Comparative ontologies and cosmologies

[Huon Wardle]

HORATIO
O day and night, but this is wondrous strange!

HAMLET
And therefore as a stranger give it welcome.
There are more things in heaven and earth, Horatio,
Than are dreamt of in your philosophy.


Much ink has been spilt in anthropology recently about 'ontology' and the ontological turn. There is a shift in attention from seemingly 'old' framing concepts like society and culture to issues concerning the language/philosophy of 'being'. A great deal of of this discussion can be traced to three figureheads - Viveiros de Castro, Marilyn Strathern and Bruno Latour. What is the significance of this turn? Since there has always been a comparative cosmology entailed in anthropological debates, is there anything distinctive or surprising in this emphasis? What, if anything, has changed?

It has been suggested that I refer readers of this thread to the GDAT debate 'Ontology is just another word for culture'. This may help make sense of the issues at hand in a number of the discussion points.

[John McCreery]

Kathleen,

Your remarks "coming from the Guarani studies angle" are fascinating and your conclusion "that this is both one of the beauties of anthropology -- using the Other as a means of self-critique -- as well of one the tendencies that might make is rightly wary of the verité of particularly exemplary ethnographies from any era" is one I can happily endorse.

Would it be fair, if reading this these remarks as an answer to Huon's questions, to conclude that, "No, there is nothing distinctive or surprising here"? Once again, it seems, the natives are being trotted out to embody the European intellectual's critique of something that the intellectual disapproves of in his own society/culture. The technique is familiar. You mention Montaigne; I recallMontesquieu's Persian Letters, Voltaire's use of China, Siam and Japan to preach enlightenment values, and Rousseau's noble savages. What is still missing, for me, is a straightforward answer to the curious student who wants to know what value wrestling with the ponderousness of "ontology" will have if and when she goes into the field to experience and to study lives alien to her own.

Consider, by way of contrast, the currently popular use of "ontology" in the information sciences. Here the aim is clear, to specify the things that comprise a system with sufficient clarity to program their relationships. The reference is to the boxes and lines in what are called Entity Relationship Diagrams (ERD) It isn't hard to understand why taking the trouble to specify what a database is intended to accomplish in this form is a recommended step before writing programming code.

One suspects that the anthropologists that we are discussing here would be horrified to see their use of "ontology" read in this highly techno-scientific sense and their writings dismissed in the way that physicists dismiss books like The Tao of Physicsas a bogus metaphorical use of what are, properly speaking, well-defined concepts. How do they defend themselves against this kind of charge?

[Kathleen Lowrey]

John, your story of infosci "flowchart ontology" is somehow really charming, it doesn't horrify me at all. but I have read and enjoyed the Tao of Pooh with no particular sense of shame and my take-home version of the Theory of Relativity is that just as humans everywhere have always suspected, time and space are seriously weird. Pop ontology is okay by me.

That's part of what bothers me about the "high-serious" version of these debates in anthropology (the tone in philosophy itself is mostly just intolerable, I find) -- we're all interested in What It's All About, Really because *no one anywhere* has a definitive handle on it; if anthro owned ontology we wouldn't talk about it anymore. What bugs me is when (and I'll use him again, because I know his work better) a de Castro more or less says, "oh, the Indians are quite sure about ontology. And it looks & has always looked like exactly what I am arguing in this debate right now". You are so right that China / "the East" gets trotted out in those debates in the exact same way. It's not that counter-examples aren't illuminating -- anthro is premised on the notion that they are -- but I don't think anthropology is collectively advanced when they are trotted out as slam-dunks in a field of human inquiry in which there are no slam-dunks.

[John McCreery]

Kathleen,

"flowchart ontology" is charming. Still, just to keep you out of trouble should you ever be talking to a database programmer, ERD are not flow charts. They are ways of ensuring that all of the relations in a database have a one-to-many or many-to-one form. Thus, in the database of creators of winning ads that I have created for my own research, I have Ads and Creators and a problem because multiple Creators can play the same role in creating an ad (there may, for example, be multiple copywriters) or, alternatively, the same Creator can play multiple roles (picking up credits as both creative director and copywriter, for instance). The solution to the problem is to create a third entity, Role, to which both Ad and Creator point in a one-to-many relationship.

Fascinating stuff, to some people....

