Sunday, 25 October 2015

David Graeber against Eduardo Viveiros de Castro and the 'ontological turn'

Further to my last post, the new issue of Hau: Journal of Ethnographic Theory has an article by David Graeber, Radical alterity is just another way of saying "reality": A reply to Eduardo Viveiros de Castro. As the title might suggest, it is a reply to de Castro's Who is afraid of the ontological wolf?, which, in turn, criticises another of Graeber's essays.

It is a long, often interesting but also often rather pedantic defence of Graeber's philosophical position against what he sees as the idealism of de Castro et al.

A fairly large portion of the piece involves 'unpacking' what I thought to be really rather stark, staring obvious: that when advocates of the so-called 'ontological turn' in anthropology use the word 'ontology' they do not mean it in the traditional philosophical sense of a formal, structured, philosophical account of being but rather use it to refer to the tacit, informal lived understandings and presuppositions that have ontological or cosmological consequences. I really don't know why it needed so much explication as, from what I've read, this is not in any way a controversial interpretation.

The reason for the longwinded explication is presumably that Graeber sincerely dislikes this redefinition (although he is relatively courteous, for the most part, with how he goes about stating this). Particularly at issue is the relativism that refuses the right of an anthropologist to posit a singular reality 'behind' the multiplicity of understandings of the world (the famous multiculturalism/mononaturalism).

For Graeber, the self-declared ontological pluralists are just radical idealists. He prefers the 'depth ontology' of critical realism that posits a singular reality that is unknowable. This unknowability is thought to leave enough room for cultural multiplicity and all the ontology talk is, at best, unnecessary.

As I see it, the accusation of idealism is the product of Graeber's own dualism. For the ontological turners, Graeber claims:
"[authoritative] statements must be treated as a window onto “concepts,” and concepts treated—through a form of “radical constructivism”—as if they were themselves realities of the same ontological standing as “things,” or indeed, constitutive of the world itself." (20)
For concepts to be of the same ontological standing as things simply means that one is not bifurcating reality – i.e. one is refusing dualism. To slip from that to "or indeed, constitutive of the world itself" is a conflation. He continues:
"The “ontological turn,” then, involves not only abandoning the project of ontology1 [i.e. traditional, philosophical ontology], but adopting a tacit ontology which seems indistinguishable from classical philosophical Idealism. Ideas generate realities. One could go even further. What they seem to be proposing is abandoning the entire project of philosophy (or at least, philosophy in anything like the forms it has historically taken in Europe, India, China, or the Islamic World.) Science, in contrast, would be preserved, but as the special property of “Westerners” or “Euro-Americans”—which if taken seriously, would amount to one of the greatest acts of intellectual theft in human history, since after all, much of what underlies what we now call “Western science” was actually developed in places like Persia, Bengal, and China, and in (dare I say?) the real world. Most scientific research is no longer being conducted by Euro-Americans at all." (21)
There are various claims here. The claim to be able to state what the entirety of the philosophical traditions of not only Europe but also India, China and the Islamic world have been all about is presumptuous bordering on pompous, to say the least. Even Western philosophy has been about rather more than magisterially postulating the furniture of the universe, although that has certainly been a part of it. The bit about science is a cheap shot that relates to nothing I can imagine de Castro et al. accepting for a moment.

The main claim, however, is that all the ontological turners are really doing is reproducing a tired, old and quite extreme form of idealism where concepts produce and/or exhaust reality. Now, I've by no means read the entirety of the relevant literature but this strikes me as a knee jerk reaction based on a misunderstanding that ultimately says more about Graeber's presuppositions than it does about anyone else's.

I do not understand the position of ontological pluralism to be claiming that concepts constitute reality without remainder. I understand the claim to be that concepts, ideas, whatever you want to call them, are on the same level of reality as everything else. That is, that it is illegitimate to say that one has concepts on one hand and things on another (bifurcation). Furthermore, that it is impossible to refer to a thing except in liaison with particular concepts. That doesn't mean that concepts are all that there is, only that to speak is to conceptualise. You cannot think without abstractions, as Whitehead put it.

