 173 the fields or the young fallows. Relishes are also supplied by leaves of trees that were once planted or protected (Adansonia digitata L., Tamarindus indica L.) or that were not cut during the clearing of the savanna (Pterocarpus lucens Lepr. ex Guill. & Perr., Ficus spp .). Some of these leafy greens can also appear spontaneously in the fields as weeds. Weeding represents the main effort and the crucial stage of Duupa farming habits (Figure 9.2). Once the rains begin, many hours of many days are spent weeding: domesticates like sorghum need human help to face competition with weeds. Humans have an intimate knowledge of the plant community; not all components of the spontaneous flora of farms are felt to be equally noxious. Leaves of “weeds” such as Luffa cylindrica (L.) M. Roem., Ipomoea eriocarpa R. Br. or Amaranthus viridis L. are edible. They will be weeded or protected, depending on their abundance and depending on the season (some are consumed while they produce young leaves but are inedible afterward). Among this category of the flora from the farm biota, some individuals of the same species can grow spontaneously as weeds while others are sown. Justicia insularis T. Anderson and Corchorus olitorius L., for example, can appear spontaneously in the fields, but if the farmer feels they are not sufficiently abundant, he/she can sow a handful of their seeds to help natural mechanisms.

Figure 9.2 Leaving in weeds. Photo credit: Éric Garine. 174 Éric Garine et al. Duupa farmers, especially women, monitor the abundance and the diversity of the flora in the fields, among which they can find leafy vegetables (Figure 9.3). Whether the plant goes spontaneously or is cultivated, its status remains the same; it will be called by the same name and used the same way. An important point is that these plants are part of the expected flora of farms. Even those species

Figure 9.3 Children sorting leaves. Photo credit: Éric Garine. Growing a shared landscape 175 that were not selected and sown – i.e., not domesticated in the strict sense of the word – are familiar and foreseeable.

Trees of the farmed parkland As noted previously, some tree species also furnish leaves for the soup of the Duupa meals. As in many agroecosystems in the savanna area of Africa, Duupa farmers manage trees as well as annual crops and weeds. Some domesticated tree species are planted, mostly in villages. They provide fruits (most of the species), medicinal products, raw materials for various purposes and shade. The leaves of some of these species are also used for cooking relishes (Moringa oleifera Lam., Tamarindus indica L.), although they are not very frequently eaten. Some of the domesticated tree species planted in villages will eventually mark the landscape for a long time if dwellings are abandoned and, after a while, are included in newly cleared swiddens. For a greater number of species, the trees are not planted but protected, especially from fire, when they sprout near houses or, less often, in fields (Adansonia digitata L., Ceiba pentandra (L.) Gaertn.). The great majority of trees of the farmed parkland receive no particular care, they are not planted and no special protection is provided to help them grow. However, they are preserved when the savanna is occasionally cleared for install- ing a field. These trees will be accompanied by the cereals sown in the field and intercropped minor species for a period of 4–5 years, before remaining on their own during the fallow period. Well-known species frequently seen in fields and fallows in West Africa include Parkia biglobosa (Jacq.) R. Br. ex G. Don (valued for its seeds used to make a fermented condiment), Vitellaria paradoxa C.F. Gaertn. (for the oil made of its fruit) and a great diversity of other trees, many of which provide edible leaves for relishes. Duupa maintain more than 20 trees per hectare and in surveys of 80 hectares, we have counted 100 species of trees (Garine et al. 2005). The leaves of 14 of these species can be cooked in relishes. Some of these are by no means second-choice ingredients for cooking, and some are prized as much as domesticated leafy vegetables (for example, Celtis integrifolia Lam., Ficus dicrano- styla Mildbr., Ficus vallis-choudae Delile, Pterocarpus lucens Lepr. ex Guill. & Perr.). Furthermore, they are fortunately available during the dry season, when few annual vegetables are at hand.

Discussion: domestication within the continuum of Duupa/plant relationships In the previous pages, as we observed which plants are eaten by the Duupa farmers of northern Cameroon, it appeared that the components of the flora that end up in the calabashes for meals come from various biota in the landscape and that they are obtained in various ways: some are gathered, others are protected, yet others are cultivated after having been sown using carefully selected seeds. All the 176 Éric Garine et al. ecological units found in the landscape (villages, fields, fallows, savanna patches) and all the means of acquisition play their parts in the adaptation of this farming society. In the shifting cultivation system of the Duupa, clearing the savanna to create a temporary field – dominated by sorghum – is an ecological disturbance that temporarily transforms the flora, as well as a symbolic shift in the symbolic value of the place itself. The grass cover is suppressed and some trees are destroyed, while others are left or pollarded. The domesticated species, especially the most desired cereals, are deliberately introduced by the farmers, and the flora of adventitious species also appears for the few years the farm is exploited and before it is moved to another place. The practice of agriculture is thus a temporary rearrangement of the flora. Some species are under the direct control of individual farmers and need to be sown and taken care of. Other species appear as side effects of the local disturbance and can either be welcomed or considered noxious. Yet others, including most of the trees, remain where they were before with little human influence. Human mastery of the production of plants in mutualistic relationships with them is clear for domesticated crops, although some retain the capacity to transform themselves into weedy or wild forms that escape humans’ plans for them. At the landscape level, the transformation and uses of patches while moving farms or habitation in the territory are explicit human collective actions, but their impacts on the flora, although anticipated, are due to ecological mechanisms that are indirectly tied to human agency. Considering both how farmers eventually talk about plants – and particularly what they, often silently, do with plants – it appears that the wild/domesticated continuum remains useful for organizing the diversity of plant uses by humans in the Duupa case. Species such as sorghum are more tightly linked to human collective actions, thoughts and desires than are weeds, or the trees of the farmed parkland. However, intensive selection and cropping of domesticated plants – agriculture – does not lead to the disappearance of the practice of gathering spontaneous species. For the Duupa, the use of spontaneous species is not confined to emergency situation due to crop failure. They are used as a “normal” component of the adaptation to seasonal variation. However, gathering, whether of weeds or of tree products, is largely conducted in places like fields and young fallows that are under human mastery. These places, not the plants themselves, are domesticated, even for the short period of time (five years) when a place in the savanna is used as a field. Places chosen for clearing are somehow “tamed” for a while. Before clearing the savanna, sacrifices to the ancestors – previous users of the place – have to be made. Genealogical connections to these deceased people are verified and their offspring, sometimes distant, are asked to proceed to the offerings to their spirits. When the rituals have been properly done, the place has “cooled down” and is appropriated by the farmers for a while. It also becomes a familiar and “safe” place because the sacrifices create a positive relation with the spirits. Thus, fields are somehow domesticated places, from an ecological as well as a symbolic point of view. This process can only be achieved through the socially normed cooperation of the humans, those alive and those dead. Ironically, a Duupa farmer needs Growing a shared landscape 177 some sorghum from his previous field to brew the beer he needs for the sacrifices to the ancestors that will allow the reproduction of the next generations of sorghum so important for his own biological and social reproduction in specific parts of the landscape. These cycles of socially bounded intertwined generations of multiple species is an important theme for describing the diversity of human/plant interactions among the Duupa agriculturalists of northern Cameroon, and trying to grasp which of these relations can be considered domestication. Not all of them are to the same degree. The old-fashioned wild/domesticated opposition is still useful for depicting some aspects of Duupa ecology as long it is viewed as a continuum of multigenerational, cooperative human/plant interactions. The species of cereals that have been genetically selected and transformed by humans are indeed the most intensively cropped and desired in the most socially valued human-made parts of the landscape where humans cooperate more intensively. The genotypic and phenotypic evolution of domesticated plants could be thought as a driver for the social and cultural organization of human collective life. In the Duupa case, if one is searching the inventory of known species to find out which are the genetically domesticated plants, the best bet is to look for the plants whose reproduction necessitates more cooperation across generations between humans.

Note 1 Sorghum bicolor (L.) Moench and Pennisetum glaucum (L.) R.Br.

References Barnaud, A., Deu, M., Garine, E., Chantereau, J., Bolteu, J., Ouin Koïda, E., McKey, D., and Joly, H., 2009. A weed-crop complex in Sorghum : The fate of new genetic combinations in a traditional farming system. American Journal of Botany , 96 (10), 1869–1879. Barnaud, A., Deu, M., Garine, E., McKey, D., and Joly, H., 2006. Local genetic diversity of sorghum in a village in northern Cameroon: Structure and dynamics of landraces. Theoretical and Applied Genetics , 114, 237–248. Barnaud, A., Joly, H., McKey, D., Deu, M., Khasah, C., Monné, S., and Garine, E., 2008. Gestion in situ des ressources génétiques du sorgho (Sorghum bicolor ssp. bicolor ) chez les Duupa du Nord Cameroun: sélection et échange de semences. Cahiers Agricultures , 17 (2), 178–182. Descola, P., 2004. Le sauvage et le domestique. Communications, Nouvelles figures du sauvage , 76, 17–39. Dumont, R., Dansi, A., Vernier, P., and Zoundjihekpon, J., 2005. Biodiversité et domestication des ignames en Afrique de l’Ouest. Pratiques traditionnelles conduisant à Dioscorea rotundata . Montpellier: CIRAD-IPGRI, Repères. Feleke, Y., Pasquet, R. S., and Gepts, P., 2006. Development of PCR-based chloroplast DNA markers that characterize domesticated cowpea (Vigna unguiculata ssp. unguiculata var. unguiculata ) and highlight its crop-weed complex. Plant Systematics and Evo- lution , 262, 75–87. Garine, E., 1996. Une bonne sauce de mauvaises herbes. Note sur les repas des Duupa du massif de Poli (Nord-Cameroun). In F. Cousin and M. C. Bataille, eds. Cuisines, reflets des sociétés . Paris: Editions Sépia/Musée de l’Homme, 77–96. 178 Éric Garine et al. Garine, E., 2001. An ethnographic account to the many roles of millet beer in the culture of the Duupa agriculturalists (Poli Mountains, Northern Cameroon). In I. Garine and V. Garine, eds. Drinking: An Anthropological Approach . Oxford: Berghan, 191–204. Garine, E., Moussa, A., Raimond, C., Dounias, E., and Kokou, K., 2005. Usages alimentaires du parc arboré sélectionné (Duupa, Massif de Poli, Nord Cameroun). In C. Rai- mond, O. Langlois and E. Garine, eds. Ressources vivrières et choix alimentaires dans le bassin du lac Tchad . Paris: IRD, 63–86. Harris, D. R., 1989. An evolutionary continuum of people-plant interaction. In D. R. Harris and G. C. Hillman, eds. Foraging and Farming: The Evolution of Plant Exploitation . London: Unwin Hyman, 11–26. Ingold, T., 2000. Making things, growing plants, raising animals and bringing up children. In T. Ingold, ed. The Perception of the Environment: Essays in Livehood, Dwelling and Skill . London and New York: Routledge, 77–88. Pasquet, R. S., and Fotso, M., 1994. Répartition des cultivars de niébé Vigna unguiculata (L.) Walp. du Cameroun: influence du milieu et des facteurs humains. Journal d’Agriculture Tropicale et de Botanique Appliquée (N.S.) , 36 (2), 93–143. Rival, L., 1993. The growth of family trees: Understanding Huaorani perceptions of the forest. Man (N.S.) , 28 (4), 635–652. Scarcelli, N., Tostain, S., Mariac, C., Agbangla, C., Da, O., Berthaud, J., and Pham, J. L., 2006. Genetic nature of yams (Dioscorea sp.) domesticated by farmers in Benin (West Africa). Genetic Resources and Crop Evolution , 53, 121–130. Zeder, M. A., 2006. Central questions in the domestication of plants and animals. Evolu- tionary Anthropology , 15, 105–117. 10 Fig and olive domestication in the Rif, northern Morocco Entangled human and tree lives and history

Yildiz Aumeeruddy-Thomas and Younes Hmimsa

Introduction

The domestication of trees and grapevine in the Mediterranean region has been studied chiefly by archaeobotanists and geneticists through their analysis of the morphological traits of fruits derived from human selection, a process that started during the Neolithic (Kaniewski et al . 2012 ; Zohary et al . 2012 ; Bouby et al . 2013 ; Gros-Balthazard et al . 2013 ; Willcox 2016 ). These studies have focused on the end results, and are especially revealing as they highlight differences between the morphological and/or genetic traits of domesticates and those of their wild relatives. Processes of domestication are linked to a wide array of sociocultural and political contexts, including the acquisition and transmission of knowledge, social practices and techniques, and conceptualizations of nature that, however, cannot be examined by archaeobotanists and geneticists due to methodological constraints. Analyses of tree macro-remains by archaeobotanists and parallel studies by archaeologists have helped track major changes in social and political organization and food culture through the investigation of material contexts that prevailed with the onset of agricultural practices in the Mediterranean region (Stordeur and Willcox 2016; Willcox 2014), as well as dissemination routes, diversification processes and transformations in the techniques involved ( Salavert 2008 ). Genetic analysis of domesticates focus on present-day DNA allelic patterns and, to some extent, on linkages to fruit morphological traits of present-day tree varieties (or cultivars), and draw inferences regarding plausible past processes. Recently, the analysis of ancient DNA has given a more accurate vision of the past, thereby helping to bridge past and present contexts. There is, however, a lack of conspicuous changes in morphological traits in fruit trees owing to their allogamous reproduction system (obligate outcrossing pollination system) that leads to segregation of characters at the next generation, and because they readily hybridize with their wild parents and have done so over millennia. For instance, some geneticists find it extremely difficult to differentiate the wild olive, derived from feral types growing spontaneously from seedlings, from hybridized wild and cultivated types (Barazani et al. 2014; Besnard and de Casas 2016). Indeed, the only conspicuous characters that can effectively be described are morphological and, in some cases, biological. For instance, in the case of Ficus 180 Yildiz Aumeeruddy-Thomas and Younes Hmimsa carica , the Mediterranean fig, such characters are varieties bearing two harvests including spring parthenocarpic fruits, known as breba (bakor in Morocco) followed by a pollinized autumn fig (kermous in Morocco). The presence of infertile parthenocarpic fruits with infertile seeds is a distinct domestication syndrome. Similarly, the sweet character of almond is driven by human selection. These characters, however, readily disappear if the trees outcross and regenerate by seeds over one or two generations. Despite these difficulties, genetic studies explain large diffusion patterns and confirm the areas of origin of wild progenitors. Major questions, however, remain unanswered regarding human modes of selective approaches, intangible ecological and technical knowledge, and the conceptualizations of nature that may explain why these trees came to be integrated into the domus , the types of relations existing between people and their trees, and how people have been influenced by their ecology. Did people really view the wild and the domesticated as separate entities, and what roles have these two poles of the same species played – and still play – in human societies? Indeed, when is a tree effectively domesticated? What type of intentionality drives domestication, and how do people perceive trees or engage with them in everyday life? To what extent may this perception or engagement be a powerful agent in social organization, how are trees effectively integrated within anthropogenic landscapes and territories, and how does indigenous thinking effectively depict their existence within anthropogenic lands governed by the moral and customary rules of human groups? In the Mediterranean region, a case study conducted on one of the last areas where traditional olive forest-orchards still persist, in the Rif in northern Morocco, shows the integration of both wild and cultivated olive trees within the same agrarian space. We have led ethnographic surveys in this area since 2006 to understand people’s behaviour regarding trees, and used local knowledge to sample trees that were genetically identified. This interdisciplinary study suggests the existence of an ongoing process of domestication (Aumeeruddy-Thomas et al. 2017b). It shows the co-existence within anthropogenic agrarian landscapes of gradients extending from strictly wild (i.e. wild olive trees from non-cultivated forest areas) to hybrids, i.e. progenitors originating from crossings between wild and cultivated trees, and strictly feral (growing spontaneously from seedlings and originating from crossing between two cultivated trees) all of which are grouped under a similar local category, el berri , that is clearly distinct from trees that are cultivated generally known as zeitun (olive). This study reveals that strict barriers between biological kinds (wild, hybrids of feral and wild, and feral), have neither biological nor cultural existence in present days. In other words, from a biological perspective, there does not seem to be any major barrier between these three types of trees, and culturally they are the same. In this chapter, we use ethnographic and ethnobotanical data collected in the Rif in northern Morocco since 2006 to bring new ideas into the debate about tree domestication relating to the following broad areas of enquiry:

1 How do people effectively engage with trees from a perspective that inte- grates the socio-political, ecological and technical frameworks involved, from the encounter of social groups with the wild to its integration into the Fig and olive domestication in the Rif, northern Morocco 181 domus (i.e. domesticated spaces including the house and its surroundings) and agricultural areas? 2 What have been – and what still are – the roles played by trees as agents in domestication processes and their different types of status in society? 3 What is the meaning of diversity in the traits and morphology of trees, how is this diversity shared between all members of the group, and what does it entail as far as the place occupied by trees within human groups is concerned?

Materials and methods We have analyzed modes of engagement between humans, trees and tree products within cultural groups of the Rif in northern Morocco and the tree species Ficus carica and Olea europaea with a typical ethnobiological approach that considers both the ecology and biology of the trees as well as people’s cultural representations, their knowledge and practices. Ficus carica , the Mediterranean fig, is a dioecious tree having distinct male and female trees existing in their wild forms throughout the Mediterranean region ( Zohary et al. 2012); their natural habitats are known to be riparian forests as well as wet cliffs. The works of Achtak (2009) and Achtak et al. (2010) in Morocco reveal significant differentiations between three major regions in Morocco that indicate localized domestication processes. These studies suggest that most varieties have developed from selected seedlings of wild Ficus species belonging to western Mediterranean genetic pools. The Rif in northern Morocco is a hotspot of fig diversity and hosts some 122 distinct genotypes (Achtak 2009) corresponding to 133 named types or varieties ( Hmimsa et al . 2012 ). Olea europaea var. sylvestris, the oleaster, known as the wild ancestor of Olea europaea var. europaea , is widespread in Mediterranean thermophile vegetations. Although natural populations of Olea europaea are seldom found except in a few restricted areas in northern Mediterranean countries, the latter are widespread throughout the Maghreb within forest types known as matorral . Genetic studies by Besnard et al . (2013a) and Diez et al . (2015 ) show that olive-producing countries have developed their own set of olive varieties, most of which contain cyto- plasmic DNA from the east with admixtures of nuclear DNA from the east and west, a sign of ancient east-west migration (Besnard et al. 2013b). In Morocco, traditional varieties follow a similar pattern, suggesting introgressions by oleasters from the west (Khadari et al . 2008 ; Haouane and Khadari 2011 ). El Bakhali et al. (2013) show that a majority of Moroccan varieties have developed from recombined types that include western oleasters. Ethnobotanical surveys show that oleasters are kept in agricultural lands and they yield a high-value medicinal oil. They also serve as rootstocks for grafting olive varieties in northern Morocco, a technique that associates a wild part including the roots and trunk base and a grafted variety (Aumeeruddy-Thomas et al. 2014, 2017b ), a chimera made of two individuals, one wild and the other domesticated. People in the Rif are Berber and Arabic-speaking groups; the latter – mostly located in the western part of the Rif – are known as the Jbala . They speak a 182 Yildiz Aumeeruddy-Thomas and Younes Hmimsa pre-Hilalian Arabic dialect that has retained traits of the language spoken by the first Arab migrants who arrived in northern Morocco in the 7th century, with strong influences from Berber dialects (Aumeeruddy-Thomas et al. 2017a). Berberophone groups, known as Riffian Amazigh, speak Tarifit, a Berber dialect that has differentiated from other Berber dialects in Morocco. We focus in this paper on data obtained among the Arabic-speaking Jbala. Berber and Arabic-speaking groups have similar social setups and social organization, which are framed according to a segmentary system. The basic social and territorial unit is the village or Douar , which belongs to a confederation of villages bearing the name of one patrilineage (Ait, or Bni, meaning son of followed by a patronym), although within one confederation and even within the villages belonging to it, people identify themselves as a set of distinct patrilineages living within the same territory but tracing their original ancestors from distinct geographical areas. They exchange wives and share a common body of knowledge and the same market areas, but exchanges with other confederations and even between Berbers and Arab-speakers may occur, although more rarely.

Entangled human, fig and olive tree ecologies and societies

Figs Figs are associated with human settlements, the house (domus) and its particular ecology, i.e. availability of water spilling from kitchens or near wells, presence of organic matter and conditions where fig seeds dispersed by birds readily grow in areas corresponding to their ecological requirements. This is true for houses scattered in agricultural areas, as is often the case in Jbala settlements. It is not rare, however, to see fig trees growing within closely connected houses in small grouped villages, because fig trees also seem to highly appreciate walls and build- ings, the latter being most often adobe dwellings made of an earth/cereal straw mortar in which small cracks provide ideal places for the tiny fig seeds to grow. In other words, fig trees “invite” themselves and begin as commensal with humans before ultimately becoming mutualist, as we shall see in this chapter. Male or female fig trees growing from seeds are named nabout by the Jbala, meaning “what grows by itself or spontaneously” as opposed to trees cultivated from cuttings known as kermous , or “figs,” and which are female trees (Hmimsa et al . 2017 ). Hmimsa (2009 ) shows that the oldest multi-centennial fig trees in the Bni Ahmed area are found near shepherd shelters that are remnants of previous forest-based pastoralist activities or old abandoned villages. People in that area currently recognize that fig leaves are used as fodder and that their shade is important for animals and humans especially during the extremely hot Mediter- ranean summer. This could represent the primary contexts for fig domestication. A second major step that characterizes fig domestication is caprification, an ancient practice in which humans accompany – and actively participate in – the extraordinary pollination biology of figs that involves its obligate mutualist pollinator, Blastophaga psenes, known as chnewila in northern Morocco. The Fig and olive domestication in the Rif, northern Morocco 183 pollination biology of the fig is the result of several million years of co-evolution. The chnewila is widely known locally to complete its life cycle inside a male fig. Only when a female fig is receptive (i.e. when the fig’s ostiole , or small hole, opens up and leads to the hidden flowers found inside this urn-shaped fruit) can the insect penetrate and pollinate the fig ( Kjellberg et al . 1987 ). Humans, since the Mesopotamian period (Marinval 2008), have practiced the art of caprification, i.e. bringing male figs onto female fig trees to favour the pollination of the female figs. This practice is well-known throughout the Mediterranean region and facilitated by the sale in markets of male figs from different origins, making them available to pollinate female figs that have been selected to come into production throughout the season (August to September or October), ranging from early to late varieties. Male figs in the Rif are seldom kept in orchards, another constant feature throughout the Mediterranean region. Male figs or caprifigs are almost always collected from places where they grow spontaneously, i.e. near rivers, in forest areas, or are cultivated in specialized villages such as in Turkey in the Meander Valley which produces dried figs of the Smyrna type that represents the major Mediterranean dried fig market (Finn Kejllberg unpublished data1 ). In the Rif, caprifigs are sold in markets by specialized vendors who provide caprifigs originating from diverse geographical regions and made available throughout the fruiting season. These activities form complex social exchange networks in which vendors and buyers negotiate the quality of caprifigs and their pollinators ( Aumeeruddy-Thomas et al . 2014 ). We followed every step of this activity during the caprification season when people (men, women or children) spend considerable time accompanying caprifigs and placing them in series of pendulums made up of 2–3 caprifigs, every two or three days on each female tree, a unique system whereby humans play a direct role in accompanying an insect pollinator to individual trees in order to pollinate the flowers of that particular tree1 (Figure 10.1). People name the caprifig dokka r (the male) and the act of placing the caprifig on a female tree is kedokkar or ted- kar , which symbolically means “to mate”, although this act is meant to help the chnewila to find their way to the female figs (Aumeeruddy-Thomas et al . 2014 ; Hmimsa et al. 2017). This highly delicate sexual “traffic” implies that humans know when a female fig is receptive thanks to its odour. Indeed, a specific smell related to volatile organic components (VOCs) is emitted by female figs and is recognized by the pollinator (Soler et al. 2012). Farmers in northern Morocco perceive this specific odour that differs from the ordinary smell of unreceptive female figs. This is the signal they use to identify when distinct varieties need to be caprified. Needless to say, within this intermingled context in which humans accompany the sexual life of figs, people develop a deep understanding of the trees’ ecology and a proximity that makes them close kin, and a member of their social group. Domestication here is not merely a matter of selecting fruit types (as it is often thought), but rather a question of complex mutual interactions between humans, figs and insects that bring these three distinct species within a common sphere produced by large socio-cultural networks (markets) underpinned by the fact that humans accompany the pollinator to the right place. 184 Yildiz Aumeeruddy-Thomas and Younes Hmimsa

Figure 10.1 Children in the Bni Ahmed region caprifying a fig tree, Rif, northern Morocco. Photo credit: Yildiz Aumeeruddy-Thomas.

Historically, extensive savannah-type areas (Figure 10.2) that include fig and olive trees as well as cereals were, according to oral memory, planted with fig trees only after the independence of Morocco in 1955, through first clearing the forest and second, planting cuttings of fig varieties that originally had developed in ancient orchards or near shepherds’ temporary dwellings ( Hmimsa 2009 ). This was a first step to appropriating forests out of fear that they would fall into the public domain following previous French colonial rules. This step implied bringing cuttings from the house to the field, a process whereby fig trees were used to signal land appropriation by humans. Located far from places where they naturally grow, they are highly cared for and protected by humans and they benefit immensely from the cultivation of cereal fields with which they now share the same niche. Basal shoots (ghattas ) are systematically cut by the farmers to encourage the tree to create a single trunk so that sheep and goats may not climb and reach their delicious leaves and fruits. People also always accompany animals in the fields after harvest of cereals, allowing animals to eat only the lowest leaves. Tree architecture and morphology is a typical result of trees being tended by hand by humans over long periods of time and of being browsed by animals. Furthermore, fig trees travel with humans and are planted everywhere they stay. In the Rif, it is said that when a young man leaves his father’s house to Fig and olive domestication in the Rif, northern Morocco 185

Figure 10.2 Large mixed fig, olive and cereal agroecosystem, Rif, northern Morocco. Photo credit: Yildiz Aumeeruddy-Thomas. build another house, he takes with him a branch of the fig tree where his mother’s placenta had been buried after his birth and or his prepuce after circumcision (Bouzidi 2002). He will plant this branch near his new house as confirmed to us by Bouzidi, although we have not witnessed this practice in person. People also perpetuate what their grandfathers had already planted, thereby using living trees and cuttings from them to maintain their ties with previous generations ( Hmimsa 2009 ). The transgenerational nature of fig trees, that typically live for more than three human generations, means that these trees (as well as olive trees) are associated historically with an important function, that of territorial markers. In the Rif, where territorial units are at the centre of social organization, as demonstrated by Jamous (1981 ), this function that favours the dispersal of trees throughout the landscape benefits both humans and the fig trees themselves. The presence of figs on the territory may also refer to myths linked to ancient periods when current populations first settled on this territory. Fig trees were already there and belong, according to current sayings, to the original inhabitants who were there before, these inhabitants being earlier human groups or intangible creatures, jnoun , that lived in this place before them (Jabiot 2008). The encounter with fig trees forms part of the inhabitants’ foundation myth. The people belong to the land where fig trees already existed and will still exist over historical times. Indeed, it is the trees that create the territory and the sense of belonging to the place because of 186 Yildiz Aumeeruddy-Thomas and Younes Hmimsa their long temporality and perennial role as spatial markers as also shown for almond populations (Prunus dulcis) in the eastern part of the Rif (Delplancke and Aumeeruddy- Thomas 2017). Although the context is very different, similar situations linking trees inextricably to people exist elsewhere in the tropics such as in Sumatra ( Aumeeruddy-Thomas 1994 ) and in Papua New Guinea ( Kennedy 2012) whereby Ficus species are perceived as civilizing agents. As emphasized by Kennedy (2012:145) “That trees might domesticate humans rather than the other way around also resonates with a common place in Papua New Guinean origin myths”. In Kerinci, in central Sumatra, a large banyan (Ficus sumatrana ) is the mediator which guides and enables the founding father of the valley to find the spring where lives the semi-divinity who eventually became the mythical mother of the Kerinci people (Aumeeruddy-Thomas 1994). Trees provide concrete support in the form of shelter and food thereby enabling humans to overcome their awe and fear of the environment. They are frequently perceived as powerful mediators between humans and supernatural beings and gods. Thus, people’s lives, from birth to adult age and death (as shown later in this chapter) are intrinsically linked to their trees. Fig trees – like many other long-lived trees – are very close to humans; they persist and help humans to fix their lives within a given territory and, as such, to forge together a common history.

Olives The role played by olive trees in people’s lives is very different from the role played by figs. Indeed, wild olive or oleasters, known locally as el berri, are common elements of natural forests throughout the Maghreb. In the Rif, especially among the Jbala, el berri oil is well-known and is highly valued as medicinal oil. It fetches a higher price in markets than olive oil, but it is mainly extracted for personal use. El berri in Jbala society is intrinsically linked to how people acquire new agricultural lands as well as to religious contexts. Access to new agricultural lands results from the clearing of the forest, designated as ghaba . While doing so, farmers save many oleasters, especially the large ones. Among these, the majority will be grafted with selected olive varieties (Figure 10.3). However, exceptionally large individuals bearing large quantities of oleaster fruits with typical shapes are named and designated either as berri rkik , meaning oleasters with thin and long fruits and berri meslal , oleaster fruits that are roundish and bigger. An the overall classification of olive varieties found throughout Morocco considers Meslal olive varieties to be those that bear large olives. They occur early during the season, have elevated water content and are not suitable for making oil. They are difficult to conserve and are therefore transformed immediately into table olives. Berri rkik and berri meslal are saved and generally develop very stout trunks that derive from the pruning of low branches and basal off-shoots, an architecture obtained through the active shaping of the tree by humans; such trees can be inherited similarly to olive trees. They are known to benefit from agricultural practices (cereal cultivation, tilling etc.) and are harvested. Grafting oleasters is the way of appropriating land from a customary perspective as well as the best option, from Fig and olive domestication in the Rif, northern Morocco 187

Figure 10.3 Oleasters grafted with olive varieties within a tended orchard in Sidi Redouane, Rif, northern Morocco. Photo credit: Yildiz Aumeeruddy-Thomas. the farmer’s point of view, in rain-fed areas to produce the best fruit and oil quality ( Aumeeruddy-Thomas et al . 2017b ). Domesticating olive trees takes another approach. Oleasters remain where they grow in the forests; it is the forest that is converted. Oleasters are kept and grafted, thereby forming a “double” individual, a dual plant that incorporates the wild and the cultivated. People say that the variety changes its “passport” and becomes stronger. This action is historically associated with the conquest of agricultural lands throughout the Maghreb. This differs from fig trees which are transported by farmers away from places where they grow or, alternatively, “visit” human settlements through germination near houses. People sometimes transplant oleaster seedlings, but this is a recent practice due to the absence of forests. The rootstock is the essential factor that links the variety to the land and is constantly presented by people as the part of the tree which is most important, a vision rarely considered in domestication literature devoted to olive trees. The orchards produced are closer to forests than fruit orchards, even though cereals and pulses may be grown between the trees. The rootstock is sometimes por- trayed as the father of the olive, the part that confers vigour and taste to the oil, a quality attributed to its taproot; but it is also a metaphor for this highly patrilocal society where the bride goes to live in her husband’s house, on land belonging to him and upon whom she will be totally reliant. Her own land is transferred to her brothers who virtually consider this land to be their own; a widow coming back 188 Yildiz Aumeeruddy-Thomas and Younes Hmimsa to her village generally finds it extremely difficult to recover her inheritance. Being a man in the Rif requires possessing land (Jamous 1981). Acquiring land from the forest is a blessing because productive land is typically considered in Islamic countries to be a way of transforming “dead land” into “living land” or a productive land, a concept and practice which we also observed in Indonesia (Jabiot 2008; Aumeeruddy 1994). Grafting olive varieties on oleasters is a widespread practice throughout the Maghreb, and whoever plants a tree or grafts forest oleasters acquires land under a regime known as m’gharrsa, meaning “that is planted” – a system that the Romans already used in antiquity to make the “indigenous workers” produce more oil and which most probably pre-dated Roman occupation ( Aumeeruddy-Thomas et al . 2017b ). As we can see, trees prove to be at the centre of political and social life and, more specifically, lie at the heart of the relationships between people and their territories (see also Delplancke and Aumeeruddy-Thomas 2017 ). Remnants of the practice of grafting on oleasters are found in Corsica, Sar- dinia, in certain Greek islands and in Turkey, but this practice has almost completely disappeared in Northern Mediterranean countries (Diez et al . 2005). This possibly occurred during Greek and Roman periods, when large scale “industrial” plantations based on plants originating from nurseries were established in rows, a change in the paradigm of allowing trees to grow spontaneously, and possibly a change in ways Mediterranean people started to conceptualize nature as a disjointed element of culture. The oldest traces of grafting practices have been found in Tunisia during the Roman period (Brun 1986) and it has been argued that the Berbers knew the art of grafting well before the Greeks ( Camps 1961 ). From a historical perspective, olive trees in agricultural landscapes symbolize at once a historical past, as well as the present livelihoods of the people. Land- scapes shaped over historical times exist in Andalusia and southern Portugal, in the Dehesa and the Montades where holm-oak and cork oak are the key features of large agricultural landscapes, providing shade and acorns to animals (Joffre et al . 1999), or in the vast Argan agroecosystem of southern Morocco, where Argania spinosa, the Argan tree, which is a forest tree, structures domesticated landscapes, intrinsically linked to the history of people who inhabit such areas (Ruas et al . 2016; Simenel and Aumeeruddy 2016). Around the world, the social lives of trees have made an immense contribution to human lives and histories as land markers and symbols of identity that are often misunderstood by distant power and hierar- chies (Rival 1998). How trees have brought about cultivated landscapes that host complex temporal and spatial practices is still little understood by foresters and agronomists ( Genin et al . 2013 ).

Humans and trees: who protects whom? Fig trees are intentionally planted near houses for their shade and are tightly knitted to cane hays (Arundo sp.), forming a boundary that protects domestic life from the outside world or from any prying eyes. Bouzidi’s studies ( 2002 ) among the Fig and olive domestication in the Rif, northern Morocco 189 Jbala highlight the fact that fig trees are seen as protectors of the house and host intangible entities which ward off evil and thunder, a system whereby men and figs are closely knitted in a mutualistic system of protection. Humans, indeed, plant fig trees and protect them from the teeth of domestic animals when they are young. Bouzidi (2002 ) also refers to the importance of figs as a staple food, representing the “araoula” which means basic food, a term sometimes also used to designate dried figs. “Araoula” is the stock of food required for each house to be able to survive, together with other staple food such as cereals, beans and olives, an observation also validated by Hmimsa (2009). In a diversity of rituals, figs serve as mediators between humans and the dead or angels during specific religious rites. Dried figs deposited at the cemetery when somebody dies are believed to accompany his soul and wash away his sins ( Aumeeruddy-Thomas et al . 2014 ). Furthermore, like many other products in Morocco, figs as well as olives, especially oleasters, may be associated with specific places where a well-known saint is believed to have lived or died, or with ancient burial grounds. Almost all mosques, cemeteries and the mausoleums of saints are associated with very large oleasters or sometimes fig or carob trees. Figs and olives are known to carry the baraka , or benediction of God, through the mediation of the saint. The baraka accompanies people who transport tree products from the place formerly inhabited by a saint back home to the village. Re-distributing figs or olive oil to kinfolk and neighbours is thus a social act that links human society to saints, God and trees through fruit tree products. The baraka is a widespread concept throughout the Maghreb ( Jamous 1981 ). In the case observed, it is an essential system by which people are linked to their territories and the Saints with the help of trees and their products, which therefore acquire agency. Indeed, in the Rif, good harvests as much as children are symbols of prosperity and major signs of baraka . In this way, trees associated with saints, individuals and social groups form part of the same social exchange network. Although figs and olives are important for the local economy and people’s livelihoods, they are not solely selected for their fruits. They are strong spiritual mediators used by people throughout their lives at specific key moments: at birth through the burying of the placenta at the foot of the fig tree seen as the protector of the house, during circumcision, through the occupation of land at the territorial level, when people die and during life after death. Perceived as people’s protectors, they are in turn protected as part of a common cultural system where they have a specific “moral” status, bringing good (or bad) influences into social life as much as humans influence the trees’ biology. Although it is difficult to know how and in what circumstances such strong linkages between humans and trees were developed, it is likely that this high density of reciprocal bio-cultural linkages has developed since prehistoric times. We suggest that this participated to the onset of domestication and could be considered as a constitutive element of tree domestication processes for these two emblematic Mediterranean tree species. Domestication could thus be defined as a mutualistic system leading to a very high level of reciprocity between humans and their domesticates, an idea developed and further highlighted by Descola (2005 : 514–525) and other authors more recently. 190 Yildiz Aumeeruddy-Thomas and Younes Hmimsa Domestication: naming, taming and reproducing Naming – which is intricately linked to reproduction techniques, another idea that has little permeated domestication “thinking” to date – varies according to the biology of specific trees as well as the socio-political organization in question. From the farmer’s perspective, in northern Morocco, the naming of a good variety of fig is the main step leading to the inclusion of a wild fig tree that is subsequently propagated by cuttings and joins the portfolio of cultivated figs ( Aumeeruddy-Thomas 2010 ; Hmimsa et al . 2012 , 2017 ). However, even though the Rif hosts a substantial diversity of fig varieties that are propagated through cuttings, people do not constantly invent new varieties. Naming individual fig varieties that correspond to specific morphological characteristics are socially recognized labels that support domestication and diversification processes. Our previous results show that these labels correspond to different semantic fields. As we have shown in an earlier study, naming follows different rationales, that of the colour of fig epidermis, analogies to animals, domestic objects, but also origins ( Hmimsa et al . 2012 ). Genetic studies conducted by Achtak et al . (2010), who simultaneously recorded vernacular names of the samples and analyzed the nuclear DNA with the help of microsatellites, show that there are synonyms and homonyms, which means that the different samples bearing vernacular names refer in some cases to the same SSR profiles (or genotype) and on the contrary, that sampled varieties bearing the same vernacular names may have different SSR profiles. However, in the Rif, Achtak (2009) found 122 distinct SSR profiles and Hmimsa’s study of known named fig types revealed 133 distinct named types. Achtak (2009) through an allelic pairwise comparison of all samples analyzed, finds more genotypes that differ by a large number of allele than genotypes varying by a few alleles (1–3), which suggests that there is a very high probability that the majority of SSR profiles found are varieties that originate from the selection of seedlings rather than the selection of new varieties emerging from somatic mutations of pre-existing varieties. People progressively create new varieties, but the creation processes of new varieties are slow in order to avoid over-creativity that may hinder a comprehension shared by all of the social significance of a given variety. Indeed, most varieties are shared, recognized and circulated throughout a very large region in northern Morocco. They play a central role in gift-making, in market social exchange networks and they need to be clearly identified for what they are. The small percentage of homonyms and synonyms found shows that farmers may select a wild fig named after an already cultivated one due to anal- ogy in morphology, thus leading to homonyms, but the percentage remains very low. Most farmers explain that their major driver is to reproduce what their fathers and forefathers had planted, a norm which tends to limit the haphazard, individual development of new varieties. The absence of naming among olive trees is highly noticeable and has posed many problems to geneticists (Khadari et al. 2008), in addition to the fact that there is one dominant variety in Morocco, the Picholine marocaine , which in some places comprise almost 80% of planted olive trees. The fact that it is Fig and olive domestication in the Rif, northern Morocco 191 grafted on a very large diversity of rootstocks in the north adds diversity to this domesticated variety thanks to the addition of the highly heterogeneous wild part ( Aumeeruddy-Thomas et al. 2016). A similar situation occurs in Corsica, where virtually a single variety is grafted onto oleasters, the sabine in all old orchards along the western coast (Aumeeruddy-Thomas et al. 2017b). In the Rif, almost all grafted olive varieties are invariably named zeitun (or zeitun beldi, i.e. “the olive from the land”), which implicitly means good for oil, unless they look like a Meslal, which is a variety used as table olives. Although there are a few other varieties related to olive morphology, the main distinction made by local farmers lies in whether the variety has been grafted onto an oleaster by them or whether the variety has been brought from elsewhere and planted by means of cuttings. In this case and even though it is the same variety, they would call the latter either nokla or lghers , meaning “the plant” or “planted”, thereby signifying that the reproduction system does not correspond to the local system. Nokla and lghers , or zeitun romi (“the foreign olive”, as opposed to the zeitun beldi) because the plants have been reproduced in faraway nurseries by specialists, generally by means of cuttings or grafting on cuttings generally proposed – if not imposed – by central authorities. They are perceived locally to be less vigorous and to yield an oil which has a bad taste compared to grafted trees. As the reproduction of the trees by the farmers is a significant part of their identity, the way the trees are named is therefore also linked to strong sociopolitical or political contexts. Indeed, trees belonging to the land are those defined by the way farmers interact with them through reproduction (with a parallel established here between how they reproduce and name trees) and how society reproduces itself socially.

