A Onça e a Diferença

Sobre o virtual e o atual na ontografia amazônica[]

Abaixo, alguns comentários críticos generosamente enviados por Martin Holbraad em mensagem eletrônica a Eduardo VC, em maio de 2004, respondendo a The Forest of Mirrors. EVC está ruminando desde então um tratamento satisfatório dos problemas levantados por MH. As conexões que MH estabelece com o ifá cubano remetem, entre outros, a um artigo de sua autoria publicado na revista Mana 9/2 (2003),Estimando a necessidade". Veja-se também seu artigo recém-publicado no Journal of the Royal Anthropological Institute 11/2 (2005), intitulado "Expending multiplicity: money in Cuban ifá cults".--Eduardo 18:32, 17 Jul 2005 (UTC)

Voltar a Notas Provisórias

Voltar à página principal


In fact, I find your argument on spirits and multiplicity so persuasive (a tour de force of indigenous ontology) that what I have to say is more in the nature of ‘what, then, is the next problem –the next step?’. After raising these questions, rather abstractly, I’ll outline some further questions regarding a possible comparison between animist and polytheistic spirits/deities.


So, to the ‘conceptual core’. This of course reminded me of that amazing lecture of yours on the Raw and the Cooked to the students in Rio when I was there. I won’t reproduce here the diagram representing the continuous with a line and the discrete with shorter interrupted segments, but, as you showed yourself in that lecture, the distinction between the virtual-molecular-infinite-differential and the actual-molar-finite(internally infinite)-identical maps on to that diagram. So, your point could be put by saying that if myth plays out this contrast quasi-diachronically (from virtual world of maximum differentiation to actual world of discrete self-identical speciation), then spirits locate the conceptual residue of this transition in synchronic fashion (“the disjunctive synthesis which connects-separates the actual and the virtual”). In my own terms what is supposed to be connected-separated is the predicative and the motile (or the conventional and the inventive sensu Wagner). This, to put it bluntly, is the $64,000 question, and one way of reading your paper is as if Amazonian spirits have given us the answer. Below I’ll say why I’m still not sure they have, but let me first say abstractly why I think it’s the $64,000 question.

Simply put, the connection between the virtual and the actual must be a second order connection, right? If it’s first order then it’s either virtual or actual, and that would beg the question. But if it’s second order then the question is what are the candidates? Again, it can’t be virtual or actual. So some relation between these two needs to be postulated that does not simply collapse into either. The temptation, of course, is to elide the question by saying something lame like ‘dialectic’, which is part of what Wagner says. Alternatively one might resort to some kind of trick, one of which is his notion of ‘obviation’, or Marilyn’s ‘eclipse’, which you also use in this paper, as do I in my Ph.D. But again, one wants to ask how the notion of ‘eclipse’ (or whatever) relates to the actual and the virtual in the first place, for the danger is that it might be encompassed by one of them, in which case we’re back to square one. Finally, I guess there might be the option of coming up with a third concept, to be placed alongside the virtual and the actual. That would be quite something, and I haven’t a clue how it might work (for myself I’m still trying to think of ‘tricks’ in various ways, ‘abstraction’ being the one I’m thinking about at the mo, as we discussed in Rio). One way or the other, the problem is a real big one!

So how does your tantalising analysis of Amazonian spirits fare in this connection? Well, it seems to me that the answer has two parts. First, like the Yanomami themselves, you consider that the virtual encompasses the actual (another way of saying it is that it’s more ‘real’). So, since spirits are the residue between the two, they are in effect on the virtual side of things. As you emphasise, they are, after all, intensive multiplicities which is the hallmark of the virtual. Indeed, some of the argument seems to me to elide the 2nd order ($64,000) question about the relationship between the virtual and the actual with the 1st order question about how best to characterise the virtual itself (i.e. as opposed to the actual). Spirits, in effect are still what animals and humans used to be (in the absolute past of myth) but seem no longer to be, encapsulated as they are as self-identical forms of affects. But –and this is a crucial ‘but’- since humans and animals still ‘have’ spirits (‘true’ animals/humans), their encapsulation is really a chimera: “But if Amazonian concepts of ‘spirit’ are not rigorously speaking taxonomic entities... it is probably just as improbable that notions of such as ‘animal’ and ‘human’ are elements of a static typology of genuses... I’m led to imagine a single cosmic domain of transductivity...”.

