Ontology, ‘hauntology’ and the ‘turn’ that keeps anthropology turning

Twentieth-century anthropology has been operating with the assumption of one nature and many cultures, one reality experienced and lived in many different ways. Its primary job, therefore, has been to render the otherness of the other understandable, to demonstrate that although different it is also the same; in short, to show that although other, others are people like us. The latest theoretical paradigm, known as the ‘ontological turn’, appears to reverse this assumption and to posit many natures and one culture. Whether it does in fact reverse it and constitutes a meta-ontology, as critics have pointed out, or it is only a heuristic, methodological device, as some of the proponents of the ‘turn’ have recently argued, the contention of my article is the same: first, this move – the ontological – is made in the hope of doing a better job in redeeming otherness than earlier anthropological paradigms; second, it fails as they did – in the same way and for the same reasons.

ethnocentrism -hopeless and impossible because, paradoxically, anthropology becomes ethnocentric in the very attempt to defeat ethnocentrism. It is this paradox that I wish to highlight here.
The debate concerns what for most of the 20th century has been the discipline's key analytical category -culture -and whether it is up to the task at hand, or whether we are not all, anthropologists and the people we study, better off replacing it with an entirely different category -ontology. 2 To turn first to the anthropological task, let us note that from its inception as an academic discipline in the 19th century, the key premise of anthropology has been the principle of human unity, the idea that human beings are essentially and fundamentally the same. Hence the problem -the 'ethnological problem', as anthropology's foremost historian, George Stocking (1987: 50), calls it -has always been how to uphold this principle of unity or sameness in the face of often radical empirical difference. The anthropological task, then, has been to demonstrate that although different, others are people like us, that their otherness is both different -for it must be taken seriously and not simply explained away -and the same at the same time. In effect, the task has been to manage, to the extent that can be managed, this very logical contradiction. As Stocking says, in the 19th century the ethnological problem was about 'demonstrating the unity of the human species' -or the 'psychic unity of mankind', as the foremost Victorian anthropologist, E. B. Tylor, famously put it; that is, affirming what is denied by racism. In the 20th century, it became, in addition, about demonstrating what we might call 'the unity of human culture'; that is, affirming what is denied by ethnocentrism. This is why, as Geertz says in a seminal essay, 20th-century anthropology sought 'a more viable concept of man', one where cultural diversity is taken seriously and not explained away, and, 'at the same time, one in which the governing principle of the field, ''the basic unity of mankind'', would not be turned into an empty phrase' (Geertz, 1973a: 36;emphasis added).
Managing the contradiction entailed in the anthropological task is easier said than done. As I have argued elsewhere (Argyrou, 2002), there is no anthropological paradigm in which cultural difference does not emerge as cultural inferiority, not a single one that has not been found guilty of ethnocentrism. No better proof of this is the impatient, radical, overarching and badly misunderstood critique of anthropology by another turnthe 'postmodern turn' of the 1980s. Much like postmodernism in other human sciences, the critique was epistemological, in this case, a critique of anthropological representations as 'fictions' and 'partial truths' (Clifford, 1986), an argument misread by many as anti-anthropological. A careful consideration of the postmodern claim however, suggests that not all representations were meant to be understood in this way, certainly not the 'governing principle of the field', the idea of human unity or sameness. Without this representation the critique of the discipline would not be possible. For how can anyone know that anthropological representations are fictions without a representation to act as the standard of truth? My argument, then, was that although not readily apparent, the postmodern critique sought to safeguard this fundamental idea of sameness from whatever representations arose in the discipline that contradicted it. It sought to do exactly what the discipline itself has been striving to do: to solve 'the ethnological problem' by demonstrating that, although different, the otherness of the other is also the same as the sameness of the self.
Such, I will contend in this article, is also the desire that motivates 'the ontological turn' -the latest push that keeps anthropology turning and going round and round in circles. This, I wish to argue, is ultimately the reason for proposing to replace the notion of culture with the notion of ontology. To put the matter rather schematically, 'ontological anthropologists', having come to the same realization as the 'postmodernists' a few decades earlier -that no matter how well intended, anthropological representations cannot avoid being ethnocentric -sought ways and means of neutralizing representations. The difference, subtle but significant, is that for these latter anthropologists, what is at fault is not representation per se; hence for some (e.g. Holbraad, 2012) it is not a matter of questioning the possibility of truth, either. For them, the problem is not epistemological but ontological.