[Huon Wardle]

ERD ensures that all of the relations in a database have a one-to-many or many-to-one form.

I won't pretend to understand ERD except to comment that if applied ethnographically it might possibly end up looking like 1960s componential analysis. Which raises an intriguing point about the 'many' and the 'one'; because if a central angle here is that there is no simple mappability of society/individual (one to many) or individual to society (many to one) - which is after all what Strathern and Latour are arguing (see Keith's comments) - then the tao of pooh may be our best resort (but not Bourdieu I am afraid). In the new turn 'persons' expand and contract and transform themselves in ways that can't be described according to the individual-is-the-part;society-is-the-whole map - myths or pandemic diseases become persons with their own social capacities and so on. It is hard to overestimate how radical this perspective is if we take it seriously, and in some ways there seem to be two approaches; we work out the ramifications of this new way of thinking intellectually and ethically or we say that it is a fad and we go back to doing what we were doing before.

I would also like to hear more about Amazonian ethnographies that do not follow the well-worn path (i.e. what are the alternatives to the picture of Amazon as mirror of the West that Kathleen indicated?). Some of the writings by Joanna Overing and her students seem to be cutting a different track for instance. I have always admired Overing's paper 'the shaman as maker of worlds' which debatably takes the argument in the opposite direction, envisaging the shaman as a scientific thinker.

[Kathleen Lowrey]

I really admire the work of Anne-Christine Taylor; for example her essay "the soul's body and its states: an Amazonian perspective on the nature of being human" (JRAI 1996). First, she is witty -- this always counts for a lot with me, although it may seem trivial, because it sets the tone for a intellectual conversation in which playfulness is welcome and towering righteousness is out of place. Second, she places a lot of emphasis on uncertainty, and discusses Amazonian perspectives on dealing with the fact that the nature of Being is fundamentally puzzling. This, I think, is the right place to start -- that every human ontology is as much about what is unknowable as what is known. Taylor has a few amazing essays, and has done quite a bit of ethnohistorical work, but hasn't ever laid out a Grand Program on the lines of Strathern, de Castro, Latour, Overing, or Descola (her husband)... but I think her suggestive pieces are much more attractive than any of those Grand Programs.

[John McCreery]

Huon Wardle said:

ERD ensures that all of the relations in a database have a one-to-many or many-to-one form.

I won't pretend to understand ERD except to comment that if applied ethnographically it might possibly end up looking like 1960s componential analysis.

Actually, no. 1960s componential analysis was, in many respects, like classic ontology (Aristotle, St. Thomas). The underlying assumption is that things will fall into neat taxonomies; the category boundaries may differ from what we expect, but the basic notion that there will be a hierarchy of more and less inclusive categories defined by necessary and sufficient properties remains the same.

The underlying premise in ERD is that there will be pointers from one entity to others, so that, for example, shoes might point to socks, socks to feet, feet to left or right feet, which, in turn might have pointers to American football place kickers who kick with one foot or the other . There is no assumption that if entity A points to entity B that they, in turn, must be part of a larger set {A, B} defined by properties that they share. No taxonomy is implied.

[Keith Hart]

Giovanni, I was worried that I was pulling the discussion in an unproductive direction and your thoughtful post (plus Huon's remark) has allayed that fear. It may take me some time to get back to you, but it will be in the same cooperative spirit. Thanks.

Giovanni da Col said:

Keith,
you are a walking encyclopedia of social sciences so I don’t think I can argue with you on the historicization of the concept of society. Let me make a few points related to my previous statement. For me society is first a topological issue. A widespread ontology (to avoid the term Western) which assume society is constituted by individuals. What you need then is to constitute a self-evident entity, defining properties like body, intentionality, emotions, desire, agency, freedom, subjectivity which constitute it.

[Giovanni da Col]

Sorry everyone, I just write down what I have in mind quickly without proofreading or anything, let me correct a couple of mistakes.