Discussion and conclusion In this chapter, we show two distinct approaches to tree domestication. The first approach uses fig trees that invite themselves spontaneously to human settlements; they may remain where they happen to grow as a result of seed disper- sion by birds, or alternatively, they are selected and dispersed through vegetative reproduction across a landscape where the tree would not settle spontaneously. The fig tree, beyond morphological selection, remains the tree that is closely associated to the domus , and to patrilinear and segmentary social organizations where territorial construction is a major feature. Fig trees symbolically and effectively accompany people who establish new houses and are planted across the agrarian landscape wherever humanity decides to appropriate new lands; they protect, and are protected by, people. The second approach involves the olive, which is primarily a forest tree. Humans build their identity not through displacing these trees but through grafting selected varieties on them, thus benefitting from the vigour of the land. Olive trees are a very strong symbol of power in Morocco, as shown by their presence in all imperial medieval Berber gardens such as the Menara or Agdal gardens in Marrakech. Olive trees, even during the French protectorate, were widely used to show the importance of French occupation. Since independence 192 Yildiz Aumeeruddy-Thomas and Younes Hmimsa and up until today, the olive is at the centre of major power-related issues; the latest agrarian policy continues to promote large-scale plantations of olive trees that do not consider the farmers’ own capacity to reproduce them, but rather imposes exogenous reproduction techniques, an approach which is disdained by farmers ( Aumeerudddy-Thomas et al . 2016 ). Both the olive fruit and oil are signs of baraka , God’s gift and benediction, and are involved in many exchanges including gift systems which we have not discussed in this chapter. Keeping trees that grow spontaneously – the most sacred being found in the gardens of a mosque and cemeteries – is a way people maintain the flow of baraka , a concept of the sacred vital force of nature found in many other societies regarding trees (Rival 1998). The fact that they come to people (figs) or provide rootstock for cultivated varieties (olives) in addition to the variety of products that they provide – including fruits, wood, and shade – are perceived as signs of benediction that lead to people’s deep engagement with them. Trees are powerful agents that mediate between humans, as well as between humans and supra-natural forces, thereby ensuring reciprocal exchanges. Their presence over transgenerational periods adds to their roles as members of socio-political collectives which have a common trajectory and history. Nam- ing or not naming are two contrasting ways of incorporating these two trees into human lives and history. On the one hand with figs, the individuality of each cultivar is valued, while on the other hand for olives, the population provided by forest-based rootstock is of greater importance. In the case of both figs and olives, the farmers’ key motivation to be a full actor in their reproduction system is a necessity, a strong element of identity through the reproduction of the trees by cuttings and caprification for figs, and grafting for olives. Both systems ensure the continuing role of the wild, with wild and cultivated clearly being two sides of the same coin, ensuring the reciprocal engagement of humans and trees beyond the divide between wild and domesticated. For figs and olives, people manipulate both wild and cultivated trees through different steps that include socially defined land appropriation systems and issues of power, moving from a wild forest tree to one that produces a new variety or cultivar with no linear intention. This ethnographical study suggests that the selection of fruit traits are not the sole drivers of domestication processes. For both figs and olives, one can understand that it is not the genetics of the trees that counts the most, but rather how they reproduce, how they participate in social life within the historical occupation of a territory and how they live in very close intimacy with humans. Archaeobotanists have focused much attention on changes in tree reproduction systems, especially the shift from sexual reproduction to vegetative reproduction. The latter was considered by Zohary et al. (2012) as the main way societies could fix the desired traits of trees which are mostly allogamous (including dioecious). The review of the diversity of practices pertaining to fruit trees across the world ( Bouby and Ruas 2014 ) shows that the sowing of fruit trees or allowing trees and woody vines (grapes) to grow spontaneously represented an important propagation Fig and olive domestication in the Rif, northern Morocco 193 system during the Bronze Age and even until recently in non-industrial and traditional settings. It is possibly a much more common practice in the Mediter- ranean region than initially thought. Indeed, this case study shows that trees growing spontaneously play major roles, both as a way to develop new varieties and provide rootstock, as well as perform roles of a religious nature, and in the co-construction of human identity and history. The “chaîne opératoire ” (successive steps in technical systems) and techniques involved in tree reproduction are crucial to understand and explain how the reproduction of trees also means the reproduction of human societies.

Acknowledgements We are thankful to different foundations and programmes that have supported this work, including the GDR Mosaïque, 3353, CEFE_INEE, CNRS, the CNRS PICS, La Montagne et Ses Savoirs (Institut de Recherche et d’Etudes sur le Monde Arabe et Musulman, Aix en Provence & Centre Jacques Berque, Rabat), the FIGOLIVDIV project of the Agropolis Foundation and more recently the BIODIVMEX programme, Mistrals, CNRS. We are also thankful to Bouchaib Khadari (SupAGRO, CBNMed, INRA, UMR AGAP Montpellier), with whom we learned about the genetic dimensions of figs and olives, and Majid Moukhli, INRA Marrakech who introduced me to the world of olives in Morocco, Moham- med Ater from Abdelmalek Essaadi University, Tetouan who has been my major administrative and research partner in northern Morocco, and Younes Hmimsa, Abdelmalek Essaadi University, Larache, with whom we did some common and joyful field work although looking at the same “objects” with very different eyes and approaches. Foremost, our heartfelt thanks go to the inhabitants of the Rif, especially in Bni Ahmed and Ouezzane, with whom we learned most of what is in this chapter, with a special word of thanks to the trees of the Rif if the expression of such thanks is permitted.

Note 1 In the case of date palms, people harvest pollens and bring them to the female date trees.

References Achtak, H., 2009. Domestication et diversification variétale chez le figuier au Maroc: bases pour la conservation et la valorisation des ressources génétiques locales . Thesis dissertation. SUPAGRO, Montpellier. Achtak, H., Ater, M., Oukabli, A., Santoni, S., Kjellberg, F., and Khadari, B., 2010. Tradi- tional agroecosystems as conservatories and incubators of cultivated plant varietal diversity: The case of fig (Ficus carica L.) in Morocco. BMC Plant Biology , 10, 28. Aumeeruddy-Thomas, Y., 1994. Représentations et gestion paysannes des agroforêts en périphérie du Parc National Kerinci Seblat à Sumatra, Indonésie . People and Plants Working Paper Series, No 3, UNESCO, Paris 46p. www.peopleandplants.org/storage/ working-papers/wp3sp.pdf 194 Yildiz Aumeeruddy-Thomas and Younes Hmimsa Aumeeruddy-Thomas, Y., 2010. Domestication des arbres en méditerranée: un continuum entre nature et culture. In Eds Claire Delhon, Isabelle Théry-Parisot and Stéphanie Thiébault, eds. XXXe Rencontres Internationales d’Archéologie et d’Histoire d’Antibes: “Des hommes et des plantes: exploitation du milieu et gestion des ressources végétales de la Préhistoire à nos jours”. Editions APDCA, Antibes, 379–390. Aumeerudddy-Thomas, Y., Bailly, A., Alleaume, S., and Hmimsa, Y., 2016. Grafted oleaster- olive agrosylvopastoral systems in Northern Morocco. In S. Thiébault and J.-P. Moatti, eds. The Mediterranean Region Under Climate Change: A Scientific Update . IRD Edi- tions, Marseille, 523–532. Aumeeruddy-Thomas, Y., Caubet, D., Hmimsa, Y., and Vicente, Á., 2017a. Les sociétés jbala et la nature. Parlers et relations à autrui dans le Rif, nord du Maroc. Revue d’ethnoécologie [online], Supplément 1. http://ethnoecologie.revues.org/3234 Aumeeruddy-Thomas, Y., Hmimsa, M. A., and Khadari, B., 2014. Beyond the divide between wild and domesticated: Spatiality, domesticity and practices pertaining to fig (Ficus carica L.) and olive (Olea europaea L.) agroecosystems in Morocco. In Eds Alex Chevalier, Elena Marinova and Leonor Peña−Chocarro, eds. Crops and People: Choices and Diversity Through Time . Earth EU, Brussels, OXFAM, London, 191–197. Aumeeruddy-Thomas, Y., Moukhli, A., Haouane, H., and Khadari, B., 2017b. Ongoing domestication and diversification in grafted olive – Oleaster agroecosystems in Northern Morocco. Regional Environmental Change 17, 1315–1328. Barazani, O., Westberg, E., Hanin, N., Dag, A., Kerem, Z., Tugendhaft, Y., Hmidat, M., Hijawi, T., and Kadereit, J.W., 2014. A comparative analysis of genetic variation in rootstocks and scions of old olive trees, a window into the history of olive cultivation practices and past genetic variation. Plant Biology , 14, 146. Besnard, G., El Bakkali, A., Haouane, H., Baali-Cherif, D., Moukhli, A., and Khadari, B., 2013b. Population genetics of Mediterranean and Saharan olives: Geographic patterns of differentiation and evidence for early generations of admixture. Annals of Botany , 112, 1292–1302. Besnard, G., Khadari, B., Navascués, M., Fernández-Mazuecos, M., El Bakkali, A., Arrigo, N., Baali-Cherif, D., Brunini-Bronzini de Caraffa, V., Santoni, S., Vargas, P., and Savolainen, V., 2013a. The complex history of the olive tree: From Late Quater- nary diversification of Mediterranean lineages to primary domestication in the northern Levant. Proceedings of the Royal Society B , 280, 20122833. Besnard, G., and Rubio de Casas, R., 2016. Single vs multiple independent olive domestications: The jury is (still) out. New Phytologist , 209(2), 466–470. Bouby, L., Figueiral, I., Bouchette, A., Rovira, N., Ivorra, S., Lacombe, T., and Terral, J. F., 2013. Bioarchaeological insights into the process of domestication of grapevine (Vitis vinifera L.) during Roman Times in Southern France. PLoS ONE , 8(5), 1–13. Bouby, L., and Ruas, M.P., 2014. Introduction of the chapter Adding diversity. Between Occasional food and speculative productions: Diversity of fruit trees. In Eds Alex Che- valier, Elena Marinova, Leonor Peña−Chocarro, eds. Crops and People: Choices and Diversity Through Time . Earth EU, Brussels, OXFAM, London, 141–149. Bouzidi, S., 2002. Le figuier: histoire, rituel et symbolisme en Afrique du Nord. Dialogues d’histoire ancienne , 28(2), 103–120. Brun, J.-P., 1986. L’oléiculture antique en provence. Les huileries du Département du Var. Revue archéologique de Narbonaise , 15, 11–301. Camps, G., 1961. Massinissa, ou les débuts de l’histoire , Imprimerie Officielle, Alger. Delplancke, M., and Aumeeruddy-Thomas, Y., 2017. Des semis et des clones. Revue d’ethnoécologie [online], Supplément 1. http://ethnoecologie.revues.org/3166 Fig and olive domestication in the Rif, northern Morocco 195 Descola, P., 2005. Par-delà nature et culture . Gallimard, Paris. Diez, C. M., Trujillo, I., Barrio, E., Belaj, A., Barranco, D., and Rallo, L., 2011. Centennial olive trees as a reservoir of genetic diversity. Annals of Botany, 797–807. https://doi. org/10.1093/aob/mcr194 . Diez, C.M., Trujillo, I., Martinez-Urdiroz, N., Barranco, D., Rallo, L., Marfil, P., and Gaut, B. S., 2015. Olive domestication and diversification in the Mediterranean Basin. New Phytologist , 206, 436–447. El Bakhali, A., Haouane, H., Hadiddou, A., Oukabli, A., Santoni, S., Udupa, S. M., Van Damme, P., and Khadari, B., 2013. Genetic diversity of on-farm selected olive trees in Moroccan traditional olive orchards. Plant Genetic Resources , 11(2), 97–105. Genin, D., Aumeeruddy-Thomas, Y., Balent, G., and Nasi, R., 2013. The multiple dimensions of rural forests: Lessons from a comparative analysis. Ecology and Society , 18(1), 27. Gros-Balthazard, M., Newton, C., Ivorra, S., Tengberg, M., Pintaud, J.-C., and Terral, J.-F., 2013. Origines et domestication du palmier dattier (Phoenix dactylifera L.). Revue d’Ethnoécologie , 0–15. http://ethnoecologie.revues.org/1524 Haouane, H., and Khadari, B., 2011. Olive diversification process in southwestern Medi- terranean traditional agro-ecosystems. In K.E. Hummer, ed. 28th International Horticul- tural Congress on Science and Horticulture for People (IHC), International Symposium on Plant Protection/3rd International Symposium on Plant Genetic Resources, Acta Horticulturae 918 , 807–812. Hmimsa, Y., 2009. L’agrodiversité de l’agroécosystème à l’arbre: cas du Rif (Nord du Maroc) . Thèse de doctorat. Tétouan, Université Abdelmalek Essaadi – Faculté des Sciences. Hmimsa, Y., Aumeeruddy-Thomas, Y., and Ater, M., 2012. Vernacular taxonomy, classification and varietal diversity of fig (Ficus carica L.) among Jbala cultivators in northern Morocco. Human Ecololgy , 40(2), 301–313. Hmimsa, Y.,Aumeeruddy-Thomas, Y., andAter, M., 2017. Une forme spontanée de figuier (Ficus carica L.), le nābūt. Revue d’ethnoécologie [online], Supplément 1. http://ethnoecologie. revues.org/3186 Jabiot, I., 2008. Devenir fellah et se perpétuer chez les Bni Ahmed. Traversée du 20e siècle et renouvellement à chaque génération. Région Jbala, Nord-Ouest du Maroc . Mémoire de Master 2, Université Paris X, Nanterre. Jamous, R., 1981. Honneur et baraka. Les structures sociales traditionnelles dans le Rif . Cambridge University Press & Editions de la Maison des Sciences de l’Homme, Cam- bridge and Paris. Joffre, R., Rambal, S., and Ratte, J., 1999. The dehesa system of southern Spain and Portu- gal as a natural ecosystem mimic. Agroforestry System , 45, 57–79. Kaniewski, D., Van Campo, E., Boiy, T., Terral, J. F., Khadari, B., and Besnard, G., 2012. Primary domestication and early uses of the emblematic olive tree: Palaeobotanical, historical and molecular evidence from the Middle East. Biological Reviews , 87(4), 885–899. Kennedy, J., 2012. Agricultural systems in the tropical forest: A critique framed by tree crops of Papua New Guinea. Quaternary International , 249, 140–150. Khadari, B., Charafi, J., Moukhli, A., and Ater, M., 2008. Substantial genetic diversity in cultivated Moroccan olive despite a single major cultivar: A paradoxical situation evidenced by the use of SSR loci. Tree Genetics and Genomes , 4(2), 213–221. Kjellberg, F., Gouyon, P. H., Ibrahil, M., Raypond, M., and Valdeyron, G., 1987. The stability of the symbiosis between dioecious figs and their pollinators: A study of Ficus carica L. and Blastpphaga psenes L. Evolution , 41, 693–704. 196 Yildiz Aumeeruddy-Thomas and Younes Hmimsa Marinval, P., 2008. Aliment végétal de la préhistoire. In F. Hallé and P. Lieutaghi, eds. Des Hommes et des Plantes , Fayard, Paris, 20–39. Rival, L., 1998. The Social Life of Trees: Anthropological Perspectives on Tree Symbolism . Berg, New Yok. Ruas, M.-P., Ros, J., Terral, J.-F., Ivora, S., Andrianarinosy, H., Ettahiri, A. S., Fili, A., and Van Staëvel, J.-P., 2016. History and archaeology of the emblematic argan tree in the medieval Anti-Atlas Mountains (Morocco). Quarterly International , 404, 114–136. Salavert, A., 2008. Olive cultivation and oil production in Palestine during the early Bronze Age (3500–2000 BC ): The case of Tel Yarmouth, Israel. Vegetation History and Archaeo- botany , 17(suppl. 1), 53–61. Simenel R. et Aumeeruddy-Thomas Y., 2016. The argan agroecosystem. Local meanings of Argan tree and bee diversity within man-made territories. In Stéphanie Thiébault and Jean-Pierre Moatti, eds. The Mediterranean Region Under Climate Change. A Scientific Update . IRD Editions, Marseille, 543–550. Soler, C. C. L., Proffit, M., Bessière, J.-M., Hossaert-McKey, M., & Schatz, B., 2012). Evidence for intersexual chemical mimicry in a dioecious plant. Ecology Letters , 15(9), 978–985. Stordeur, D., and Willcox, G., 2016. Jerf el-Ahmar. In Youssef Kanjou and Akira Tsuneki, eds. A History of Syria in One Hundred Sites . Archaeopress, Oxford, 47–50. Willcox, G., 2014. Food in the Early Neolithic of the Near East. In Milano Lucio, ed., Paleonutrition and food practices in the ancient near east: Towards multidisciplinary- approach (Vol. XIV). SARGON Editrice e Liberia, Padova. Willcox, G., 2016. Les fruits au Proche-Orient avant la domestication des fruitiers. In M.-P. Ruas, ed. Des fruits d’ici et d’ailleurs . Omniscience, Paris, 41–54. Zohary, D., Hopf, M., and Weiss, E., 2012. Domestication of Plants in the Old World . Oxford University Press, Oxford. 11 Cooperating with the wild Past and present auxiliary animals assisting humans in their foraging activities

Edmond Dounias

mo è jεε mbεlεkɔ yiê pɔkì kɔ If you hear the lesser honeyguide singing, honey is near! (Baka proverb in southeastern Cameroon expressing evidence)

Introduction Ethnozoological literature has continually documented the incredible diversity of relationships that human beings have woven with other sentient creatures within the animal kingdom. Following a wide spectrum of scenarios that elapsed over extremely variable periods of time, some of these relationships have led to particularly achieved forms of domestication,1 whereas others were more fleeting and opportunistic. Some have probably persisted a certain amount of time without leading to domestication, whereas some were rapidly abandoned and forgotten in the meanders of the evolution of human societies. Some are motivated by practical and material uses, whereas others are carried by considerations of a spiritual, artistic or psycho-cultural nature. Intrigued by the fact that early hominids had already begun to establish privileged relationships with certain animals (for instance, scavenging vertebrates such as vultures and hyenas), many scientific disciplines – archeology, history, linguistics, population biology, functional ecology, behavioral sciences and cultural anthropology, just to name a few – have combined their efforts to try and understand the diverse trajectories that took animals from a free-living wild state to states that engaged them in diversely constrained partnerships with humans. So far, attempts to cross domestication trajectories, on one hand, with functional categories of human-animal interactions, on the other hand, have not received the attention they deserve (Clutton-Brock 1981). The aim of this chapter is to consider a particular category of interactions between humans and animals – the use of animals as assistants for foraging activities – and to look at how these animals, hereafter qualified as ‘auxiliaries’, are situated along the broad spectrum of relationships between humans and animals. A prominent feature of this category of human-animal interactions is that the auxiliary function eventually addresses a limited number of domesticated animal species. The majority of contributions 198 Edmond Dounias as assistants are in fact fullfilled by wild-born animals that are purposedly tamed in captivity to become auxiliaries. Furthermore, this category includes a smaller number of cases of human cooperation with wild, non-captive and non-tamed animals. In spite of their natural avoidance of humans, these animals deliberately assist (or assisted) humans in very specific foraging activities. Emphasis will be placed on these cases of wild auxiliary animals – be they kept captive or in the wild – in order to consider how they can enlighten our views of the history and prehistory of animal domestication.

Auxiliary animals The term ‘auxiliary’, derived from auxili ārius (assisting, aiding, helping), implies a wide range of capacities to refer to animals that assist, serve as an aid or function in a supporting capacity to humans in their production activities. ‘Auxiliary’ does not encompass situations where animals are trained as vectors of a proximate profit-making activity, such as animals in a circus or for racing, fighting, gam- bling, exhibiting, dressage and other competitions or contests. Nor does ‘auxiliary’ refer to signal guides that incidentally inform humans about the presence of a given resource, without any formal intention to attract the attention of humans. For instance, whereas dogs (Canis familiaris) and pigs (Sus scrofa) are truly auxiliaries of truffle (Tuber melanosporum ) harvesters ( Chazoule 2004 ), Suillia flies and leiodid beetles that also track the mushroom in order to lie their eggs inside it and whose presence is accordingly observed by the same truffle harvesters ( Pérez Andueza et al. 2015 ) are not auxiliaries. Tamed auxiliary animals – be they domesticated or wild-born captives – intervene in a vast spectrum of usages:

• carrying loads and doing heavy labor (horse, donkey, camel, dromedary, elephant, yak, onager, mule, llama); • serving as mounts (horse, donkey, camel, reindeer, elephant); • serving as draught animals for farming and transport (dog, horse, water buf- falo, donkey); • keeping and protecting livestock (dog); • aiding and rescuing humans (dog); • assisting disabled persons (dog); • serving as therapeutic companions (all kinds of pets); • clearing of water bodies (manatee); • hunting (dog, cheetah, caracal, raptors, chinkara, blackbuck); • fishing (otter, cormorant); • gathering (macaque, baboon, pig, dog).

The last three categories of usages – hunting, ﬁshing, gathering – are what we group under the term ‘foraging’, which refers to the acquisition from the wild of edible resources, be they of animal or vegetal origin ( Danchin et al. 2008 ). Foraging activities have been carried out without interruption from the time of early hominids Cooperating with the wild 199 up until today; they have been consubstantial of Homo sapiens evolution, both in its biological and sociocultural dimensions, over the past 200,000 years. Foraging activities remain predominant among the last modern-day hunter-gatherer societies. As will be further discussed later in the chapter, a few of these activities – to be counted on the ﬁngers of one hand – are carried out in collaboration with non- domesticated auxiliary animals. It is worth briefly mentioning warfare, which throughout the history of humankind has been a propitious context for the training of auxiliary animals, to respond to offensive as well as defensive purposes. Historians report that during the First World War alone, 14 million mammals were pressed into service, 10 million were killed and 120,000 were decorated for exploits of war ( Lasserre 2014 ).

Auxiliary domesticates for foraging activities Animal domestication occurred along variable trajectories that were contingent on various locally-shaped biological and cultural parameters. According to Zeder (2012), these various trajectories can, however, be grouped in three domestication pathways that seem to encompass the broad range of known situations: the commensal pathway, the prey pathway and the directed pathway.2 Zeder’s commensal pathway addresses animals that came into contact with humans to feed on refuse or to prey on other animals attracted to human settlement. At some point, these animals developed with their human hosts close social or economic bonds that brought them into a domestic partnership with humans. Very few animals that help as auxiliaries were domesticated along this pathway, but they include the most important and ubiquitous of all auxiliary animals: the dog. In all types of human societies, the dog is mostly used as an assistant for hunting. But it also intervenes in almost all categories and is even the sole auxiliary in several domains of intervention (for instance, aid and rescue, or assistance to disabled persons). The prey pathway proposed by Zeder likely began when humans developed hunting strategies designed to increase prey availability. Over time and under certain circumstances, these game management strategies developed into actual herd management and, eventually, the controlled breeding of managed animals. The wether (Ovis aries ) trained to lead a sheep flock ( Tani 1989 ) and the reindeer ( Rangifer tarandus) that initially served as mount in hunting activities (Clutton- Brock 1981 ) before becoming a major source of meat ( Stépanoff et al. 2017 ), are prominent examples of auxiliary animal species that were domesticated along this pathway. The directed pathway is viewed by Zeder as a fast track to domestication that begins when humans use knowledge gained from the management of already domesticated animals to domesticate a wild species that possesses a desirable resource. Good examples of auxiliary animals that were domesticated following this pathway are equids – the horse (Equus caballus ), the donkey (Equus asinus ) and the onager (Equus hemionus) – that are used as mounts during hunting expeditions, even though their service as mounts is far from being limited to foraging activities. 200 Edmond Dounias Auxiliary animals among tamed wild-born captives Many more auxiliary animals are found among wild-born animals that are tamed in captivity. These animals are captured and deliberately trained following a human-directed conditioning to become tolerant of humans and to respond favor- ably to their human master during their activities as auxiliaries. But, by contrast to domesticated animals, these tamed captives are not bred and have not undergone any genetic modification aiming to foster the inheritance of favorable predispositions towards humans, even though there is increasing evidence that a potential for domestication may occur as a by-product among wild animals being bred in captivity ( O’Regan and Kitchener 2005 ). Several birds of prey, ungulates, elephantids and felid carnivores have been trained as hunting assistants. Traced back in Central Asia to the 2nd millennium BC ( Soma 2012), horseback falconry is still actively performed today. Golden eagle (Aquila chrysaetos ), northern goshawk (Accipiter gentilis ), European buzzard (Buteo buteo ) and several falcon species (Falco cherrug , F. peregrinus , F . vespertinus ) are obligate partners used by Kyrgyz and Kazakh equestrian hunters to track various mammal prey including hares (subgenus Eulagos of the genus Lepus ), red and corsac foxes (Vulpes vulpes , V. corsac ) and gray wolves (Canis lupus ). Tamed blackbucks (Antilope cervicapra ) and chinkaras (Gazella bennettii ) assisted Indian hunters as decoys, following a somewhat unusual yet clever technique: a tame antelope was sent into the wild herd with nooses attached to its horns. This unwanted guest would inevitably be provoked in fight and hunters would easily capture the wild fighter whose horns were entangled with those of the tamed fighter ( Menon 2000 ). African elephants (Loxodonta africana) and Asian elephants ( Elephas maximus ) were tamed predominantly as mounts for war, for the carriage of heavy loads and, in the early 20th century, for serving as draught animals in colonial farms ( Bannikov and Popov 2014 ). This practice, still vivid in Asia (Lainé, this volume), has been abandoned in the African continent. Although attested in the Congo Basin (Bennett 1957), taming of elephants in Africa was predominantly located in the northern part of the continent. Elephants were rarely used for foraging activities, except maybe as log carriers in artisanal logging. As mentioned earlier for blackbucks and chinkaras, tame female elephants serve as decoys in Sri Lanka, India, Myanmar, Cambodia and Thailand, to attract wild individuals into places where they can more easily be trapped (Baker and Manwell 1982). Hunting with the help of felids is no longer practiced. The asiatic cheetah (Aci- nonyx jubatus venaticus) – now declared critically endangered in the International Union for Conservation of Nature (IUCN) red list of threatened species – has been extirpated from nearly all of its range (Nowell and Jackson 1996). A major cause advanced for this extirpation is the live capture of cheetahs to be trained for sport hunting of deer and gazelle or to be kept as pets (Divyabhanusinh 2000). This has also been the fate of the caracal (Felis caracal ), which was similarly tamed for hunting in India and ancient Egypt ( Sunquist and Sunquist 2002 ). Cooperating with the wild 201 In most cases, tamed mammal captives assisting the hunts were luxury goods that only rich members of the aristocracy and monarchy – in ancient Egypt and Assyria, the Mogul Empire of Central Asia, African kingdoms, medieval Euro- pean elites – could possess. The fall of each of these various dynasties at some point in history naturally caused the subsequent loss of these collaborations. In contrast to hunting, fishing activities benefited from the assistance of tamed captive animals that were not in the exclusive hands of the elites. Their accessi- bility to poorer traditional fishers may explain why these partnerships have more frequently persisted until today. Two major auxiliaries tamed for fishing are worth mentioning. The first remarkable one is the smooth-coated otter (Lutrogale per- spicillata), whose use as an assistant in fishing is still in vogue in southern Ban- gladesh (Feeroz et al. 2011). The oldest records of this practice are situated in the Yangtze River in China during the Tang dynasty, 6th century AD ( Simoons 1990 ). In India, otter fishing was practiced in the Indus and Ganges river basins, in Ben- gal and in southern India along the Coromandel Coast. Otter fishing is reported to have existed in Central and South America (Gabriel et al. 2008); it was also known in Europe (Svanberg et al. 2016) from as early as the 15th century, with a first mention in the British Isles that is dated 1480 ( Walton 1653 ). The second astonishing fishing auxiliary is the great cormorant (Phalacrocorax carbo). The first scientific records of this partnership were published in the late 1920s, and according to Laufer (1931), the earliest historical mention of the use of tame cormorants is dated AD 607. Great cormorants were bred and reared in captivity. The eggs of captive cormorants were rapidly taken away from their mothers, transferred for brooding to domestic fowl, and the young birds benefited from a special diet. Their intensive training could take up to eight months before they became fully obedient and docile. Auxiliary cormorants have a ring placed around the base of their neck that prevents the deglutition of large fish. After catching a fish, the bird is forced to disgorge it. Egremont and Rothschild (1979 ), who provide the first detailed description of the fishing procedure, evoke, however, a curious and yet unexplained limitation to the cormorant’s total submission to their fisher master: as if they were able to count, the cormorant will refuse to obey after having captured seven fish, if its master does not remove the ring and let the bird have access to a reward (by letting it eat the eighth prey or fish for itself). The sole case currently known of gathering activity in which a tamed captive animal operates as an auxiliary is that of coconut harvesting from the tops of planted coconut palm trees with the help of the southern pigtailed macaque ( Macaca nemestrina). A salient aspect of this Southeast Asian activity – practiced mainly in Indonesia, Malaysia and Thailand – is its strong integration into com- merce of a globally important commodity. The coconut market is so lucrative that coconut-picking monkeys have acquired a commercial value per se . In Thailand, the price for a trained coconut-picking macaque is nowadays negotiated between $1,450–2,900 US ( Florescu 2014 ). A past gathering activity also involving a primate as auxiliary animal is depicted on a few Egyptian tombs by paintings that illustrate the use of Hama- dryas baboons ( Papio hamadryas ) as harvesters of figs (Ficus sycomorus ) and 202 Edmond Dounias doum palm nuts (Hyphaene thebaica ). However, evidence that the baboon was purposely trained to accomplish this fruit picking for a human master remains questionable ( Gudger 1923 ).

Non-captive wild auxiliary animals The functional category of auxiliary animals provides rare but enlightening cases in which the animal partner is neither a domesticate nor lives under human care in captivity. Although not tamed, these animals may have developed some tameness predispositions, in the sense that they may occasionally be welcoming towards the presence of humans or even towards an interaction with them ( Geist 2011 ). We have been able to identify only four attested cases: two concern marine and riverine mammals that belong to the sub-order of toothed whales (Odontoceti) and that are represented by the Iniidae (a family of nearly extinct river dolphins) and the Delphinidae (the family of currently persisting orcas and dolphins); the two other cases concern bird species belonging to two very distinct orders: Passeriformes (the raven, Corvidae ) and Piciformes (the honeyguide, Indicatoridae ).

Marine and riverine mammals: dolphins and orca

Dolphins Several species of marine and riverine dolphins of the genera Inia , Tursiops , Sota- lia and Orcaella collaborate with fisherfok. This collaboration has been reported in distinct parts of the world:

• with the bottlenose dolphin (Tursiops truncatus) along the Mauritanian coast (Busnel 1973; Pelletier 1975 ), in the Mediterranean Sea as described very early by Pliny the Elder (Bostock and Riley 1855), and in Laguna (Pryor et al. 1990; Peterson et al. 2008) and Barra de Imbé/Tramandai (Zappes et al. 2011) in southern coastal Brazil; • with the tucuxi dolphin (Sotalia fluviatilis) in Santa Catarina, Brazil (Monteiro- Filho 1995 ); • with the Amazon river dolphin (Inia geoffrensis) in the Araguaia River, state of Tocantins, Brazil ( Gravena et al. 2008 ); • with the Irrawaddy dolphin (Orcaella brevirostris) in Myanmar (Smith et al. 2009 ; D’Lima et al. 2014).

Collaboration with marine, bay and estuary dolphins is mainly related to the seasonal migrations of mullets, a family of fishes (Mugilidae) that share the behavior of leaping out of the water to escape predators. The massive and dense concentration of fish following a typical ‘predator satiety’ strategy ( Holling 1965 ) is a strong incentive for top predators to converge and merge their forces in order to catch a maximum number of prey in a minimal amount of time. The collaboration between fishers and dolphins is fleeting, and both sides react promptly to the signs Cooperating with the wild 203 sent by their ephemeral partners. In these opportunistic situations, both partners participate in numbers. The fishers possess an extensive panel of techniques to emit acoustic signals and attract the attention of the dolphins. In return, dolphins use various gestural signs to alert the fishers when a school of fish is entrapped. River dolphins ( Inia geoffrensis in Brazil, Orcaella brevirostris in Myanmar) take part in more regular collaborations to capture a much broader range of fishes and crustaceans. River dolphins generally live as single individuals or couples, more rarely as small groups. They are also more territorial than their marine relatives. For these various reasons, the partnership with the fishers is more intimate and is backed up by a more sophisticated communication between partners (cf. Table 11.1). Dolphins come into closer contact with the fishers and their boats. Individual dolphins are more frequently personified with a name and can easily be recognized by the shape of their fins, skin pigments and scars, or their personal behavior. Some dolphins really act as pets, seemingly appreciating direct physical contact with their fishers. In Myanmar, the proximity between river dolphins and their human partners even led to situations in which river fishers claimed exclusive associations with particular dolphins. Some fishers would even bring conflicts into native courts to recover a share of the fish captured by a rival fisher with the alleged help of the claimant’s dolphin. Collaborations with dolphins have shaped positive perceptions of the cetaceans by the local fisher communities with whom they interact. The intelligence and sense of sharing of dolphins are strongly put forward and mediated by local mythologies that frequently cast dolphins as symbolic emissaries between humans and supra- natural forces (Patel 1994). The fact that resources are shared with dolphins has also contributed to draw visions of the world in which shared spaces between humans and non-human sentient creatures strongly drive local cultural identity. Nevertheless, collaboration between dolphins and fishers is globally unstable and uncertain, particularly with marine dolphins. Sometimes, dolphins do not respond to the calls of the fishers; in other occasions, their unwanted presence will disturb fishers and even compromise their fishing success. When it is effective and successful, this joint exploitation of fish resources fluctuates between an opportunistic commensal relationship and an interspecific collaboration.

Orcas The orca is one of earth’s most intelligent animals. It has remarkably sophisticated hunting methods, languages and cultures, and even long-term memories (Neiwert 2015). Not surprisingly, orcas play a pivotal role in the mythology and contemporary popular culture of sea-mammal hunting peoples. This role reflects a close relationship of mutual exchange between humans and orcas, especially in Norway and the Chukchi Peninsula (Bering Strait between Alaska and Russia). Holzlehner (2015), who has traced human-orca relationships along the Pacific Rim, gathered distinctive epistemologies among whaler communities, which draw, however, on surprisingly similar mythological substrates in places as diverse as Tierra del Fuego (Argentina), Peru, Alaska, Okhotsk Sea (Russia) and Hokkaido (Japan). 204 Edmond Dounias Collaboration of humans with orcas as auxiliaries for whale hunting was less frequent than that with dolphins for mullet fishing, and certainly less resisted change, but it was based on a more solid partnership and greater involvement of the non-human partner. It is in Twofold Bay (southeastern Australia) that Clode (2002) compiled the most detailed descriptions of the peculiar behavior of orcas as whale- hunting auxiliaries. Between 1840 and 1930, orcas would prowl the entrance of the bay, and ambush migrating humpback (Megaptera novaeangliae ), blue (Balae- noptera musculus), southern right (Eubalaena australis) and minke (Balaenoptera acutorostrata ) whales. Working in groups, with individuals taking on different roles, orcas would shepherd the whales into the bay. Ripping at fins and diving over the whales’ blowholes, they would take advantage of the bay’s unique geography and force the whales into shallower waters. While the pack kept the whales cornered, one of the orcas would station itself at the mouth of the river near the whaling station and would attract the whalers’ attention by breaching and lashing the water with its tail. The whalers would finish the whales off with harpoons. Some witnesses report how the predators towed the whaleboats to the flailing whales by tugging ropes with their teeth. Once a whale was dead, whalers would let orcas feast on the lips and tongue, before hauling the carcass ashore. Some of the orcas would even grab the ropes in their teeth and aid the whalers in hauling. The orcas would then also feed on the many fish and birds that congregated. The partnership always involved the same pod of orcas, most of which could easily be recognized by the shape of their dorsal and caudal fins. In Twofold Bay, all orcas were personified by names given after whalers who had died. The most popular orca, named ‘Old Tom’, was the very distinctive male in charge of alert- ing the whalers.

Terrestrial birds

Ravens Corvids (jays, jackdaws, magpies, crows, rooks and ravens) are large-brained social animals that understand their physical and social worlds ( Marzluff and Angell 2005 ). They are capable of causal reasoning, flexibility, imagination and prospection, and elaborate solutions to social and physical problems. They share these complex cognitive abilities with monkeys and apes, which have for long been the preferred subjects of studies of the evolution of intelligence because of their close evolutionary relationship to humans (Seed et al. 2009). Ravens ( Corvus corax ) are ubiquitous and omnipresent year-round human settlements in the North Pacific. They are generalist feeders and predators of small animals, as well as scavengers on carcasses of salmon and large mammals. Ravens also have kinship-based social systems: they are monogamous and live in extended family groups that show complex networks of cooperation and strong social hierarchy ( Heinrich 1999 ). Ravens’ vocalizations and gestural playfulness are the most immediately per- ceptible expressions of a sophisticated language (Pika and Bugnyar 2011) that is Cooperating with the wild 205 used not only to communicate with conspecifics, but also with mammal partners. In its intelligence and its faculties for communication and social interactions, the raven appears very humanlike and is cast as a central mythological figure by North Pacific first nations ( Munday 2013 ; Nelson 1983). It symbolizes mystical connections between native peoples of northwestern North America – The Koyukon, the Tlingit, the Haida – and their natural world, and is a pivotal element of the vibrant Inuit ontological system, their understanding of animals in relation to the wider environment and the cultural substance of being a hunter. For the Inuit, the raven has many social roles: as creator of the present world, as trickster and possessor of knowledge, and as a creature that can be helpful at times, but dangerous at others ( Laugrand and Oosten 2015 ). According to myths of the Chukchi and the Koryak, it is the raven that brought fire to humans ( Bogoras 1904 ; Serov 1988 ). Among the Yukaghirs of Siberia (Russian Republic of Sakha), the raven is considered as a person ( Willerslev 2007 ). This prominent cultural value attributed to the raven is probably a long story: the burial conditions of two raven skeletons that were excavated from Charlie Lake Cave (British Columbia) clearly suggest that the two birds were deposited deliberately by the Paleoindian occupants, whose presence was dated at about 10,500 BP and 9,500 BP ( Driver 1999 ). Besides hunting cooperatively in groups with each other, especially for small prey such as squirrels (Heinrich 1999), the raven is a prominent scavenger of ungulate kills made by gray wolves (Canis lupus ). Ravens call wolves to dead animals so they will make the carcasses more accessible to the birds ( Zahara and Hird 2015 ). They are also quick to locate and harass an injured wapiti (Cervus elaphus ) and draw the attention of wolves ( Stahler et al. 2002 ). This relationship seems not to be just an incidental and proximate by-product of the presence of fresh meat. Instead, ravens seem to look for the companionship of wolves independently of the presence or the absence of food ( Kaczensky et al. 2005 ). Vuce- tich et al. (2004) demonstrate that the presence of scavenging ravens leads to an increase in gray wolves’ group size and to higher per capita gains in the largest observed packs. Inuit hunters also reported that ravens lead polar bears (Ursus maritimus) to dead seals (Phocidae) (Heinrich 1999): as they do with gray wolves, ravens accompany the bears and scavenge leftovers from the carcasses. In the Pacific Northwest of North America, the raven is the most easily visible animal in natura . But it is also the most highly salient in local culture, and has often formed complex relationships with hunters. The Koyukon say that ravens bring luck if sighted during a hunt and will lead hunters to their prey by dipping a wing. Koyu- kon hunters look to the flight of a passing raven as a sign of whether the hunt will be successful ( Nelson 1969 ; Heinrich 1999 )3 . Among the Inuit, newborn boys are clothed in raven skin to help them become successful hunters ( Munday 2013 ). The Inuit also mimic the raven’s dance to attract polar bears in hunting ( Munday 2013 ). Wolves are said to exploit types of prey similar to those exploited by humans. Interestingly, where wolves are abundant, ravens are not considered to have close relationships with humans or to be equivalent to humans. Conversely, ravens more often assist humans in locating prey and are more embedded with mystical functions in places where wolves are less abundant (Pierotti 2011). This gives 206 Edmond Dounias credence to Heinrich’s assumption (1999 ) that humans, as later-arriving predators who also usually hunt in packs, were probably just surrogate wolves, and have become substitute partners to wolves in areas deserted, or never occupied, by the canid predators.

