Wednesday 27 February
1. Nicholas Evans, Stephen Levinson & Kim Sterelny: Introductory remarks (20 minutes)
2. Joan Silk: The phylogenetic roots of kin bonds
In order for evolution to favour nepotistic biases in behaviour, individuals need to have opportunities to interact with their relatives and to be able to identify them. Both these requirements impose constraints on the evolution of kinship bonds in other species of primates. For non-human primates and many other mammals, maternal care generates close ties between mothers and offspring, and among litter-mates. For species in which mother-offspring bonds extend beyond weaning, associations with the mother also generate familiarity among siblings of different ages. For species with female philopatry, mother-offspring ties provide the foundation for dense matrilineal kin networks and lifelong relationships.
3. Rob Wilson: Incest, incest avoidance, and attachment: Revisiting the Westermarck Effect
Over 120 years ago, Edvard Westermarck defended the idea that there was a developmental mechanism present in humans that facilitated incest avoidance. In virtue of this mechanism, those who grow up together come to experience sexual aversion or inhibition when they are adults. The claim was lampooned by cultural anthropologists for most of the twentieth-century before being revived (and the term 'Westermarck Effect' named) by Robin Fox in the 1960s and explored more systematically by Arthur Wolf and others more recently. In this talk I articulate and defend a version of the Westermarck Effect, integrating existing clinical, phylogenetic, and philosophical dimensions to the literature in the biological and social sciences on incest avoidance and raising some broader questions about explanation in cultural anthropology. By focusing on care-based attachment in the primate lineage, my formulation of the Effect suggests the power of a phylogenetic argument that is already widely accepted by primatologists but not by cultural anthropologists. In effect, this talk will aim to draw together insights of the psychiatrist Mark Erickson with the explorations of the “deep structure” of kinship that the primatologist Bernard Chapais has engaged with over the past ten years.
4. Robert Layton: Lines of evidence for something that leaves no trace?
The aim of this workshop is to test the hypothesis that ‘The evolution of more complex systems is likely to have played a key role in expanding the social universe of early humans or their hominin forebears.’ How does one talk about kinship in deep time, and in particular the archaeological signatures of the elaboration of the complex forms of kinship known in human social life, and of how they might have emerged from simpler systems. This is a difficult topic, but I will discuss two methodological steps through which to approach the question of recognising archaeological symptoms for the elaboration of complex forms of kinship:
(i) How do we extend kinship: Different modes of fictive kinship: relatives by marriage (hunter-gatherers generally and Chinese villagers), namesakes (Ju/’hoansi, Netsilik), extended genealogies (West Africa - Daryl Forde’s hypothesis and the Tiv), section systems (Australia).
(ii) When might those uses have become functional in the course ofhominimevolution? I will briefly review the adaptive advantages of social living > of fission-fusion behaviour in social groups > changes in the length of time members of the community spend apart (comparing chimpanzees and humanhunter gatherers) and the scale of interaction. Here I will draw on Layton, O’Hara and Bilsborough (2012).
Although both chimpanzee and human hunter-gatherer social groups display fission-fusion behaviour, a key contrast between chimpanzees and humans lies in the difference between their social behaviour as communities. The chimpanzee community occupies a small area and, although members frequently separate into small foraging parties, all members encounter each other frequently, perhaps even on a daily basis. The human hunter-gatherer community lives at a much lower population density and its members are dispersed between different bands, which meet much less frequently.
Bernard Chapais (2017) has recently published a paper which notes that chimpanzees show much less kinship recognition than primate species that are more distantly related to humans. It’s tempting, then, to take the approach that I’ve used before, of using archaeological and bio-anthropological evidence for when modern human patterns of social interaction may have (gradually) appeared in the course of hominid evolution. Unlike Chapais and many others, I will not speculate on structures of social organisation that no longerexist(disregarding questions such as whethermatrilinyprecededpatriliny: see Layton  on method).InsteadI will focus on the value of kinship recognition in face-to-face situations in contrast to recognising kinship with people we meet less often.
