Steve Pinker gets the memo (sort of)

The cover story of the the August number of the New Republic is a piece by Steven Pinker entitled “Strangled By Roots: The Genealogy Craze in America”:http://www.tnr.com/doc_posts.mhtml?i=20070806&s=pinker080607. Pinker ought to be given credit as an academic who writes for a popular audience, and there is no doubt that his work is easy to read and always has a clear take away message. These days, though, he is venturing further and further afield from his area of expertise and one gets the feeling that he is suddenly encountering brand new intellectual territory. Those of us in the social sciences for whom this is well-worn ground are, of course, happy that he has finally gotten the memo, but disappointed that hasn’t read it very carefully.

The overall plot of “Strangled By Roots” will be familiar to any one familiar with evolutionary psychology: a New Field Of Research has been opened up that sheds Scientific Light on a previously untheorized and salaciously quirky bit of human life. The Social Scientists, of course, with their Social Science Models, have got it wrong, but luckily New Experiments have revealed the hidden evolutionary basis of said quirky behavior. Unfortunately — alas! — however adaptive this behavior once was, it no longer suits the rigors of modern life and is currently the source of many social woes.

This time around its kinship. In the article Pinker claims that “for all its fascination, kinship is a surprisingly neglected topic in the behavioral sciences.” While “many social scientists have gone so far as to claim that kinship is a social construction with no relation to biology” others disagree. “Genetics and evolutionary theory,” Pinker says, “predict that the biology of kinship should have biased our thoughts and emotions about relatives in several ways” — for instance, that we like to share resources with them (this helps perpetuate their genes, including the genes we share with them).

Ancient humans, of course, could not do DNA testing to find people who shared genes with them, so “instead we rely on cues that in the evolutionary past tended to correlate with relatedness” such as living together. And indeed, “recent experiments by Debra Lieberman, John Tooby, and Leda Cosmides have shown that two kinds of life experiences are crucial in triggering family feelings towards siblings… one consists of observing the infant being cared for by one’s mother…. the other is having grown up in the same household as the sibling”.

Unfortunately, says Pinker, this previously-adaptive method is now problematic for two reasons. First, an ‘obsession’ with genealogy it is actually a pretty lousy method of reckoning kin in a world where the amount of genetic material you share with collateral relatives plummets exponentially the higher you ascent a family tree to locate a shared ancestor. Second, “successful [nonkin] coalitions often try to co-opt family feelings by tricking the brain into perceiving collateral kin” so that, for instance, the Mafia uses kin-inspired idioms of brotherhood to cement ties between unrelated people. In sum, Pinker writes, “Outside a small family circle, the links of kinship are biologically trifling, vulnerable to manipulation, and inimical to modernity.”

Pinker’s argument sounds plausible at first — especially if you don’t know anything about the centuries-old literature on kinship or lack in-depth knowledge of the cultural complexity of ours species. In Pinker’s case the problem is mostly naivete. “A martian reading a textbook in psychology would get no inkling that human beings treated their relatives differently from strangers,” writes Pinker, as if to demonstrate that this is a sign of how “surprisingly neglected” kinship is as a field. But obviously, the lack of kinship in pyschology textbooks points to a problem with psychology and not to science. Pinker’s failure to review the literature on the topic can be blamed on many things, but our failure to write it is not one of them.

Of course this is just a popular article and so Pinker can perhaps be excused from citing “Nonagnates Among The Patrilineal Chimbu.” But simply because your work lacks a scholarly apparatus doesn’t mean that you don’t have to do your homework. And it is clear that Pinker and his colleagues are slowly trying reinvent the wheel, spoke by spoke. I have not read the research by Lieberman et. al. that Pinker cites, but if they are seriously conducting ‘experiments’ to deduce that people raised together feel related, I wouldn’t be surprised if their next ‘experiment’ proves that human beings eat by first chewing and then swallowing. I understand the impulse to make bench science the gold standard of scientific objectivity, but do we really need grant money going to work of this sort?

But let me get to the main point: there are two main problems with Pinker’s argument. First, there is that we have no evidence of what social organization was like deep in our evolutionary past. Of course we can imagine what they might have been like, but speculation is not science — especially for someone sufficiently serious about intellectual rigor that they feel the need to conduct experiments to prove the obvious fact that people who are raised together feel related. So his claim that feelings of kinship were once nontrivially adaptive in the evolutionary past but no longer are is in fact based on speculation. There is nothing wrong with speculation — indeed, it is all we have to go on with in some cases — but this point needs to be flagged.

The second problem is with Pinker’s claim that kinship is currently no longer adaptive. The problem here is that Pinker, as philosophers say, ‘proves too much’. For, as he himself shows and anthropology has already demonstrated, folk theories of relatedness and accurate biogenetic reckoning are so loosely coupled as to be only tenuously connected. In fact they are so tenuously connected that one wonder why he thinks they are or should be connected at all, except for his assumption (based on speculation) that they must have been in the past. Let’s take a closer look.

