The normative Western post-Enlightenment understanding of the world is that whereas there are many different cultures, both now and in the past, there is, and indeed only can be, one underlying nature. This modern Western position can be called ‘naturalism’.Footnote ¹ But what if the nature/culture dichotomy is not universally valid? What if in some places, or at some times, various peoples have inhabited a natural world that is specific to their culture? In the case of the ancient Greeks, it would be a nature in which they shared the world with supernatural beings of many different kinds, one in which even trees and springs might be alive. If this were true, it would obviously have a huge impact on the way we study, understand and describe ancient Greek religion.

Since its inception in the late 1990s, the ontological turn in anthropology has asked us to take seriously what is specific to a particular culture’s lifeworld, while at the same time seeking to collapse the traditional dichotomy between the multiplicity of culture and the uniformity of nature.Footnote ² Advocates of the ontological turn in its original, and most controversial, articulation maintain that pre-modern and contemporary traditional cultures inhabit different worlds, each existing within an alternate reality. In other words, it is not only cultures that may vary but even nature itself may vary.Footnote ³ As expressed in a very influential edited volume, ‘If we are to take others seriously, instead of reducing their articulations to mere “cultural perspectives” or “beliefs” (i.e. “worldviews”), we can conceive of them as enunciations of different ‘worlds’ or ‘natures’, without having to concede that this is just shorthand for “worldviews”.’Footnote ⁴

In other words, the ontological turn differs from cultural relativism in the claim that it is not only epistemologies (forms of knowing and understanding) that may vary between cultures, but even ontologies (forms of being or existing).Footnote ⁵ According to the Greeks’ ontological conception of how reality is put together, the gods took an interest in the welfare of human beings and were both willing and able to interact with them either directly (through epiphanies, dreams, and oracles) or more indirectly through omens and signs.Footnote ⁶ The most radical form of the ontological turn would take this a step further and posit that in the world of the ancient Greeks their various gods and other supernatural beings actually existed. By contrast, the traditional anthropological approach to cultural difference has been to speak of different worldviews, all of which share the same underlying reality. At this point it is worth stressing that even if that traditional approach is actually the correct one, it is still fundamentally necessary to take seriously the ontologies of other cultures as viable ways of experiencing their worlds.Footnote ⁷

In their most recent treatment of the subject the anthropologists Martin Holbraad and Morten Pederson have significantly modified the radical claims mentioned above, asserting that the ontological turn in anthropology is ‘decidedly not concerned with the “really real” nature of the world or any similar metaphysical quest. Rather, it is a methodological project that poses ontological questions to solve epistemological problems’.Footnote ⁸ To put it simply, they argue that anthropologists need to adjust their analytical categories and concepts in order more accurately and appropriately to account for their ethnographic data and its underlying ontological assumptions. Their assertion that the ontological turn has never been about metaphysical claims may seem somewhat disingenuous, given their own previous scholarship.Footnote ⁹ Nevertheless, in this discussion I accept their revised position that taking different worldviews/lifeworlds seriously does not necessarily entail the conclusion that nature is itself mutable.

This essay pursues an ontological understanding of consultations at Dodona by employing the second wave of ontological theory as advocated by Holbraad and Pedersen. My premise is that if we are to understand a divinatory consultation as the Greeks themselves did, then we need to put aside our own Western post-Enlightenment (largely secular) ontological assumptions and view the world through the ontological assumptions of the Greeks themselves.Footnote ¹⁰ This is a much more radical suggestion than the traditional injunction of putting on the cultural filters of the ancient Greeks, in as much as that step is then invariably followed by an act of cultural translation.Footnote ¹¹ That is, I wish to sidestep the translation of their cultural notions into our own Western Euro-American explanatory terms; for most translations of this kind are invariably mistranslations, since the terms involved are incommensurable. In effect, I am suggesting that the practice of divination should be both analyzed in emic terms and then described in those terms as well, rather than being redescribed in our own terms.

