sign in sign up
<<
Home , Churl, Thegn

The Retrospective Methods Network

Newsletter

Interdisciplinary and Comparative Methodologies

№ 14

Exploring Circum-Baltic Cultures and Beyond

Guest Editors: Joonas Ahola and Kendra Willson

Published by Folklore Studies / Department of Cultures University of Helsinki, Helsinki

1

RMN Newsletter is a medium of contact and communication for members of the Retrospective Methods Network (RMN). The RMN is an open network which can include anyone who wishes to share in its focus. It is united by an interest in the problems, approaches, strategies and limitations related to considering some aspect of culture in one period through evidence from another, later period. Such comparisons range from investigating historical relationships to the utility of analogical parallels, and from comparisons across centuries to developing working models for the more immediate traditions behind limited sources. RMN Newsletter sets out to provide a venue and emergent discourse space in which individual scholars can discuss and engage in vital cross- disciplinary dialogue, present reports and announcements of their own current activities, and where information about events, projects and institutions is made available.

RMN Newsletter is edited by Frog, Helen F. Leslie-Jacobsen, Joseph S. Hopkins, Robert Guyker and Simon Nygaard, published by: Folklore Studies / Department of Cultures University of Helsinki PO Box 59 (Unioninkatu 38 C 217) 00014 University of Helsinki Finland

The open-access electronic edition of this publication is available on-line at: https://www.helsinki.fi/en/networks/retrospective-methods-network

Interdisciplinary and Comparative Methodologies: Exploring Circum-Baltic Cultures and Beyond is a special issue organized and edited by Frog, Joonas Ahola and Kendra Willson.

© 2019 RMN Newsletter; authors retain rights to reproduce their own works and to grant permission for the reproductions of those works.

ISSN 2324-0636 (print)

ISSN 1799-4497 (electronic)

All articles in the main section of this journal have been subject to peer review.

2

Contents

Editor’s Column ...... 6

INTERDISCIPLINARY AND COMPARATIVE METHODOLOGIES EXPLORING CIRCUM-BALTIC CULTURES AND BEYOND Interdisciplinary and Comparative Methodologies: An Introduction ...... 7 Frog, Joonas Ahola and Kendra Willson Abstract: The core of this special issue engages with the constructedness of ideas and understandings rooted in the discourses of scholarship. This introductory article explores perspectives on discourses and practices of interdisciplinarity. The discussion of topics related to these issues develops in dialogue with individual contributions to the collection in ways that make their interconnections and complementarity apparent. Theory and Methodology for Assembly Site Identification and Analysis ...... 17 Alexandra Sanmark Abstract: This article presents a new interdisciplinary methodology successfully developed for the identification and detailed analysis of Viking-Age outdoor assembly sites. This methodology draws on archaeological evidence, topographic information and a wide range of written sources, from laws to eddic poetry. It is hoped this approach will inspire a new way of thinking, leading to the identification of outdoor assembly-sites in areas of the world where they are not yet known. in the Social Order of Anglo-Saxon England and Viking-Age : Outlines of a Methodological Reassessment ...... 25 Denis Sukhino-Khomenko Abstract: The article addresses the possibilities for a methodological reassessment of the phenomenon of the thegns in England and Scandinavia in the late (ca. 900–1066). The historiographical overview reveals that the thegns have never been examined for their own sake, and that the recent developments in source studies open new methodological prospects in anthropological research. The case study of the thegns hopes to outline some of them. Castrén’s Lectures on Finnish Mythology: Methodological Reflections ...... 51 Joonas Ahola and Karina Lukin Abstract: M.A. Castrén’s Föreläsningar i finsk mytologi [‘Lectures on Finnish mythology’] represents an important phase in the study of mythology and religion. It is the first comparative study on Finnish and related mythologies that draws on contemporary international scholarship. The Lectures are here reviewed critically both as an emblem in the history of the discipline and as a case study on challenges in combining methodologies from different disciplines and approaches. Great Goddess Theory in Ancient Germanic Studies ...... 70 Joseph S. Hopkins Abstract: The present article discusses a historical tendency among scholars to focus on male-gendered deities in the Germanic corpus, a boom period of goddess studies in the 1990s under the influence of Marija Gimbutas, and the place of Great Goddess theory in the field of ancient Germanic studies today. Great Goddess theory is proposed to remain an influence in ancient Germanic studies, particularly in the field’s tendency to identify goddesses as extensions of a single entity. Runo Revisited: Borrowing and Semantic Development ...... 77 Kendra Willson Abstract: The Finnish and Karelian word runo [‘poem’] has been viewed as reflecting a loan from Proto-Germanic *rūnō, the ancestor of rún [‘rune’]. However, given its limited distribution in Finnic and the meanings attested in Germanic and Finnic languages, the word was more likely borrowed at a later stage, from Early Norse

3

near the start of the Viking Age, probably in the meaning of ‘incantation, verbal charm’ and in connection with incantational magic. On the Analogical Comparison of Performance Environments: Lament Poetry’s Soundscape as a Case Study ...... 91 Eila Stepanova and Frog Abstract: This paper confronts the research tendency to treat a performance tradition as a semiotic phenomenon in isolation from its performance environment. Karelian funeral lament offers an example of a tradition with a customary soundscape that contextualizes and reinforces the performance arena, while performance participates in the soundscapes of additional ritual activities. The case’s analogical value is illustrated through comparison with Old Norse examples.

COMMENTS, PERSPECTIVES AND REPORTS Scandinavian–Finnic Language Contact and Problems of Periodisation ...... 112 Johan Schalin Abstract: This paper explores the periodisation of pre-documentary Scandinavian language and its synchronisation with that of Finnic, the era in focus being the first millennium CE. Stratification of loanwords is discussed both as a means to refine periodisation and as an aim in itself. The problems as concerns the latter half of the first millennium are highlighted. The account is built on dissertation research defended in September 2018. Archaeological and Lexical Relevant Indicators of South Sámi Prehistory ...... 123 Minerva Piha Abstract: This paper complements my model for combining archaeological and lexical material with relevant indicators, a discipline-neutral tool that provides information on cultural processes enabling comparison of research results of different disciplines. It shows how the model is applied to South Sámi material, and how relevant indicators should be used in the data. I suggest that a degree of probability be added to the concept, distinguishing strong and weak relevant indicators. In the Hollow of Tursas? – An Overview of an Archaeological Work-in-Progress in Southwest Finland ...... 131 Sami Raninen Abstract: Since 2012, an archaeological excavation has continued on a Late Iron Age (AD 800–1050) settlement site and a medieval and early modern village plot of Tursiannotko in Pirkkala (near Tampere). The very well-preserved materials found at the site give a lot of information regarding the modes of subsistence, building techniques, material culture, ritual practices and trade contacts of a relatively wealthy farmer and hunter community. Jómsvíkinga saga – Recent Research: Focus on Genre ...... 136 Sirpa Aalto Abstract: This report offers an overview of recent research being conducted on Jómsvíkinga saga. Russian Laments of the Vologda Region in Modern Records: Distinctiveness and Relations to Other Traditions ...... 140 Elena Jugai Abstract: This article introduces Russian laments of the Vologda region in relation to other lament traditions of the Baltic Sea region. The tradition of this region and lament collection there are briefly introduced as well as the Russian Lament database developed and used in connection with the current research. The language and poetics of the laments are outlined followed by a rhythmical analysis and comments on parallelism, alliteration and practices. New Research on Skolt Saami Laments ...... 151 Marko Jouste Abstract: This report briefly presents on evidence of a lament tradition among the Skolt Saami. The lament tradition was recognized when documented by earlier scholars but has never entered discussion of lament traditions and has remained generally unrecognized in discussions of lament traditions in the Baltic Sea region. 4

CONFERENCES AND EVENTS Conference Report – Austmarr VI: Religion, Language, Practice, with a Workshop on Late Iron Age Mortuary Behaviours ...... 154 Kendra Willson Conference Report – Austmarr VII: Crossing Disciplinary Borders in Viking-Age Studies: Problems, Challenges and Solutions ...... 156 Joseph Ryder Conference Announcement – Austmarr IX: Genius loci in the Prehistory of the Baltic Sea Region ...... 160 Vykintas Vaitkevičius Conference Report – Folklore and Old Norse Mythology ...... 161 Jesse Barber Conference Report – Network of Early Career Researchers in Old Norse: Trends and Challenges in Early Career Scholarship Workshop ...... 166 Luke John Murphy Conference Announcement – Methodology in Mythology: Where Does the Study of Old Norse Religion Stand, and Where Can We Go from Here? – The 2019 Aarhus Old Norse Mythology Conference ...... 168 Eldar Heide

RECENT PUBLICATIONS Contacts and Networks in the Baltic Sea Region: Austmarr as a Northern Mare nostrum, ca. 500–1500 AD ...... 169 Maths Bertell, Frog and Kendra Willson (eds.) Atlantic Outlooks on Being at Home: Gaelic Place-Lore and the Construction of a Sense of Place in Medieval Iceland ...... 170 Matthias Egeler Viking Law and Order: Places and Rituals of Assembly in the Medieval North ...... 173 Alexandra Sanmark

PHD DISSERTATION PROJECTS Communicative Registers, Enregisterment and Indexicality in Viena Karelian Kalevala-Metric Poetry (working ) ...... 176 Tuukka Karlsson

MASTER’S THESES Óðinn: A Queer týr? A Study of Óðinn’s Function as a Queer Deity in Iron Age Scandinavia ..... 178 Amy Jefford Franks

CALLS FOR PAPERS The Formula in Oral Poetry and Prose: New Approaches, Models and Interpretations ...... 179

5

S.J. Semple. BAR International Series. Oxford: åren 1319 till 1521: Ett bidrag till den historiska Archaeopress. Pp. 107–119. geografien. 3rd edn. Stockholm: Norstedt. Semple, S.J., & A. Sanmark. 2013. “Assembly in North Ståhle, C.I. (ed.). 1960. Rannsakningarna efter West Europe: Collective Concerns for Early antikviteter, band 1: Uppland Västmanland, Dalarna, Societies?”. European Journal of Archaeology 16(3): Norrland, Finland, I. Stockholm: Kungl. Vitterhets- 518–552. akademien. Semple, S.J., A. Sanmark, N. Mehler & F. Iversen, with Taranger, A. 1924. “Alting og lagting”. Historisk A.T. Skinner, H. Hobæk & M. Ødegaard. Forthcoming. tidsskrift utgitt av den Norske Historiske Forening 5: Negotiating the North: Meeting Places in the Middle 1–45. Ages. The Society for Medieval Archaeology Turén, S. 1939. “Om ‘rätt tingsplats’ enl. Västmannalagen”. Monograph Series. Abingdon: Routledge. Västmanlands fornminnesförenings årsskrift 27: 5–16. Styffe, C.G. 1911. Skandinavien under unionstiden: Med Wildte, F. 1931. “Västergötlands medeltida tingsställen”. särskildt afseende på Sverige och dess förvaltning Rig 14: 174–184.

Thegns in the Social Order of Anglo-Saxon England and Viking-Age Scandinavia: Outlines of a Methodological Reassessment Denis Sukhino-Khomenko, University of Gothenburg

Abstract: The article addresses the possibilities for a methodological reassessment of the phenomenon of the thegns in England and Scandinavia in the late Viking Age (ca. 900–1066). The historiographical overview reveals that the thegns have never been examined for their own sake, and that the recent developments in source studies open new methodological prospects in anthropological research. The case study of the thegns hopes to outline some of them. If the period from ca. 1800 to the 1950s could to give it space in their limited media (be it be described as the ‘age of grand narratives’, to parchment, stone or an oral skaldic stanza). a certain extent promoted through European Understanding categories behind medieval National Romanticism and the search for vernacular terminology and associated texts is ‘origins’, then recent decades might be termed a fundamental research problem for a number ‘a period of revision of received wisdom’. of disciplines. The concept of a as a Calling it, in Thomas Kuhn’s terminology, a social institution characteristic of the early ‘scientific revolution’ or ‘paradigm shift’ is, medieval pre-state tribes, variously termed in perhaps, going too far, but earlier frameworks the (Lindow 1976: 10– are undoubtedly being constantly revised. At 84), reached a virtual research stalemate until the close of the 20th century, there appeared recently reassessed by Petr Stefanovich, who important ad fontes works, many of which in his 2012 book applied it to the Old Russian sought to re-evaluate established terminology lexeme дрѹжина (družina) [‘(military) in order to challenge and clarify existing retinue’] (Stefanovich 2012), proving its concepts of the medieval social order (e.g. on terminological potential (at least in Slavonic the much-debated problem of , see studies). Doubts about the often-postulated Abels 2009). Closely intertwined Nordic, prevalence of blood feud and hence the ‘Germanic’, and Viking Studies followed suit. meaningfulness of the term wergild [lit. ‘man The present article is positioned on the price’] in early Germanic and later trajectory of such re-evaluations. It concentrates Scandinavian societies have been expressed on a particular empirical case that arguably (Sawyer 1987). Even before then, Alexander deserves a reassessment, the case being the Murray, it seems, overthrew the hitherto phenomenon of the thegns (also spelled ‘’ prevailing notions of an agnatic kinship among in Modern English) in late Anglo-Saxon England the early Germanic peoples (Murray 1983). and Viking-Age Scandinavia ( of the The very pages of this journal saw a dispute /Old Norse þegn are also recorded between Rudolf Simek (2010), questioning the in continental Germanic languages). When the concept of vanir as a family of Nordic gods, and available evidence is pieced together, it is Clive Tolley (2011), and Frog and Jonathan impossible to deny the thegns their historical Roper (2011), who critiqued Simek’s scepticism. role: both in England and Scandinavia, literate Many further examples of such debates and people considered this role important enough reassessments could be enumerated – holmgangr and óðal in the Nordic studies, hall