Honeyguides Honeyguides comprise the bird family Indicatoridae, which counts four genera and 16 species, 14 of which are from sub-Saharan Africa and two from Southeast Asia. Some African honeyguides are involved in an astounding collaboration with humans. They live in forests and woodlands, but also in streamside trees near desert areas of Namibia and Botswana. The English term ‘honeyguide’ and the Latin root of the family name Indicato- ridae clearly reflect the unique habit observed in a few species, which consists in leading honey hunters to honeybee nests. This habit concerns only some African honeyguides of the genus Indicator . The guiding behavior is motivated by the fact that honeyguides have specialized in consuming beeswax. African bees often nest in inaccessible cavities and fortify the entrances to their hives, making access nearly impossible for the bird. Honeyguide species that have developed a specialized diet by feeding almost exclusively on wax must team up with a partner that can calm down the bees and that is equipped to break into the hives. Only humans meet these requirements. There is so far no evidence of guiding behavior by either of the two Southeast Asian species of Indicator nor by any species of the other three genera, Melichneutes , Melignomon and Prodotiscus . These all have more generalized diets. Although they eat beeswax, all these honeyguides also eat small scale insects4 as well as other insects such as caterpillars, flies, bugs, termites, ants, beetles, aphids, grasshoppers and butterflies. Partnerships between honey hunters and some honeyguide species are thus highly conditioned by the predominance of beeswax in the bird’s diet. Wax-eating by honeyguides is apparently made possible by the bird’s mutualistic relationship with bacteria located in its gut. Friedmann and Kern (1956) suspected Micrococcus cerolyticus and Candida albicans to be the symbionts facilitating wax digestion, but this assumption was later questioned (Downs et al. 2002 ) and is currently still debated. Guiding behavior has been most frequently described in the greater honeyguide ( I. indicator ). Friedmann (1955 ) reports that the East African scaly-throated honeyguide (I. variegatus) displays a similar guiding behavior, but other authors con- test this assertion (Short and Horne 2001). Guiding behavior has also thus far not been documented, apart from our own observations (Dounias 2009), in the Congo Basin rainforest species of Indicator , the lesser honeyguide (I. minor ) and – with less certainty – the least honeyguide (I. exilis) and/or Willcocks’s honeyguide (I. willcocksi). The two latter species are peculiar in feeding on sticky wax exudates produced by stingless bees (Apidae: Meliponinae), and the Baka hunter-gatherers of southern Cameroon will look for hives of stingless bees when they hear the call of this bird that they name ‘stingless bee honey bird’ . Least and Willcocks’s Cooperating with the wild 207 honeyguides are regularly confused with each other, and data are lacking to clarify which of these two birds, if not both, is the right guide. Nevertheless, since these two birds are generalized in their diet, their accidental collaboration with honey hunters may only reflect an opportunistic commensal relationship. Numerous and detailed descriptions of greater honeyguide behavior are available in the literature to ascertain the sophistication of the efforts made by the bird to gently guide the honey hunters. These efforts combine cyclic vocalizations, undulating flight, tail feather spreading, and perching in trees in order to wait for the human followers. They are modulated according to distance to the hive and difficulties imposed by the terrain. Hunter-gatherer and pastoralist societies such as the Maasai, the Borana, the Yao, the Hadza and so-called ‘Pygmy’ and ‘San’ societies have each developed their own mode of communication with the honeyguides of their territories. Interestingly, the sounds produced by these various human societies differ significantly – the Borana use calling devices made of clasped fists, snail shells or doum palm nuts (Hyphaene thebaica); the Maasai, the Hadza and the San whistle, the Yao emit a loud trill followed by a grunt, the Baka shout; in each case, the bird has adapted its capacity to respond to the specific calling procedure. None of these honey-hunter societies try to reproduce in any way the vocalization of the honeyguide. Instead, they have engaged in elaborat- ing a communication system that is specific to them. Spottiswoode et al. (2016) demonstrate that the bird will less efficiently guide the Yao honey hunters if these hunters emit a sound that differs from their usual call. The help of honeyguides multiplies the probability of locating a bees’ nest by a factor of four or five for the Borana of Kenya (Isack and Reyer 1989), the Hadza of Tanzania (Wood et al. 2014 ), and the Yao of Mozambique ( Spottiswoode et al. 2016 ).

Discussion Figure 11.1 synthesizes per category of use the various auxiliary animals – domesticates as well as non-domesticated – that we managed to census, according to the various domestication pathways. However, a few animal species may have escaped our attention.

Interspecific communication While it is far from our intention to minimize the importance of communication in the relationships between humans and their domesticated animals, it certainly also plays a prominent role in the interspecific cooperation between humans and their non-domesticated auxiliary animals. It is highly likely that the wild auxiliary animal took the leading role in engaging in the partnership and was the more active in establishing the first terms of communication that transformed the simple converging exploitation of the same resource into a more elaborate cooperation. The necessity that the auxiliary animal takes the first step may help explain why such elaborate partnerships between humans and wild animals are extremely rare. Figure 11.1 Domesticated and non-domesticated auxiliary animals assisting humans in their foraging activities. Credit: Edmond Dounias. Cooperating with the wild 209 Birds are likely candidates for wild species apt to initiate communication with humans. They combine the capacities to sing, to fly and to interact with a great diversity of life forms (Terashima 2007). In most nature-dependent human societies, birds have a privileged position in cosmogonies and folk beliefs: the combination of the three capacities mentioned previously inevitably cast birds as emissaries between humans and the supra-natural forces that are the masters of natural resources (Ichikawa 1998). In animist religions, gaining the good will of these spirits is a prerequisite before capturing or killing a wild creature. The presence of dolphins and orcas among the few wild auxiliary animals certainly has to do not only with their capacity for echolocation, which is nothing short of a sixth sense ( Neiwert 2015 ), but also with their ability to perform very explicit gestural dances – nodding head, shaking fins, leaping out of the water. Although not evoked in the descriptions of these partnerships, the capacity of toothed whales for vocal sounding and teeth chattering should also not be ignored. As summarized in Table 11.1, hunter-gatherers and fishers of mullet are quite proactive in their communication with honeyguides and dolphins, respectively, whereas whale hunters and North Pacific ungulate hunters are much more passive vis-à-vis orcas and ravens, respectively, which spare no effort explicitly to catch the attention of their human partners. The degree of reciprocal communication is not correlated with the degree of sophistication of the interactions, although the most sophisticated honeyguide-honey hunter partnerships are also those in which reciprocal signaling is the best developed. While communication between mullet fishers and dolphins appears seemingly advanced, their collaboration never reaches the level of symbiosis abusively reported by Busnel (1973) and Cousteau and Diolé (1975 ), and rightly contested by Robineau (1995 ).

Solitary versus grouped partners A corollary aspect to communication that is worth considering is the solitary versus group habits of the wild partners and the incidence of this distinction on the degree of interactions. Dolphins, orcas and ravens are certainly acknowledged for their high level of intelligence and for their sophisticated social relations, but they are surpassed by the honeyguide in terms of intricacy of relationships with the human partner. The most elaborate mutualistic relationships with a wild animal auxiliary occur when the animal contributes as an individual, not as a member of a pack. Dol- phins perfectly illustrate the contrast between isolated versus in pack assistance: marine dolphins that are collectively engaged in mullet fishing in cooperation with groups of humans are not always trustworthy partners in comparison to the riverine Irrawaddy and Amazon dolphins, which have more personalized ties with individual fishers. Honeyguides are always solitary when they interact with honey hunters. The situation is more equivocal for orcas and ravens: although orcas always operate in packs, one member of the pack is in closer thus more individualized interaction with humans when it comes near the whale station to catch the attention of the whalers; although the scavenging birds always hunt in Table 11.1 Profiles of wild and untamed auxiliary animals assisting humans in their foraging activities.

Wild Number of Geographic Foraging Target Payoff (P), Intervention Communica- Communica- Personifica- Resilience of untamed involved location of reported activity resource Reward (R) by auxiliary tion mode tion mode by tion of animal cooperation auxiliaries species * cooperation by animal humans partner Aquatic Marine/ 2 Atlantic coasts, Gillnet and Mullet P Group Gestural Shouting/Water + Persisting mammals estuarine Mediterranean Sea, cast net Mullet drumming or only in dolphins Bay of Bengal, fishing slapping/Boat Brazil and Coral Sea side tapping Mauritania

Riverine 2 Amazon and Gillnet and Various P + R Individual or Gestural Shouting/ +++ Almost dolphins Ayeyarwady rivers cast net fishes and Various small group Guttural extinct in fishing crustaceans fishes and calling/Water Amazon; crustaeans drumming or Receding in slapping/Boat Irrawady side tapping

Orcas 1 Pacific Rim Hunting Whales P Individual Gestural None +++ Extinct Whale lips Group ** and tongue

Birds Ravens 1 Northern Pacific Hunting Ungulates R Individual Gestural & None/Gunshot +++ Persisting Carcass Group ** vocal

Honeyguides 3 or 4 Sub-Saharan Africa Gathering Wax (bird); P + R Individual Gestural & Whistling/ ++ Receding Honey Wax and bee vocal Shouting/Wood (human) larvae knocking/ Calling device

* See Figure 11.1 for species names ** Hunting in packs, but communicating individually with human partners Cooperating with the wild 211 great flocks, engagement into communication with the hunters is most of the time performed by individual birds. The fact that orcas and ravens hunt in groups but individualize communication with their human allies deserves further investigations. In both cases, however, communication is a one-way dialogue in which hunters seldom ‘exchange’ with the cetaceans and the bird. Exception made of human-marine dolphin interactions that occur in groups, individualized partnerships are often reflected in a personification of the animal partner through the attribution of a name. Naming reinforces the psycho-cultural bonds that sometimes result in friendship, complicity, mutual respect and exclusivity of service between allies. Personhood of wild animals is a common cultural trait among hunter societies of the North Pacific and Siberia ( Willerslev 2007 ), and is accordingly more pronounced for ravens and orcas.

Payoffs and rewards In the relationships between marine dolphins and orcas auxiliaries and humans, the animal partners directly receive a benefit from performing the activity: dolphins eat mullets while pushing them to the fishers’ nets; orcas devour the tongue and lips of the freshly harpooned whale. In these cases, the payoff is a by-product benefit: there is no cost to cooperating with a partner, and therefore no selective incentive to cheat ( Mayer et al. 2014 ). Benefits obtained are slightly different for auxiliary honeyguides, ravens and river dolphins. Honeyguides and ravens will get their share only after humans have finished taking their part: honeyguides gain a direct benefit by their capacity to take advantage of fire and smoke that keep the bees calm or distant, and to penetrate into the nest cavity left open by the honey hunters. But in supplement, they will also receive honeycombs as a gift that is decided and controlled by the human partner. We term this gift ‘reward’ by distinction with ‘by-product benefit’ defined previously. Ravens will feed on the carcass that is deliberately left behind by the hunters. This is again a reward that is ‘offered’ to the animal partner, since the hunters can easily decide to withhold access to a carcass. River dolphins also receive a reward as a supplement to the by-product benefit they obtain in the course of cooperative fishing: when the catch of fish is over, auxiliary dolphins come in close contact with the boats to receive extra fish from the hands of the fishers. Long after the fishing, dolphins may also obtain leftovers of fish being cut and eviscerated on the shore by women processing fish for cooking. This distinction between by-product benefit of the mutualistic partnership and reward is reflected in mutualism theory (Bronstein 1994). A significant proportion of the total benefits gained by the wild animal auxiliary often remains under the arbitrary decision of the human partner. Mutualism theory predicts that under such circumstances, the ‘controlling’ partner may reduce the cost of the relationship by withholding or reducing rewards, i.e., by ‘cheating’. However, this risk is mitigated by cultural safeguards: wild animal auxiliaries occupy a pivotal position in the cosmogony of the human societies that they interact with, and humans 212 Edmond Dounias believe they will be exposed to sanctions by supernatural entities if they do not reward their animal partner. Among hunter-gatherer and pastoralist societies that attribute great value to honey, similar due respect for the honeyguide is an intangible rule, and rewarding the honeyguide with a gift of honeycomb is a constant concern. The Maasai exhibit a reverential attitude when they deposit portions of combs for the bird. For the San, honey that is obtained with the help of the honeyguide provides power for taming the animal world and ensuring rain. The San believe that not thanking the bird may lead its follower to a lion, bull elephant, or venomous snake as punishment (McGovern 2009). Similar tales of misfortune befalling honey hunters who do not repay the honeyguide are found among various Pygmy groups (Bergier 1941). Although not explicitly reported, it is likely that fishers and hunters who see dolphins, orcas and ravens5 as imbued with specific cultural attributes similarly believe they will be exposed to punishment if they offend these animals. Interestingly, rewarding the honeyguide has become a critical yet revealing issue among the Hadza. These hunter-gatherers of Tanzania were observed to thank the greater honeyguide in the past6 . They used to explicitly throw the combs onto the ground in the direction of the bird perched in a nearby tree and waiting for its share. Today, they no longer reward their honeyguides. Worse, they even withhold by-product benefits by purposely burning the comb leftovers as a means to keep the bird hungry and consequently ensure its collaboration (Wood et al. 2014 ). Clarifying the reasons for such a change in Hadza attitude and the consequences of withholding reward (and even by-product benefits) on the fate of this mutualistic cooperation requires further investigation. Rewarding is present in the cooperation with auxiliary domesticates, and is even a strong incentive for tamed captives that assist in foraging activities. It is a key component of the training stage during which humans establish their master- ship. However, as evoked with the great cormorant, the reward may sometimes be demanded by the tamed partner, otherwise it may no longer obey its master. The only attested case of a tamed auxiliary that is engaged in a foraging activity without being rewarded is that of the southern pigtailed macaque for coconut picking. Distinct sources concur to describe the macaque training as coercive in Malaysia and Thailand, where it is based exclusively on punishment and avoidance of punishment ( Bertrand 1967 ).

A tritrophic interaction Another important yet poorly explored aspect of the interspecific partnership between humans and a non-domesticate auxiliary animal is that it constitutes a tritrophic interaction. This three-poled relationship requires further investigation because one out of the three interactions is generally overlooked: the relationship between the human forager and the coveted resource is generally well documented; lesser known are the links that tie human foragers with their auxiliary animals; most ignored in the triptych is the relationship between the auxiliary animal and the resource that it helps humans obtain. Beyond the trivial acknowledgement Cooperating with the wild 213 of a predator-prey link between orcas and whales, between dolphins and mullets, between ravens and the scavenged dead or weakened animal, between the honeyguides and honeybees, much still needs to be explored to better clarify the motivation of the auxiliary animal for initiating a partnership with humans. In the case of the best developed human-honeyguide partnership, the capacity to use fire confers a decisive advantage to human honey hunters in comparison to the few other mammals that look for honey (genet, mongoose, baboon, chimpanzee, bonobo, honey badger), and that could thus be potential candidates for a partnership. Non-human honey consumers generally enhance the honeybees’ aggressiveness, and this is counterproductive for the honeyguide, which is sensitive to bee stings. Fire and the smoke it produces keep the honeybees distant and reduce their aggressiveness. A final interesting aspect of the more elaborate cooperation between honey hunters and honeyguides is the fact that the humans and the birds are not interested in exactly the same resource. Honeybees constitute the third member of the tritrophic relationship, and they are not directly sought as prey, but rather for what they produce. And whereas humans are interested in honey, the honeyguide is focused on wax. The mutualistic relationship is facilitated by the absence of direct competition between the two partners: each obtains a resource that is of minor interest for their ally.

Do cases of wild and untamed auxiliary animals prelude domestication processes? The possibility that domesticated animals serving as auxiliaries could originally cooperate with humans prior to their domestication should not be excluded. Many herder societies of tundra reindeer acknowledge that wolves help keeping together their flocks and tolerate in return wolf-kills of the weakest reindeer (Stépanoff et al. 2017 and references herein). Such auxiliary-sounding scenarios may possibly have taken place around wild reindeer flocks towards the Pleistocene and accordingly influenced the dog domestication process: as pointed out by Germonpré et al. (this volume), wolves assisted to drive scared game to the hunters. Similarly, the concomitant appearance of cats and rodents around grain silos during the Middle Ages is probably not neutral in the process of cat domestication as a means to control pests in an emerging context of crop storage. One may wonder, however, what benefits non-domesticates could conceivably gain in voluntarily collaborating with humans outside of a captive situation. Domesticated reindeer provide a good illustration of such benefits: they are adamant to breeding in captivity and most traditional herders, like the Tozhu of the Sayan Mountains (southern Siberia), do not watch their herds. Reindeer find a secure advantage in staying and collaborating with herders who satisfy their fondness for salt and human urine. In the absence of this payoff, bred reindeer would refuse to be drown back to the camp and would instead join wild herds (Stépanoff et al. 2017). Nevertheless, none of the cooperation with wild and untamed auxiliary animals detailed previously should be considered as a vestige of domestication attempts 214 Edmond Dounias or as a state of domestication in progress. Instead, the cases described should be viewed as remarkably well-developed exceptions that have survived the passing of time and persisted until recently. Three out of the four cases – dolphins, orcas, ravens – concern animals that are remarkably intelligent, live in kin-based social groups and have acquired the capacity to hunt in socially organized bands and flocks. The fourth case is unique in that honeyguides have conquered a specific dietary niche based on eating wax. They have no competitors for this resource, which can only be digested by specialists, and have developed two mutualistic relationships to achieve such exclusivity: one with bacteria hosted in their gut to facilitate wax digestion; the other one with humans as a means to gain access to wax, taking advantage of this partner’s interest in a resource associated with wax: honey.

Conclusion Animals that assist foraging human societies as auxiliaries provide new insights into the history of domestication processes but more importantly into human relationships with wild-born animals, whether these are kept in captivity or not. The category of tamed wild-born captives has provided the largest diversity of auxiliary animals. Captivity concerns just a few individuals taken out of a wild population. These tamed individuals never alter the behavior of their wild relatives vis-à-vis humans, even after their eventual release from captivity. None of the animal species that formerly served as captive auxiliaries have conserved vestiges of their servile assistance to humans. The major targets of this chapter have been the few existing cases of wild and untamed animals that assist humans. These mutualistic interactions were initiated and evolved independently of domestication processes. The rare wild and untamed animal species involved as auxiliaries are remarkable for their intelligence, socialization faculties and sophisticated communication system, or for their very atypical diet. They most likely were the instigators of these partnerships. In return for their help, these animals obtain by-product benefits, sometimes supplemented by a reward, from their human partners. They legitimately occupy a lead position in the cosmogony of the foraging societies with which they interact, and these interactions thrive on an obligation of mutual respect. Unfortunately, most of these interspecific collaborations between humans and wild auxiliary animals are declining nowadays because of dramatic environmental – and sometimes social and cultural – changes. There is no evidence of recent mutual collaboration between whalers and orcas, and the fairly recent ban of whale hunting has encouraged many orca populations to adapt to new sources of food (Parsons et al. 2013). Today, it is only off the Pacific coast of North America that pods of transient orcas are still reported to attack migrating gray whales (Barrett-Lennard et al. 2011). Most marine dolphins dropped their partnership with fishers because of the rarefaction of mullet schools, the intensification of industrial fishing, the arrival of new fishing instruments replacing traditional cast-nets, increasing marine pollution and other causes. The partnership with riverine dolphins is apparently Cooperating with the wild 215 more resilient to change. This partnership has influenced the emergence of taboos against killing dolphins, and has positively engaged the human partner in preserving these aquatic animals. However, many freshwater dolphin species are nearly hunted to extinction by fisher communities that are not involved in such cooperation ( Gravena et al. 2008 ). The interactions between honeyguides and honey hunters has developed over millennia, possibly as far back as our distant hominid ancestors, and constitute the most elaborate mutualistic relationship so far established with a wild and untamed animal. However, some troubling signs indicate that this relationship is receding. On the honey-hunter side, hunter-gatherer and pastoralist societies are forced to become sedentary; they are losing their land rights or are not allowed to circulate in lands declared to be protected areas. On the honeybee side, sub-Saharan Afri- can colonies are increasingly affected by the Colony Collapse Disorder, which is a major cause of worldwide decline of honeybees. On the honeyguide side, species that are highly specialized on wax eating are more incidentally impacted by the various environmental threats affecting honeybees. The changing behavior of Hadza honey hunters who have decided to keep their guiding birds hungry is one more worrying sign that the survivorship of the bird is probably compromised. Only ravens seem to find efficient new solutions for maintaining their scavenging strategy with the help of wolf or human partners. Ravens display an apparent fear response to large carcasses that were not killed by wolves ( Heinrich 1988 ). This fear behavior is characterized by a cautious approach to a carcass and retreating without feeding. Stahler et al. (2002 ) suggest that this fear response is inhibited upon discovering large carcasses attended by wolves. Although not documented when they interact with humans, the suppression of ravens’ innate fear when they are in the company of their mutualistic partner could be a means to broaden their access to novel food sources and in adapting to a changing environment. As stated by White (2005), ravens have also learned the usefulness of gunshots and now react positively to this new stimulus, by contrast to wildlife prey that typically avoid gunshots, which to them are signs of danger.

Acknowledgements The author addresses his heartfelt thanks to Doyle McKey for his careful editing of the English writing, and for his insightful comments on the first draft. Charles Stépanoff is also warmly acknowledged for his remarks that noticeably enriched the final version of this manuscript.

Notes 1 In this chapter, animal domestication is understood as a permanent genetic modification of a bred lineage that leads to an inherited predisposition toward humans (Driscoll et al. 2009). 2 These pathways should not be viewed as mutually exclusive. For instance, horse and reindeer domestications may be considered concomitantly along the prey and directed pathways. 216 Edmond Dounias 3 Tuvan hunters of southern Siberia (Russia) similarly reported to C. Stépanoff that the raven alerts them on the presence of the game by whirling in the sky before eventually perching in a nearby tree. Among the Even people of Kamchatka in eastern Siberia (Rus- sia), hunters told Stépanoff that the raven guides them by croaking in a special manner (Charles Stépanoff, personal communication, 2017). 4 Numerous scale insects produce wax as a protective covering (Doyle McKey, personal communication, 2017). Although data are lacking concerning the part that scale insects represent in honeyguide diet, the phylogenetic distribution of honeyguide diets ( Fried- mann 1957 ) suggests the hypothesis that eating wax produced by scale insects (and possibly other sources) was the primitive condition, and that some Indicator species may have specialized on a much more concentrated wax source, honeybees’ nests. 5 After the hunt, the Tuvan hunters ritually deposit meat in the trees to the ravens, saying “Let this be the part of the birds of the mountain country!” (Charles Stépanoff, personal communication, 2017). 6 As shown in the film by Hudson and Woodburn (1966) and confirmed by James Wood- burn (personal communication, 2015).

References Baker, C.M.A., and Manwell, C., 1982. Man and elephant, the ‘dare theory’ of domestication and the origin of breeds. Zeitschrift für Tierzüchtung und Züchtungsbiologie , 100, 55–75. Bannikov, A.V., and Popov, A.A., 2014. War elephants of Indian and African origin in classical literature. Middle-East Journal of Scientific Research , 21 (2), 396–400. Barrett-Lennard, L., Matkin, C.O., Durban, J.W., Saulitis, E.L., and Ellifrit, D.K. 2011. Predation on gray whales and prolonged feeding on submerged carcasses by transient killer whales at Unimak Island, Alaska. Marine Ecology Progress Series , 421, 229–241. Bennett, C.F., Jr., 1957. A brief history of trained African elephants in the Belgian Congo. The Journal of Geography , 56 (4), 168–172. Bergier, E., 1941. Insectes comestibles et peuples entomophages . Avignon: Rullières Frères. Bertrand, M., 1967. Training without reward: Traditional training of pig-tailed macaques as coconut harvesters. Science , 155 (3761), 484–486. Bogoras, W., 1904. The Chukchee, vol. 7. New York, NY: Memoirs of the American Museum of Natural History. Bostock, J., and Riley, H.T., 1855. Pliny the Elder, the Natural History . London: Taylor and Francis. Bronstein, J., 1994. Conditional outcomes in mutualistic interactions. Trends in Ecology and Evolution , 9 (6), 214–217. Busnel, R.G., 1973. Symbiotic relationship between man and dolphins. Transactions of the New York Academy of Sciences , 3 (5), 112–l31. Chazoule, C., 2004. L’histoire inachevée de la domestication truffière. Ruralia , 15, 1–22. Clode, D., 2002. Killers in Eden: The True Story of Killer Whales and Their Remarkable Partnership With the Whalers of Twofold Bay . Crows Nest: Allen & Unwin. Clutton-Brock, J., 1981. Domesticated Animals From Early Times. London: British Museum of Natural History. Cousteau, J.Y., and Diolé, P., 1975. Les dauphins et la liberté . Paris: Flammarion. Danchin, E., Giraldeau, L., and Cezilly, F., 2008. Behavioural Ecology . New York, NY: Oxford University Press. Divyabhanusinh, 2000. The End of a Trail: The Cheetah in India. Delhi: Oxford University Press. Cooperating with the wild 217 D’Lima, C., Marsh, H., Hamann, M., Sinha, A., and Arthur, R., 2014. Positive interactions between Irrawaddy dolphins and artisanal fishers in the Chilika lagoon of Eastern India are driven by ecology, socioeconomics, and culture. Ambio , 43 (5), 614–624. Dounias, E., 2009. Keeping an Eye on the Sentinel: Mutualistic Relationship Between the Honeyguide Bird and Honey Hunters. Poster, Apimondia International Congress, Mont- pellier, 15–20 September 2009. Downs, C.T., van Dyk, R.J., and Iji, P., 2002. Wax digestion by the lesser honeyguide Indi- cator minor. Comparative Biochemistry and Physiology , Part A, 133, 125–134. Driscoll, C.A., MacDonald, D.W., and O’Brien, S.J., 2009. From wild animals to domestic pets, an evolutionary view of domestication. Proceedings of the National Academy of Sciences , 106, 9971–9978. Driver, J.C., 1999. Raven skeletons from Paleoindian contexts, Charlie Lake Cave, British Columbia. American Antiquity , 64, 289–298. Egremont, E., and Rothschild, M., 1979. The calculating cormorants. Biological Journal of the Linnean Society , 12, 181–186. Feeroz, M.M., Begum, S., and Hasan, M.K., 2011. Fishing with otters: A traditional conservation practice in Bangladesh. In Proceedings of XIth International Otter Colloquium . Gland: IUCN Otter Specialist Group Bulletin, 28A, 14–21. Florescu, B., 2014. Monkey school in Surat Thani, Thailand. The Thailander, 21 January 2014, http://thailander.eu/thailand/thai-people/monkey-training-school-surat-thani-thailand/ Friedmann, H., 1955. The honey-guides. Bulletin of the United States National Museum , 208, 1–292. Friedmann, H., 1957. Zenker’s honey-guide. Journal of Ornithology , 109, 276–283. Friedmann, H., and Kern, J., 1956. The problem of cerophagy or wax-eating in the honeyguides. The Quarterly Review of Biology , 31, 19–30. Gabriel, O., Lange, K., Dahm, E., and Wendt, T., 2008. Fish Catching Methods of the World . 4th ed. Hoboken, NJ: Wiley-Blackwell. Geist, V., 2011. Wildlife habituation: Advances in understanding and management application. Human-Wildlife Interactions , 5, 9–12. Gravena, W., Hrbek, T., Da Silva, V.M.F., and Farias, I.P., 2008. Amazon River dolphin love fetishes: From folklore to molecular forensics. Marine Mammal Science , 24, 969–978. Gudger, E.W., 1923. Monkeys trained as harvesters: Instances of a practice extending from remote times to the present. Natural History , 23, 272–279. Heinrich, B., 1988. Why do ravens fear their food? Condor , 90, 950–952. Heinrich, B., 1999. Mind of the Raven: Investigations and Adventures with Wolf-birds . New York, NY: Cliff Street Books. Holling, C.S., 1965. The functional response of predators to prey density and its role in mimicry and population regulation. Mémoires de la Société d’Entomologie du Canada , 45, 1–60. Holzlehner, T., 2015. Werewolves of the sea: Human-killer whale (Orcinus orca) relationship in the North Pacific. Oral Communication at the session “Animal auxiliaries”, 11th Conference on Hunting and Gathering Societies, Vienna, Austria, 7–11 September 2015. Hudson, S., and Woodburn, J., 1966. Hadza, The Food Quest of a Hunting and Gathering Tribe of Tanzania . Melbourne: Hogarth Film, 42 mn. Ichikawa, M., 1998. The birds as indicators of the invisible world: Ethnoornithology of the Mbuti Hunter-gatherers. African Study Monographs , supplement issue, 25, 105–121. Isack, H.A., and Reyer, H.U., 1989. Honeyguides and honey gatherers: Interspecific communication in a symbiotic relationship. Science2 , 43 (4896), 1343–1346. Kaczensky, P., Hayes, R.D., and Promberger, C., 2005. Effect of raven Corvus corax scavenging on the kill rates of wolf Canis lupus packs. Wildlife Biology , 11 (2), 101–108. 218 Edmond Dounias Lasserre, A., 2014. Le sort des animaux requis dans l’enfer de 14–18 . Paris: Editions Edilivre. Laufer, B., 1931. The domestication of the cormorant in China and Japan. Field Museum Natural History (Anthropological Series) , 18 (3), 200–262. Laugrand, F., and Oosten, J., 2015. Hunters, Predators and Prey: Inuit Perceptions of Ani- mals . New York, NY: Berghahn Books. Marzluff, J.M., and Angell, T., 2005. In the Company of Crows and Ravens. New Haven, NY: Yale University Press. Mayer, V.E., Frederickson, M.E., McKey, D., and Blatrix, R., 2014. Current issues in the evolutionary ecology of ant-plant symbioses. New Phytologist , 202, 749–764. McGovern, P.E., 2009. Uncorking the Past: The Quest for Wine, Beer, and Other Alcoholic Beverages . Berkeley, CA: University of California Press. Menon, R.K., 2000. The quintessential antelope – Life of the blackbuck. Resonance , November 2000, 69–79. Monteiro-Filho, E.L.A., 1995. Pesca interativa entre o golfinho Sotalia fluviatilis guianensis e a comunidade pesqueira da região de Cananéia. Boletim do Instituto de Pesca, 22, 5–23. Munday, P., 2013. Thinking through ravens: Human hunters, wolf-birds and embodied communication. In E. Plec, ed. Perspectives on Human-Animal Communication: Inter- natural communication . New York, NY: Routledge, 207–225. Neiwert, D., 2015. Of Orcas and Men: What Killer Whales Can Teach Us . New York, NY: Overlook Press. Nelson, R.K., 1969. Hunters of the Northern Forest: Designs for Survival Among the Alas- kan Kutchin . Chicago, IL: Chicago University Press. Nelson, R.K., 1983. Make Prayers to the Raven: A Koyukon View of the Northern Forest . Chicago, IL: Chicago University Press. Nowell, K., and Jackson, P., 1996. Wild Cats: Status Survey and Conservation Action Plan . Gland: Switzerland, Cambridge: UK, IUCN/SSC Cat Specialist Group. O’Regan, H.J., and Kitchener, A.C., 2005. The effects of captivity on the morphology of captive, domesticated and feral mammals. Mammal Review , 35 (3–4), 215–230. Parsons, K.M., Durban, J.W., Burdin, A.M., Burkanov, V.N., Pitman, R.L., Barlow, J., Bar- rett-Lennard, L.G., LeDuc, R.G., Robertson, K.M., Matkin, C.O., and Wade, P.R., 2013. Geographic patterns of genetic differentiation among killer whales in the northern North Pacific. Journal of Heredity , 104 (6), 737–754. Patel, K.C., 1994. Women, earth, and the goddess: A Shäkta-Hindu interpretation of embodied religion. Hypatia , 9, 69–87. Pelletier, F.X., 1975. Symbiose entre l’Amrig et le dauphin sur la côte mauritanienne. In L’homme et l’animal, Premier colloque d’ethnozoologie . Paris: MNHN, 171–176. Pérez Andueza, G., Morcillo, M., Sanchez, M., and Vilanova, X., 2015. Truffles’ flies and beetles – Suillia and Leiodes . In M. Morcillo, M. Sanchez and X. Vilanova, eds. Truffle Farming Today, a Comprehensive World Guide . Barcelona: Micofora, 285–299. Peterson, D., Hanazaki, N., and Simões-Lopes, P.C., 2008. Natural resource appropriation in cooperative artisanal fishing between fishermen and dolphins (Tursiops truncatus ) in Laguna, Brazil. Ocean & Coastal Management , 51, 469–475. Pierotti, R., 2011. Indigenous Knowledge, Ecology and Evolutionary Biology . New York, NY: Routledge. Pika, S., and Bugnyar, T., 2011. The use of referential gestures in ravens (Corvus corax ) in the wild. Nature Communication , 2, 560. Pryor, K., Lindbergh, I., Lindbergh, S., and Milanol, R., 1990. A dolphin-human fishing cooperative in Brazil. Marine Mammal Science , 6 (1), 77–82. Cooperating with the wild 219 Robineau, D., 1995. À propos de la prétendue symbiose entre les pêcheurs Imraguen et les dauphins. Mammalia , 59 (3), 460–463. Seed, A., Emery, N., and Clayton, N., 2009. Intelligence in corvids and apes: A case of convergent evolution? Ethology , 115 (5), 401–420. Serov, S.L., 1988. Guardians of the Spirit-masters of Siberia. In W.W. Fitzhugh and A. Crowell, eds. Crossroads of Continents: Cultures of Siberia and Alaska . Washington, DC: Smithsonian Institution, 241–255. Short, L.L., and Horne, J.F.M., 2001. Toucans, Barbets and Honeyguides . Oxford: Oxford University Press. Simoons, F.J., 1990. Food in China: A Cultural and Historical Inquiry . Boca Raton, FL: CRC Press. Smith, B.D., Tun, M.T., Chit, A.M., Win, H., and Moe, T., 2009. Catch composition and conservation management of a human – Dolphin cooperative cast-net fishery in the Ayeyarwady River, Myanmar. Biological Conservation , 142 (5), 1042–1049. Soma, T., 2012. Ethnoarhchaeology of horse-riding falconry. In Proceedings of the Asian Conference on the Social Sciences . Osaka, 3–6 May 2012, 167–182. Spottiswoode, C.N., Begg, K.S., and Begg, C.M., 2016. Reciprocal signaling in honeyguide- human mutualism. Science , 353 (6257), 387–389. Stahler, D., Heinrich, B., and Smith, D., 2002. Common ravens, Corvus corax , preferentially associate with grey wolves, Canis lupus, as a foraging strategy in winter. Animal Behaviour , 64, 283–290. Stépanoff, C., Marchina, C., Fossier, C., and Bureau, N., 2017. Animal autonomy and intermittent coexistences. North Asian modes of herding. Current Anthropology , 58 (1), 57–81. Sunquist, M., and Sunquist, F., 2002. Wild Cats of the World. Chicago, IL: University of Chicago Press. Svanberg, I., Kuusela, T., and Cios, S., 2016. Sometimes it is tamed to bring home fish to the kitchen: Otter fishing in northern Europe and beyond. Svenska landsmål och svenskt folkliv , 138, 79–98. Tani, Y., 1989. The geographical distribution and function of sheep flock leaders: A cultural aspect of the man-domesticated animal relationship in southwestern Eurasia. In J. Clutton-Brock, ed. The Walking Larder: Patterns of Domestication, Pastoralism, and Predation . London: Unwin Hyman, 185–199. Terashima, H., 2007. The status of birds in the natural world of the Ituri forest hunter-gatherers. In E. Dounias, M. Durham and E. Motte-Florac, eds. Animal Symbolism: Animals, Key- stone in the Relationship Between Man and Nature? Montpellier: IRD Editions, 191–229. Vucetich, J.A., Peterson, R.O., and Waite, T.A., 2004. Raven scavenging favours group foraging in wolves. Animal Behaviour , 67, 1117–1126. Walton, I., 1653. The Compleat Angler: Or, the Contemplative Man’s Recreation . London: T. Maxey for Rich. Marriot. White, C., 2005. Hunters ring the dinner bell for ravens: Experimental evidence of a unique foraging strategy. Ecology , 86, 1057–1060. Willerslev, R., 2007. Soul Hunters: Hunting, Animism, and Personhood Among the Sibe- rian Yukaghirs . Berkeley, CA: University of California Press. Wood, B.M., Pontzer, H., Raichlen, D.A., and Marlowe, F.W., 2014. Mutualism and manipulation in Hadza-honeyguide interactions. Evolution and Human Behavior, 35 (6), 540–546. Zahara, A.R.D., and Hird, M.J., 2015. Raven, dog, human: Inhuman colonialism and unset- tling cosmologies. Environmental Humanities , 7, 169–190. 220 Edmond Dounias Zappes, C.A., Andriolo, A., Simões-Lopes, P.C., and Madeira Di Beneditto, A.P., 2011. ‘Human-dolphin (Tursiops truncatus Montagu, 1821) cooperative fishery’ and its influence on cast net fishing activities in Barra de Imbé/Tramandaí, Southern Brazil. Ocean & Coastal Management , 54, 427–432. Zeder, M.A., 2012. Pathways to animal domestication. In P. Gepts, T.R. Famula, and R.L. Bettinger, eds. Biodiversity in Agriculture: Domestication, evolution, and sustainability . Cambridge: Cambridge University Press, 227–229. 12 Why did the Khamti not domesticate their elephants? Building a hybrid sociality with tamed elephants

Nicolas Lainé

The domestication of Asian elephants (Elephas maximus ) is a subject of much debate. The species has had representatives in a free-ranging state, as well as living among human populations across Asia for millennia.1 According to anthropologist J.-P. Digard (1990), this particularity makes elephants a limit case for domestication. Such animals are not considered domesticated because they are tamed but not bred, so that even though thousands of individuals have lived and are still living in human societies, this has not led to genetic modifications of the species from one generation to the next.2 Other than in rare cases, the renewal of the population is carried out by collection – via capture – from the forests, or by mating between village and forest individuals. Historically, the choice of (non)-domestication of the species has been justified by the abundant availability of pachyderms in forests and the use of the animals linked with various economic factors. Since elephants are not put to work before they reach their adult size, it is cheaper and quicker to socialise a weaned forest animal (from the age of 5 years) rather than raising it from birth. The long gesta- tion period (20–22 months) is another justification for not attempting the domestic breeding of elephants. As highlighted by Trautmann (2015), this has been notably the case in India since ancient times. The institutionalisation and diffusion of war elephants throughout Asia (starting from 1000 to 500 BC ) encouraged monarchs to create specific areas called mahal in order to secure the viability of pachyderms and their capture. Later on, during colonial rule and the advent of timber elephants, the economic cost was the main reason for not breeding elephants. There is no doubt that such interpretations, still admitted nowadays, reflect the vision and demands of a dominant group – namely, to serve a colonial policy or to maintain a contingent of war elephants. But, apart from royal and/or imperialist motivations, it is difficult to admit that techno-economic considerations were the only reasons that prevented local populations from domesticating the elephants with whom they lived. Contrary to what Trautmann (2015 ) argues, it can easily be assumed that local pachyderm management systems existed prior to the introduction of war pachyderms, and that such systems have co-existed though the age-old history of human-elephant relationships in Asia up until today. Anthropological literature has shown that the choice of domesticating or not a specific species responds to several factors and cannot be reduced to a utilitarian 222 Nicolas Lainé argument. For example, Philippe Descola ( 1994 ) studied the relationship between peccary and Amazonian populations. His study revealed that the availability of a “domesticable” species and the presence of the necessary skills among local populations to manage this species are not sufficient conditions to trigger a domestication process. Despite the fact that these two conditions were present, his work shows that throughout South America, the peccary has never been domesticated notably because of the relational status this animal assumes. When being hunted in the forest, the animal is considered akin, like an alter-ego, to humans, who have to mediate and pacify the peccary’s spiritual master in order to take their lives. When living in a village as pets, tamed peccary are too close to humans to be killed and eaten. Actually, the non-domestication of the peccary is not a matter of lack of skill and techniques, but more crucially such technical processes would call into question the entire set of relations that humans share with any living being, reflecting a specific cosmo-ecological view of the world (in the case studied by Descola, this corresponds to animism). Keeping in mind Descola’s demonstration on the peccary, this chapter aims to highlight the perceptions of a local population, the Khamti,3 who live and work daily with elephants in Northeast India, forgoing domestication. The Khamti do not breed but capture elephants in the forest and integrate them into the village. Why do the Khamti not domesticate their elephants? This question raises a further one: indeed, what bond exists between tamed elephants and the Khamti if not a process of domestication? What kind of hybrid sociality? By studying the different steps in the constitution of Khamti-elephant communities, this chapter intends to examine the logic underlying the relationships within these interspecies collectives. This case may shed light on the permeable limits between taming and domesticating and on the variety of interspecies entanglements which surround domestication and may or may not favour the development of this relationship. I will first describe the process through which a newly captured elephant is integrated into the village; I will then turn my attention to the capture operations using village elephants. In these two sections, I will focus my attention on the role of adult village elephants, named konkies . From a practical point of view, I am interested in the role and interactions between these konkies , the Khamti, and wild elephants. From a more symbolic point of view, I will also study their relationship with different supernatural forest and village entities.