So, yes, Amerindians and Europeans have different perspectives, different cosmologies, etc. But the trees and ants and spiders and so on presumably have their perspectives, too. Of course, the anthropologist qua anthropologist is not so interested in the perspectives of those things in themselves but in what their informants say about those things (although this is far from straightforward, admittedly). Hence why it might look a bit like idealism. But it simply does not follow that these things are equivalent. Methods and philosophies overlap and might confuse each other a little bit but a sympathetic reading can, I think, tell them apart.

The question hanging around the entire article is 'okay, but you don't really believe it when they say that, do you?' The implication is that it may be good manners (or even, depending on the circumstances, good politics) to rhetorically accept the beliefs of others that strike the observer as being wholly untrue but that there is a deeper and more fundamental sense in which one can, and should, declare them obviously unbelievable. Belief is absolutised, underlying all other modes of action. The epistemic underlies and underpins the moral, the political and so on. Belief is sacred. To fail to give belief its sovereignty is simply incomprehensible.

The thing that Graeber really can't get away from, I think, is the idea that either you accept that there is a real world (I won't add 'out there' because I think he is sophisticated enough to avoid that artefact of common sense stupidity) or you deny the existence of a world apart from concepts altogether. In other words, he seems to presume that to say that there are 'many worlds' rather than just a singular world is to reduce reality to concepts. This just doesn't follow and doesn't grasp what I think ontological pluralists are claiming.

It's a tough thing to explain so I'll approach it from a bit of an oblique angle (I'm thinking this through more for myself than for anyone else, to be honest!).

When I used to live in London, I realised quite quickly that nobody actually lives 'in London' – it's an abstraction. London is just far, far too big to live 'in' it in any meaningful sense as a totality. You live in this or that street, that area. Perhaps you have friends here and there. You take this tube line, etc. It is quite obvious that you live inside quite a narrow and confined network and that the vast, overwhelming majority of the city falls between the lines that you travel. Most of it you will never see nor even think about.

The fact that the epistemic techniques of geography allow us to place these networks within a larger container that can be defined as the totality 'London' (and the fact that this may be a very useful and sensible thing to do navigationally, administratively and so on) should not determine the social ontology by which one understands this city as it is being lived in. In other words, the common sense fact that all these streets, offices, bars, pipes, drains and so on that make up an individual's city-as-lived reside within a definable, connected region does not mean that London is one, cohesive, singular entity – like a body of which one was a cell.

The claim I would make is that there is no one way of joining these various things up that can be said to be the 'reality' that would have parts from which particular perspectives could be assembled (not the planner's bird's eye view, nor the individual's lived experience, etc.). That singularity or totality is a superfluous hypothesis. The crucial thing to emphasise is that the bollards, the rails, the bins, the foxes, the concepts and so on – all of these things are part of this not-one, part of this multiplicity that cannot be assembled in any one way that can be declared the most real. Because they are not bifurcated into different realms, it is perfectly sensible and consistent to hold that there is no one way of joining them up, just connected and overlapping ways that never add up to a whole. Or, rather, when a particular manner of joining things up forms a totality, this is also just one kind of entangled complex among others. It may be more or less important, more or less useful relative to a particular task or problem but it can be given no transcendent, automatic priority.

What is true of London is also true of the world, indeed the planet. Yes, in a certain sense we all live on 'one' planet. It would be silly do deny this but in either affirming or denying it we have tacitly accepted certain premises with regard to the form of connection that matters most. Nothing obvious about how things are proves that we should think about this planet as a singular entity rather than as an intensively connected but nevertheless fragmented and discrete bundle of bundles – that is if we think about anything beyond our immediate locale at all (that might also be okay).

The banal fact that everything is connected to everything else in some fashion does not mean that there is one world rather than many. It is in no way obscure or contradictory to claim that there are many worlds, all of which overlap and are related in some way but some of which are quite distinct and should be understood and dealt with as such. Nothing obvious about the way things are licenses a 'real world' which would reside behind all merely cultural understandings of said world. It's a dogmatic presumption, which doesn't make it wrong but doesn't make it right either.