25 and bretwalda in the Anglo-Saxon spring to 112). But who were those ‘western thegns’ mind, to name but a few. Yet despite the who also played an essential role at one of variety of philological, theoretical, and Northern Europe’s crucial points in history by analytical tools brought to bear on these unanimously changing their allegiance and questions, a current, coherent methodology for paving the way for a new, albeit short, English addressing issues of this type remains wanting. dynasty that in the closing years before the It seems that academic discourse currently arguably reoriented the displays a tendency to inadvertently turn Anglo-Saxon culture and politics towards the certain historical lexemes, such as družina, North, imparting them with a ‘Nordic’ flavour? barbari, barones, víkingar, væringjar, or When answering this question, most indeed þegnas/þegnar, into somewhat ideal, anglophone authors rely on a contemporary self-sufficient concepts, often as analytical text, traditionally called Geþyncðu [‘Ranks’ or constructs, that are then retrospectively ‘Dignities’], which maintains: projected back onto the culture in question. As And gif ceorl geþeah, þæt he hæfde V hida a result, such constructs may converge with the fullice agenes landes, [kycenan], bellan ⁊ actual usage in the sources and receive a status burhgeat, setl ⁊ sundornote on cynges healle, of a ‘common knowledge’. With those þonne wæs he þanon for þegenrihtes wyrðe. problems in sight, the global goal of this paper And se þegen þe geþeah, þæt he þenode is, by bringing a particular case study to the cynge ⁊ his radstæfne rad on his hirede, gif se fore, to suggest a possible methodology with þonne hæfde þegn, þe him filigde, þe to cynges [utware] V hida hæfde ⁊ on cyninges which those may be overcome, or at least sele his hlaforde þenode ⁊ þriwa mid his compensated for. Readers are kindly asked to ærendan gefore to cynge, se moste siððan bear in mind, though, that before them is only mid his foraðe his hlaford aspelian ⁊ his an illustration of this methodology in ongoing onspæce geræcan mid rihte, swa hwar swa he research, rather than final historical þorfte. (Geþyncðu, §2–3: Liebermann 1903: conclusions about the thegns. 456; Rabin 2015: 68–69.) [...] if a layman prospered so that he had fully Sketching Thegns’ Presence in the Sources five hides1 of his own property with a church Thegns’ importance in both England and and kitchen, a bell-house and fortified gate, a Scandinavia is difficult to overestimate and is seat and an appointed role in the ’s hall, highlighted by an abundance of sources then he was worthy of a ’s rights ever after. And if a thegn prospered so that he referring to them, and the role thegns play waited upon the king and rode on his business therein. Thus, reporting the nadir of the Old among his retinue; then, if he had a thegn who English monarchy, the Anglo-Saxon Chronicle followed him, who had five hides for the complains: king’s service, and had waited upon his lord in the king’s hall, and had gone three times on ⁊ com Æþelmær þyder ⁊ þa his business to the king, then [his thegn] westernan þegenas mid him, ⁊ bugon ealle to afterwards might represent his lord in various Swegene ⁊ hi gisludon. Þa he ðus gefaren obligations with his initial oath and handle his hæfde, wende þa norðweard to his scipum ⁊ litigation, wherever he must. eal þeodscype hine hæfde þa for fulne cyng [...] (ASE, manuscripts C, D, E: s.a. 1013 AD; Such a clear-cut explanation is rather appealing Whitelock 1979: 246.) to those modern encyclopaedic views that Ealdorman Æthelmær came there, and with favour monosemantic definitions. However, it him the western thegns, and all submitted to poses a great many problems: no instance of Swein, and they gave him hostages. When he such upward social mobility is known for sure; had fared thus, he then turned northward to this is but one unique source’s testimony, and his ships, and all the nation regarded him as unus testis non testis est [‘one witness is not a full king [...] (Unless otherwise indicated, all witness’]; the author of this text was translations are my own.) established in 1950 to be Wulfstan, archbishop Few doubts exist concerning the position of a of York (d. 1023), a famous homilist and late Anglo-Saxon ealdorman, a provincial civil political figure with a markedly strong interest governor and military leader, appointed by a in the moral ‘deterioration’ of his age; king (see e.g. Molyneaux, 2015: 66–67; 111– Geþyncðu’s genre is hardly legislative but

26 rather polemic (see a further elaboration and skaldic verse, for almost a century, while below) – to name but a few concerns. Moreover, agreeing on the thegns’ local prominence, Wulfstan’s statement contradicts other sources scholars have been arguing about whether that either undermine thegns’ connections with Scandinavian thegns bore any similarity to the monarchy or directly oppose them to it their Anglo-Saxon counterparts and served in (such as in the quote from the ASE, where the ’ military retinues or assumed the status thegns abandon their supposed lord). At first of provincial aristocrats / petty chiefs / clan glance, Wulfstan’s vivid description finds leaders. To this must be added the much good support in at least 188 diplomas from the weightier arguments archaeologists wield in years 900–1066 that attest land grants to the Scandinavian academia when it comes to the thegns (see below):2 faithful servants to kings, pre-medieval period. The most recent who sit on their councils, get rewarded with examination of the matter was undertaken estates, where they surely take over existing within an archaeological survey (see below) manor complexes or build new ones. But if one and aimed at reconciling the conflicting views: looks beyond the West Saxon perspective, a Danish and West Swedish late 10th-century different picture emerges. It has not escaped magnates voluntarily chose to join Kings Sweyn scholars’ attention that the Domesday Book and Cnut’s raids in England, where they adopted records hundreds of taini (a Latin rendering of the of thegns (though not the lexeme thegn) – according to Michael Costen’s (2007) itself) and brought them back to Scandinavia. calculations, about 820 (sic!) of them in This was done in line with a long-standing Somerset, Wiltshire, Devonshire, and Dorset tradition of outlining the process of early alone – whose positions seem hardly medieval state building in Scandinavia, in distinguishable from those of top-tier – which thegns have frequently played one of the something bizarre for the accepted pivotal roles, being interpreted as the agents monosemantic understanding of the term. and officers of the new political units that were Thanks in large part to the multicultural forming. background, local differences might equally The attention in the sources as well as the worry an aspiring student: thus, the so-called supposed level of influence thegns’ probably Regulation of Thegns’ Guild in Cambridge exerted within Anglo-Scandinavian society/-ies (Whitelock 1979: 603–605) meticulously warrant us to apply to these people a modern describes the circumstances in which blood sociological concept of a [lay] elite (< Fr. élite feud should ensue, something strictly prohibited [lit. ‘chosen person’]) or an echelon thereof. For by the royal legislation of the period.3 the purpose of the present discussion, I engage Compared to England, far fewer individual the definition of an elite adopted by the French thegns are known east of the North Sea, yet historians Laurent Feller and Régine Le Jan: their regional prominence, reflected in 46 th th tous ceux qui jouissent d’une position sociale known 10 –11 -century Danish and Swedish élevée [...] [ce qui signifie] la détention de la commemorative runic inscriptions, is still clear fortune, du pouvoir et du savoir ainsi que la in the landscape today. In Södermanland, for reconnaissance par autrui (Feller 2006: 6). instance, one reads: all those who enjoy a high social position [...] styrlaugʀ : auk : hulmbʀ : staina : raistu : [which means] the possession of wealth, at : bryþr : sina : brau(t)u : nesta : þaiʀ : power, and knowledge as well as recognition entaþus : i : austruiki : þurkil : auk by others. sturbiarn þiaknaʀ : kuþiʀ (Sö 34.) Unlike original native terms like nobilitas, Styrlaugr and Holmr raised the stones next to proceres, optimates, æþelboren or góðir menn, the path in memory of their brothers. They which could at times demonstrate a judgemental met their end on the eastern route (Austrvegr, character, this word elite’s emotional and moral i.e. in Russia), Thorkell and Styrbjörn, good neutrality gives it flexibility and scientific thegns. applicability. Perhaps, because a Geþyncðu-like text is Interpreting the thegns’ position within the absent from the Old Norse corpus and due to social environment has always been regarded the vagueness of the relevant runic material as a tool to paint an intricate portrait of early 27 medieval communities in England and across lexis in a given language or a group of related the North Sea. Among the many questions, languages (see e.g. Lindow 1976), but also understanding the thegns’ standing in the linguistic loans, which in turn can help to form socio-political orders allows us to broach the a dynamic view on contact patterns or subject of power distribution in Anglo- mechanisms, social systems and their variation Scandinavian polities. The debate in the throughout recorded history. Scandinavian historical community, briefly alluded to above, serves as a prime example. If The State of Research one agrees with proponents (mostly historians Before proceeding to the methodological case and archaeologists) of Svend Aakjær’s idea study, I will briefly review the extant that, in the runic inscriptions’ thegns, we see scholarship on the topic, highlighting some of kings’ retainers, then one advocates a top- its inner controversies that, I believe, advocate a down power structure in which rulers establish reassessment. Furthermore, since thegns have, their control by asserting their presence as far as I have been able to explore the through dependent agents. Should one take the relevant works, not been researched for their side of his critics (mainly runologists and own sake, there currently is no similar philologists), a bottom-up hierarchy emerges, comprehensive historiographical overview. where the power resides with the wealthy local Naturally, it is hardly possible to cover all ‘clan’ leaders. This is just as relevant for the relevant monographs, book chapters and Anglo-Saxon situation. On the one hand, articles in just one paper, so only the most historians working on the development of important ‘milestone’ works shall be mentioned kingship normally treat the thegns from the below. Because of the methodological character charters as a certain extension of the court in of the current article, multiple related texts had the provinces (e.g. Larson 1904: 103); on the to be left out for the sake of brevity. Paralleling other, Domesday scholars emphasise the thegnly the geographical distribution of the source class’ territorialisation and embeddedness in material, the following historiographical sketch the social fabric (e.g. Hollister 1962: 64–65). But is organized through the developments of the important role of the thegns and their place discussions in scholarships of different in society were studied in the aforementioned languages. ‘age of grand narratives’, and the subsequent progress in source criticism, methodology, and Anglophone Historiography overall discourse calls for revisiting the The first British ‘scientific’ examination of the received wisdom, as well as bringing the thegnly phenomenon perhaps began with the national and disciplinary foci of research into a opening volume of the Constitutional History dialogue. of England by William Stubbs (1874), where In the narrow sense, this paper aims at the word thegn appears a minimum of 160 outlining a feasible methodological solution to times. Stubbs’ description largely stems from this case study, as well as its historiographical Wulfstan’s treatises. In a nutshell, his view is background and state of current research. that thegns used to be warrior members of the Besides casting new light on the social and royal comitatus, but that this group split by political arrangements of the Viking Age and about the 930s: some of its members absorbed their interconnectedness, the suggested approach the upper crust of the free population, while may contribute to scholarly knowledge by others consumed the former aristocracy by finding a wider application in (pre-)medieval blood and became the new ‘’ : studies and, possibly, beyond. However, on a On the one hand the name is given to all who larger scale, the suggested solution might also, possess the proper quantity of land whether where sufficiently large corpora exist, provide or no [sic] they stand in the old relation to the an applicable tool for investigating socially king; on the other the remains of the older constructed categories linked to historical nobility place themselves in the king’s terminology in other languages and periods. service. (Stubbs 1874: 156.) This application of the methodology can be For a commoner to become ‘thegn-worthy’, he extended to not only the stock of an indigenous needed to “acquire five hides of land, and a

28 special appointment in the king’s hall, with received universal acceptance. It was, other judicial rights” (Stubbs 1874: 155). The nevertheless, adopted by one of the leading idea of social mobility is of prime importance Domesday scholars, David Roffe, who, to Stubbs: “[...] there is no impassable barrier contrary to Reid’s focus on the judicial between the classes: the ceorl4 may become privileges, spotlights the bocland-type of land thegn-worthy, and the thegn, eorl-worthy” (ibid.: tenure (Roffe 1989; 2000: 28–46). Roffe’s 81), and he briefly mentions that a thegn’s ideas have seen some support (Reynolds wergild was 1,200 shillings (ibid.: 157). 1992), but his and Reid’s rather dogmatic Stubbs’ condensed overview has probably assertions have been sharply criticised both by served as the primary point of departure for students of the law (Hudson 2012: 58–62) and generations of English-speaking historians.5 by specialists in lordship (Baxter 2001; Baxter For example, whereas Stubbs was content to & Blair 2006). mention thegn’s 1,200-shilling wergild in just Sir Frank Merry Stenton developed his one sentence, when Henry Chadwick wrote his predecessor’s connection of the post-Conquest Studies on Anglo-Saxon Institutions (1905), he feudal to the Domesday taini (an idea dedicated almost 40 pages to proving this was much later promoted by John Gillingham, statement arithmetically (Chadwick 1905: 76– esp. 1995) and Norman – with the royal 114), and ever since then this idea has acquired ministri, though both strata enjoyed a 1,200- the status of a common knowledge. In order to shilling wergild “irrespective of the duties prove these claims, Chadwick came up with which he might happen to owe to his lord” “one of the most insightful and influential (Stenton 1932: 130): essays on the Anglo-Saxon monetary system” It is more than probable that many thegns of (Naismith 2015: 143) that served as a “prelude the eleventh century were country gentlemen, to the main business, which was using the data with no special aptitude for war. In most in law codes to shed light on aspects of Anglo- cases, the estates of a thegn of 1066 must Saxon society” (ibid: 146). At the same time, have come to him by inheritance, and not by Chadwick also laid the foundation for the the gift of a king or any other lord. But his tradition of studying kings’ thegns from the obligation to military service represented the charter evidence, and was echoed by his ancient duty of attending a lord in battle. American contemporary Laurence Larson (Stenton 1932: 119.) (1904). Following Stubbs, Stenton further explained Medievalist Rachel Reid (1920) joined the the differences between a thegn and a ceorl discussion and offered an original view on the also through their relations to military service: problem at a time when the ongoing debate had thegns served in the army due to their rank, been developing for a few dacades surrounding whereas ceorls reported to the army due to the continuity/discontinuity between the alleged old Germanic custom (Stenton 1932: Anglo-Saxon and Norman periods. In a 38- 116–118). page article, she brought extensive Anglo- Three decades after Stenton’s monograph, Saxon, Norman, and Scottish sources from Hollister, working with the Wulfstanian over a millennium and postulated that kings’ texts and the Domesday Book, opposed part of thegns were officials endowed with the judicial his interpretation that concerned marshal rights of sake and soke, toll and team, and duties (Hollister 1962). He claimed that, in late infangenetheof 6 “and duties akin to a sheriff’s” pre-Norman England, military obligation (Reid 1920: 172). After 1066 (the Norman stemmed not from one’s rank but depended Conquest), these privileges were appropriated upon the so-called ‘five-hide principle’, which by Norman barons, but the institution’s predicated army service for one man from essence remained the same, so “the Norman every five hides of land, regardless of one’s clerks who identified the king’s thane with the status. Characteristic of many works, Hollister must have had regard chiefly for the one uses the word thegn at least 216 times in his that they had in common” (Reid 1920: 170-page book (an impressive number, given 173). By the end of the 20th century, this idea that it is not his main focus), but not once does of institutional continuity, it seems, had not he actually explain what a thegn is. And the

