There was a time in our discipline of anthropology when we thought we knew how to talk about subjects and objects, people and things, and the use the former made of the latter, materially or imaginatively. This may no longer be so. As anthropology has come to undo its own intellectual legacies—Christian, Cartesian, Hegelian, and what have you—“matter” has come to matter in largely unanticipated ways. This clearly is not simply a default development arising out of the belated realization that “subjectivity” is not all that it was once vaunted to be, as Foucault and others have taught us. Nor does it merely reflect a newfound interest in the “non-” or “post-human” (e.g., Cerullo Reference Cerullo2009; Kirksey and Helmreich Reference Kirksey and Helmreich2010; Pyyhtinen and Tamminen Reference Pyyhtinen and Tamminen2011). It is an issue that cuts to the core of a discipline that has long staked its claims on its capacity to illuminate worlds in which “apparently irrational beliefs” blur the conceptual bright lines that modern Western ontological predications tend to draw, if not always consistently, between subjects and objects, agents and patients, people and things.

The problem itself is by no means new. Largely inspired by E. E. Evans-Pritchard's Witchcraft, Oracles and Magic among the Azande (Reference Evans-Pritchard1937), the so-called “rationality debate” of the 1960s and 1970s (Wilson Reference Wilson1977; Hollis and Lukes Reference Hollis and Lukes1982) centered on whether worlds whose furniture, for example, included agentive forces such as mangu (causing misfortune, in that case), could be understood to possess a rationality of their own. It was Evans-Pritchard's merit to have argued that, even though based on erroneous premises, such worlds could be shown to have their internal logical consistency and, yes, rationality.Footnote ¹ Rather than coming to a genuine conclusion, the debate simply fizzled out in the 1980s, arguably under the impact of the so-called “crisis of representation” in anthropology. Still, it had generated surrogate solutions in, for example, symbolic interpretations (statements such as “twins are birds” or “cucumbers are oxen” are true once subjected to the right kinds of hermeneutics), or in a resurgent universalism grounded, by and large, in gestures of ethnographic charity executed with varying degrees of good faith. We can take the impact of cognitive science on anthropology as exemplary here.

How interesting, then, that this debate has recently been retrieved from the dustbin of disciplinary history and repurposed in various manifestos that appear to invert its very terms. As Henare, Holbraad, and Wastell Reference Henare, Holbraad, Wastell, Henare, Holbraad and Wastell2007 (see also Holbraad Reference Holbraad2009; Reference Holbraad2012; Holbraad and Pedersen Reference Holbraad and Pedersen2017; Viveiros de Castro Reference Viveiros De Castro2012) have argued in a deliberate reversal of the premises of the original debate, it is “us” rather than “them” who got things wrong, and not because (as went the original argument championed by Gellner, Lukes, Hollis, Horton, or even Evans-Pritchard himself) “we” proceed from premises that are supported by scientific truth claims, and “they” do not. Instead, or so the argument for an “ontological turn” goes, it is we who have proceeded from arrogantly universalized, “provincial” conceptual premises (in the sense of Chakrabarty Reference Chakrabarty2000; see Trouillot Reference Trouillot2003). As proponents of such a “turn” tend to argue, these prior assumptions precluded us from comprehending worlds structured by forms of “radical alterity” resistant to our own ontological presuppositions. The result was that ethnography tended to devolve into a machine productive of just the kinds of conceptual paradoxa and aporia which the “rationality debate” had aimed to solve, but really did not (see Scholte Reference Scholte1984 for an incisive, but very different critique). The result, or so I understand the argument for an “ontological turn,” is that now the anthropologist's onus is no longer to interpretatively account for statements such as “stones are persons,” but instead to see what transformations taking them at face value might wreak upon our own conceptual apparatuses and theoretical languages. The goal, then, would no longer be the Malinowskian ideal of grasping the “native's point of view,” as Holbraad and Pedersen (Reference Holbraad and Pedersen2017: 7) have recently put it, but instead to be “grasped by it”: to take incommensurability and the resulting impression of radical alterity not as a limit, but as a challenge, an invitation to renew the mission of our discipline by turning its relativist inclinations onto itself.

To my thinking, this may all be well and good. But might there not be an alternative to this peculiar revival of the very concept of “radical alterity,” harking back as it appears to do (or so one might argue) to the theoretical apparatus associated with Lévy-Bruhl that Evans-Pritchard, however respectfully, once set out to attack? Mindful of the fact that contemporary calls for “ontological turns” are anything but internally coherent,Footnote ² I take courage from Philippe Descola's recent elaborations (Reference Descola2013; Reference Descola2014) on what he calls “ontological hybridities”—the locally and historically specifiable coexistence and partial overlap of divergently composed systems of inferences about the world and its furniture. I have no investment in Descola's by now well-known, and certainly debatable terminology concerning four ideal types of such largely pre-discursive modes of “ontological predication” (animist, totemic, naturalist, and analogist) and will not refer to it in what follows. But what I would say is that the following “thought experiment”—in Holbraad's (Reference Holbraad2009) sense—concerning an American anthropologist's perplexities in mid-twentieth century Cuba might throw into relief what happens when highly divergent “modes of ontological predication”Footnote ³ about the nature of matter, personhood, animacy, agency, and sociality coincide in an ethnographic space. What is more, in this case they coincided in a setting that was neither fully “modern” (as Latour [1993] might have it) nor “traditional” in any self-evidently defensible sense.