Keith,

you are a walking encyclopedia of social sciences so I don’t think I can argue with you on the historicization of the concept of society. Let me make a few points related to my previous statement. For me society is first a topological issue. A widespread ontology (to avoid the term Western) ASSUMES society is constituted by individuals. The problem is how to constitute a self-evident entity called individual through properties like body, intentionality, emotions, desire, agency, freedom, subjectivity. Correct me if I am wrong Keith (you are the classicist) but Greeks did not have a concept of society but private vs. public, oikos/polis but most interesting idioteuein vs. desmoseuein, people concerned only with their private life (the idiotes) and people concerned with the public one. In this case, the topology changes, the individual is not contained within society but you have a Venn diagram in which members of one set could belong to the other but not necessarily. The great thing is that people like Socrates who defines himself at some point as idiotes, could decide to stand apart from the community. While ‘our’ individual could not leave society. I am not a Melanesianist neither I consider myself an ANT experct but my reading and elaboration of Strathern and Latour is the following.

S argues that you don’t have individuals are not logically prior to society but beings-with-relations. Every entity is already connected to others and the problem is how and where to introduce the ‘partition’ and the ontological caesura. Also a clan could operate as a singularity (or ‘individual unit’) and then transform to another configuration. Cuttings and partitions are enacted in certain transactions or ‘aesthetic events’ which make visible the relationships embedded within the object of transactions and the transacting subjects. The latter are nevertheless neither subject or objects since each ‘thing’ contains them and is therefore a form of fetish (more akin to Lacan’s view than to Marx). Of course the problem of scales applies but apparently Wagner suggests that the relationships between different levels are holographic.

The problem for the observer and the entities involved that they both enact the schismogenetic process (Bateson) yet they entertain a generative indigenous epistemology at the same time which also have schismogenetic effects. Every time they/I scale or focus they/I make visible or bring forward the ‘figure’ or the ‘ground’ of the relationships involved. If after all humanity is constituted through a series of partitions we don’t really need the concept of society anymore but to understand the configurations of entities which operate in certain instances and their principle of visibility or concealments.

Latour would say that the problem is how to define the constitutive properties of the thing called society and especially the links, that thing called ‘social’ that links its elements. He does not regard the social as being a connection different from glue or mud. We can see or introduce ties and groupings everywhere so the social would be the movement of what he would call enrollment of entities according to an arbitrary set. But I don’t know Latour in detail and might be wrong.

We always encounter multiplicities first and then we introduce ontological caesuras: separate the desk from the pens on it, see a room as singularity or a container for multiple objects. So it’s not like one vs. many or many vs. one like Huon suggests but question of perspectives which coexists and work according to positions, counting operations and partitions (both from the point of view of the actant/person and the observer).

Tarde would instead say that entities imitates each other so one is mirrored by others and an inter-relationality is established through process of borrowing, contamination and fascination. Society would be established through a kind of epidemiology of affects and social contracts or fellowships established to pursue common goals would be replaced with a theory of viral interactions where selves imitate each other and form a new ‘body’ (my Spinozian reading here) which he may call society (although I think Tarde sees society in non-humans entities such as molecules). Seems to me a ‘hive’ idea of society as a form of collective intentionality achieved through memetic synchronization. I guess in this sense we would need to understand how certain ideas suddenly spread and provoke massive changes (similarly to what Malcom Gladwell in the Tipping point with his theory made of connectors, salesmen, mavens, etc.). Interestingly enough, this idea of contamination is also discussed in Sloterdjik's untranslated magnum opus, Spheres. It was also present in Marsilio Ficino who believed that the eyes could send out rays and love was a contamination by human radiation.

Maybe we should make a discussion group only on comparative cosmologies and social ontologies...

[John McCreery]

Giovanni writes,

In this sense we would need to understand how certain ideas suddenly spread and provoke massive changes (similarly to what Malcom Gladwell in the Tipping point with his theory made of connectors, salesmen, mavens, etc.).

Just as a point of information, there is quite a lot of work on this question in the network analysis literature from which Gladwell took his inspiration. A lot of the mathematics are taken from phase-transition physics. I don't pretend to understand the math, but one point that might be worth considering is the role of (I may be making up this term) susceptibility thresholds.

In one amusing example, it turns out that in network terms, nuclear reactions and bar fights have a lot in common. So long as the explosive elements are separated by other elements that dampen their interactions, no explosion occurs. In the bar fight example, in bar A, those who pick fights are part of a crowd composed largely of people who will jump into the fight once it erupts. In bar B, they are part of a crowd made up mostly of people who don't want to fight. The results are predictable.

In a somewhat more serious vein, epidemiologists have a major interest in this stuff, trying to sort out the implications of different mixtures of those vulnerable, carriers, and immunes in different network configurations.