5. Kim Sterelny: Kinship and the origins of multi-level society
Abstract: There is a very striking difference between even the simplest ethnographically known human societies and those of the chimps and bonobos. Chimp and bonobo societies are closed societies: with the exception of adolescent females who disperse from their natal group and join a nearby band (never to return to their group of origin), a pan band is the whole social world of the agents who make it up. That is not true of forager bands, which have fluid memberships, regular associations with neighbouring bands, and which are components of a larger social world. The open and fluid character of forager bands brings with it many advantages, so the stability of this more vertically complex form of social life is not difficult to explain, once it establishes. But how did it establish if, as is likely, earlier hominin social worlds resemble those of our close pan relatives in the suspicion (even hostility) of one band toanother.How did hominin social organisation transition from life in closed bands, each distrustful of its neighbours, to the much more open social lives offoragers.In thistalkI will discuss and attempt to synthesise two approaches to this problem: one ecological, based on the work of Robert Layton and his colleague; and another that is organised around an expansion of kin recognition, an idea primarily driven by Bernard Chapais.
Thursday 28 February
1. Alice Gaby: Tutorial on Kuuk Thaayorre kinship
A Kuuk Thaayorre speaker can use kinterms to refer to any other member of Thaayorre society. What’s more, they can select between at least four different kin terms for any given relationship. For instance, to talk about my brother (Jack) to a third party, I would describe him as pam kanam (which can only mean ‘older brother’). But to address Jack directly, I would say Wanhn! (which means ‘older brother, son’s son or brother's son's son'). Following the death of our father, I’d refer to Jack as kaal-mangk (which means ‘one bereaved of a father, father’s brother or father’s sister’). And lastly, I can refer to Jack with a hand sign, touching my shin (which means ‘older brother, younger sister, father’s father’s brother, brother’s son’s daughter’, among other relationships). This session will explore the genius of the Kuuk Thaayorre kinship system; the logic of how these four kinds of kin terms map onto one another, and how they relate to — and are exploited for — social regulation and behaviour.
2. Ian Keen & Ron Planer: Referring to kin - the evolutionary pre-requisites of kinship terminologies
Kinship systems play a foundational role in organising human social life by underpinning many of our basic rights, privileges and obligations. That remains true even in the modern industrial world, but it is especially true in hunter-gatherer societies. Hence, explaining the evolutionary origins of human kinship systems is integral to the project of explaining human social evolution more generally. In this paper, we sketch an account of the minimal cognitive, linguistic, and socio-ecological requirements needed for the establishment of basic kinship terms in hominin groups. Such an account provides clear signposts for the construction of detailed and empirically supported theories of the evolution of human kinship systems.
3. Steve Levinson (with Marisa Casillas): Ruminations on the functions and cognitive prerequisites of kinship
The evolution of kinship systems might, one supposes, have something to do with their efficacy, e.g. as contributions to evolutionary fitness. However, those of us who do not live in kinship-basedsocieties,may have restricted intuitions here. The same goes for what it takes to mentally encompass complex kinship systems. I propose to throw some light on this by examining the role of a complex double-descent system used on Rossel Island, PNG, and some new data on how long it takes for kids to acquire this system.
First, the functions. Proximately, the nuclear family relations (M, F, B, Z, S, D, H, W) structure authority, incest avoidance, inheritance, while giving a unique identity to individuals. Extended reckoning beyond the natal family has many and various functions depending on the system. Thus unilinear descent systems uniquely work to delineate non-overlapping but nested descent groups – the basis of ‘segmentary opposition’, as mobilised in warfare and feud. Systems that specify marriage partners (e.g. cross-cousins) produce closed worlds with exchange relations. Systems that specify marriage outside clans (Lévi-Strauss’s ‘complex systems’) produce distributed ties that can constrain conflict, and offer friends in dangerous territory.
On Rossel Island, where sorcery is an omnipresent danger, having safe harbours elsewhere (your exported sisters if you are a man) is essential. Clan totems even allow the identification of pseudo-kinsmen onneighboringislands, a kind of passport system. All these functions have been well explored in the now unfashionable ethnographies of the first half of the twentieth century.
Second, the cognitive prerequisites. Kinship terms are anomalous linguistic systems: they look like nouns (“uncle”) denoting persons (like proper names or titles, e.g. “chief”), but infactthey are relational predicates with deictic properties – my uncle isn’t yours. This makes the acquisition of kinship by children potentially difficult. Add to that the fact that kin-terms offer a recursive calculus of indefinite extension. Western children hardly understand this system at age five or six. IncontrastRossel Island children can already trot off ancestors to as much as 10 generations deep, and calculate the kin-term appropriate for e.g. MMBD, or triangulate from dad’s usage to the term they should apply.