Let me begin with a bit of historical background and some ethnographica. In the summer of 1948, William R. Bascom was dispatched by his mentor Melville Herskovits to test the strength of African survivals in Cuba. It was an obvious choice on the part of Herskovits: Bascom had done doctoral work in British Yoruba-land in 1937–1938 and was married to a Cuban woman, Berta Montero, who could be counted on to help Bascom ease his way into the field. After a brief time in Havana and some crisscrossing of the island in search of a field site, the Bascoms settled in the small town of Jovellanos in the province of Matanzas for just under four months (an entirely acceptable period by Herskovits's own ethnographic standards).Footnote ⁴ Jovellanos was an interesting choice, since contrary to Herskovits's emphasis on rural locations, presumably as sites of “traditionality,” it was a relatively new settlement that had grown out of Cuba's belated nineteenth-century sugar boom (it was only incorporated as a municipality in 1866). Moreover, its economic mainstay was not just agro-industrial sugar production, but also the manufacture of industrial equipment. As Bascom himself noted, by the time of his fieldwork a toothpaste and cosmetics factory (Laboratórios Gravi, S.A.) was not only one of the largest employers in Jovellanos’ urban core; it was also—as he was repeatedly told (e.g., on 13 and 16 August 1948)—a site of rampant mutual witchcraft suspicions among its employees. Jovellanos also already had Baptist and Methodist churches, both presumably American-sponsored. It was, in sum, a rather peculiar place to conduct fieldwork on Cuba's “African traditions.”

Bascom's mission was to parse out what was Catholic and what African about the religious practices he would find there. Since Herskovits had argued that “religion” was a “focus” of “African cultures,” and so more resistant to “acculturative pressure” than other domains of social life, Bascom sort of knew what he wanted to find. Things did not go as easily as he had expected, not least because the locals had been aware of anthropological interests in their practices for quite some time (García Reference García2014). That is a longer story, which I will tell some other time in appropriate detail. Suffice it to say for now that Bascom found what he had come to “discover” (Palmié Reference Palmié2013): while everyone was baptized in church, and Catholic statuary and chromolithographs of saints abounded, the true “focus of Cuban santeria” consisted in the ritual use of stones, herbs, and blood, or so Bascom argued two years later (Reference Bascom1950). The herbs and blood seemed to present no problem for someone who had done fieldwork in Nigeria, but the stones did. Not only was Bascom not aware of any Yoruba antecedent (or parallel, as I would argue), but the way these stones were described to him seemed unusual:

Bascom noted that this “miraculous power is given to the stones by treating them with the two other essentials of santeria, herbs and blood,” and that stones that have not been treated this way—which his informants referred to as a baptism—were said to be “jewish”; that is, completely powerless. But he had little more to add, except to hypothesize that the ritual importance of stones had either been overlooked among the Yoruba,Footnote ⁵ or that “the focal elements of Cuban santeria may not represent a carry-over of the focus of West African religion, but a shift in emphasis which has occurred as a result of culture contact” (ibid.: 68). Since all “informants, without exception, stated unqualifiedly that they were Catholics, yet … stressed the importance of those very elements of their faith and ritual which set it apart from that of the Catholic Church,” this, Bascom surmised in conclusion, “would seem to be another illustration of Herskovits’ concept of ambivalence in New World Negro cultures” (ibid.).Footnote ⁶

Clearly, Bascom did not quite know what to make of such walking, growing, progeny-producing, and otherwise “lively” stones, especially since his own experience in Yorubaland provided no ready-made “origin” for conceptualizing the animacy, and even sociality, of mineral objects. But he was right. Some stones were certainly very much alive in pre-revolutionary Cuba, and so they are today in Cuba and its religious diaspora. They live, eat, work, propagate, are born, and even die.

Fortunately, Bascom's fieldnotes from Jovellanos are more instructive than what he chose to publish. As a woman named Florencia Baró told him in August 1948, the stones active in mid-twentieth-century Jovellanos had a history:

Perhaps as little as a century had transpired between Casimiro/Adekpęlę’s and Bascom's advents on the island.Footnote ⁸ As to the continued liveliness, willfulness, and even recalcitrance of these stones, a woman named Benita told Bascom on 21 August 1948:

As Benita and others told Bascom, such dreams were communications from the gods, adding oneiric data to what would otherwise, or subsequently, be confirmed by divination. Sometimes, for instance upon the death of their previous owners, Benita said, “you ask the gods with dilogun [cowrie divination] whether the stones are to be taken, put out in the field, or given to someone else to carry on and who it should be. When the stones are told to be thrown away, the gods already know who will come along and find them and take care of them” (ibid.). In other words, in terms of her own predicament, Benita had no choice in the matter. It was the stone who found her, and came back, not the other ways around: not a case of an objet trouvé, but a femme trouvée.

