Biological Theory https://doi.org/10.1007/s13752-021-00379-6 THEMATIC ISSUE ARTICLE: EVOLUTION OF KINSHIP SYSTEMS Kin Against Kin: Internal Co‑selection and the Coherence of Kinship Typologies Sam Passmore1 · Wolfgang Barth2 · Kyla Quinn2 · Simon J. Greenhill2,3 · Nicholas Evans2 · Fiona M. Jordan1 Received: 4 December 2020 / Accepted: 12 April 2021 © The Author(s) 2021 Abstract Across the world people in diferent societies structure their family relationships in many diferent ways. These relationships become encoded in their languages as kinship terminology, a word set that maps variably onto a vast genealogical grid of kinship categories, each of which could in principle vary independently. But the observed diversity of kinship terminology is considerably smaller than the enormous theoretical design space. For the past century anthropologists have captured this variation in typological schemes with only a small number of model system types. Whether those types exhibit the internal co-selection of parts implicit in their use is an outstanding question, as is the sufciency of typologies in capturing variation as a whole. We interrogate the coherence of classic kinship typologies using modern statistical approaches and systematic data from a new database, Kinbank. We frst survey the canonical types and their assumed patterns of internal and external co-selection, then present two data-driven approaches to assess internal coherence. Our frst analysis reveals that across parents’ and ego’s (one’s own) generation, typology has limited predictive value: knowing the system in one generation does not reliably predict the other. Though we detect limited co-selection between generations, “disharmonic” systems are equally common. Second, we represent structural diversity with a novel multidimensional approach we term kinship space. This approach reveals, for ego’s generation, some broad patterning consistent with the canonical typology, but diversity (and mixed systems) is considerably higher than classical typologies suggest. Our results strongly challenge the descriptive adequacy of the set of canonical kinship types. Keywords Anthropology · Cross-cultural kinship · Linguistics · Semantic systems · Typology Introduction in English).1 The kinship system of any language can thus be seen as a particular matrix of equivalences or “syncretisms” In the languages of all societies, the vast number of logically demonstrating which kin types use identical or similar terms, distinguishable kin types (e.g., categories such as father’s and which are diferentiated. The taken-for-granted nature of younger sister, father’s older sister, mother’s younger sister, how we use language means that the confation of such cat- mother’s older sister) are mapped onto a smaller and more egories seem obvious to speakers of the language that uses tractable set of kin terms (e.g., aunt for all the above terms them (here English, for aunt), but counterintuitive to those whose own languages ofer diferent conceptual cuts (e.g., the Australian language Kayardild, which would reserve the * Fiona M. Jordan term marrkathu for father’s sisters but group mother’s sisters [email protected] with one’s mother, as ngamathu (Evans 2011)). Linguists 1 Department of Anthropology and Archaeology, University and anthropologists must thus move between the language- of Bristol, Bristol, UK internal conceptual groupings that seem obvious to their 2 ARC Centre of Excellence for the Dynamics of Language speakers (aunt; marrkathu), generally called emic in the (CoEDL), Australian National University, Canberra, ACT , literature, and the fner language-independent categories Australia 3 Department of Linguistic and Cultural Evolution, Max Planck Institute for the Science of Human History, Jena, 1 Useful papers surveying this topic are Dole (1969) and Hage Germany (2001). Vol.:(0123456789)1 3 S. Passmore et al. (e.g., “father’s older sister”)2 which form an etic grid against anthropological puzzle for decades. Situated at a nexus of which language-specifc categories can be calibrated. Kin language, cognition, and social reproduction, kinship sys- types are etic; kin terms are emic, and the cross-linguistic tems ofer a multifaceted domain through which to explore study of kinship terminology focuses on the diferent ways cognitive and cultural evolution. languages map their kin terms onto the universally-defnable Since the pioneering work of Morgan (1871) and Fison etic grid to produce kinship systems. and Howitt (Gardner and McConvell 2015) in the 19th cen- Notably, the enormous design space of possible kinship tury, scholars have worked to establish an encompassing systems is thinly populated amongst vast empty spaces. The typology of kinship systems for the languages of the world. pioneering study of sibling terms by Nerlove and Romney By compiling large sets of terms, across many languages, (1967) found that among 4,140 possible sibling term sys- for hundreds of kin types (over 260 in Morgan’s case), these tems,3 only 15 systems occurred more than twice in