The problem

The question of the exact nature of the Pythia's expertise has been the subject of academic debate for a very long time. It would indeed not be an exaggeration to say that this has been, and continues to be, one of the most controversial questions in the study of ancient Greek religion. Modern scholars are sharply divided over whether any inspired female oracles, and especially the Pythia at Delphi, had the ability to prophesy in hexameter verse without male assistance.Footnote ¹ During the classical period the two most famous oracles were those of Zeus at Dodona in Epirus in north-western Greece and of Apollo at Delphi, which was located on the south-western spur of Mount Parnassus. According to Plato (Phaedrus 244), the Delphic priestess, as well as the priestesses at Dodona, prophesied in a state of altered consciousness (which he calls mania), and were practitioners of ‘inspired prophecy’ (mantikē entheos).Footnote ²

Although Plato has a particular agenda (to discredit the ability of itinerant seers in comparison to that of the priestesses employed by sanctuaries), I do not think that his testimony here should be doubted, even though other techniques are attested at both Delphi and Dodona (including the use of lots).Footnote ³ At Dodona in particular the varied testimony of our sources strongly indicates that the nature of the divinatory rituals may have changed over time or that different techniques may have been in use either concurrently or in conjunction.Footnote ⁴ Although Herodotus has some interesting things to say about the priestesses at Dodona, the best known of all female seers was the Pythia at Delphi, who served as the mouthpiece of the god Apollo. Apollo's oracles were themselves sanctioned by Zeus (Aesch. Eum. 17–19, 616–18; Hymn Hom. Herm. 532–40). Apollo was thought to possess the Pythia and to speak directly through her; the voice was hers, but the words were his.Footnote ⁵ Thus the Pythia saw all time and space as one. For Apollo, as Pindar expresses it (Pyth. 3.29), ‘has the mind that knows all things’.Footnote ⁶

Needless to say, the bibliography on Delphic prophecy is enormous. Part of my previous contribution to this topic consisted in building on the suggestion of W. Geoffrey Arnott in this journal in 1989 that a modern parallel might be found in the Chief State Oracle of Tibet.Footnote ⁷ This oracle was located at the monastery of Nechung until the flight of the Dalai Lama from Tibet in 1959, and now functions at the re-established Nechung monastery in exile at Dharamsala in northern India.Footnote ⁸ The Nechung oracle, a male priest, is called the Kuten (which means ‘receiving body’) and acts as the mouthpiece of the counselling spirit Dorje Drakden (‘the Renowned Immutable One’). He works himself into an altered state of consciousness, aided by the stimuli of incense, the sound of horns, cymbals, and drums, heavy clothing and armour weighing more than seventy pounds, controlled respiration, and the chanting of invocations and prayers by a choir of monks. After he has entered this altered state of consciousness, a huge helmet is placed on his head, weighing some thirty pounds.

Since new evidence has emerged and older evidence has not yet been fully exploited, I would now like to revisit the comparison between Greek and Tibetan oracles in more detail. Such a comparison includes methods of consultation, the importance of divinatory advice for public and private decision-making, the kinds of questions put to oracles, and the forms of oracular answers (in prose and verse) that were given. Most classical scholars assert that the Pythia 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 (no more so, I would say, than improvising freestyle rap).Footnote ⁹ Moreover, it has recently been pointed out that if Greek women were competent orally to compose funerary laments in verse (very likely in elegiac metre), then so too the Pythias must have been competent to compose verse oracles in hexameters.Footnote ¹⁰ The modern denial of the Pythia's competence is at least partly due to gender bias (that is, that women were insufficiently educated to compose poetry) and partly to an assumption that divination must have been subordinated to elite male control. One standard argument is that male priests were the ones who turned the Pythia's ‘gibberish’ into poetry; another is that, in reality, almost all oracular responses were in the form ‘yes’ or ‘no’, and the verse oracles quoted in literary texts are later elaborations, perhaps being the work of professional male poets who were associated with the sanctuaries.

The dominant scholarly orthodoxy, repeated in study after study, is that at Delphi, as at Dodona, the vast majority of questions and answers took a very simple form.Footnote ¹¹ The consultant posed his or her question in this way: ‘Would it better and more profitable for me to do x’, or ‘Which god or gods should I sacrifice to in order to do x successfully?’ 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 a list of deities to whom one should sacrifice. At most, so the story goes, the question, while still being simple, might require a slightly fuller answer. For example, the question ‘Would it better and more profitable for me to do x or y?’ requires the Pythia, at a minimum, to respond ‘do x’ or ‘do y’.