More puzzling still, as one of Bascom's most theoretically minded informants, Pedro Peñalver, known as Ito, said on 23 August, “The stone is not just a stone, it is an alive thing. The first thing that God fed Christ was a stone; the bible says so. San Juan was the godfather of Jesus, (Dios he said), Because he baptized him” (ibid.: 199). Ito continued with a rather peculiar theory of presence and representation that, if Bascom understood him correctly, may help explain why god would have fed a stone to Jesus: “The stone is not the same as the fleche [sic] of Oshosi [a bow and arrow representing this hunter deity]; in this sense it is not the symbol. The stone is like our body in that we can talk and see and act, and the stone has all the power of Yemaya and other saints. It is not like the home of the saint, but more like its body. More like the body than like the image, home, or symbol” (ibid.).

I will return to Bascom's notes and Ito's theory, but let me first adduce some more contemporary evidence. Writing some forty years after Bascom, David Brown (Reference Brown1989: 253) noted a remarkable practice among the members of a New Jersey casa de santo or ilé ocha (cult group) named Templo Bonifacio Valdés. Again, it involved stones. Cutting Brown's ethnographically rich story short, what many Anglophone santeros nowadays refer to as the “birthing process”—a gloss on the logic of the initiatory process whereby a new initiate and the objects of his or her orichas “come into the world” (nascen) simultaneously—was anything but metaphorical for Brown's informants in the 1980s. Hence, “If for any reason the approaching asiento [initiation ceremony] is cancelled during the ‘gestation’ period, after the orichas have been ‘conceived’ but before they are born in the asiento [initiation ceremony], they cannot come to term. An itutu [funerary rite] must be performed for them, the funerary ceremony in which, in ordinary cases, the orichas of the deceased priest are [to] be sent from this world, or inherited by another. Here, however, the itutu carries connotations not of a funeral but of an abortion” (ibid.: 253–54).

To be sure, this is highly idiosyncratic. No one seems to have documented a comparable praxis anywhere else. Practitioners of regla de ocha will readily speak of the birth of objects, or even situations, in signs of the Ifá-oracle that speak to, or rather document, the coming into the world, or birth, of matters such as, say, fraternal strife or adultery. Yet a failed coming-into-being demanding funeral rites for “unborn” orichas seems unique. What should interest us here, though, is the logic. What Brown's informants called orichas in gestation were ultimately a set of objects, including the kinds of stones that so puzzled Bascom.

In my first book (Reference Palmié1991: 241–340), I devoted considerable space to what practitioners call the fundamentos of their orichas, a term only poorly glossed as “materializations,” or perhaps better “groundings” of their orichas. These are usually housed in porcelain soup tureens known as soperas, and invariably consist of stones (otanes) and a variety of other objects called herramientas (i.e., tools).Footnote ⁹ It is there, in such fundamentos, that the orichas “eat”—receive sacrifices—and become agentively manifest in two other ways: First, they “work” (trabajan) in the natural, social, or bodily domains and qualia over which an oricha rules, such as when, for example, Ochun's power becomes evident in fresh water, sweetness, the circulation of money, sexual pleasure, the color yellow, and so forth. Second, they manifest through their sporadic indwelling in another vessel prepared during the initiatory process in a fashion paralleling the ritual manufacture of the fundamento. That vessel is the head of the oricha’s initiate, which during possession will be similarly filled with the deity's presence. Dramatically abbreviating matters, such fundamentos, the otanes contained in them, and the possessed practitioner's body may be said to constitute material tokens of an immaterial type: the oricha.Footnote ¹⁰

Another story about such stones comes from my own fieldwork in Cuba in the first decade of this century: the stones in question are obviously not just any old stones. On the contrary, to effectively be inhabited by the oricha—be the oricha—the stones must reveal themselves as such through divinatory procedures. They need to tell you who they are.

Some, for instance, have to be found in rivers, and once they are, they will be asked by oracular means if they are what everyone hopes they are. On one of the days of the three-day-long initiation ceremony, the neophyte (iyawo) is sent on a mission to find stones that “are” the oricha he or she is going to be initiated for. The problem is that the Almendares and the other rivulets emptying into the Bay of Havana are so silted up that no stones could possibly be found in them. So, I was told, “we go to the river the night before, and we carefully drop the stones, so they can speak to the iyawo the next day.” At least in theory, this does not preclude the possibility that the stones may refuse to speak, thereby remaining mere stones rather than stones that might come to act in the ways Bascom described. In Alfred Gell's (Reference Gell1998) terms, plenty of divinatory abduction needs to precede the confirmation of any old stone as an index of divine presence.