At several points, Graeber pulls out the old incommensurability argument, which suggests that the de Castro et al. think that the collectives for which they are describing ontologies exist as some kind of definite, bounded, billiard ball-type entity. That, again, is a cheap shot against a position that I doubt anyone involved in this debate is really claiming to adhere to. No collective, certainly not one that has had a Western anthropologist living in it for years, is absolutely pure and separated from the rest of the world. Of course these things overlap, of course they are complicated. The limitations of language sometimes brings us to suggest that such things are definite entities in the manner of classical physics (or ball-based sports) but that does not make it necessarily so. Every attempt to name a collective is in a sense an attempt to verbally lasso a loose and self-assembling bundle of bundles that may well shrug off that signification, that may even violently reject it. The degree to which one is successful in articulating a bounding that is acceptable to the larger proportion of those persons is the basic barometer by which the naming can be judged a success. Of course, this is complicated by all sorts of power dynamics – who gets to object? whose objections are heard?, etc. – but such is life. These are the problems that are to be negotiated by the trials of experience.

I add by way of conclusion shutting up and going to sleep that I'm not unquestioningly or naively on board with the ontological pluralist agenda as it's being articulated in anthropology (for a start, the things I'm describing above may well not be fair representations of or addenda to their arguments). First, I'm not an anthropologist, just someone with an interest in, for want of a better term, the comparative sociology of knowledge. Graeber's argument that it should be wholly legitimate for anthropologists to adopt the role of an instigator of dialogue between collectives rather than just someone who continually emphasises and safeguards radical alterity seems perfectly acceptable to me. However, it really depends on the collective in question. Making general ethical rules with regard to this seems not only impossible but also deeply unwise.

The really important question would be which ethical and political roles are appropriate depending on what circumstances. That isn't the conversation that is happening at the moment but there does seem to be room for it. I'm sure that this conversation will roll on, in any case.


  1. I didn't say that "ontological pluralists" would have to be arguing that words create things, I said that a specific group of people ARE arguing that words create things, because they literally say they do. I was speaking of a specific tradition in anthropology called the "ontological turn" and I quote the founding text of that tradition, the introduction to a book called "Thinking Through Things." What's more I not only summarised their argument but provided a footnote that said "just in case the reader thinks I'm exaggerating..." and then provided an actual quote from that text literally saying that words produce things.

    To just completely ignore that, and pretend that I'm instead making an unreasonable accusation against some imaginary group of "ontological pluralists" that I never mention, is incredibly sloppy scholarship. I am referring to a specific tradition, with specific authors, and I cite their exact words.

    If you want to say "sure, but there "could" be a form of ontological pluralism that doesn't fall into this trap; or even that such a school exists, then fine, make your case.

    This essay makes that same kind of move over and over,

  2. I would never refer to my blog as 'scholarship' (proto-scholarship at best) but fair enough, I should have been more careful. I think ontological pluralism is a fair term to ascribe to these authors. But you're right, you didn't use that term and I should have recognised that fact.

    I must also admit that I am more familiar with Viveiros de Castro's work than some of the others you are citing and so I am not perhaps fairly understanding the group to which your arguments have been posed.

    However, to the quote in question, since this is apparently exhibit A:

    "Though Foucault would say that discourse creates its objects, he still works from the presumption that there is some real-world fodder out there. For example, while a body may not be male or female until a discourse of gender invokes this as an operative distinction, there is still a body to which the discourse refers. By contrast, what is advanced here is, if you like, an entirely different kind of constructivism—a radical constructivism not dissimilar to that envisaged by Deleuze. . . . Discourse can have effects not because it “over-determines reality,” but because no ontological distinction between “discourse” and “reality” pertains in the first place. In other words, concepts can bring about things because concepts and things just are one and the same” (Henare, Holbraad, Wastel 2006: 13).

    You interpret this to mean that things are just concepts – more precisely "words create things" (without resistance or remainder). In other words, the elimination of the opposition word/thing means the absorption of things by words.

    While this excerpt is clunky and badly constructed (and I would not like to defend it), I do not understand them to be saying what you are interpreting them as saying. However, I will need to go away and re-read this in context before I can say anything more intelligent about it than that.

  3. Okay, so I've been back to the book and re-read the above from Henare, Holbraad and Wastell. I am led to the conclusion that I was probably a bit unfair to David's argument because there are some serious weaknesses in this chapter (and perhaps, then, in the so-called 'turn' also – but I am not fully up to date, as I admitted previously) and because his critique was more precisely focused than I originally acknowledged. However, in general I stand by what I wrote. I am by no means prepared to defend HH&W because I found their piece to be rather conflicted in itself (and they can presumably defend themselves). But I am still happy to abide by my interpretation of the 'turn.'