What exactly is the phenomenon we are examining here? Throughout the article Pinker vacillates between providing an account of the American craze for genealogy using DNA based testing and a more universal “fascination with ancestry [which] has long been part of the human condition”. But which is it? Americans obsession with roots is clearly distinct and different from obsessions with roots in other times and places. What accounts for this local variation? Pinker cannot simultaneously claim American obsession with roots is both culturally distinct and based on species-wide evolved psychologies.

And is an obsession with genealogy a cultural universal? Pinker moves rather to quickly from ‘experiments’ proving the importance of siblingship and shared biogenetic substance to a focus on lineage. How, evolutionarily, does this slippage occur? And does it occur? Pinker should provide some evidence of the universality of this train, but in fact the only main ‘non-Western’ sources that he draws on are ‘the Bible,’ a description of cousin marriage in Iraq published in The American Conservative, and scenes from The Godfather. The best thing that can be said of this ‘evidence’ is that its juxtaposition of Abrahamic genealogy and Muslim marriage helps remind us how ‘Western’ the Islam is — Isaac and Jacob, after all, married their cousins.

But are the begats of the bible a failed attempt at biogenetic reckoning? As Nancy Jay writes in Throughout Your Generations Forever: Sacrifice, Religion, and Paternity, the genealogies in the Hebrew Bible (and the Gospel of Matthew, for that matter) manage to portray men producing male heirs without any female intervention whatsoever — thus Abraham begets Isaac, who begets Jacob who begets Isaac. Sarah, Rebecca, Rachel, and Leah — who did the physical begetting, let’s not forget — completely disappear. It is for this reason that reform Jews add the matriarch back into their daily prayers. Even if you dislike Jay’s mix of furniture-chewing feminisim and stratospheric social theory (I think it is delightful, personally) the point has been made by researchers ranging from Emrys Peters to Andrew Shryock to J. David Schloen: kinship in this area of the world not only fails to reckon genetics accurately, the whole point of it is to elide how reproduction actually works and emphasize agnatic ties to the exclusion of collateral ones.

For an even clearer example of this fact, consider the material from Sport of Kings, Rebecca Cassidy’s delightful ethnography of British thoroughbred breeding. Although ostensibly about improving the performance of their horses, Cassidy demonstrates that breeder’s theories of blood and pedigree have rather more to do with the British class system than optimizing their horse’s time. Here we have a cultural system in which hundreds of years of breeding have failed to produce increased performance in their horses but have done a wonderful job of reproducing the social structures of the world of high-end racing. This is an example of a group explicitly attempting accurate biogenetic reckoning which end up doing something else altogether.

Both of these examples suggest what all anthropologists have long known — that there is a ‘hidden force’ at work in people’s obsession with relatedness. But that force is culture, and distinct and recognizable cultural formations influence thinking about relatedness.

Consider the Simpsons, for instance. In classic ‘eskimo’ style kin reckoning (what Americans use) Bart and Lisa are siblings, and Ling, Selma’s daughter is his cousin (Selma is Marge’s sister, thus making Ling Bart’s mother’s sister’s daughter or, as we say in the business, his MZD). But in a ‘Hawaiian’ style system all members of the generation above you (the ‘ascending generation’) are classified as siblings themselves, so Selma would be Bart’s mother and Ling would be his sister. In an Iriquois-style system Selma would be Bart’s mother and her children would be his siblings but Abbie (Homer’s sister) would be considered his aunt and her children his cousins. Why? Because Iriquois-style kinship practices bifurcate merging, in which same-sex siblings are identified but cross-sex siblings are not. Father and father’s brother are both called ‘father’ and mother and mother’s sister are both called ‘mother’, but mother’s brother is called ‘uncle’ and father’s sister is called ‘aunt’.

And of course people can even have elaborated theories of relatedness that aren’t tied to genetics at all. James Leach has shown that on the north coast of New Guinea imagine people being grown, like plants, out of the territory where they live. For them relatedness is about rootedness in land. The Malay fishermen studied by Janet Carsten explicitly understand themselves to be related by shared nurturance, not biogenetic substance. These sorts of examples could be multiplied.

They say that everything looks like a nail to a guy holding a hammer, and Pinker’s obsession with evolutionary adaptiveness means that refuses to see this bigger picture. He claims that it is a “paradox” that kinship should be so out of step with his hypothesis about biology and kinship. But the big question is not why our systems of relatedness are bad reckoners of biology, but why Pinker thinks that they are or ought to be reckoners of biology at all. Relying on an unproven assumption about our deep past, Pinker imputes a goal to kinship systems which they themselves have never claimed to try to solve. To a scientist — who prefers elegant, parsimonious explanation — it appears not that there is a ‘paradox’ between the data and Pinker’s hypothesis to be explained away, but simply that his hypothesis is incorrect.