But speaking of indexes: at first glance, this case might qualify for treatment as an ethnographic index of the failure of (Western) epistemology to come to terms with the radical alterity that—or so Henare, Holbraad and Wastell (Reference Henare, Holbraad, Wastell, Henare, Holbraad and Wastell2007) argue—ought to override the old, relativist, knee-jerk reaction (one world, many representations), and force us to face up to the existence of ontologies (worlds, not world views!) in which “powder is power” (Holbraad Reference Holbraad, Henare, Holbraad and Wastell2007; Reference Holbraad2012) or stones can be alive. But is this really such a case? I sympathize with the congeries of projects that are nowadays rather self-consciously subsumed under the moniker “the ontological turn,” but only to a degree. I will use the rest of this essay not so much to critique the presuppositions and potential consequences of such a “turn” (there already is a plethora of such critiques), but rather to suggest an alternative scenario for confronting what current pluri-ontological speculations have made into latter day pseudo-ethnographic surrogates for the subject matters that drove the so-called “multiple realities,” “other minds,” and “rationality” debates among philosophers from the 1940s into the 1970s (Schuetz Reference Schuetz1945; Wisdom, Austin, and Ayer Reference Wisdom, Austen and Ayer1946; Wilson Reference Wilson1977; Hollis and Lukes Reference Hollis and Lukes1982).

Bascom's fieldnotes are as fine an entry point into such a discussion as there could be. This is so because they allow us to not only conceive Afro-Cuban religious worlds as far from self-enclosed “ontologies unto their own,” but also expose anthropology's complicity in making them so, even if in the service of radical critiques of Western “humanism” and “representationalism.”Footnote ¹¹ Here I once more find myself in sympathy with Descola's (Reference Descola2014) argument that anthropology is, or at least has long been, part of a set of instituted practices growing out of a “naturalistic” (hence also humanistic, if not anthropocentric) mode of apprehending and generating knowledge about the world that gained ground in “the West” from the seventeenth century onward. I can think of no compelling argument against Descola's position. If he is right, then what I have called “the ethnographic interface” (Palmié Reference Palmié2013) is also the interface between potentially different ways of composing worlds, rather than a barrier of incomprehension at the borders of incommensurable ontologies.Footnote ¹²

To elucidate this point let me turn to a different body of thought—Actor Network Theory—if only to see what cargo it delivers in the case at hand. I should emphasize here that I do not take recourse to ANT to solve the riddle of Bascom's animated stones, but merely to experiment with an approach that I find illuminating. I should add another caveat: contrary to Pels (Reference Pels2015), I fail to see how Latour, Callon, Law, Mol, and others could have been mistaken in extending the Edinburgh School's “strong programme” into a methodology (mind you: not an epistemology) of “generalized symmetry”: a building up from the ground of how, for example, anthrax spores, farms and state bureaucrats, scallops, scientists and fishermen, and for that matter American anthropologists, Cuban santeros, chromolithographs, stones, blood, and herbs are recruited into more or less stable assemblages that, if successful, produce palpable effects in the world.Footnote ¹³ Or how, in another inflection, different sets of clinicians carve up, and reconnect, the human body in specific ways, with different, but equally inescapably real effects. In Mol's (Reference Mol2002) terms, the question is not what the world is like—really and truly—but how it is done, how practices intersect (or not), and how the resulting ontologies are stabilized (or fade away).Footnote ¹⁴ That, I think, is entirely appropriate to the empirical case I am dealing with here, the process of hacer santo: to make the oricha.

Many agents and actants must be mobilized in order for stones thrown into a river to become abodes for indwelling deities, and at the same time, much has to be done to turn the heads of humans into equivalent receptacles. A cast of ritual actors and witnesses has to be mobilized; financial resources and esoteric competences pooled; sacred potions must be prepared to wash both the otanes and the novice's body; ancestors of a line of ritual descent need to be invoked; libations poured; animals sacrificed; chickens plucked; their intestines roasted; heads shaved and painted; incantations made—all to render the novice's body and the fundamentos he or she will receive functionally equivalent or even partially consubstantial. The success or failure of each and every step in the procedure is determined by divination. Depending on the tradition of the casa de santo or ilé ocha (cult group) in question, the outcome is calibrated and ratified by either the obí, diloggun, or ifá oracle. Failure can occur at any step. Remedies can be sought, and compromises made, albeit within limits. The procedures are at all points highly methodical, and subject to their own modes of proof and falsification.Footnote ¹⁵

What this process brings into being is a set of linkages or nodes that extend the network that we might call “santería” or “regla de ocha” in both social space and historical time. It does so, firstly, by placing the initiate into a genealogy of ritual descent (rama) that ideally reaches back to a relatively small set of fundamentos—that is, African or first-generation creole initiators active before or around the turn of the twentieth century, such as Adekpęlę, Florencia Baró’s enslaved bringer of lively, procreative stones (note the use of the term fundamento). The process does so secondly by reproducing the casa de santo or ilé ocha as a social formation, which nowadays can span several continents and has moved to the Internet.Footnote ¹⁶ Thirdly, it potentially enables initiates to become vectors of its further extension, should they participate in or direct initiations themselves. Finally, the process prolongs the deities’ own leases on life.

It is in regards to the latter set of nodes that my ANT thought experiment might appear at first glance to begin to grow tenuous. But to view matters in this way is to remain beholden to what Fenella Cannell (Reference Cannell2005) rightly called the “Christianity of anthropology”: a suitably secularized vision of instances of “the religious” that nonetheless carries the weighty intellectual burden of distinctly Christian ideas of immanence and transcendence, body and spirit, the phenomenal and the numinous. This is, of course, just about the worst perspective from which to approach Afro-Cuban ritual formations since the oricha, though immensely powerful, are decidedly not transcendent. On the contrary, they are as dependent on material “immanence” (and mediation) as you and I when we go about our business of preparing dinner, mailing a letter, or writing an essay.

