Cultures of academic (dis)agreement

There is more to being an anthropologist with a strong interest in psychology and natural and cultural evolution than experiencing the imposter syndrome in several disciplines. One of the perks of transdisciplinarity is attending conferences in different fields. While I have very little expertise in the social studies of sciences, I like to observe people, and conferences are great places to see science at work by watching its practitioners interact with each other.

Sian White and Robert Aunger at EHBEA 2016 in London / Photo: Anne Koerber/LSHTM

This year I have attended two conferences, one anthropological, the other on evolutionary theories of human behavior. Both were awesome in the quality of scholarship, yet so different in the way attendants received and responded to presentations. The moment which sparked my interest was an afternoon session of an anthropological panel.

I wont pinpoint one particular exact presentation, but things went something like this. I collate from different papers, since the content is less important than the style of academic conversation. The speaker presented her fieldwork and a set of theoretical ideas, ending with a discussion of how the two are mutually relevant. Many anthropological panels are organised around concepts, emerging theories or avenues of research, so every presenter paid at least lip service to the panel theme, and often engaged substantively with it.

The fascinating moment came afterwards with the questions and comments from the audience. Each speaker congratulated the presenter (a politeness I believe to be forthcoming in all academic conferences) and then latched on to a bit of ethnography and/or theory from the paper. And here is the rub. The comment usually proposed another concept, or interpretation. Something along the lines of “This is great, but I wonder if ‘class” is not a good lens to interpret these findings” or “I find (insert name of French Continental Philosopher)’s concept of XYZ quite useful to make sense of this fact” or “what do you think of another way to look by the work of (insert contemporary anthropologist, usually a big wig or rising star) which seems particularly relevant for your work). More often than not, audience sounds and head nods suggested agreement with the proposal

My fascination continued when the speaker responded by usually acknowledging the comment and answering the question with a conciliatory stance, promising to look into the suggestion and thanking the audience for their supportive feedback. Contrast this kind of academic interaction with what I experienced among evolutionary social scientists.

As with anthropologists, presentations of facts and scholarly interpretations were followed by audience reactions that often drew attention to other empirical cases, theories, and relevant scholars. But there is something different between these two arenas.

With evo-scientists, peer comments often challenged interpretation, sometimes methodology or analysis; they suggested that another theory makes better sense of the data. Quite unsurprisingly, I noted that the commentators’ own work fell exactly in the paradigm they were presenting as superior. Many were thus not innocent by-standers, but scholars socially- and intellectually-motivated to defend an alternative approach. Peer arguments were pitted against the arguments of presenters and it was up to the entire audience to decide which argument was more convincing. The way I saw it, for evo-scientists but not for anthropologists, ideas entered into conflict with one another. Better said, some actors had ideas that conflicted with the ideas of other actors.

The impression I had as a participant observer in the anthropological conference was not that of witnessing a conflict. Most scholars in all fields are nice people in conference interaction, but anthropologists are especially nice during presentations. Almost never was a speaker challenged directly in terms of findings or interpretations. At worst, the audience expressed that they did a good job, but it could be even better if they did something else : additionally, not instead of what they had done.

I call this the “agglutinative style of academic argumentation.” An argument is not intended to displace another argument. As anthropologists are fond of saying (and not without a large dose of truth), social reality is complex. Many things are happening at once, real existing societies are different from lab settings. Informers are whole persons with social, political, economic, religious sides, with various positions, motivations, and social embeddings.

This ontological and epistemological commitment makes anthropology unique among neighbouring disciplines. I think this also causes the difference in style of academic argumentation. Since there are so many stories to tell about a single event – the Rashomon-effect we may call it – , it is very probable that one account does not exhaust all possible causal mechanisms. Thus, one concept or theory can very well live alongside others. Surely, attention to affect or the morality of kinship can help us understand better a marriage practice, but so would social hierarchies or an analysis of language or symbols. It need not be one or another. And there are so many more valuable possibilities, that it comes to be that no idea is really bad or useless.

Like anthropologists are wont to do, I will enhance my ethnographic vignettes with a theoretical interpretation. Alas, it’s based on evolution. In The Enigma of Reason, Hugo Mercier and Dan Sperber maintained that human reasoning evolved to produce and evaluate arguments. Its function is primarily social rather than individual, to persuade others and to avoid being manipulated or misled. And we usually reach better reasoning outcomes in interaction and communication with others than we do on our own.