In fact, given the data, it makes much more sense to say that culture organizes biology, rather than the other way around. Or, to be more specific, that there is a complex interaction between folk theories of relatedness and genetics. Saying this is not a call for postmodern epistemological nihilism or wooly-headed feel-good relativism, but a demand for serious research. The problem is that we are not likely to get it from Steven Pinker and his colleagues. However titilating Pinker’s brand of pop evolutionary psychology is, it lacks what he prides himself on most — scientific rigor and a reliance on the extant data on human social organization. This work is being done and it is important — however it is being done by anthropologists, biologists, and researchers in a host of other fields. We’ve been doing it for over a century now…. ever since we got the memo.

Rex

Alex Golub is an associate professor of anthropology at the University of Hawai‘i at Mānoa. His book Leviathans at The Gold Mine has been published by Duke University Press. You can contact him at rex@savageminds.org

26 thoughts on “Steve Pinker gets the memo (sort of)

  1. Argh, WordPress ate my comment. In brief:
    Nice thoughtful piece. Can’t get to Pinker’s, but he and the hardline EP’s don’t speak for all evolutionarily-minded anthropologists. There *is* good work on evolution and kinship out there, it doesn’t involve “well, duh” science, and sadly, the PR machine of EP once again works against those of us doing (for example) behavioural ecology, cultural evolution, anthropological demography etc. But it’s nice to see you recognise that in your final paragraph.

    Teach me to spend half an hour on a long reply.

  2. BLOCKQUOTE
    “But let me get to the main point: there are two main problems with Pinker’s argument. First, there is that we have no evidence of what social organization was like deep in our evolutionary past. Of course we can imagine what they might have been like, but speculation is not science…”
    /BLOCKQUOTE

    I’m going to nitpick a little:

    We do have evidence of what social organization was like deep in our evolutionary past. You are overstating the case. Take an obvious example: we know that deep in our evolutionary past there were no states. We also have a lot of reason to believe that it was based on kinship.

    BLOCKQUOTE
    “Rex writes: especially for someone sufficiently serious about intellectual rigor that they feel the need to conduct experiments to prove the obvious fact that people who are raised together feel related.”
    /BLOCKQUOTE

    Its a little easy to point fingers and laugh. I haven’t read the work so I can’t make any specific comments, but both the science in general and anthropology in particular have been extraordinarily useful precisely because they do not take the obvious for granted: Of course mothers do X. Of course kinship is about biology. Of course non-whites are inferior to whites. I think that we should tread carefully when engaging in this kind of criticism.

    Consider the Simpsons, for instance. In classic ‘eskimo’ style kin reckoning (what Americans use) Bart and Lisa are siblings, and Ling, Selma’s daughter is his cousin (Selma is Marge’s sister, thus making Ling Bart’s mother’s sister’s daughter or, as we say in the business, his MZD). But in a ‘Hawaiian’ style system all members of the generation above you (the ‘ascending generation’) are classified as siblings themselves, so Selma would be Bart’s mother and Ling would be his sister. In an Iriquois-style system Selma would be Bart’s mother and her children would be his siblings but Abbie (Homer’s sister) would be considered his aunt and her children his cousins. Why? Because Iriquois-style kinship practices bifurcate merging, in which same-sex siblings are identified but cross-sex siblings are not. Father and father’s brother are both called ‘father’ and mother and mother’s sister are both called ‘mother’, but mother’s brother is called ‘uncle’ and father’s sister is called ‘aunt’.

    Although you (correctly) criticize biological or genetic interpretations of modern kinship, your examples of different kinship terminologies here is still based upon (or overlaid upon) a single common genealogical grid of kintypes, one usually taken as implying some system or theory of (biological) relatedness via the idiom of descent. Kintype products like MZD are elucidated by genealogical tracing.

    So unless you depart from interpretation, you risk sneaking in what you threw out a moment before. Merely replacing common kinship terms like sister and daughter with letters like Z and D, doesn’t change anything of substance (and one wonders why american/british kinship terms are used as its basis?).

    One can argue of course that this common genealogical grid is not precisely a theory of biological descent and relatedness. But then what is it? The only other real option on the table is that it is a cultural construct (perhaps only loosely coupled to biology, if at all). But if that is the case, what is the basis for using it as universal grid? It is true that it is very useful anyway, but analytically its on shaky ground.

    One might argue that kin type products like MZD don’t have to be anything about genealogy (much less biology). Instead they are something much more abstract. But then again, what is it? Why should it be used as the universal standard? And for that matter, why should we bother with it at all?

    If it is universal, then we are lead to accept that kinship terminologies are merely specific encodings of a biological theory of relatedness. If its culturally specific, then our analytic apparatus collapses.