Mythology tells us why this is so: after the oricha decided to stop living among humans (a long story, that one), there came a time when humans stopped paying attention to them. The consequences were disastrous. Rain stopped falling, fields bore no fruit, nor women children, men quarreled, nothing ever got done, people died in droves, and everything was very bad. But the gods went hungry, too. Finally, they decided to send one of their number, named Elegguá, down to earth to teach humans how to ascertain the orichas’ needs by divination. Ever since then, humans have known how to respond to the orichas’ wishes, which they express by visiting calamities upon us: people fall ill, have accidents, run into trouble with the law, or with their spouses, fail to get exit visas to the United States, or are denounced for black market activities by their neighbors. All kinds of rational remedies are sought, but to no avail. It is only when a visit to your friendly neighborhood babalao discloses, by divinatory means, that the gods want you to become initiated into their cult that effective steps can be taken to alleviate the situation. This is so because—as Marcio Goldman (Reference Goldman2009) argues in the largely analogous Brazilian case—the gods had already claimed you and the stones that will come to fill your soperas all along (you simply did not know it),Footnote ¹⁷ and so you, and these stones, become what you were always meant to be: vessels and vectors of the oricha.

If I may mention the name of the arch-relativist/representationalist in this context, practitioners of regla de ocha are Durkheimians in the strong sense. They know, in other words, that they are dependent (or co-dependent, as contemporary American psycho-babble would have it) on what Durkheim might have called the collective representations that rule their world and, consequently, decide their destinies, whether they like it or not. Indeed, the whole ritual tradition of regla de ocha, and hence the oricha themselves, only exist because humans and their gods are joined at the hip.Footnote ¹⁸ One can do without the other only at their peril. Diana Espírito Santo (Reference Espírito Santo, Santo and Tassi2013: 40) puts it well when she argues, “Cuban spirits expect to be ‘made.’” By this she means “not just that a person's relation to a spiritual entity must be discerned and achieved—temporally, material, phenomenologically—but also that this relation is in some way constitutive of the entity itself, which paradoxically both preexists its relationship to people and depends on it for continuous efficacy, presence, and importantly, substance” (ibid.; her emphasis).

The logical corollary to this is that unless “made and remade” over and again—materialized, if you will, but better: extended in time and social space—the gods themselves may lose their grip on the world. While many fundamentos are inherited, people know that some oricha nowadays are merely vague memories: objects embodying them may still exist, but since elderly priests took their knowledge of how to interact with them to their graves, no one even knows how to properly feed them, let alone initiate anyone into their cult. They are former nodes in the network that have failed to be activated and extended. As a result, their influence is no longer felt, nor diagnosed in divination (at least not with prescriptions for appropriate remedies)—which pretty much amounts to the same thing. They have lost their anchorage in human praxis, and so their claim on materiality and worldly efficacy as well.Footnote ¹⁹ They are dead gods.

This brings me to the final analytical point that I want to make in this essay, and I preface it by saying that, unlike colleagues who have become santeros, and santeros who have become colleagues, I have no particular stakes in the matter (Palmié Reference Palmié2013). In fact, I cannot know what they may know about what it is to live—and here is the point: to share one's body—with the awesome but needy entities that may well rule our world as long as we play along. I myself do not, but that does not matter in the larger scheme of things. This last point directly speaks to the title of my essay: when is a thing a human, a deity, a stone, or what else? I am sure such a question never crossed Bascom's mind. For him, people believed that stones walked, grew, and had children, and what counted was whether he could find an African precedent for such beliefs. What Bascom worried about was the extent to which the circulation of chromolithographs of Catholic Saints had come to induce a “confusion of theological concept” such as his mentor Herskovits had diagnosed in his Haitian research (Reference Herskovits1937a), after reading Arthur Ramos' Reference Ramos1934 take on “syncretism” in Brazil. To his credit, Bascom satisfied himself that the same was not the case in provincial Matanzas. None of his informants would have dreamt of sacrificing a chicken or goat to a statue of St. Barbara; it would have been Changó’s stones over which the blood was poured.