What ontology is supposed to mean, exactly, in this context is not entirely clear, not least because, as one anthropologist associated with the 'turn' admits, it is possible 'that some ontologists themselves are unclear about what it is they are advocating' (Salmond, 2014: 159). In general, the argument has been read as suggesting the existence of different worlds -our world and the worlds of the people we study. As we shall see, this reading is consistent with the work of one of the leading figures of the 'turn', Viveiros de Castro, and the argument of 'multinaturalism'. And while some 'ontological' anthropologists dispute this reading and argue that the notion of ontology is a heuristic, methodological device (e.g. Holbraad, 2012), others accept it as a theoretical innovation to be proud of (Pedersen, 2012: [online]). Whatever the case, they all seem to agree that the problem is ontological rather than epistemological. This is because of the existence of different 'ontologies' in the world -which is to say, different definitions of reality, or, better still, different truths about reality and not simply different representations of it (Salmond, 2014).

Taking seriously and doing justice
In my own work I also argued that anthropology's problem is not epistemological but ontological. Because Viveiros de Castro (2003) referred to this argument, and because what I meant is very different from what 'ontological anthropologists' mean, I feel it is necessary to clarify my position before proceeding further.
In the paper that Viveiros de Castro refers to and the book that followed (Argyrou, 2002), I was raising the question as to whether the problem that anthropology strives to solve is solvable. And I was saying, drawing on authorities of the likes of Hegel and Derrida, that it is not solvable, because difference is an inescapable part of the world. Put simply, without difference the world would be Pure Being, which, as Hegel argued, is nothing. This is what I meant by 'ontological problem'. I argued that the kind of unity that the discipline presumes would appear if only ethnocentrism was not in the way -'Pure Humanity', as I called it elsewhere (Argyrou, 2005), after Hegel's notion of Pure Being -cannot appear. Whether we like it or not, in society and history there can be no such unity.
Had I been aware at the time of Derrida's (1994a) book Specters of Marx, I would have said that anthropology's problem is 'hauntological', as this idea captures its nature more precisely. As others have noted (e.g. Peim, 2005), this notion is a variation on earlier Derridean themes such as trace, supplement and diffe´rance. In broader terms still, one can speak of hauntology rather than ontology because, as Derrida and before him Heidegger have shown, Being does not appear in the empirical world and cannot be present. Yet although it is nowhere to be found in society and history, Being is not 'nothing'. After all, we think it, name it, speak about it: every time we use the word 'is', says Heidegger, we name Being, even though we have no idea what it might be. My contention, then, is that a similar case can be made about the state of human unity, purity and innocence that anthropologists implicitly posit in their struggle against ethnocentrism. Following Derrida's playful use of the theme of the spectre or ghost that exists and does not exist, is present and not present, we may say that anthropology's problem is hauntological or spectral. What it considers to be real is a ghost, present as a disembodied, insubstantial, immaterial being; but precisely because of this, it is not present, since a body without substance, materiality, flesh and blood is not a real being, but rather a vision, a phantom, an apparition.
Yet liminality -being neither here nor there, neither present nor absent, dead or alivedoes not exhaust the idea of the hauntological. Unlike liminality, it is unsettling, the return of the repressed that disturbs and spooks the living. It is this aspect of the hauntological that has been especially influential in the human sciences; and it is this, too, that has come under attack by critics, for being, for example, apolitical -sacrificing specificity and the possibility of critique for a generalized, 'vacuous' structure (Luckhurst, 2002). The structure is no doubt general, but this does not make it vacuous or apolitical. It certainly helps us to understand why anthropologists are possessed by the idea of Pure Humanity and obsessed with it, why we seem unable to rest until we accomplish what it is demanding of us: to demonstrate that it is a real being rather than a phantom, a possible experience rather than a figment of the imagination. Hence the fact that every failure to do so acts as a spur for another attempt, the latest being the ontological. It should be clear, then, that something is haunting us -something dark, lurking in the shadows behind or below the state of plenitude promised by Pure Humanity. I shall return to this in the last section of the article.
In contrast to this reading of anthropology's ontological problem as a problem of haunting, there is the meaning attached to it by the anthropologists of the 'ontological turn'. For them, the discipline has an ontological rather than epistemological problem because it fails to take into account other realities or, at any rate, other truths about reality. It assumes that there is one reality or nature shared by everyone, and merely different ways of representing, understanding, living it; in short, that there are different cultures. It further assumes, without explicitly saying so, that our understanding of this common reality -and this is where ethnocentrism rears its head -is superior to the understanding of everyone else. As we have seen, the ontological argument -to the extent that there is only one -is that there are different ontologies in the world, not simply different representations of it. This could mean that there is one reality and different truths about it, in which case, as with different representations, a decision needs to be made as to which truth is 'the truth'. 'Different ontologies' could also mean different realities, or, as Viveiros de Castro (1998,2004,2013,2014) seems to suggest with the term 'multinaturalism', different natures or worlds. Whatever the case, the ontological argument implies that contrary to what I am suggesting here, the ethnological problem may be solvable. What needs to be done, it seems, is 'simply' to recognize that our representations are limited and need to be reconceptualized, redefined and refined until the contradictions that 'tempt you to say that your informants are being ''irrational''' (Holbraad, 2012: 255) are removed. 3 A happy end to the story, one might say. But perhaps there is no end, let alone a happy one; perhaps the end is also the beginning, and the more we strive the faster we go round and round in circles. Let us look at the ontological argument in greater detail.