Integrating forest elephants in village life A newly captured elephant from the forest will learn new living conditions in the village among men and other congeners. According to the Khamti, the wild forest elephant (chang thun ) becomes a village elephant (chang man ) at the end of a two-step process: the elephant must first get used to the human presence at its side; the animal must then learn some of the basic commands. This primary socialisation lasts for only a few weeks. Then, and for the next few years, the animal will mainly live in the company of his assigned mahouts (elephant handlers), villagers, and other congeners. Why did Khamti not domesticate elephants? 223 Taming the wild elephant The first step aims at making the animal accept a human presence at his side. This includes familiarising it with the human voices he hears in addition to tactile and visual contact, human smells, and fire. This phase of acceptance is essential before considering teaching it commands, which will constitute the second step of this process. These steps begin soon after bringing the captured elephant back to the village. Before beginning the operation, the Khamti set up a camp some distance from the village and close to a river to construct the lak chang , a wooden structure comprised of two tree trunks sufficiently spaced to place the animal at the centre (Figure 12.1). Once the lak chang is built, the first operation consists of placing the animal at its centre. To do this, a village adult elephant is necessary. Without such an animal, which was usually involved in capturing the young elephant, humans would not be able to place it in the lak chang. The juvenile will refuse to be driven. The fact of being connected to an adult elephant not only forces it to enter the lak chang , it also helps keep the young one calm. At first, the elephant is tied in such a way as to prevent it from moving freely. Its hind legs are tied together and pulled back, while its forelegs are directed forwards to throw it off balance and prevent it from using its strength. During the day, the sequences consist of petting, massaging, and rubbing every part of the elephant’s body while reciting a chant.4 The animal generally trumpets

Figure 12.1 Elephant tied to the lak chang. Photo credit: Nicolas Lainé (2009). 224 Nicolas Lainé loudly and tries to break free from its tethers. As the sessions continue, the animal gradually accepts being touched, and one of the men starts loosening some of the ropes attached to the back legs, followed by those in front. During these sessions, the konkie who helped put the juvenile into the lak chang is tied to a tree nearby. During the sessions, it sometimes trumpets at the same time as the young elephant. At night, the juvenile remains attached to the lak chang . During this period, the trainer and his assistants carry flaming torches around the elephant to get him used to the heat and teach him to not fear fire. In the forest, elephants flee fire.5 But in the present situation attached to the lak chang , the juvenile has no choice but to face the flames and accept the situation. Moreover, according to a Khamti belief, forest spirits (phi thun ) live on the skin of the animal and cling to its hair ( khon chang). Thus, by passing torches near it, the Khamti set out to singe the elephant’s hairs to make the spirits return to the jungle. After successive sequences in the morning, afternoon, and during the night, the trainer can eventually sit on the back of the pacified animal. The first step of the training process is then considered complete. It is then time for the elephant to continue its apprenticeship by learning some commands. The treatment inflicted on the elephant may appear violent in many respects, as the animal is completely immobilised during the first step. When I questioned them, it was difficult for my Khamti informants to justify these coercive methods. They answered that this period should last as short a time as possible, generally from 7 to 10 days. In fact, the duration depends not only on the nature and size of the elephant, but also on the trainer’s experience. On that point, it must be said that the Khamti recognise a good trainer if he manages to complete this step in less than ten days. During this short period of time, the animal must accept the treatment meted out to it and learn not to react aggressively to the actions taken by humans, who must constantly remain alert and ensure that the animal is not in danger. Other- wise, the consequences can be dramatic: some elephants die because they refuse to eat. In fact, one sign that the familiarisation process is going well is when the animal is willing to be fed from a human hand. Socialisation can cause several behavioural changes that affect an elephant’s biology. In ethological terms, this period refers to a habituation in the sense that it consists of inhibiting the elephant’s escape behaviour when faced with humans and fire: in other words, a desensitisation to certain stimuli. Habituation constitutes a form of non-associative learning leading to the progressive and reversible disappearance of stimulus response, when that stimulus no longer contains rein- forcement value for the organism ( Lassalle, 2004 ). In the present case, the exterior stimuli are represented by the human who the animal must accept at his side. This first step is marked by the repetition of gestures and human contact with the animal, so that it gradually gets used to people. The animal learns to live in the village within human society. The objective is not that the elephant should become dependent on a single individual who it has learned to respond to and obey . The village elephants keep a certain degree of autonomy: they mostly feed by themselves on the abundance of plants available in the village and forest area. The Khamti provide them with some specific nutrition complement (such as rice balls and salt) on particular occasions only; before Why did Khamti not domesticate elephants? 225 going to work, for example. Elephants also maintain some initiatives during the course of work in which they are engaged with humans and congeners, such as during timber operations ( Lainé, 2016 a).

Apprenticeship of commands Accepting a human presence at its side is only the first step in the integration of a newly caught elephant in the village. The animal will then continue its socialisation by learning commands through an apprenticeship, forming the second step of its integration. This second step clearly marks the beginning of the animal’s life in the village. It is lengthier than the previous step, and does not take place in a confined space (the lak chang), but along a field. It involves making the elephant move and teaching it to respond to simple commands such as agaat (“go ahead”), pischu (“stop”), dhaat (“stop”), soi gum (“turn”), and boit (“sit”). To force the animal to comply, it is first surrounded by and bound to two adult village elephants. The presence of a konkie here is doubly important: on the one hand, without the konkie, the Khamti would be unable to control the animal since it would be totally unconstrained in its movements and, on the other hand, because adult village elephants provide an example for the young animal to follow. During the sessions, two men are always necessary regardless of the number of village elephants present (Figure 12.2). A trainer sits on the juvenile animal’s back and a second

Figure 12.2 Learning commands. Photo credit: Nicolas Lainé (2009). 226 Nicolas Lainé man holds it by a rope, like a leash. The trainer on the elephant’s back guides the operation. With his feet, he adds gestures to oral communication, according to the taught command; he supports himself with his feet behind the ears. Once they arrive at the end of the field, a treat is systematically given to the animal as a reward (a piece of sugar or salt rice ball wrapped in a banana branch). Gradually, as the young elephant assimilates the commands, only one konkie will be necessary to continue the apprenticeship. The second step is considered complete when the elephant responds to commands. The Khamti consider this period a learning process for the elephant. It is clear that we are seeing a typical form of an associated learning process through positive and negative conditioning. This learning process includes a double reinforce- ment and invites the individual to discover independently the right attitude to adopt and the right answer to provide. Contrary to more classical conditioning, such as Pavlovian conditioning, such learning emphasises the ability of the learner to have some form of control over the consequences of its actions. One can also consider, however, that an imitation process is at play during this step. It is by imitating other village elephants that the apprentice responds positively to the commands given to it. In fact, at this step, the elephant probably has insufficient significant knowledge to understand what is expected of it. The juvenile elephant may be reassured, however, to see another elephant accomplish a task that it will, in turn, learn to achieve. In this view, the elephant integration process cannot be reduced to a personal relationship engaging a man with a single animal. It must be borne in mind that the objective is not that the elephant should become dependent on a single individual, to whom it will simply have learned to respond, but to the entire village community. It is also noteworthy that the elephant is introduced to the village with the help of another elephant. During the sequences, the congeners are used both as a means of control and security for the animal and a way of learning by imitation. Thus, elephant socialisation is not only a human-animal interaction but also a man-animal-animal interaction. At the end of these two steps, the owner will give the elephant a name. It becomes a member of the village community, and belongs to the household of its owner. In this respect the animal will be protected by the domestic spirit of his house, the phi hun . A mahout will be assigned to it, followed by an extended period of mutual socialisation.

Cooperating in the capture of wild elephants Among the Khamti, capture operations are led by two coordinated teams each composed of a village adult elephant and its owner (the phandi ), as well as the mahout employed by the owner.

Elephant protectors From a symbolic point of view, these operations require the team to conciliate the wild spaces inhabited by supernatural forces, called pa . Among the Khamti, the Why did Khamti not domesticate elephants? 227 world of the forest is a hostile area for humans and is thought to be in opposition to the village, which is protected. For such perilous operations as catches, the men first have to ensure their protection, but also to conciliate the forest entities in charge of the wild pachyderms. Among the different associated rituals two concern elephants directly: it is first by creating a ritual exchange with the spirit taking care of forest elephants, Chao Pling Chang , and second by being associated with the Elephant Lord, Utingna , that the Khamti obtain permission to capture a wild elephant. Chao Pling Chang literally means “the one who takes care of elephants”. He is considered to be the mahout of forest elephants. During each capture expedition, the owner must offer a hen and a cock to Chao Chang Pling. By offering him the pairs of animals, the owner hopes to receive an elephant in return: the human owner therefore attempts to establish a ritual exchange with the divinity. The offering takes place at the beginning of the expedition precisely when a member of the team (including the konkie ) confirms the presence of elephants in the sector. It may be, for example, the sound of an animal trumpeting , the presence of traces or faeces on the ground, or if a konkie sniffs the odour of one of several forest congeners. Some owners prefer to perform the ritual when they pass a migration route (hati dandi in Assamese ) taken by wild elephants. The intended ritual exchange with Chao Pling Chang is as follows: the owner will first throw some uncooked rice grains on the ground and place the pair of fowls nearby; once the birds are released, each team member carefully observes their behaviour; before disappearing into the forest, the two animals must con- sume the rice and not fight with each other since this would be a sign that Chao Pling Chang refuses the offering. In this case, some would consider the expedition to be in vain, and that no animal will be brought back to the village. One or more events can take place preventing the catchers from bringing an elephant back to the village. This may include an attack on the team by the mother of the captured baby elephant, an encounter with a tiger, an accident during their stay in the forest or quarrels between team members. However, if the animals begin to peck before disappearing, it means that the offering is accepted and the exchange will take place without major incident and be successful. Although this ritual may only be performed once, its interpretation is often subject to discussions between the owner and his teammates. The Khamti consider Utingna to be the lord and king of wild forest elephants. He is represented as a man sitting on an elephant playing a musical instrument, a flute or lyre (Figure 12.3). The catcher believes that the music played by Utingna has the power to attract and hypnotise wild elephants, thanks to which they can come closer to elephants without being attacked. According to a story well known to elderly catchers, an unsuccessful owner would approach Utingna to ask him if he could help in capturing elephants by ensuring him that elephants living among humans are well treated. Following this meeting Utingna , as the lord of forest elephants, would personally reassure the elephants living in the forest, and ask them not to flee the humans who come to capture them. Because of this, elephant catchers always carry a small statue of Utingna with them on their chest. Utingna also protects them against wild animals. Thus, this statue has a double function 228 Nicolas Lainé

Figure 12.3 Ivory statue representing Utingna . Photo credit: Nicolas Lainé (2008). of protection and ensuring success. According to some catchers, by bringing the statue along with them, dangerous animals such as tigers will make a path for the konkie and the team.

Stalking and capturing wild elephants Let us now examine the practical role of the mounted elephant during capture operations. The owner sits on the neck of his elephant, while the mahout is seated behind him on the back. The team arrives in the forest and goes to the temporary camp to stalk. The primary objective of this is to roam into the forest to detect the presence of an elephant herd. During all these exploratory sessions, the owner decides on an initial direction (north, south, east or west). Nevertheless, once they go deeper into the forest, the rest is determined by the konkie . The first task of the konkie is to be a tracker. With its trunk, it seeks out the presence or a route taken by a herd. It smells each and every branch. If some wild elephants have recently passed nearby, their traces will be on trees and branches. The konkie also looks for footsteps on the ground. Stalking is done in absolute silence. The team must make as little noise as possible in order to not frighten the potential catch. On their konkie , the mahout and the owner communicate by min- ute gestures or by touch. To communicate with the rest of the crew, they whistle. When I interviewed6 an expert catcher about how his konkie indicates the presence of wild elephants, he explained the intimate relationship between the owner Why did Khamti not domesticate elephants? 229 and his elephant. To be able to do this, the human must know and be able to accurately interpret his elephant’s movements. He added that the reaction of each konkie is different depending on whether it detects a bull elephant, a herd, or even a tiger. When the konkie is on the ground or between branches, the owner observes its conduct. The owner looks for signs such as ear flapping or non-movement, the way it uses its trunk from left to right or up and down to smell out for more clues as to the presence of such or such an animal, or even the way the konkie breathes. In the forest, all tracking sessions take place at night. To find and know what area to explore, the capturer has no choice but to rely on the animal. Once the presence of a herd is confirmed by one konkie , the other men on other elephants are informed and come together in one place. They then set up a strategy to get closer to the herd, observe its composition and identify who, among the herd members, will be their target. They can then decide whether to follow the herd and wait for a propitious moment. From that moment on, every second is precious. Once the signal is given by the owner, all the men begin to incessantly prick the ears and thighs of the konkie to get them to run as fast as possible. The animals will try to isolate the target animal. It is then the turn of the owner to throw his rope and lasso it around the leg or neck of the target. According to interviewees, the konkie know that the capturers are only interested in juvenile elephants and try to avoid adults. When the animal is lassoed, the konkie assist the men by catching the target elephant’s tail with its trunk to bring it as close as possible to the hunter. This precious help lets the men immobilise the calf. As soon as the elephant is tied by the owner, the second team comes to immediately block the slipknot so that the juvenile won’t be strangled or escape. This also prevents the rope from getting tangled up. At the same time, the other team’s task is to fire several gunshots into the air to scare the herd and prevent it from coming back to rescue the juvenile. It is then again with the help of the konkie that the calf elephant is first brought to the forest camp. It remains constantly and closely linked to the konkie ’s rope that was used to capture it. The captured animal is thus forced by the konkie to move forward. Without the konkie , it would be impossible for men to make the animal move, even if it is a juvenile. The konkie’ s job, as a catcher told me, is to contain the captured elephant thus rendering more coercive and possibly painful methods unnecessary. As soon as they arrive at the camp, it is time to gather up all their equipment and return very quickly to the village. Again, on the way, the konkie may at times push the juvenile forward.

Revisiting the (non-)domestication of Asian elephants The description of the capture and integration of forest elephants highlights the essential roles of the konkie in replenishing the stock of village elephants. Adult village elephants are present from the very first moment, and are also involved in the more active role of teaching basic commands later on. Konkie are thus indispensable for the Khamti in the process of integrating captured elephants into the 230 Nicolas Lainé village. By living in the village with humans, the konkie become indispensable for their interactions with forest elephants. Because they were born in the forest where they lived until their capture at the age of 4–5 years, the konkie know well where to locate and how to approach their congeners. Without them, it would be impossible for any phandi to approach a wild herd without being detected. On the other hand, elephants born in the village would not risk approaching forest congeners and would not possess the skills needed to locate them in the forest. In capture operations, village elephants serve as intermediates enabling their human partners to detect, approach and capture forest elephants. Symbolically, rituals and practices specifically associating Chao Pling Chang and Utingna constitute the necessary condition for taking an elephant out of the forest. While primarily conceived of as a protection for humans, such practices also give indications regarding the status conferred to village elephants and their capacity to interact with these divinities. For the Khamti, forest elephants are under the symbolic responsibility of the mahout Chao Pling Chang. By a ritual exchange, this divinity will accept that one of his wards be taken and adopted by a human mahout. For the Khamti, ritual practices are one aspect of a proper management which must ensure respect and health granted to the elephants living amongst humans. Figure 12.4 reveals the different dynamics underlying Khamti-elephant relations. It shows that according to the Khamti, the existence and reproduction of village elephants cannot be considered without the existence of forest elephants. While in a naturalistic ( Descola, 2005 ) and “modern” conservation science-based view, these two populations of the same species are always considered as distinct, the Khamti both engage and think of these animals as one. The particularity of

Figure 12.4 The relational dynamics between the Khamti and elephants. Why did Khamti not domesticate elephants? 231 elephants living among the Khamti is that they move between the two worlds: the village (man ) and the forest (pa ). This allows constant and dynamic interactions between the two animal populations. Figure 12.4 reflects the transformation process of a forest elephant (chang thun) into a village one (chang man), including the various entities implied in such processes, and the nature of relationships that humans or elephants share with them. It also highlights the circulation of animals between these two social spheres, the village and the forest. Figure 12.4 also emphasises the dynamic of interactions between village and forest elephant populations. In order to secure the reproduction of the forest elephant population, Khamti deliberately let their village elephants (especially males) roam freely in the nearby forest at night, when they are not working, or when they know that a forest herd is passing near the village, in order to mate with their forest congeners. Doing so implies a degree of human intervention in elephant reproduction which, for the Khamti, guarantees the maintenance of the forest elephant population for later capture. This fact has been confirmed on several occasions during fieldwork. Thus, contrary to what anthropologist Piers Locke (2014 : 13) wrote that “elephants have traditionally been kept without mankind playing a consistent and decisive role in their breeding until recently”, we note here that for a long time, the Khamti have certainly played an important role in elephant breeding. Such a role may indeed be described as what Haudricourt (1962 ) called an indirect action on nature. What should be retained from the present case is that neither the economic aspect, the time required, nor the social prestige of catchers are arguments that prevail in the Khamti’s relations with the elephants they bring into the village. The Khamti do not try to exert control over every aspect of the elephants with whom they live, and their intervention on their reproduction with forest elephants is very indirect. Capturing a 5-year-old animal means owning an elephant who knows the forest and how to survive in it. This would not be the case of a village-born elephant. The forest elephant who lives in the village knows both worlds. It knows how to behave in the forest, and this can be of benefit to humans since the animal knows how to locate and approach its forest congeners during the catching operations. It also knows how to heal itself, and may have participated in the co-construction of medical knowledge.7

Conclusion: an elephantine culture at the village level? Although they are not domesticated, village elephants compose a specific group behaviourally distinct from their forest congeners. Their role, notably their role in transmitting behaviour and attitudes to newly captured elephants, opens the door to some elements of a culture transmitted among village elephants. As noted, while the Khamti create a context allowing such transmission, they are far from able to accomplish it by themselves. The imitation process at play during the second step of socialisation is based on a form of transmission of practices and attitudes within the village elephant 232 Nicolas Lainé cohort. Imitation of congeners is crucial to lead a juvenile to understand what is expected of it. Later, during capturing or timber operations, no specific training is given to village elephants who become konkie . The animal learns by experience, i.e. they have learnt how to learn and to respond to what is expected of them. To a certain extent this learning process deals with what Bateson (1972 ) called deu- tero-learning: village elephants have learnt to learn . They are capable of acting , sometimes creatively and autonomously in a variety of contexts and situations, to achieve various tasks ( Lainé, 2016 a). We can therefore consider that the konkie constitute a behavioural sub-population sharing adaptations to life with humans, although the transmission of these adaptations is not mainly genetic – as is the case for domestic animal populations – but rather a cultural transmission. And if like Christophe Boesch (2012 ) we were to narrow down the definition of culture as applied to chimpanzees it would contain three essential aspects: learning, collective aspects and the sharing of meaning within a specific distinct population group, then one can assess the existence of an elephantine culture among the konkie at the village level. On the other hand, these roles and attitudes of adult village elephants towards the young and newly caught also can refer to the process of enskilment , as highlighted by Tim Ingold (1980 , 2000 ). According to this author, enskilment above all involves observing and paying active attention to the movements of others; imitation is aligning that attention to the movement of one’s own practical orienta- tion towards the environment. As seen in the present case, forest elephants learn to become konkie by imitating their fellows and by engaging in actions with them. Learning is a process of enskilment and cannot be separated from doing, embedded in the context of a practical engagement with the world. Finally, this atypical case makes it possible to highlight the importance of social transmission in the construction of an animal community associated with humans. This collective of socialised but non-domesticated animals does not share distinctive biological aspects, but rather a set of communication attitudes and routines that are partly transmitted between the animals themselves during the taming process, and will continue throughout their lifetime as village elephants.

Notes 1 Globally, the most recent estimate for the population size of the Asian elephant was between 40,000 and 50,000 animals, including about 15,000 living in human societies ( Choudhury et al ., 2008 ). 2 Interestingly, this is not the case for wild elephants. Research has shown that Asian elephants have become genetically adapted to poaching. For example, new generations of pachyderms no longer possess huge tusks as in the past (Kurt et al., 1995). In Sri Lanka, for example, about 98 per cent of the male elephant population is tuskless (Chelliah and Sukumar, 2013 ). 3 The Khamti form a small Tai group in the Indian states of Assam and Arunachal Pradesh, made up of tens of thousands of individuals (GoI, 2010). In the middle of the 18th century, they separated from the other groups of Khamti populations scattered mainly in the present Kachin State in Burma. Their relationship with elephants was the object of my PhD dissertation (Lainé, 2014). I have specifically analysed the conditions and Why did Khamti not domesticate elephants? 233 implications of what I have termed “living together,” studying the capture, socialisation and work performed by and with elephants in Khamti society. 4 Music is an essential aspect of this process. It serves as an attention support to interact with the young elephant (see Lainé, forthcoming). 5 This well-known behaviour is used to prevent wild elephant incursions into fields. 6 Interview conducted in Namsai district in September 2010. 7 Such is the case with other animal species. For example, Krief and Brunois-Pasina (2017 ) have recently tested the hypothesis of an interspecific genesis on the ways humans and chimpanzees cure themselves by selecting the same plants in Uganda.

References Bateson, G., 1972. Step to an ecology of mind . New York: Ballantine Books. Boesch, C., 2012. Wild cultures: A comparison between chimpanzee and human cultures . Cambridge: Cambridge University Press. Chelliah, K. and Sukumar, R., 2013. The role of tusks, musth and body size in male-male competition among Asian elephants, Elephas maximus. Animal Behaviour, 86 (6), 1207–1214. Choudhury, A. et al ., 2008. Elephas maximus . The IUCN Red List of Threatened Species. Available from: http://dx.doi.org/10.2305/IUCN.UK.2008.RLTS.T7140A12828813. en [ Access on 08 June 2017]. Descola, P., 1994. Pourquoi les Indiens d’Amazonie n’ont-ils pas domestiqué le pécari. Généalogie des objets et anthropologie de l’objectivation. In: B. Latour and P. Lemon- nier, eds. De la Préhistoire aux missiles balistiques. L’intelligence sociale des techniques . Paris: La Découverte (collection Recherches), 329–344. Descola, P., 2005. Par-delà nature et culture . Paris: Gallimard. Digard, J.-P., 1990. L’homme et les animaux domestiques: anthropologie d’une passion . Paris: Fayard. Government of India (GoI), 2010. Arunachal Pradesh: Data Highlights: The Scheduled Tribes, Census of India 2001. Available from: http://censusindia.gov .in/Tables_Published/ SCST/dh_st_arunachal.pdf [ Access on 06 June 2017]. Haudricourt, A.-G., 1962. Domestication des animaux, culture des plantes et traitement d’autrui. L’Homme , 2 (1), 40–50. Ingold, T., 1980. Hunters, pastoralists and ranchers: Reindeer economies and their transformations . Cambridge: Cambridge University Press. Ingold, T., 2000. The Perception of the environment: Essays on livelihood, dwelling and skill . Okon/New York: Routledge. Krief, S. and Brunois-Pasina, F., 2017. L’interspécificité du pharmakôn dans le parc Kibale (Ouganda): savoirs partagés entre humains et chimpanzés? Cahiers d’Anthropologie Sociale , 14 (‘Guérir, Tuer’), 112–135. Kurt, F., Hartl, G. B. and Tiedemann, R., 1995. Tuskless bulls in Asian Elephant (Elephas maximus). History and population genetics of a man-made phenomenon. Acta Therio- logica , Supp. 3, 125–143. Lainé, N., 2014. Vivre et Travailler avec les Eléphants. Une Option Durable pour la Survie de l’Espèce. Enquête sur les Relations entre les Khamti et les Eléphants dans le Nord- est Indien. PhD Dissertation Thesis in Ethnology. Université Paris-Ouest Nanterre- La Défense, Nanterre. Lainé, N., 2016a. Conduct and collaboration in human-elephant working communities of Northeast India. In: P. Locke and J. Buckingham, eds. Conflict, negotiation, and coexistence: Rethinking human-elephant relations in South Asia. New Delhi: Oxford Univer- sity Press, 180–205. 234 Nicolas Lainé Lainé, N., 2016b. Pratiques vocales et dressage animal. Les mélodies huchées des Khamtis à leurs éléphants. In: N. Bénard and C. Poulet, eds. Chant pensé, chant vécu, temps chanté: Formes, usages et représentations des pratiques vocales. Paris: Éditions Dela- tour, 187–205. Lainé, N., 2017b. Surveiller les Animaux, Conserver l’Espèce. Enjeux et Défis du Contrôle de la Tuberculose des Éléphants au Laos. Revue d’Anthropologie des Connaissances , 11 (1), 23–43. Lainé, N., forthcoming . Voice, Sound, Melody. Music as a training tool and a way to interact with elephants in Northeast India. In : E. Emery, ed. The Elephant Reader . London: RN Books/SOAS Publication. Lassalle, J.-M., 2004. Apprentissage, adaptation et cognition. In : J. Vauclair and M. Kreutzer, eds. L’Ethologie Cognitive. Paris: Ophrys /MSH éditions, 49–77. Locke, P., 2014. The anomalous elephant: Terminological dilemmas and the incalcitrant domestication debate. Gajah , 41, 12–19. Trautmann, T. R., 2015. Elephants and kings: An environmental history . Chicago: The University of Chicago Press. 13 Cognition and emotions in dog domestication

Sarah Jeannin

Through domestication, distinct animal species share a physical space, have an everyday life, sometimes cooperate in joint tasks and create affectional bonds: how can that be when their cognitive systems and means of communication seem to differ significantly? The dog is the domestic species that shares the oldest bond and interaction with humans. How do we understand each other? How do we manage to create such emotional bonds? How is it possible to share reciprocal emotions?

Domestication The domestic dog (Canis familiaris) is a member of the canidae family. Despite their great diversity in terms of both morphology and behaviour, all modern dog breeds would seem to share the same ancestor, the grey wolf (Canis lupus ). The dog is the domestic animal that has lived among human populations for the longest period of time (about 15,000 years) (Frantz et al. 2016) and virtually anywhere there are humans, one can also find dogs. The origins of dogs and the modalities of their domestication are developed by Germonpré, this volume. For many thousands of years, humans have subjected dogs to intensive artificial selection ( Hare et al . 2002 ), making it possible to create lineages of working dogs bred to assist humans as well as lineages of pet dogs. It has been possible to develop this wide variety of ways in which dogs are used thanks to their remarkable morphological and behavioural plasticity.

The human-dog relationship Although the human-dog relationship was originally cooperative and mutualistic in nature, archaeologists have found dog cemeteries all around the world, dating back between 12,000 and 14,000 years, near human burial grounds, evidence of a long-standing emotional relationship ( Morey 2006 ). Several studies suggest that artificial selection has made it possible to promote the predisposition of dogs to form affiliative bonds with humans ( Topál et al . 2005 ). In fact, animals that are physically, behaviourally or cognitively close to humans tend to be preferred and arouse more positive affects in humans ( Batt 2009 ). 236 Sarah Jeannin In modern Western societies, dog owners display strong emotional bonds to their pets which play a wide variety of roles: a friend, a confidant, a child, an image their owners want to embody, etc. Dogs become the receptacle of emotional needs; they soothe, secure and contribute to improving the quality of life of their owners. They are now often considered “part of the family” by their owners. According to Askew (2003), the behaviour of modern owners towards their pets can be likened to parental attitudes directed to a member of another species. Serpell (2004) suggests that humans develop feelings and positive forms of behaviour when taking care of dogs because this emotional relationship is similar to the one created between a mother and her child. Many authors have attempted to understand the underlying dynamics of the emotional bond between humans and dogs. A large number of these researchers have referred to, and adapted, the experimental protocols used in human psychology to study the parent-infant bond. The results of these studies highlight numerous overlaps between the two types of relationships, from both a physical (neuronal and physiological data) and behavioural perspective.

Cerebral homologies Human functional magnetic resonance imaging studies show that the presentation of the faces of human and canine family members activate the anterior cingulate cortex, a region deeply affected by oxytocin systems, whose activity is associated with affective and emotional aspects of social cognition (Shinozaki et al . 2007). Moreover, when exposed to a picture of their own child or of their own dog, women display a similar neuronal activation in brain regions involved in emotions, reward/affiliation (amygdala, the periaqueductal grey, ventral tegmental area, insula, thalamus), visual processing and social cognition (fusiform gyrus and superior temporal gyrus) (Stoeckel et al. 2014). The image of her child, however, induced a greater activation of the mother’s mesencephalic regions, especially the ventral tegmental area and the substantia nigra, both involved in the reward and affiliative processes, whereas the image of their dog induced a greater magnitude and extent of activation in the fusiform gyrus. This region is key to visual and face processing and social cognition. Facial expressions and gaze play a more central role in the human-dog relationship than in human-to-human communication. This is probably because dogs have a limited understanding of human language ( Stoeckel et al . 2014 ).

Identical hormonal climate Oxytocin, commonly called “the bonding hormone”, is both a hormone and a neuro-modulator that promotes affiliative behaviours and facilitates the bond between the mother and her child (Heinrichs et al. 2009). Oxytocin also promotes social cognition and the interpretation of social signals and, as such, provides a greater predisposition to empathy and encourages approach behaviours in social contexts ( Heinrichs et al . 2009 ). Positive and calm interactions between humans and dogs lead to an increase in oxytocin concentration in both protagonists ( Nagasawa et al . 2015 ). This Cognition, emotions in dog domestication 237 phenomenon is similar to the one observed in human parent-infant interactions. The mutual gaze between the parent and the infant has been described as affiliative behaviour and a sign of social commitment whose principal role is to regulate social bonds, notably through the release of oxytocin ( Feldman et al . 2007 ). A recent study shows that the variations of oxytocin concentrations in dog owners are correlated to the frequency of exchange behaviours initiated by a dog’s gaze. The authors concluded that humans may experience an affectional bond with their animal similar to the bond they feel for a member of their family (Naga- sawa et al. 2015). This system of oxytocin release might be the factor underlying the psychological and physiological benefits of the human-dog relationship. This phenomenon is not present, however, in the interactions between humans and hand-raised wolves. It is likely, therefore, that over successive generations dogs have developed the ability to make use of the way humans communicate a social bond, through gazing behaviour, for example ( Nagasawa et al . 2015 ). This heredi- tary learning capacity may have helped to establish the special affectional bond between humans and dogs ( Nagasawa et al . 2015 ).

The emotional bond Many authors have attempted to explore whether the owner-dog relationship constitutes an attachment bond similar to the one formed between parents and their babies ( Payne et al . 2015 ). The owner-dog relationship has consequently been studied using the theoretical framework of the ethological attachment theory (Bowlby et al . 1989; Ainsworth and Bell 1970 ) and, more specifically, a modified version of the Ainsworth Strange Situation initially used to assess types of human infant attachment. The results of these studies show that dogs demonstrate behaviours towards their owners that closely resemble those reported in human infants: they explore and play more in the presence of their owners, and demonstrate stress signals in their absence and an increase in social behaviours towards them when reunited (Topál et al. 1998; Palmer and Custance 2008). On the basis of this research, it has been argued that owners represent a secure base for their dogs and that, according to the definition given by Bowlby et al. (1989), dogs are linked to their owners by an attachment bond ( Mariti et al . 2013 ). In contrast, hand- raised wolves, despite their preference for their carers over unfamiliar individuals during reunions, do not express the same behavioural pattern ( Ujfalussy et al . 2017). These behavioural traits might, therefore, be specific to dogs. However, it is important to emphasize, first, that the criteria used to define the owner-dog bond are based on the theory of human-to-human filial attachment and, second, that this yardstick may not necessarily be appropriate in the context of human-dog relationships ( Crawford et al . 2006 ).

Communication between humans and dogs Moreover, the social environment and experiences of dogs partly correspond to those of human infants. A comparison between infants and dogs makes it possible to explore how two species with very different evolutionary paths behave after 238 Sarah Jeannin exposure to similar social environments ( Gómez 2005 ). The experimental para- digms initially used in development psychology to study the social and cognitive capacities of preverbal children have therefore been adopted to evaluate the ability of dogs to understand and use human communication cues.

Sensitivity to human emotions Dogs can discriminate between affiliative and agonistic human postures and adapt their emotional and behavioural responses accordingly ( Vas et al . 2005 ). Simi- larly, dogs are able to discriminate between different emotional tones in the human voice. For example, they display more interest when listening to an infant crying than when they hear an infant laughing (Ruffman and Morris-Trainor 2011). Dogs rely primarily on facial expressions to assess humans’ emotional states (Gácsi et al . 2004 ) and read human expressions in the same way as humans do: the gaze is first directed to the eyes, irrespective of the emotion concerned ( Somppi et al . 2016 ). Both puppies and adult dogs recognize, react to and select significantly more emotional faces than neutral faces ( Deputte and Doll 2011 ). They are not only able to discriminate between human emotions but also to recognize them ( Albuquerque et al . 2016 ): for example, they match a threatening facial expression to a vocalization possessing a negative emotional valence. This aptitude to recognize human emotions allows them to adjust to humans. For example, they will approach and fetch a target significantly more willingly if their owner expresses a positive emotion towards this object compared to a negative (disgust or fear) or neutral one ( Merola et al . 2014 ; Turcsán et al . 2015). The emotions expressed by humans consequently have an important value for dogs and play a major role in the relationship. Furthermore, humans exert a social influence on how a dog performs: in one study, dogs were found to rely more on a human pointing gesture than on their own olfactory abilities when attempting to obtain hidden food ( Szetei et al . 2003 ). Moreover, research has shown that whereas dogs preferentially choose the larger of two food quantities when alone, they can be induced to abandon that preference when their owner (or a stranger) expresses ostensive cues such as gaze alterna- tion or a high-pitched voice to express their interest in the smaller food quantity ( Marshall-Pescini et al . 2012 ). Dogs can therefore make counterproductive choices because they are very sensitive to human ostensive cues. The authors conclude that social learning may in some cases be maladaptive ( Marshall-Pescini et al . 2012 ). Dogs present behavioural responses significantly more oriented towards an individual pretending to cry compared to the same person humming or talking: they look at him/her, approach him/her and touch him/her more. These reactions are similar whether it is their owner or an unfamiliar person ( Custance and Mayer 2012 ). Dogs also present physiological responses to human emotional states: their cortisol concentration increases significantly when listening to the cries of a human infant (Yong and Ruffman 2014); when reunited with a familiar person after a few minutes of separation, dogs show an increase in oxytocin concentration and a decrease in cortisol level ( Rehn et al . 2014 ). Cognition, emotions in dog domestication 239 Sensitivity to human attentional states Dogs are sensitive to human attentional states and adapt their behaviour accordingly: they will more readily steal forbidden food if the experimenter cannot see them (Kaminski et al. 2009) or if the experimenter is not attentive to the situation (eyes closed, back turned or distracted by a video game) ( Schwab and Huber 2006). In contrast, they will ask for food preferentially to an individual who can see them compared to an individual having an obstructed view (Gácsi et al . 2004 ), which suggests that dogs are also sensitive to the human’s visual perspective. A recent study shows that dogs are quite skilled in adopting the human visual perspective ( Catala et al . 2017 ).

Sensitivity to human visual communication In the last two decades, dogs have become major research subjects. For psycholo- gists, dogs may be the new chimpanzees. Indeed, dogs are very competent in the use of human gestural communication. For example, they understand and follow a human pointing towards a hidden food reward better than chimpanzees do (Hare and Tomasello 2005); even puppies succeed and do so at their first attempt, which suggests that it is not the result of learning during the task (Dorey et al . 2010 ). Dogs are also able to use more subtle cues such as orienting, nodding or shaking the head, even a simple glance (Udell et al. 2008). Wolves are less capable of using human visual signals, probably because it is much more difficult to establish eye contact with a wolf. Dogs are more likely to look at people’s eyes ( Miklósi et al . 2003 ). It would therefore appear that the long-standing cohabitation between humans and dogs has allowed these species to form a close relationship and develop common ways of communication such as eye-to-eye contact (Hare and Tomasello 2005). Indeed, when puppies cannot obtain or reach a food reward, they gaze at humans – whereas wolf cubs, even when hand-raised by humans, will persever- ate. Gazing at humans is also found in adult dogs but not in hand-raised wolves ( Miklósi et al . 2003 ; Passalacqua et al . 2011 ). Dogs also use human social referencing, which can be defined as the seeking of information from another individual to form one’s own understanding and guide action. For example, when encountering ambiguous situations or a potentially frightening object, human infants will look to see how their parents react: if the parent communicates happiness or enthusiasm through facial expressions, vocalizations or body language, the infant will be more likely to approach the object or react in the same way. If the parent shows the opposite emotions or behaviour, this will probably be reflected in the infant’s behavioural choices. Social referencing, from an evolutionary perspective, is an important survival tool, especially when one needs to assess what is, or is not, a threat. In their study, Merola et al . (2012 ) tested social referencing in dogs using a potentially frightening object: an electric fan with plastic green ribbons attached to it. They found that dogs looked to the fan and immediately after to their owners. In this way, dogs were similar to human 240 Sarah Jeannin infants. Moreover, dogs receiving a positive emotional message from their owners (through both voice and facial expressions) were more likely to approach and spend time close to the fan. Finally, the owners’ behaviour contributed to whether or not the dogs approached the fan.