Consulting the Gods at Dodona

The best case study for understanding Greco-Roman divination in this way may well be Dodona with its vast corpus of surviving questions. So far, over 4,000 oracular inscriptions, inscribed on some 1550 lead tablets called lamellae, have been published. Some of these lamellae contain answers as well as questions.Footnote ¹² The longstanding debate about the nature of Delphic verse responses that are preserved in literary sources will probably never be settled; but even if we cannot know for certain how the lamellae were used at Dodona, we nevertheless have objects to investigate, which are both artifacts of the consultation and texts contemporary with it.Footnote ¹³ According to the most common reconstruction, the lamellae were written either by the consultants themselves or by someone acting on their behalf. The lamellae would then be handed over to some official of the shrine or to one of the priestesses. Either before or after the conclusion of the consultation, they would have been rolled and folded, and some, or perhaps all, of them would have been left in the temple as a votive. As things inscribed with texts they participated in a network that linked gods (Zeus and his consort Dione), one or more priestesses and possibly priests, a human consultant, and perhaps other actors (such as scribes) as well.

Nonetheless, a caveat is in order over and beyond our uncertainties about the mechanics of the consultation and the uses of the lamellae. The individuals who chose to consult oracles occupied a particular standpoint in their society. Others may have been less likely to consult seers or oracles for a variety of reasons ranging from skepticism to outright rejection of their truth claims. The lamellae represent the particular ontology and belief system of one (and an undoubtedly very large) set of situated observers – those who both believed in and felt the need to consult oracles. For my own part, I have no doubt that the number of agnostics and atheists in classical Greece, primarily to be found among the sophists and their elite students, was relatively small.Footnote ¹⁴ After all, our main eyewitness for the whole range of Greek divinatory practices is Xenophon, a member of the international elite who was a follower of Socrates and perhaps a student of Prodicus.Footnote ¹⁵ However that may be, the situated knowledge of the skeptics and doubters, whatever their actual number, would have been very different from that of the many thousands of consultants at Dodona. By ‘situated knowledge’ I mean the proposition that all forms of knowledge are affected by the specific historical, cultural, linguistic and value contexts of knowing persons and reflect their social identities and social locations.Footnote ¹⁶ As the anthropologist Sherry Ortner has observed in relation to her own intellectual development, by the end of the twentieth century anthropologists had begun to think differently about ‘cultures’, regarding them as ‘more disjunctive, contradictory, and inconsistent’ than she had been taught.Footnote ¹⁷

Ontology, needless to say, is not the only theoretical lens at our disposal, even if it is one of the newest. Functionalist, symbolic and performative perspectives all have value in revealing what is at stake and at play in divinatory rituals. But they are etic, rather than emic perspectives. As Martin Holbraad has astutely expressed the situation:

The crucial point is that the Greeks, even if their culture was not completely homogenous (as no culture ever is), had a different understanding of reality, of what constituted the ‘really real’, than our own. And the methodological problem is the way that their ontological assumptions and categories have been mistranslated into modern secular ones. Agnosticism, for instance, has been re-described as atheism and belief as ritual practice.

The problem, however, with giving a purely emic account is its entanglement with cultural relativism as a way of coping with cultural difference. One can imagine this kind of mindset, ‘How quaint – the Greeks believed that they were communicating with their gods – I respect their belief, but I know that it’s wrong because there weren’t any gods there to communicate with. What they were doing, whether they knew it or not, was this – resolving indecision, reaching consensus, negotiating a solution through performance, and so on.’Footnote ¹⁹ A better way to approach cultural difference is through an ontological perspective which takes seriously that the Greeks had both a culture and an ontology that was different from our own. So by putting both etic and emic accounts within an ontological framework I hope to reach an understanding of a consultation at Dodona that would be consistent with the world that the Greeks actually inhabited rather than with the one that most modern scholars do.

Let us start with a very basic question. Why would a person invest their time and money in order to consult the oracle of Zeus at Dodona, or indeed other oracle centers, such as Delphi? Plutarch (Mor. 386c) implies that inquirers at Delphi typically asked ‘if they shall be victorious, if they shall marry, if it is advantageous for them to sail, to farm, to go abroad’. This type and range of question are confirmed by the lamellae from Dodona. Inquirers seek answers to matters that affect their personal lives, including marriage and children, health and well-being, business dealings and the manumission of slaves, employment and military service, theft and adultery, forthcoming court cases and athletic contests.Footnote ²⁰ Twentieth-century functionalist anthropology was not wrong to insist that when a person made a consultation it was to resolve such dilemmas. In his Memorabilia, Xenophon attempts to show that Socrates’ religious attitudes and practices were both traditional and completely normative by having him say (4.3.12): ‘In so far as we are unable to foresee what is advantageous for the future, the gods themselves work with us, indicating through divination to those who consult them what is going to happen and teaching them how to obtain the best results.’