A key claim of the ontological turn is that anthropologists do not take native cultures seriously and do not do them justice (e.g. Viveiros de Castro, 2003;Henare, Holbraad and Wastell, 2007;Pedersen, 2011) -a claim that itself appears not to be taken all that seriously by critics (e.g. Heywood, 2012;Laidlaw and Heywood, 2013). This may be partly owing to the odd, and, as far as I can tell, odd one out, self-admission that 'the ontological turn takes itself seriously' only to a 'limited degree' (Pedersen, 2012: [online]). But it is also partly owing to our assumption -not unjustifiable -that if there is an academic discipline that has been treating native cultures with seriousness and respect this is anthropology. It seems, however, that there may be more substance to the ontological claim than appears at first sight.
It concerns the way we treat the otherness of the other when it blatantly contradicts what we take to be true. The easy way to deal with such contradictions is to say that what others think, say, or do is wrong, or irrational, or absurd, or some such. But anthropologists are not in the business of dismissing otherness as lack and cultural inferiority. As we have said, ethnocentrism is the arch enemy and needs to be defeated at all costs. The only way out of this difficulty -assuming there is one -is to argue that what others think, say or do refers to something other than what they think it refers to; that it is rational and meaningful, but in a way they are unaware of, that it does make sense, but in a way they do not understand. In this way, the contradiction is removed, but -and this is a big 'but' -ethnocentrism remains firmly in place. Unwillingly and unwittingly, anthropologists reproduce it by claiming for themselves superior knowledge of native life. It is correct, then, to say that anthropologists do not do justice to native culture. Insofar as they explain it as something other than what natives take it to be, they explain it away. Yet this is not intentional, as ontological anthropologists seem to suggest. It is the inevitable outcome of the impossibility of human unity or sameness. Let us look at a few examples of how anthropologists explain native cultures away used in the ontological literature.
Take the Nuer claim that 'twins are birds', says Holbraad (2010: 184). Apparently the Nuer 'assert what we take to be false. If we know what ''twin'' means and what ''bird'' means at all we also know that twins are not birds. No amount of relativist fudge can get us out of the fact that, as far as we are concerned, the Nuer are saying something wrong.' In fact, as far as we are concerned, they are saying something so obviously wrong that the anthropologist may be tempted to say what she is not supposed to say -that 'they are being irrational'. As they cannot/must not be irrational, they must be saying something other than what they are saying, something they are unaware of saying. As Holbraad notes, anthropological paradigms have come up with all sorts of suggestions as to what natives might be saying when they say (or do) apparently irrational things, such as that 'twins are birds': Is it because thinking in this way serves some purpose for them (functionalism)? Is it because of something about the way the brain works (cognitivism)? Is this because such a view makes sense in the context of other views that they hold (interpretivism)? Or are they just being metaphorical in some way (symbolism)? (Holbraad, 2010: 183) We can wholeheartedly agree with Holbraad that all these paradigms -and others he does not mentioned, such as structuralism -explain native culture away, and that in this sense they do not take it seriously and do not do it justice. This is also what Viveiros de Castro argues, but in a way that I find highly problematic. For it implies intentions that are not present and draws attention away from the real problem, which is structural.
What makes the native a native is the presumption, on the part of the anthropologist, that the native's relationship with his culture is natural, which is to say, intrinsic, spontaneous, and, if possible, nonreflexive or, even better, unconscious. Thus the native gives expression to her culture in his discourse. The anthropologist does so too, but if he hopes to be something other than a native, he must also be able to express his culture culturally, which is to say, reflexively, conditionally, and consciously. (Viveiros de Castro, 2013: 475;emphases added) Thus, Viveiros de Castro goes on to say, although the notion of culture places the native and the anthropologist on equal footing -as they both have a culture -this equality is only de facto and 'does not imply equality de jure -that is, an equality with regard to their respective claims to knowledge. The anthropologist tends to have an epistemological advantage over the native ' (2013: 474). Viveiros de Castro is right in saying that the anthropologist emerges as someone who knows native culture better than the native. What is problematic is the presumed 'presumption' and 'hope' of the anthropologist. It imputes an intention that seems to me to be totally alien to how anthropologists think -the intention of distinguishing themselves and asserting the superiority of their culture vis-à-vis native cultures. Holbraad (2010: 184) makes a similar claim. He diagnoses lack of humility on the part of anthropologists and considers 'the culturalist take on alterity [to be] downright presumptuous'. As he goes on to say, the 'culturalists imagine for themselves unlimited powers of comprehension'.