And how could he have thought otherwise? Every one of his informants seemed to have some opinion or other regarding an iconic or symbolic (in Peircian terms) relation between orichas and saints. Some of these opinions were probably elicited by Bascom, since his informants appear to have found the questions themselves counterintuitive. In any case, they all insisted that the images, whether pictures or statues, were mere adornos to their shrines, had no animacy or efficacy in themselves, had been adopted only fairly recently, and indexed nothing. The statement on this that is both most sophisticated and most puzzling comes, once more, from “Ito” Peñalver, who told Bascom on 23 August:

Ito seemed to contradict himself (his quote on p. 791, above, comes from the same interview) when he said that the “stone was the saint's body,” suddenly appearing to veer toward a theory of presence and representation that, initially, appears oddly Protestant (Keane Reference Keane2007; Engelke Reference Engelke2007),Footnote ²⁰ or from a different perspective, even outright Derridean. Bascom, unfortunately, let the matter go, and there is no telling if Ito might simply have cooked up this theory in a desperate attempt to give Bascom an analogy that the Yanquí ethnographer, obsessed as he seemed to be with Catholic saints,Footnote ²¹ might finally understand. Surely, a photograph is only an index of its subject (say, a deceased mother-in-law) when regimented by a specific semiotic ideology (Keane Reference Keane2003; but see Gunning Reference Gunning2004). Just so, a stone is only a portal to the divine under certain specifications. And these specifications, for Ito, might have included a sense in which individual otanes—just as the heads and bodies of initiates—were merely material tokens (akin to datable and locatable photographic indexes) of a type the reality of which ultimately escapes such predications: the divine “woman” Yemaya.Footnote ²²

Bascom, in turn, seems to have been satisfied that Ito drew any kind of distinction that would affirm what he wanted to hear, viz., the nonidentity between statuary, chromos, and the stones that so fascinated him. And so Bascom broke, however unwittingly and inconsequentially, the stranglehold on the ethnographic imagination that Ramos and Herskovits's theories of syncretism as an essentially “unconscious” by-product of religious culture contact had lastingly imparted to an emergent African Americanist anthropology of religion. Whatever Ito meant to tell Bascom, there was nothing unreflexive about it.

Fair enough, I would say. But not enough. In fact, the whole reason for my little ANT thought experiment is to suggest that what Bascom saw as a form of attenuated cultural transmission and hence transformation of belief (then known as “acculturation”) was, in fact, evidence of and for what I will call an ongoing process of transduction within a relational network, configured in a way that divine input would trigger human output, though always precariously so. I mean transduction here in a simple, straightforward way, such as when mechanical energy is transformed into heat or electricity.Footnote ²³ Central to this is a “mana-like” (or electricity-like) notion subsumed under the term aché. If you have read Holbraad (Reference Holbraad, Henare, Holbraad and Wastell2007; Reference Holbraad2012) on “powder and power,” then you already know what I mean. It is the stuff that makes things happen, and it is a substance that can be accumulated, transferred, dissipated, and so forth. It is, if you like, regla de ocha’s reply to the laws of thermodynamics.

As in the latter case, energy—or aché—manifests in distributed form, in material states of manifestation, if you will, that can become subject to mediated transduction into other states, manifestations, or distributions. This is nothing less than what the initiation ceremonies known as kari ocha, asiento, or hacer santo aim to achieve. Like a relay or switch that is thrown here or there, these ceremonies direct the processes by which aché is variously manifested in media like stones, or bodies possessed by the oricha. To some degree, this is also what happens in public possession ceremonies (tambores or bembés). As Bascom's informant Candito Pérez put it on 14 August 1948, “In a ceremony, the gods are all around the house. When a santo has a caballo Viejo [experienced medium], they know how to do it very well, they pick up the god from the stone and dances [sic] with it, and when it goes away, it goes back to the stone. The sube [from Spanish subir, a gloss for taking possession] is from the stone into the head of the person” (Bascom Reference Bascom1948: 144).

Note here how attributions of agency shift back and forth, as if it were a current oscillating between two poles: it is the deity that has a caballo (not the other way around), but it is the medium who picks up the god from the stone and dances with the oricha in his or her head, until the god returns to the stone. What effects such transductions is a network of multiply articulated mediators: a ceremonial display (trono) of the orichas’ soperas luxuriously draped in cloth; a spread of food offerings (plaza); the sound of batá drums (themselves consecrated containers of a divinity); an order of chants (suyeres) that have to be sung in order to coax the oricha to manifest; the presence of’ subidores (possession mediums); dancing crowds, colors, sweat, tobacco smoke, noise, and heat. Once all agents and actants have been mobilized, all circuits connected, and all conditions met—which is neither a mechanical process nor a foregone conclusion—what results is the visible, audible, and haptic presence of a god.Footnote ²⁴ Nor is this a matter of “belief.” As Paul Johnson puts it, from a phenomenological angle, we are dealing with “the bundling in ritual practice of an interwoven sensory load that is named, when done according to code, ‘a spirit’” (Reference Johnson and Johnson2014: 9). “Ver para creer,” santeros will say, but such “seeing” is conditional on both a sensorium responsive to cues of the workings of aché (what Wirtz [Reference Wirtz and Johnson2014] in a related context calls “perspicuity”) and what Johnson calls “code”; that is, the meeting of conditions under which mundane perceptual signs transduce into indices of the divine.