    These concerns are what motivated David Schneider to write his damning A Critique of the Study of Kinship in 1984 with his analysis of the Yapese fak/citamangen relationship. Many have ignored this critique, some have used it to declare kinship dead or to move onto other matters in the field, but not all. There is a way out of the conundrum. It requires mathematics and some hard work, and a lot of rethinking. But it is rich in possibilities. It is the algebraic view of kinship terminologies as cultural structures. For example: http://www.sscnet.ucla.edu/anthro/faculty/read/PDF_Files/Papers/Read_WhatIsKinship.PDF

  3. Sorry, i can’t get the html to work? i’ve separated out.

    BLOCKQUOTE
    “Consider the Simpsons, for instance. In classic ‘eskimo’ style kin reckoning (what Americans use) .[…] or, as we say in the business, his MZD). But in a ‘Hawaiian’ style system […] In an Iriquois-style system Selma would be Bart’s mother and her children would be his siblings but Abbie (Homer’s sister) would be considered his aunt and her children his cousins. [etc.]”
    /BLOCKQUOTE

    Although you (correctly) criticize biological or genetic interpretations of modern kinship, your examples of different kinship terminologies here is still based upon (or overlaid upon) a single common genealogical grid of kintypes, one usually taken as implying some system or theory of (biological) relatedness via the idiom of descent. Kintype products like MZD are elucidated by genealogical tracing.

    So unless you depart from interpretation, you risk sneaking in what you threw out a moment before. Merely replacing common kinship terms like sister and daughter with letters like Z and D, doesn’t change anything of substance (and one wonders why american/british kinship terms are used as its basis?).

    One can argue of course that this common genealogical grid is not precisely a theory of biological descent and relatedness. But then what is it? The only other real option on the table is that it is a cultural construct (perhaps only loosely coupled to biology, if at all). But if that is the case, what is the basis for using it as universal grid? It is true that it is very useful anyway, but analytically its on shaky ground.

    One might argue that kin type products like MZD don’t have to be anything about genealogy (much less biology). Instead they are something much more abstract. But then again, what is it? Why should it be used as the universal standard? And for that matter, why should we bother with it at all?

    If it is universal, then we are lead to accept that kinship terminologies are merely specific encodings of a biological theory of relatedness. If its culturally specific, then our analytic apparatus collapses.

    These concerns are what motivated David Schneider to write his damning A Critique of the Study of Kinship in 1984 with his analysis of the Yapese fak/citamangen relationship. Many have ignored this critique, some have used it to declare kinship dead or to move onto other matters in the field, but not all. There is a way out of the conundrum. It requires mathematics and some hard work, and a lot of rethinking. But it is rich in possibilities. It is the algebraic view of kinship terminologies as cultural structures. For example: http://www.sscnet.ucla.edu/anthro/faculty/read/PDF_Files/Papers/Read_WhatIsKinship.PDF

  4. Fiona — Thanks for this. I think the point of this article was to insist on the legitimacy of the work your doing, and the continued relevance of what we in the states call ‘four field anthropology’

    Jacob — Are you familiar with the work of Janet Carsten (who I cite in this piece)? I think it will answer most of your questions… as well as offer a trajectory out of Schneider’s work which is less… shall we say… algebraic than what you have in mind. My invocation of James Leach’s book “Creative Land” is also meant to signal an awareness of this work.

  5. A small, and quibbling point, really an aside from your overall argument:

    “The best thing that can be said of this ‘evidence’ is that its juxtaposition of Abrahamic genealogy and Muslim marriage helps remind us how ‘Western’ the Islam is—Isaac and Jacob, after all, married their cousins.”

    Rex, I can only hope this is a sarcastic aside, the tone of which did not come through in print. As we know, similarities between Abrahamic geneologies and what you call “Muslim marriage” should surprise very few, Islam being, after all, an Abrahamic religion.

    More than that, descent from Abraham himself, and from the Prophet Muhammed, as validated by adherence to (or, crucially, the representation of adherence to) codified laws of proper descent (e.g. the Sharia) is not a trivial matter in some “Muslim” (for example, Shi’a) corners.

    Of course, a shared commitment to Islam in general, or the Sharia (and paticular interpretations thereof) doesn’t commit anybody to a particular specific model of descent, nor to any form of cousin marriage, preferential or otherwise, no matter who one takes to be one’s ‘cousin’ in the first place.

    Not that I aim to make some kind of real geneological connection between “the Jews” and “the Muslims,” as we know that “the Muslims” are a, dare I say, ‘culturally’ and ‘ethnically’ diverse group who define themselves by something other than strict descent. I only mean to point out that those practicing “Muslim marriage” in Iraq know their patriarchs too, and they appear to think they are the same patriarchs Reform Jews have supplemented.

    Or, did you mean to say, rather, that it reminds us how “Western” the Iraqis are? Not to make any kind of diffusionist argument, AT ALL, but modern day Iraq and Israel are NOT FAR from each other, as US foriegn policy never ceases to remind us. Neither are Hebrew and Arabic very different, typologically.