Where modern scholarship, both anthropological and in the disciplines of Classics and ancient history, has encountered difficulty is in explaining who was responsible for giving the requested advice and how the answers were delivered. It is precisely here where ancient Greek and modern ontologies, emic and etic understandings respectively, become incompatible. If you were able to interrogate a Greek standing in line at Delphi or Dodona for an explanation of how their question was going to be answered, you would be told that a god was going to deliver it. And if you asked them how they knew that, they could quote a number of texts, most famously perhaps the Homeric Hymn to Pythian Apollo (287–293), where Apollo says that he intends to establish an oracular center where he will give ‘unerring advice’ to all who consult it. If pressed further, they might say that this is also what they were told by their parents, and was common knowledge. And finally, if asked if they really believed this, you might get a blank stare.

Such things were not a matter of high intensity creedal belief, as they might be for a Christian living within a religious marketplace, but were rather a matter of common knowledge and, most importantly perhaps, of direct experience.Footnote ²¹ As Bruno Latour has argued in his essay, ‘“Thou Shalt not Freeze Frame” or how not to misunderstand the science and religion debate’, Christianity is not at all about belief (‘Belief is simply immaterial for any religious speech-act’: p. 122), but about experience (in the form of immanence rather than transcendence).Footnote ²² At Delphi the priestess known as the Pythia served as the mouthpiece of the god Apollo. The Greeks who actually consulted her, if not the modern scholars who merely rationalize the entire procedure, had no doubt that Apollo possessed her and spoke directly through her; the voice was hers, but the words were his.Footnote ²³ And whoever put the following question to the oracle at Dodona in the early fifth century BCE must have been certain that the gods would not only give him an answer but could even know his thoughts: ‘By praying to which of the gods would he accomplish what he has in mind’ (tini ka theōn euxamenos praxai ha nooi eche;).Footnote ²⁴ As I have argued before, it was in the context of the divinatory ritual that the real presence of the divine was commonly to be experienced by the Greeks. Footnote ²⁵

Divinatory Truth as Transformative

All of this, however, raises the thorny question of what kind of truth-claims Greek divination was predicated upon. As Martin Holbraad has shown in his paradigm shifting study of Ifa Divination practiced in Cuba:Footnote ²⁶

Likewise, the truth that Greek oracles pronounce may strike us as irrational or primitive because it is a nonrepresentational truth insofar as it cannot be doubted or falsified. This is a different kind of truth from normative common-sense notions of truth because its source, the god Zeus, knows all things. His declarations, whether delivered at Dodona by himself, or at Delphi through this son Apollo, are both true and indubitable.Footnote ²⁷ The Greeks conceived of oracular truth as not only being indubitable but also as true for all time. This conception is well illustrated by an episode recorded by Plutarch in Nicias (13.1). Before the Sicilian expedition of 415 BCE the oracle of Zeus Ammon at Siwah oasis delivered an oracle that ‘The Athenians shall capture all the Syracusans’. Plutarch later comments (Nicias 14.6) that the seers serving the Athenian forces were distressed by the possibility that this prophecy was accomplished when the Athenians captured a Syracusan ship conveying tablets recording the complete list of Syracusan citizens; but Plutarch then adds that it was also said that the oracle was fulfilled later (353 BCE) when Kallipos, an Athenian, killed Dion and took possession of Syracuse. The interesting point is the normative belief that the fulfillment of oracles was inevitable, either in the short term or even many decades later.

But we may go even further and say that the oracles delivered by Zeus are not just predicative: they are also simultaneously transformative, insofar as they irrevocably change the life circumstances of the consultants who receive them. They are a special transformative kind of truth that is fundamentally different from most of the truth-claims uttered by mortals.Footnote ²⁸ Given that the Greeks considered Zeus and Apollo’s verdicts to be indubitably true, if it turned out otherwise, one of the main pillars of their ontology would collapse.Footnote ²⁹ This is not to say that no Greek consultant ever expressed doubts – but those doubts centered on the competence and honesty of freelance seers and on the authenticity of specific oracles. General statements of disbelief are exceedingly rare, and in literary texts quickly and decisively disproved.Footnote ³⁰ As mentioned above, Bruno Latour has suggested that in Christianity religious speech-acts are not really about belief but rather about transformative experience; they are not so much informational as transformational.Footnote ³¹