29 notion of a Germanic ‘nation in arms’, inherited link’. That at least some kings’ thegns, as from Stenton7 and his Victorian precursors,8 attested by the charters, were members of the was later largely debunked by Richard Abels, social elite cannot be denied. Neither can the who, elaborating on the preceding work of Eric Domesday evidence of the numerous petty John (1960), conclusively argued that Anglo- thegns be discarded. But how does one bridge Saxon military institutions had always been the gap between these enormously different rooted in lordship and land tenure (Abels groups? Viewing them as strata of one and the 1988). same social group requires a further A major breakthrough came in an article by explanation. Hollister’s generous equation of Henry Loyn, who applied philological methods the Domesday taini with the peasantry, in to the Latin–Old English translations of the late contrast to Senton’s identification of the 800s and tackled the semantic evolution of the former with the gentry, also inadvertently term thegn in Old English: reaches a theoretical deadlock: if thegns and [...] as late as the end of the ninth century, a ceorls alike performed military duties on the thegn was still, customarily, one who served same principle, why did the former allegedly a lord in a personal capacity. By the time of enjoy a more advantageous social position? the Norman Conquest eorls and thegns were The ‘whiggish’ notion of upward ceorl-to- the two recognized social divisions between thegn social mobility hinges on a somewhat freeman and king. (Loyn 1955: 543.) problematic source. As already mentioned, its In Britain, the most recent assessing of the oft-reiterated form stems from but one thegnly phenomenon was undertaken by Ann authority: Archbishop Wulfstan of York. Williams (2008). Unlike many, she dedicates Sadly, in assessing Geþyncðu, its first modern the whole of her nine-page introductory publisher, Felix Liebermann, might have done chapter to setting the scene and establishing the a great disservice to subsequent research by position of a thegn. Her study may be seen as putting Geþyncðu in his edition of the Anglo- emblematic of the British historiographical Saxon laws, presumably on the basis of its empirical tradition of dealing with the subject. opening phrase (Liebermann 1903: 456–458). For one thing, Williams would rather describe More and more, it has been realised, to quote a thegn than suggest a definition, and, for Andrew Rabin, that “many of the practices it another, she would restate some of the describes are unsupported by contemporary established views: the division of the free evidence” (Rabin 2015: 67). It might be argued populace into ceorlas and þegnas, and upward that a very similar passage can be found in an mobility by virtue of land ownership and actual legal code, Norðleoda laga [‘The Laws possession of military ammunition (Williams of the Northern People (presumably the 2008: 2, 4). Northumbrians)’]. The problem is that this Despite this well-developed scholarship, legal code was also edited by Wulfstan and there persist some contradictions that, it may therefore cannot be treated as wholly be argued, are difficult to eliminate within the independent evidence. Wulfstan’s relation to prevailing paradigm. The historiographic these documents is also why it took Chadwick ‘elephant in the room’ is the apparent lack of almost 40 pages to prove the prevalence of a synthesis. Preceding paragraphs exemplify 1,200-shilling thegnly wergild: though such only some of the various terms employed to rate is attested in the laws on multiple describe the phenomenon of thegns (aristocracy, occasions, only one legislative piece actually retinue/comitatus, vassal, class, nobility, connects it with the thegns – the so-called official, et al.), as well as the multitude of Mircna laga [‘The Law of the Mercians’], approaches (through comitatus, military history, which is also attributed to Wulfstan (Rabin land tenure, legal status, Domesday social 2015: 71). Even after Wulfstan’s authorship patterns, etc.). Such variance clearly stems from has been established, these texts are normally discrepancy between the sources, but it seems the ‘go-to’ reference material in general works that no coherent mechanism of explaining such (e.g. Molyneaux 2011: 266–267; O’Brien dissimilarity has been suggested, save the oft- 2011: 86). The problem with such a point of repeated claim for wergild as the ‘missing departure is twofold: a) even within the Viking

30

Age proper, these are somewhat late texts; b) his ground-breaking article. He drew attention one author’s testimony should not be to the frequent appearance of the Old Norse uncritically taken as historically accurate, þegn and drengr in runic commemorative especially in view of Wulfstan’s strong inscriptions throughout the “old Danish king- personal interest in orderly societal dom”, and concluded that, “though the meaning organisation and the turbulent epoch in which of these various expressions shows through but he was writing. To polemise: Wulfstan’s vaguely, there seems nevertheless little reason assertions might well not be as representative as to believe that they should only stand for ‘man’ one would wish, and potentially a politically- pure and simple” (Aakjær 1927: 9). Aakjær driven construct. As Ryan Lavelle said of postulated that thegns and drengs were Wulfstan, “social mobility was a long-standing members of the kings’ comitatus, known in concern for conservative churchmen” (Lavelle Scandinavia as hirð. Noticing that Old English 2010: 63). Naturally, nothing of this was also featured words thegn and dreng (the latter known at the time of Stubbs and Liebermann, being a loanword from Old Norse), he extended and so these developments in source criticism their meanings, borrowing them from Stubbs’ call all the more for a comprehensive revision and Chadwick’s interpretations. Curiously, of the common interpretations of Anglo-Saxon though Aakjær never mentions Reid’s article, social practice, especially given the scepticism his comparative methodology is very similar; on the function of wergild some scholars have another parallel between the two authors is that expressed (Sawyer 1987). neither was a specialist in the early medieval In sum, very seldom have the Anglo-Saxon period, and both specialised in the High thegns been the direct interest of academic Middle Ages. inquiries, often being reduced to a subsidiary Not everyone has agreed with Aakjær’s role in one theoretical argument or another. interpretation. His greatest opponent on this This shortcoming sometimes reads a bit like a matter, the Danish runologist Karl Martin Picasso – the same object has been addressed Nielsen, pointed out that, looking at the runic in fragmented pieces which occasionally material as it is while not having comitatus in contradict each other to the point that they mind, it is impossible to render the terms in become incompatible. Surprisingly, Anglophone question as the kings’ retainers. Moreover, scholarship sometimes reveals little Nielsen and his German colleague Hans Kuhn knowledge of the occurrence of the lexeme argued forcefully that the empirical observations þegn in Old Norse in the same period, and this on skaldic poetry (the earliest datable Old may be characteristic of its prevalent “interests Norse coherent texts) do not support Aakjær’s in England’s particularities”, in the words of interpretation. Thus, Kuhn found that only six Ryan Lavelle (2010: 64). of the 72 examples he identified of the lexeme þegn in skaldic verse were used as a Scandinavian Historiography Rangbezeichnung [‘designation of rank’], In Scandinavian academia, the discourse has distinguished as a combination of a noun for been very different but no less intricate. For a ‘man’/‘warrior’ (maðr, rekkr, þegn and drengr) while, scholars showed little interest in the with a possessive pronoun or genetivus lexeme þegn in Old Norse, and treated it just possesivus of another noun rather than forming like any other entry in a dictionary. For a kenning (Kuhn 1944: 105–106, 110–111). example, when analysing Norwegian laws in Nielsen conceded that drengs could have 1890, legal historian Ebbe Hertzberg become chieftains’ followers, but this certainly understood it as “[...] fri og uafhængig udøver was not the case by default, whereas the þegnar af alle en fuldtberettiget persons rettigheder” were the ‘backbone’ of the Viking-Age Danish (Hertzberg 1890: 266) [‘a free and independent society, the well-to-do bœndir (Nielsen 1945). practitioner of all rights of a person, fully Today, however, Martin Syrett characterises vested with liberties’], deriving this meaning such “notions of a free independent class of from the oldest Norwegian law codes farmer-chieftains” as “outdated”, and calls the (Gulaþingslǫg and Frostaþingslǫg). “speculation concerning the independent status Everything changed when, in 1927, Danish historian and archivist Svend Aakjær published 31 of the late Viking-Age freeman” romantic . In a 1980 monograph, Klavs Randsborg (Syrett 1998: 249, 252). interpreted runic þegns as agents of the Jelling By and large, these two standpoints still kings who granted ‘fiefs’ to their in hold in the Scandinavian historical narrative up exchange for various services. Rune stones in till now. Scholars with a thorough philological his explanation in a way served as titles to land, background hold fast to the Norse literary since such land-tenure praxes were hitherto sources that do not support Aakjær’s claims. absent from the Scandinavian societal practice Such, for example, was the opinion of John and required further support (Randsborg 1980: Lindow (1976), who agreed with Kuhn that in 29–44). To Randsborg, this is an important the Viking Age the word þegn did not convey point in his argument for the state formation in the sense of a social rank. Today, this view is Viking-Age : the elite position of the supported by Judith Jesch (e.g. 2013) in her king’s followers in the localities speaks for a comparison of the runic inscriptions and the strong central power. skaldic poetry. On the other hand, some In settling the opposing views, Carl historians have seen Aakjær’s argument as Löfving, a Swedish archaeologist and lawyer, conclusive and supported it from a theoretical took Randsborg’s views further. One of the perspective. Niels Lund, for instance, aimed at main arguments of his doctoral thesis (2001), a reconciling the two opinions, suggesting that product of nearly two decades of research, is “ leaders, such as Alle at Glavendrup that, at the turn of the millennium, Götaland [...] with an armed force of their own” had to was governed and influenced not by a ruler acknowledge the growing power of the Danish from Uppsala but by the Danish king, who , thus becoming, “at least in theory, from 1018 to 1035 was King Cnut. Bearing in members of his and, thus, his thegns,” mind that he was also the English monarch, even if they never were his retainers in the and in England (including in Cnut’s own laws) narrow sense (Lund & Hørby 1980: 62). It has kings’ followers were known as thegns, Löfving also been pointed out that philologists generally maintains that this interpretation should also be tend to overlook place-name evidence: apart accepted for the Scandinavian toponymical from two instances in south-eastern , and runic material (Löfving 1984; 2001: 79– Sweden has preserved a minimum of 12 102). The rune stones’ þegn inscriptions, in toponyms Tegneby/Tängby/ Tägneby (from this explanation, therefore mark both the Old Swedish ÞægnabyR) (Elgqvist 1947: 113), authority the Danish monarchy wielded and which can hardly mean ‘settlement of the social support it could recruit. One has to men/people’ (Strid 1986: 305). This dichotomy give Löfving credit as apparently the first to try between the philological and historical to put those rune stones in their proper approaches was first noticed already by Martin historical context. Peter Sawyer (1988: 34), Syrett (1998: 249). describing politogenesis in Sweden, followed Some prominent runologists joined Aakjær’s in Löfving’s footsteps and also identified the stance. Erik Moltke, a towering authority on runic þegns as ‘under lordship’ of Cnut. The the Danish runic material, wrote that since the idea of borrowing the sense of a title for the earliest appearance of the word þegn in runic pre-existing lexeme þegn from Old English inscriptions was: was, however, briefly criticised as unlikely by associated with liþ, host, warband or the like, Eric Christiansen, who pointed out that, “to we may reasonably assume it denoted a kind Nordic intruders, they [thegns] can only have of military status. Thegn is then a title of rank appeared as local bosses, district defenders; [....] Private individuals do not seem to have which is how they appear in the skaldic verse had thegns, so it must have been the ruler’s of ca. 1030 onwards [...] not [as] officials or prerogative to appoint thegns (and certain royal retainers, but sometimes of their drengs). We thus come to the same opponents” (Christiansen 2002: 335).9 conclusion as Svend Aakjær – or not far off Closely resembling the situation in Britain, it. (Moltke 1985: 285–286 [1976: 235–236].) the debate between Aakjær’s opponents and Moltke’s expertise has been keenly accepted supporters has to a certain extent overlooked by some archaeologists both in Denmark and the source work. A few examples will suffice.

32

Aakjær, as one would expect, relied on (Fritzner 1867: 774) without referring to or secondary English works, but even his most checking it, which, granted, was standard dedicated challengers have not noticed that a practice of scholarship at the time.11 These weighty chunk of his argument was invalid faults, of course, do not undermine the already in 1927. To substantiate his point, historical importance of his article: after all, he Aakjær had to prove that drengs were a was the first to raise the question of how Old subclass of the Anglo-Saxon thegns, so he English and Old Norse handle the same social retreated to a text known as Constitutiones de term, and an allowance for the state of the Foresta [‘Forest Regulations’], which he methodology at the time must be made. interpreted as the forest law of King Cnut and To recap, the heated discussion has, to a combined this with the evidence from 13th- large extent, not revolved around the century Northumbrian charters as retold by interpretation of the sources per se but around Frederic Maitland. However, Constitutiones de a retelling of an interpretation. Note that in Foresta was written in the 12th century by an various discourses (Lund, Löfving, Randsborg, anonymous Norman clerk who probably did Sawyer), mirroring the British counterparts, not even speak English and has nothing to do þegns themselves were not the object of study with Cnut (Harris 2014).10 Even though Felix but served as yet another methodological Liebermann pointed this out as early as 1894, element in the general sketch of early medieval Aakjær mentions Constitutiones de Foresta as state building in the North. Martin Syrett was a reliable source in his last published piece probably the first to subtly identify the likely (Aakjær 1962) and thus was likely unaware of stimulus for such a persistence: the argument. In another case, Aakjær’s 1927 That historical approaches have tended to article takes an erroneous reference to an link the thegns and drengs of the runic original source for granted: he defines inscriptions with the growth of a royally þegngildi as the “fine paid to the for sanctioned aristocracy derives largely from having killed his thegn (a free man in the the necessity of positing some royal officers king’s service),” and refers to Eiðsivaþingslǫg somewhere to account for the development of 1, 28 [‘The Law of Eidsivathing’] (1927: 11). the Danish state in the tenth and eleventh However, the paragraph in the actual source centuries. As Peter Sawyer put it, ‘kings must runs as follows: have had agents ... not only to lead local defences but also to gather royal resources’. En ef þionn mannz etr kiot i langu fastu. þa er (Syrett 1998: 268.) hann utlægr. oc skal uera i uallde skapdrottens hans er a. huart hann uill løysa Aakjær’s explanation was simply too good for hann undan þui giallde. halft þægngilldi the “state-formation addicts”, as Christiansen kononge. oc kaupa honum sua larønzuist. eða (2002: 335) pejoratively called the disciples of hit ælligr. at hann fare af larønde brot. this school of thought, to subject it to a critical (Eiðsivaþingslǫg 1, 28; Halvorsen & Rindal source-study test: theory prevailed, the cart 2008: 28.) was put before the horse. And if a man’s slave/servant eats meat during All these circumstances once again call for the long fast, then he is an outlaw, and his a methodological return ad fontes and predicate lord, who owns him, should decide whether the necessity for a reassessment. he wishes to release him from punishment [by paying] half a þegngildi to the king and Methodological and Source Overview [thereby] buying him the right to stay in the To break free from at least some of the country, or lets him leave the country. constructions of the discourse described above, Though Aakjær’s definition of þegngildi is a wide range of options are available. In my generally correct, Eiðsivaþingslǫg 1, 28 reveals opinion, the three cornerstones among them little about the nature of a þegngildi and certainly are as follows: says nothing about murder exculpation. My 1. Context analysis and methods from Corpus initial suspicion of fact juggling proved Linguistics incorrect: Aakjær, it turns out, copied this 2. A discriminant approach, chronological passage from Johan Fritzner’s dictionary analysis, particular attention to inner features