It may be argued of course that this is merely a hyperbolic way of speaking about the 'culturalists'. The 'ontologists', it could be said, are well aware that what motivates anthropologists is not the desire to distinguish themselves and their culture but, on the contrary, the desire to take native cultures seriously. But if that is the case, the 'ontologists' must do two things: first, recognize that there is a paradox and highlight it; second, explain in some detail how their own efforts are able to overcome it. This is something that to my knowledge no one in the ontological camp (or anyone else in anthropology for that matter) has done. Hence the need for an article such as this that does highlight the paradox and raises the question of whether anyone can overcome it. My contention is that no one can. Anthropologists, including those of the ontological turn, have no choice in the matter. Whenever something in native culture contradicts what they take to be true, they cannot afford to take it seriously and must explain it away. Either this, or give in to the temptation that Holbraad identifies -to brand it irrational and in this way become consciously ethnocentric. Since the latter is not an option, they must presume that what natives say or do has no intrinsic value in and of itself and that the real value is to be found in what what they say or do says or does behind their back. Such is the anthropological predicament.
As we have said, anthropologists of the ontological turn think that by positing different worlds or different truths about the world, as the case may be, they can overcome this predicament. If, say, the Nuer truth about reality is different from ours, then perhaps what they (or anyone else for that matter) say about twins and birds (or anything else that strikes us as irrational) may not contradict what we know to be the truth. 'The Nuer may appear to be asserting that twins are birds but may in fact be saying something quite different -something we fail to grasp, not because it contradicts what we assume to be true about twins, but rather because it goes beyond our own assumptions.' In other words, if the Nuer truth about reality is different from ours, what counts as twins when they say that twins are birds may be very different from what it counts as when we say they are human beings. If so, the task of anthropology becomes rather different. 'Instead of explanation or interpretation, what is called for is conceptualization' (Holbraad, 2010: 183-4). Interpretation and explanation will simply explain native life away and not get us very far. If we truly wish to understand other societies, and if we wish to do them justice, we must be prepared to refine, redefine and reconceptualize our categories. And we should keep doing so until claims like 'twins are birds' no longer strike us as irrational -until the reconceptualized categories render definitions 'of powder as power reasonable, rather than absurd' (Holbraad, 2011: 14). At that point, the anthropological task would have been completed and the ethnological problem solved.

'The ethnological problem'
Viveiros de Castro's argument of multi-naturalism draws inspiration from what he calls 'Amerindian perspectivism', indigenous ideas about how humans, animals and spirits see themselves and one another. As we have said, this theory is a symmetrical inversion of western '''multiculturalist'' cosmologies' -'perhaps too symmetrical to be more than speculative ', Viveiros de Castro admits (1998: 470). While western ontology posits the unity of nature and the plurality of cultures, Amerindian perspectivism posits the unity of culture and the plurality of natures. Because of the plurality of natures, the main protagonists in this drama see each other differently: humans see themselves as humans, animals as animals and spirits as spirits, while from their perspective animals (predators) and spirits see humans as animals (prey) and animals (prey) see humans as spirits or animals (predators). Because of the universality of culture, on the other hand, all the protagonists see themselves in the same way: animals and spirits see themselves as humans and experience what they do in the form of culture.