  6. GP —
    This was not meant as a sarcastic aside but a reminder of the points that you make so well, that wee need to distinguish two related but separate things: 1) Islam is an ‘Abrahamic’ religion, and that 2) the ‘west Semitic’ kinship systems in the Hebrew Bible are similar to the ‘Arab’ ones described by anthropologists and share very deep roots (I meant to gesture towards this by citing J. David Schloen’s work).

    My decision to combine religious and ethnic practices was, in retrospect, confusing and inauspicious. I think I only really did it because I thought two words in a row starting with “M” (‘Muslim Marriage’) sounded good! The typos don’t help either 🙁

    I mention this because I really think that most people forget it. Middle Eastern politics don’t exactly encourage people to see the commonalities pointed to above, and narratives of “The Western Tradition” tend to move it steadily west from Jerusalem to Athens to Rome and then (for Americans) to Philadelphia.

    People interested in learning more about Islam as a ‘Western’ or ‘Abrahamic’ tradition should take a look at the work of F.E. Peters, who goes into this in detail.

    So overall GP, I think we agree!

  7. A propos of what you have written, here is a take-down of Pinker by Louis Menand that while it deals with a different book, nevertheless brings up many of the same problems with Pinker’s MO that you raise. It’s a great shame that someone with Pinker’s obvious communicative talents chooses to make an ass of himself on such a regular basis. It’s an even greater shame if more people don’t realize what an ass he is…

  8. Rex: I am not familiar with Janet Carsten’s work; thank you, I’ll have to add that to my reading list.
    Schneider’s critique is an important one, and one that should have been integrated into existing frameworks rather than used as an excuse to abandon formal approaches to kinship altogether, a response I find especially unfortunate since these approaches were beginning to mature precisely at the same time. But, and correct me if I am wrong, Schneider’s critique was not directed at formalism per se, but at a particular approach to the study of kinship that presumed too much. I might add that Dwight Read, just as Schneider’s critique was being published, published a, in my view, damning critique on some contemporary formal approaches (like rewrite analysis) on formal grounds (more or less). And I suspect that anthropology’s abandonment of these approaches has a lot to do with other, less scientific, factors. Regrettably, anthropologists are not regularly trained in mathematics (except some statistics, perhaps) and formal approaches to kinship demand mathematics, so many anthropologists did (and do) not know what to do with this research. And maybe there was a touch of PoMo distaste for objective approaches to anthropology which came into play too.

    Yet, I am glad that Schneider’s critique has afforded more than one trajectory into kinship studies. I look forward to reading what Carsten has to say. Schneider opened up space for the study of aspects of kinship which had until then been more neglected than they ought to have been (for example, gender and power). Perhaps it was the time for it. And whatever you might think about more formal approaches to kinship (and I hope I am excused for thinking that they are not being given a fair shake), I am sure that we can agree that kinship is a rich topic affording a multiplicity of approaches and perspectives.

    Best,

    Jacob

    If this is bold then HTML just might be working!

  9. Rex:

    Obviously not all Arabs are Muslims, nor are all Muslims Arabs, and tons of Muslims live outside the Middle East, and have done so for centuries. And all of those Muslims have been participating in an Abrahamic religion which richly references Classical philosophy, for a long time. The world is, and has long been, a complex cosmopolitan mess. Much messier than one reading Pinker would suspect!

    THX

  10. GP: Exactly!

    Jacob: Kinship is dead?!? Its a very ‘hot’ topic in anthropology at the moment! With all due respect I’d urge you not to assume that it is not a topic of discussion simply because this work is not on your radar 🙂 I’d suggest you take a look at Carsten’s “Cultures of Relatedness” volume. For highlands PNG, I’d suggest “Ku Waru” by Merlan and Rumsey for a nice digest of the literature. But none of this work is ‘formal’ in the sense that you use the word. In fact if anything I’d say one of the signs of ‘progress’ in anthropological theorizing is the subsmption of theories of kinship into more general models of meaning making coming out of linguistic anthropology. So I’d say that if anything the issue has now become how macrolevel cultural structures are used in microlevel interaction to assign people identities in one form or another — including ‘kin’ types. This is quite complicated (as Ku Waru shows) and the focus has ben how these identities are created in interaction, rather than modelling some sui generis system. Or… well that’s the best way I can describe it briefly at this time of night 🙂

  11. Rex:

    Ridiculous. I never said that kinship is dead. I am aware that there has been a revival of interest in kinship. You might say that I’m part of that revival, though sidetracked at the moment doing a MS in Computer Science. So, with all due respect reciprocated back to you…

    What I did say was that the result of Schneider’s critique was to shift the emphasis of kinship research away from the approaches advocated by folks like Anthony Wallace, John Atkins, Dwight Read, and Ward Goodenough. In fact, their work has been more or less abandoned except by a few. I think that it is unfortunate, not because there are no other worthwhile approaches to kinship (there are), but because these approaches have theoretical value, pure and simple.