33

3. Approaching the historical reality behind a following section. The linguistic presupposition fleshed-out concept for contextual analysis seems to reside with an It is my belief that such methodological enquiry into the etymology of the lexeme thegn devices have a wider range of application in in order to conceivably reconstruct its possible historical examination of actual social Proto-Germanic semantics. Thegn and its phenomena, and in fact elements of this triad cognates are present not only in Old English have indeed been used in previous research in and Old Norse, but also in one way or another. For instance, context and Old Saxon, as well as in Scots (though here 12 analysis was used by Lindow when he went it is almost certainly a borrowing). Knowledge through most of the relevant skaldic stanzas of a word’s origin may help us understand the and then generalised his conclusions. An subsequent development of the term in the example of an approach through corpus different societies in historical times, though it is linguistics is observable in an article by Jane worth bearing in mind that such reconstructions Roberts (2000), in which she searched the always remain hypothetical. Consequently, Thesaurus of Old English, a massive project etymology per se should not be used as an hosted by King’s College London and the absolute tool in historical research, as is University of Glasgow, for Old English words warned by Hans Kuhn (1944: 120). for nobility, and she analysed their usage in the The general methodological premise of the language. Similar methodology was employed subsequent study should manifest itself in a by Petr Stefanovich in his reassessment of the fundamental alteration of the hierarchy of the Old Russian družina: instead of relying on the research procedures. Contrary to previous most telling evidence, he meticulously went approaches, rather than taking for granted the through nearly all extant and available source definition of the lexeme þegn as used in current evidence, organized by chronology, region and academic discourse, an investigation that seeks genre. The major but essential advancement to bring a historical phenomenon behind this proposed here lies in the ‘triangulation’ of word into focus should first and foremost these methods, i.e. bringing elements together critically review the contexts of the lexeme’s in one investigation. Combining these occurrence in the sources. Only then may techniques is aimed at a multidimensional definitions or concepts arrived at inductively be presentation: we will probably be able to brought back into dialogue with conventional follow the evolution of both an actual social views in scholarship and the sources which phenomenon and its reflection in a given have dominated those definitions. language, thus abandoning the ‘synthetic’ With Old English, the need for such an picture. In doing so, I hope my undertaking can approach has already been alluded to above: serve as a sui generis case study, and, should it because Wulfstan’s Geþyncðu and similar prove to be successful, this methodology could texts offer, at first sight, a very straight- receive a wider application in the broader forward definition of what an Anglo-Saxon anthropological studies. I do not propose that thegn was, modern explanations follow that applying this methodology will realise the definition closely. Nevertheless, an assumption Rankian dictum about reconstructing ‘how it that all uses of a word by all users in all really happened’ (wie es eigentlich gewesen), contexts will mean the same thing or refer to much cherished in the positivistic school of the same category contradicts both common thought. At best, the methodology enables us sense and current sociolinguistic theory. It has to approach the reflection of ‘how it really not escaped historians’ attention that the word happened’ as preserved in the extant sources. thegn can describe different social realities with observable gaps between them: Context Analysis Among the thegns, at one end of the scale, The first cornerstone on the list above is were men who possessed estates in many context analysis, i.e. examination of the shires acquired through generations of royal environment to which an object of study service, and, at the other, were men belongs, coupled with quantitative methods indistinguishable from land-holding freemen from Corpus Linguistics, introduced in the

34

except by their rank. (Barlow 1988: 6; cf. untranslated. The other is that it is not always Reid 1920: 170.) clear which connotation is applicable in each The reasons for such discrepancies are not hard particular case. Some instances are transparent, to come by and include such factors as regional for example in the Norwegian alliterative legal peculiarities in both language and social formula þegn ok þræll [‘thegn and ’] conditions, social progress through history, þegn stands in an opposition to a þræll potential variations in register and/or type of [‘slave’] and should be interpreted as ‘free- 14 discourse, etc. born’. For instance, Gulaþingslǫg [‘The Law What is proposed instead is that we turn to of Gulaþing’] asserts compensation for injuries less loaded sources, using context analysis to for both the ‘free’ and ‘slaves’: deduce one or multiple definitions and to trace Aller eigu sarbøtr iamnar þegn oc þræll. Nu the changes these definitions had most likely ef maðr Særer þræl mannz. þa scal hann undergone in relation to regional peculiarities hallda upp føðzlo við hann meðan hann liggr and chronological evolution. Should the results í sarom. oc verclaunum ollum við drotten more or less match Wulfstan’s formula, proving hans. oc lækníngar kaupi. (Gulaþingslǫg 215; its validity, the archbishop’s description can be Eithun & Rindal & Ulset 1994: 129.) integrated into the case, though every unique All have equal right to compensation for detail therein should still be taken with a pinch wounds, a thrall, as well as a thegn. If a man of salt.13 wounds another man’s thrall, he shall provide Unlike in England, Old Norse authors did victuals for him as long as he lies wounded; not leave us detailed definitions of who was he shall also [pay] leech money and considered a þegn in their society. The only compensation to the master for loss of labor. (Larson 1935: 149.) definition-like stance is to be found in the early-13th-century Skáldskaparmál [‘Language On the other hand, however, in skaldic poetry, of Poetic Art’] section of The Prose Edda, due to its metrical constraints and linguistic attributed to Snorri Sturluson, where it is stated registers, the meaning can be rather elusive, that Þegnar ok hǫlðar, svá eru búendr kallaðir since it is part of the poetic equivalence (Skáldskaparmál 81; Faulkes 2005: 106) vocabulary rather than used to distinguish a [‘Freeholders are called Thanes and Yeomen’ social category per se. Granted, types of (Brodeur 1916: 234)]. This laconic and hardly equivalence category and associated construc- illuminating phrase being the only relevant tions (cf. Kuhn’s concept of a Rangbezechnung) historical definition, scholars have often could potentially offer insights into aspects of tended to take instead the interpretation of Old a word’s significance: patterns of verse use Norse dictionaries as the point of departure for likely reflected the perception of the category, subsequent investigations (e.g. see: Aakjær to which the lexeme þegn pertained (at least in 1927: 4; 16; Nielsen 1945: 112; Strid 1986: the period of the genesis of the skaldic art). 301–302; Syrett 1998: 247–249; et al.). The Finally, though it has been noted early on that, dictionaries’ renditions could be summarised in Old Norse, þegn can denote both a male as follows (in order of frequency) (see also person in general and a monarch’s subject in Fritzner 1867: 774; Cleasby & Vigfusson particular, few attempts have been undertaken 1874: 732; Jónsson 1931: 637): to explain this development, especially since neither of these meanings are attested in other 1. A free-born man, man in general 2. A monarch’s subject Germanic languages. 3. A husbandman, good man Corpus Methods 4. A lord’s servant In order to avoid at least some of the theoretical Unfortunately, two hidden dangers can easily perils and offer a possibly new reading, methods delude a scholar. One is that none of these of Corpus Linguistics are proposed. Corpus dictionaries actually use the much earlier runic Linguistics is an approach developed in the late material, which is probably why, in different 1960s and made ever more efficient today due editions of the runic inscriptions, þegn is to advances in computer software and mass rendered as ‘man’, ‘warrior’, ‘’ or left digitalisation. The underlying premise of this

35 field is that large-scale analysis of a given database of all known (and currently edited) language’s collected corpus should (at least in skaldic poetry that began as a digital theory) minimise a researcher’s bias or workspace for editors of the Skaldic Poetry of interference. the Scandinavian Middle Ages publication Previously, scholars researching the early series (2009–present). medieval period lacked such corpora for their These databases enable research to reach a respective ancient languages. To borrow the qualitatively new level, allowing an in-depth words of Judith Jesch: search and providing a catalogue of all Most dictionary-makers, whether dealing mentions of a given word throughout the with living or dead languages, have an respective corpora. enormous body of material on which to base The initial search of the DOE, ONP, and their definitions, and have to be selective. [...] The Skaldic Database17 has yielded the Historical dictionaries can further restrict the following numbers: in Old English before ca. material through the sources they use [...] 1150, the word thegn appears 1,793 times, with (Jesch 2013: 78.) its related and compound lexemes including a Research of primary sources made a great leap further 314 entries (Table 1). The figures for forward at the end of the 20th century and Old West Norse before ca. 1550 are 237 and beginning of the 21st: devoted teams of 137 respectively, to which are added the results linguists and philologists have compiled such for the Old East Norse languages (Old 18 corpora for both Old English and Old Norse Danish and Old Swedish), not indexed in the that have been digitised and made them databases above: 4 and 9, 21 and 3 respectively accessible to the general public: (Table 2). Finally, supplementary data have been taken from the texts in continental  Dictionary of Old English (DOE) is an on- Germanic languages, Old High German and going project, first conceived back in 1968 Old Saxon, which feature at least 73 and 77 and developed by the Centre for Medieval words (theg(a)n/deg(a)n and their Studies at the University of Toronto. As of derivatives) respectively (Table 3), predomin- 2017, the entries for letters A–G have been th 19 published (letter H is at present available antly from the 9 century. The relevance of online only). Though the dictionary itself is continental lies, first of all, in a general far from completion: as of 2009, the team linguistic comparison, and, secondly, it released the DOE Web Corpus which is a provides additional material for substan- database of “at least one copy of every tiating the etymology of the lexeme þegn, surviving Old English text”; “as such, the since these texts are fairly early yet DOE Web Corpus represents over three seemingly independent of either Old Norse or million words of Old English and fewer than Old English usage. However, despite their a million words of Latin” (DOE). This allows number, their value for the sheer historical researchers to conduct lexical surveys already analysis of the feasible social relationships is now. largely undermined by the religious and poetic  Ordbog over det norrøne prosasprog – Dictionary of Old Norse Prose (ONP) is a nature of most of these texts. (N.B.: the figures similar enterprise started by the the do not reflect the actual number of unique Arnamagnæan Commission in 1939 and occurrences, as many can be found in one and taken over by the Department of the same text but different manuscripts or Scandinavian Research at the University of fragments thereof). Copenhagen in 2010. The first volume To conclude, a final immense aid is the appeared in 1995, and in the years 2005–2008 Scandinavian Runic-Text Database, created in “all of the unedited dictionary citations [were 1993–1997 by Lennart Elmevik, Lena made] available on the Internet” (ONP), thus Peterson, Henrik Williams et al. of the making possible searches in the entry index Department of Scandinavian Languages at the (though the creators warn that the current University of Uppsala (Peterson 1994). stage of the project is still preliminary).  The Skaldic Project Database, developed by According to their log, the first public version Tarrin Wills and Hannah Burrows at the was launched online in 2001. Today, version University of Aberdeen, is an electronic 3.1 is available for download, with the latest

36

Table 1. The word thegn in the Old English corpus. þegn and its inflected þegn and its inflected Spelling þegn and its Spelling þegn and its forms, compounds and forms, compounds and variant inflected forms variant inflected forms related lexemes related lexemes þegn 849 649 ðegn 646 596 þegen 216 188 ðegen 66 57 þægn 50 50 ðægn 14 9 þægen 6 5 ðægen 3 3 þeign 4 4 ðeign 22 21 þæign 1 – ðæign – – þein(g) 31 26 ðein 4 3 þen 169 161 ðen 26 21 TOTAL 1326 1083 TOTAL 781 710

Table 2. The word þegn in the Old Norse corpus.

‘Old Norse’ (Old Icelandic + Old Norwegian) Old Danish Old Swedish TOTAL Prose Skaldic verse Eddic poetry15 4 (thæghn; 21 (þäghn; ca. þegn 136 (ca. 8816 (ca. 900– 13 (date 253 1432–1511) 1220–1386) 1200–1543) 1300) uncertain) 3 (þiængs gæld, 9 þegn- 136 (ca. 1 *þäghns bani, – (thæghngjald; 149 compounds 1225–1542) (þegnskapr?) brödhtäghn; ca. 1241–1515) 1220–1300) TOTAL 272 89 13 13 24 411

Table 2. The word theg(a)n/deg(a)n in continental Germanic languages. Old High German: Hildebrandslied Old Saxon: Heliand (ca. 800– (ca. 840); Otfrid von Weissenburg (ca. 850); glosses on Prudentius’ TOTAL 860s); Ludwigslied (881); Notker the Psychomachia (ca. 1000s) German (ca. 950–1022) ‘thegan’ 46 30 76 ‘thegen’ 4 – 4 ‘thegn’ – 43 43 ‘degen’ 2 – 2 ‘degan’ 4 – 4 ‘thegan’-compounds 17 (theganheiti, thegankind, 4 (theganlîc, theganscepies, 21 and related lexemes theganlicho, druttthegan) theganskepi, suerdthegan) TOTAL 73 77 150

update being from 19th January 2018. The personal name Þægn in Södermanland and 13 database contains records of at least 46 in Uppland” (Strid 1986: 302),20 and this fact mentions of thegns on the Viking-Age Danish should be further researched in collaboration and Swedish rune stones ca. 970–1050 with the specialists in onomastics. (Table 4). Three exceptions to this chronology exist. The famous Glavendrup stone’s (DR Table 4. The word thegn in runic inscriptions. Province Number of 209) dating is extremely convoluted (Thrane & Country TOTAL Nielsen 1998), but due to allegedly heathen inscriptions references it is usually placed before the 960s, Jutland 10 Skåne 4 as is the Gunderup stone (DR 143), whereas Denmark Lolland 1 17 Åker stone 3 (DR 372), in contrast, belongs to Fyn 1 the years 1050–1150. Also, two inscriptions Bornholm 1 (DR 129 and DR 150) and one coin (DR M94) Västergötland 18 have been excluded from this list due to the Södermanland 8 Sweden 29 corruption in their preserved condition. Småland 2 Finally, “there are two certain records of the Östergötland 1