They see their food as human food (jaguars see blood as manioc beer, vultures see the maggots in rotting meat as grilled fish, etc.) . . . their bodily attributes (fur, feathers, claws, beaks, etc.) as body decorations or cultural instruments . . . their social system as organized in the same way as human institutions are (with chiefs, shamans, ceremonies, exogamous moieties, etc.). (Viveiros de Castro, 1998: 470) The theory has been criticized on all sorts of grounds, some of them more justifiable than others. If one were to take stock, one could perhaps first mention criticisms of an empirical nature. Thus, experts in Amazonian ethnography have dismissed Viveiros de Castro's portrayal of indigenous ideas as a 'misrepresentation and mistranslation of the form, content and meaning' of Amazonian cultural systems. This, the argument goes on, is the result of developing the ontological claims not on the basis of ethnographic data but through a kind of 'philosophical dialogue between ideal-typical formulations' of modernist and Amazonian ideas (Turner, 2009: 18). In a similar vein, Ramos (2012: 482) criticizes Viveiros de Castro for giving the 'false impression' that the Amazon is a homogenous cultural area, thereby slipping into 'reductionism, oversimplification and overinterpretation' (ibid.: 483). In addition, Ramos criticizes de Castro for diminishing 'the intellectual value of indigenous thinking' through the use of such terms as 'cosmology ', 'myth' and 'cannibalism' (ibid.: 485). For others (e.g. Bessire and Bond, 2014: 449-50) the major problem with the theory is that it 'defers' the critique of 'actually existing politics' and 'real-world collisions and contradictions' in favour of 'a message of messianic redemption' of indigenous populations -as if, we may note in disagreement, the redemption of otherness is not a political issue and not part of the real world. More justifiable is Bessire and Bond's observation that the theory 'polices' the Great Divide between the modern and the non-modern, although the use of the term 'polices' is problematic, as it suggests that this is done intentionally. Still other critics (e.g. Vigh and Sausdal, 2014) raise methodological questions about the possibility of anthropologists ever being able to understand something that is posited as completely other. They also express concerns about essentializing difference and how such images of radical alterity could be exploited outside academia for political reasons -as if, we may note again in disagreement, the inability to control the uses of one's discourse ought to prevent it from circulating in the first place. This is by no means an exhaustive summary of criticisms, but nor is it meant to be. It is, I think, a fairly representative sample of what has been said about the ontological turn, and hence a good indication, also, of what remains to be said. And there is much to be said, as my concern here is not to criticize the 'ontological turn' (and not to praise it either). Rather, it is to show, first of all, that it constitutes the latest attempt to grapple with the 'ethnological problem' in its 20th-and 21st-century version -the redemption of the otherness of the other as the elimination of ethnocentrism; and, second, it is to examine if it can solve this problem and succeed where all other anthropological paradigms have failed.
It may be clear perhaps from what has just been said that my view of the ontological turn is very different from the view of Bessire, Bond and others in the same camp. For them, the 'turn' is merely an epiphenomenon of a much more 'substantial' reality: 'we assert that the soteriological figure of ontological alterity is a crucial metanarrative of late liberalism' (Bessire and Bond, 2014: 450; emphases added). In 'asserting' this conviction, however, they miss the mark completely, since it can be easily shown that the redemption of 'alterity' has been the 'crucial metanarrative' of all anthropology since its inception as an academic discipline. As I have already noted, what I wish to do here is to examine whether ontological anthropology can succeed in its aim to redeem the otherness of the other. In this respect, I find more relevant the argument made by a number of other critics, even if for reasons different from my own: namely, that in positing multiple realities or natures (even if, I will argue below, only as a methodological, heuristic device) the theory presupposes a meta-ontology, a definition of a world that contains different worlds (e.g. Laidlaw, 2012;Heywood, 2012;Keane, 2013;Laidlaw and Heywood, 2013).
I shall return to this point below. Here, I should note that in a certain sense some of the critics discussed above have already answered the question that I am posing herenamely, whether ontological anthropology can avoid ethnocentrism. As we have seen, it has been criticized for reproducing the Great Divide between the modern and the nonmodern, essentializing alterity, diminishing the value of indigenous knowledge through the use of 'inappropriate' vocabulary, and so on and so forth. Yet although these criticisms are not wrong, they are piecemeal responses to what I think is a more general and systemic problem in which all anthropology is implicated. The issue, then, is not simply that ontological anthropology cannot avoid ethnocentrism. It is, rather, that it cannot avoid it for the same reasons that every other anthropological paradigm before it could not. Because these critics do not recognize or do not wish to recognize the nature and extent of the problem, they fail to raise the question as to whether it is solvable at all, and if not, why not -which is what I wish to do here. Yet the problem cannot be swept under the carpet. That is why every few decades there appear heroic attempts to solve it -the most recent being the postmodern, and now the ontological turn.
For some of the critics discussed here, taking others seriously and learning from them can be done without positing multiple worlds. 'We should be open to the possibility that what we learn from our ethnography can tell us something we don't already know about what kinds of things there are in the world', says Laidlaw (2012: [online]) in his review of Pedersen's book on Mongolian shamanism. 'This is an important proposition for anthropology', he goes on to say, but there is no need to posit other ontological realms 'where things can be true without having to be true for the rest of us'. Doing so is both 'unsustainable' and 'unnecessary'. Perhaps; but the issue, here, is not whether we can learn something that we do not already know. Rather, it is what to do in all those cases where what we already know to be true is contradicted by what we learn from our ethnography, where what is true for them is not true for the rest of us. This is the question that the 'ontological turn' is raising.