This is impressive because the system makes one switch perspective every time you traverse a link through an opposite-sex relation – thus a male ego must switch to the mother’s way of reckoning MB, because the ‘brother’ term actually means cross-sex sibling: for a male linking relative (as in FB) mbwo means ‘brother’ but for a female linking relative (e.g. MZ) the term means ‘sister’. Compounding the difficulties, this is a system with Crow-type skewing, so MMB = ‘mbwo’ = B for a man. The very early acquisition of this complex system both underlines its functional importance and the essential role of extended childhood in human evolution. Everyfive-year -old already knows who are not possible marriage partners, but even 17-year-olds can make calculation errors. We tested 70 kids between five and 17. As far as I know, this is amongst the first systematic data on the acquisition of a complex kinship system. Data like these help to constrain our speculations about the origins of human kinship systems.
4. Patrick McConvell: The open end of kinship dynamics
Kinship evolution studies have tended to be based on models of closed, symmetrical systems, especially in speculations about early systems. In fact certain types of kinship systems are based on migration, exogamy and open interaction of groups; and this characteristic is even more apposite when we try to look at early systems during the spread of humans across the globe. This paper looks first at how the ancestors of kinship studies in Australia, Fison and Howitt (Gardner & McConvell 2015) , contributed to the general theory of kinship evolution, although rarely acknowledged. It then moves on to how certain types of kinship system are attributable to phases of migration and open interaction, using historical linguistic evidence.These include
These are illustrated from Australia, but are also to be found elsewhere across the world, such as in North America, Africa etc. Dousset (2019 in press) points out that the first colonisation of Sahul must have been largely based on open systems, not the closed Dravidian-like systems that are favoured by many anthropologists discussing such questions.
Dousset, Laurent. 2019. Systèmes ouverts et systèmes fermés: reconstruire l’organisation des sociétés de la préhistoire, Chapter 3 of Dousset, Park & Gulle-Escuret Parenté, écologie et historie. STE editions
Gardner, Helen & Patrick McConvell. 2015.Southern Anthropology: a HistoryMi of Fison and Howitt’s Kamilaroi and Kurnai. Palgrave MacMillan.
Friday 1 March
1. Patrick Heady: The cognitive organisation of kinship practice - universals and variation in the ethnographic present
Kinship networks can be represented algebraically and graphically in two different ways: as individual people connected by procreative unions, and as procreative unions connected through particular people. The second kind of representation turns out to be convenient when analysing some aspects of descent and marriage. In this paper I argue that these alternative representations reflect two distinct but complementary conceptions of kinship ties, and that the ability to operate with either, and to switch between them, is a fundamental feature of human kin cognition. The paper considers the symbolic and practical contexts in which each conception is relevant. The final part looks at how these non-linguistic conceptions are reflected in terminology – with particular reference to Morgan's distinction between ‘descriptive’ and ‘classificatory’ terminologies.
2. Fiona Jordan: The cultural evolution of kinship diversity
Humans have a variety of mating and marriage systems, and societies vary in the ways in which they classify family. This variation is not without limits, and explaining constraints and diversity invites a multi-level evolutionary perspective. I will describe the VariKin project, a multidisciplinary approach to understand the patterned variation in human kinship systems. The team brings together theory and method across anthropology, linguistics, evolutionary biology, cognitive science, and computational methods. We have results from large cross-cultural analyses using comparative phylogenetic methods, where we infer ancestral states and test classic anthropological theories of adaptive drivers. We use web, spoken, and written corpora of over 40 Indo-European languages to study patterns of kin-term frequency of use, particular the relationship between use and change. FInally, we explore insights into child acquisition and proficiency from fieldwork in a Datooga community in Tanzania, as well as child acquisition corpora across Indo-European languages. Our framework takes inspiration from Tinbergen's framework to consider proximate and ultimate drivers of diversity, with the aim of illuminating this longstanding enquiry into human variability.