37

Whereas highly specialised and poorly found en masse, may undergo a randomised attested vocabulary (such as already-mentioned selection (for example, manual picking of vanir) may allow detailed evaluation of each every tenth occurrence, or similar) to create a and every example of the word, examples of representative yet manageable corpus. Here many such terms, þegn among them, are so the study can also benefit from a collaboration numerous that reviewing them all in like with linguistic research of ‘distant reading’. manner would be impractical. The brief This is a relatively new method, pioneered by the overview above reveals, first, that this would Italian literary scholar Franco Moretti (2013), be too strenuous a task,21 and, second, that it who suggested that understanding literature is might not produce an expected pay-off. As possible not only through reading individual with plenty of Old Norse and Old English texts, but by aggregating and processing massive linguistic units, prevalent mentions are found data thereof with the aid of computational in: a) late saga literature that, unlike the methods. Alternatively, one could turn to the theoretically archaising laws, might represent minimal context the databases provide for the medieval usage; and b) Old English religious search: the brief verbal surroundings that prose, to a certain extent dominated by Ælfric accompany each hit (e.g. the DOE Web Corpus of Eynsham (d. ca. 1010), who “maintains his provides the sentence in which a lexeme occurs focus [...] assiduously on enduring and and sometimes the preceding and following important spiritual matters” (Amodio 2014: sentences). Conceivably then, the list of results 130). Since these sources, in spite of their could be briefly skimmed through during the abundance, are not as illustrative of social initial analysis, and all examples where the realities of the Viking Age, this limiting factor usage appears to diverge from common calls for a discriminant approach with a patterns could be separated for a more in-depth particular chronological emphasis. review. Having offered the solution to the challenge Discriminant Approach of the volume of the sources, we are still left The sheer volume of the assortment above with the problem of distinguishing the primary (2,668 hits in all corpora) calls for a discrete from the secondary sources. Traditionally, and balanced approach, and a separation of the when conducting social inquires, early primary sources from the secondary ones medieval historians show marked, somewhat underlies the suggested study just as in any ‘positivistic’ preference for the following texts other anthropological undertaking. In this as the most representative: particular case, Old English laws and diplomas, prose secular texts, and the Domesday Book, 1. Whenever possible written in vernacular, if and Old Norse place names, runic inscriptions, there are any, or employing vernacular terminology, for it is often believed that older skaldic verse and the oldest provincial Latin lexis might obscure the actual social laws are believed to belong to the former circumstances of the period22 category and therefore should be examined in 2. Of secular, normative and/or documentary their entirety. In contrast, Old English religious nature, that is to say customary law, royal texts, secular poetry, and Old Norse medieval ‘doom books’, capitularies, land titles, land royal legislation, sagas and eddic poetry fall property surveys, and the like into the latter category and may rather be subject To a certain extent, this choice is not without to a more selective analysis. Nevertheless, justification, and, when identifying the though the proposed methodology advocates primary sources for the current study, I suggest for a possibly maximal inclusion of all available beginning with similar criteria but conjugated Old English and Old Norse vernacular sources, with a strictly chronological principle. For this due to the overwhelming volume of the material, reason, for example, Gulaþingslǫg, though secondary sources require a more nuanced recorded in the 13th century but hypothetically treatment for the sake of manageability. It retaining some archaic layers (note, however, seems only logical that the most instructional the debate between Elsa Sjöholm and her of them ought to be treated in a similar way to opponents23), seems more representative for the primary ones – i.e. assessed in full. Less the Viking Age than the Norwegian Landslov telling attestations, especially should they be 38

[‘The Law of the Land’], issued at the medieval usage and later influences, not to instigation of King Magnus VI (r. 1263–1280) mention the local peculiarities and language of between 1274 and 1276 (Agishev 2015). This composition. All in all, Latin and Scottish is also why Scandinavian Viking-Age rune sources should perhaps be treated in a similar stones, older skaldic poetry and place names way to Wulfstan’s works: discreetly and have been categorised here due to their continuously checked against the vernacular apparent contemporaneousness with the period evidence. under examination. Much of the criticism of the state of In this connection, as stated above, Latin research laid out above has to do with the narrative texts, even when meeting the criteria exclusion of the manifold additional sources, for the primary sources, pose something of a called here secondary. Numerous details and methodological dilemma. On the one hand, insightful pieces of evidence can be found they are indeed abundant, and they frequently therein. On the one hand, they are perhaps more belonged, and were produced in, similar or informative of the ways the societal composition even the same social environments – hence was actually perceived by its various members, they should of course be taken aboard the study. from the high clergy, such as Ælfric, to the On the other hand, in line with John Kemble’s anonymous saga-writers. On the other, logically assertion, they resist the corpus approach connected to the previous surmise, they might presented above owing to their lexical question the acceptance of the views on the variance. Though we can be relatively positive social structures, imposed by legislators, that the normal rendition of the Latin minister should the actual sources present alternative was the Old English thegn (see: Loyn 1955), pictures (see above). Hence, ‘secondary’ as the opposite is far from being universally the they may be, they ought to be included in the case: on the whole, Latin translations from Old study. At the same time, as elucidated by the case English show a great deal of instability, of the Wulfstanian texts, no information warrants rendering thegn as tainus, minister, optimas, being taken at face value, so when being proceres, nobilis, comes et al. (cf. Thacker processed, all sources must be checked against 1981). Word choice might vary even within their provenance in spirit of the Weibull one individual text. For example, where other brothers, which in turn calls for breaking them manuscripts of the Anglo-Saxon Chronicle into categories of one kind or another. unanimously mention thegns, manuscript F (ca. Many a study with large sets of data 1100) has quidam perdives for an þegen in 465 necessitates various groupings for its data. (Baker 2000: 21), nobiliores for godan ðegenas Organising principles may vary from region to in 1010 (ibid.: 102) and primi Occidentalium region, period to period, field to field and Saxonum for ealle ða westernan ðegenas in depend on the type of records, scribal material, 1013 (ibid.: 106). In short, this variability provenance, genre, notions of authorship, requires a certain degree of caution in dealing preservation, circulation in, and indigenousness with the Latin sources, especially those that are to, a given culture, as well as numerous post-Conquest, due to an erosion of the object additional criteria. One of the arbitrary of study of sorts. When it is conceded that Old descriptive patterns for the Old English and Old English thegn used to be an ‘umbrella term’ for Norse literary sources is presented in Table 5. any member of the lay elite (cf. Roberts 2000) Table 5. Possible categorisation of Old English and Old and therefore enjoyed multiple Latin Norse sources. equivalents, then a thorough analysis of the Parameter Value vernacular sources is required before turning to Language of Vernacular Latin those in Latin. The same holds true for the composition Scottish evidence, first pioneered by Reid Discourse type Secular Religious (1920). There can hardly be much doubt that Type of this material preserves some older traits from Prose Poetry the society the lexeme þegn was borrowed composition Society the source ‘Germanic’ Mythical/foreign from (either Norse or Anglo-Saxon), but it is describes often challenging to discern them from

39

Applying this subjective scheme to Table 6. The word thegn in the Old English laws ca. 690–1066.24 the sources does not necessarily Non-royal codes Royal Private impose an evaluative hierarchy, it Date (recorded TOTAL codes compilations merely assists in the assessment of the ca. 1000–1025) clues they yield in relation to their 690–900 3 – – 3 types. One can expect that, for 900–975 10 7 – 17 instance, the secular and prose Anglo- 978–1012 11 2 – 13 Saxon Chronicle, written in Old 1012–1066 10 6 9 25 English and covering historical events TOTAL 34 15 9 58 in England, should be regarded as more suggestive of an actual society than, say, knowledge about thegns’ legal status is the eddic poem Rígsþula which, though also therefore of late origin, a point which seems to composed in a vernacular, retells the have been underestimated by many historians. ‘mythological’ origins of a similar society. One could argue that this proportion correlates That being said, both sources present answers with the overall chronological preservation of to scholarly questions, but the questions the Old English corpus, hence such a distribution themselves must be essentially different, and is to be expected. To this may be countered that, this notion must be accounted for in the actual first, the Alfredian legislation (late 9th century), historical research. In this case, the Anglo- despite its extensive volume, bizarrely contains Saxon Chronicle is perhaps more illuminating but a single explicit reference to the thegns of the everyday language used to describe the (Liebermann 1903: 126), and, second, that the contemporary society, whereas Rígsþula later asymmetry is largely the effect of apparently highlights the symbolic categories Wulfstan’s legislative activity (be it his own through which a community perceived itself. unique texts or the remastering of supposedly older customs). Indications exist that elevation Chronological Analysis and spread of the thegns as a socially elite group Of paramount importance is relating the chronologically coincide with the elaboration described sources to a timeline, which for our of its features in the legal codes. For instance, purposes has been framed twofold. Due to the King Æthelstan’s (r. 924–939) legislation arbitrary nature of establishing dates for a stipulates for the first time ever that king’s period such as the Viking Age, and in order to thegns act in the judicial capacity in the courts treat the subject in its proper historical and also as ‘officials’ charged with the authority contexts, the study cannot avoid employing to provide shelter for persecuted criminals sources from earlier and later periods, though (Liebermann 1903: 168, 171) – at the same this ought to be done with great caution. This time, Æthelstan’s diplomas are exceptionally is especially true for the Old Norse laws: well-witnessed by the royal ministri, their th th written down in the 13 –14 centuries, they overall number known being 127, which is an purport to be a codification of the earlier legal absolute record (Keynes 2002: Table 39). It tradition, but assessing how much of their seems likely that Wulfstan’s personal interests material actually dates from before 1100 in the orderly society alone might not explain remains a dodgy problem (Brink 2013). the aforementioned chronological disproportion, The methodological danger in this connection perhaps, his attention to the thegns was a has been described above: a tempting prospect response to some actual social shifts (see also of projecting the later information onto the below). Whatever the cause for the numerical earlier epochs. Nowhere is it more evident than skew in Figure 7, the bottom line is that when dealing with the Anglo-Saxon sources. projecting the later evidence onto an earlier Figure 6 demonstrates the chronological period needs meticulous argumentation. distribution of the mentions of thegns in the Old English legislative texts ca. 690–1066. Attention to Inner Features Easily observable is the prevalence of the Closely linked to the chronology of the sources younger pieces in this selection: 43% of all is the problem of their provenance, genre and mentions are concentrated in the fifty years circulation in society. The Anglo-Saxon kings’ before the Norman Conquest. Most of our law codes seem to be the easiest to tackle, as 40 we are usually aware of their promulgators and that the lexical choice “may have been the circumstances of their appearance. It is conditioned not simply by its semantic content much harder to establish the nature of the so- but also by the necessity of finding a term to called ‘private compilations’ and non-royal alliterate with þér,” and that “had the verse legislation. As a matter of fact, they contain been composed in either the first or third just as much relevant evidence, but a great person rather than the second, then phrases many details for contextualizing the sources such as *mér at mǫnnum or *honum at hǫlðum are wanting. We often know very little about might have been equally acceptable with an who wrote them down, when and why.25 The equivalent semantic force” (Syrett 1998: 262). discriminant analysis referred to above These observations, naturally, do not destroy suggests a careful examination of each source’s the value of these strophes for the aims of the credibility. Particular emphasis should be placed suggested study, but they nevertheless must be on recent developments in source criticism taken into account. On the other hand, just as advanced by philologists and linguists.26 The with the case of the ‘secondary’ sources, the same holds true for the Old Norse sources: the skaldic verse sheds light perhaps not on the formulaic and poetic nature of the skaldic verse hypothesised social structures per se, but on must always be considered and given due weight the shifting connotations a lexeme could when observing the Nordic þegns’ position. acquire in its changing cultural environment. It Two examples can illustrate the case in point: is very possible that both King Harald harðráði Þegi þú, Þórir! Þegn est ógegn; and Óttarr svarti consciously chose þegn not heyrðak, at héti Hvinngestr faðir þinn. only for the sake of meter, but for the overtones (Hharð Lv 3.) and morphing senses it conveyed. Last but not least, the local peculiarities of Be quiet, Thorir! You’re an unreliable man the region under study need to be observed: (þegn); I heard that your father was called Hvinngestr (‘Thief-guest’). (Gade 2009: 45– neither England nor Scandinavia were 46.) anything like socially or culturally unified entities, especially the Nordic lands. Though In this stanza, King Harald harðráði uses the England was indeed politically united by the word þegn among the variety of poetically West Saxon dynasty by 954, this never meant equivalent words for ‘man’ in order to meet a societal levelling throughout the new alliteration with þegi and Þórir required by the country. For example, the Domesday Book meter, and simultaneously to rhyme with attests that shires west of the Danelaw were ógegn. Did Harald mean that Þórir was an almost totally devoid of sokemen, and the free unreliable ‘man’ in a general sense, or that he peasantry (liberi homines) made up less than was unreliable in a more specific sense as a 10% of the western counties’ populations; in ‘retainer/warrior’? The second instance is in the West Saxon territories proper and their the Hǫfuðlausn (ca. 1023) by Óttarr svarti, dependencies annexed before the Great where þegn again rhymes with gegn: Heathen Army’s invasion in 870s, only a few Gegn, eru þér at þegnum enclaves contain small percentages of such a þjóðskjǫldunga góðra free population (Darby 1977: 61–94). These haldið hæft á veldi differences, of course, have a direct bearing on Hjaltlendingar kenndir. (Ótt Hfl 20.) the suggested research, as there are some Trustworthy one, you hold fittingly onto the records (e.g. the Domesday Book itself) from power of good kings of the people; the the Danelaw that speak of thegns, and it is Shetlanders are known to you as your thanes. methodologically hardly warranted to explain (Townend 2012: 739.) them with the West Saxon texts’ aid alone. Unlike in the previous stanza, here the meaning Scandinavian sources also need to be seems more transparent: Martin Syrett (1998: meticulously analysed against their regional 262) and Judith Jesch (1993: 168) comment background. The most obvious dissimilarity is that the sense conveyed here is that of a associated with distinctions between the three ‘vassal’ or ‘subject’ (Syrett prefers the former, Nordic kingdoms. Norwegian and Icelandic Jesch the latter). However, Syrett also remarks manuscripts contain the overwhelming majority

41 of the literary mentions of þegns and provide Saxon charters and the Domesday Book – to a the most enlightening context, although, they descriptive statistical analysis. Their distinction are all of a medieval origin. In contrast, from all other sources lies in their documentary Denmark and Sweden seem to have lost this nature: the charters’ (that is, royal diplomas’) term in their day-to-day language rather early primary function was that of a land title29 and on, but it is here that the late Viking-Age rune whatever the purpose of the Domesday Book stones commemorating certain þegns are to be originally was (for discussion, see e.g. Roffe found. Comparing their distribution with the 2000: 1–16), the Great and Little Domesday Swedish Tegneby/Tängby/Tägneby poses further and their ‘satellites’ do present a nation-wide questions: land survey, even if their data omit a lot of complexities. As such, these sources are, at Does the absence of þægnaR in the Upplandic runic inscriptions imply that these warriors first glance, not particularly instructional, as were not svear by origin? What was the their information is rather repetitive and seems difference between a þægn and a rinkr?27 Is it conceptually mostly uniform. On the other possible that the term þægn was originally hand, such information permits adding a used in the Götarike, believed to have been practical dimension to the cultural phenomena; subdued by the svear sometime in the late for example, we can inquire about the extent of Iron Age? (Strid 1986: 305). royal land donations in favour of the kings’ After all this has been done, the final thegns, or investigate the presence of the comparative stage – a search for probable Domesday taini in the English localities, both parallels and influences between the two West Saxon and Scandinavian. societies, separated (or, perhaps, united?) by Anglo-Saxon England has left us at least the North Sea – can be commenced. 1,875 Latin and vernacular documents, collectively referred to as charters. A full Approaching the Historical Reality behind a catalogue was put together in 1968 (Sawyer Fleshed-Out Concept 1968), and later the Centre for Computing in An often-recurring distress in historical research the Humanities at King’s College London used is trying to relate a social or cultural category, it to create an online version, which first identified in the literary sources, with its actual appeared in 2007 as a beta version (Electronic existence and operation in the society of a Sawyer). It is by far the fullest and most particular time in terms of factual praxes. Not fundamental annotated list of the extant Anglo- infrequently are scholars forced to “draw wide Saxon documents and an essential tool for any conclusions from detailed case studies” (Lavelle student of the Anglo-Saxon charters. The 2010: 63) for the lack of ‘tangible’ data – earliest specimens date from the late 7th century Lavelle’s example from the quote being that of and these initially served as land titles: kings la mutation feodale.28 At times this would record donating estates to the Church shortcoming can be remedied by adding and (later) to laymen in their service, the essentially alternative material. Exemplary are majority of which were styled thegns. As time archaeology as well as further ancillary progressed, other types of documents appeared: historical disciplines (epigraphy, heraldry, wills, marriage agreements, lease contracts, etc., numismatics, onomastics, prosopography, etc.) though the single largest portion of the extant that can ‘put flesh on the bones’ of a literary corpus is still made up of solemn diplomas phenomenon, thus offering case of a ‘happy (1,060 originals and/or cartulary copies). This marriage’ between ‘theoretical’ and ‘empirical’ is an astonishingly rich material that permits a approaches (cf. Chadwick 1905). Fortunately, thorough descriptive statistical analysis: how, thegns are of this sort. After the conceptual where, how often and in what amounts was framework for the term thegn and its evolution land granted to kings’ thegns? Though this in time has been elucidated, the study may approach has been at least partially practiced endeavour to augment one of the possible before (e.g. Keynes 2005 [1980]; Lavelle 2011; social realities behind the literary texts. This Snook 2015), as far as I am aware, diplomas en venture might be achieved by subjecting an masse have so far not been used a source for inherently different type of source – the Anglo- social history. The application of such