This leads to the second part of my reconstruction of your argument. For if Aristotelian speciation is –for you and the Amazonians- derivative from and, therefore, ‘less true’ than Deleuzian difference, then the old $64,000 question is cast in this way: how might the derivation of actual from virtual best be conceptualised? The most relevant passage in the paper states:

“The originary transparency or infinite complicatio where everything seeps into everything else bifurcates or explicates itself, from this point on, into a relative invisibility (human souls and animal spirits) and a relative opacity (the human body and the somatic animal ‘clothing’) which determine the makeup of all present-day beings. [Again:] Relative invisibility and opacity because reversible, and reversible since the ground of pre-cosmological virtuality is indestructible or inexhaustible.”

In other words the pure positivity of infinite difference resolves itself (in the visualist key of Amazonia) by the appearance of negativity: the ‘real’ becomes invisible, hidden behind the opaque clothing of the speciated body. It is as if the negativities that are constitutive of the discrete (actual), viz. the ‘gaps’ of the structuralist diagram, are themselves derived from the negation of the virtual truth (god, I’ve made you sound like a Marxist!). Despite my obscure expression, I think this is quite a telling move. Effectively, this part of the argument states that the virtual and the actual relate to each other in the same way as different species of the actual relate: they are separated by extensive ‘gaps’ of negativity.

So, it seems to me that your argument can be set up in a somewhat paradoxical light. In your first move, the 2nd order relation between virtual and actual is itself virtual, to the extent that spirits are, like other mythical beings, multiple and intensive links that connect-separate (this paradoxical expression is indicative) the visible with the invisible (the actual with the virtual). In the second move the 2nd order relation is itself actual, to the extent that spirits are either visible or invisible, and this either/or business (negation) is the extensive armoury of the actual world of speciation. To me this paradox seems perfectly natural, precisely because I can’t imagine what third element (concept) could serve to articulate the passage from virtual to actual. But you see why I don’t think the problem is resolved?

Interestingly, there’s the sketch of a third possible move later on in the paper, I think, which would be to formulate a ‘hybrid’ of the actual and the virtual. I’m referring to the section where you explain the ‘anomal’ nature of the spirits:

“a kum-being is at once an archetype and a monster, model and excess, pure form and hybrid reverberation... in a single figure.”

Of course, hybridity is, again, a form of virtuality, so we return to the position that the virtual always trumps the actual (like in your first move above). Yet, logically speaking, if the virtual can indeed be hybridised with the actual, that would seem to indicate that there would be some kind of difference between senses of virtuality here. After all, in what sense would the virtual be hybridised with the actual, if it thus just remained virtual in the same sense as before? In other words, we do, after all, need some distinction between virtual and hybrid if the former will be shown to become the latter.

If this sounds too abstract, let me put the idea a bit more ethnographically. You say that spirits in a certain sense can be understood as being –to put it too simply- both ‘normal’ and ‘abnormal’ (hence hybrids of the actual with the virtual?). In this case you are speaking of the ways in which Yanomami and Yawalapit? people imagine spirits (tiny and gigantic respectively), and you conclude that “we are faced with the two complementary extensive (spatial) schematisms of the idea that a spirit is an intensive and ‘anomalous’ multiplicity”. The problem here, for me, is that again the virtual is trumping the hybrid, as it were. So I was wondering whether there might be a way of pre-empting that consequence, i.e. a way to show that spirits are a genuine ‘function’ (in your sense) of normality as well as abnormality.

Of course I don’t have a solution (that’s what the $64,000 are there for!). But, predictably enough, my inclination is to explore whether my general and still vague ruminations about ‘abstraction’ might be useful in this respect. This, in my view, might be possible because the way I define abstraction is actually quite intimately connected to the notion of ‘negativity’, and, as I was saying above, negativity is a key trope of the actual. Hence if, as I outlined in my e-mail to Alberto Corsin Jimenez (the one we discussed in Rio), abstraction can be formulated in motile (virtual) terms and contains the possibility of ‘the negative’ within it, then perhaps abstraction might turn out to be some kind of hybrid of the virtual and the actual.

So, let’s imagine what abstraction might refer to in relation to Amazonian spirits. In very general terms, the problem to me seems to be that if we define the world of the spirits in purely virtual terms, we arrive at a pretty anarchic picture which actually does not correspond to the way Amazonians talk about it. If spirits were just virtual potentialities then it would seem as if the message was that they could actualise in any way possible, they would be –to use your terms- signs of any vehicle and vehicles of any sign. Or, to use your other idiom, of ‘infinity’, to say just that spirits are virtual multiplicities is to stop short of distinguishing different orders of infinity (I remember spending the best part of a night as an undergraduate with a mathematician friend, trying to understand how distinctions between infinite series can be manipulated in thought –not that I succeeded mind you). But it seems that ethnography of spirits is more sophisticated than that. Certain spirits do certain things, other do not; some of them transform in certain ways, others in others, etc.