    I still think that it is incorrect to interpret them as saying that in breaking down the concept/thing distinction that things are therefore absorbed into concepts – classic idealism (in places they err close to this but I still don't think that this is correct; there is more going on here). I think that it would be more accurate to say that, for them, concepts *are* things – if anything it's the opposite move. Although I'll happily admit that they don't articulate this very consistently (perhaps indicating differences in interpretation amongst themselves). In the passage quoted above they refer to Deleuze & Guattari's 'What is Philosophy?' with regard to the constructivism of concepts. There are many readings of D&G but the importance of the plane of immanence in that book doesn't seem to chime with their argument – at least not as they've made it here. I understand why they might want to refer to it in terms of declaring philosophical allegiances but it raises more questions than it answers with regard to their argument.

    The main source of conflict/confusion in their chapter, for me, is that they start out (and indeed finish off) by referring to what they are proposing with regard to ontology as an 'analytical' or 'methodological' proposal, as distinct from a 'theoretical' one. So, anthropologists are to strictly bracket their own understandings of the world in order to do justice to the things that they are encountering rather than trying to explain them in terms of a pre-existing, superimposed scheme. But at times they seem to suggest that this bracketing would basically wipe the researcher's own understandings of the world clean. So, method overrides theory not relatively but absolutely.

  4. Continued…

    Where I would lean more towards what David Graeber was arguing is in that this 'bracket' (much like Michel Callon's famous 'frame') must necessarily overflow. There is no way of leaving out one's own ontological assumptions altogether. (I suspect that hardly anyone involved in this debate would disagree with this but it bears stating.) They can be minimised but they cannot be eliminated. The very act of bracketing is an act with ontological presuppositions. It could even be understood as being ontologically creative in its own right. There are no more blank slates with regard to ontology than there are with regard to gender, race, class, etc. The opposition between method and theory they set up is too strong and that, I think, is the principal source of conflictedness and, at times, incoherence.

    So, I would argue that everything we do has ontological presuppositions attached to it and, further, that *that is okay*! That isn't necessarily a problem. However, the trick is to refuse the bifurcationism that posits a singular reality or nature 'behind' all of us that is only broken up with different interpretations (mononaturalism/multiculturalism, again). One does not need to go all the way to idealism or to the refusal of all positive ontology or philosophy in order to do that. The process, as with the ethnographic method generally, is that of translation. It is about setting oneself (I must again affirm that I am not an anthropologist so this is all strictly from the armchair) up as an agent imbued with the necessary practices and dispositions so as to achieve the task at hand (it's essentially pragmatic). In this case that means not assuming a unified reality upon which are imposed a multiplicity of representations. I fail to see how theory, philosophy, method and ontology in all its senses can be pulled apart so easily.

    That said, I think HH&W actually anticipate a lot of David's objections and while they don't answer them quite satisfactorily, I think that the collective project in general has enough about it philosophically to sustain the claims that I am making. Worth noting that although this is apparently the 'foundational text' of the 'turn,' they are in turn drawing on various works going back some years, including Viveiros de Castro, Latour, etc. It was with that larger intertext in mind that I was writing and referring to 'ontological pluralism' – which is a term some of these authors also use.

    I don't think that this will satisfy anyone who thought that my original post was crap but it satisfies my sense of blog propriety. Were I to write something on this more formally of course I would try to do a better job. As it is, these are just my scribblings. (The fact that my scribblings are public is an artefact of an age the consequences of which I try not to think about as it makes me nauseous.)

  5. I've also been trying to navigate the ontological twists and turns! It seems to me like there is more than one turn. Besides Viveiros de Castro and HH&W, I would suggest another scholar as foundational for new takes on ontology, Annemarie Mol, especially with her chapter in "Actor-Network Theory and After" (1999) titled "Ontological politics. A word and some questions".

    Her book "The Body Multiple. Ontology in Medical Practice" (2002) focuses on practices and she even suggests calling her approach a praxiography rather than ethnography. The attention to practices and not just concepts seems to me to distinguish this approach from idealism. Mol explicitly wants to challenge the primacy of the realm of ideas for knowing the world. In contrast, doesn't the title "Thinking Through Things" sound like it is still caught in a primacy of thinking?