Despite Latour’s protestations against extending this insight to other religious traditions, it does seem relevant to how the Greeks interacted with and communicated with their gods. In most religious systems objects play an important role in activating the experience of immanence, and for the Greeks this role was played most obviously to us by statues, images on vases and relief sculptures, but also undoubtedly by the lamellae themselves. They are not simply records of a consultation, to be taken home or hung in the sanctuary. They are also the consultation’s embodiment. To touch a lamella, to read it or hear it read, is to experience divinity. And even the most banal of answers are not merely or simply informational. They are also transformative. Or, as Holbraad has argued in his ground-breaking work on Ifa divination, they change the very nature of reality. Perhaps the views of Latour and Holbraad, although dealing with different religious traditions, converge in an unexpected way. Hearing the words of the Priestess at Dodona as she gives the god’s response, or (re-)reading the response on the tablet, is transformative in that it unalterably and unquestioningly alters the course of one’s life.

Such must have been the experience of an inquirer named Arizelos when, sometime in the fourth century BCE, he asked what occupation he should undertake: no alternatives are given and the question is essentially open-ended:Footnote ³²

The existence of essentially open-ended questions belies the common modern assumption that questions were always carefully formulated so as to restrict the range of possible responses. To be sure, they often were, but that does not foreclose the possibility that some consultants had so much confidence in the gods that they were willing to let their lives be utterly transformed in a way that the gods judged best.

Epilytos also was willing to let the gods transform his life, as we know from a question that he posed in the middle of the fourth century BCE:Footnote ³³

The complexity of Epilytos’ question is not unparalleled. Isodemos also poses a three-part question, dealing with marriage, children and citizenship (fourth–third century BCE).Footnote ³⁵

Epilytos and to a slightly lesser extent Isodemos were amazingly brazen in the number of questions they posed at the same time. Epilytos, in particular, was definitely taking a risk that the whole course of his life might be altered: for instance, he might have gone home to a new wife and new profession. The latter is not as unlikely as it may sound.Footnote ³⁶ These three Greeks (Arizelos, Epilytos and Isodemos) were willing to let the gods, speaking through the medium of their oracle, transform their lives. Zeus is not giving advice here – he is being asked what they should do, and they were willing to be transformed by an answer that was not subject to falsification or doubt.

The How and Why of Divine Communication

Although modern anthropological studies have revealed that the divinatory ritual has consequences that are social, political and psychological,Footnote ³⁷ it is essential not to conflate or confuse the by-products with its fundamental purpose.Footnote ³⁸ First and foremost, divination is a system of communication that is intended to expand the range of ordinary human knowledge.Footnote ³⁹ But, like other religious practices, divination also has various direct and indirect consequences that impact individual consultants in profound ways – such as alleviating anxiety, reducing uncertainty, managing risk and resolving indecision. These are, in effect, divination’s secondary functions. The same can be said of Greek religion generally – that the purpose of all of its rituals and practices was communicative in that they were the means whereby communities and individuals attempted to seek advice from the gods, win their support and favor, establish reciprocal relationships and even to gain their friendship.Footnote ⁴⁰

Classical scholars understand Delphic prophecy through the lens of a modern Western ontology, not least of all in their sureness that the Pythia herself, being an uneducated woman, could not have generated spontaneous verse oracles in hexameter verse (the verse of Homeric epic), even though it is not especially difficult to do so and despite comparable evidence from contemporary non-Western cultures.Footnote ⁴¹ The same problem holds for Dodona, and some scholars have gone so far as to substitute ‘priests’ for ‘priestesses’ in their discussions, even though only Homer and Pindar mention a male priesthood (the Selloi).Footnote ⁴² All other sources refer to one or more priestesses. Herodotus tells us that the priestesses at Dodona expressed to him their own speculations about the origins of the gods and were well informed about the mythical origins of Dodona.Footnote ⁴³ He clearly sees them as the local authorities, he names them and he implies that any other officiants at the site had positions of lesser importance and prestige: ‘The priestesses of Dodona said these things, of whom the oldest was Promeneia, the next Timarete, and the youngest Nicandra. And the other Dodoneans who are involved with the temple agree with them.’