The evo-science conference seems to fall closer to what may be called the “proper” domain of reasoning. Participants argued against each other, each showed the blind spots of other arguments and tried to show why their idea was better. The anthropological conference was slightly different. True, people tried to persuade others of a point of view, of a personal proposal, but they seemed less interested in rejecting alternative ideas. Neither speaker nor peer comments were challenged. At most, they were painted as slightly less interesting than another brilliant interpretation. Ideas don’t seem to die under the crush of counter-arguments, at most their relevance is comparatively downgraded.

The anthropological tradition of praising and seeking complexity, inter-relatedness of phenomena and the messiness of social life explains part of the conference interaction. Another cause, I think, comes from the structure of ideas in anthropology. Simply put, there is no relevant opposition between theories in the discipline. This was not always the case, as the descent versus alliance debate in kinship or the formalist-substantivist divide in economic anthropology illustrate. But this was back in the 1940s-to-1970s. Since then, no real battlefield appeared in anthropology. If anything, the biggest challenge was is one could anthropology and ethnography in the first place.

Language is instructive in this case. In anthropology, debates have been replaced by “turns”. The “ontological turn” comes to mind as a recent development. But notice the metaphor. Turns are not conflicts. They evoke more the turn of attention to something else, a diversion to a new road, rather than a break with the past, refusal or denial of past or current ways.

Another effect comes from another peculiarity of anthropology. By and large, anthropologists are very kind towards each other’s ideas. They certainly do not destroy them publicly in conferences and seldom express the desire to do so in writing. But one thing unites them: a deep feeling that all other academic disciplines are missing the point. Surely, economists, psychologists, political scientists or biologists do their job well. But they cannot reach the depth of anthropologists with their immersion in real societies, the emic perspective of taking the native point of view, of experiencing first-hand the social life that other disciplines are trying to gauge in labs, surveys, or quantitative data.

For some but not all anthropologists, the attitude is more than just bemoaning the naivete of other disciplines. It is in fact squarely opposed to attempts to explain societies and people by recourse to (perhaps the most widely shared bête noire) nature, biology, or natural evolution. The figure of Marshall Sahlins, one of the most insightful anthropologists for my academic formation, looms large above such an entrenched position of putting “culture” or “nurture” first.

Thus, anthropologists have fewer intellectual enemies within than outside. It is small wonder then that conference presentations are moments of benevolent exchanges of compatible ideas rather than mutually exclusive arguments. Different interpretations agglutinate towards that elusive but powerful dream of describing and interpreting a society and its people holistically. The cost is, I fear, a risk of keeping many bad ideas alive and often kicking. The benefit is one that has been provided by anthropology since its inception. Whatever theory is proposed, another theory could add something to it. How the cost and benefit balance is a subject for another blog post.


  • comment-avatar
    Thom Scott-Phillips 19 November 2018 (17:29)

    Different degrees of fit with the proper domain of human reasoning
    This is a super post, very interesting indeed. I don’t have the ethnographic expertise that Radu does, but my own experiences match what he describes here.

    One (agglutinative!) comment. Radu writes that “The evo-science conference seems to fall closer to what may be called the ‘proper’ domain of reasoning”. Indeed it does, but there are several other fora of human reasoning that fall even more tightly into the proper, argumentative domain of human reasoning. The clearest example is law, where it is the legal and professional duty of both prosecution and defence to mount the best case possible for their side, regardless of their own views and preferences.

    This thought has caused me to daydream, sometimes, what science would be like if it was structured more like the legal system. You would have several different categories of scientist. One would be charged with developing ideas, and others with arguing, as best they possibly can, either in favour or against these ideas. A jury would then decide what should go in the textbooks (which can be revised in the future, of course, subject to some mechanism to revisit consensus opinion).

    It will never happen of course, but it’s curious to think about such a model, as part of a continuum of different degrees of fit with the proper domain of human reasoning. Radu’s observations about anthropology (at the other end) and evolutionary social science (in the middle) fill out the continuum.

  • comment-avatar
    Hugo Mercier 20 November 2018 (06:38)

    Thanks Radu, very interesting post!

    Experimental evidence (work by Nemeth and others) suggests that brainstorming doesn’t work — i.e. people are less good at generating good new ideas in groups than on their own. This seems to happen in part because in typical brainstorming sessions, people aren’t allowed to criticize each other. As a result, they interfere with one another’s thought process, but without having the positive effect of pruning bad ideas. Once these rules are relaxed, brainstorming seems to become a bit more efficient (and we also know that argumentation works well in allowing good ideas to spread).