Imitation Finally, dogs can learn to imitate humans. Indeed, authors of various studies have taught dogs to learn, not a specific command (e.g. “lie down”), but a non-specific rule: “do it!” which means “do as I do” ( Topál et al . 2006 ). Thanks to this method, it is possible to teach the dog new actions that he has never performed before.

Verbal communication between humans and dogs In the presence of another individual, and especially when visual information is not available, we use acoustic correlates of emotional and motivational states to modulate and predict social behaviour. The accurate perception of affective information in the vocalization produced therefore plays a critical role in both conspecific and heterospecific interactions. Belin et al. (2000) measured human cerebral activity during stimulation with human and animal affective vocalizations. They found that positively versus negatively valenced vocalizations from cats and monkeys elicited different cerebral responses despite the participants’ inability to differentiate the valence of these animal vocalizations by overt behavioural responses. Moreover, the comparison with human non-speech affective vocalizations revealed a common response to the valence in the orbitofrontal cortex. The authors concluded that the neural mechanisms involved in processing human affective vocalizations may be recruited by heterospecific affective vocalizations. This result supports claims of shared emotional systems across species. More recently, studies have showed that when listening to dogs’ vocalizations and without any visual cue, humans are able to identify the dog’s emotional state as well as the context in which the recording was made (Pongrácz et al . 2005 ). It seems that even 6-month-old infants are able to match aggressive and non- aggressive barks with the corresponding facial expression (Flom et al . 2009 ), which suggests that this sensitivity develops early in their ontogeny. Reciprocally, dogs are sensitive to human vocalizations and our intonations influence their performance. For example, a dog will be more motivated to look for a hidden food reward if the experimenter asks him to find it using a high- pitched friendly voice rather than an imperative tone ( Scheider et al . 2011 ). Dogs are not only sensitive to voice intonations, but also to speech content. Indeed, two studies, both carried out with border collies, showed that dogs are able to recognize and use several hundred words (Kaminski et al . 2004; Pilley and Reid 2011) via deduction learning. These behavioural observations have been further objectivized by new technologies: a neuroimaging study demonstrated that voice areas exist in dogs showing a similar pattern to anterior temporal voice areas in Cognition, emotions in dog domestication 241 humans, and that sensitivity to vocal emotional valence cues engages similarly located non-primary auditory regions in dogs and humans (Andics et al . 2014 ). More recently, the same team ( Andics et al . 2016 ) found that dogs pay attention to both what humans say and how they say it: the dogs present neural mechanisms that analyze and integrate word meaning and intonation separately. These neural similarities could be the reason why communication between humans and dogs is so efficient and why dogs are very sensitive to their owner’s emotions ( Custance and Mayer 2012 ). Authors exploring verbal communication in the context of affective relationships between humans and dogs highlight that owners not only behave with their dogs as with infants, they also speak to their dogs using pet-directed speech (PDS), a register that strongly resembles the infant-directed speech (IDS) used by humans when talking to young children ( Trainor et al . 2000 ) characterized by a high pitch, wide pitch modulations, short phrases, simple grammar, repetitions, etc. ( Hirsh-Pasek and Treiman 1982 ; Burnham et al . 2002 ). In a recent study (Jeannin et al. 2017a), we analyzed the speech of French female owners, recruited from the waiting room of the preventive medicine unit of the Centre Hospitalier Universitaire Vétérinaire d’Alfort (CHUVA) and via veterinary students’ social networks, addressing their dog in four experimental conditions designed to solicit different emotions and intents from the owners: before a brief separation, after a reunion, during play, and giving commands. We showed that women owners adapt the verbal and non-verbal features of their speech to the context of the interaction in order to provide the dog with information about their intentions and emotions (i.e. they use a high-pitched voice when they want to express praise or affection, and a low-pitched voice when they want to control the dog). We highlighted that owners also take account of the dog’s limited ability to understand human verbal signals and adjust their mix of speech components accordingly, using appropriate verbal content reinforced by nonverbal cues (e.g. affective sentences, verbal praising associated with a high-pitched voice) in order to enhance the dog’s comprehension and, to some extent, to teach the dog some basic utterances. Research has shown that naturally spoken IDS is associated with increased infant attention and social responsiveness compared to adult-directed speech (ADS) (Dunst et al. 2012), and that IDS leads to an increase in neuronal activation (Golinkoff et al. 2015) as well as to positive effects on infant language development (Song et al. 2010). In the context of human-dog communication, there is evidence that dogs display an increased neuronal activity in the auditory cortex when listening to vocalizations with positive emotional valence compared to negative or neutral emotional valence (Andics et al. 2014), but no study had explored dogs’ preference for PDS until 2017. In another study (Jeannin et al. 2017b), we investigated the perceptual relevance of PDS features from a dog’s perspective in order to evaluate the effect of PDS on the dog’s attention, mood and ability to learn new commands. We video recorded 44 adult pet dogs and 19 puppies listening to the same phrase enounced either in ADS, PDS or IDS. The phrases were previously recorded and were broadcasted via a loudspeaker placed in front of the dog. The total gaze duration of the dogs 242 Sarah Jeannin towards the loudspeaker was used as a proxy of attention. Our results show that adult dogs are significantly more attentive to PDS than to ADS and that their attention significantly increases along with the rise of the underlying frequency of human speech. The same pattern of results was found only in puppies by Ben- Aderet et al. (2017). In our study, puppies tend to show a greater reaction to all vocal stimuli compared to adult dogs and remain alert, whatever the type of sound and the intensity of the acoustic parameters. The preferences of human infants and dogs for exacerbated prosodic features may have an evolutionary explanation. Indeed, mammalian species use particular acoustical signals to signify motivations, intentions and emotional states that share similar acoustic features ( Ohala 1984 ): high tonal sounds are associated with affiliative or submissive motivation probably because they mimic the sounds produced by infants (leading to an appeasing effect on the receiver); these sounds are generally produced in fearful or appeasing contexts (Morton 1977). In contrast, low-frequency sounds – which increase the perceived size of the caller – are generally produced in hostile contexts, during hostile interactions and associated with aggressive motivation (Morton 1977). This preference for PDS may also be promoted by learning: adult dogs probably learnt to associate PDS with positive greeting contexts, as PDS is exacerbated during friendly interactions (Jeannin et al . 2017a), and it is well established that dogs have a well-developed ability to associate prosodic cues of human speech with specific contexts ( Scheider et al . 2011 ). Thus, IDS and PDS appear whenever there is a willingness to create a bond. Although PDS is originally the consequence of a particular affective context, this means of communication leads to heightened attentiveness. It is likely that the exaggerated prosody of PDS is used by owners as an ostensive cue for dogs that facilitates the effectiveness of their communication, and should represent an evo- lutionarily determined adaptation that benefits the regulation and maintenance of their relationships. Considering a dog to be part of the family is probably linked to improved living conditions and, especially, to modern urbanization; its generalization in the Western world is a reflection of these changes. It is likely that the use of PDS, as described in the previous paragraphs (increased prosodic cues, use of affective sentences, nicknames, etc.) is a recent phenomenon in the human-animal relationship. However, the use of particular frequencies and voice modulations when addressing dogs is probably an old and universal custom: hunters use high frequency variations to encourage their dogs to chase after prey; herders in Mongolia use human verbal language to address their dogs whereas they use non-verbal vocalizations for cattle ( Marchina 2015 ). It would be interesting to compare the way Western owners talk to their dogs with the way shepherds, hunters, soldiers and dog handlers communicate with their service dogs, as well as conduct cross-cultural comparisons.

Conclusion Overall, these studies show that dogs have a remarkable capacity to use gestural and oral modes of human communication as well as to decrypt human visual Cognition, emotions in dog domestication 243 attention (Topál et al . 2009 ; Kaminski et al . 2009 ; Lakatos et al . 2012 ). These abilities do not seem to have been inherited from their wild ancestors, because a comparison between hand-raised wolves and dogs show that wolves are not as competent as dogs in understanding and using human social communication cues ( Miklósi et al . 2003 ). What are the possible origins of these exceptional aptitudes? According to Topál et al. (2014), two non-exclusive hypotheses could explain the emergence of these abilities:

1 The by-product hypothesis claims that the selection of dogs based on their docility paved the way to the evolution of other cognitive competences, including the ability to read human communication signals. This hypothesis is supported by experiments carried out on foxes, the former initiated by the Russian geneticist Dmitry K. Belyaev in the late 1950s and subsequently reported by other researchers (Trut 1999), the results of which demonstrate that after an intensive selection based on docility criteria, foxes showed behavioural similarities with domestic dogs: they actively asked for human contact, they sniffed and licked humans, and groaned to draw their attention. The results also show that tamed foxes developed human cognitive skills that were not found in the control group. 2 The alternative hypothesis is the adaptation hypothesis that stipulates that humans actively selected dogs precisely on their exceptional aptitude to use human communication signals. This hypothesis is supported by the fact that puppies with only a little experience of humans manage to use human communication signals (e.g. gaze or pointing gesture) as easily as puppies used to human contact. Other authors claim that there is little scientific evidence that dogs’ sensitivity to human actions and intentions is a particular adaptation or a case of co-evolution: indeed, this sensitivity is found neither among the whole canine population nor, as outlined previously, among socialized wolves. Thus, authors such as Udell et al . (2010 ) suggest that the aptitude of certain dogs and a few wolves to respond to human behaviour is more the expression of a basic operant conditioning process on animals that have been socialized and released totally dependent on human beings.

The studies conducted on communications between owners and their dogs show that human beings have also adapted to dogs: PDS is used by owners when addressing their dogs as a means of communication to transmit their emotions and intentions, and may be used to some extent with the conscious or unconscious intention of teaching them the bases of human verbal language. Owners adapt to dogs’ cognitive and language skills using behavioural patterns partly derived from those used in the context of parent-infant communication. In this way, humans and dogs – two species distant at a phylogenetic level – have developed during the course of their evolution the ability to adjust to each other, an identical adaptive response to analogous environmental constraints. Thanks to these mutual adaptations, humans and dogs manage to cohabit. 244 Sarah Jeannin References Ainsworth, M.D.S. and Bell, S.M., 1970. Attachment, Exploration, and Separation: Illus- trated by the Behavior of One-Year-Olds in a Strange Situation. Child Development , 41 (1), 49–67. Albuquerque, N., Guo, K., Wilkinson, A., Savalli, C., Otta, E., and Mills, D., 2016. Dogs Recognize Dog and Human Emotions. Biology Letters , 12 (1), 20150883. Andics, A., Gábor, A., Gácsi, M., Faragó, T., Szabó, D., and Miklósi, Á., 2016. Neural Mechanisms for Lexical Processing in Dogs. Science , 353 (6303), 1030–1032. Andics, A., Gácsi, M., Faragó, T., Kis, A., and Miklósi, Á., 2014. Voice-Sensitive Regions in the Dog and Human Brain Are Revealed by Comparative fMRI. Current Biology , 24 (5), 574–578. Askew, H.R., 2003. Treatment of Behavior Problems in Dogs and Cats (second ed.) Berlin- Vienna: Blackwell Verlag GmbH, 8–16. Batt, S., 2009. Human Attitudes Towards Animals in Relation to Species Similarity to Humans: A Multivariate Approach. Bioscience Horizons: The International Journal of Student Research , 2 (2), 180–190. Belin, P., Zatorre, R.J., Lafaille, P., Ahad, P., and Pike, B., 2000. Voice-selective Areas in Human Auditory Cortex. Nature , 403 (6767), 309–312. Ben-Aderet, T., Gallego-Abenza, M., Reby, D. and Mathevon, N. 2017. Dog-Directed Speech: Why Do We Use It and Do Dogs Pay Attention to It?. Proceedings of the Royal Society of London B , 284 (1846), 2016–2429. Bowlby, J., May, D.S. and Solomon, M., 1989. Attachment Theory . Los Angeles, CA: Lifespan Learning Institute. Burnham, D., Kitamura, C., and Vollmer-Conna, U., 2002. What’s New, Pussycat? On Talking to Babies and Animals. Science , 296 (5572), 1435–1435. Catala, A., Mang, B., Wallis, L., and Huber, L., 2017. Dogs Demonstrate Perspective Tak- ing Based on Geometrical Gaze Following in a Guesser – Knower Task. Animal Cogni- tion , 20 (4), 581–589. Crawford, E.K., Worsham, N.L., and Swinehart, E.R., 2006. Benefits Derived From Com- panion Animals, and the Use of the Term ‘attachment’. Anthrozoös , 19 (2), 98–112. Custance, D. and Mayer, J., 2012. Empathic-like Responding by Domestic Dogs (Canis Familiaris) to Distress in Humans: An Exploratory Study. Animal Cognition , 15 (5), 851–859. Deputte, B.L. and Doll, A., 2011. Do Dogs Understand Human Facial Expressions? Jour- nal of Veterinary Behavior: Clinical Applications and Research , 6 (1), 78–79. Dorey, N.R., Udell, M.A.R., and Wynne, C.D.L., 2010. When Do Domestic Dogs, Canis Familiaris, Start to Understand Human Pointing? The Role of Ontogeny in the Develop- ment of Interspecies Communication. Animal Behaviour , 79 (1), 37–41. Dunst, C.J., Gorman, E., and Hamby, D.W., 2012. Preference for Infant-directed Speech in Preverbal Young Children. Center for Early Literacy Learning Reviews , 5, 1–13. Feldman, R., Weller, A., Zagoory-Sharon, O., and Levine, A., 2007. Evidence for a Neu- roendocrinological Foundation of Human Affiliation: Plasma Oxytocin Levels Across Pregnancy and the Postpartum Period Predict Mother-infant Bonding. Psychological Science , 18 (11), 965–970. Flom, R., Whipple, H., and Hyde, D., 2009. Infants’ Intermodal Perception of Canine (Canis Familairis) Facial Expressions and Vocalizations. Developmental Psychology , 45 (4), 1143–1151. Frantz, L.A.F., Mullin, V.E., Pionnier-Capitan, M., Lebrasseur, O., Ollivier, M., Perri, A., Linderholm, A., Mattiangeli, V., Teasdale, M.D., Dimopoulos, E.A., Tresset, A., Cognition, emotions in dog domestication 245 Duffraisse, M., McCormick, F., Bartosiewicz, L., Gál, E., Nyerges, É.A., Sablin, M.V., Bréhard, S., Mashkour, M., Bălăşescu, A., Gillet, B., Hughes, S., Chassaing, O., Hitte, C., Vigne, J.-D., Dobney, K., Hänni, C., Bradley, D.G., and Larson, G., 2016. Genomic and Archaeological Evidence Suggest a Dual Origin of Domestic Dogs. Science , 352 (6290), 1228–1231. Gácsi, M., Miklósi, A., Varga, O., Topál, J., and Csányi, V., 2004. Are Readers of Our Face Readers of Our Minds? Dogs (Canis Familiaris) Show Situation-dependent Recognition of Human’s Attention. Animal Cognition , 7 (3), 144–153. Golinkoff, R.M., Can, D.D., Soderstrom, M., and Hirsh-Pasek, K., 2015. (Baby) Talk To Me: The Social Context Of Infant-Directed Speech and Its Effects On Early Language Acquisition. Current Directions in Psychological Science , 24 (5), 339–344. Gómez, J.-C., 2005. Species Comparative Studies and Cognitive Development. Trends in Cognitive Sciences , 9 (3), 118–125. Hare, B., Brown, M., Williamson, C., and Tomasello, M., 2002. The Domestication of Social Cognition in Dogs. Science , 298 (5598), 1634–1636. Hare, B. and Tomasello, M., 2005. Human-like Social Skills in Dogs? Trends in Cognitive Sciences , 9 (9), 439–444. Heinrichs, M., von Dawans, B., and Domes, G., 2009. Oxytocin, Vasopressin, and Human Social Behavior. Frontiers in Neuroendocrinology , 30 (4), 548–557. Hirsh-Pasek, K. and Treiman, R., 1982. Doggerel: Motherese in a New Context*. Journal of Child Language , 9 (1), 229–237. Jeannin, S., Gilbert, C., and Leboucher, G., 2017a. Effect of Interaction Type on the Char- acteristics of Pet-directed Speech in Female Dog Owners. Animal Cognition, 20 (3), 499–509. Jeannin, S., Gilbert, C., Amy, M., and Leboucher, G., 2017b. Pet-directed Speech Draws Adult Dogs’ Attention More Efficiently than Adult-directed Speech. Scientific Reports , 7 (1), 4980. Kaminski, J., Bräuer, J., Call, J., and Tomasello, M., 2009. Domestic Dogs Are Sensitive to a Human’s Perspective. Behaviour , 146, 979–998. Kaminski, J., Call, J., and Fischer, J., 2004. Word Learning in a Domestic Dog: Evidence for ‘Fast Mapping’. Science (New York , N.Y.) , 304 (5677), 1682–1683. Lakatos, G., Gácsi, M., Topál, J., and Miklósi, A., 2012. Comprehension and Utilisation of Pointing Gestures and Gazing in Dog-human Communication in Relatively Complex Situations. Animal Cognition , 15 (2), 201–213. Marchina, C., 2015. Faire communauté. Etude anthropologique des relations entre les éleveurs et leurs animaux chez les peuples mongols (d’après l’exemple des Halh de Mongolie et des Bouriates d’Aga, Russie) , PhD thesis. Paris: INALCO. Mariti, C., Ricci, E., Zilocchi, M., and Gazzano, A., 2013. Owners as a Secure Base for Their Dogs. Behaviour , 150, 1275. Marshall-Pescini, S., Passalacqua, C., Miletto Petrazzini, M.E., Valsecchi, P., and Prato- Previde, E., 2012. Do Dogs (Canis lupus familiaris) Make Counterproductive Choices Because They Are Sensitive to Human Ostensive Cues? PLoS ONE , 7 (4), e35437. Merola, I., Prato-Previde, E., Lazzaroni, M., and Marshall-Pescini, S., 2014. Dogs’ Com- prehension of Referential Emotional Expressions: Familiar People and Familiar Emo- tions Are Easier. Animal Cognition , 17 (2), 373–385. Merola, I., Prato-Previde, E., and Marshall-Pescini, S., 2012. Dogs’ Social Referencing Towards Owners and Strangers. PLoS ONE , 7 (10), e47653. Miklósi, Á., Kubinyi, E., Topál, J., Gácsi, M., Virányi, Z., and Csányi, V., 2003. A Simple Reason for a Big Difference: Wolves Do Not Look Back at Humans, But Dogs Do. Cur- rent Biology , 13 (9), 763–766. 246 Sarah Jeannin Morey, D.F., 2006. Burying Key Evidence: The Social Bond Between Dogs and People. Journal of Archaeological Science , 33 (2), 158–175. Morton, E.S., 1977. On the Occurrence and Significance of Motivation-Structural Rules in Some Bird and Mammal Sounds. The American Naturalist , 111 (981), 855–869. Nagasawa, M., Mitsui, S., En, S., Ohtani, N., Ohta, M., Sakuma, Y., Onaka, T., Mogi, K., and Kikusui, T., 2015. Oxytocin-gaze Positive Loop and the Coevolution of Human-dog Bonds. Science , 348 (6232), 333–336. Ohala, J.J., 1984. An Ethological Perspective on Common Cross-Language Utilization of

F 0 of Voice. Phonetica , 41 (1), 1–16. Palmer, R. and Custance, D., 2008. A Counterbalanced Version of Ainsworth’s Strange Sit- uation Procedure Reveals Secure-base Effects in Dog – Human Relationships. Applied Animal Behaviour Science , 109 (2), 306–319. Passalacqua, C., Marshall-Pescini, S., Barnard, S., Lakatos, G., Valsecchi, P., and Prato Previde, E., 2011. Human-directed Gazing Behaviour in Puppies and Adult Dogs, Canis Lupus Familiaris. Animal Behaviour , 82 (5), 1043–1050. Payne, E., Bennett, P.C., and McGreevy, P.D., 2015. Current Perspectives on Attachment and Bonding in the Dog – Human Dyad. Psychology Research and Behavior Manage- ment , 8, 71–79. Pilley, J.W. and Reid, A.K., 2011. Border Collie Comprehends Object Names as Verbal Referents. Behavioural Processes , 86 (2), 184–195. Pongrácz, P., Molnár, C., Miklósi, A., and Csányi, V., 2005. Human Listeners Are Able to Classify Dog (Canis Familiaris) Barks Recorded in Different Situations. Journal of Comparative Psychology (Washington, D.C.: 1983) , 119 (2), 136–144. Rehn, T., Handlin, L., Uvnäs-Moberg, K., and Keeling, L.J., 2014. Dogs’ Endocrine and Behavioural Responses at Reunion Are Affected by How the Human Initiates Contact. Physiology & Behavior , 124 (Supplement C), 45–53. Ruffman, T. and Morris-Trainor, Z., 2011. Do Dogs Understand Human Emotional Expres- sions? Journal of Veterinary Behavior: Clinical Applications and Research, 6 (1), 97–98. Scheider, L., Grassmann, S., Kaminski, J., and Tomasello, M., 2011. Domestic Dogs Use Contextual Information and Tone of Voice When Following a Human Pointing Gesture. PLoS ONE , 6 (7), e21676. Schwab, C. and Huber, L., 2006. Obey or Not Obey? Dogs (Canis Familiaris) Behave Differently in Response to Attentional States of Their Owners. Journal of Comparative Psychology (Washington, D.C.: 1983) , 120 (3), 169–175. Serpell, J., 2004. Factors Influencing Human Attitudes to Animals and Their Welfare. Ani- mal Welfare , 13. Shinozaki, M., O’Day, S.J., Kitago, M., Amersi, F., Kuo, C., Kim, J., Wang, H.-J., and Hoon, D.S.B., 2007. Utility of Circulating B-RAF DNA Mutation in Serum for Moni- toring Melanoma Patients Receiving Biochemotherapy. Clinical Cancer Research: An Official Journal of the American Association for Cancer Research , 13 (7), 2068–2074. Somppi, S., Törnqvist, H., Kujala, M.V., Hänninen, L., Krause, C.M., and Vainio, O., 2016. Dogs Evaluate Threatening Facial Expressions by Their Biological Validity – Evidence From Gazing Patterns. PLoS ONE , 11 (1). Song, J.Y., Demuth, K., and Morgan, J., 2010. Effects of the Acoustic Properties of Infant- directed Speech on Infant Word Recognition. The Journal of the Acoustical Society of America , 128 (1), 389–400. Stoeckel, L.E., Palley, L.S., Gollub, R.L., Niemi, S.M., and Evins, A.E., 2014. Patterns of Brain Activation When Mothers View Their Own Child and Dog: An fMRI Study. PLoS ONE , 9 (10), e107205. Cognition, emotions in dog domestication 247 Szetei, V., Miklósi, Á., Topál, J., and Csányi, V., 2003. When Dogs Seem to Lose Their Nose: An Investigation on the Use of Visual and Olfactory Cues in Communicative Context Between Dog and Owner. Applied Animal Behaviour Science, 83 (2), 141–152. Topál, J., Byrne, R.W., Miklósi, A., and Csányi, V., 2006. Reproducing Human Actions and Action Sequences: ‘Do as I Do!’ in a Dog. Animal Cognition , 9 (4), 355–367. Topál, J., Gácsi, M., Miklósi, Á., Virányi, Z., Kubinyi, E., and Csányi, V., 2005. Attach- ment to Humans: A Comparative Study on Hand-reared Wolves and Differently Social- ized Dog Puppies. Animal Behaviour , 70 (6), 1367–1375. Topál, J., Gergely, G., Erdohegyi, A., Csibra, G., and Miklósi, A., 2009. Differential Sen- sitivity to Human Communication in Dogs, Wolves, and Human Infants. Science (New York , N.Y.) , 325 (5945), 1269–1272. Topál, J., Kis, A., and Oláh, K., 2014. Dogs’ Sensitivity to Human Ostensive Cues: A Unique Adaptation? A Unique Adaptation? Hungarian Academy of Sciences , 319–346. Topál, J., Miklósi, A., Csányi, V., and Dóka, A., 1998. Attachment Behavior in Dogs (Canis familiaris): A New Application of Ainsworth’s (1969) Strange Situation Test. Journal of Comparative Psychology (Washington, D.C.: 1983) , 112 (3), 219–229. Trainor, L.J., Austin, C.M., and Desjardins, R.N., 2000. Is Infant-Directed Speech Prosody a Result of the Vocal Expression of Emotion? Psychological Science , 11 (3), 188–195. Trut, L.N., 1999. Early Canid Domestication: The Farm-Fox Experiment: Foxes Bred for Tamability in a 40-Year Experiment Exhibit Remarkable Transformations that Sug- gest an Interplay Between Behavioral Genetics and Development. American Scientist , 87 (2), 160–169. Turcsán, B., Szánthó, F., Miklósi, Á., and Kubinyi, E., 2015. Fetching What the Owner Prefers? Dogs Recognize Disgust and Happiness in Human Behaviour. Animal Cogni- tion , 18 (1), 83–94. Udell, M.A.R., Dorey, N.R., and Wynne, C.D.L., 2010. What Did Domestication Do to Dogs? A New Account of Dogs’ Sensitivity to Human Actions. Biological Reviews of the Cambridge Philosophical Society , 85 (2), 327–345. Udell, M.A.R. and Wynne, C.D., 2008. A Review of Domestic Dogs’ (Canis Familiaris) Human-Like Behaviors: Or Why Behavior Analysts Should Stop Worrying and Love Their Dogs. Journal of the Experimental Analysis of Behavior , 89 (2), 247–261. Ujfalussy, D.J., Kurys, A., Kubinyi, E., Gácsi, M., and Virányi, Z., 2017. Differences in Greeting Behaviour Towards Humans With Varying Levels of Familiarity in Hand- reared Wolves (Canis Lupus). Royal Society Open Science , 4 (6), 160956. Vas, J., Topál, J., Gácsi, M., Miklósi, Á., and Csányi, V., 2005. A Friend or an Enemy? Dogs’ Reaction to an Unfamiliar Person Showing Behavioural Cues of Threat and Friendliness at Different Times. Applied Animal Behaviour Science , 94 (1), 99–115. Yong, M.H. and Ruffman, T., 2014. Emotional Contagion: Dogs and Humans Show a Similar Physiological Response to Human Infant Crying. Behavioural Processes , 108 (Supplement C), 155–165.

Part IV Ongoing transformations

1 4 Domestication and animal labour

Jocelyne Porcher and Sophie Nicod

Domestication processes have long been described as power relationships, which only consider animals for their productive or commercial functions. The 19th century’s zootechnics (animal production science) embodied this trend. “Zootech- nics is the science for the appropriation and industrial exploitation of animals to serve our needs, in order to produce various goods and services in the economy” ( Roland, 1877 , p 3). In the first zootechnicians’ minds, animals must be regarded as animal machines, not because they are really machines but since they fulfil the same functions in the economy. In 1888, André Sanson wrote: “Some domesticated animals are described as animal machines, precisely because of their economic functions; these functions distinguish them from other animals, also living alongside men, in his home (domus ) through the various services they produce or provide” ( Sanson, 1907 , p 19, first edition 1888). The framework of domination, coupled with an instrumental vision of our relations to animals, have inhibited our ability to consider domestication from the perspective of labour and led to creation of a permanent opposition between animals with an economic and therefore productive value: farm animals considered in a generic way – i.e. bovine for tractive power, sheep for wool, cow for milk, pig for meat and fat – as too often described in children’s books, and animals without any economic functions, who would only remain unproductive at their master’s home without ever bringing anything back, or pets. Pets are not considered by zootechnicians as having an economic function as they do not contribute to the production of marketable goods (meat, wool, eggs and traction power.). This divide between animals “for production” and animals “for companionship” exposes for a large part the relations between cities and the countryside, but above all else it shows the opposition of social classes between the 18th century aristocrats (Mac- Donogh, 1999) and farmers, between the 19th century bourgeois and proletarians with the expansion of industrial capitalism later on ( Porcher, 2009 ; Serna, 2016 ). Animals with an economic function (for transport, horses in mines, bovines for traction, farm dogs, etc.) are used by farmers and workers, and those with no economic function (leisure horses, pets, etc.) belong to aristocrats and the bourgeois. It is important to stress that ever since the 19th century, the economic function of farm animals has been asserted and theorized from the sole point of view of humans, leaving out of the debate the issue of animal labour. Yet, the economic 252 Jocelyne Porcher and Sophie Nicod function of these animals lies indeed within the broader issue of production and therefore the issue of labour. If labour is undoubtedly a rational mode of production, production is not its first and sole purpose. This rational mode of production is backed by relational interactions of all involved, just as the secondary sociality – working relationships or charity work – is backed by a primary sociality – family ties, friendly and love relationships ( Caillé, 1994 ). Working is producing, but it is mainly living together (Dejours, 2012 – among humans, but also among animals. Domesticated animals produce goods; we produce foods from farm animals we breed, but also services with the help of dogs, horses and elephants, among others. Millions of animals are now at work with humans. These animals contribute to human work in many aspects and activities ( Estebanez et al., 2017 ; Mouret, 2017 ; Porcher, 2017a ). That is to say that beyond the restrictions of the organisation of labour, they are engaged individually and collectively in labour. Labour is what historically defines and secures the domestication processes of animals by men. The same applies to livestock farming, as far as cows and sheep are concerned (Porcher and Schmitt, 2012 ). We can argue that when the working relation collapses, so does the domestic relationship with the animals.

A monstrous cohabitation? When you consider that animals actually work, this has a great impact on our way to approach domestication processes as well as their actual and future effectiveness. Especially because of the fact that nowadays, in Western countries but not only in Western countries, our relations with domesticated animals are threatened. Domesticated animals have been singled out in the environmental crisis (livestock breeding would have an important responsibility in the production of greenhouse gases, water pollution and the reduction of biodiversity; Steinfeld et al., 2006 ). They have also been singled out because of the profound changes in the balances of power in the economy of food products; industrial agriculture is deemed obsolete by multinational companies and foodtech 4.0 startups which promote a cleaner, more profitable and socially acceptable agriculture (Porcher, 2010a. Finally, our relations with domesticated animals are threatened by an ideology which aims at “liberating” animals, and “abolishing meat consumption” and beyond that, an agriculture without livestock farming. The idea that domestication is a monstrous cohabitation, as described by Sloterdijk (2000 ), and that our relations to animals only refer to the register of exploitation and violence, leads some to defend the end of livestock breeding, the “liberation” of animals and the return of animals to wildlife. Among many theoricians of this movement, a surprising distinction is established, in spite of expressed anti-speciesist intentions, between farm animals and pets, although both are domesticated animals. In a caricatural way, we can argue that they want to free cows, i.e. send them back into the “wild”; give identification cards to dogs and recognize their common citizenship. We should give up the “forced involvement” of animals in human activities. Consequently, Donaldson and Kymlicka Domestication and animal labour 253 (2011), for example, defend the idea of animals in politics overlooking almost completely the highly political question of labour. Yet, labour is the main vector of social relations among humans, the main vector of relations between humans and animals, the main vector of animals’ agency and the most important place for political representation of animals. Animal labour is not a task free of purpose, and this constitutes an upsetting difficulty to reflect on animal labour for those who want to “free” animals. Animals are involved in labour activities, and without their collaboration, no work could really happen (Porcher, 2017b). Think- ing about the place of animals in social relations should therefore not elude the issue of animals’ place in labour. This issue must be taken seriously in all its complexities.

Back to the wild Farm animals’ “liberation” from human exploitation implies the notion that these animals will enjoy the benefits of their “wild state” and a fulfilled existence. This proposal is notably to be found a Lévi-Strauss article written in 2001, p 14 in the midst of the “mad cow” disease incident:

Agronomists will find a way to increase the protein content of food plants, chemists will produce synthesised proteins in industrial quantities (. . .) livestock breeding, a non-profitable activity, having disappeared completely, we will buy this luxury meat which will be the result of hunts. Our former cattle, delivered and abandoned to themselves, will be game and preys in a wild countryside handed over to Nature.

One might argue that Levi-Strauss’s wishes are being accomplished. Chemists are indeed working at creating alternatives to meat-based foods, in vitro meat for example (Porcher, 2010b), and animals are indeed being returned to the wild in ever-larger numbers. This reversion to their wild status is de facto the consequence of animal release from labour. As we wrote previously, taming is “working with”. When the labour relation ceases to exist, the domestication relation is weakened or even broken. Mechanization is one of the reasons for this release. That is what was observed with French donkeys that owe the maintenance of the species within the social space to their reconversion in tourism activities. That is also what happened in India as far as elephants are concerned (Lainé, 2016, this volume). The economic context environment can also weaken the relation. That is the case for Irish horses. If Irish horses do benefit from a tourism value and advertising, the reality of their existence is far from what is promoted by Irish tourism offices. Garrano ponies found themselves in a similar situation in Portugal. Before 2008, horses in Ireland enjoyed a very high value and sold for great amounts of money due to an Irish passion for horses. They then suffered when the economic crisis hit the country very hard. Horse owners could no longer afford to undertake financially the cost of their care and maintenance. They often decided to 254 Jocelyne Porcher and Sophie Nicod get rid of them either by selling them cheaply to the meat industry or by “releasing” them. Irish horses then became lost or astray, and suffered from hunger. Forestry officers who regard them as a nuisance can shoot them. They sometimes find shelter in farming families who need to re-initiate a taming process (Dayan, 2014).

The example of Garrano ponies in Portugal What happened to horses in Ireland also happened to Garrano ponies in Portugal, whose status through time has fluctuated from a feral animal caught by breeders and whose genotype has mutated thanks to domesticated relations (Cortés et al., 2017 ) to a domesticated animal “returned” to the wild. The Garrano is a small and brown horse standing about 1.35 meters to the withers with a dark brown robe living in the Pénéda Gères National Park in northwestern Portugal. It is a medium-sized mountainous site crossed by three valleys where eucalyptus and cork oaks grow. Broom, gorse, scrub and thorny plants form the food base of this equine animal. They have strong lower jaws, and their moustaches made of many strong whiskers protect them from stings of thorny plants. They have completely adapted to their natural environment. The Garrano pony has been a tireless and helpful collaborator, present in all aspects of human activities in Portugal. Its work has generated economic wealth and contributed to the country’s grandeur. It still works nowadays in equine tourism activities and one traditional activity: the “passo travado” races. This is a “restrained pace” race. This speed is quicker and more comfortable than the natural pace of the pony, but slower than the trot speed. It would allow farmers to travel long distances in mountainous regions without excessive tiredness for the pony. However, the place of this small horse in the Portuguese society has considerably changed. Their number went from 9,000 domesticated individuals before the 1960s to just 2,000 in the year 2011.1 In 1994, the CEREOPA (Centre d’Etudes et de Recherche sur l’Economie et l’Organisation des Productions Animales, or Centre for Studies and Research on the Organisation and Economy of Animal Production) classified the Garrano pony as an endangered breed. This classification was however decided more as an objective to protect a wild animal participating in the natural park’s food chain, rather than protecting the animal that took a massive part in the economic, social and cultural life of the country and which was for so long involved in labour alongside humans. As in some other European regions, European wildlife protection programs were first designed to finance the preservation of the breed in the form of subsidies and compensations to breeders for each identified animal. This headcount was first carried out by a simple statement from farmers. As explained by one Garrano breeder (June 2016): “EEC technicians would come, we would go in the mountain. They would point and say: ‘Is this your horse?’ ‘Yes’ would answer the farmer, and he would get the subsidy.” However, from 2000 onwards, microchipping became compulsory in order to identify any horse and its owner in the region. Owners had to declare and identify their own horses in order to get the European subsidies. But in 2008, the economic Domestication and animal labour 255 crisis hit Europe very hard – and especially Portugal, which received the injunc- tion to reduce its public debt. The consequences on the preservation program were twofold. On the one side, European subsidies amounts shrunk, and on the other, the Portuguese government added new taxes on these European subsidies. As a result, owning Garrano ponies became less and less financially attractive. Plus, subsidies became scarcer as the horse became less and less protected by their former owners. Ultimately, an increasing number of these animals were “released” and found themselves without an owner, without work and therefore without any protection whatsoever. These “wild” horses are now considered a nuisance and threatened by farmers to be shot as they are seeking fresh food in cultivated fields when they starve during the cold winter period. Breeders accuse them of grazing on the grass destined for their goats or cows, and often use their rifles to make them flee or shoot them dead. In order to avoid these killings, the Pénéda Gérès National Park has appointed a controller, who monitors horse families and push them back into the mountain when they come too close to the city. From co-worker, the Garrano pony has become a pest to be eliminated. What is more, it became an easy target for wolves to prey on.

The promise that labour holds The issue of labour is paramount in our understanding of domesticated animal- human relations. Domestic animals are not involved just randomly in domesticated relations with humans; they undoubtedly take advantage of the situation. Survival is the first advantage they draw from it since the first bound between animals and humans is protection. In both cases explained earlier, it is this “promise of protection” offers by labour which disappears in the first place. The horses are no longer protected, being out of the breeders’ scrutiny. They have become game for wolves and prey for humans who consider them as no longer wanted. Let us remember that the concept of labor does not refer only to relationships of coercion, exploitation and domination, even if the working relationships between bosses and workers and between humans and animals are asymmetrical. Work- ing with humans as with animals is first of all living together and being engaged together in a productive activity of value (Dejours, 2007. Labor therefore concerns not only draft animals, and does not refer only to the production of impel- ling energy. Nor does it depend on the presence of a contract or remuneration. Our research results (Porcher, 2017d) show that animals engage their subjectivity and their affectivity in the work. They demonstrate their agentivity, this thanks to which we can work them. Without their involvement in the work, none of the productions we realize through them could be exist. As we have described and contrary to liberationist theories, putting animals out of work is putting them out of their relations with humans and sentencing them to a costly freedom with many consequences for the individuals. Everyone knows about the tale of the wolf and the dog, highlighting the higher value of freedom over domestication. The dog’s collar around his neck would not 256 Jocelyne Porcher and Sophie Nicod be accepted by the wolf under any circumstances (Porcher and Lécrivain, 2017). The vision of feral horses galloping in Irish or Portuguese hills could also lead us to believe that the “re-wilding” of domesticated animals would benefit them more than being domesticated. But when you look closer, the reality is somewhat more complicated: horses being hunted down by predators, falling sick, getting hurt or starved, etc. The example of “re-taming” processes led by some breeders and the reintegration of horses within a human space show that horses do accept or even “wish” to give up this freedom which they did not even ask for. The involvement of animals in human labour is undoubtedly a constraint, an undeniable loss of freedom, but it is potentially also an original source in the liberty to act. Labour holds promises, for humans and animals alike. Creativity, self-fulfilment and self-betterment are among them. Domesticated horses are no longer “natural” beings. Their world is made up of their own world, the world of their species generated by their own senses ( Straus, 2000 ), but also by the human world through labour. It is in this interface between their own world and the human world built by labour activities that domesticated animals find purpose in their existence and enjoy a richer, more interesting, surprising and challenging life compared to a life outside of the human. It is labour’s fundamental promise. However, just like for humans, there is sometimes a great difference between promises and the reality of labour. For the vast majority of domesticated animals, labour does not hold its promises. It holds its promises neither for farm animals nor for almost all other animals like dogs, circus animals, natural parks animals, etc., involved in labour. Even after three decades of research funding on “animal welfare” designed supposedly to engage changes in the life conditions of animals at work, this issue of “animal welfare”, as it belongs to the field of “natural” sciences, overlooked from the start the question of labour and that of the intersubjectivity of animal-human relations at work, and reduced animal conduct when they are involved in labour to mere behaviours. Theoretical frameworks of biology and applied ethology do not provide an accurate toolbox to tackle the issue of labour, the subjectivity and intersubjectivity – notions that are radically attached to social sciences. Abandoned horses in Ireland and Portugal were dismissed from work by their owners without any form of closure or compensation. One might argue that it happens every day to familiar animals like cats, ferrets or others, which get “released” into the woods by their owners. That is the true nature of “animal liberation”: a return to freedom, which hides the fact that our social, political and environmental constraints, as well as the economic system in which we live as a whole – capitalism – leaves no place for animals. Only by taking into account animals in the field of labour can we change their situation in a lasting way. Our research on animal labour shows how labour organizes our relations to animals, and how it transforms animals and humans. In order to work together, and to cooperate, labour rules need to be laid down, and as we noted, these rules are not laid down only by humans. Being able to communicate, being in a trustworthy exchange, giving and getting recognition is key. Continuing to live alongside animals, not renouncing calls to be humane with animals, implies profound changes Domestication and animal labour 257 in our relation and in their relation to labour, even if this proposal may seem offset under the current social and economic situation. Living with animals, working with them, is about maintaining the cycle of gift and reciprocity, which started with the first domestication bonds. We are involved with animals in the cycle of give-receive-giveback, as described by Mauss (1999 [1923 ]). However, the main difficulty is that the organizations of labour as well as the economic context of our relations to animal do not allow us to, individually and collectively, give back what the animals have given us. We owe them a debt. And just as some breeders say, we are not living up to our responsibility (Porcher, 2002 ).