Santeros speak of meeting such conditions as expending “work”—trabajar or trabajar ocha. A good deal of time, labor, money, and savoir faire goes into staging a successful tambor. Once done correctly, though, ritual not only renews the world, but potentially transforms it by adding to its furniture, refurbishing it, or presenting what was already there in new light. Was this what Bascom's organic intellectual informant Ito Peñalver meant when he told him that the Bible said that Jesus was ritually consecrated (“baptized”) by San Juan and then fed a stone by God? Was this the key to the supreme Catholic mystery of incarnation? Did St. John wash and prepare (“make”) the body of Christ so that it became a human conduit for the aché of “Dios”? Was not San Juan quite literally the padrino (initiator) of Christ?Footnote ²⁵

But whatever else such ceremonies may achieve, or help explain, most of all they establish the connectivity that drives onward the whole system of mutual entanglements, exchanges, and transductions that we might call a closed circuit, or more sociologically, a human-divine commonwealth. Tambores or bembés are history in the making, and as such they are perennially open to “structures of the conjuncture” (Sahlins Reference Sahlins1981) that may allow for replication or transformation, including the transformations that may have introduced a new iconic register, that of Catholic chromolithographs, into the relation between humans and their gods in Jovellanos. Who knows? At one point or another, the human vessels of the orichas Changó or Ogún may have expressed a wish to be iconized as Santa Barbara or San Juan (which would be all but inconceivable from today's ethnographic vantage point). Divination may have confirmed it. What counts is that it is in the scope of such ritual encounters that deities, human bodies, and stones meet on equal ground. They are “symmetricalized” to each other, as it were, and in mutual states of coming under each other's influence. Paraphrasing both Marx and Durkheim, we might say that men and gods make each other, though not under conditions of their own choosing.

No wonder, then, that the stones and the people who have been ritually fused to them are what Lucy Suchman (Reference Suchman2011), in the context of robotics, calls “subject-objects,” neither unqualified humans, nor unqualified things, but terms of a relation. What Bascom did not realize was that his informants and the stones they told him about were part and parcel of just such a relation, one in which people, things, and deities are labile potentialities rather than starting points. Such relationality, we might say, is constituted by a kind of “intra-activity” in the sense of Karen Barad's “agential realism”—a perspective that instead of assuming “that there are separate individual agencies that precede their interaction … recognizes that distinct agencies do not precede, but rather emerge through, their intra-action.” As such, Barad continues, “‘distinct’ agencies are only distinct in a relational, not an absolute, sense, that is, agencies are only distinct in relation to their mutual entanglement; they don't exist as individual elements” (Barad Reference Barad2007: 33, her emphasis).

And so it was in Jovellanos in the summer of 1948. Gods, stones, people, herbs, blood, plaster statues, and chromolithographs were all part of an “intra-active” continuum that had begun to unfold from the moment that Adekpęlę had put his stones to work on Cuban soil, causing them to bring new stones—and newly consecrated heads—into the world, and so giving the oricha a new lease on life on another continent. We might call this a massively “hybridized ontology” in Descola's sense (including the witchcraft accusations flying in the toothpaste factory). Yet I do not really see what is gained by such a description. Guided by some wrongheaded, preconceived questions, Bascom just tried to cut up the continuum in front of his face along a line that was dictated more by his epistemological priorities than by what he heard and saw.Footnote ²⁶ Much like avatars in evolving multiplayer online games, the people and orichas that Bascom encountered in semi-rural Matanzas in 1948 were making and remaking each other. We might say they were “running on the same platform,” and so driving forward in time the serious game of Afro-Cuban religion. Bascom, in turn, was making his social and intellectual moves within a different game. The name of that game was Herskovitsian acculturation research, and while the two games came to intersect for a few months that summer, it may well be just now that we may have at our disposal the media, and the analytics they afford (Peters Reference Peters2015), that allow us to fathom the connection, and multiple disconnections, between them. Minerva's owl, after all, takes wing at dusk. Once “old” media are superseded by “new” ones, we suddenly realize how thoroughly naturalized the former once were.Footnote ²⁷ The same holds for dead theories. On which more below.

Stones and initiated bodies were and are the vehicles and vectors through which tokens of the oricha materially reproduce the abstract type, in the very same way that concrete instances of “parole” come to constitute, reproduce, and potentially transform “la langue.” None can be prior to the others. They form a becoming-together that reaches into the past as it reaches into the future, brimming with contingency across the spectrum of time. This represents, among other things, a solution to the pseudo-problem of “acculturation” that clouded Bascom's mind in that summer of 1948.

This is not the place to hazard more than a guess about the underlying epistemic infrastructure that guided, or misguided, Bascom's research during that summer. Chances are that what made for his ambivalence were the linguistic ideologies underlying the analogy with comparative philology that he had inherited from his teacher Herskovits's teacher Franz Boas. To be sure, Boas's program was effective in combatting the racism of his time by demonstrating the fundamentally historical—and so radically contingent, and physiologically entirely undetermined—nature of human life-ways. But it retained the scientistic objectivism within which Boas or Herskovits simply had to operate, thus unwittingly reifying “cultures” and their “traits” as they sought to install the concept of culture as a politically relativist antidote to racial determinism in the American public sphere (Trouillot Reference Trouillot2003). What they did not, and indeed could not see was that the ontological (never mind “psychological”) weight they attributed to what all of them occasionally acknowledged to be a pure abstraction—that is, “culture”—was in itself an integral component of a moment in the social and intellectual history of their discipline (Wagner Reference Wagner1981).Footnote ²⁸ It thrived upon a historically specific “semiotic ideology” (Keane Reference Keane2003) that was part of a no-less-historical “language game” (in Wittgenstein's sense) in its own right.