For others who recognize the aim of ontological anthropology in more explicit terms -that 'taking seriously' ultimately means 'taking indigenous systems of knowledge on equal intellectual terms' -this aim has not been achieved because 'the voice we hear is not indigenous, but an alien verbalization' (Ramos, 2012: 490). This claim seems to me to show little understanding of the magnitude and complexity of the problem with which we are dealing. If one were to take it seriously, one would have to assume that indigenous voices are not already indebted to their imputed intellectual superiors, hence have not lost the game before it has even begun: that they can recognize and find relevant and meaningful the issue of intellectual and more broadly cultural inequality without prior schooling; that they will be heard, understood and taken seriously in a language other than the dominant academic language; that they will express ideas that are not simply the dominant ideas of the dominant party but somehow universal and belonging to them as much as to anyone else. Ramos, as if to confirm that none of these conditions applies, and hence that her argument cannot be taken seriously, gives as an example of an indigenous voice that can speak authentically namely the 'auto-ethnography' of a Baniwa Indian 'who recently received his doctoral degree in anthropology' (ibid.).

'The politics of ontology'
Let us turn, then, to the politics of ontological anthropology and look at it in more detail. The first thing to say is that there are no surprises here. As Candea (2014) rightly points out, for what comes across as a highly unconventional paradigm -one that posits the existence of different worlds or different truths about the world and the need for a 'permanent' conceptual 'revolution' to understand them (Holbraad, Pedersen and Viveiros de Castro, 2014) -its politics is highly conventional. What is more, this politics is not subject to any sort of re-conceptualization, let alone revolution. It is highly conventional and is bound to remain so.
'The politics of ontology' (Holbraad, Pedersen and Viveiros de Castro, 2014) is not the politics of the other, another kind of politics -let us say for the sake of argument, the politics of 'hierarchy' (Dumont, 1980). It is our very own politics of equality and freedom, which presumes that this state of plenitude can be achieved by means of thinking. This is the politics of the Enlightenment that Kant captured so well with the motto of the Enlightenment: think for yourself. Think for yourself and take nothing for granted (as a grant or gift) if you wish to be a self-determining, autonomous being. Hence the battles over minds, ways of thinking and definitions of reality that are characteristic of this politics and the theories that seek to explain how minds are captured and how they can be set free -ideology, hegemony, or, more recently, Bourdieu's theory of symbolic violence. As Hage (2012: 303) rightly points out, Viveiros de Castro's theory of multiple realities is quite similar to Bourdieu's theory, at least insofar as in both cases 'certain realities come to dominate over others, so much so that they become simply ''reality'''. This is why the 'mission' of ontological anthropology 'is to promote the ''permanent decolonization of thought''', and why the discipline is envisaged as '''the science of the ontological self-determination of the world's peoples''' (Holbraad, Pedersen and Viveiros de Castro, 2014: [online]; Viveiros de Castro, 2014). It is because our definition of reality dominates over the definitions of the rest of the world to such an extent that it becomes 'reality' as such. Hence the need to have the minds of the world's peoples decolonized and their thinking set free. If it is set free, they can begin to think and decide for themselves the nature of reality: they can become 'ontologically self-determining'. If they become ontologically self-determining, they become autonomous like us. If they become like us, their otherness is redeemed, as it is now shown to be the same as the sameness of the self. The ethnological problem is solved and anthropology's job is done.
The 'conventionalism' of the 'ontological turn' should not be at all surprising. It is what one would expect from an anthropological paradigm and a discipline whose rationale and reason for existence are the redemption of otherness. Nor, as we have said, can these ideas become 'anthropology's business' if by that we understand what the protagonists of ontological anthropology mean: the 'business to generate alternative vantages from which established forms of thinking are put under relentless pressure by alterity itself, and perhaps change' (Holbraad, Pedersen and Viveiros de Castro, 2014: [online]). Perhaps some of those assumptions and ideas of ours that render things in native life irrational or absurd are put under pressure to change. But this process has a limit and this is where it is reached: when it comes to our ideas about freedom and equality. For these ideas are the very rationale for whatever reconceptualization is supposed to take place, and are not, themselves, subject to it. Without these ideas, there would be no ethnological problem, and hence no need for ethnology (whether ontological or of any other kind).