    I have no problem with kinship studies being subsumed into more general models of meaning (and interaction). They should be! But we’ve learned things that we ought not to ignore, eg:

    Kinship terminologies (all of them have analyzed!) have algebraic structures (or close approximations). By this I mean two things:

    1) An ordered pair of kin terms (*not* kintypes) can be consistently associated with a third kin term, forming a tuple, by asking a question like, “The X of your Y is your ___? This has everything to do with culturally specific *words* and their meaning (or structure) in relation to other *words* from a particular finite set. No need for some presumed universal grid of kin types.

    2) The set of tuples (the ordered set of kin-terms elicited) have relationships between each other, forming a graph (a kin term map). This structure (a structure of sentences if you prefer) can be generated from a much smaller core structure (a set of a few atomic kin terms and some rules for how they are put together). A random graph on the other hand might take a distinct rule to describe every relation in the graph. But not so the structure of a kinship terminology.

    This is an empirical fact, but a fact that has not been yet satisfactorily explained. Such structures are highly non-random; they don’t just happen.

    And so while I agree with you that we need to explain how “macrolevel cultural structures are used in microlevel interaction to assign people identities in one form or another”, I must point out that this clearly is a distinct issue from the one I have outlined above. In fact, the formal approach I am advocating distinguishes between terminological structure(s), and the assigning of terms onto people. Dwight Read calls the latter ‘rules of instantiation’. Instantiation of terms is much more visible and much more contested than a terminology. People may argue about who is mother, who is grandmother, but they don’t argue about whether the mother of your mother is your grandmother, or whether the child of your aunt and uncle is your cousin. As a general rule (with exceptions).

    jacob

  12. Jacob,

    I think your comments point to why `formal’ studies of kinship were largely abandoned by cultural anthropologists: the more detailed and careful they became, the clearer it became that this area of research was basically a form of lexical semantics. Cultural anthropologists gave up this kind of study of kinship terminology as only marginally relevant to their concern: the role of kinship relations in social action.

    This coincided with, and probably partly fueled (on the cultural anthropologists’ side), the intellectual estrangement between students of `culture’ and those of `language’ (the so-called `linguistic turn’ notwithstanding!). Ideally, what would have happened at this point would have been for linguists to take up the study of kinship terminology and bring it to some relatively mature point, as happened with other related areas, such as color term research and ethnozoological/botanical terminology research. But with a number of lonely exceptions, linguistics was at the time gripped by the formal, cognitivist (read: Chomskian) revolution of which Pinker is a direct heir, and `cultural’ matters like kinship terminology simply did not attract attention in that research climate.

    But as much of the shine is wearing off formal linguistic theory, I think that kinship terminology will become an area of interest for typological linguists and those interested in areal linguistics. I was recently at a gathering of these types, and a workshop on kinship terminology was proposed and received very positively by everyone. So Jacob, cultural anthropologists may not end up being very receptive to your formally-inclined interests, but you may be able to find interested colleagues nevertheless!

  13. All of this argument brings to mind the question of what determines non-biological kinship. Would my roommate who I am not legally or genetically bound to in any way be considered a part of my family? We share living space and resources, have a certain sense of responsibility and protectiveness towards each other. A sense of sibling connection is present in our home. Yet, when asked about my (biological) family I have little to say and feel little connection to them. They are widespread, and much of the familial feelings that we shared as I was growing up have dwindled as we’ve grown further apart.

    Is the increasing separation of modern living affecting what anthropologists can consider a kinship group? Were I to die tomorrow I would probably prefer my friends and roommate handle my remains, rather than the family that I am biologically related to.

    If one can consider the sharing of living space, experiences, familial feeling, and resources as ties as binding as shared genetic material (funny, isn’t that all that marriage is?) then it brings up a multitude of questions about the laws in the US regarding same-sex marriages, and who can legally be considered “family”?

    Sorry, bit of a tangent. Just brought to mind a lot of interesting questions.

  14. Ariel, you ask good questions. Perhaps a good starting point would be Robert Frost’s famous observation that “Home is where when you have to go, they have to take you in.” That observation resonates with British Social Anthropologist Meyer Fortes’ proposition that the boundary of kinship is where the “Axiom of Amity” ceases to apply. Inside that boundary, people are presumed to operate on the basis of open reciprocity; we do things for each other without demanding an immediate quid pro quo.

    These are the people who pay for our weddings, attend our funerals, give us a place to go when serious help is needed, in general behave as if the old wedding vow, “for better or worse, in sickness or health, til death do us part” applies becaise–“we’re family.”