    Make of that what you will 🙂

  • comment-avatar
    Radu Umbres 20 November 2018 (22:47)

    Reply to Thom and Hugo
    Thank you both for your very keen comments.

    Thom, the legal model is perhaps emulated by the peer-review system, although there are clear differences. And I have a feeling that anthropology might not even be at the end of the continuum, there being even more generous, non-conflictual academic arenas. Someone mentioned the Religious Studies on Facebook, but I cannot vouch for this.

    Hugo, the brainstorming comparison is spot on, and I think its negative effects are felt in anthropology. But here I speculate on another effect of non-conflictual anthropological conversations: getting a feel of the field, receiving feedback and suggestions about what is “in” and what is “out”, what people expect to hear, the connections which audiences feel are afforded by the paper. Of course this does not improve bad arguments (quite the opposite, I fear) but it helps authors fine-tune their work according to social expectations.

  • comment-avatar
    Pascal Boyer 22 November 2018 (00:42)

    I know the answer !
    [Anyone who is deluded or pretentious enough to open their contribution to a debate with a resounding “I know the answer” should re-read and ponder Daniel Nettle’s ironic reflections on “Big Idea Papers” [1]. But bear with me]
    The style of interaction observed by Radu Umbres in his cultural anthropology conference is, I must say, sadly typical of the discipline. And I say “sadly” because it is the symptom of a malady. There is no reason why it should be so.
    I tried to offer a diagnosis a while ago in a chapter contributed to Ted Slingerland’s and Mark Collard’s collection Creating Consilience in the social sciences [2].
    The idea was that there are two traditional modes of scholarship in many disciplines, and a third, more recent, style that is very largely a pathology.
    The two traditional styles are erudition and science. Erudition is the accumulation of facts, what people do, e.g., when they list species of ants, describe the evolution of languages, chart the prices of wheat in Roman history, and catalogue Korean portraits. Science is what you do when you test models and hypotheses, e.g., explain ant cooperation by inclusive fitness, deduce laws of phonetic assimilation, model the micro-economics of agrarian societies, or propose that cultural evolution will favor some kinds of portraits (e.g., with subjects gazing at the spectator) over others.
    In many healthy disciplines, the two styles feed into each other. The interaction of natural history and scientific biology is a case in point.
    That was the case also in cultural anthropology, for many decades (and it still is the case in some places). Theoretical disputes about descent vs. filiation were nourished by comparative erudition in Asian vs. African ethnography. Theories of witchcraft were proposed on the basis of comparisons between many ethnographic studies.
    So, what about the third style? I described it as “relevant connections”. It is a kind of scholarship in which one does not contribute new facts (as in erudition) or new models and explanations (as in science), but simply connections with previously known elements: Shakespeare as colonialism, Balinese government as choreography, New Guinean initiation as a metaphor, witchcraft as text, kingship as theater, and so forth. [There are many more, and funnier examples in my original chapter, for people who like that sort of thing].
    The “relevant connections” mode of scholarship does not add to our knowledge or understanding of the world. Incidentally, this may explain why cultural anthropology, which used to be a salient contributor to public debates (think of Mead, Leach, Ruth Benedict, others) is completely absent from such debates these days.
    This may also explain the kind of “discussion” observed by Radu. In the “relevant connections” style, there is simply no way you could agree or disagree. All you can do is acknowledge people’s inventive linkages, and suggest your own connections. You see Verdi’s operas as the denial of homoeroticism, I see them as the first trumpet sound of triumphant neoliberalism. Let a hundred flowers bloom!
    So, it seems to me that the “tolerant” style of discussion is the symptom of an etiolated or stagnant research tradition. Far from mobilizing our argumentative capacities, that kind of academic discipline shifts them into neutral.
    At the other conference Radu attended (I was there too), many empirical claims (erudition) were challenged, as people argued that the evidence was clearly insufficient. Several theoretical models (science) were politely addressed as possibly incoherent or circular, or too unclear to explain anything. Isn’t that the way things ought to be?