The most commonly cited example of a typical question and response involves the Athenian philosopher and historian Xenophon when he consulted Delphi about joining the expedition of Cyrus the younger in 401 bc (An. 3.1.5–7). This Cyrus was a renegade Persian prince who enlisted an army of some 10,000 Greek mercenaries in an ill-fated attempt to overthrow his brother the king. It was Socrates who advised Xenophon to consult Delphi, but, as every student of Greek divination knows, Xenophon famously asked the wrong question of the god. He asked which of the gods he should sacrifice and pray to in order to make his intended journey successfully and then return safely. The question as posed was a standard formulation, but not the most appropriate one given the circumstances. Associating with Cyrus was politically risky for an Athenian and Xenophon was subsequently exiled. Indeed, Socrates foresaw this and censured Xenophon for not first having asked whether it was better for him to go on the expedition or to remain at home.Footnote ¹²

Now it is possible that a lot oracle was used at both Delphi and Dodona as a supplement to inspired divination, perhaps as a way of dealing more expeditiously with a large number of inquirers who came to ask basic binary questions.Footnote ¹³ Xenophon's question, for instance, could conceivably have been answered by a system of drawing lots for the names of gods. But if lots were employed at Delphi, as a recent study has brilliantly demonstrated, then they were not utilized by the Pythia herself, but only in the Corycian cave located on the slopes of Mount Parnassus, where some 24,000 knucklebones (used for divinatory rituals) have been discovered.Footnote ¹⁴ Moreover, there are examples of answers from both Delphi and Dodona that pass the test of reasonable doubt in terms of their authenticity and which could not have been answered by any system of lottery. A few of the lead tablets from Dodona seem to give both question and response,Footnote ¹⁵ and there is one in which an inquirer named Arizelos asks what occupation he should undertake: no alternatives are given and the question is essentially open-ended: ‘Gods. Good fortune. Arizelos asks the god by doing or making what thing, it will be better and more good for him and there will be a good acquisition of property.’Footnote ¹⁶ In these examples, furthermore, the nature of the answer entails that the oracular priestess was aware of the precise question being posed to her.

In the case of Delphi, Thucydides (5.16.2) tells us that in 427 bc the exiled King Pleistoanax was restored to Sparta because the Pythia (here called by her title promantis) proclaimed to all Spartan inquirers that they should ‘bring back the seed of the demigod son of Zeus from the foreign land to their own; otherwise they would plough with a silver ploughshare’ (Διὸς υἱοῦ ἡμιθέου τὸ σπέρμα ἐκ τῆς ἀλλοτρίας ἐς τὴν ἑαυτῶν ἀναφέρειν, εἰ δὲ μή, ἀργυρέᾳ εὐλάκᾳ εὐλαξεῖν). We are further told that his enemies accused Pleistoanax and his brother of having bribed the priestess to give this response. This is one of several cases in which the Pythia is said to have been bribed, and it is reasonable to wonder what the point would have been in bribing her unless one thought that she had the agency to formulate very specific responses.Footnote ¹⁷

One can also cite passages in Greek authors that strongly suggest that oracles in verse, from both Delphi and Dodona, were considered normative and unexceptional. To give but one example, an inscription from Delphi (probably dating to 344/343 bc) records a paean to Apollo written by Aristonous of Corinth, which includes these lines (9–16):

Although poetic discourse of this type has its own conventions and creates a world beyond human perception, it is clear that Apollo was understood to prophesy in song while playing his lyre. That in turn entails, if Apollo's words were to reach the ears of mortals, that the Pythia herself channelled his words in verse. It is important to point out that the audience for this poem consisted not only of mortals present at the sanctuary but also, from an emic point of view, of Apollo himself. And Apollo could be expected to know how his own oracle functioned. Yet arguments of the sort presented above, based on our classical sources, can only take us so far and are insufficient to persuade modern sceptics. If progress is to be made, then we need new evidence and a new approach.

New evidence and a new approach

I am looking to Tibet for an example of what is possible in another society, in the hope that this might clarify what was possible in ancient Greece. At worst this might appear to be a naïvely unsophisticated example of comparative religion; but at best it might help to convince some of the sceptics who refuse to accept what our literary sources claim about the nature of the Pythia's agency. After spending many years scouring the anthropological literature for a parallel for the extemporaneous composition of verse oracles, the living Tibetan tradition is the closest that I have discovered to date.Footnote ¹⁹ Of course, I am well aware that both the political history and the theology underpinning divination are different in these two cultures.

In Tibet the status of oracles (sku rten), and of the Dharma Protectors (chos srung, bstan srung) whom they channel, has a long and complicated political history that continues today. The history of the Nechung oracle is inextricably tied to the rise of the Dalai Lamas as religious–secular leaders under the Qing Dynasty in the seventeenth century. Other oracles, and the Dharma Protectors who ‘chose’ them, are linked to alternative histories of political and religious authority, and that explains why some of the most violent schisms in Tibetan Buddhism today surround permissible and prohibited oracles and Dharma Protectors.Footnote ²⁰ The history of Greek oracular sites, as well as the nature of Greek divinities, is substantially different; and yet, as we shall see, there is considerable overlap both in the theological conception of possession and in the details of the oracular consultation.

One essential similarity between Delphic and Tibetan possession is worth stressing at the outset: they are examples of the same type of possession. In both cultures, possession is ritually instigated by the person who becomes possessed (as opposed to spontaneous, uninvited, and potentially hostile possession) and has a positive, socially useful result.Footnote ²¹ From an emic (or ‘insider’) point of view, the possessing deity provides advice that is advantageous for the consultants. In other words, the formal similarities make comparison worthwhile despite the distance in time, place, and cultural context: in both Delphi and Tibet, oracular possession is controlled, induced, ritualized, performative, and socially beneficial.