    Another angle is the one neglected by purely formal or cultural accounts of kinship, the tangled jural (legal) issues associated with property, inheritance and succession to office. You might ask yourself, for example, if you and your roommate own something in common, such that if one of you dies it passes, all other things being equal, to the other automatically. You might ask, if you are lovers, what happens if one of you is hospitalized. Does the other have the right to visit you in hospital or make medical decisions on your behalf?

    Finally, of course, there is the great divide in definitions of citizenship in nation-states, between those who have a “natural” right to citizenship, through birth to citizen parents or birth on the nation’s soil, and those who do not . What about your passports?

  15. In some cultures, such as the US, ancestry is a very important part of how we think about race, while in other cultures (such as Brazil) skin color may trump ancestry. Given these differences, it would be interesting to know if Brazilians are as interested in ancestry as Americans are.

    Regarding HTML issues: Please use “preview” before you post!

  16. Kerim,

    Who knows how indicative this is, but a few weeks ago I got some Brazilian spam offering to research my genealogical background, complete with an example of the kind of the kinds of results their service produces. The example documented someone’s principally Italian heritage.

  17. Rex: Thanks for writing this! That piece by Pinker has been annoying me for months. Actually, I think Pinker is the mortal enemy sociocultural anthropology for all the reasons that Menand makes so brilliantly clear and that Rex here signals with the nail/hammer idea. I was a little surprised that you did not invoke Marshall Sahlins’s critique of ‘relationship coefficients’ for decoding the meaning of cultural systems. You know his work, right? 🙂

    The genealogy craze is receiving attention by anthropologists with a more subtle heuristic than the sledge hammer of evo psych. See for example JD Faubion’s article on sumptuary kinship in Anthropological Quarterly from a while back.

  18. Another (to my mind) exciting piece of work in progress about the American kinship craze, from a different angle: Fenella Cannell’s work on kinship in Mormon communities in the US. What I’ve seen so far is concerned with Mormon geneological projects as related to salvation and longing…. Fascinating work to watch unfold, made more interesting by the fact that Cannell writes with a little distance on the US, as she’s British.
    It’s delightful work, and relevant to this Pinker business, because it highlights the entanglements of religious value, specific cosmological stakes, and geneology as practice. A central point here is love for the dead, articulated in their ‘salvation’ via posthumous baptisms (for which much geneological work is required).
    I think her 2005 Malinowski lecture, published in JRAI as “The Christianity of Anthropology,” is the first published evidence of this work (JRAI 11(2): 335-356).
    It’s funny that as kinship slid off the front burner of anthropology it became such a vital topic in the US, animated by geneological fascinations.
    In a francophone metropole of West Africa, where I work, geneology is obviously important (think griots etc.) but kinship is a big popular topic too. While this makes Evans-Pritchard very easy to teach to university first years (my students LOVED his piece on kinship terms, for example), it also brings out something that might vex Pinker & co. Discussions of the family/kinship often center, in conversation and in popular media, around the dangers of family life. That is, rather than the (tropic) place you flee to in times of economic and other trouble, the family is often figured as the source of these kinds of disruptions. So, in these cases, it is people who putatively share your genetic material who are your number one rivals and enemies. At least that’s how they are often represented in popular drama and general gossip!

  19. As someone with whose training has included several courses in language structure and social organization this has been an interesting discussion to follow. While it is true that language can be disassociated from culture and analyzed as a logical system (pace Everett) the application of formal logic to the analysis of culture and society requires a great deal more care. Schneider’s essay “Some Muddles in the Model: Or, How the System really Works” in the collection The Relevance of Models for Social Anthropology deals wonderfully with this issue.

  20. Heh. There are two questions here. The first is, can ‘cultural systems’ be studied as formal systems the way language can be (that is, not ‘is it hard’ but ‘is it possible at all’). The second question is… can language be studied as a formal system the way the first question assumes.

    For those of you who are interested in something that is less than four decades old and less muddled than Schneider’s “Muddles” paper I’d recommend William Hanks’s “Language and Communicative Practice” for the second question, and Silverstein’s “Languages/Cultures are Dead! Long Live the Linguistic-Cultural!”

    And GP is right on — I’d include Sarah Franklin’s work to GP’s list.

    Strong: I basically feel like it is every anthropologist’s job to provide an up-to-date rewrite of “Use and Abuse of Sociobiology” every three years since it seems that’s about how often its needed!

    As I hope I’ve made clear, and as Strong points out, the problem with Pinker’s not the what of it (I think studying the interaction of biology and culture is a good and important project) but the how of it (Pinker is a lousy scientist). Thumbs up on integrated science. Thumbs down on shoddy work. Four field anthro ftw!