According to Plato in the Phaedrus, both the Delphic priestess and the priestesses at Dodona prophesied in a state of altered consciousness (which he calls mania), and were practitioners of ‘inspired prophecy’ (mantikē entheos).Footnote ⁴⁴ Euripides similarly links the Pythias at Delphi and the priestesses at Dodona in a fragment of his lost play Melanippe, but without specifically mentioning ecstatic possession by a god. Melanippe asserts that women are better than men for many reasons, including their essential role in worshipping the gods: ‘In Phoebus’ halls women proclaim (prophēteuousi) Apollo’s will; and at the holy seat of Dodona, beside the sacred oak, womankind conveys (poreuei) the will of Zeus to all Greeks who wish to know it.’Footnote ⁴⁵

Rather than dismissing these emic accounts out of hand, it is methodologically preferable to work from the reasonable assumption that, as in other cultures, more than one method of divination might be employed simultaneously both at the same shrine and even by the same person.Footnote ⁴⁶ At Dodona in particular, the varied testimony of our sources strongly indicates that different divinatory techniques may have been in use either concurrently or in conjunction. Sarah Iles Johnston plausibly reconstructs how the priestesses at Dodona might listen to and interpret the sounds made by doves, ringing cauldrons, rustling leaves or a murmuring spring, while simultaneously being in an altered state of consciousness.Footnote ⁴⁷ Indeed, there are cultures in which a diviner will both become possessed and at the same time practice an empirical form of divination such as ornithomancy (observing the flight of birds) or cleromancy (drawing lots).Footnote ⁴⁸

There is, in fact, an abundance of evidence, especially from literary sources, but also from the lamellae, that the priestesses at Dodona sometimes gave fully articulated answers.Footnote ⁴⁹ In 367 BCE the Spartans won a battle against the Arcadians that came to be known as the Tearless Battle, because not a single Spartan was killed. Diodorus, undoubtedly drawing on the fourth-century BCE historian Ephorus of Cyme, claims that this had been foretold by the priestesses at Dodona: ‘The Priestesses at Dodona proclaimed to the Spartans that this war would be tearless.’Footnote ⁵⁰

Xenophon does not name the battle in his Hellenica, nor does he report any oracle in connection with it.Footnote ⁵¹ For Parke this was sufficient reason for rejecting the historicity of the oracle and postulating that it was later invented in the form in which Ephorus recorded it.Footnote ⁵² Even leaving aside the implicit, and unproved, assumption that Ephorus was not a very careful researcher, no scholar alive today would reject as unhistorical all of the things that Xenophon does not tell us that are mentioned in other sources.Footnote ⁵³ Yet Parke’s full explanation, although written in 1967, reveals the still dominant way in which scholars erase the evidence for verbal responses: ‘Of course, it would be possible to argue that the oracle itself was an invention, particularly as the method of divination used at Dodona did not lend itself readily to these verbal details in the responses.’Footnote ⁵⁴ The circularity of the reasoning is transparent – if one dismisses all of the evidence for ‘verbal details’ both in the literary testimonies and in the preserved lamellae, then of course this is no evidence for them.

Although the evidence is late, it seems that before their invasion of Sicily in 415 BCE the Athenians consulted three different oracles: the oracle of Zeus Ammon at Siwah oasis in Libya,Footnote ⁵⁵ the oracle of Zeus at DodonaFootnote ⁵⁶ and the oracle of Apollo at Delphi.Footnote ⁵⁷ Pausanias reports the answer from Dodona as follows:

The fact that Thucydides does not mention these consultations is hardly an objection to accepting their authenticity. The luminous historian shows a very limited interest in oracular responses, which is not to say that he never reports them. The important thing is that his silence, as also with Xenophon’s, can never be used to prove that something did not take place, as the debate over the existence of the Peace of Callias all too well illustrates.

Finally, it would be hard to justify a total rejection of the testimony of Demosthenes, who in three of his speeches cites oracles from Dodona.Footnote ⁵⁸ To take just one of these, in his speech Against Meidias of 347/6 BCE, Demosthenes quotes two oracles from Delphi and two from Dodona, and these texts are included in our manuscripts. Their authenticity as genuine oracles has been much debated, but the first of the two Dodona oracles has every appearance of being an actual answer (as opposed to being a later pastiche or outright invention):