42 methodology may cast a lot of light on the donation being about 9.47 hides (1,819 hides thegns’ economic standing and relationships recorded in 192 authentic and/or based upon with the monarchy (Sukhino-Khomenko 2016). authentic diplomas) (Sukhino-Khomenko 2016: Ideally, a full prosopographical dimension, i.e. 279–280). As mentioned in the introduction, studying the careers and lives of all individuals numerous Domesday taini (colloquially, today of whom we have documented knowledge, could they are usually called ‘minor thegns’) therein contribute considerably to our understanding seem to have enjoyed only limited economic of the composition of the late Anglo-Saxon lay prosperity: for example, according to Michael elite. However, reasonable doubts persist as to Costen’s (2007: 65) figures, in Wiltshire, an whether this voluminous task alone, which has average tenure of an anonymous thegn in 1066 sometimes been a topic for entire PhD theses was 1.39 hides. This has usually been (see e.g. Senecal 1999), will not require a explained as a sign of the thegnly group’s whole team of specialists. diffusion and erosion, a process which led to It is especially intriguing to compare these many of its members to amalgamate with the statistics against similar Domesday data. A non-aristocratic strata while still preserving an preliminary search has returned hundreds of elevated legal status, expressed in the wergild hits for thegn in its Latin form tainus (pl. taini) rate (cf. the opinions of Reid and Barlow above). in the Domesday Book, and looking at the pair However, a possibility must be entertained of similar information sets can prove that, similarly to Old Norse þegn, the Old English beguiling. This, of course, does not mean that cognate possessed more than one meaning, there lived only a few hundred thegns at the attested by typologically different sources, time of the inquest: in most cases, the word is which makes this comparison all the more used in the plural form, and few thegns’ names needed and warranted. Practically speaking, a are actually recorded. Regretfully, to my question of why two groups that were so knowledge, existing Domesday databases are economically incomparable came to be referred not designed to search for anonymous persons to by one and the same term requires further of whom only their social status is known contemplation. As said, the 1,200-shilling (Palmer 2016), and that precludes ‘old-school’ wergild is often proclaimed to be the uniting manual searching patterns, although this task is factor (see Stenton’s opinion above), though it nevertheless much facilitated by modern is hardly imaginable that a petty tenant would computer technologies. possess adequate resources, comparable to those Two vital methodological shortcomings must of a king’s man, to secure such demanding nonetheless always be borne in mind during legal privileges (cf. Sawyer’s scepticism on the this exercise. First of all, chronologically, the functioning of the wergild). Other criteria have charters and the Domesday Book provide very also been suggested: particular types of land different evidence – the diplomas demonstrate ownership/tenure or, maybe, military service a dynamic picture; the Domesday Book, in (Costen 2007: 62; cf. Maitland 1921 [1897]: contrast, shows merely a static one. Second, 164). What seems yet to have received a the Domesday Book was written 20 years after methodological articulation is the possible the Conquest by and for the Norman notion of symbolic capital. Kings calling the government. Hence, we are dealing largely recipients of land grants their thegns might with a rather Norman, not an Anglo-Saxon, have implied some certain types of personal perception of society. bonds that entailed military service for the Moreover, even a brief overview of the donees. But what the country-folk might have charter and Domesday evidence raises a instead observed in their everyday routines is suspicion that, though both sources belong to a that their affluent secular lords, clothed in documentary type, they might essentially lavish garments and armed with expensive concern starkly different social groups. The weapons, referred to themselves as [king’s] royal diplomas addressed to the royal ministri thegns. Therefore, it could be hypothesised recorded kings’ grants to a rather narrow (see that, to the populace, any holder of such the statistics in Keynes 2002) group within the symbolic objects was essentially a thegn and a monarchs’ inner circle, with the average member of the lay elite. Unexpected support

43 for this argument may come from none other reveal consistent ‘oddities’ in this region not than Wulfstan, who not only continuously present in the ‘English’ part of the politically stressed the importance of the royal service, united Anglo-Saxon kingdom, they might be but, in Norðleoda laga, went so far as to deny a cautiously compared to the Scandinavian thegnly wergild to a ceorl who owns “a helmet counterpart, although this shall demand a lot of and a coat of mail and a gold-plated sword” comparative studies and very discreet (Rabin 2015: 71), unless he also possesses land approaches: as argued by Dawn Hadley (2000), to acquit obligations for the king. Particularly Scandinavian incursions and colonisation of thought-provoking is the apparent wider North-Eastern Britain cannot explain away all circulation of swords in the Nordic lands in the regional peculiarities in a deus-ex-machina Viking Age than in England (Androushchuk fashion. At the end of the day, fleshing out the 2009), which yet again advocates for framing social reality of the Scandinavian þegn-hood the study of Old English and Old Norse may yet have some resources to fall back on. evidence in tandem and plies the question of whether there has been influence either Conclusion regionally or in certain social arenas. The current paper’s main goal has been to Unfortunately, no similar statistics exist for bring forward one feasible methodology for Viking-Age Scandinavia, which is a great pity. reassessing certain original historical lexemes Rune stones are very tangible traces of that in order to better form modern notions of the social reality, but they can only tell so much, categories behind them. Achieving this is and the immediate context is often wanting. suggested in a threefold way. Despair is, however, probably premature, as First, one might wish to consider abandoning there are yet alternative ways, be they even the search for meanings of a studied lexeme in perhaps not as exhaustive, to at least partially dictionaries and instead deduce them from contextualise the possible socio-cultural concrete literary contexts. This can at first glance circumstances of the thegnly phenomenon also seem a duplication of an actual dictionary, but in Scandinavia. First, the very runic modern progress in digitalisation of the inscriptions, laconic in their content as they historical sources enables applying methods of are, exist on a wider background and their corpus linguistics in historical research, which general cultural surrounding has seen decades used to be inaccessible for the dictionary- of investigations (e.g. Moltke 1976; Löfving makers of the past. 2001). Though not as vocal as literary pieces, Second, a corpus search can yield an the prestige and symbolic significance embodied unmanageable number of results, which in turn by rune stones confer the notion of a calls for prioritising the sources. Those commemorated person’s prominence in the considered ‘primary’ ought to be studied in full locality, if even Danish kings – Gorm and while the ‘secondary’ ones may receive a more – adopted this medium to selective approach. However, as a running communicate powerful messages. The sheer phrase teaches us: cost of transporting, raising and ornamenting There are no good or bad sources, there are such a monument sheds light on the wealth of only good or bad ways to use them. (Janson the corresponding family or clan. Second, 1999: 71.) Scandinavia features not only Tegneby place Let the word ‘primary’ not delude a researcher, names, absent in England, but other toponyms for it labels only those texts that do not always with a similar meaning and coined in a similar yield the answer to the immediate question fashion as well: Sveneby, Karleby, Rinkeby. asked. Instead they can illuminate further Comparing their patterns of distribution with aspects of a problem under investigation that archaeological data and putting them in the might get omitted by the ‘primary’ source context of landscape can yet yield relevant material. Therefore, all sources ought to material for the studies of the social orders in receive their maximum due attention, even if it these regions (Brink 1999). Finally, given the entails a less exhaustive approach. In this Danelaw’s uniqueness, should the statistical connection, it is necessary to remember the analysis of the Domesday and charter evidence probable chronological, regional, genre and

44 other distortions of search results. Hence, inner historiska studier for inviting me to the winter Internat- source criticism, not unlike the one hotly 2018, where I was offered an opportunity to deliver an propagated and defended by the Weibull earlier draft of this article. brothers (Janson 1999: 70–75; Torstendahl Notes 1981: 117–126), remains as relevant as ever. 1. A hide (Old English hīd, hiwisc) was a plot of land, Third, upon tracing the probable meanings capable of providing for one household throughout a of a lexeme and their historical evolution, one year; later a fiscal unit (see Charles-Edwards 1972). can try ‘putting flesh on the bones’ of the 2. That is not to say that there were 188 thegns in reconstructed concept(s). This is usually England in 900–1066. The total number of supposed land donations in favour of the thegns in this period carried out with the aid of the ancillary is 227, but only 188 of them are unanimously historical disciplines and by associating some considered authentic by modern scholars (see ‘practical’ information with the study. That is Sukhino-Khomenko 2016). Some charters might be to say, the third stage presupposes changing grants to one and the same person. Every Anglo- the focus of the inquiry from establishing the Saxon diploma normally contains a list of witnesses, many of whom are titled ministri, and minister is the multifaceted nature of a historical concept to standard equivalent for Old English þegn (for more adding an actual practical, ‘tangible’ detailed statistics see: Keynes 2002; see, however, dimension to it. above). Therefore, we may preliminarily conclude This paper has tried to demonstrate how that the overall number of the Anglo-Saxon thegns, such a methodology might be applied to the whose existence is recorded in the 900–1066 diplomas alone, must be around the figure of a few hundred. study of the Viking-Age Anglo-Saxon and 3. The problem of the over-reliance on Wulfstan’s and Scandinavian thegns. It is expected that the further ‘monarchocentric’ sources when examining possible conclusions of the suggested the late Anglo-Saxon aristocracy was touched upon approaches could touch on many problems in by Christine Senecal in her doctoral thesis (1999: current medieval scholarship in England and 17–28) and a subsequent article (2000: 251–252). 4. In this social context, a ceorl (modern English , Scandinavia. It is also humbly hoped that this German Kerl, Old Norse karl, etc.) is usually case study can outline one feasible way to understood as a ‘commoner’, a ‘rank-and-file’ transcend disciplinary borders and add to a member of the Anglo-Saxon society. Rosamond more holistic view of the . Faith (1997: 127) characterises ceorlas as “a large Re-approaching and bringing all available data and loosely defined social category [...], which included all those who were neither unfree nor of together in the light of current episteme, as aristocratic birth,” which also “may preserve well as comparing it with contemporary vestiges of a social class of a type which escapes our material from other geographical regions, can modern typologies, a class in which both peasant help us better understand both the political and farmers and lesser landowners were to be found.” structural organisations of early medieval 5. To give Stubbs’ predecessors credit, most elements of his scheme had been laid out in the preceding societies, and, perhaps, human societies more seven decades. Sharon Turner (1768–1847) was the generally. first popular writer to transform the thegns from an antiquated curiosity akin to the Scottish thane from Denis Sukhino-Khomenko (denis.sukhino-khomenko Macbeth into an object of historical interest and [at]gu.se), Department of Historical Studies, study (Turner 1801: 76–83, 222–231). And John Eklandagatan 86, Box 200, Göteborg 405 30, Sweden. Mitchell Kemble (1807–1857) contributed the

notion of the Germanic comitatus to Stubbs’ Acknowledgements: I would like to thank professor Ian description (Kemble 1849: 162–184), for Kemble Wood (University of Leeds), Steven McGrath, Matthew was the pioneer of the German 19th-century Delvaux (Boston College) and Frog (University of historical scholarship on British soil. Helsinki) for proof-reading this paper, as well as 6. As indicated by Reid (1920: 174) herself, “It is not professors Henrik Janson and Lars Hermanson easy to determine with exactness what rights were (University of Gothenburg) and my fellow doctoral- actually conveyed by a grant of sac and soc, toll and student colleagues Frida Esoplin Norstein (University team, and infangthef.” David Roffe (1989) and she of Gothenburg), Lewis Webb (University of Gothenburg) understand sake and soke to be synonymous with what and Peder Flemstad (University of Lund) for their was later known as ‘leet jurisdiction’, infangentheof much-valued comments, critique and suggestions. I as ‘the summary judgement of hand-taken thieves’, would also like to thank the organisers of the 7th and toll and team as the rights to pursue trade in one’s Austmarr Symposium at the University of Tartu (1st–3rd estate and to warrant cattle purchases, respectively. December 2017), where I had the pleasure to present 7. Cf. Stenton’s (1932: 117) metaphor of “a true national the historiographical section of the present paper, as levy”. well as the Swedish Nationella forskarskolan i