  21. -For those of you who are interested in something that is less than four decades old and less muddled than Schneider’s “Muddles” paper I’d recommend William Hanks’s “Language and Communicative Practice” for the second question, and Silverstein’s “Languages/Cultures are Dead! Long Live the Linguistic-Cultural!”-

    Heh. Perhaps we can convince a contemporary rock star to restate Schneider’s arguments in opaque and jargon-laden prose that bemoans his lack of political engagement and gratuitously cites Julia Kristeva, all while offering no new insights. Then it would truly be anthropology for the 21st century. But the basic points of the original paper would still stand or fall on their own merit.

    I’m curious as to what you find muddled about Schneider’s paper? I assume it’s the arguments rather than the prose, because Schneider is Hemingway vis-à-vis Silverstein in that department.

  22. Rex asks:

    “… can ‘cultural systems’ be studied as formal systems the way language can be (that is, not ‘is it hard’ but ‘is it possible at all’). The second question is… can language be studied as a formal system the way the first question assumes.”

    Answering the second question first: `language’ encompasses phenomena that are wholly or largely independent of the strategic interests of social interaction, ones that are heavily dependent on those interests, and ones that lie in between. The former extreme, which Hanks call the `formal’ aspect of language and linguists call `grammar’, is exemplified by phenomena like the rule for allomorphy of the English nominal plural (you know, the /-s/~/-z/~/Iz/ alternation). The latter extreme, where interactional strategy and language ideology plays a major role, which Hanks calls the `relational’ aspect of language, and which linguists call `pragmatics’, is exemplified by phenomena such as honorification. Then you have in-between phenomena, such as grammaticalized evidentiality, which show a mixture of substantial formal and relational (grammatical and pragmatic) aspects. Those language phenomena that lie towards the formal/grammatical end can be insightfully (though not necessarily exhaustively) analyzed as `formal systems’ (depending, in part, of what one *means* by this term — this is a point of vigorous debate). Hanks is very clear about this when he argues against relationist reductionism. A large body of work in descriptive, typological, and functional linguistics shows the fruitfulness of this approach. Of course, there are important aspects of `language’ (really, communicative practice), that simply cannot be treated formally, but this does not mean that there is not a lot that can be treated formally — pick up your average descriptive grammar to see what I mean.

    Looking closely at what domains can be treated formally in language shows, I think, how completely unworkable this approach is for `cultural systems’. It is, after all, the very looseness of the connection between grammar and strategic social action that makes the formal approach feasible. Can one imagine a cultural system which is loosely connected to strategic social action?

    And, hey, MTBradley, if you haven’t read Hanks, especially the work Rex cited, you should give it a go, its clear and well-reasoned. I can understand if Silverstein’s style sticks in your throat, but you shouldn’t tar all linguistic anthropologists with that brush.

  23. “I can understand if Silverstein’s style sticks in your throat, but you shouldn’t tar all linguistic anthropologists with that brush.”

    Silverstein is obviously a brilliant scholar, I just wish he would write more clearly! The input/output ratio of work put into reading versus insight gained from his texts is much lower than it could be. And the Hanks book does look promising. I had a look at the Google Books preview and found his topics interesting. Anyone who knows who Floyd Lounsbury is has already gone a long way towards winning me over! I find it interesting that linguistics departments (or at least the Linguistics Department at my university) do not generally offer any sort of “History of” course. I know that individual courses often have a kind of historical trajectory through the semester, but this creates its own problems (such as “Advanced Phonology” being equivalent to “a semester of OT”).

    This is steering the discussion away from where it began, but I think the mention of Silverstein’s work offers a good opportunity to mention what I think is a particular problem with the subfield of Linguistic Anthropology. Silverstein is both a fine linguist (by which I mean to say that he can do linguistic analysis at a high level) and cultural (or sociocultural or whatever one’s prefered adjective for the subfield is) anthropologist. I know that a handful of programs- like those at Chicago and Texas- exist which are designed to give a good foundation in both hard linguistics and cultural anthropology, and that individual students at other institutions find their own way. But doesn’t the criticism of many linguists that linguistic anthropology is really a kind of cultural anthropology making use of speech as data rather than its own subfield ring true? And yes, I know, this gets back to the point of Silverstein’s “Sacred Bundles” paper. But the fact remains that Silverstein can engage equally well with both linguists and anthropologists. How often is that true of younger linguistic anthropologists?

    I don’t mean to place any blame here. I suspect the issue is one of funding as much as anything. One has to learn twice as much to have the skill set to do it right, but one rarely has twice the time and money. And I am not suggesting that Linguistic Anthropology is not the only place this phenomenon pops up. Ever since David Schneider offered his innovations it has apparently become OK to write about kinship despite lacking the ability to do traditional kinship analysis.

    “Can one imagine a cultural system which is loosely connected to strategic social action?”

    This is the justification for treating culture and society as distinct but related entities, isn’t it? I mean, you can argue against that distinction, but I think the reason for making it is to be able to posit both a cultural system that is truly a system as well as account for the fact that social behavior is strategic.

Comments are closed.