Conclusion Against all odds, social theories on labour and gift exchange offer relevant frameworks to shape our thinking on animal relations with men and conditions for their durability by describing their complex and often contradictory dynamics. Working with animals in the future might not be obvious anymore. Gradually, our relations with animals in labour have become difficult and even impossible to maintain, particularly through the regulations on “animal welfare” which impose rules going against the autonomy of breeders and their animals.2 This trend has been accentuated by the actions of “liberators” whose real project is to free themselves from domesticated animals in order to settle the issue of the moral debt we owe animals, and through the robotization of labour ( Porcher, 2017c ). We therefore stand today at a very unique moment in our millennial history with animals. For the first time since we tamed them, we consider breaking the bonds and creating a society without animals – this even before we really perceive and understand the wealth that they have been giving us, and have taken the measure of their intelligence and abilities. When all animals will have been “released”, when robots will have replaced them in our daily lives and when we will only have food substitutes in our plates, just as Philip K. Dick’s characters, we will probably wonder when exactly we made this major mistake for our future.

Acknowledgement Many thanks to Guillaume Jeandenans for the English translation.

Notes 1 Approximately 1,500 females and 150 stallions. This is the number declared by owners (statistics given by the Portuguese Association portugaise for the protection of Garranos). 2 Regulations on “animal welfare” aim firts at making acceptable industrial animal production systems (industrialized and intensified); however, they greatly impact livestock breeding, breeders and their animals, as these regulations do not take into account the differences between livestock breeding and animal production, between breeding animals and producing them. It should be noted that scientits and drafters of these legisla- tions are not experts in livestock breeding. 258 Jocelyne Porcher and Sophie Nicod References Caillé, A., 1994. Don, intérêt et désintéressement . Paris: La Découverte. Cortés, O., Dunner, S., Gama, L.T., Martínez, A.M., Delgado, J.V., Ginja, C., Jiménez, L.M., Jordana, J., Luis, C., Ooom, M.M., Sponenberg, D.P., Zaragoza, P., Consortium, B., and Vega-Pla, J.L., 2017. The Legacy of Columbus in American Horse Populations Assessed by Microsatellite Markers. Journal of Animal Breeding and Genetics , Online early. doi:10.1111/jbg.12255 (IF2015 1,745; Q1 Agriculture, Dairy & Animal Science). Dayan, E., 2014. Boom Horses . Film. Kepler Production. https://vimeo.com/193899162 Dejours C., 2007. Subjectivity, work and action. In Jean-Philippe Deranty, Danielle Peth- erbridge, J. Rundell and Robert Sinnerbrink (eds.), Recognition, Work, Politics. New Directions in French Critical Theory . Leiden and Boston: Brill, 71–88. Dejours, C., 2012. From the Psychopathology to the Psychodynamics of Work. In Nicholas H. Smith and Jean-Philippe Deranty (eds.), New Philosophies of Labour – Work and the Social Bond . Leiden and Boston: Brill, 209–250. Donaldson, D. and Kymlicka, W., 2011. Zoopolis: A Political Theory of Animal Rights . Oxford: Oxford University Press. Estebanez, J., Porcher, J., and Douine, J., 2017. Travailler à faire semblant: les animaux au cinéma. Ecologie et Politique , 54, 103–124. Lainé, N., 2016. Conduct and Collaboration in Human-Elephant Working Communities of Northeast India. In P. Locke and J. Buckingham (eds.), Conflict, Negotiation, and Coexistence: Rethinking Human-Elephant Relations in South Asia. New Delhi: Oxford University Press, 180–205. Lévi-Strauss, C., 2001. La leçon de sagesse des vaches folles. Études rurales , 157–158, 9–14. MacDonogh, K., 1999. Reigning Cats and Dogs: A History of Pets at Court Since the Renaissance . USA: Saint Martin’s Press. Mauss, M., 1999 (1923). Essai sur le don, dans Sociologie et anthropologie. Paris: Presses universitaires de France. Mouret, S., 2017. Apprendre à prendre soin. La centralité du travail dans l’éducation des chiens guides d’aveugles. Ecologie et politique , 54, 87–102. Porcher, J., 2002. Eleveurs et animaux, réinventer le lien . Paris: Presses universitaires de France. Porcher, J., 2009. Bêtes de somme. Ravages , 3, 115–135. Porcher, J., 2010a. Ce que les animaux nous donnent en nature. Revue du MAUSS , 42, 49–62. Porcher, J., 2010b. La production de viandes in-vitro, stade ultime? La Revue Politique et Parlementaire , 1057, 97–104. Porcher, J., 2017a. The Ethics of Animal Labour: A Collaborative Utopia. USA: Palgrave. Porcher, J., 2017b. Animal work. In L. Kalof (ed.), The Oxford Book of Animal Studies . Oxford: Oxford University Press. Porcher, J., 2017c. Elmo et Paro. Pourquoi l’un travaille et l’autre pas, et ce que cela change. Ecologie et Politique , 54, 17–34. Porcher, J., 2017d. Le programme ANR COW: l’ouverture d’un front de recherches inédit sur le travail animal. The Natural Sciences Society , 25 (2), 172–179. Porcher, J., and Lécrivain, E., 2017. La louve et la chienne patou: la liberté et le travail. Ecologie et politique , 54, 65–78. Porcher, J. and Schmitt, T., 2012. Dairy Cows, Workers in the Shadows. Society and Ani- mals , 20 (1), 39–60. Domestication and animal labour 259 Roland, F., 1877. Programme d’un cours de zootechnie . Paris: Bibliothèque Nationale. Sanson, A., 1907. Traité de zootechnie . Tome I. Paris: Librairie agricole de la Maison Rus- tique, 5ème édition. 1ère édition, 1888. Serna, P., 2016. L’animal en République. 1789–1802 Genèse du droit des bêtes . Toulouse: Anacharsis. Sloterdijk, P., 2000. Règles pour le parc humain . Paris: Mille et une nuits. Steinfeld, H. et al., 2006. Livestock Long Shadow . Environmental issues and options. Rome: Food and Agriculture Organization of the United Nations. www.fao.org/docrep/010/ a0701e/a0701e00.HTM . Straus, E., 2000. Du sens des sens: contribution à l’étude des fondements de la psycholo- gie . Grenoble: Jérôme Million.

15 Human-dog-reindeer communities in the Siberian Arctic and Subarctic

Konstantin Klokov and Vladimir Davydov

Most research on animal domestication focuses on the relationships between humans and one particular species. However, only the dog, the first species to be domesticated, enjoyed the exclusivity of being the sole domestic animal present when it was integrated into human societies. All other animals domesticated thereafter had to adapt to a new environment peopled not only by humans, but also by other species, even if only the dog. Anthropized environments were more exactly anthropo-canified environments. In this chapter, we investigate domestication processes through the interactions between three species: humans, dogs and reindeer in North Asia. How can these three species negotiate a shared livelihood, coexist and even cooperate in spite of their differences in needs and behavior? How does each species change the biosocial environment of the other two? What kind of different hybrid communities do they shape in the different cultural and ecological contexts of the North? The purpose of this chapter is to investigate the synergy of collaborative activities of humans, dogs, and reindeer in hybrid communities (hereafter HDR communities) in northern Russia. To do this, we utilize data from the 1926–1927 Polar Census, a unique project initiated by the Soviet Central Statistical Administra- tion to gather primary data on the whereabouts, economy, and living conditions of the population living in the Arctic and Subarctic (Pokhoziaistvennaia perepis’ 1929 ; Anderson 2011 ); we have supplemented this with information from literary sources from the early 1900s and data from our own fieldwork in different regions of Siberia. These sources of information allow us to present a fairly complete picture of the coexistence and mutual activities of animals and humans in HDR communities in tundra, taiga (boreal forest), mountainous, and coastal (maritime) landscapes. The Polar Census contains data on just over 33,000 nomadic and settled local households, 270,000 people, 100,000 dogs, and 1,800,000 reindeer ( Table 15.1 ). The Polar Census primarily registered households involved in reindeer herding, fishing, and hunting for fur-bearing animals, sea mammals, ungulates, waterfowl, and upland game birds (Tetraonidae ), as well as gathering berries, mushrooms, birds’ eggs, and many other items. The area covered by the Polar Census was a special region, which in Russia is usually called the North or Far North. Accord- ing to the data from the Polar Census, 35.4% of its population was nomadic, and 262 Konstantin Klokov and Vladimir Davydov Table 15.1 Number of people, dogs, and reindeer in the households of the Russian North. Calculated on the basis of data from the 1926–1927 Polar Census ( Pokhoziaistvennaia perepis’ 1929 ).

Settled households Nomadic households Totals

Households 33,641 23,429 57,070 People 164,587 106,263 270,850 Households with dogs 13,419 8,354 21,773 Adult sled dogs 48,666 5,817 54,484 Adult hunting dogs 9,626 4,982 14,608 Adult pastoral dogs 1,267 10,284 11,551 Dogs under 1 year of age 16,828 5,380 22,208 Households with reindeer 4,218 9,654 13,872 Gross reindeer holdings 78,540 1,732,231 1,810,771 Transport reindeer 36,458 364,800 401,258 the remaining 64.6% were residents of small villages and trading posts. These included communities from more than 20 different indigenous northern peoples speaking different languages, and small groups of Russians who had intermar- ried with them. The latter partly adopted the aboriginal populations’ lifestyles. We argue that in these northern communities, cooperation between people, dogs, and reindeer, through which various forms of traditional economy developed, successfully existed, and were transformed through various historical processes. At the same time, the role played by people, reindeer, and dogs, and the functions that each category carried out in these symbiotic communities, changed in response to political, economic, and social challenges, some of which emerged from the center of Russia. Cooperation was achieved through a variety of innate and acquired abilities and practices of each of the three participants in the HDR community. Some of these abilities are probably genetic, while others have been acquired in the course of their joint activities ( Stépanoff 2012 ). Particular abilities were tied to each individual’s intrinsic characteristics, while others were worked out in the course of their joint activities. The development of relevant capacities allows everyone to contribute to the overall synergistic activity of the community, namely to perform a specific set of functions forming an integrated part of every type of community in various forms and combinations. On the one hand, these combinations depended on local conditions – relations of the community with the natural “sustaining” landscape. On the other hand, they were also contingent on the historical traditions of each ethnic group. As a result, even within very similar types of environments, the structure of HDR communities and the roles of each of the three participants were often significantly different.

Human-dog-reindeer household typological groups The maps which in Figures 15.1 and 15.2 made it possible to characterize HDR communities in all of the 56 census areas, are based on the data provided by the Figure 15.1 Ratio of reindeer, sled, pastoral, and hunting dogs in the nomadic households of the Russian North, according to the data from the Polar Census of 1926–1927. Figure 15.2 Ratio of reindeer, sled, pastoral, and hunting dogs in settled households of the Russian population of the Russian North, according to the data from the Polar Census of 1926–1927. Human-dog-reindeer communities in Siberia 265 Polar Census. Nomadic ( Figure 15.1 ) and settled (Figure 15.2) households were presented in the materials of the Polar Census separately, while settled households were further divided into sedentary indigenous and settled Russian groups. The ratio of dogs and reindeer in nomadic and settled households significantly varied, but the differences between Russian and indigenous settled households were rather small. Nomadic households in the tundra possessed by far the largest percentage (95.7%) of all reindeer. They also had the largest reindeer herds, the majority of which were kept largely as sources of meat and skins. In contrast, the nomadic peoples of the taiga, as well as many settled native and Russian households, kept reindeer primarily as transport animals. Since they required relatively few transport reindeer, their herds were not as large as those of the tundra nomadic households. Sedentary households owned the largest percentage of dogs (74.3%), and these were categorized in the census data as sled dogs. The nomadic groups employed reindeer sleds or rode on reindeer, and in turn had a small number of sled dogs, or sometimes none at all. In the census data, most nomadic households also possessed pastoral dogs, while hunting dogs were present both in nomadic and settled families ( Table 15.1 ). The analysis of the data from the Polar Census in a regional context where we used just one formal criterion – the ratio of dog and reindeer numbers – allowed us to distinguish six main typological groups of HDR communities ( Table 15.2 ). The first group includes households of sea mammal hunters and fishers (mainly Chukchis, Eskimos, Koryaks, and Russians) living on the coasts of the Pacific and the Arctic oceans. As a rule, these households had no deer, and possessed many sled dogs, which form the primary means of winter transportation. Dogs were also particularly critical in hunting seals and polar bears. To date, however, sled dogs have almost entirely disappeared in these areas. In Chukotka and Kamchatka dog sledding has now become mostly a hobby. Since 1991 dog sled races have regularly taken place in which 10–20 sleds travel several hundred kilometers along the coast of the Pacific Ocean (Bogoslovskaia 2011: 38–65). Furthermore, dog sled races have also become an important part of Reindeer herder’s day celebrations in Dolgan villages in Taimyr (Davydov – Taimyr Peninsula field notes). The second group of HDR communities is based on the large herds of reindeer, which people keep primarily as sources of meat and that take part in annual migrations across the tundra, sometimes traveling hundreds of kilometers. This is common in the western part of the Russian Arctic in Nenets and Komi-Izhemtsy nomadic households, where dogs are commonly actively involved in herding large groups of deer. The other groups engaged in reindeer herding in the tundra possessed much smaller numbers of herding dogs. However, by the mid-twentieth century, reindeer herding dogs had become widely used in eastern tundra areas ( Baskin 2009 : 251, 255). The third group characterizes nomadic households of eastern Chukotka and the Koryak okrug (district) of the Kamchatskii krai (region), northern Kamchatka. These households have large reindeer herds, but almost no herding dogs, while Table 15.2 Main typological groups of HDR communities.

Types of HDR Landscape Way of Region Number % of house- Number of Incl. % of house- Sled dogs per Hunting Herding communities life of households with reindeer per transport holds with household dogs per dogs per holds reindeer household reindeer dog household household

1 No reindeer, dogs only Seashore Settled Mostly 4,878 6.0% 8.4 2.4 80.7% 9.0 0.2 0 for transportation Far East 2 Large reindeer herds Tundra Nomad All 4,316 96,5% 209.7 52.3 77.7% 0.1 0.2 2.1 with herding dogs regions 3 Large reindeer herds Tundra not Nomad Northeast 2,233 94.3% 312.8 30.9 88.0% 2.9 0.7 0.5 without herding dogs, far from reindeer and dogs for the ocean transportation 4 Few reindeer for Taiga Mostly All 11,250 46.2% 20.6 8.3 69.3% 0.1 1.5 0.1 transportation and nomad regions dogs for hunting 5 Reindeer for Taiga Settled Kola 982 69.6% 9.0 4.1 5.55 % 0 0.5 0.5 transportation Peninsula (Pomors) 6 Few reindeer for All All All 4,032 6.1% 15.7 8.4 43.4% 0.3 0.6 0.3 transportation and few dogs for different aims Human-dog-reindeer communities in Siberia 267 some sled dogs are to be found nearly everywhere. Although this looks strange at first glance, this general pattern can be explained. Historically, Chukchi can be divided into nomadic reindeer herders of the tundra, who traveled with the help of reindeer, and sedentary sea hunters and fishers who lived on the coast and used dog sleds as the main means of conveyance. Households from these two Chukchi groups maintained ties with each other, including through family visits to each other’s homes (Vaté 2005). Sled dogs in the coastal households were not accustomed to reindeer, and could attack and bite the visiting sled deer while they were tethered in the marine communities. Therefore, reindeer herders often traveled with dog sleds instead of reindeer sleds when visiting the sea coast (Klokov – Chukotka field notes). The fourth group of HDR households is characterized by nomadic populations in the taiga. These communities included a small number of well-tamed reindeer, most of which were used as riding and/or cargo (pack) reindeer. While herding dogs were absent, hunting dogs played critical roles in these households. Human hunters in these households, riding on deer, would follow the hunting dogs who would lead them to animals such as squirrel and sable. Hunting fur animals represented the main source of income for nomadic taiga households. Furthermore, dogs helped to hunt elk and bear. In these types of hunts, transport reindeer helped to cover large distances in the taiga. This type of HDR household, widely spread across the Siberian taiga in the past, is now practiced only in particular areas. Mechanical transport, in particular snowmobiles and tracked vehicles, has nearly nullified the benefits of hunting with reindeer. The number of reindeer herder families who continue to nomadize the taiga is now extremely small. Such households still can be found in parts of Yakutia, as well as in some areas of the Irkutskaia and Amurskaia regions ( oblasti), the republics of Buryatia and Tuva, and Khabarovskii and Zabaikal’skii regions ( kraia ). The fifth group includes sedentary households that held a significant number of reindeer (so-called izbennoe olenevodstvo; Koz’min 2003: 95–122). This type of household was common mostly for Russian populations (Pomors) in the Kola Peninsula. Similar households can also be found in some areas of Yakutia. People in these households kept a small number of dogs mostly for hunting. This type of farm disappeared in the middle of the twentieth century, when reindeer were replaced by mechanical transport. Communities where reindeer and dogs were present in small numbers we ten- tatively attributed to the sixth, and last, group.

Interactions and roles in HDR households Here we will describe the complementary roles and the requirements of the three species involved in these communities. Humans overall played a series of roles in relation to dogs and reindeer. Herd- ers of course supervised and guided the movement of reindeer herds, as a kaiur , or sled driver, and people guided reindeer and dogs who pulled sleds, transporting 268 Konstantin Klokov and Vladimir Davydov people or cargo. People also directed riding deer that carried them upon their backs, as well as transport reindeer, which were loaded with cargo. Hunters worked in combination with dogs and reindeer in pursing fur-bearing animals as well as ungulates, with the reindeer carrying the person, and the dog leading both to the prey. All of these ways of interacting with reindeer and dogs involved human attentiveness and acts of care and protection. Both reindeer and dogs sometimes required protection from predators, and both would have to be appropriately interacted with (“tamed”) to help ensure they were suitably prepared for their required roles. Dogs and reindeer were selectively bred, which was achieved through isolating choice animals with one another, and by removing animals from the breeding pool by killing or castrating them. Further, reindeer (far more so than dogs) had to be killed and transformed into food, clothing, dwellings, and other things. All of these practices involved yet other sets of things – sleds, harnesses, saddles, fences, knives, containers, fires, dwellings, entire landscapes – many of which required construction and maintenance ( Anderson et al . 2017 ). The spatial dimension of people’s relationships with these animals is of great importance. However, the explanatory models of human-animal relations often focus on human choices and underestimate the role of animal agency in this process ( Stépanoff 2017 ). Generally, people in various ways try to limit and direct the movement of dogs and deer while, at the same time, move in a particular geographical area, taking into consideration the interests and needs of their animals. To a large extent people “adjust” their seasonal rhythms to the grazing requirements of reindeer, as well as to the needs and capabilities of dogs. Reindeer provided humans a range of raw materials such as meat, skins, antlers, and milk, but also gave them labor by pulling and carrying people and goods, and acted as decoy animals (Rus. manshchiki ) for “wild” deer during hunting. In the tundra, reindeer can supply people with food, clothing, and materials for nomadic dwellings, and, moreover, every year transported nomadic families and their belongings for many hundreds and even thousands of kilometers. In the taiga, reindeer made the same contributions, but to a smaller degree. In the tundra, reindeer herding could be the primary and even the sole human occupation, but in the forest, herding was always combined with other activities, most often hunting, which as a rule was the main source of income. By the early twentieth century, tundra reindeer herding was already heavily involved in trade relations. Thus, the average herd size in a reindeer Komi Izhemtsy herder household in the tundra documented by the Polar Census was 755.7 animals and in western Chukotka 622.8 head. In contrast, in the taiga people bred reindeer primarily as transport animals, rather than as sources of meat, and the average size of a herd usually did not exceed 50–70 head. Partially as a result of these radically different herd sizes, reindeer-human daily and long-term interaction also differed between the tundra and taiga. In the forest, the process of domestication had an individual character: humans closely worked with each calf, starting from the day it was born. To do this, in some areas of the taiga calves are tethered with ropes immediately after their birth (Davydov 2014). This close daily deer-human socialization meant that reindeer came to know Human-dog-reindeer communities in Siberia 269 humans relatively well and showed little fear of them. Such taiga reindeer can be left for “free-grazing” for a few months without human surveillance with little fear that they will become “wild” prior to rejoining humans later in the year. In the tundra, where interactions with reindeer are far less direct and intensive, herders have to constantly monitor their herds. To do this, people take turns keeping watch over a herd around the clock (this is typical for Nenets and Komi-Izhemtsy) or at certain intervals, perhaps once a day or every 2–3 days, when the whole herd is collected together. In a sense, many tundra deer are managed as a single unit, while only sled reindeer are taught and trained individually. The places and architectures of interacting with reindeer in the taiga differ significantly from those used in the tundra. For example, it was common in the taiga to use smudges, or smoky fires, to protect reindeer from insects. In the warm season, some people built special sheds consisting of poles and branches to provide shade for reindeer, since reindeer are very sensitive to overheating. Gathering reindeer near a smudge serves as a daily means for people to check on their presence and condition, but this is not a one-sided affair – the reindeer benefit from this, too, acquiring shelter from the sun and insects. In the taiga, people traditionally also used fences to keep reindeer, but not all year round: they did this only for a few months. Fawns are pets both among tundra and taiga reindeer herders, and can sometimes enter human dwellings. Such pets are most often deer that were left without their mothers to nurse them. They eat bread from human hands and generally do not live near the herd, but rather stay near people. After they have matured and can feed wholly on their own, they continue to approach dwellings, seeking out human food. Reindeer dietary needs also were intertwined with human activities. Salt was an important element of interacting with reindeer that gathered near smudges. Rein- deer require some salt in their diet, but it is difficult for them to procure it on their own, and the deer in effect use people to obtain it for them. Salt is poured on large stones or roots of trees, put in wooden troughs made from logs, or even provided to them directly from the hand. Reindeer consumption of this salt keeps them in close proximity to humans (and sometimes also dogs), helping to maintain their familiarity. Over the past decades, domestic reindeer have unwittingly undertaken a new and important role in Russia – a political one. These animals have become an ethnic marker and a cultural symbol of indigenous peoples of Siberia, and this role in ethnic politics is rapidly becoming increasingly important. Images of reindeer can be seen in the crests of several northern regions of Russia on the arms of the Murmansk oblast’, Yamalo-Nenets Autonomous District, as well as on several dozen arms of municipalities (districts, cities) of northern Russia. In this regard, reindeer herding in Russia is referred to as the “ethno-saving” branch of the economy, as the nomadic way of life of reindeer herders is said to allow them to maintain some cultural distance from the rest of society, helping to avoid assimilation. Thus, as a result of the mutual adaptation of humans and reindeer to each other, thousands of reindeer herder families live almost all year round with their animals and do not have direct contact with other human groups. 270 Konstantin Klokov and Vladimir Davydov Dogs in HDR communities performed a variety of functions. They pulled sleds, sometimes carried loads on their backs, helped people manage reindeer herds, and participated in various types of hunting. Among the latter, the most important were: hunting fur-bearing animals (sable, squirrel, etc.), large animals (ungulates and large predators), upland game, waterfowl, and seals on the ice. Dogs developed special skills while undertaking these activities, which they learned by working in concert with humans, other dogs, and reindeer. “Universal” dogs that simultaneously acquired several such skills were rare. Sometimes dogs and people also pulled loaded sleds together, dogs and people slept next to one another to share each other’s body heat. In some cases, dogs’ bodies were used as sources of food and their fat as a medicine for lung diseases. However, the needs and capabilities of dogs impose serious limitations on the mobility of people. For example, for a trip on a dog sled, one needed to consider beforehand how to provide food (fish or meat) for the animals, which exert significant amounts of energy during such trips. When hunting for sable or squirrel in the fall (which is the most productive period for hunting these animals), dogs must run through the snow, which tends to become progressively deeper as the season unfolds. When the snow becomes too deep for dogs, a hunter either completely abandons hunting and returns home, or continues hunting using traps. Living together with humans, dogs instinctively perform the signal-guard function, warning of approaching bears, other predators, and strangers, as well as driving predators away. At the same time dogs use humans for protection and defense from the same predators. In the taiga and tundra, dogs are not just watchdogs who guard dwellings; they usually also accompany everyone who goes hunting, fishing, and picking mushrooms or berries outside the village, and warn them of danger by barking. In recent years, their importance in such roles has increased along with the growing number of brown bears, which now represent a more common threat to residents across the Siberian taiga and even tundra.

Reciprocal learning The relationships between dogs and deer are especially interesting and highly variable. Dogs from settlements, as well as hunting dogs, which are unaccustomed to reindeer, can injure reindeer. Dogs living together with herders pass a rigorous selection process with regard to their loyalty to reindeer. People severely punish dogs which show aggression to reindeer, and if this is not enough, kill them. The classical reindeer herding laika is a dog of small size. According to Nenets, a small dog is convenient for the safety of calves and because it is easier to transport it on sleds. One might assume that the training of reindeer herding laikas requires a significant number of special techniques. However, Nenets reindeer herders usually do not agree with such statements, but rather emphasize the innate ability of the Nenets laikas and their ability to learn from older dogs. They do not teach dogs specially: a dog either is born with certain abilities, and in part improves upon them or learns them from older dogs. Thus, Nenets reindeer herders said: Human-dog-reindeer communities in Siberia 271 Nenets reindeer herding dogs are able to pasture reindeer as if they got this from God. Russian dogs, as well as the Russian people – they do not know how to do this. The older dog teaches the younger. Even a bad dog is better than nothing, as reindeer see a dog and gather themselves in a group. (Klokov – Yamal Peninsula field notes)

These insights of reindeer herders are confirmed by observations of ethologists, who argue that all highly skilled herding dogs emerge as a result of intrinsic characteristics and learning (Baskin 2009: 152–153; Coppinger and Feinstein 2015). This is in part learned from watching other dogs, and is to some extent taught by the herders, whose initial task is to prevent a puppy from driving deer aimlessly. The main way in which herding dogs engage with reindeer is to push deer that have drifted from the main group back towards the herd. This task the dog performs itself. This performance is facilitated by the fact that the frightened reindeer in most cases turns toward, and merges into, the herd, regardless of which way a dog initially chases them. As soon as a reindeer herder is sure that the reindeer have turned and moved to the herd, he calls the dog back to him. In addition, the dog, just like a wolf beginning to prey on reindeer, usually does not break into the herd, which now presents itself as a dangerous mass of swiftly moving feet and antlers, but rather rushes along its edges, making the reindeer stay close to each other. This is exactly what reindeer herders need ( Baskin 2009 : 152–153). All reindeer herders emphasize that people do not play a leading role in the training of a young dog, but rather that the young dog primarily learns by mimicking the behavior of older, more experienced dogs. The same opinion exists among taiga hunters concerning the training of hunting dogs. However, if he does not possess an experienced dog, a hunter trains the young dog himself, performing the actions usually carried out by a more experienced dog. For example, he can find a sable in a tree in order to show it to the young dog, or he can run along with the dog following the tracks of a sable. Despite the fact that the dogs teach each other, reindeer herders believe that the working qualities of a dog also depend on his or her master: “There are people who cannot train the dogs, and the dogs work well with some others. If a person has a dog which worked badly, bit reindeer, a new dog will not work well, and vice versa” (Klokov – Yamal Peninsula field notes). The principle of “self-teaching” in HDR communities applies not only to dogs, but also includes reindeer. Thus, when answering questions about the teaching of reindeer, hunters and reindeer herders in Tofalaria often deliberately answered simply: “learning to use the pack is easy – laid them, tied and that is all” or “saddled and rode” (Klokov – Tofаlaria field notes). Through more detailed and persistent interviews, it became clear that in fact the teaching of a reindeer occurs through efforts to ensure that it is docile or “tame”. The main thing is to ensure that reindeer are accustomed to being on a leash, carrying cargo or riders on their backs, and behaving calmly, without jumping or attempting to escape or loosen their cargo. It is also important that taiga reindeer are not afraid of hunting dogs. The main method of such training is recurrently placing both animals in physical proximity to one another. This is accomplished by tying the deer and dog together 272 Konstantin Klokov and Vladimir Davydov in various different ways when they are not working. It should be noted that the training and upbringing of children in the families of the nomadic peoples of the north is based on similar principles of imitation or mimesis. Children are not forced to learn anything; they learn alone by imitating adults ( Wulf 2001 ), and this imitation plays an important role in human-animal relations. The connection between people, reindeer, and dogs is not only material, but also emotional and spiritual. Thus, among reindeer herders, ethnographers note a special type of perception of the surrounding world, time and space called “reindeer thinking” – the ability of people to perceive and to observe the world from this animal’s perspective ( Golovnev et al . 2015 : 16–17). Evenki hunters believe that there is a special relationship between dogs and humans. It is even possible communicate with dogs in dreams, and dogs are able to tell their master where game may be found ( Brandišauskas 2017 : 201).

Conclusion The relationships between humans and animals in HDR communities can be conceptualized as variable forms of asymmetrical interdependence. On the one hand, people to a certain extent act as masters and organizers of human-animal cooperative activities. On the other hand, humans have to synchronize their daily and seasonal rhythms with the needs of their animals, upon which they are heavily reliant. In most northern Russian indigenous communities, the relationships between humans and reindeer are paramount. For example, the needs of reindeer heavily influence the selection of seasonal campsites and nomadic routes, as well as herders’ daily sleep and rest schedules. Large numbers of either reindeer or dogs cannot be economically fed everywhere across the North. Large reindeer herds need vast pastures rich in reindeer moss and with a chilly climate and the possibility of migration over long distances. This combination of ecological conditions can be observed mainly in the tundra, and the Polar Census data clearly show that larger herds were found in the tundra than in the taiga. Relatively small numbers of dogs, perhaps only 2–3 individuals, are needed for hunting or for controlling even a large herd of reindeer. Larger numbers of dogs are needed in settings where these animals are the primary means of transport. One needs an average of 5–10 strong and obedient dogs for one sled. People need fish or meat year-round to feed their dog teams, although they mostly work only in winter. In winter, each dog needs around 1.5–2 kg of fish or meat per day, while in summer they eat far less. To feed one team consisting of 10 sled dogs for one year, one needs 3–4 tons of fish ( Chikachev 2004 : 18). Such quantities of protein may only be provided by households of fishers and marine hunters. A contributing factor is that the sea often throws ashore corpses of walruses and whales, whose meat can help to feed a dog team for a few months. Thus, the symbiosis of humans, reindeer, and dogs in different types of landscapes was based on the use of different types of biological productivity provided by natural ecosystems. In the mainland tundra, the main component of the HDR systems was large herds of reindeer, consuming plant resources – forage pasture plants. Humans and pasture dogs (where they could be found) depended on reindeer as the main source of food, and, in the conditions of the market, as income. Human-dog-reindeer communities in Siberia 273 In coastal landscapes, the dog was the main companion of humans, who used it to harvest biological resources of the aquatic ecosystems. Only in some areas, for example, in the lower reaches of the Yenisei River (Klokov 2000) and in the delta of the Ob River (Klolov – Yamal Peninsula field notes, 2013), sled dogs were partially replaced by reindeer, which were used to transport fish caught for sale. It is important to mention that at these places reindeer were often fed on fish, which was also eaten by people and dogs. The combination of these two types of HDR communities made it possible to harvest the maximum of bio-resources in the continental and aquatic landscapes of the Arctic zone as a whole. In the Subarctic (in the taiga zone), the symbiosis of humans, reindeer, and dogs had a different structure. Here it was aimed at collecting the maximum result from hunting game animals: fur-bearing animals and wild ungulates. Humans achieved this goal jointly with a riding reindeer and a hunting dog. Thus, each of these three main groups of HDR systems was focused on the use of three different types of biological productivity of natural landscapes: plant resources in the tundra and forest-tundra, fish and marine mammals in coastal landscapes, and wild animals in the taiga. The resilience of such systems can be explained from the standpoint of environmental synergy. Reindeer can be considered as the central component of the HDR community, since it is the only one of the three who can live in the tundra and taiga without the two other members. First of all, the role of humans is to stabilize such systems. In natural conditions, three main factors regulate the number of wild reindeer populations: predators (most often, the wolf), diseases, and feed base. Herders can at times weaken the effect of these destabilizing factors by regulating the number of reindeer in their herds by increasing or decreasing the number of slaughtered animals. Therefore, catastrophic fluctuations in the numbers of wild reindeer populations occur frequently but rarely in the case of domestic reindeer (Syroechkovskii 1986: 150–156; Baskin 2009: 24–63). The reindeer herders’ dog, unlike the wolf, affects the number of reindeer indirectly, acting as an assistant to a human, i.е. increasing his ability to maintain the homeostasis of an HDR system. A riding dog, acting partly as a competitor of a riding reindeer, also increases homeostasis, since a reindeer and a dog, if necessary, can partly replace each other. To be more specific, the duplication of systemic links increases the stability of the system. By helping humans, a hunting dog plays the role of an amplifier since it increases the efficiency of the hunter’s work. With the help of their dog and reindeer friends, humans have managed to “tame” the harsh landscapes of the Arctic and Subarctic regions, making them their home. For a man without a dog and a reindeer, the northern environment has been and remains a hostile territory, which can only be conquered with iron, gasoline, and electricity.

Acknowledgments This research was supported by the Russian Science Foundation (project No. 18-18-00309). Fieldwork in 2012-2016 was sponsored by the European Research Council (project ADG 295458 Arctic Domus). The authors are especially grateful to David Anderson, Rob Losey, Peter Loovers, Laura Siragusa, Dmitry Arzyutov, 274 Konstantin Klokov and Vladimir Davydov and Charles Stépanoff for their comments which were especially helpful in improving the manuscript of this chapter.

Sources Field notes: Davydov – Taimyr Peninsula, 2014–2016. Field notes: Klokov – Tofаlaria, 2013, 2014 (Irkutsk region (oblast’ ), Eastern Siberia). Field notes: Klokov – Yamal Peninsula, 2013. Field notes: Klokov – Chukotka, 2016.

References Anderson, D. G. (ed.). 2011. The 1926/27 Soviet Polar Census Expeditions . Oxford and New York: Berghahn Books. Anderson, D. G. et al. 2017. Architectures of domestication: On emplacing human-animal relations in the North. Journal of the Royal Anthropological Institute 23(2), 398–416. Baskin, L. M. 2009. Severnyi olen’. Upravlenie povedeniem i populiatsyiami. Olenevod- stvo. Okhota . Moscow: Tovarishchestvo nauchnykh izdanii KMK. Bogoslovskaia, L. S. (ed.). 2011. Nadezhda – gonka po kraiu zemli. Moscow: Institut Naslediia. Brandišauskas, D. 2017. Leaving Footprints in the Taiga: Luck, Spirits and Ambivalence Among the Siberian Orochen Reindeer Herders and Hunters . New York: Berghahn. Chikachev, A. G. 2004. Ezdovoe sobakovodstvo Iakutii. Iakutsk: IaF GU ‘Izdatel’stvo SO RAN’. Coppinger, R., and M. Feinstein. 2015. How Dogs Work. Chicago: University of Chicago Press. Davydov, V. N. 2014. Coming Back to the Same Places: The Ethnography of Human- Reindeer Relations in the Northern Baikal Region. Journal of Ethnology and Folklor- istics , 8 (2), 7–32. Golovnev, A. V., Ye V. Perevalova, I. V. Abramov, D. A. Kukanov, A. S. Rogova, and S. G. Usenyuk. 2015. Kochevniki Arktiki: tekstovo-vizual’nye miniatiury . Ekaterinburg: Alpha-Print. Klokov, K. B. 2000. Nenets Reindeer Herders on the Lower Yenisei River: Traditional Economy Under Current Conditions and Responses to Economic Change. Polar Research , 19 (1), 39–47. Koz’min, V. A. 2003. Olenevodcheskaia kul’tura narodov Zanadnoi Sibiri. St. Petersburg: Saint-Petersburg State University. Pokhoziaistvennaia perepis’ Pripoliarnogo Severa SSSR 1926/27 goda. Territorial’nye i gruppovye itogi pokhoziaistvennoi perepisi . 1929. Moscow: Statizdat TsSU SSSR. Stépanoff, C. 2012. Human-Animal ‘Joint Commitment’ in a Reindeer Herding System. HAU: Journal of Ethnographic Theory , 2 (2), 287–312. Stépanoff, C. 2017. The Rise of Reindeer Pastoralism in Northern Eurasia: Human and Ani- mal Motivations Entangled. Journal of Royal Anthropological Institute , 23, 376–397. Syroechkovskii, E. E. 1986. Severnyi olen’ . Moscow: Agropromizdat. Vaté, V. 2005. Maintaining Cohesion Through Rituals: Chukchi Herders and Hunters; A People of the Siberian Arctic. Senri Ethnological Studies, Vol. 69. In K. Ikeya and E. Fratkin (Ed.), Pastoralists and Their Neighbors in Asia and Africa. Osaka: National Museum of Ethnology, 45–68. Wulf, K. 2001. Anthropologie der Erziehung. Eine Einführung . Weinheim and Basel: Beltz. 16 Domesticating the machine? (Re)configuring domestication practices in robotic dairy farming

Séverine Lagneaux

“It works on its own; there’s no farmer!” exclaim surprised visitors during an open day on a dairy farm. They hurry to get a look at the hydraulic arm connected to the cow that the farmer is showing them in the robot-operated milking stall. The brochures of the firm that manufactures this robot claim that this is “freedom”. This oscillation in the register of expressions, between worry and fascination, driven by the intrusion of the milking machine, reflects the Western imaginary constructed around the robot. Since its beginnings, robotics has been associated with the myth of a society with no work. Machines, as both friends and slaves, are supposed build a world of leisure for us, profoundly marked by a state of hedonism in which humans have rid themselves of the most laborious tasks. Yet the enslavement of machines and the superhuman powers that they possess constitute the very condition of the destruction of mankind. The freedom permitted by delegating tasks to machines comes hand in hand with the replacement and even obsolescence of humans, as well as animals (Porcher, 2017). The milking machine’s position at the very heart of the age-old domestication relationship ( Vigne, 2012 ) that brought humans and cattle together under the same roof is transforming livestock farming.1 Is it destroying it? Or, on the contrary, might it be an unavoidable technological development required by precision work (Hostiou et al., 2014) and the optimization of the performance of hybrid systems combining the activity of the human and nonhuman actors composing them? Whereas Pol, a Walloon breeder, for example, sees the robot as the “last beast to domesticate”, Marc, another dairy farmer, argues that it implies “talking with your machine”. Human labour is no longer essential for milking. Since robots place teat cups on cows, what does this delegation of tasks imply in the daily practice of livestock farming? By establishing distance between the livestock farmer and animals in the stable, what transformations are there in the daily practice of domesticating dairy cows? Does the installation of a milking robot on a livestock farm (re)configure domestication, or is it nothing more than another technical means added to the already wide array of tools in the modernization of livestock farms? How can we conceive of this additional separation in the relationship with the animal, which Lizet and Ravis- Giordani (1995 ) describe, in the title of their book, as “playing on distance”? To answer this question, I will proceed in three stages:

1 Based on analysis of the content of brochures published by a milking robot manufacturer, referred to as “the Firm” here (it is the manufacturer from 276 Séverine Lagneaux

Figure 16.1 The view from Marc’s office of the entirety of the robotic stable, Belgium, August 2016. Photo credit: Séverine Lagneaux.

which Pol and Marc purchased their equipment, and a leader in this dairy market), I will highlight some of the most salient characteristics of the main actors in robotic livestock farming. I will not dwell on the details of relating the story of the innovation ( Hennion, 2003 ) in the usual way. 2 I will focus in more detail on Marc’s daily practice of robotic livestock farming (Figure 16.1), which is the subject of my ethnographic research started in March 2016, and to the centre of which I would like to transport the reader. Marc is one of Europe’s pioneers in automated milking machines. After acquiring his first robot in 2000, he participated in fine-tuning it. Marc is moreover very knowledgeable on the milking robot studied here. He shares his experience with livestock farmers who have purchased the tool: alongside them, he contemplates the spatial layout of robotic stables, helps them to become familiarized with the software, and helps the herd to get used to this new milking environment. He also trains livestock farming advisers, some of whom present themselves as “robot specialists”. In addition to the two milking robots, this Walloon livestock farm also has a robot that pushes the food and a scraper robot. Marc recently equipped his two stabling systems (one for dairy and another for veal, heifers, and dry animals) with eight cameras directed at points to monitor, such as the calving area or the entrance to robots. 3 Last of all, I will go over, probably too briefly, this issue of the (re)configuration of domestication in this specific hybrid community.