This oracle, it must be admitted, is irrelevant to Demosthenes’ case that Meidias had committed impiety by assaulting him when he was a choregos (‘chorus-producer’) at the Greater Dionysia, except insofar as it reinforces the general point that it is essential to perform festivals in the proper way in order to avoid angering the gods.Footnote ⁵⁹ The Athenians failed to send a sacrifice and sacred embassy to a Dodonian festival at the appropriate time and now they have to make amends. There is no reason at all to doubt the genuineness of this particular oracle either on philological grounds or on the basis of its content: ‘The deities, sacrificial animals, and dedication all suit a genuine occasion and oracle.’Footnote ⁶⁰ If one follows the methodology of judging each of the documents (legal or religious) cited in the Attic orators on its own merits, then there is absolutely no rational reason for dismissing this one as a much later insertion.Footnote ⁶¹ But even if one took the extreme position that all of the oracles quoted in the texts of Demosthenes’ speeches are later ‘forgeries’,Footnote ⁶² that still leaves us with the fact that Demosthenes quoted fully articulated oracles in his speeches and that must entail the further fact that he expected his audience to accept the existence of such oracles.

Nevertheless, most scholars reject the historicity of these and other prose and verse responses from Dodona that are quoted in literary texts, and they do this despite the fact that a wide range of Greek and Roman writers take their authenticity for granted. The way that Parke rationalizes the evidence is typical of the standard modern approach to the literary evidence, which rejects as literary fiction or explains away as subsequent embellishment anything that seems impossible in terms of a Western secular ontology. This approach is nothing less than an act of cultural mistranslation, whose only purpose is to make explicable in modern terms that which seems inexplicable in its own terms. Reality as experienced by the Greeks was not the same as that experienced by us, no matter how absolutely certain one is that the secular, post-Enlightenment intelligentsia knows better than the subjects of their inquiry, whether those subjects be ancient Greeks or the members of contemporary traditional societies and religious groups.

The dominant position in modern scholarship is that at Dodona, as at Delphi, the vast majority of questions and answers took a very simple form. The consultant posed his question in this way: ‘Would it better and more profitable for me to do x’, or ‘which god or gods should I sacrifice to?’ We do have examples of questions put in this form. Nonetheless, this scenario allows the comfortable and rational conclusion that most answers were correspondingly in the form of a ‘yes’ or ‘no’, or very occasionally of a ‘do x or y’, or of a list of deities to whom one should sacrifice, all of which could potentially have been answered by the priestess drawing one or more lots. Additionally, it has been argued that a system of lottery was employed in which the consultant sometimes submitted two tablets containing alternatives.Footnote ⁶³ Yet, it should be noted, the single literary reference to the use of lots at Dodona does not conform to these standard modern reconstructions of the procedure.

According to the fourth-century BCE Greek historian Kallisthenes of Olynthos (as paraphrased by Cicero), when the Spartans were enquiring ‘about victory’ before the battle of Leuktra in 371 BCE, the Spartan ambassadors themselves ‘set up the vessel containing the lots’ (legatique vas illud, in quo inerant sortes, collocavissent) and when a pet monkey overturned it, ‘the priestess who is in charge of the oracle said that the Spartans should think about safety, not about victory’ (sacerdos dixisse dicitur de salute Lacedaemoniis esse, non de victoria cogitandum).Footnote ⁶⁴ Even in our single extant description of how a lottery system may have worked, the priestess, however exceptionally, articulates the answer on her own.

Now most oracular responses at both Dodona and Delphi were probably fairly prosaic (in both the literal and figurative senses of the word). Nonetheless, a few of the handful of recorded answers at Dodona give very specific answers to specific questions. There is no ambiguity or wiggle room in these answers. They close down options rather than letting the questioner negotiate a range of possible actions. Here are two examples. The first (middle of the fourth century BCE) is inscribed:Footnote ⁶⁵

Side A:

Side B:

The single word ‘sea’ would certainly have been a satisfactory answer, and perhaps the one that Lysias was hoping for. But the god has gone further and directed that Lysias do nothing by land. Considering how risky it was to ‘take a share of a ship’, especially for purposes of sea-borne trade, this is far from being a safe or conservative injunction. One wonders if Lysias had contemplated the possibility that he could be given so restrictive a response, one that effectively eliminated the possibility of economic diversification.