What remains to be explored is how ontological anthropology fails to solve the ethnological problem. Let us note, to begin with, that the 'ontological self-determination' of the world's peoples and the 'permanent decolonization of their thought' is a giving, therefore, also a taking -not only by those who are at the receiving end but also by those who are doing the giving. It is a giving, not of something material, of course, but something symbolic -the gift of 'ontology', the 'ontological' as a gift, which is to say, the empowerment of the world's peoples to think and decide for themselves the nature of reality. 'The idea of an ontological self-determination of peoples should not be confused with supporting ethnic essentialization [and] Bult und Boden primordialism . . . It means giving the ontological back to ''the people'', not the people back to ''the ontological''' (Holbraad, Pedersen and Viveiros de Castro, 2014: [online]; emphasis added). It should not be confused, because this sort of essentialization divides people, which is hardly what ontological anthropology is striving to do. Yet even when this self-determination is not confused with something else, division is unavoidable. It has always already happened and there is nothing anyone can do to stop it from happening. In empowering the 'world's peoples', ontological anthropology is also disempowering them. It gives but it also takes, and it takes more than it gives.
Having been given 'the ontological', the world's peoples are now empowered to think for themselves. Yet even if they begin to think for themselves, they would still not be thinking for themselves. The paradox, not readily apparent perhaps but massively present, is that to think for themselves, they must not think for themselves. They must take the motto of the Enlightenment for granted as an ontological grant or gift, which is to say, they must accept it without questioning it -otherwise, it would mean that they were already enlightened and not in need of this gift. If they take it for granted, they are by definition not thinking for themselves. If they do not take it for granted and refuse to accept it, they are also by definition not thinking for themselves. Any way one looks at it, there can be no such thing as thinking for oneself. As I have argued elsewhere (Argyrou, 2013), Kant, in promoting the motto of the Enlightenment, clearly wished to liberate people from their 'guardians'. Yet he could do so only by becoming their guardian himself. To put it in terms of the present discussion, he maintained an 'epistemological' advantage over them. He knew what they did not know and had to be taught -that the only way to be autonomous is to take nothing for granted. This is also the case with ontological anthropologists and how they end up being ethnocentric. Unwillingly and unwittingly, no doubt, they become the guardians of the world's peoples. They seek to liberate them in the only way possible -by teaching them what they do not know, hence making them dependent. The equality between anthropologist and native achieved through the decolonization of native thought is instantly transformed into an asymmetrical relation; the epistemological advantage of the former over the latter is maintained.
There is another way of looking at the workings of this sort of inadvertent ethnocentrism. As a number of critics have pointed out, quite rightly so, ontological anthropology operates with a meta-ontology -the definition of a reality in which there are different realities, a world that consists of many worlds (Laidlaw, 2012;Heywood, 2012;and Laidlaw and Heywood, 2013). As Heywood (2012: 146) puts it, ontological anthropologists 'neglect to acknowledge that insisting on the ''reality'' of multiple worlds commits you to a meta-ontology in which such worlds exist'. For Heywood, this is problematic because it is inconsistent with the ontological anthropologists' empirical claims -that it is things themselves rather than theories that point to the existence of multiple ontologies. But there is another problem, here, which is far more important: the way in which this meta-ontology impacts on the relationship between anthropologist and native. Put simply, it reproduces, albeit at another level, the very asymmetry between native and anthropological discourses that ontological anthropology strives to eliminate.
As we have seen, anthropologists, in their effort to show that natives are people like us, must do something about all those things in native life that contradict what we know to be true. They must reinterpret them and argue that in 'reality' there is no contradiction, or that contradiction exists only because natives are not aware of the 'real' meaning and purpose of their beliefs and practices. Positing lack of this sort of awareness, howeverlack of reflexivity, not thinking for oneself but taking for granted -makes natives inferior to us and defeats the object of the exercise. With ontological anthropology, this move is reversed. If there are things in native life that appear to contradict what we know to be true, this is only because we do not know the real meaning and purpose of these things. Our categories of understanding are not up to the task of making sense of other lives and other worlds. Once again, the contradictions emerge as more apparent than real, but in this case the integrity of native discourse is preserved. We must assume that natives know the true meaning and purpose of what they say and do, much like we assume that we know the true meaning and purpose of what we say and do. The equality with regard to our respective claims to knowledge that was only de facto now becomes at long last also de jure. But does it?
Despite the ontological efforts, the anthropologist's and the native's claims are still not equal, whether de facto or de jure. The anthropologist still has an epistemological advantage over the native and can claim superior knowledge. And this is true irrespective of how one reads the ontological argument -as a claim about different worlds or different truths about the world. The claims may be equal when it comes to the knowledge each has of his or her respective worlds/truths, but if what ontological anthropology says is anything to go by, this knowledge does not exhaust reality/truth and what can be known about it. The anthropologist's world/truth and the native's world/truth do not exist in isolation. If the former can encounter the latter and is able to study him, it must be because they share some sort of common ground, a world/truth or universe that contains their respective worlds/truths. Yet although common to both, it is only the (ontological) anthropologist who knows anything about this reality/truth, even if the only thing she may know is simply that it exists. This is the knowledge she claims to have when she speaks of multinaturalism/respective truths, the meta-ontology she neglects to acknowledge but is nonetheless 'committed' to, as Heywood says.