A configuration on paper: the business story of a new partner in livestock farming The use of milking robots has been spreading to more and more farms since their appearance on the European market during the 1990s. The number of farms equipped with this device doubled in 2010. V. Corbet (2010) estimates that by Domesticating the machine? 277 then 5,000 farms across the globe had this milking machine, 80% of which were in Northwestern Europe.2 While according to the five main equipment manufacturers that share this market, milking robot sales3 are currently declining, their numbers within single farms are increasing. This market is therefore clearly not saturated; it remains buoyant, with an average growth of around 10–20%, according to a robotics manager at the Firm.4 In addition to automats, N. Hostiou et al. (2014 ) also point to growth in the number of electronic sensors associated with information processing software in northern European cowsheds. It is therefore appropriate today to speak of “precision livestock farming”. According to D. Berckmans (2004), this consists in monitoring animals and automatically managing their farming by using technology that combines advances in ICTs as well as mathematical and biological models. The milking robot discussed in this study contains both an automaton composed of a cage which adjusts to the size of each cow that it contains, and which distrib- utes an additional amount of food at an adaptable speed, in order to keep the cow calm. At the same time, a hydraulic arm moves across the ruminant’s abdomen to clean its teats and place the teat cups, and to pick up and clean these teat cup liners once that section of the udder has been emptied. To bring the cow in for milking, a circulation system is set up. It can be open or guided, depending on the architecture of the cowshed and the farmer’s choice. In the guided system, “smart” doors direct the animals, based on a “milking authorization” code programmed by the farmer. Two types of directed systems exist. One prioritizes milking, whereas the other focuses mainly on feeding animals. In addition to this labyrinth of tubes, we find massage brushes, drinking troughs, food distributors, and automatic scrapers, as well as different heat and mobility sensors, for example. A circuit of lacto-ducts sends the milk from the robot to the tank via a cooler and a milk analysis laboratory (the progesterone level allows farmers to detect heats that are not yet visible, a cell counter allows to them ensure the good health of each animal, etc.). As the Firm’s brochures explain, the robot removes the twice-daily constraint of milking and thus allows livestock farmers to free up more time for activities such as family holidays. In addition, it supposedly results in increased comfort not only for the farmer but also for the animals. Unlike humans, the machine carries out milking in a routine and constant way: no fatigue, no conflicts, no mood swings, tiredness, or imprecision. The cows are said to enjoy the mechanical regularity of the hydraulic arm that places cups on teats through the triangu- lation of the lasers contained in the camera attached to the arm. They are also supposedly granted more natural expression through individualized milking said to be “voluntary”, and which is no longer managed on the level of the herd. As a result, each animal supposedly chooses to go for milking of its own will. Taking advantage of this decision-making freedom results in an increase in the amount of milk per cow in 24 hours: from 2 to 2.8 litres on average. “Voluntary participation”, the term used by robot manufacturers and which is understood here as the set of means and incentives used to get cows to go to the milking robot, is claimed to have a direct impact on the cows’ performance. A self-regulated “voluntary” or willing cow lives at its own pace and produces better quality and larger quantities 278 Séverine Lagneaux of milk. It is also in better health. Moreover, there is another advantage: such an animal, which is “free to fully express its milking potential”, will allow livestock farmers to focus their attention on working on the problematic animals indicated by the system. The information collected and distributed by the sensors installed throughout the cowshed, including on the bodies of animals, enables the livestock farmer to focus on herd management. In addition to the milking that it carries out for livestock farmers, the machine also provides a set of data in the form of graphs and tables designed to “advise” farmers in the daily management of their herds. In particular, the machine alerts5 farmers when the measurements for an individual animal do not coincide with the modelled numbers and requirements pre- programmed in the system. Supporting the decision making of the “data manager” through advice based on the rigorousness and accuracy provided by statistics and measurements, the robot is a sort of “partner” for farmers. It appears to effectively cause a disconnection between livestock farmers and their herd, instead connecting farmers, through the screen, to individual problems converted into data. Under these circumstances, is it not the machine that is now “domesticating” the living beings that “circulate” within it, submitting them to scheduling recommended by and for the system? Answering yes to this question would mean completely dis- regarding the adaptations carried out daily by farmers in their cowsheds, such as Marc’s practices analyzed below.

Practical configuration Marc emphasizes the contrast between the promises made at the time of the sale of the machine and its daily use in robotic livestock farming:6

You adapt to the robot and to your facilities, but the robot won’t adapt to you. It will milk whatever comes its way, nothing more. They went too far in promising that with the barriers, all of the cows would go to get milked, that it would happen entirely on its own, and that we wouldn’t have to do anything else. Freedom! As if the machine replaced livestock farmers. Whenever you put up a barrier, pay close attention to what might happen behind it. We have to start with the cow, as we don’t get anywhere by forcing animals, whereas technicians and salespeople base themselves on numbers. But even when these numbers are correct, they can’t replace observation.

Hence, this livestock farmer sees many problems arising from robotic livestock farming, for multiple reasons. First, he points out that the animals have been forgotten, claiming that certain farmers focus only on dairy yields and their cutting- edge technology, and no longer observe their cows. He moreover associates a lack of in-depth knowledge of the technique with two forms of misuse: either excess trust in the machine causes farmers to entrust more than just milking to it, namely their herd; or a lack of trust in the robot’s milking ability prevents farmers from making the most of its advantages, thus limiting potential gains for him and his animals. According to Marc, it is necessary for farmers to “step into the cow’s Domesticating the machine? 279 skin”, to consider the machine as well as their actions with regard to it, in order to adapt it to their animals. To do so, he asks the cow why it behaves in a certain way. To answer this, the farmer must “talk” with the machine. I propose to analyze these two points: what do “stepping into the cow’s skin” and “talking with the machine” mean?

In the cow’s skin “The question that I often ask myself is: ‘how does the cow think?’” Marc explains to me. To do so, he attempts to adopt the point of view of a generic cow in a particular situation: “If I were a cow, what would I be willing to do or not? What would I accept or not accept? What constraints would I tolerate, and what wouldn’t I?” This ability to shift his stance supposedly allows him to enter the world of cows:7 it is a “world” that is specific to the cows, especially because, as the farmer points out, the animal’s body is completely oriented towards the ground; they therefore conceive of their shared environment in a particular way. This shift is described as being essential in robotic livestock farming. It is acquired through reciprocal and asymmetric immersion in both the animals and the machine, through their shared life under the same roof (which clearly evokes the domus of domestication). In fact, the farmer’s shift is twofold. The first shift pertains to empathy with his animals. In this respect, Marc explains and regularly insists on his belief that he shares their feelings. He remembers situations that he experienced personally, his impressions at the time, and constraints that resulted in his discomfort. When the waiting area is occupied by a number of animals that he considers to be excessive, Marc remembers, for example, the oppression felt on a crowded bus, which despite being full, continued to fill up instead of setting off: “If I feel that the animals are being jostled around, restricted, or cornered, that makes me suffer in a certain sense. It’s hard for me to stand it for long”. For example, observing what is taking place from the patio that overhangs the stable, Marc points to cow number 4525. She remains still in front of the door of the most recent robot, while the clicking of the barrier has indicated that the spot is vacant.

If there’s a delay in the milking, the cow that’s there isn’t going to go in any quicker than normal. Some might go in quicker, but not her. She will be milked, but she isn’t going to worry about what happens after her. There isn’t necessarily solidarity among the herd, where they are going to say, “Every- body’s in a hurry, so let’s hurry up”.

Other recent events nevertheless make Marc believe that his cows demonstrated their cohesion as a herd and their masterful use of the circulation system. When intervening in the waiting area on a primiparous cow that had been standing there for what he deemed to be too long for her well-being, Marc used an electric prod to chase the animal towards the robotic stall. He had first attempted in vain to stimulate her as usual by making animal noises and the associated gestures, or 280 Séverine Lagneaux using imitation between fellow animals. As she received the shock, the young cow mooed in a very specific way. Following what Marc describes as a “call out”, he observed a movement of the herd within the stable. Rather than attempting to flee, the cows moved in a way that revealed their ability to use the different doors of the circuit in order to move closer to the waiting area and the frightened animal. Marc observed their knowledge of the devices and the facilities as well as their concern for a new member of the herd, which machine measurements subsequently confirmed.

It was as if, at a given moment, there was a sort of solidarity with the animal that seemed to be using a suffering tone. It might even have been a cry for help. In any case, it really did get a response. I could see it, because everything is more or less divided into squares at my farm, so you can see the movements that take place following the paths. I saw animals that did a full loop around the circuit to get to the place where it was happening. These are things that I don’t want to call surprising, but it’s still astonishing to see this sort of empathy, as I could call it, although for certain things, they don’t have any empathy at all. Maybe when in danger, they do have empathy, whereas not when it comes to our preoccupations, our need to always gain time and make things go quicker.

The second shift that the livestock farmer carries out, is an imagined projection into the body of moving cows. Based on a blueprint of the stable system designed by a livestock farmer asking for his help, or alternatively to improve the spaces rarely frequented within his own stable, Marc imagines that he is a cow following a path. In addition to comparing the events experienced, he also tries, as far as possible, not to think like a human being. In reality, the projection does not consist of a superimposition: Marc does not attempt to become a cow.

I try to ignore who I am, but in the end, in terms of some of our behaviours, cows operate like us. We don’t like to go into places that won’t be very pleasant for us; we don’t like to wait. And sometimes I think, “It’s funny; I would be in more of a hurry than her”, but then I tell myself, “Of course I would: she has cow time. She has nothing else to do. Our concept of time is different”.

This shift is not an attempt to put himself in the other’s position, but rather to partially understand a different experience undergone out of choice and feelings, and not only in reaction to a stimulus. This (dis)position leads Marc to accept the idea that his cows voluntarily contribute to milk production, not only because they resist in certain situations, as J. Porcher and T. Schmitt (2012 ) have shown, and without denying that the system also forces them to “give” their milk. Marc moreover emphasizes what A.G. Haudricourt (1962 ) has already highlighted: “I really do think that the way I behave with animals can be transposed to human relations”. Marc can adopt the point of view of a generic cow or a specific individual. He thus gives me an explanation of why cows do not want to go into the milking stall: Domesticating the machine? 281 For cow 4525 over there, milking is painful, as she has a longer udder. For another one who has just calved, she has oedema, so it hurts. Others get bored. So, it’s not a place that they’re keen to get to. There are also cows who are further along in lactation. They no longer give a lot of milk. They’re often force-fed, because it’s a single ration. The simplest thing to do is [he says, assuming the point of view of an imaginary cow]: “I’m not giving any more. I’m lying down. I’m eating”.

Marc will stimulate his cows to get them to go to the robot, without the necessary adaption having an impact on the other cows in the herd. He claims that he therefore always has to think about both the individual and the group. He will, for example, momentarily suspend the cleaning of the rear left teat of 4525, who repeatedly raises her leg during this stage of the milking. He will adapt the speed of the distribution of “treats” to cows at the time of milking so that “impatient” cows will calm down throughout the operation, or in order to spark their curiosity in the machine by taking advantage of their “gluttony”. For animals at the end of lactation, he will milk them more quickly or will stimulate them by modifying the initial programming of the distributor to complement the robot. He tries to limit constraints from having a negative effect on animals in order to maintain their curiosity in the robot and their desire to be milked. Marc is attentive to not reinforcing certain behaviours among his animals. He aims not to directly modify their habits but rather to act indirectly via the machine. To do so, he needs to speak the language of his machine.

Speaking machine language With the expression “speaking the machine’s language”, Marc is referring to a dual process. First, it is necessary to be familiar with the computer program in order to extract the desired information and to understand what it covers, and thereby to understand how it measures animal behaviour. This necessary learning feeds the farmer’s trust in the tool, to which he delegates milking, as he under- stands how it works. He can let it do the milking while he ensures the smooth function of the process, albeit furtively by listening to the regular noise of the pumps, for example. Afterwards, once this language has been acquired, it is necessary to be able to use it to adequately program the actions carried out by the farmer or another party involved (nutritionist or veterinarian, for example). This daily programming enriches the data and enables an adjustment in the livestock farming schedule, as well as the adaptation of the machine to the animals. This consists not only of formatting the system, but also circumventing or tricking it. Sitting in front of his computer (Figure 16.2), Marc scrolls through the lactation graphs for some of his cows to show me how the software presents a “drop in lactation”, and the way in which livestock farmers must act faced with this problem. This sudden drop in the milk production curve for each cow is, in this case, shared by a dozen primiparous animals. Initially the farmer thought that this drop was because another animal on which he had placed a nose ring was suckling on 282 Séverine Lagneaux

Figure 16.2 Marc at his office computer, which shows a lactation curve in decline, Belgium, July 2017. Photo credit: Séverine Lagneaux. the teats. However, the curves did not increase, the number of animals afflicted remained stable, and when Marc observed diarrhoea among these dairy cows despite them maintaining a good appetite, he asked the veterinarian to take a blood sample. These animals were infected by a small “mollusc” similar to a worm, but which instead of establishing itself in the liver, infected the intestines. He tested a treatment, but also had to address the machine. The difficulty of managing these animals is a result of the robot’s specific way of classifying a decrease in lactation, which it likens to “incomplete milking”. Hence, as the system does not classify a cow whose udders are empty within the category of milked cows, she will be sent back through the smart doors, to circulate within the stable towards the waiting area. She will once again have to pass through the robot, which will place the teat cups on her udders, even though they are empty, and will be required to undergo the constraints of the machine. Moreover, as per Marc’s programming, animals that have not been milked completely are sent back to the cowshed and do not have access to the grazing areas. The inadequate identification of a difference in production statistics (between what is expected, that is, what is recorded by the machine based on an animal’s previous yields, and the quantity of milk effectively given) results in discomfort and the failure of the system to adapt to the animals in question. As their lactation does not comply with the required data, the robot will distribute less additional food to these “deficient” animals. Whereas they should benefit from an additional ration to increase their strength, the machine plans a completely different type of treatment by linking different measurements: less production, less food dispensed. Financial profitability prevails in the “original” Domesticating the machine? 283 process of the machine. The thermodynamic animal converted into data is not the cow that Marc has inside him. To get around the system, the farmer must give these specific cows an “exit authorization”. He has to deactivate this category of cows – “cows directly outside” – once the parasite is under control.

(Re)configuration of domestication? It is not rare for robots to “precipitate men and animals into a hole”, as Marc says: the former are cornered both economically and emotionally, whereas the latter are perishing all but physically. Marc observes the dysfunction and harmful impact of this through new expressions circulating in the world of robotic milking, where “cows are thrown out” (“[on] jette les vaches ”) and moreover “go wild” (“s’ensauvagent ”). A farmer “who loves his animals” (“l’amoreux de ses bêtes”) therefore steps aside in favour of the “milker”. In many ways, technology makes domestication in livestock farming more fragile, by establishing a distance which, when ignored, results in a rupture. Let us take note of two interlinking elements contained in the mentioned expressions: the productivism of the milker, and the manufacturing of cows. We could initially reflect on the “wild” nature of the undomesticated machine. In fact, the farmer insists on the fact that he does not want to be dependent on this autonomous machine and its designers. Because of this, he must master the tool or risk being subject to it himself. This association between domestication and mastery is classic. In 1989, J. Clutton-Brock defined domestication by insisting on the control exerted by humans and the conversion of animals into objects of ownership (p. 7). According to her, this practice consists in controlling vital processes in order to exploit them through the implementation of technical means, to derive benefits from them. Productivism and profitability are the end goals, and the biological transformation of domesticated animals is the result. According to the configuration of robotic stables on paper, livestock farmers follow recom- mendations made by the software, which manages the livestock farming schedule in relation to statistics on animals. The relationship embodied by the livestock farmer and his cows is suspended through the metric precision and financial purpose contained within robotic facilities. Even more so nowadays, through the connection of multiple cowsheds to its network, the Firm is collecting a large quantity of data allowing it to re-assess its models. Through an iterative process claiming to increase the profitability of the livestock farmer’s personal operation, the farmer is participating in shaping an ideal animal manufactured according to mean statistics collected for every animal passing through every robot at every robotic stable across the globe. Is this not a way of putting additional pressure on livestock farmers and their animals? On a macro-economic level invisible on the scale of a single farm, is this not a case of constraining livestock farmers to a somewhat greater extent and reconfiguring domestication by completely elimi- nating the bond underpinning the practice of livestock farming, for the purpose of maximizing control? Does this not consist in manufacturing an ideal animal whose algorithm is calculated based on the average of the statistics collected? 284 Séverine Lagneaux This calculated cow would be perfectly adapted to the machine, while the farmer would be relegated to the ranks of a labourer, applying models designed outside of his or her cowshed, with the aim of maximizing the amount of milk produced. Under the pretext of defending livestock farmers’ wallets, is it not the profitability of the global economic model based on technologization and acceleration ( Rosa, 2012) that is being consolidated, all the while remaining well out of reach within a black box? Nonetheless, Clutton-Brock’s reference to ownership, beyond the idea of “property”, can also refer to the process of “establishing belonging” through learning, and namely learning a “language”, as Marc put it. The central focus of J. Porcher’s work, the relationship at the heart of livestock farming, is also recognized as the basis of domestication. While N. Vialles (2004 : 287) prefers the term “asocial sociability”, the bonds between livestock farmers and animals could be considered the main element of the continuous domestication process ( Digard, 1990 ). D. Haraway (2003) also insists on the profound pleasure of sharing life with different beings, while S. Budiansky, in the subtitle of his book The Covenant of the Wild: Why Animals Chose Domestication (1999), insists on the co-evolution of humans and animals. Last of all, J. Porcher (2002 ) and V. Despret (2012 : 216) insist on the “invisible work” of animals caught within human activities. Domestication is also a relationship, a familiarization which, despite being asymmetrical, proves to be reciprocal. According to Marc, talking to the machine requires an understanding of the data provided by the machine and of the framework of rules presiding over this quantified agencement , so that livestock farming is not reduced to the application of orders and plans managed by the robot. It is therefore the duty of livestock farmers to immerse themselves in the technical process in order to adapt it to living things:

When I walk past my robots, in thirty seconds, I’ve already taken it all in. I’ve listened; I’ve watched; it’s automatic. I never walk past the machine without listening to whether the cattle cake is falling, if there are any air leaks, etc., and all of that takes a minimum amount of time. I know the program: there’s a data package, lists. . . . I feel that I’m more intuitive. That’s how I run my ship. Immediately, without checking the lists, I’m able to say that things are going well over there but not over there. I feel it. I can’t explain it. That’s why it’s difficult to replace yourself. It becomes a bit magical to understand things in the end. Fully immersed.

According to Clark (2007: 62), the development of this ability to “read the other” is the driver of domestication and the mutual acknowledgement that it implies and requires. Learning the language of the machine precedes the immersion that then allows Marc to take a step back from numerical data and concentrate on dairy cows and the questions that they raise through their behaviour. Marc’s observations lead him to reﬁne his knowledge of the world of cows, as well as their individual behaviour and personalities, in such a way that he is able to put himself in the place of the other as an individual, a member of a group (herd), and/or a species. Domesticating the machine? 285 The machine allows him to increase the time he spends observing his cows. It also provides him with data that conﬁrms his intuitions, compliments them, and gives nuance to them. Under no circumstances do these measurements invalidate the farmer’s observations. His knowledge through immersion is expanding and not obsolete. It is through his ability to observe and shift his stance, and the acknowledgement of the ability of cows to ask him questions, along with the subjectivity of animals not reduced to a living mechanism, that Marc can circumvent the machine to maintain the balance of his animal-centred system. Does this make him, to some extent, a sort of hacker of this robotic system? Domestication is thus not reconfigured by the robot: it combines control, familiarization, attachment, and detachment. It does not consist of a single “blueprint”, a series of actions resolving a problem ( Suchman, 1985 ), as the brochures claim when they present the milking robot as an intermediary and interface between cows and the livestock farmer, who are “freed” from one another. The logic of livestock farming practices is shifting despite everything. According to Marc, is necessary to anticipate:

You can hear people incorrectly say that farmers with milking robots no longer know their cows, whereas in reality, they know them very well and individually. We no longer manipulate herds but rather individuals. In a classic milking room, you milk a herd for an hour. It’s an intensive job and you don’t stop at a single animal. Here, we have all the time available to understand what’s happening for the cow in the stall. Lots of information is transmitted by the computer, but I think that if you don’t have a livestock farmer’s eye, you can miss a lot of things with a milking robot. If you forget to watch the animals, you will make mistakes. I tend to trust what I see and then go to the machine afterwards to confirm. You can free up time for observation. It’s obvious that you notice lots of things, but for that happen, you have to want to do it, and to have that affinity.

Even though the Firm’s salespeople will insist on the precedence of numerical data over the “feelings” of livestock farmers, which are reduced to vague impressions, farmers also highlight the shift of the livestock farming logic towards anticipation. This shift, which calls for in-depth analysis, appears to be moving toward the planning of actions; yet, as Marc’s livestock farm shows, this anticipation requires perception and immersion. The detachment associated with numbers or the common idea of replacing humans with machines is counterbalanced. Despite its negative publicity ( Candea et al., 2015 ), this detachment cannot be considered a clear-cut separation, a harmful distance.8 It requires attachments in multiple forms. The actors of livestock farming are intermediaries: the cow informs the livestock farmer of modifications to program into the system, the system informs the livestock farmer of an intervention to carry out on a given animal, the livestock farmer programs the system to guide animals in a given direction, cows circulate to get milked or to look for an animal that is suffering, etc. The routines are specific to each actor but are also intertwined, as each must adapt and/or be adapted to the 286 Séverine Lagneaux other. Anticipation is not prescription: it consists not in applying rules, but rather in reacting by adapting the system, based on the evidence perceived and the experiences undergone, not only by targeting good production in terms of quantity, quality, profitability, and efficiency, but also by targeting the equilibrium of the system for the harmonious cohabitation of each actor. Each of them acts, is acted upon, and reacts in order to maintain the fragile equilibrium of a system fine-tuned by the livestock farmer with his tools for his animals, highlighting the continuous yet temporary nature of domestication and the hybrid community established.

Acknowledgments I wish to thank the Fyssen Foundation for the grant awarded to me, which enabled me to further my work on milking robots and to apply it within the framework of the “Humanimachine” project.

Notes 1 The domestication of ungulates began in the Near East during the Neolithic, between 12,000 and 10,000 years B.C. Thanks to research combining zoology and genetics, milking and dairy production are no longer considered a secondary form of domestication (see Roffet-Salque et al., this volume). 2 Six to seven percent of French livestock farms have these machines, compared to 40% in Sweden and Norway. 3 Not including building modifications, machine maintenance fees, and additional materials (food distributor, drinking trough, brushes, mats, detectors, tank), the average cost of one of the Firm’s robots is around 150,000 euros. 4 The Institut de l’élevage estimates that 4,806 French farms had at least one robot in 2015. http://idele.fr/rss/publication/idelesolr/recommends/robots-de-traite-le-deploiement- continue.html 5 Alarms to indicate the breakdown or dysfunction of the system itself operate day and night. 6 A filmed extract of Marc’s stables showing the automated milking process, entitled “circulation,” can be found at www.youtube.com/watch?v=gDvdL7cykWc&feature=youtu. be . To film it, we used GoPro cameras attached to the different actors in robotic livestock farming. We aim to demonstrate the interconnection of their specific routines in a longer documentary, the filming of which was enabled by the Fyssen Foundation. This film has been created with the collaboration of Joffrey Becker, an anthropologist specializing in humanoid robotics. 7 This expression brings to mind two concepts created by von Uexküll (1909), of which Marc was unaware. I am referencing the Umwelt , the world external to a subject, which constitutes its environment, and the Innenwelt , the internal world correlated with the subject. 8 For example, according to M-C. Pouchelle (2007 ), patients associate the use of a robot for heart surgery with an additional guarantee of the success of their upcoming heart operation. Surgeons see it as either a cutting-edge technique or as the distancing of actions on the body that are at the heart of their profession.

References Berckmans, D. 2004. Automatic Monitoring of Animal by Precision Livestock Farming, in F. Madec and G. Clement ed., Animal Production in Europe: The Way Forward in a Changing World , Saint-Malo: International Society for Animal Hygiene, 27–30. Domesticating the machine? 287 Budiansky, S. 1999. The Covenant of the Wild: Why Animals Chose Domestication , London: Yale University Press. Candea, M., Cook, J., Trundle, C. and Yarrow, T. 2015. Detachment: Essays on the Limits of Relational Thinking , Manchester: Manchester University Press. Clark, N. 2007. Animal Interface: The Generosity of Domestication, in R. Cassidy and M. Mullin ed., Where the Wilde Things Are Now: Domestication Reconsidered , New York: Berg, 49–71. Clutton-Brock, J. ed. 1989. The Walking Larder: Patterns of Domestication, Pastoralism and Predation , London: Unwin Hyman. Corbet, V. 2010. Installation de traite robotisée, présentation, www.interreg-aocfromages. org/fileadmin/Documents/Action_2/Presentation_Vincent_Corbet.pdf Despret, V. 2012. Que diraient les animaux, si . . . on leur posait les bonnes questions , Paris: La Découverte. Digard, J.-P. 1990. L’homme et les animaux domestiques. Anthropologie d’une passion , Paris: Fayard. Haraway, D. J. 2003. The Companion Species Manifesto: Dogs, People, and Significant Otherness , Chicago: Prickly Paradigm Press. Haudricourt, A.-G. 1962. Domestication des animaux, culture des plantes et traitement d’autrui, L’Homme , 2: 40–50. Hennion, A. 2003. L’innovation comme écriture de l’entreprise. Récits d’innovation au sein d’une entreprise de services, in P. Mustar and H. Penan ed., Encyclopédie de l’innovation , Paris: Economica, 131–152. Hostiou, N., Allain, C., Chauvat, S., Turlot, A., Pineau, C. and Fagon, J. 2014. L’élevage de précision: quelles conséquences pour le travail des éleveurs? Productions Animales , 27: 113–122. Lizet, B. and Ravis-Giordani, G. ed. 1995. Des bêtes et des hommes. Le rapport à l’animal: un jeu sur la distance , Paris: Editions du CTHS. Porcher, J. 2002. Eleveurs et animaux, réinventer le lien , Paris: Presses Universitaires de France. Porcher, J. 2017. Elmo et Paro®, pourquoi l’un travaille et l’autre pas, et ce que ça change, Ecologie & Politique , 54: 17–34. Porcher, J. and Schmitt, T. 2012. Dairy Cows: Workers in the Shadows? Society and Ani- mals , 20: 39–60. Pouchelle, M.-C. 2007. La robotique en chirurgie cardiaque, Communications , 81: 183–200. Rosa, H. 2012. Aliénation et accélération, Vers une théorie critique de la modernité tar- dive , Paris: La découverte. Suchman, L. 1985. Human-machine Reconfigurations: Plans and Situated Actions , Cam- bridge: Cambridge University Press. Vialles, N. 2004. La nostalgie des corps perdus, in F. Héritier and M. Xanthakou ed., Corps et affects , Paris: Odile Jacob, 277–289. Vigne, J.-D. 2012. Les débuts de l’élevage , Paris: Le Pommier. Von Uexküll, J. 1909. Umwelt und Innenwelt der Tiere, Berlin: Verlag von Julius Springer.

17 From parasite to reared insect Humans and mosquitoes in Réunion Island

Sandrine Dupé

Introduction Some species of mosquitoes have developed direct links with humans. Females of Aedes albopictus, especially, have a preference for human blood to mature their eggs and lay in small volumes of clean water. Changes in consumption patterns and the extension of urban areas have significantly multiplied the number of breeding sites close to humans. Added to their taste for human blood, this change in the urban ecosystem has developed new generations of mosquitoes with a “fondness for man” (Galton, 1863: 131), a caracteristic Galton attributes to animals which prefer keeping a link with humans rather than avoiding them if they are neglected. Galton placed this characteristic in second position on the list of conditions for animal domestication. Humans who, unlike mosquitoes, do not benefit from these interactions, have consequently found themselves in a relationship of parasitism. In other words, mosquitoes are bound to humans, their host, but at their hosts’ expense. Beyond the nuisance associated with the itching bites resulting from this parasitic relationship, emerging viruses transmitted by mosquitoes increasingly engage researchers and public health authorities in finding new mosquito vector control strategies at a time when international and national regulations have considerably hampered the use of insecticides. While chemical–based products constituted the quasi–exclusive strategy for vector control in the 20th century, public health issues related to these substances have become so acute that it is necessary to develop new control techniques. As a result, researchers and public health authorities all over the world are experimenting with new approaches. The range of strategies deployed rely on the a priori indefectible character of the relationship that binds vector mosquitoes to humans. The two main strategies are: the rearing and the release of sterile mosquitoes to weaken the reproduction potential of wild ones, and the enlistment of local communities in order to make them break the parasitic “domestic” mosquitoes’ reproduction cycle. These vector control changes are both led and accompanied by a profound transformation of the relationship between humans and mosquitoes ( Dupé, 2015 ). Based on a Réunionese case study, this chapter aims to explore the destabiliza- tion of the relationship between humans and mosquitoes induced by these new forms of control. Since the outbreak of the chikungunya epidemic in 2005–2006, 290 Sandrine Dupé mosquitoes and humans have been experiencing new forms of interaction. Never considered as a threat, and even unnamed by locals before the outbreak, Aedes albopictus suddenly became the target of intense collective vector control, based on the exploration of the possibilities offered by the “fondness for man” (Gal- ton, 1863 ) exhibited by mosquitoes. Could these parasites become the allies of humans through domestication? As Nigel Clark says (2007 : 52), “what a body ‘owes’ to other bodies defies any final reckoning or settling of accounts. Any opening between bodies is always, to some degree, unpredictable”. By divert- ing the privileged link between parasites and their hosts to better distance them, researchers and public policy invite us to explore the boundaries between domestication, parasitism and mutualism. The case study is based on ethnographic fieldwork that ran from 2010 to 2013 in Réunion Island. As an anthropologist involved in a research program on the interest of releasing millions of sterile mosquitoes,1 I could observe as closely as possible the ongoing implementation of a biotechnology. This status facilitated my access to the state’s vector control service. At the same time, I questioned some inhabitants of the island about their relations with mosquitoes. The material and ideal interactions between humans and mosquitoes were thus observed in several areas: in a research insectarium, in gardens, in gullies, in meeting rooms, and so on. The collection of data was mainly based on long–term participating observations, semi–directed and undirected interviews with approximately one hundred persons. To explore the boundaries between domestication, mutualism and parasitism, we will compare several grounded categories of mosquitoes through an analysis grid resulting from Mason’s work on domesticated animals ( 1985 ). This analysis grid will lead to a reflection on the bridges that connect these types of interrelations between living beings.

The domestication as a grid of analysis of the evolution of human–mosquito relations The domestication of harmful, even parasitic, species seems to be almost oxy- moronic. Does it make sense to express interest in it? If, like Micoud (2010 ), we mean domestication as an interdependent relationship between humans and animals, whose balance of power is favorable to humans, it seems counter–intuitive to think of the voluntary inclusion of undesirable living beings in our city. If, like Clark (2007 ), we consider that domestication is no longer a system of domination, but rather a consequence – often uncontrolled – of proximity between humans and living beings, sometimes risky, often unpredictable, sometimes liberatory or costly for the species in question, then we can undertake this reflection. The question of domesticity is debated so much today that the two poles of “domestic” and “wild” appear to be notions emptied of their meaning. However, they still help us to think about the relationships between humans and animals. This is why we will now turn to this exercise, drawing on Mason’s work in 1985. Although his book is dated, it has the virtue of not taking agriculture or pets as priority analytical prisms. His review of the domestication situation is relatively From parasite to reared insect 291 systematic to the effect that it characterizes domestic animals in a fairly broad and inclusive way. Mason (1985 ) uses four main characteristics to define domesticated animals: (1) their breeding is under human control, (2) they provide a product or services useful to humans, (3) they are tame (4) and they have been selected away from the wild type. To refine this definition of domestication, Mason includes some categories of exceptions: he considers animals which are reared but not bred in captivity (such as cormorants, used for fishing), those which are bred in captivity to be released into the wild (such as pheasants, used as game), and castrated animals. Finally, Mason excludes three types of animals from the frame of domestication: laboratory animals (unless they are selected far away from their wild type – which constitutes a subjective category), animals in experimental research (arguing that they are not “truly brought into the house”) (Mason, 1985: 2) and animals caught wild and tamed. He specifies that domestic animals can be considered as domesticated ones, without stating the perimeter of the domus , or the type of animals considered. This vague definition, which refers to anthropolihy, does not allow us to consider this characteristic relevant to a definition of domesticated animals. However, it highlights the vagueness of the boundaries of domesticity, raised by Clark (2007). Can a fly, a mouse or a mosquito, by virtue of their proximity to human dwellings, be considered domestic, as well as cats? This anthropophilia, or “fondness for men”, is a characteristic which may equally refer to domestic animals as well as to commensals or to parasites. As it is not relevant, and to avoid any confusion, this characteristic will not be considered in this analysis. To put Réunionese Aedes albopictus to the proof of this framing of domesticity, four grounded categories of mosquitoes will be taken into consideration. As the redefinition of the links between humans and Aedes albopictus in Réunion island 2 is the work of researchers and policy makers, we will adopt the four categories that emerge from their discourse: “mosquitoes” as an encompassing category, “mosquitoes in the gardens”, “mosquitoes of the insectarium” and “released mosquitoes”, to reflect on the bridges that can link parasitism and domestication.

“Mosquitoes”: an encompassing category The first one comprises Réunionese Aedes albopictus in general, as an encompassing category. These mosquitoes do not necessarily live close to humans. They bite humans if they encounter them, or other animals if they do not encounter humans, and lay their eggs in clean water, collected in holes formed in rocks or trees, in human waste, or gutters. This category includes mosquitoes whose life cycle does not necessarily depend on humans, as well as those living in gardens. Flows of Aedes albopictus can link gullies to houses, as some gullies cross the cites, but the extent of these movements is unknown. Thanks to this general category, it can be seen that the privileged link between humans and mosquitoes, if not univocal, can be structuring, to the extent that mosquitoes are likely to become the parasites of any humans they happen to encounter, whether in the vicinity of places infested by mosquitoes or not. There is no question of domestication here (Table 17.1) , and that of parasitism is debatable, 292 Sandrine Dupé Table 17.1 The domesticity of mosquitoes, as an encompassing category, according to Mason’s categories ( 1985 ).

Mosquitoes in La Réunion

BASIC DEFINITION 1/ Its breeding is under human control − 2/ It provides a product or services useful to humans − 3/ It is tame − 4/ It has been selected away from the wild type − EXCEPTIONS It is reared but not bred in captivity − It is bred in captivity to be released into the wild − Castrated animals − EXCLUDED Laboratory animals (unless they are selected far from the wild type) Animals in experimental research Animals caught wild and tamed

− Characteristic excluding mosquitoes from the Mason definition as mosquitoes do not need humans to live and to reproduce. The painless bites (for people accustomed to them) and the inocuity of mosquitoes whose lives were related to humans made them more commensals than parasites before the chikungunya epidemic – or even momentary allies, in spite of themselves, when their sting, painful for tourists and new arrivals on the island, helped to make their island hostile to “foreigners”, transforming mosquitoes into a “paradoxical figure of the useful pest” ( Claeys–Mekdade, 2002 ). The discovery of its vectorial potential has made the strong link between humans and mosquitos more tangible, and mosquitoes have suddenly become parasites in the discourses and practices of policy makers and, gradually, in the general discourse, even if the threat of this relationship took a long time to be taken seriously. “Mosquitoes do not bite me” has become a slogan that people use both as an assertion and, ironically, as a way to subvert the discourse of the authorities emphasiz- ing the responsibility of the mosquito vector in the spread of chikungunya when other causes were invoked locally to explain the epidemic, and to distinguish themselves from tourists who feared mosquitoes.

“Mosquitoes in the gardens”: a political category to enjoin people to control mosquitoes The second category, which forms part of the first, refers to mosquitoes present in gardens in Réunion Island ( Table 17.2 ). It describes a category of practices more than a named category. Since the chikungunya epidemic in 2005–2006, public From parasite to reared insect 293 Table 17.2 The domesticity of mosquitoes from the gardens, according to Mason’s categories ( 1985 ).