45

8. Such as William Stubbs, John Green, Edward saga ok Heiðreks [‘The Saga of Hervǫr and Freeman, and Frederic Maitland (see further Lavelle Heidrek’]. I would like to thank Frog for suggesting 2010: 47–55). incorporating these numbers into the present study. 9. I would like to thank professor Elena Melnikova 16. It should be noted that, as in the case with Old from the Institute of World History (Russian Swedish, Old High German and Old Saxon, these Academy of Sciences) for pointing this book out to figures are likely incomplete, for the Skaldic Poetry me. Melnikova was in fact also the first colleague to of the Scandinavian Middle Ages project is ongoing. direct my attention to Carl Löfving’s works. Thus, when the database was first consulted in June 10. A similar paradigmatic critique of Aakjær’s 2016, it listed only 72 occurrences; the extra 20 approach comes from Martin Syrett (1998: 246), instances in 2018 originate from the 2017 volume who justly points out Aakjær’s heavy dependence on Poetry from Treatises on Poetics. Even today, not all the English post-Conquest sources, though even he listed occurrence have made it into the printed did not consider Constitutiones de Foresta’s editions. Furthermore, though the headword search invalidity for the proposed argument. returned 93 results for the lexeme þegn and two for 11. For a more complete and exhaustive treatment of the its compounds, certain allowances concerning problems of Aakjær’s influential article, see various dubious readings had to be made for the sake Sukhino-Khomenko 2018. of the chart’s clarity (plus the search did not yield the 12. Beginning from the 1860s, linguists have linked lausavísa by Óláfr bjarnylr Hávarðarson, though the thegn with the Greek τέκνον [‘child’] and derived lausavísa itself is indexed in the database). both from PIE *tek- [‘to beget’]. In this case, the Therefore, it is correct to speak of 89 confirmed semantic evolution of the word would be recon- occurrences (88 for þegn and one for þegnskapr) and structed: ‘to beget’ → ‘child’ → ‘boy’ → ‘servant’ six that remain dubious. (e.g. Aakjær 1927: 18–19). However, linguist Guus 17. As disclaimed by the editors, the database is still Kroonen (2013: 536) has convincingly connected being constructed, therefore its “material is þegn with Proto-Germanic *þegjan- [‘to request’] incomplete and is for reference only” (Skaldic (cf. Old Norse þiggja, Old English þicgan), thus Project). The search was executed throughout the rooting it in the meaning ‘retainer’ rather than headwords of the dictionary relevant to the editions. ‘child’. When the present article was published, a 18. For Old Danish, the material from the Gammeldansk paper on this matter involving the word’s historical Ordbog [‘Old Danish Dictionary’], developed by background and its analysis was being drafted by Det Danske Sprog- og Litteraturselskab [‘The Guus Kroonen and myself for a further publication Danish Language and Literature Society’], has been (preliminary results were delivered by me at the used. workshop Perspectives on the Nordic Middle Ages 19. The figures for Old Swedish should still be at Aarhus University on 4th May 2018). considered preliminary and may be subject to 13. A similar methodology has been recently used by revision. They have been extracted from both the Aleksei Shchavelev (2017) when reconstructing the accurate word register of Carl Schlyter (1877: s.v. earliest historical lineage of the Russian Rurikid ‘þægn’), first editor of the Swedish Landskapslagar dynasty in the 10th century. The problem has always [‘Provincial Laws’], and a potentially less complete been that the only extant Russian annalistic source list of examples in Knut Söderwall’s Ordbok öfver for this period, the so-called Primary Chronicle, svenska medeltidsspråket [‘Dictionary of Swedish compiled only shortly before 1120, includes many Medieval Language’] (Söderwall 1884–1918: s.v. folkloric elements and dates events retrospectively. ‘þäghn’). In case of Old High German and Old Shchavelev set the Primary Chronicle aside and Saxon, Thesaurus Indogermanischer Text- und instead called upon synchronous foreign evidence – Sprachmaterialien (TITUS), a joint project of the Byzantine, Latin, Jewish-Khazar and Arabic. His Goethe University Frankfurt, Charles University, investigation demonstrated that evidence from these Prague, Copenhagen University, and University of sources closely matched the Primary Chronicle’s Oviedo, was used. Note that Kuhn (1944: 109–110) genealogy, thus affirming the Primary Chronicle’s provides lower figures, though he probably did not (or its prototype’s) reliability in this matter. A similar seek to produce an extensive list. However, some procedure was proposed by Curt Weibull in his 1921 compounds with degan listed in the dictionaries essay on the study of Saxo Grammaticus’ works: (such as heridegan) are for some reason absent from Weibull insisted that events of Danish 12th-century TITUS. This is probably due to the ongoing nature political history be reconstructed off contemporary of the electronic project (although I cannot be certain documentary evidence, and Saxo’s troubled narrative here), which is why the figures for these languages be applied later on (Torstendahl 1981: 126). must for now be considered preliminary. At any rate, 14. As pointed out by Frog, this formula is also found in it is believed that the acquired statistics might some saga texts, e.g. in Sverris saga. already be sufficient and, more importantly, 15. For eddic poetry, indices by Hugo Gering (1903) and illustrative of the suggested methodology. Robert Kellogg (1988), which for the most part 20. “Sö 237, 349, U 34, 131, 201, 353, 363, 372, 456, overlap, have been consulted. The only differences 935, 937, 991, 990, 999, and a stone found in 1978 are that Kellogg lists the occurrence of the name at Ösby, Lunda parish, as yet unnumbered” (Strid Þegn in Rígsþula [‘The Song of Ríg’] with personal 1986: 314). names, and he includes two occurrences in Hervarar

46

21. Similarly to the method used by Shchavelev, such DR 150 = Rune stone located in Hurup, Hurup sn., Refs ‘brute-force’ context analysis has also been inspired hd, North Jutland, Denmark. In Jacobsen & Moltke by Petr Stefanovich (2012). However, two major 1942: 603. obstacles hinder its complete adoption when dealing DR 209 = Rune stone located in Glavendrup, Skamby with thegns: for one thing, the number of individual sn., Skam hd, Fyn, Denmark. In Jacobsen & Moltke cases Stefanovich meticulously went through 1942: 606. amounts to merely hundreds, and, for the other, even DR 372 = Rune stone located in Aakirkeby, Åker sn., that many instances turned his research into a 656- Sønderhd, Bornholm, Denmark. In Jacobsen & page monograph. Moltke 1942: 615. 22. In the Anglo-Saxon studies, Lavelle (2010: 64) Eiðsivaþingslǫg. In Halvorsen & Rindal 2008: 1–117. suggests that this may be at least partly due to the Eithun, Bjørn, Magnus Rindal & Tor, Ulset (eds.). 1994. “result of the Germanist scholarship of the nineteenth Den eldre Gulatingslova. Oslo: Riksarkivet. century, in which J.M. Kemble argued that Faulkes, Anthony (ed.). 2005. Snorri Sturluson, Edda, Continental Latin terminology should not be used in Skáldskaparmál I: Introduction, Text and Notes. 2nd the study of Anglo-Saxon England and that ‘we use edn. London: Viking Society for Northern Research. no words but such as the Saxons themselves used’.” Gade, Kari Ellen (ed.). 2009. “Haraldr harðráði 23. In the 1970s through the 1990s, heated polemics Sigurðarson, Lausavísur 3”. In Poetry from the unfolded between Elsa Sjöholm, who challenged the Kings’ Sagas 2: From c. 1035 to c. 1300. Ed. Kari hitherto prevailing germanischen Urrecht theory and Ellen Gade. Skaldic Poetry of the Scandinavian maintained the Swedish provincial laws’ medieval Middle Ages I:2. Turnhout: Brepols. Pp. 45–46. continental origin (and, hence, their inapplicability Geþyncðu. In Rabin 2015: 68–69. for studying Viking-Age social structures), and her Gulaþingslǫg. In Eithun et al. 1994: 31–206. academic adversaries (Sverre Bagge, Thomas Halvorsen, Eyvind, & Magnus Fjeld Rindal (eds.). 2008. Lindqvist, and Ole Fenger), who saw Sjöholm’s De eldste østlandske kristenrettene: Tekst etter methodology as too rigid (for a current but not håndskriftene, med oversettelser. Oslo: Riksarkivet. entirely impartial overview, see Brink 2014). Jacobsen, Lis, & Erik Moltke. 1942. Danmarks 24. This chart is based on Sukhino-Khomenko 2014. runeindskrifter: Text. København: Ejnar 25. A masterpiece of an overview and assessment of Munksgaard Forlag. these problems is Wormald 2001. Larson, Laurence Marcellus (trans.). 1935. The Earliest 26. See e.g. a thorough contextualisation of Wulfstan’s Norwegian Laws: Being the Gulathing Law and the writings in Ponz-Sans 2007. Frostathing Law. New York: Columbia University 27. Unlike Tegneby, place names with the element Press. *rink- can be found not only in modern-day Sweden Liebermann, Felix. 1903. Die Gesetze der (18 instances, predominantly in the Mällaren region) Angelsachsen, I. Halle: Max Niemeyer. but also in present-day Denamark (8 instances) (Hald Rabin, Andrew (ed. & trans.). 2015. The Political 1933). Writings of Archbishop Wulfstan of York. 28. ‘Feudal mutation’, also termed ‘transformation’ – a Manchester: Manchester University Press. controversial theory in francophone historiography, Sawyer, Peter H. 1968. Anglo-Saxon Charters: An first formulated by Georges Duby in 1953 and later Annotated List and Bibliography. London: Offices much debated in the journal Past & Present (Abels of the Royal Historical Society. 2009: 1018–1020). Sö 34. = Rune stone located in Tjuvstigen, Vagnhärads 29. Note that, as cultural objects, charters fulfilled far sn., Hölebo hd., Södermanland, Sweden. In Brate & more diverse functions (see e.g. Keynes 2005 Wessén 1924–1936: 26–28. [1980]; Snook 2015). Townend, Matthew. 2012. “[Introduction to] Óttarr svarti, Hǫfuðlausn”. In Poetry from the Kings’ Works Cited Sagas 1: From Mythical Times to c. 1035. Ed. Diana Sources Whaley. Skaldic Poetry of the Scandinavian Middle Ages I:1. Turnhout: Brepols. P. 739. ASE = The Anglo-Saxon Chronicle: A Collaborative Whitelock, Dorothy (ed.). 1979. English Historical Edition, V–VIII: MS C, D, E, F. Cambridge: D.S. Documents I: c. 500–1042. 2nd edn. London / New Brewer. York: Oxford University Press. Brate, Erik, & Elias Wessén (eds.). 1924–1936. Sveriges runinskrifter, III. Stockholm: Almqvist & Wiksells Boktryckeri. Dictionaries Brodeur, Arthur Gilchrist (ed. & trans.). 1916. The Cleasby, Richard, & Gudbrand Vigfusson. 1874. An Prose Edda. London / New York: Oxford University Icelandic-English Dictionary. Oxford: Clarendon Press. Press. DR 129 = Rune stone located in Durup, Durup sn., Fritzner, Johan. 1867. Ordbog over det gamle norske Gislum hd, North Jutland, Denmark. In Jacobsen & sprog. Kristiania: Feilberg & Landmarks Forlag. Moltke 1942: 602. Gering, Hugo. 1903. Vollständiges Wörterbuch zu den DR 143 = Rune stone located in Gunderup, Gunderup Liedern der Edda. Germanistische Handbiblithek sn., Fleskum hd, North Jutland, Denmark. In VII:4:5, Die Lieder der Edda II: Wörterbuch. Halle: Jacobsen & Moltke 1942: 602–603. Waisenhauses.

47

Jónsson, Finnur. 1931. Lexicon poeticum antiquae Brink, Stefan. 2013. “The Creation of a Scandinavian linguae septentrionalis. 2nd edn. København: S.L. Provincial Law: How Was It Done?” Historical Møllers. Research 86(233): 432–442. Kellog, Robert. 1988. A Concordance to Eddic Poetry. Brink, Stefan. 2014. “The Origins of the Swedish East Lansing: Colleagues Press. Medieval Laws: A State of the Art, from My Kroonen, Guus. 2013. Etymological Dictionary of Perspective”. In How Nordic are the Nordic Proto-Germanic. Leiden / Boston: Brill. Medieval Laws? – Ten Years After: Proceedings of Söderwall, Knut Fredrik. 1884–1918. Ordbok öfver the Tenth Carlsberg Academy Conference on svenska medeltidsspråket, II. Lund: Berlingska Medieval Legal History 2013. Ed. Per Andersen et Bocktryckeri & Stilgjuteri-Aktiebolaget. al. Copenhagen: DJØF Publishing. Pp. 129–144. Schlyter, Carl Johan. 1877. Samling af Sweriges gamla Chadwick, Henry Munro. 1905. Studies on Anglo-Saxon lagar, XIII: Ordbok till Samlingen af Sweriges Institutions. Cambridge: Cambridge University gamla lagar. Lund: Berlingska Boktryckeriet. Press. Charles-Edwards, Thomas M. 1972. “Kinship, Status Literature and the Origins of the Hide”. Past & Present 52: 3– Aakjær, Svend. 1927. “Old Danish Thegns and Drengs”. 33. Acta Philologica Scandinavica 2: 1–30. Christiansen, Eric. 2002. The Norsemen in the Viking Aakjær, Svend. 1962. “Torperne og de danske drenge”. Age. Oxford: Blackwell Publishing. Politiken (København) 24 (December): 8–9. Costen, Michael. 2007. “Anonymous Thegns in the Abels, Richard P. 2009. “The Historiography of a Landscape of Wessex 900–1066”. In People and Construct: ‘Feudalism’ and the Medieval Historian”. Places: Essays in Honour of Mick Aston. Ed. History Compass 7(7): 1008–1031. Michael Costen. Oxford: Oxbow Books. Pp. 61–75. Abels, Richard P. 1988. Lordship and Military Darby, H.C. 1977. Domesday England. Cambridge: Obligation in Anglo-Saxon England. London: Cambridge University Press. Publications. Elgqvist, Erik. 1947. Ullvi och Götevi: Studier rörande Agishev, Sergei. 2015. “Zemskoy Zakon (Landslov) Götalandskapens införlivande med Sveaväldet. Magnusa Ispravitelya Zakonov: Utrachennoe Lund: Olins Antikvariat. edinstvo?”. Vox Medii Aevi 2–3: 49–89. Available Faith, Rosamond. 1997. The English Peasantry and the at: http://cyberleninka.ru/article/n/zemskiy-zakon- Growth of Lordship. Leicester: Leicester University landslov-magnusa-ispravitelya-zakonov- Press. utrachennoe-edinstvo (accessed 27th April 2018). Feller, L. 2006. “Introduction”. In Les élites au haut Amodio, Mark C. 2014. “The Writings of Ælfric of moyen âge: Crises et renouvellements. Ed. F. Eynsham”. In Mark C. Amodio, The Anglo-Saxon Bougard et al. Turnhout: Brepols. Pp. 5–22. Literature Handbook. Malden: Wiley-Blackwell. Frog & Jonathan Roper. 2011. “Verses versus the Pp. 116–134. ‘Vanir’: Response to Simek’s ‘Vanir Obituary’”. Androushchuk, Fedir. 2009. “ and Farmers: RMN Newsletter 2: 29–37. Some Remarks on the Social Interpretation of Gillingham, John. 1995. “Thegns and in Swords and Long-Distance Contacts During the Eleventh-Century England: Who Was Then the Viking Age”. In The Martial Society: Aspects of Gentleman?” Transactions of the Royal Historical Warriors, Fortifications and Social Change in Society (Sixth Series) 5: 129–153. Scandinavia. Ed. Lena Holmquist Olausson & Hadley, Dawn M. 2000. “‘Cockle amongst the Wheat’: Michael Olausson. Stockholm: Archaeological The Scandinavian Settlement of England”. In Social Research Laboratory, Stockholm University. Pp. Identity in Early Medieval Britain. Ed. William O. 93–104. Frazer & Andrew Tyrrell. London / New York: Barlow, Frank. 1988. The Feudal Kingdom of England, Leicester University Press. Pp. 111–135. 1042–1216. 4th edn. London / New York: Longman. Hald, Kristian. 1933. “Om rekkr og hǫldr i danske Baxter, Stephen. 2001. “Review of Domesday: The Stednavne: En Oversigt”. Danske folkemaal 7: 73– Inquest and the Book, (review no. 216)”. Reviews in 80. History. Available at: Harris, Sara. 2014. “Tam Anglis Quam Danis: ‘Old http://www.history.ac.uk/reviews/review/216 Norse’ Terminology in the Constitutiones de (accessed 1st February 2018). Foresta”. Anglo-Norman Studies 37: 131–148. Baxter, Stephen, & John Blair. 2006. “Land Tenure and Hertzberg, Ebbe. 1890. “Efterskrift angaaende Royal Patronage in the Early English Kingdom: A tvivlsomme ord i Norges gamle love”. Arkiv för Model and a Case of Study”. Anglo-Norman Studies Nordisk Filologi 6: 262–271. 28: 19–46. Hollister, Charles Warren. 1962. Anglo-Saxon Military Brink, Stefan. 1999. “Social Order in the Early Institutions on the Eve of the Norman Conquest. Scandinavian Landscape”. In Settlement and Oxford: Clarendon Press. Landscape: Proceedings of a Conference in Århus, Hudson, John. 2012. The Oxford History of the Laws of Denmark, May 4–7 1998. Ed. Charlotte Fabech & England II: 871–1216. Croydon: Oxford University Jytte Ringtveg. Moesgård: Jutland Archaeological Press. Society. Pp. 423–439. Janson, Henrik. 1999. Till frågan om Svearikets vagga. Gällstad: Västergötlands Hembygdsförbund.