I am 100% convinced by your argument that these organisational features should not be taken as Aristotelian taxonomic ‘constraints’. But nor should they be ignored as irrelevant residues of representational thinking. Rather, it seems to me that they intimate a rather profound possibility: that the motile-virtual continuum is not necessarily at odds with the possibility of distinction, i.e. (in some uncommon sense) ‘negativity’. Of course, the major danger when one begins to think like this is that if the Scylla is virtuality trumping actuality, the Hayrides is actuality imposing itself on the virtual becoming of the spirits. In other words we don’t want the use of negativity to end up making heterogenic becoming look like an ordinary predicative ‘change’ or ‘process’, as you say. So the question is: what does negativity look like when it isn’t a predicative ‘negation’ (p, ~p)? And the answer that I’m wondering about has to do with the directionality of becoming. After all, the ethnographic problem, as it were, was precisely that the becoming of the spirits, their transformativity, is in some non-trivial sense organised, and this organisation comes down to the fact that each motion of becoming is potentialised in a determinate way, i.e. towards one particular direction (e.g. feline to human and back). Directionality, in other words, is constitutive of the very concept of virtuality (insofar as the virtual is, as I would say, motile). And directionality in this sense entails a re-definition of negation which makes ‘the negative’ definable in terms of the central virtual trope, namely difference. More specifically, this definition of the negative gives exactly that 2nd order concept that we were looking for: differences (1st order becomings from jaguar to human, say) are different from each other (2nd order differences between becomings). It seems to me that this passage to the second order (which of course is itself intensive and infinite) expresses the form that a hybrid of the virtual and the actual might take. The hybrid is virtual inasmuch as it implies infinite becoming. And it is actual inasmuch as this infinite becoming works in 2 dimensions, by which difference between differences can be articulated. Hence I would say that if abstraction is the possibility of 2nd order becoming, then what looked like ‘speciation’ in actuality, looks like abstraction in the hybrid order of the virtual-actual. And if spirits are supposed to avail us of the link from virtual to actual, then they would have to work at this 2nd order level of the hybrid of virtual-and-actual that abstraction implies.

It occurs to me that another way of communicating this point would be to speak in terms of the fact that virtual becomings are always, so to speak, ‘thick’. Might this be a way of rehabilitating L-S’s theses about the logic of the concrete –bricolage? If bricolage implies an infinite becoming, it also implies an infinity which is qualitatively demarcated by the ‘content’ (to use an actualist word) of its ‘raw materials’ –the jaguar, the human, etc.). Unfortunately L-S contrasts this with the ‘abstract’ infinity of the engineer, but this may well be just a terminological problem. My point would be that abstraction (2nd order difference) need not operate only in the ‘mathematical’ space of the engineer, but also in the ‘striated’ one (sensu Deleuze, I think) of the mythologist, and of the spirits.


Re. multiplicity and the possibility of perhaps writing something together along comparative lines. Three fascinating projects spring to my mind, each of which relates in different (though probably connectable) ways to your argument about the multiplicity of spirits and mine about the multiplicity of powder and money. Before sketching these out, it’s worth saying that our respective perspectives on multiplicity seem to me highly compatible. As I understand it, your central claim is that, due to their irreducible multiplicity, spirits (and shamans) are best interpreted not as self-identical (non-human) subjectivities, but as regions, say, of ontological conversion. This because multiplicity implies intensive difference, and intensive difference is ontological in character: as multiplicities, spirits are (as oppose to ‘have’) the potential to become different things. On the other hand, my point about the ‘pure’ multiplicity of powder and money (a fractal with no structure, the unadulterated ‘many-ness’ of particles) is basically the other side of the same coin –or, better, ground to (your) spirit-figures. Ontological conversion, in your view, implies a wholesale overhaul of our analytical frame, from actual self-identical entities to potential (virtual) differential relations. To this I add that it also requires as a corollary the logical priority of motion rather over rest. So, if spirits are ontological conversions they are also trajectories, i.e. vectoral relations. But clearly this logical reversal must require a further corollary, regarding the ‘space’ within which these vectors move. For they cannot move in the extensive space of, say, a map (i.e. any form of Cartesian space). This would not only be analytically inconsistent, but also ontologically anomalous, no? So we need to ‘potentialise’ space too. And that’s my argument regarding the ‘pure’ multiplicity of powder and money in Ifá. If every act of consecration requires these two substances for its performance (as is the case in Ifá), that’s because each of them constitutes precisely a space that is eminently (immanently?) intensive: it is a space that moves inwardly (reconstituting its own particles) in a way that couples perfectly with the vectoral motion of the spirits and the shamans (well, the orichas and the babalawos).