Mosquitoes of the gardens

BASIC DEFINITION 1/ Its breeding is under human control ? 2/ It provides a product or services useful to humans − 3/ It is tame − 4/ It has been selected away from the wild type − EXCEPTIONS It is reared but not bred in captivity ? It is bred in captivity to be released into the wild − Castrated animals −

EXCLUDED Laboratory animals (unless they are selected far from the wild type) Animals in experimental research Animals caught wild and tamed

− Characteristic excluding mosquitoes from the Mason definition ? Paradoxal characteristic of mosquitoes within Mason definition of domesticity authorities have concentrated their efforts on this category of mosquitoes. People leaving saucers under flower pots and other rainwater–collecting objects increase the number of breeding sites. As the radius of movement of mosquitoes during their lifetime is considered restricted, mosquitoes bite more humans when they are born near their houses. Their behaviour is more parasitic than anthropophile. The risk of disease transmission is therefore increased. As the use of insecticides is now heavily regulated, public authorities seize the mosquitoes’ “fondness for man” to break their life cycle, enjoining people to control mosquitoes laid in their own gardens. Local authorities have developed a discourse asserting that Aedes albopictus only reproduce as a result of human domestic practices. The change in the status of mosquitoes – from insignificant to threatening, and from ubiquitous to “domestic” – is a moral lever for local authorities to involve the population in the struggle. Since 2007, state agents have been coming into gardens to teach people how to destroy the eggs laid by female mosquitoes in their gardens, a lesson partially appropriated by the inhabitants. Réunionese people are considered by local authorities as mosquito breeders: “ This house is a major producer of mosquito breeding sites in this zone”, says a supervisor after field action conducted by state agents. “Réunionese people raise mosquitoes”! became a common statement, circulating between civil service offi- cials and entomologists . One researcher even considers the responsibility of the Réunion Islanders in the mosquito breeding process to exceed the domestic space, with the production of anthropic deposits in spaces that he considers natural: “If 294 Sandrine Dupé there were not the anthropic breeding sites, we would remove between 75 and 90% of the mosquitoes, so it is not negligible. It is not necessarily in people’s gardens. But if the Réunion Islanders are breeding mosquitoes? Yes, 100%”! The category “mosquitoes bred by Réunionese people” could be more exact as it is often referred to by researchers and public authorities, but it does not overlap practices, targeted on gardens. In spite of this discourse, “mosquitoes from the garden” does not correspond to Mason’s definition of domesticity (1985 : Figure 17.2) even if some of their characteristics are paradoxical. Is their breeding under human control? But the health policy makers’ discourse is an aesthetic lever ( Foucault, 1984 ; Nading, 2012) which is supposed to moralize the gestures of everyday life. In fact, this “breeding” is not intentional. Forgetfulness, an impossibility to clean one’s garden every week, disinterest and a taste for plants in pots are factors involved in passive breeding. The Réunionese share with mosquitoes a habitat that “changes the microcli- mate experienced by its inhabitants” ( Leach, 2005 : 353): the processes of avoidance and tracking impact the space and behaviour of these two species. People use fewer pots and drain off clean water while mosquitoes lay in new places, in dirtier water. This process of mutual acquaintance is more a relationship of attraction–avoidance more typical of parasitism than domestication. The nature of the relations induced by domestication (the responsibility of humans towards domesticated animals) supports an action–oriented political discourse, but the garden is not where mosquitoes are domesticated in Réunion Island.

“Insectarium mosquitoes”: the first step towards the domestication of mosquitoes The third category, here called “insectarium mosquitoes”, represents mosquitoes used in the research programme on the feasibility of the sterile insect technique to reduce the risk of vector–born epidemics. This technique consists of altering the ability of mass–raised mosquitoes to reproduce, modifying their reproduction competency. Once sterilized, millions of male mosquitoes are released in order to compete with the males already present in the territory. The more females mate with the released males, the more their offspring is non–viable. Releases are made until the population of mosquito vectors is low enough to no longer constitute a public health problem. Only males are sterilized and released, to avoid increasing the population of biting mosquitoes,3 which would mean an increased risk of epidemic . Several processes can make the offspring of modified mosquitoes non–viable. In this programme, the selected method of sterilization was Gamma irradiation.4 In 2010–2013, when the ethnographic research took place, releases were not made in field conditions. Entomologists tried to breed and rear a large number of mosquitoes in an insectarium, to develop mass breeding techniques, carry out experiments to determine the characteristics of the mosquitoes (flight, reproduction, From parasite to reared insect 295 genetic structuring, etc.) and test the advantages of the sterile insect technique in controlled and semi–field conditions. To set up a mass rearing programme and to understand mosquitoes’ behaviour, scientists have to submit to their Umwelt ( von Uexküll, 2010) in order to benefit from their cooperation, at the same time as they impose on mosquitoes new rules of selection and survival in very constrained living conditions – small cages, artificial light, new food – which challenge the bond between mosquitoes and humans. In the insectarium, among the three species of anthropophile mosquitoes raised – Aedes albopictus , Aedes aegypti and Anopheles arabiensis – only Aedes albopictus was able to adapt to this new ecosystem. If the “fondness for man” is a characteristic facilitating domestication (Mason, 1985: 2), it is not the only condition ( Rader, 2007 ). Mosquitoes are invited to cooperate with humans by mating, laying eggs and producing scientific data. In return, scientists are compelled to take care of them and to adopt a part of the mosquitoes’ frame of reference ( Rennesson et al ., 2012 ), although the relationaship is devoid of intersubjectivity: mosquitoes are not bred as individuals, but as cages (a group of mosquitoes destined to experimentation). The rhythm of work at the insectarium is determined by the rhythm of life of mosquitoes – no weekends, no holidays, early and late presence to take care of the mosquitoes – in order to maximize the mosquitoes’ reproduction and number of experiments. On the other hand, circadian cycles can be programmed to correspond to scientists’ preferences. The aim for both species is to adapt to mutual constraints, to benefit as much as possible from each other. In the insectarium, a paradoxical link is established with mosquitoes, as we can see here. In order to encourage the multiplication of mosquitoes, laboratory technicians prefer to give their own blood rather than to feed the females with sheep blood, in spite of the inconvenience caused by repeated mosquito bites. This feeding “on the arm” continues until the genetic adaptation of the mosquitoes to insectarium life is sufficient to impose sheep blood on the offspring of the females captured in Réunionese gullies and cemeteries. Moreover, a high level of reproduction is always considered a good thing, but mosquitoes are killed in their hundreds as soon as they have outlived their usefulness: “When you have a cage of 100 or 200 individuals, and you have to go to the next generation, you take a cage, you put it at −80°C for five minutes, it’s done!” says an entomologist. In this environment, mosquitoes represent another aspect of the “paradoxical useful pest” ( Claeys– Mekdade, 2002 : 154), where the care provided, the conditions in which the mosquitoes are bred ( Porcher, 2002 ), go hand–in–hand with scientific rationality. Mosquitoes reared in the insectarium correspond much more closely to Mason’s definition of domesticated animals ( Table 17.3 ): they are bred under human control, they make it possible to produce scientific data and they are gradually selected away from the wild type (the first generations are only reared in the insectarium). It is hard to say if they are tame. If we consider their ability to live in a constrained ecosystem, and to adopt an expected behaviour, they are tame. But nothing allows us to consider their individual behaviour, as insects and humans do not share a large frame of reference (Rennesson et al., 2012). Mason chooses to 296 Sandrine Dupé Table 17.3 The domesticity of experimental mosquitoes, according to Mason’s categories ( 1985 ).

Insectarium mosquitoes

BASIC DEFINITION 1/ Its breeding is under human control + 2/ It provides a product or services useful to humans + 3/ It is tame ? 4/ It has been selected away from the wild type +

EXCEPTIONS It is reared but not bred in captivity first generations It is bred in captivity to be released into the wild – Castrated animals some of the males EXCLUDED Laboratory animals (unless they are selected far from the wild type) − Animals in experimental research − Animals caught wild and tamed ?

− Characteristic excluding mosquitoes from the Mason definition ? Paradoxal characteristic of mosquitoes within Mason definition of domesticity + Characteristic falling within the Mason definition of domestic animal exclude laboratory animals used in experiments. According to him, although they are fed, reared and confined by humans, they have not found their place in the city (1985 ). This viewpoint can be the subject of extensive discussion (Silver, 2001) if we consider the biopower exerted on and by engineered animals as a means by which the city acts, at least indirectly ( Rabinow, 2006 ; Haraway, 2008 ). Indeed, the insectarium is an interstice where the manipulation of the living is aimed at transforming piblic health policies. According to Rader, engineering life in labo- ratories “actively seeks to obscure the boundaries between human and animal domestication, at the same time that they renaturalize domestication itself as a practice” ( Rader, 2007 : 195).

“Released mosquitoes”: human auxiliaries to domesticate all mosquitoes? The category “released mosquitoes” (Table 17.4) refers to a speculative entity during the ethnographic study. Since the first experimental releases occurred in 2016, it might cover another reality for entomologists now. If the experiments are conclusive, these released mosquitoes could be mass reared in a factory. In order to better deceive female mosquitoes, male sterile mosquitoes must maintain reproductive behaviour similar to that of their natural counterparts. At the same time, they must undergo breeding conditions imposed by the researchers prior to their release. This is why mosquitoes from the natural environment From parasite to reared insect 297 Table 17.4 The domesticity of released mosquitoes, according to Mason’s categories (1985 ). Released mosquitoes BASIC DEFINITION 1/ Its breeding is under human control + 2/ It provides a product or services useful to humans + 3/ It is tame ? 4/ It has been selected away from the wild type ?

EXCEPTIONS It is reared but not bred in captivity It is bred in captivity to be released into the wild + Castrated animals + EXCLUDED Laboratory animals (unless they are selected far from the wild type) Animals in experimental research Animals caught wild and tamed ?

? Paradoxal characteristic of mosquitoes within Mason definition of domesticity + Characteristic falling within the Mason definition of domestic animal

are regularly introduced into the breeding programme, in order to “feralize” the population, similar to practices pursued for over 2,000 years in the breeding of pheasants and partridges intended for hunting ( Digard, 1988 ). By imitating wild insects, male sterile mosquitoes can create an illusion of community with their counterparts, in order to be effective lures. By removing one of their vital functions – reproduction – researchers make them hybrid beings whose lives are diverted to the benefit of humans. The release of sterile mosquitoes aims to nur- ture nature ( Artaud, 2013 ). Finally, the potentially released mosquitoes fall more clearly under the category of domestic mosquitoes, although areas of uncertainty remain (Table 17.4). The domestication process is complete for released mosquitoes, as it involves their castration. The modification of the rules of mosquitoes’ reproduction is not trivial: scientists make these insects pass from the status of harmful insects to the status of auxiliary insects, and even auxiliary sacrificial insects. As discussed previously, considering these mosquitoes to be tame raises a number of problems. Like genuine fleas, these insects are encouraged to cooperate with humans, but their cooperation consists of reproducing under mass breeding conditions and then adopting a mating behaviour similar to that of wild mosquitoes. This ambivalence also precludes any decision about the degree of selection away from wild types, because the interest of these hybrid mosquitoes lies in their ability to play this in–between role. However , the fact of releasing sterile mosquitoes plays a role in domestication beyond the released mosquitoes themselves: controlling their reproduction 298 Sandrine Dupé implies controlling the mosquito population as a whole; in short, the beginning of Aedes albopictus domestication. By mating with females, they become the auxiliaries of humans seeking to control the population of all their homologues on the island. From the sterilization of male mosquitoes, there emerges a co–dependency between mosquitoes and humans. Mosquitoes reared in an insectarium depend on humans for their food and reproduction, and humans are dependant on them to control the vector risk. This mutual dependancy is, according to Budiansky (1999), the hallmark of domestication (cf. Rader, 2007 ).

Mosquitoes’ “fondness for man”: a path to domesticated parasites? The grid of analysis which results from the crossing of the domestication characteristics defined by Mason and the grounded categories of mosquitoes (Table 17.5) shows that none of these categories of mosquitos can be considered to be wholly domesticated. Does it mean that this frame of analysis is inappropriate? On the contrary, by giving us a broad and constant framework for interactions involving different actors, in different places, the results highlight the intense renogotiations of the terms of the relationship between humans and mosquitoes (Clark, 2007). When they considered the danger posed by mosquitoes, policy makers and scientists created a symbolic and operational segmentation of space through which new forms of interaction have been developed. Each space of interaction has been invested by different policy issues and different modes of inter–relation. Mosquitoes in general are attracted to humans, which officially includes them in the category of parasites, and determines the strategy of vector control. Mosquitoes in gardens became, in the political discourse, bred and reared by humans, in order to direct the forms of interactions between humans and mosquitoes towards birth control. In the insectarium, mosquitoes were indeed reared and bred by humans, but in order to control the mosquitoes’ behaviour directly. In this context, the relationship becomes favourable to humans; the mosquitoes’ behaviour is consistent with what humans expect. They are fed and reproduced in order to be involved in the production of knowledge and to participate in the development of a common political project with humans. In the last (theorical) place, sterile mosquitoes transgress what is known of domestication: they are not a tool of production, but they are transformed in such a way as to cooperate with humans in the struggle against their own species. It is a form of paradoxical domestication, which is not intended to form a domesticated population. Moreover, they are domesticated yet are required to behave as wild mosquitoes. They are confined only subsequently to be released. Nevertheless, they have this in common with all domestic species, which are the product of many mediations and reciprocal transformations with humans.

Conclusion A parasite is in a situation of dependence on its host, which gives it a form of availability that the host can then divert to its own benefit to varying degrees Table 17.5 The domesticity of Réunionese Aedes albopictus , according to Mason’s categories ( 1985 ).

Mosquitoes in Mosquitoes in Insectarium Released La Réunion gardens mosquitoes mosquitoes

BASIC DEFINITION 1/ Its breeding is under human control – ? + + 2/ It provides a product or services useful to humans – – + + 3/ It is tame – – ? ? 4/ It has been selected away from the wild type – – + ?

EXCEPTIONS It is reared but not bred in captivity – ? first generations It is bred in captivity to be released into the wild – – – + Castrated animals – – some of the males +

EXCLUDED Laboratory animals (unless they are selected far from the wild type) – Animals in experimental research – Animals caught wild and tamed ? ?

– Characteristic excluding mosquitoes from the Mason definition ? Paradoxal caracteristic of mosquitoes within Mason definition of domesticity + Characteristic falling within the Mason definition of domestic animal 300 Sandrine Dupé (O’Connor, 1997; Budiansky, 1999), ranging from simply fighting against the parasite to the emergence of a mutualism or domestication: “humans may select, but only from a set of options determined by forces beyond their control” ( Budi- ansky, 1999 : 50). Domestication would comprise a variety of resource management strategies whose spectrum ranges from exploitation to accommodation or mutuality ( Cassidy, 2007 ). The mosquitoes’ penchant for humans, which is particularly highly developed in Aedes albopictus , can only be resolved through acceptance and proximity, and then through appropriation, by sacrificing the mosquitoes involved in this approach. To adapt to this unbroken link, to turn it to their advantage, researchers, agents of the anti–vectorial service as well as locals have much to gain from sharing the world of mosquitoes, and to know them better. This penchant for humans is first used in social power relations, to enjoin the inhabitants to get rid of their mosquitoes by themselves, but also as a means of turning mosquitoes into auxiliaries of the struggle against them. Beyond the category’s intrinsic lability and the domestication process, if mosquitoes are so reticent to enter this category, it is probably because they are central to the creation of unprecedented relationships with humans and not because they have never cohabited, bitten and transmitted diseases to humans. But it is rather their place in society which is new. With the sterilization of mosquitoes, humans are trying to subsume the management of harmful fauna within an intraspe- cific problem that gives the illusion of preserving all non–harmful living beings ( Lezaun and Porter, 2015). This makes it possible to engineer nature as desired, to maintain a non–threatening wild world and to keep mosquitoes at a proper distance. Mosquitoes are not invited to enter the world of Man.

Notes 1 This research program was led by the IRD (Institute for Developpment Research) and the IAEA (International Atomic Energy Agency), and financed by the French Ministry of Health, the European Union and the Réunion Region. 2 There are 12 species of mosquitoes in Réunion island. Here, we will only consider one of the vector mosquitoes, Aedes albopictus . Boussès et al . (2013 ). 3 Only females bite. 4 This resarch programme is supported by the International Atomic Energy Agency.

References Artaud, H. ed., 2013. Leurrer la nature. Cahiers d’anthropologie sociale, 9. Paris: L’Herne. Boussès, P., Dehecq, J.S., Brengues, C. and Fontenille, D., 2013. Inventaire actualisé des moustiques (Diptera: Culicidae) de l’île de La Réunion, Océan Indien. Bulletin de Société de Pathologie Exotique , 106 (2), 113–125. Budiansky, S., 1999. The Covenant of the Wild: Why Animals Chose Domestication . Yale: Yale University Press. Cassidy, R., 2007. Introduction: Domestication Reconsidered. In : Cassidy, R. and Mullin, M. eds., Where the Wild Things Are Now: Domestication Reconsidered . Oxford and New York: Berg, 1–27. From parasite to reared insect 301 Claeys–Mekdade, C., 2002. Les controverses relatives à la démoustication de la Cam- argue: rapports à l’animal et au territoire. In : Staszak, J.F. ed., La place de l’animal . Numéro thématique de la revue Espaces et Sociétés. Paris: L’Harmattan, 147–165. Clark, N., 2007. Animal Interface: The Generosity of Domestication. In : Cassidy, R. and Mullin, M. eds., Where the Wild Things Are Now: Domestication Reconsidered . Oxford and New York: Berg, 49–70. Digard, J.P., 1988. Jalons pour une anthropologie de la domestication animale. L’Homme , 28 (108), 27–58. Dupé, S., 2015. Transformer pour contrôler. Humains et moustiques à La Réunion à l’ère de la biosécurité. Revue d’Anthropologie des Connaissances , numéro spécial, la Surveil- lance des animaux, 9 (2), 213–236. Foucault, M., 1984. The Foucault Reader . P. Rabinow ed. New York: Pantheon. Galton, F., 1863. The First Steps Towards the Domestication of Animals . London: Ethno- logical Society of London. Haraway, D., 2008. When Species Meet (Posthumanities). Minneapolis: University of Minnnesota Press. Leach, H.M., 2005. Human Domestication Reconsidered. Current Anthropology , 44 (3), 349–368. Mason, L., ed., 1985. Evolution of Domesticated Animals . London and New York: Longman. Micoud, A., 2010. Sauvage ou domestique, des catégories obsolètes? Sociétés 2 (108), 99–107. Nading, A., 2012. Dengue Mosquitoes Are Single Mothers: Biopolitics Meets Ecologi- cal Aesthetics in Nicaraguan Community Health Work. Cultural Anthropology, 27 (4), 572–596. O’Connor, T.P., 1997. Working at Relationships: Another Look at Animal Domestication. Antiquity , 71 (271), 149–156. Porcher, J., 2002. Eleveurs et animaux: réinventer le lien. Paris: Presses Universitaires de France. Rabinow, P., 2006. Biopower Today. BioSocieties , 1 (2), 195–217. Lezaun, J. and Porter, N., 2015. Containment and Competition: Transgenic Animals in the One Health Agenda. Social Science and Medicine , 129, 96–105. Rader, K., 2007. The Metaphor of Domestication in Genetics. In : Cassidy, R. and Mullin, M. eds., Where the Wild Things Are Now: Domestication Reconsidered . Oxford and New York: Berg, 183–204. Rennesson, S., Grimaud, E. and Césard, N., 2012. Insect Magnetism. The Communication Circuits of Beetle Fighting in Thailand. Hau: Journal of Ethnographic Theory, 2 (2). Available from: www.haujournal.org/index.php/hau/article/view/136/208 [Accessed 07 Jul 2017]. Silver, L., 2001. Thinking Twice, or Thrice, About Cloning. In : Klotzko, A.J. eds., The Cloning Sourcebook . Oxford: Oxford University Press, 60–63. Von Uexküll, J., 2010 (1934). A Foray Into the Worlds of Animals and Humans . J.D. O’Neill transl. Minneapolis: University of Minnesota Press. Index

Note: Page numbers in italic and bold format indicate figures and tables respectively. acquired immune deficiency syndrome baraka concept 189 , 192 (AIDS) 115 , 121 – 122 biodiversity 1 , 40 , 252 Aedes albopictus 291 , 292 , 293 , 300 birds: communication initiated by 209 ; agriculture 2 , 11 , 12 , 85 – 88 , 148 terrestrial 204 – 207 agrosystems 98 , 102 , 104 – 107 , 168 , 169 blackbucks 200 , 208 animal labour see labour Black Sea 148 , 154 , 157 animal production 251 , 257 n2 bone collagen 84 , 86 , 87 , 90 , 91 animal remains 82 , 129 , 134 bottlenecks 24 – 27 , 30 , 31 , 32 animal(s): change in the perception of 68 ; breeding: about 1 , 2 , 4 , 6 ; directed 23 , 24 ; dairy 127 , 136 ; empathy with 279 ; fats livestock 67 , 252 , 253 , 257 n2 131 , 132 , 133 ; fodder 99 , 101 , 103 ; Bronze Age 88 , 104 , 106 – 107 , 145 , 193 ; Late “liberation” of 252 , 253 ; living 8 , 14 , Chalcolithic-Early Bronze Age 106 – 107 127 , 150 ; products from 48 – 49 , 50 – 53 , 132 ; sacrifice of 54 , 146 , 154 , 176 , Cameroon see Duupa farmers 177 ; shifting stance for understanding canid products 48 – 49 , 50 – 53 279 – 281 ; see also domestic animals; caprification practice 182 , 183 , 184 wild animals caprines 134 , 136 , 137 , 138 “animal welfare” 256 , 257 carnivores 40 , 46–47 , 53 anthropized environments 9 , 12 , 261 ; Çatalhöyük 104 – 105 , 107 anthropogenic landscape 172 – 175 , 180 cattle 9 , 14 , 26 , 28 , 66 , 67 , 83 , 86 , 91 , 104 , Anthropocene concept 1 , 23 , 145 , 146 , 147 105 , 127 , 128 , 128 – 131 , 132 , 134 – 137 anthropophobia 71 , 72 cereal: crops 98 , 99 ; fields 100 , 103 , 106 ; antiretrovirals 122 – 124 harvest 107 , 184 archaeobotanical remains 98 , 104 – 107 chickens 27 , 30 , 32 , 83 , 84 artificial selection 24 – 25 , 30 – 31 , 235 chikungunya 289 , 292 auxiliary animals: about 197 – 198 ; among China 85 – 88 tamed wild-born captives 200 – 202 , 208 ; co-evolution 9 – 10 , 119 , 145 , 183 , 243 , 284 conclusion about 214 – 215 ; description commensal pathway for domestication 23 , of 198 ; discussion about 207 , 209 , 27 , 81 , 82 , 85 , 199 211 – 214 ; and domestication 208 , common vetch 98 , 102 213 – 214 ; for foraging activities 199 , communication: between humans and dogs 208 , 210 , 214 ; and hunting and fishing 237 – 242 ; interspecific 207 , 209 ; with activities 198 , 201 ; and interspecific solitary versus grouped partners 209 , communication 207 , 209 , 211 – 213 ; 211 ; visual 239 – 240 and marine and riverine mammals companion species 13 , 124 202 – 204 ; non-captive 202 – 207 ; profile conscious selection 25 – 26 , 97 , 172 of 210; and terrestrial birds 204 – 207 ; Convention on the Conservation of and tritrophic interaction 212 – 213 ; uses European Wildlife and Natural Habitats of 198 of the Council of Europe 72 Index 303 Convention on International Trade in reproduction isolation in 28 ; see also Endangered Species (CITES) 72 wild animals cormorant 201 , 212 domesticated plants 4 , 6 , 30 , 130 , crop mimics 12 , 99 – 101 , 104 , 107 165 – 167 , 176 – 177 crops: annual 100 , 175 ; and weeds 11 , 12 domestication: about 1 ; and ancient crossbreeding 6 , 68 , 69 genomes 31 – 32 ; and animal labour Cruciferae/Brassicaceae families 104 , 105 251 – 257 ; artificial selection for 30 – 31 ; and auxiliary animals 213 – 214 ; dairy animals see animal(s ) common ground of 13 – 14 ; configuration dairying: evolution of 136 – 137 ; husbandry of 283 – 286 ; defined 291 ; description 129 , 134 – 136 ; products 128 , 130 , 131 , of 3 , 4 , 23 ; and domestic mammals 134 ; see also animal(s), dairy 40 – 41 ; as domination 7 – 8 ; and Duupa/ deleterious mutations 31 plant relationships 175 – 177 ; excluded dietary needs: of dogs 272 ; of pigs 83 – 85 , animals from 291 ; of fig and olive trees 88 , 89 ; of reindeer 269 , 272 ; of wolves 179 – 193 ; and genetic diversity 26 – 27 ; 46 of HIV 122 – 124 ; how to recognise Dioscorea abyssinica 167 , 168 5 – 10 ; and human mosquito relations Dioscorea cayenensis-rotundata 167 , 168 290 – 291 ; importance of 23 ; livestock directed pathway for domestication 23 , 81 , 66 , 129 ; and livestock farming 283 – 286 ; 199 , 215 n2 models of 24 – 26 ; as a monstrous DNA ancient sequences (aDNA) 30 , 32 cohabitation 252 – 253 ; of mosquitoes DNA sequencing technologies 24 , 26 , 292 , 293 , 294 – 296 , 297 , 299 ; motives 30 , 31 for 54 ; and nature-culture divorce dogs: about 7 , 8 , 11 ; ancestors and 3 – 4 ; and non-human agencies 11 – 13 ; descendants 65 – 66 ; and artificial as ongoing transformation 10 – 11 , 13 ; selection 235 ; behavioural responses of other narratives about 4 – 5 ; pathways 238 ; categories of 39 ; communication to 81 , 199 ; points of view about 2 ; with 237 – 242 ; conclusion about process of 179 , 180 ; reciprocal 124 ; 242 – 243 ; contributions made by 270 ; as reproduction control 5 – 6 , 10 , 123 ; crossbreeding issues 68 – 69 ; dietary and reproduction isolation 27 – 30 ; and needs of 272 ; domestication of 41 , 55 , reproduction techniques 190 – 191 ; 66 , 235 ; emotional bond with 236 – 238 ; syndrome 6 – 7 , 13 , 23 , 40 – 42 , 97 , 180 ; feral 39 , 70 , 71 , 72 ; for foraging and technology 283 – 286 ; of ungulates activities 208 ; forms of appropriation of 13 , 14 , 40 , 67 ; of virus 123 ; of wolf 40 53 ; free-ranging 39 , 69 , 71 , 72 , 73 ; gene domination: about 1 , 2 ; and hierarchy flow from and into 28 , 29 ; and genetic 148 – 150 ; paradigm 7 – 8 diversity 26 ; herds of 271 ; human-dog domus 3 , 13 – 14 , 180 , 182 , 279 , 291 relationship 66 – 68 , 73 – 74 , 235 – 237 ; Duupa farmers: about 165 – 166 ; cereals hunting 262 , 263 , 264 , 267 , 270 – 271 , produced by 166 , 169 , 170 , 172 , 175 , 273 ; and interbreeding 29 ; origin of 176 ; discussion about 175 – 177 ; and 39 ; purebred dogs 26 , 67 ; relationship leafy vegetables 172 – 175 ; relationship with humans 66 – 68 , 73 – 74 , 235 – 237 ; with plants 175 – 177 ; seeds’ selection by relaxed selection hypothesis 31 ; sled 167 – 172 ; and tree species 175 262 , 265 , 266 , 267 , 272 , 273 ; social referencing used by 239 – 240 ; see also early nomads 13 , 146 , 147 wolves elephants: about 221 – 222 ; African 200 ; dolphins 202 – 203 , 209 , 210 , 211 , 214 ; and apprenticeship of commands river dolphins 203 , 208 , 211 225 – 226 ; Asian 200 , 221 , 229 – 231 , domestic animals: about 2 , 4 , 6 – 9 ; as 232 n1, 232 n2; capturing and stalking animal machines 251 ; definitions of of 226 – 229 ; conclusion about 231 – 232 ; 92 – 93 ; features of 291 ; for foraging forest 222 , 224 , 227 , 229 – 232 ; and activities 208 ; gene flow from 28 , 81 , imitation process 226 , 231 – 232 ; and 90 ; genetic diversity in 27 ; humans’ Khamti population 230 , 231 ; taming of relationship with 127 – 128 ; and 223 – 225 ; tied to lak chang 223 ; village interbreeding 29 ; labour issues 251 – 257 ; 222 , 224 – 226 , 229 – 232 ; wild elephants, 304 Index capturing and stalking of 226 – 229 ; wild 73 ; and HDR communities 272 ; levels elephants, taming of 223 – 225 of 41 ; and Upper Palaeolithic period 40 Endangered Species Act (1966) 68 human control: about 4 , 9 ; hypothesis endogenous retroviruses 117 – 119 , 125 n5 52 , 53 ; of wolves 45 ; see also environmental crisis 2 , 252 domination human-dog-reindeer communities 14 , 261 , farm animals 81 – 82 , 93 , 251 – 252 262 , 265 , 266 , 267 – 270 , 271 – 273 feral animals: dogs 39 , 70 , 71 , 72 ; pigs human immunodeficiency virus (HIV) 88 – 93 115 , 117 – 124 , 125 n13, 125 n16 fig trees 179 – 186 , 188 – 190 humans: and communication with dogs food: products 127 , 252 ; storage 45 , 51 , 52 237 – 242 ; and ecological association foraging activities 3 , 15 , 198 – 199 , 208 , with wolves 42 ; and emotional bond 210 , 214 with dogs 236 – 238 ; and lactose founder effect 24 , 25 intolerance 129 – 131 ; and non-human agencies 11 – 13 ; relationship with dogs Gamma irradiation 294 and wolves 66 – 68 , 73 – 74 , 235 – 237 ; gathering activity 201 – 202 relationship with domestic animals gene flow 25 , 26 , 28 , 29 , 81 , 90 127 – 128 ; relationship with mosquitoes genetics: and ancient genomes 31 – 32 ; 15 , 290 – 291 ; relationship with plants architecture of traits 24 , 30 ; and 97 – 108 ; and trees 188 – 189 ; and viruses artificial selection 30 – 31 ; conclusion 115 – 124 ; see also auxiliary animals about 32 ; diversity 24 – 27 , 30 – 32 , hunter-gatherers: about 6 , 8 , 15 ; and 67 ; and domestication 26 – 27 ; and auxiliary animals 209 , 210 ; and human-animal relationships 73 ; and domestication 44 , 45 , 46 , 48 ; and reproduction isolation 27 – 30 honeyguides 207 ; Upper Palaeolithic genomes (ancient), rise of 31 – 32 period 39 , 40 , 49 , 55–56 geometric morphometrics (GMM) 88 , 93 hunting and fishing activities 198 , 201 goats 7 , 14 , 26 , 66 , 67 , 101 , 104 , 105 , husbandry practices 82 – 88 , 92 , 137 127 , 128 , 128 – 131 , 134 , 137 – 138 , 184 , hybrid communities 4 , 12 , 14 – 16 , 56 , 255 123 , 261 grafting practice 181 , 186 , 188 , 191 , 192 hybridization between dogs and wolves 4 , guiding behaviour 206 – 207 11 , 68 – 74

HDR communities see human-dog- Indicatoridae 206 – 207 reindeer communities Inner Asia 145 , 147 herds/herders: dogs 271 ; nomadic 14 , 15 ; interbreeding 27 – 29 , 66 , 68 , 70 pigs 83 – 85 ; reindeer 261 , 265 , 269 – 271 International Union for Conservation of Herodotus 148 , 149 , 156 , 157 Nature (IUCN) 68 , 200 hierarchy and domination 148 – 150 intimacy 8 – 10 , 14 , 192 hominids 65 , 197 , 198 , 215 introgression 69 – 72 , 82 , 97 , 181 honeyguides 206 – 207 , 208 , 209 , 212 – 213 , Iron Age 90 , 98 , 145 , 148 , 152 , 157 215 horses: abandoned and domesticated 256 ; Jbala settlements 181 – 182 , 186 about 7 , 8 , 9 , 13 , 14 ; and assumption of hierarchy 150 – 151 ; bridle structure 153 ; Khamti population: about 222 ; and capture and deleterious mutations 31 ; equine operations 226 – 229 ; relationship with domestication 145 , 148 ; equine sociality elephants 222 – 225 , 230 , 231 , 232 n3 13 , 151 , 152 ; Garrano ponies 254 – 255 ; genetic diversity issue 32 ; management labour: about 251 – 252 ; animal release of 152 – 154 ; way of 150 – 152 ; wisdom from 253 – 254 ; and “animal welfare” of 156 – 158 ; and wolves 42 , 44 , 46 256 , 257 ; conclusion about 257 ; and human-animal relationships: about 10 ; and mode of production 252 ; promise of farm animals 81 , 82 , 93 ; and genetics protection offers by 255 – 257 Index 305 lactase persistence (LP) 127 , 130 neotenic theory 7 , 73 lactation issues 128 , 131 , 134 , 135 , 281 , Northern peoples 40 , 45 , 47 , 50 , 262 282 landscape: agricultural 188 ; anthropogenic Olea europaea 181 172 – 175 , 180 ; coastal 261 , 273 ; Duupa olive trees 186 – 193 ; oleaster fruits 166 , 172 ; sharing of 67 186 – 188 , 191 leafy vegetables 172 – 175 orcas 203 – 204 , 208 , 209 , 210 , 211 Linearbandkeramik complex 105 – 107 outcrossing 28 , 29 , 169 , 179 livestock farming: about 252 , 275 ; oxytocin systems 236 , 237 breeding 67 , 252 , 253 , 257 n2; domestication 66 , 129 , 283 – 286 ; Palaeolithic see Upper Palaeolithic practical configuration for 278 – 283 ; Palaeolithic dog morphotype 42 robotic 276 – 278 parasitism 15 , 289 , 290 , 291 , 294 pastoralist societies 207 , 212 , 215 meat production 14 , 83 , 84 , 85 Pazyryk culture and people: about 145 ; Mesopotamia 106 – 107 and armor and weaponry 155; burial milk and milking: about 13 , 14 ; adoption mounds 146 , 147, 148 ; conclusion of 132 ; calf present during 135; and about 158 ; and evidence of violence ceramic containers 131 , 132 , 136 ; 154 , 156 ; and human-horse connections composition of 128 ; conclusion about 152 – 154 137 – 138 ; cow’s milk 128 , 137 ; and peccary, non-domestication of 222 dairy evolution 136 – 137 ; and dairy pet-directed speech (PDS) 241 , 242 , 243 husbandry 134 – 136 ; human 128 ; pigs: about 7 , 11 , 15 ; and commensal lactation issues 128 , 131 , 134 , 135 , 281 , pathway 27 ; conclusion about 92 – 93 ; 282 ; and milk production 134 , 136 ; dietary needs of 83 – 85 , 88 , 89 ; feral and milk use 131 – 134 ; as a nutritious 88 – 93 ; herds of 83 – 85 ; husbandry food 128 ; role of 129 ; understanding practices 84 – 87 ; and outcrossing 29 ; animal’s behaviour for 281 – 283 ; use of reliance on farmed food 87 robots for 276 – 278 ; see also domestic plants: domesticated 130 , 165 , 166 , 167 , animals 176 , 177 ; human/plant interactions 165 , millet cultivation 85 – 89 166 , 172 , 177 ; and interbreeding 28 ; and mosquitoes: about 289 – 290 ; Aedes notion of fondness 12 ; relationship with albopictus 291 , 292 , 293 , 300 ; breeding humans 97 – 108 of 289 , 291 – 294 , 296 – 297 ; conclusion Portugal, Garrano ponies in 254 – 255 about 298 , 300 ; domestication of 292 , prey pathways 81 , 82 , 199 293 , 294 – 296 , 297 , 299 ; and “fondness productivism 283 – 284 for man” 290 , 291 , 293 , 298 ; in the pulses: cultivation of 168 – 169 ; fields 99 , garden 292 – 294 , 298 ; maintaining 100 reproductive behaviour 296 – 297 ; purebred dogs 26 , 67 relationship with humans 15 , 290 – 291 ; released 296 – 298 ; sterile 289 , 290 , ravens 204 – 206 , 209 , 210 , 215 296 – 298 ; vector control strategies reciprocal learning 270 – 272 289 , 290 reindeer: about 5 , 6 , 14 ; contributions mullet fishing 204 , 209 made by 268 – 269 ; dietary needs of 269 , mustard 104 , 105 272 ; domesticated 213 , 215 n2; herds of mutualism 15 , 56 , 118 , 211 – 212 , 290 , 261 , 265 , 269 – 271 ; human interactions 300 with 267 – 268 ; ratio of 263 , 264 ; see also human-dog-reindeer communities nature-culture divorce 3 – 5 relaxed selection 30 – 31 Nenets laikas 270 reproduction: control 5 – 6 , 10 , 123 ; Neolithic Age 1, 3 , 6 , 8 , 11 ; farmers isolation 24 , 26 – 30 , 32 ; techniques and farming 11 , 86 , 107 , 134 , 136 ; 190 – 192 Revolution 3 , 4 , 5 , 13 retroviruses, endogenous 117 – 119 , 125 n5 306 Index Réunion island 289 – 294 , 300 n2 ungulates domestication 13 , 14 , 40 , 67 robots for milking 276 – 278 Upper Palaeolithic period: about 39 – 40 ; rootstocks 181 , 187 , 191 , 192 , 193 bear-related rituals in 47 – 48 ; canid ruminant species 127 , 128 , 129 , 132 , 135 morphotypes in 41 – 42 ; hunter-gatherers in 39 , 40 , 49 , 55–56 sacrifice of animals 54 , 146 , 154 , 176 , 177 Saharan rock art iconography 135 viruses 115 – 124 Scythian warriors 157 – 158 visual communication 239 – 240 seed crops 97 , 98 , 105 selection 10 , 13 , 25 , 30 , 31 , 41 , 45 , 53 , weeds: arable weeds 97 –108; and crops 11 , 54 – 55 , 97 – 98 , 166 , 170 , 176 , 179 – 180 , 12 ; hand-weeding 98 – 100 , 102 , 104 , 190 , 191 – 192 , 243 , 270 , 295 , 297 ; 105 ; leafy greens as 173 ; use of 98 , 103 , artificial 6 , 24 – 26 , 29 , 30 – 32 , 235 ; 104 , 107 conscious 25 , 25 – 26 , 172 ; methodical weedy crop relatives 97 , 99 9 ; natural 6 , 9 , 28 , 130 , 137 ; relaxed whale hunting 204 , 214 30 – 32 ; unconscious 9 wild animals: about 6 , 7 , 8 , 9 ; self-domestication 42 , 44 – 46 , 51 , 66 appropriation of 53 ; and artificial sheep 14 , 66 , 67 , 83 – 85 , 84 , 85 – 86 , 91 , selection 30 – 31 ; and deleterious 104 , 105 , 128 , 128 , 129 – 131 , 134 , 137 , mutations 31 ; gene flow from and into 138 , 184 , 251 – 252 , 295 25 , 26 , 28 , 29 , 30 ; genetic diversity in signs, animal products related to 49 , 50 27 ; and interbreeding 29 ; reproduction smallpox 115 – 119 , 124 isolation in 28 ; see also domestication sorghum: cultivation of 169 – 170 , 176 ; wild boars 27 , 29 , 31 , 82 , 86 – 90 , 92 diversity of 171 wolves: ancestors and descendants 65 – 66 ; Spanish conquest 117 – 118 appropriation of 53 ; black coat colour in speciation theory 26 , 27 , 28 , 29 , 30 , 32 74 n2; Canis lupus 70 , 71 ; crossbreeding stable isotope analysis 82 – 85 , 87 , 89 , issues 68 – 69 ; dietary needs of 46 ; 90 – 93 , 106 , 137 domestication 40 , 42 , 44 – 48 , 50 – 55 , subsistence economy 83 , 98 , 107 , 132 , 134 68 – 74 ; and ecological association with suids 29 , 83 , 88 , 90 , 91 , 92 humans 42 ; forms of familiarisation symbiosis 9 – 10 , 15 , 209 , 272 , 273 of 51 – 53 ; fossil wolves 39 , 44 ; gene flow from and into 28 , 29 ; grey 65 , taiga people 268 , 269 66 , 235 ; and horses 42 , 44 , 46 ; human technology and domestication 283 – 286 control of 45 ; human-socialized 54 ; tree domestication: about 179 – 181 ; hyper-sociability hypothesis 55 ; and approaches to 191 ; conclusion about interbreeding 29 ; interest in the skin of 191 – 193 ; and ecologies and societies 49 ; products from 48 – 49 ; and ravens 182 – 188 ; and fig trees 182 – 186 , 205 – 206 ; rehabilitation campaign 68 ; 188 – 190 ; and fruit trees 179 , 189 , 192 ; relationship with humans 66 – 68 , 73 – 74 , and materials and methods 181 – 182 ; 235 – 237 ; self-domestication of 42 , 44 – 46, and olive trees 186 – 191 51 ; as a symbol of the wilderness 68 tritrophic interaction 212 – 213 Tuvan hunters 216 n3, 216 n5 yams 167 – 168 Tuva nomadic herders 14 , 15 Yellow River Basin 85 , 86