Among anthropologists, it is, I believe, Dan Sperber who has done the most with these sorts of ideas.

Hope this is helpful.

[Giovanni da Col]

John McCreery said:

Giovanni writes,

In this sense we would need to understand how certain ideas suddenly spread and provoke massive changes (similarly to what Malcom Gladwell in the Tipping point with his theory made of connectors, salesmen, mavens, etc.).

Just as a point of information, there is quite a lot of work on this question in the network analysis literature from which Gladwell took his inspiration. A lot of the mathematics are taken from phase-transition physics. I don't pretend to understand the math, but one point that might be worth considering is the role of (I may be making up this term) susceptibility thresholds.

In one amusing example, it turns out that in network terms, nuclear reactions and bar fights have a lot in common. So long as the explosive elements are separated by other elements that dampen their interactions, no explosion occurs. In the bar fight example, in bar A, those who pick fights are part of a crowd composed largely of people who will jump into the fight once it erupts. In bar B, they are part of a crowd made up mostly of people who don't want to fight. The results are predictable.

In a somewhat more serious vein, epidemiologists have a major interest in this stuff, trying to sort out the implications of different mixtures of those vulnerable, carriers, and immunes in different network configurations.

Among anthropologists, it is, I believe, Dan Sperber who has done the most with these sorts of ideas.

Hope this is helpful.

Thanks John. I am familiar with Sperber and his relevance theory but not with phase-transition physics. What you say is fascinating, I will definitely look into that.

[Giovanni da Col]

Giovanni da Col said:

John McCreery said:

Giovanni writes,

In this sense we would need to understand how certain ideas suddenly spread and provoke massive changes (similarly to what Malcom Gladwell in the Tipping point with his theory made of connectors, salesmen, mavens, etc.).

Just as a point of information, there is quite a lot of work on this question in the network analysis literature from which Gladwell took his inspiration. A lot of the mathematics are taken from phase-transition physics. I don't pretend to understand the math, but one point that might be worth considering is the role of (I may be making up this term) susceptibility thresholds.

In one amusing example, it turns out that in network terms, nuclear reactions and bar fights have a lot in common. So long as the explosive elements are separated by other elements that dampen their interactions, no explosion occurs. In the bar fight example, in bar A, those who pick fights are part of a crowd composed largely of people who will jump into the fight once it erupts. In bar B, they are part of a crowd made up mostly of people who don't want to fight. The results are predictable.

In a somewhat more serious vein, epidemiologists have a major interest in this stuff, trying to sort out the implications of different mixtures of those vulnerable, carriers, and immunes in different network configurations.

Among anthropologists, it is, I believe, Dan Sperber who has done the most with these sorts of ideas.

Hope this is helpful.

Thanks John. I am familiar with Sperber and his relevance theory but not with phase-transition physics. What you say is fascinating, I will definitely look into that. I guess the conundrum we are facing as social scientists is how to bridge the parallax between a systemic view (à la Luhmann) and a subjective one.

[Huon Wardle]

I guess the conundrum we are facing as social scientists is how to bridge the parallax between a systemic view (à la Luhmann) and a subjective one.

I think this last point is critical given that the 'death of the individual' is taken as an apriori by the thinkers in question. I couldn't agree more with Kathleen that these topics demand a sense of humour, though, and there seems to be room for thinking more about laughter and its place. Giovanni's speedily written epics indexing everything from bateson to bar fights also made me think about how these debates come to centre on an imagery of relentless metamorphosis. Of course this is one of the 75 metaphysical constructs John mentioned early on - noone can 'step in the same river twice'. The decentring of the individual and the self-assembly of these leviathans and gargantuas are elements of the same liquid picture. No wonder these are our current concerns perhaps; even Latour could not have envisaged Steven Hawking suddenly becoming the personification of a world event concerning the British national health service, the American far right and innumerable other actors simultaneously pulling apart and reassembling a healthcare programme. Latour quite explicitly states that the relevance of the Amazonian ontologies is that they are 'slow' while the phenomena he is probing are fast: the reason that the Western ontological condition is starting to look like Amazonia is because of this unstoppable transformativity. But the West can do slow too (allow me to be deliberately crude). Keith points to Goody's studies of very tortoise-like global economic transformations, for instance.