So, the first idea that occurs to me is to explore this compatibility between your point and mine in a proper article. The potential for doing so is ethnografically obvious, since –remarkably!-, in Kopenawa’s words: the spirits “are as tiny as specks of sparkling dust and to be able to see them you must inhale the powder...” (my emphasis obviously). I suspect that powder might be pretty important in many more contexts in the Amazonian literature (I don’t know), but even this is enough to go on, no?

But this leads me to the second area for a possible project, which might, in my opinion, be quite important. For it does seem to me that the differences between Amazonia and Afro-America (to put it very broadly), might be very interesting indeed. I say this because from my point of view, your debate with Descola, Ingold. Morten etc., about animism, totemism, analogism, and naturalism could be rather amplified by attending to the peculiarities of what people like myself are dealing with, namely ‘polytheism’. I’m in no way talking about fitting Ifá into Descola’s taxonomy –I’m not particularly excited by this kind of project. But it seems to me that many of the features that you describe in this paper and elsewhere pertain to the ‘animist’ mode, and do not necessarily translate to the ‘polytheistic’.

I won’t begin to list these here (I’ve been writing this for days now, and I’m supposed to be doing fieldwork!), but take multiplicity as a possible focus. Amazonian spirits, it seems to me, are multiple in a much more disorganised and token-specific way than the orichas. And this, as far as I can tell, is because the orichas’ potential for mutation (which of course is supreme) is not tied to a ‘flat’ perspectivist field, but a highly hierarchical order of being (which does recall some elements of Descola’s ‘analogist’ cases, though none of his analysis of analogism). Hence, for example, the classic and pertinent distinction between shamanism and spirit-possession: shamans in a sense become spirits (or, qua shamans, are spirits), whereas babalawos cannot become orichas, they can only let them speak through the oracles (and santeros can be possessed by them). In general, deities’ human ‘mutations’ are invariable spoken of as ‘descents’. Indeed, this verticality (again, interesting to think of this in terms of ‘abstraction’) is a very prominent feature of Ifá cosmology, and babalawos emphasise that the whole pantheon is ordered in a hierarchy from the sky to the earth, with Eleggua (the trickster deity) mediating between the different levels. So multiplicity and hierarchy are the order of the day here...

Now, as you can probably gather, I am not at all clear how to theorise all this in terms of the previous discussion. But one idea which to me seems confluent with your sense of multiplicity, is that ‘many’ does not only refer to the hierarchically ordered different orichas, but to each oricha him/her self as well: each oricha is said to have many different ‘paths’ (caminos, often referred to in the Cuban literature as ‘avateres’). So one cannot speak of, say, Yamayá, other than in very general terms, because in each case she appears as a particular ‘camino’ of Yemayá, which has particular characteristics in terms of behaviour, likes and dislikes, age, proficiency in magic, etc. A very obvious point, then would be to say that what we have here is something like a ‘bundle theory’ of identity (re. the philosophical literature on tropes we spoke about). There is no identity of an oricha outside the characteristics associated with its various ‘paths’ –so, many predicates and no subject, as it were. This might be a starting-point for exploring how our common interest in multiplicity might be modulated in contrasting ways in Ifá and in Amazonia...

The third area which really grabbed me in your paper was your fascinating aside about words (“it is the words of the xapirip? which augment our thoughts”, and your “if studious reason is the hallucination proper to whites, then writing is their shamanism”). For it seems to me that Kopenwa’s analogy between powder and words is actually spot on, in so far as words really are multiplicities, as are sentences (was it Quine who suggested sentences rather than words are the proper units of meaning?). It so happens that the project I’m ending up pursuing right now here in Cuba is a comparison of the different ways the babalawos’ ‘imagination’ is constituted by ‘reading’ Ifá (from the papers where myths and recipes are exchanged between initiates), teaching it in newly formed ‘schools’ (again, using notebooks etc.) and the ‘traditional’ way, which is to ‘speak Ifá’, i.e. long conversations babalawos have with each other in various contexts, which are basically something like knowledge contests. And the power of the word in this context is supreme, and to me very much worth exploring.