My second example (first quarter of fourth century) is similar to this one.Footnote ⁶⁶

Side A:

Side B:

Here again, the answer is restricted, but this time to the safer option, since just as commerce by sea was much riskier than that by land, so too naval warfare was more dangerous than a land campaign. In addition, there was the very real danger that if you died at sea your corpse might not be retrieved for a proper burial. It is noteworthy that the questioner assumes that Zeus and Dione do not need to be told his name or the circumstances of any particular military campaign that he may be contemplating. The gods simply know these things and there is no presumption at all that such information might also be useful to the priestesses who were involved in the consultation.

There is something especially striking, and so far unprecedented, about this particular response. The use of the Attic-Ionian form of the dative (gēi) and of the verb (skethe), which is rare in prose but frequent in Homer, lends it an epic flavor and tone.Footnote ⁶⁷ This may seem surprising if viewed strictly in the context of the other tablets, but it is less so in the context of the literary tradition (Pausanias 10.12.10, for instance) and of Herodotus’ depiction of the priestesses as highly educated individuals.Footnote ⁶⁸ In any case, even if no other example of Homeric or poetic language is ever found on a lamella, that is not a reason for reinterpreting the status of this one (DVC 108B) as an answer.Footnote ⁶⁹ The gods, as well as the inspired priestesses who communicated the gods’ answers, did not follow some sort of rule book dictated by modern scholars. The gods could, and indeed did, respond as they wished. And inspired seers, as we know from well-documented Tibetan examples, sometimes speak in highly poetic language and sometimes not.Footnote ⁷⁰ We have no warrant to deprive Zeus and Dione, nor the cult personnel, of their agency.

Gods as Objects in an Object-Oriented Ontology

Of course, someone might object that for the ontological model to work, the Greek gods needed to really exist. Otherwise, no matter how cleverly Holbraad and Pederson nuance the claims that they and others had previously made about the existence of alternative realities, we are right back where we started, with priestesses and priests formulating the recorded responses and being fully in charge of the disposition of lots. Agency, therefore, is still in their hands, as well as in the hands of chance depending on how one reconstructs the use of lots.

The vast majority of modern scholars have written the gods out of the script in their efforts to reconstruct how an oracle consultation ‘really worked’.Footnote ⁷¹ Taking the gods out of the script, denying them any agency in the divinatory ritual, has serious consequences for how the ancient evidence is interpreted: spontaneous verse oracles are generally considered to be ontologically impossible; a system of lots has been imposed upon the Pythia, in the absence of evidence;Footnote ⁷² and, as noted earlier, binary lots have been read into the lamellae where the Greek does not necessary carry this meaning. What we are then left with is a thoroughly etic reconstruction that takes no account of emic explanations – an account that is written by and for ourselves, composed from the high perch of our own supposedly value-free, objectively valid and scientifically more advanced understanding of the ‘really real’. That is how we reach the truth, or is it? By transferring agency from the gods to the calculations of religious personnel and the randomness of the lot, don’t we moderns run the risk of radically misunderstanding the actual lived experience of the subjects of our inquiry, thereby transforming the ethnographic enterprise into a flagrant act of cultural mistranslation?Footnote ⁷³

In any case, whether the Greek gods were real or not is surely the wrong question to ask, and not just because it is unanswerable.Footnote ⁷⁴ I want to stress that I am not making the strong ontological claim that the Greek gods ‘really’ existed, only the weaker claim that the Greeks interacted with them, experienced them visually and aurally, and were influenced by them as if they did. Conversely, I cannot assert categorically that the Greek gods were not really there, in one form or another.Footnote ⁷⁵ Cognitive science can only tell us which parts of the brain are activated during a ‘spiritual’ experience – it cannot tell us that the brain in and of itself is the cause of that experience.Footnote ⁷⁶ Indeed, the very same neuroscientific data can be used to argue both for and against the existence of god.Footnote ⁷⁷ But there is something of which I am more certain. The anthropologist Maurice Bloch has stressed the importance of the anthropologist being a participant observer and of providing an ethnographic account of the conceptualization of a society that makes sense to one’s native informants.Footnote ⁷⁸ This essay has been an attempt to write an account that an ancient Greek consultant at Dodona would have understood, one that would, in her or his understanding of reality, make ‘sense’.