If so, we might as well turn the tables once more and speak of the actants and agencies with which mid-twentieth-century American anthropology populated local ethnographic worlds in order to reproduce itself. “Factishes,” as we now realize they were, entities like “cultural traits and patterns” or “foci of culture,” mysterious processes like “enculturation” or “culture contact,” forces of “acculturation,” or postulated tendencies such as “cultural conservatism” or “reinterpretation,”Footnote ²⁹ all functioned like primitive search engines, crawling local webs of social relations and picking up ethnographic data in accordance with their built-in algorithmic specifications so as to recursively describe such webs as “a culture” under observation. In all of this, care had to be taken to isolate the “cultural signal” from its surrounding social noise, and amplify its redundancy at a certain not always clearly specified bandwidth and in accordance with a protocol that aimed to render the ethnographer a seemingly discrete inscriptive device, and the ethnographic interlocutor a similarly unrelated producer of streams of code about—in this case—stones, gods, chromolithographs, blood, and spirit possession, rather than, say, wage levels in the cane fields and toothpaste factory, baseball scores, games of dominos, the dawn of the Cold War, or the election of Carlos Prío SocarrasFootnote ³⁰ to the Presidency of the Cuban Republic.

Of course, in parsing this complex stream of information, with so many things happening at once and at so many scales, users of programs like “Herskovitsian Acculturation Research,” such as William Bascom, had to mobilize a host of actors and actants: not just santeros, oracles and deities, but grant monies, tape recorders, typewriters, payments to informants, fieldnotes, Africanist ethnographies, language skills in Yoruba, a Spanish-speaking wife, Cuba's relations to the United States, national and local patronage in Cuba, hotel rooms, automobiles and drivers, letters to Northwestern University, theories of culture and personality, concepts such as “syncretism” or “socialized ambivalence,” cooks and maids, and the rest. What a staggering amount of “interessments” and “enrolments” (Callon Reference Callon and Law1986) to be engineered!Footnote ³¹ When all of these came together, what resulted were a couple of articles and sometimes a monograph. Life came to be transduced into and materialized as text, behavior and speech transformed into data, data coded into patterns, and patterns abstracted as “culture.” This was the closest that ethnography comes to spirit possession in creating a sense of immediacy, however painstakingly mediated, that effectively black-boxes all the labor that went into making the mediators vanish into the manifest message constituted by an entity otherwise as rarefied as the oricha: “culture” (Keane Reference Keane2013).Footnote ³² To much the same effect, just like hierophanies, ethnographies are always open to post-facto critique. The point, however, is that just as successful “materializations” of the oricha in human bodies drive the game of Afro-Cuban religion onward into the future, so do successful “materializations” of culture in ethnographic texts reproduce and advance the no less serious game of anthropology.Footnote ³³ In both cases, transduction is the name of the game.

The result in the case at hand was Bascom's utterly nonsensical proposition that his interlocutors in Jovellanos were suffering from what Herskovits (Reference Herskovits1937b) had diagnosed as “socialized ambivalence” as a result of “culture contact.”Footnote ³⁴ That this strikes us as wrongheaded today may reflect that Bascom was unable to run the program for long enough to work out the bugs inherent in it, or had earlier failed to note the presence of stones in Nigeria, and thus employed the wrong algorithm in Cuba. But in a deeper and theoretically more interesting sense, it reveals how “history cooks us all” (Palmié Reference Palmié2013)—subjects and objects of ethnography, programs and platforms, concepts, theories, and social worlds alike. What our contemporary discomfort with the entire project that Bascom's notes and scant publications attest to is that the search engine and inscriptive program on which it relied now looks to us like an app whose platform has become obsolete. Unsupported by any operating system still available to us, the paper-born traces—indexes in the true sense—of Bascom's intellectual labors are still available in journals and in Berkeley's Bancroft Library. They are a bit like the records of 1969 Apollo moon landing that languish on obsolete hardware that no computer system today is equipped to process, or, for that matter, like the otanes of an oricha who no one knows how to properly feed anymore.Footnote ³⁵ All we can do with them now, it seems, is to marvel at their “radical alterity” (or perhaps not so radical), and reconcile ourselves with the fact that all “ontologies,” including whatever we think of as “ours,” are inescapably historical entanglements of matter, praxis, and language. Reason, perhaps, for a sigh of relief.

In sum, what none of the anthropologists at the time realized—and this is one lesson to take from my thought experiment—is that the ethnographic “naturalism” within which they framed their questions was just one, necessarily historical cultural formation among others (Viveiros de Castro Reference Viveiros de Castro2003; Descola Reference Descola2013). The ambivalence was on the American anthropologists’ side, not that of the people they encountered in Cuba or Haiti. Some of Bascom's Cuban interlocutors tried their best to teach him otherwise during that summer break from Northwestern back in 1948, but so it goes. Might it not behoove the proponents of an “ontological turn”—and indeed all of us—to take that lesson to heart?