A concrete example would perhaps make this asymmetry clearer. In a recent paper, Holbraad uses the example of Maori gifts to argue against reading Viveiros de Castro's 'perspectivism' as positing different worlds. The usual way of relating the encounter between the Maori and the anthropologist, Holbraad says, is to say that while the former sees a spirit in the gift (that of the owner), the latter sees merely an object. The perspectivist account, by contrast, argues that the difference is not between two ways of 'seeing' the gift but of conceiving what the gift is. The analogy with Amerindian perspectivism, Holbraad (2016: [no pagination]) goes on to say, is 'strict'. The 'anthropologist and native ''see in the same way'' inasmuch as they both ''see'' gift'. But for the anthropologist, the gift is a mere object, while for the Maori it is an object imbued with spirit. In effect, the gift is two different things, and this is what it means to say that the difference is ontological.
Whatever we say about this reading of 'perspectivism', one thing is certain: it does not solve the problem we are dealing with. The (ontological) anthropologist still has an epistemological advantage over the native. It should be clear that the (ontological) anthropologist has a perspective on this encounter that is neither the perspective of the Maori nor of the (non-ontological) anthropologist. It is a perspective that contains botha meta-perspective -and makes possible a truth that contains the truth both of the Maori and of the (non-ontological) anthropologist. In short, the ontological anthropologist knows what the other two actors do not know: that there are different truths as to what gift is or that the gift can be different things and that, presumably, neither the Maori nor the (non-ontological) anthropologist should dismiss each other's truth.

Conclusion: The 'hauntological'
I have said above that in giving the ontological as a gift to the world's peoples, anthropologists also take. They take from the world's peoples themselves, no doubt -symbolic things like recognition, admiration, gratitude and so on -but not only from them. By way of concluding this article, I would like to pursue the idea of taking by means of giving a bit further because it will help us clarify the 'hauntological' a little more. This, as we have said, is the claim that anthropologists are possessed by, and obsessed with, the idea of human unity, so much so that they will not rest until they demonstrate that it is an empirical reality, not simply a figment of the imagination.
The first thing to note here is that anthropologists take something in return long before they give the ontological as a gift. They take as soon as the thought crosses their minds that the ontological could be given to the world's peoples as a gift. At that very instant they take, and they take far more than what they will eventually give. This is to say, as Derrida (1994b) does in his discussion of Mauss's classic The Gift, that giving is taking with a certain capitalization; in effect, that there is no such thing as a gift. What do anthropologists take, then, by means of this giving? I will be thoroughly anthropological here and turn to a well-known essay on religion by Clifford Geertz (1973b). Religion, says Geertz, is a cultural system that maintains a vision (or the illusion) of conceptual, ethical and emotional order in the world. It does so by explaining the exception and the anomaly, and thus by bolstering the belief that we can understand whatever arises, make sound moral judgements in every circumstance, deal effectively with pain and suffering -in short, that we are in control. Can this help us at all to make sense of the 'ontological turn' and more broadly of what has kept anthropology turning for more than a century, now? I think so. The need to believe in the existence of an ethical order seems to me to be the stake in giving the world's peoples the 'gift' of recognition. If we wish to 'do justice' to their way of life, it is because we need to believe that there is justice in the world, that profanities like ethnocentrism exist not because they are inherent in reality, not because there is no moral order and anything goes, but because of regrettable but happily rectifiable things like misunderstanding, ignorance, arrogance, greed and so on. What we take through this giving is the assurance that our world is morally grounded and that we are safe.
Perhaps not many anthropologists will be prepared to accept this reading of what we do and why we do it. If there is any truth in it, however -and I think there is -it goes some way in explaining the 'haunting', the fact that anthropologists are possessed by, and obsessed with, human unity, the plenitude of human purity and innocence. We must persist in our efforts to demonstrate this unity, despite all past failures. What is at stake is nothing less than the moral order of the world. This is to say that what is haunting us is the suspicion 'that perhaps the world . . . has no genuine order at all . . . that life is absurd and the attempt to make moral, intellectual, or emotional sense out of experience is bootless' (Geertz, 1973b: 108). We may take comfort in the knowledge that we are not alone in this. This spectre is haunting others as well.

Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.