If we wish to gain real insight into how Greek consultants might have experienced a consultation at Dodona, there needs to be an active role for the gods. In terms of Bruno Latour’s actor-network theory the assemblages would have consisted of the interaction of consultants, priestesses, gods (Zeus and Dione) and lamellae. What we then have at Dodona is a human/nonhuman hybrid relational network consisting of things, people and supernatural beings. The things are lamellae, the sacred oak tree, cauldrons, urn; the mortals are inquirers, priestesses and their various assistants; and the supernatural beings are the gods Zeus and Dione. All of these participants, human and nonhuman alike, exert agency. Agency is not the sole product of human intentions but rather the product of this particular network of relations – agency is effectuated when the various relational elements interact with each other.Footnote ⁷⁹ The gods cannot be left out of this network or given minor roles as mere social constructions or figures of the imagination. Rather, in any given divinatory session, consultant, lamella, god(s) and priestess(es) all share agency as objects. We need to steel ourselves to think in terms of an object-oriented ontology, one that postulates that the world is composed not of ‘subjects on one hand and noumenal objects on the other, but rather of nothing but objects, animate and inanimate, human and non-human, all of which have to be taken as agents’.Footnote ⁸⁰ The Greek gods, therefore, must be taken seriously as objects that are equivalent to all other objects in their ability to both affect and be affected by those other objects, human and otherwise, that they becomes entwined with.Footnote ⁸¹

Needless to say, there was no Greek articulation of ‘network theory’. Nonetheless, the application of higher order concepts can reveal relationships that are not consciously articulated by people themselves. Here especially an ethnographic parallel can help to illustrate this point, and a particularly pertinent one can be found in Navaho religion. The Navaho are the largest Native American tribe in the United States, with more than 400,000 enrolled members, many of whom live on their reservation (officially called the Navaho Nation).Footnote ⁸²

The contemporary Navaho medicine man (hatałii, or ‘Singer’) Perry Robinson describes the casual relationships between gods, medicine man and patient in Navaho ceremonials, in terms of what can fairly be called a human/nonhuman hybrid relational network consisting of things, people and supernatural beings.Footnote ⁸³ During Navaho healing ceremonies, such as the Nightway Chant (Yei’ Bi’ Chei) that takes place over nine consecutive nights, the patient, medicine man and Holy People (Diyan Diné, the Navaho Gods), all have to cooperate and fulfill their roles if healing is to take place.Footnote ⁸⁴ The patient must have positive thoughts and the desire to be healed; the medicine man conducts the ceremony, sings the songs and offers the prayers that summon the Holy People to the ceremony; the sandpainting (on the floor of the hogan in which the curing takes place)Footnote ⁸⁵ outlines the origin story of the ceremony, prescribes the steps to be taken, and holds the power to help the patient; and the Holy People effect the cure if they are satisfied that the entire ceremony has been conducted in precisely the correct way and that all in attendance are in the right frame of mind.

So too in ancient Greek culture, religious specialists (priests, priestesses, seers) need to pray using the correct words and conduct sacrifices in the prescribed ways; worshippers need to provide unblemished, suitable and pleasing gifts and sacrificial animals; and the gods need to be both pleased with and willing to accept the gifts and sacrifices and consequently to grant whatever request was being made (such as for advice, healing, health, children, a good harvest, success in war and so on). The effectiveness of divine-human-material networks is not automatic either in Navaho or Greek society. In Book 2 of the Iliad (402–420), Agamemnon sacrifices a bull to Zeus, with the prayer that Troy be captured before the setting of the sun. Zeus accepts the sacrifice, but rather than granting the prayer’s request for an immediate victory, he increased the toils of the Greeks.

Latour, for his part, explicitly employs network theory to explain relationships that he takes seriously as including the agency of supernatural beings. His own ontological commitments are revealed when he imagines a scenario that all too well captures the current methods of most scholars of ancient religions:Footnote ⁸⁶

Critical scholars, in other words, know that this is not what is really going on for the seemingly simple reason that it is ontologically impossible to be ‘called by the Virgin Mary’. ‘Impossible’ for them perhaps, but not for the person who believes in the power and presence of supernatural beings.

In the universe inhabited by the Greeks, and many other cultures past and present, the gods made themselves present to mortals in many different ways – oracles, dreams, portents and even, if more rarely, through epiphanies. The records of these manifestations still exist – through inscriptions, votive reliefs and literary texts. When a pilgrim, tablet in hand, read his question aloud at Dodona, he stood in the presence of supernatural powers. Those powers could read his mind, they knew his situation and they were able to enact a transformation in the circumstances of a human life through their response. That, at least, is how a Greek would have explained it to an anthropologist from the future.