48

Jesch, Judith. 1993. “Skaldic Verse and Viking and Elsewhere. Copenhagen: The National Museum Semantics”. In Viking Revaluations: Viking Society of Denmark.) Centenary Symposium, 14–15 May 1992. Ed. Molyneaux, George. 2011. “The Ordinance Concerning Anthony Faulkes & Richard Perkins. London: the Dunsæte and the Anglo-Welsh Frontier in the Viking Society for Northern Research. Pp. 160–181. Late Tenth and Eleventh Centuries”. Anglo-Saxon Jesch, Judith. 2013. “ and Words: Runic England 40: 249–272. Lexicography in Context”. Futhark: International Molyneaux, George. 2015. The Formation of the Journal of Runic Studies 4: 77–100. English Kingdom in the Tenth Century. New York: John, Eric. 1960. Land Tenure in Early England. Oxford University Press. Leicester: Leicester University Press. Moretti, Franco. 2013. Distant Reading. London / New Kemble, John Mitchell. 1849. The Saxons in England: A York: Verso Books. History of the English Commonwealth till the Period Murray, Alexander C. 1983. Germanic Kinship of the Norman Conquest, I. London: Longman, Structure: Studies in Law and Society in Antiquity Brown, Green & Longmans. and the Early Middle Ages. Studies and Texts. Keynes, Simon. 2002. An Atlas of Attestations in Anglo- Wetteren: Universa. Saxon Charters, c. 670–1066. Cambridge: Naismith, Rory. 2015. “H.M. Chadwick and the Anglo- Department of Anglo-Saxon, Norse, and Celtic, Saxon Monetary System”. In H.M. Chadwick and University of Cambridge. the Study of Anglo-Saxon, Norse and Celtic in Keynes, Simon. 2005 [1980]. The Diplomas of King Cambridge. Ed. Michael Lapidge. Aberystwyth: Æthelred ‘the Unready’ 978–1016: A Study in Their Department of Welsh, Aberystwyth University. Pp. Use as Historical Evidence. 2nd edn. Cambridge: 143–156. Cambridge University Press. Nielsen, Karl Martin. 1945. “Var Thegnerne og Kuhn, Hans. 1944. “Altnordisch rekkr und Verwandte”. Drengene Kongelige Hirdmænd?” Aarbøger for Arkiv för Nordisk Filologi 5: 105–121. nordisk oldkyndighed og historie 1945: 111–121. Larson, Laurence Marcellus. 1904. The King’s O’Brien, Bruce R. 2011. “Authority and Community”. Household in England before the Norman Conquest. In A Social History of England, 900–1200. Ed. Julia Madison: Reprinted from the Bulletin of the C. Crick & Elisabeth van Houts. New York: University of Wisconsin. Cambridge University Press. Pp. 76–97. Lavelle, Ryan. 2010. Alfred’s Wars: Sources and Palmer, J.J.N. 2016. “A Digital Latin Domesday”. In Interpretations of Anglo-Saxon Warfare in the Domesday Now: New Approaches to the Inquest and Viking Age. Warfare in History. Woodbridge: The the Book. Ed. David Roffe & K.S.B. Keats-Rohan. Boydell Press. Woodbridge: Boydell Press. Pp. 61–80. Lavelle, Ryan. 2011. “Royal Control and the Peterson, Lena. 1994. “Scandinavian Runic-Text Data Disposition of Estates in Tenth-Century England: Base: A Presentation”. In The Twelfth Viking Reflections on the Charters of King Eadwig (955– Congress: Developments around the Baltic and the 959)”. The Haskins Society Journal 23: 23–50. North Sea in the Viking Ages. Ed. Björn Ambrosiani Liebrmann, Felix. 1894. Über Pseudo-Cnuts & Helen Clarke. Stockholm: Typografen Text & Constitutiones de foresta. Halle: Max Niemeyer. Rubrik. Pp. 305–309. Lindow, John. 1976. Comitatus, Individual and Honor: Pons-Sanz, Sara M. 2007. Norse-Derived Vocabulary in Studies in North Germanic Institutional Vocabulary. Late Old English Texts: Wulfstan’s Works, a Case London / Berkeley / Los Angeles: University of Study. Odense: University Press of Southern California Press. Denmark. Loyn, Henry R. 1955. “Gesiths and Thegns in Anglo- Randsborg, Klavs. 1980. The Viking Age in Denmark: Saxon England from the Seventh to the Tenth The Formation of a State. London: Duckworth. Century”. The English Historical Review 70(277): Reid, Rachel Robertson. 1920. “Barony and Thanage”. 529–549. The English Historical Review 35(138): 161–199. Löfving, Carl. 1984. “Förändringar i maktstrukturen Reynolds, Susan. 1992. “Bookland, Folkland and Fiefs”. kring Östra Skagerrak vid kristendomens införande: Anglo-Norman Studies 14: 211–227. En tvärvetenskaplig utgångspunkt för en arkeologisk Roberts, Jane. 2000. “The Old English Vocabulary of studie”. Fornvännen: Tidskrift för Svensk Nobility”. In Nobles and Nobility in Medieval Antikvarisk Forskning 79: 93–103. Europe: Concepts, Origins, Transformations. Ed. Löfving, Carl. 2001. Gothia som dansk/engelskt Anne J. Duggan. Woodbridge: Boydell Press. Pp. skattland: Ett exempel på heterarki omkring år 1000. 69–84. Göteborg: Göteborg Universitet, Institutionen för Roffe, David R. 1989. “From Thegnage to Barony: Sake Arkeologi. and Soke, Title, and Tenants-in-Chief”. Anglo- Lund, Niels, & Kai Hørby. 1980. Samfundet i vikingetid Norman Studies 12: 157–176. og middelalder 800–1500. København: Gyldendal. Roffe, David R. 2000. Domesday: The Inquest and the Maitland, Frederic William. 1921. Domesday Book and Book. Oxford / New York: Oxford University Press. Beyond: Three Essays in the Early History of England. Sawyer, Peter H. 1987. “The Bloodfeud in Fact and 2nd edn. Cambridge: Cambridge University Press. Fiction”. In Tradition og historieskrivning: Kilderne Moltke, Erik. 1976. Runerne i Danmark og deres til Nordens ældste historie. Ed. Kirsten Hastrup & oprindelse. København: Forum. (English trans.: Preben Meulangracht Sørensen. Aarhus: Aarhus Moltke, Erik. 1985. Runes and Their Origin. Denmark Universitetsforlag. Pp. 27–38.

49

Sawyer, Peter H. 1988. The Making of Sweden. Evropy 28. Moscow: Dmitri Pozharski University. Alingsås: Viktoria Bokförlag. Pp. 277–286. Senecal, Christine Kathryn. 1999. “The Regional Sukhino-Khomenko, Denis. 2018. “Historiographical Aristocracies of Late Anglo-Saxon England”. Wishful Thinking and Debates on Some Early Unpublished Ph.D. thesis, Boston College. Medieval Social Phenomena in the North Sea Region”. Senecal, Christine Kathryn. 2000. “Keeping up with the In Vox Medii Aevi 2(3): 38–69. Available at: http:// Godwinesons: In Pursuit of Aristocratic Status in voxmediiaevi.com/2018-2-sukhino-khomenko/. Late Anglo-Saxon England”. Anglo-Norman Studies Syrett, Martin. 1998. “Drengs and Thegns Again”. 23: 251–266. Saga-Book of the Viking Society for Northern Shchavelev, Aleksei Sergeevich. 2017. “Derzhava Research 25: 243–271. Rurikovichej v pervoj polovine X veka: khronologija, Thacker, Alan T. 1981. “Some Terms for Noblemen in territorija, socialnaja struktura”. Studia Slavica et Anglo-Saxon England, c. 650–900”. Anglo-Saxon Balcanica Petropolitana 1(21): 82–112. Available at: Studies in Archaeology and History 2: 201–236. http://slavica-petropolitana.spbu.ru/en/arkhiv/179- Thrane, H., & Michael Lerche Nielsen. 1998. studia-slavica-et-balcanica-petropolitana-2017- “Glavendrup”. In Reallexikon der Germanischen 1/miscellanea-miscellanea/402-shchavelev-aleksei- Altertumskunde XII. Ed. Heinrich Beck et al. Berlin the-rurikids-polity-in-the-first-half-of-the-10th- / New York: Walter de Gruyter. Pp. 195–200. century-chronology-territory-social-structure.html Tolley, Clive. 2011. “In Defence of the Vanir”. RMN (accessed 6th December 2018). Newsletter 2: 20–29. Simek, Rudolf. 2010. “The Vanir – An Obituary”. RMN Torstendahl, Rolf. 1981. “Minimum Demands and Newsletter [1]: 10–19. Optimum Norms in Swedish Historical Research Skaldic Poetry of the Scandinavian Middle Ages, I–IX. 1920–1960: The ‘Weibull School’ in Swedish 2007–present. Turnhout: Brepols. Historiography”. Scandinavian Journal of History Snook, Ben. 2015. The Anglo-Saxon Chancery: The 6(1–4): 117–141. History, Language and Production of Anglo-Saxon Turner, Sharon. 1801. The History of the Anglo-Saxons: Charters from Alfred to Edgar. Woodbridge: Comprising the History of England from the Earliest Boydell Press. Period to the Norman Conquest, III. London: Stefanovich, Petr S. 2012. Boyare, otroki, družiny: Longman, Hurst, Rees, Orme & Brown. Politicheskaya elita Rusi v X-XI vekakh. Moscow: Williams, Ann. 2008. The World before Domesday: The Indrik. English Aristocracy 900–1066. London: Continuum. Stenton, Frank Merry. 1932. The First Century of Wormald, Patrick. 2001. The Making of English Law: English Feudalism, 1066–1166. Oxford: Clarendon King Alfred to the Twelfth Century, I. Oxford: Press. Blackwell Publishers. Strid, Jan Paul. 1986. “Runic Swedish thegns and drengs”. In Runor och runinskrifter: Föredrag vid Websites and Internet Databases Riksantikvarieämbetets och Vitterhetsakademiens Dictionary of Old Norse Prose. Available at: http://onp.ku.dk symposium 8–11 september. Stockholm: Almqvist & (accessed 20th January 2018). Wiksell International. Pp. 301–316. Dictionary of Old English Web Corpus. Antonette Stubbs, William. 1874. The Constitutional History of diPaolo Healey with John Price Wilkin & Xin Xiang England, in Its Origin and Development, I. London: (compilers). Toronto: Dictionary of Old English Clarendon Press. Project, 2009. Available at: Sukhino-Khomenko, Denis. 2014. “Sociology of Late http://tapor.library.utoronto.ca/doecorpus/ (accessed Anglo-Saxon thegns: From ‘Servants’ to 20th January 2018). ‘Highborn’?” Unpublished paper presented at the Electronic Sawyer. Available at: http://www.esawyer.org.uk first stage of the PhD and post-doctoral training (accessed 20th January 2018). th th school Dynamic Middle Ages II (Moscow, 7 –10 Gammeldansk Ordbog. Available at: October 2014), organised by the Centre for Medieval http://gammeldanskordbog.dk (accessed 20th January Studies and the Department of History of the 2018). National Research University Higher School of Samnordisk runtextdatabas. Available at: Economics (Moscow) in collaboration with the http://www.nordiska.uu.se/forskn/samnord.htm Heinrich Heine University (Düsseldorf, Germany) (accessed 20th January 2018). and the White Rose University Consortium (UK). A The Skaldic Project. Available at: http://skaldic.abdn.ac.uk version of this paper has been submitted for a (accessed 20th January 2018). publication of proceedings from the training school. Thesaurus Indogermanischer Text- und Sprach- Sukhino-Khomenko, Denis. 2016. “Poluchateli, materialien (TITUS). Available at: http://titus.uni- topografia i dinamika proizvodstva anglo- frankfurt.de (accessed 20th January 2018). saksonskikh korolevskikh gramot 900–1066 v svete statistiki”. In Drevnejshie gosudarstva Vostochnoj

50

Would You Like to Submit to RMN Newsletter?

RMN Newsletter is an open-access annual Reports. The venue emerged with an publication that sets out to construct an orientation toward presenting information informational resource and discourse space about events, people, activities, developments for researchers in diverse and intersecting and technologies, and research which is disciplines. Its thematic center is the ongoing or has been recently completed, with discussion and investigation of cultural many contributions oriented to discussion phenomena of different eras and the research and/or engaging in discourse opened in earlier tools and strategies relevant to retrospective issued of RMN Newsletter or in other methods. Retrospective methods consider publications. some aspect of culture in one period through The success of this publication as both a evidence from another, later period. Such resource and discourse space is dependent on comparisons range from investigating the participation of its readership. We also historical relationships to the utility of recognize the necessity of opening contact analogical parallels, and from comparisons with, and being aware of, the emerging across centuries to developing working generation of scholars and welcome models for the more immediate traditions summaries of on-going and recently completed behind limited sources. RMN Newsletter MA and PhD research projects. welcomes and encourages its readership to If you are interested in making information engage in this discourse space and it also about your own work available or promotes an awareness that participation will participating in discussion through comments, support, maintain and also shape this responses or article contributions, please send emergent venue. your contributions in *.doc, *docx or *.rtf The journal features peer-reviewed articles format to Frog at: under the heading Communications or the title editor.rmnnewsletter[at]gmail.com. of a thematic number, while discussions, For more information and access to earlier reports on current research, projects, issues, please visit our web-page at conferences and so forth are organized under https://www.helsinki.fi/en/networks/retrospe the heading Comments, Perspectives and ctive-methods-network.

Published by Folklore Studies of the Department180 of Cultures, University of Helsinki