3. Nicholas Evans, Wolfgang Barth, Simon Greenhill & Kyla Quinn: Evolution and coevolution in kinship systems - testing dependencies against global data
In the languages of all societies, the vast number of logically distinguishable 'kin types' (e.g. father's younger sister, father's older sister, mother's younger sister, mother's older sister) are mapped onto a smaller and more tractable set of kin terms (e.g. 'aunt' for all the above terms in English). The kinship system of any language can thus be seen as a particular matrix of 'equivalences' or 'syncretisms' characterisable by which kin types are named by identical or similar terms, and which are differentiated. Since the pioneering work of Morgan, Fison and Howitt in the nineteenth century, there has been a great deal of work on the typology of kinship systems with detailed surveys of languages from around the world, running from the hundreds to the thousands (depending on how comprehensive the number of kin terms surveyed is). Two common threads to this enterprise, each relevant to evolutionary models of kinship, have been:
(a) the assumption of system constraints leading to internal co-selection between sub-parts of the system (e.g. if FB=F≠MB and MZ=M≠FZ there is likely to be an opposition between 'parallel cousins' like FBS, treated as siblings, and 'cross cousins' like MBS, which are terminologically distinct. In the maximal case, this led toessentialising characterisations of systems such as 'Dravidian' (the case just given), 'Eskimo' (as in English), 'Sudanese' (FB≠F≠MB), Hawaiian (FB=F=MB) etc. In these putative types, large numbers of sub-system properties are supposed to coexist (e.g. patterns of parent/uncle/aunt terms with patterns of cousin terms with patterns of grandparent terms).
(b) the assumption that particular patterns of syncretism can be related to marriage rules (e.g. cross-cousin marriage with a Dravidian kinship system). Here co-selection occurs between a marital practice and a pattern of kinship terminology.
Beyond this, evolutionary interactions of other types have long been proposed, e.g. in terms of the order of development of one system from another, and the correlation of particular patterns of kinship terminology with types of social organisation (e.g. matrilineal vs patrilineal clan structures).
As in most studies of evolution, the complex patterns of selection, whether internal or external, operate with time-lags and in a stochastic manner. In order to evaluate claims about (co)evolutionary causality, it is therefore crucial to use a largedata-basewhere elements within each language are maximally differentiated, where the effects of phylogeny, areal contact and other possible causes of convergent evolution can be registered, and where claimed co-selection effects between different parts of the system can be tested against known phylogenies. In thistalkwe outline the Parabank:Kinproject, a large database of kinship terms from languages around the world, and illustrate its use to test a range of hypotheses about co-selection effects within kinship sub-systems, and to reconstruct back to prior systems.
4. Francesca Merlan. The devolution of kinship
Australian Aboriginal societies were famous for the universalism of kinship: every person one came into contact with could be categorized as a kinsman of one type or another. Kinship was/is classificatory, and persons were of course not equally close to all those whom they called by the same category term. As these societies change, as Aboriginal people come into wider and more regular contact with others originally more geographically and socially distant – and especially, as indigenous people become small groupings within a much larger society – the universalism of kinship changes. More kinds of relationship are understood on other bases than kinship (even as there may continue to be some classification of obvious non-kin as kin). Changes go on, at the same time, in kin usage among Aboriginal people, not only in relation to outsiders. Kin terminologies change as (many) indigenous languages evanesce and are substituted by English-based forms; even more, the nature of kin practice (sharing, obligation, etc.) changes.
In Australia, kinship was partly supported by, and intersected with, social category systems (subsection, sections, moieties and the like). What forms of change have been noted by observers? What happens to social category terms? What happens to kin terminological uses? What kinds of other changes occur in social practice alongside and as part of changes in both of those classifying systems (e.g. in avoidance practices of any kind; in relative age or usages marking juniority/seniority; in distinctions oflineality/collaterality, and so on)?
There has been little discussion of how to view and analyse what occurs as formerly universalist kinship becomes less so, despite some recent work on the replacement, and retention, of kin term usages (Dickson 2015). There is some documentation of changes in social practices involving the classification and designation of others in kin and other terms, and (related, or at least accompanying) changes in social practice (e.g. J.H. Bell, 1959).
One task of this paper will be to survey information concerning such changes, and to discuss, perhaps move towards systematising, the terms in which one might look at shifts in universalism versus more limitedkinusage, understood to involve shifts in category system as well as in other structures and practices. How may one understand the implications for persons and collectivities of these changes?