their researchers discerned recurrent patterns in which diferent sample of 245 languages, and these 15 occupied 95.5% of parts of the system appear to correlate. This led to the classic the sample; moreover, there were only 23 systems altogether typologies of the early to mid-20th century (see the “Classic within their sample. Once we move from sibling terms to Typologies of Kinship” section), which assigned languages parents, uncles and aunts, cousins, grandkin, and in-laws, to a canonical “system” (e.g., Iroquois/Dravidian,5 Eskimo,6 the number of possible systems rapidly climbs into unim- Hawaiian) on the basis of diagnostic syncretisms and distinc- aginably large numbers.4 The same pattern holds: there are tions in one crucial subset of the terms (typically siblings a strikingly limited number of attested systems with respect and/or parents and nuncles7). Important questions of exter- to those that are logically possible. Some of these unat- nal co-selection between particular systems and other social tested possibilities, it has been argued, are due to trade-ofs phenomena, could then be pursued empirically: for example, between simplicity and informativeness (Kemp and Regier whether the presence of cross-cousin marriage is correlated 2012; Kemp et al. 2018). These can take the form of mark- with Dravidian or Iroquois kinship systems (Godelier et al. edness constraints, favoring the greater elaboration of terms 1998). Recurrent and reliably identifable systems of social for nearer rather than more distant kin, or a preference for and cultural organization make human diversity tractable conjunctive characterizations (“elder AND male referent, to cross-cultural analysis, and typologies of (e.g.) subsist- i.e., elder brother”) over disjunctive ones (“elder OR male ence, political organization, and color naming have allowed referent,” i.e., brother (younger or older) or older sister) researchers to ask broad questions about cultural change and (Greenberg 1966). However, studies like these only identify statistical regularities. Logically however, the validity of the some of the constraints on attested systems, and we focus answers depends on the coherence and operationalizability here on a further potential constraint—the consistency of of the systems themselves, which we can only evaluate once where distinctions or mergers are made—as a selector on we have good data on how far kin types really do cohere why so few systems populate the enormous design space. across the whole system of terminology, as well as across We will term this internal co-selection since it implies that a broad and phylogenetically well-balanced sample of lan- a design feature of one part of the system (say, the treatment guages. While we are certainly not the frst to suggest that of uncle terms) has selective consequences for another part subsystems will vary independently (e.g., D’Andrade 1971; of the kin system (say, the treatment of cousin terms): there Murdock 1970), we here advocate a root-and-branch review is co-selection, by hypothesis, between diferent syncretisms of these types, capitalizing on new data aggregation and within the overall system. Understanding the constraints on computational methods, with the goal of a more accurate these semantic systems of social life, how they emerge and and fexible description of how we carve up the social world. change, and why we see some but not others, has been an The aim of this article is the frst step in this endeavor: to evaluate whether systems of kinship terminology do in fact exhibit the terminological coherence that the classic typolo- 2 For brevity, we use the following abbreviations for the rest of gies presume. We draw on a large global sample of 1,107 the article: B = brother, D = daughter, e = elder, f = female refer- ent, F = father, G = sibling, m = male referent, M = mother, S = son, y = younger, X = cross, Z = sister, // = parallel. These are chained together as appropriate, e.g., MB for “mother’s brother.” 5 Morgan, whose interest in kinship began with various Iroquoian 3 Nerlove and Romney considered eight possible kin types, defned languages, used the term Iroquois for this type, but subsequent inves- by the three-way binary matrix “sex of referent” versus “sex of ego” tigators have often used the term Dravidian for what is essentially the versus “referent older or younger than ego.” Some further distinctions same system. See the second section. are possible but are rare. The fgure of 4140 is the Bell number for 8 6 We apologize to those Inuit peoples who fnd this term ofensive. (the number of ways a set of 8 cells can be partitioned into groups). We regretfully retain this term here simply because it is the one that 4 The authors’ database gives terms for 98 kin types (still just a sub- articulates with the literature, and we advocate for an alternative. set of all possible kin terms–it excludes, e.g., great-uncles and great- 7 For brevity, we adopt the widespread term “nuncle” for “uncle or aunts) and the Bell number for these 98 terms is 5.55 × ­10112.