  • comment-avatar
    Radu Umbres 23 November 2018 (12:01)

    The relevance of “relevant connections”
    Pascal, this is absolutely fabulous, and indeed your chapter has a very “relevant connection” to my ethnographic vignette 🙂

    The interesting question is why certain connections are seen as “relevant” and other not. To add something from my experience, I also tried to provide comments to certain anthropological presentations, usually mentioning works from psychology or philosophy, yet these were hardly seen as relevant. In fact, the awkward silence or sidestepping my comments made me feel quite irreverent to the whole enterprise. The reason, I assume, was by reference to natural evolution and naturalistic approaches to society, “positivism” or “psychologism”. This signalled my problematic affiliation to a group of scholars perceived as external if not outright inimical to socio-cultural anthropology. Something similar happened when I mentioned classical anthropological scholarship (say before 1980s). On the contrary, other connections linking papers to recent publications by well-placed and highly quoted scholars seemed quite relevant to the author and audience.

    To conclude, it is not quite the case that “everything goes” as you argued in the case of salient connections in anthropological theory. There are selection effects based on political or intellectual ideology/dogma, disciplinary boundary-keeping, the politics of hiring practices, signalling affiliation to one or another “school of thought”, and probably many more besides.

  • comment-avatar
    Burt 11 December 2018 (08:37)

    Disciplinary AssumptionsThis discussion makes what to me are some very interesting points. Coming from a background in physics and mathematics I’ve been more use to the “science” view of conference interaction, but have also had experience of the other two forms. I agree with Radu that while the “relevant connections” form can be stagnating (recalling Kuhn’s pre-paradigmatic phase of a developing science), they can also be helpful, for example in showing how an unsuspected analogy might allow cross-fertilization. Of equal interest to me, though, is to speculate on the underlying disciplinary assumptions driving the observed forms of conference (and other) interactions. I’ve spent some time observing the writing style that shows up between “hard” science, “soft” science, and humanities texts, and how different nuances of style can lead to misunderstandings when writing in one discipline about results in another. As a specific example, some time ago I was reading a text in history of science, written by a first rate historian, in which one chapter dealt with the seventeenth century debate over whether or not heavy objects fell faster than light objects. It was fascinating to learn that many people were conducting experiments on this, and that the results were mixed so that it took a number of years to reach the final conclusion that all objects fell at the same rate. At the end of the chapter, however, I felt just a hint of the idea that if only the proponents of the “heavy objects fall faster” position had been more persuasive in their argument, then this would have become scientific consensus. I’m quite sure that if I were to ask the author if he intended to give that impression, he would have said no, of course not. But it was there, or at least I felt it as a ghostly presence. And in mentioning this to a few physicists I got a reaction that “of course, historians are all perspectival relativists.” What seems to me to be the case, however, is that an unstated disciplinary assumption was involved: historians (and humanists in general) do not assume that there is a single canonical interpretation to be discovered, physicists do. And underlying meta-disciplinary assumptions such as this can exert unnoticed influence on writing style or, for the current discussion, on conference interaction styles. Which is a long way of leading into the question of what sort of underlying assumptions are there in cultural anthropology, as contrasted to evolutionary behavioral theories, and how do these influence what interactions are assumed as normal and proper in the respective fields?

  • comment-avatar
    Thom Scott-Phillips 21 March 2019 (15:30)

    F’ing nuance
    Following Pascal’s comments above, I read his article on the studious irrelevancy of much anthropology. His observations about modes of scholarship – of which Radu’s blogpost does indeed provide examples – came back to my mind this week, when I re-read Kieran Healy’s provocatively titled essay on modes of scholarship in sociology: “F*** nuance” [1]. The links are clear and unmissable. Here’s one key passage on “connoisseurship”, which echoes some of the themes of Pascal’s essay:

    “With social theory in this mode, a logic of sophisticated appreciation prevails, combined with a hierarchy of taste based on one’s alleged capacity for subtlety. It resembles the discourses surrounding fine wine, cuisine, or art, because connoisseurship thrives best in settings where judgment is needed but measurement is hard. This favors the development and expansion of specialist vocabularies that are highly elaborated but only loosely connected to measurable features of what is being talked about. One cannot simply say any old thing (mostly), but the rules governing use of the vocabulary are hard to learn and tend not to be written down anywhere. One may be confident that there is some reliable connection between the vocabulary and the object, but competence still requires certification by another expert in the field. A Master Sommelier probably knows a lot more about wine than the average person, but it is reasonable to be skeptical whether very detailed wine-talk has a fully codifiable connection to the taste of wines. Theory in this style is carried on in a similar cloud of terms, allowing for rich verbal expression that also carries a clear signal of the speaker’s sophistication.”