    Another thing to note is that Mol (pp. 83-84) actually explicitly rejects the notion of pluralism: "But the ontology at comes with equating what is with what is done is not of a pluralist kind. The manyfoldedness of objects enacted does not imply their fragmentation. Although atherosclerosis in the hospital comes in different versions, these somehow hang together…This, then, is what I would like the term multiple to convey: that there is manyfoldedness, but not pluralism. In the hospital the body (singular) is multiple (many)."


  6. Yes, there have been 'ontological turns' all over the place! I am a little sceptical of the designation. I have nothing against the notion of a 'turn' as a signpost on a much more complicated landscape of debates but they're not always very useful signposts and frequently get taken as destinations in their own right.

    Part of the disagreement here is that I was thinking of these various debates in their interconnection (because they're not separate; Viveiros de Castro, Mol, Latour, etc. are all in conversation together). Graeber was writing about this specifically in anthropology (to what extent it can be cordoned off in this way is debatable and I've clearly been operating on the admittedly underspecified assumption that it can't be).

    Mol's work has inspired a lot of people beyond her immediate disciplinary niche (insofar as she is located in one). She is probably closer to Viveiros de Castro and the anthropologists in terms of her insistence upon the locality and specificity of practices. Latour has in recent years before much more of a speculative philosopher. He is trying to work out those tensions between specificity and generality but they are very much still there, I think.

    The question of pluralism, realism, relativism, etc. is a torturously contentious one, for sure. Mol gives an important answer to it.

  7. In case anyone in this discussion can read French, that dying language, may I suggest the paper below by Patrice Maniglier, which explains my own "take" on the subject much better than I could - or, to be frank, would care to now?à_Eduardo_Viveiros_de_Castro_

  8. I'd give your analysis a C+.
    You really need to learn how to write without making straw man arguments - at the moment you seem to find this almost impossible. I did not state that "things are absorbed into concepts" I said what I said, which is that they say that concepts generate things, because that is, literally, what they said. Whether there is a "remainder" or whatever other additional distraction you are throwing in to make it sound like I am making an unreasonably strong case is irrelevant to this point. I don't attribute a position to them on this matter one way or another, because it has no bearing on my conclusion, which is that this is a form of philosophical idealism.

    If you are going to critique people in public, it really think you should learn how to critique them for their actual arguments, rather than continually making ones up. That's not critique. That's just making shit up.


  9. "I did not state that "things are absorbed into concepts" I said what I said, which is that they say that concepts generate things [...]."

    If 'concepts generate things' is not the same as 'things are absorbed into concepts' then presumably this is a 'philosophical idealism' that is not absolute; i.e. not 'without remainder.' In other words, it is an idealism that is presumed to rest on some conception of noumena or some such. Since, otherwise, there would be literally nothing but ideas because things would be reducible to ideas. In which case, there would be no meaningful difference between the above statements. They would be two different ways of stating the same thing: that there is nothing but ideas. And if that is not what was being claimed then it certainly seemed like it:

    "[...] this is not just Idealism—it is about as extreme a form of Idealism as it is possible to have." (p.23)

    I think it is quite possible that the authors of the ontological turn piece will not find your interpretation to be a 1:1 correspondence either.

    So, my interpretation, made up shit or not, remains one that I am happy to stand by. I'm sure that the handful of web wanderers who stumble across this will make up their own minds.

  10. I see, so your position is now that this is not the most extreme possible case of philosophical idealism because even though they explicitly assert that ideas do generate objects directly, rather than constructing anything that is already there, they MIGHT believe that there actually is something already there even though they say they don't, or they MIGHT think there's some other material trace somewhere involved at some point even though they never say there is, and therefore it's legitimate to critique someone for simply quoting what they say and taking it at face value because they MIGHT think something else they're not saying and I can't prove they don't.

    By this logic you can't criticise anyone for anything of course; if you applied the same logic to my own words, you'd be guilty of exactly the same thing you're accusing me of, because how do you know I don't also have a different theory I'm not saying that contradicts what I just said?

    I'd say that depends from C+ to F.

    I really do hope you're not thinking of an academic career.

  11. Nope, that isn't what I was saying at all.

    Have a nice day, though.

  12. As one of those web wanderers, I find Mr. Graeber's retorts to be ill-judged and a little spiteful. Philip, keep up the good, careful and even-handed work.