First of all, I would like to consider the role of Tibetan female oracles. There was a woman state oracle in Lhasa until 1959, and she arguably has much in common with the Pythia. Her name is Lobsang Tsedron. Oddly, this particular oracle seems to have been the subject of only one scholarly study, a long article by the anthropologist Hanna Havnevik, published in 2002. Havnevik, who interviewed Lobsang Tsedron after she had given up her vocation as a medium and become a farmer, describes her erstwhile oracular trances as follows:

Unfortunately, Havnevik does not quote any of Lobsang Tsedron's prophecies, whether in verse or prose. This is undoubtedly because Lobsang was unable to remember them after she emerged from trance. Even if she had remembered some part of a response, being illiterate, she would not have been able to write it down. Here we have both a similarity and a difference with the Pythia. The similarity is that, at least according to Plutarch writing in the first century ad, the Pythia was an uneducated woman from a poor peasant family (although one should be careful not to assume that this was also the case centuries earlier).Footnote ²³ The difference is that the evidence, partial as it is, indicates that Delphic oracles were written down at the time of their utterance and that it was considered impious to tamper with them.Footnote ²⁴

Another social anthropologist and Tibetan specialist, Hildegard Diemberger, does provide the text of a short oracle in a long article on freelance women oracles in contemporary Tibet (published in 2005). The context is a generalized description of a typical divinatory session:

First impressions can be misleading and this passage must be read with care. The reference to a ‘translator’ who acts as an intermediary does not lend support to the theory that attendant priests were the actual authors of the Pythia's oracles. It is clear that the ‘translator’ is merely helping the consultants to interpret the words of the oracle. It is not unlikely that male priests (called prophētai) acted in a similar capacity at Delphi, as has often been suggested.Footnote ²⁶ This scenario, in which the translator helps to interpret the message, but does not generate the actual oracular response, is verified by the explicit evidence of a highly placed native informant.

This informant is the Tibetan lama and scholar Lobsang Lhalungpa (1926–2008), himself the son of a former Chief State Oracle who served from 1912 to 1918. In 2001 he was interviewed on the very subject of possible parallels between the Delphic and Nechung oracles.Footnote ²⁷ He makes two interesting claims: first, that all Tibetan oracles, including those delivered by the Chief State Oracle, lesser state oracles, and even ordinary oracles, were transcribed exactly as they were delivered and without any interpretation. Secondly, when asked whether the Chief State Oracle's words were in ordinary or in highly coloured language, Lobsang Lhalungpa replied:

Similarly, it is highly probable that, whereas many Delphic oracles may have been uttered in ordinary straightforward language, some Pythias had a special talent for formulating poetic responses. More recent, and even more reliable, testimony confirms Lobsang Lhalungpa's claims on both counts. But before turning to that, let us consider an example of a short enigmatic verse oracle quoted by Herodotus that is comparable to the Tibetan oracle cited by Diemberger above (‘There is nothing which cannot be realized, you ought to keep to the very roots of Tibet’).

According to Herodotus, in 481 bc four oracles were delivered to the Greeks by the Pythia on the subject of the imminent Persian invasion of Greece led by King Xerxes. Two were given to the Athenians, one to the Spartans, and one to the Argives. Herodotus quotes all four of them, probably some forty to fifty years later. All are in verse and all are highly enigmatic. By far the shortest of the four was the one given to the city of Argos.Footnote ²⁹ The Argives knew that the other Greeks were intending to invite them to join the anti-Persian alliance, and so they sent envoys to Delphi who were to inquire what course of action would be best for them. This is the oracular response to their query:

As in the other Persian War oracles recorded by Herodotus, two things are striking. One is the poverty of vocabulary (the verb meaning ‘to guard’ is used twice: πεφυλαγμένος/πεφύλαξο); the other is the obscurity of the subject of the main verb. Who is the addressee? Is it Argos personified as a male person, holding a spear, sitting on the ground, and protecting his head? Or is it the body politic, the political community, which is here being personified? The oracle seems to be recommending that Argos remain neutral in the war and not join the Spartan-led alliance of Greek cities that had formed to resist the Persians. After some failed diplomatic manoeuvring to obtain a joint leadership with the Spartans, the Argives duly decided to remain neutral.

I now want to turn to the most famous and best-documented of all Tibetan oracles, the Nechung oracle. Although, as we all know, human memory is frail, we do possess direct testimony for the nature of prophetic utterances and their role in decision-making that comes from someone who is in an unimpeachable position to know these things first hand. As I noted in my previous investigation, in his autobiography of 1990, Freedom in Exile, the fourteenth Dalai Lama of Tibet describes his own direct consultations of the Nechung oracle:

What I failed to discuss previously, however, is his account of the consultations that took place in 1959 and how the advice of the oracle was the determining factor both in his decision to flee Tibet and in the timing of that flight.

In November 1959 the situation in Tibet was critical. There was a serious danger that Chinese forces would attack the crowd of Tibetans who were protecting the Norbulingka palace in Lhasa where the Dalai Lama was residing. The Dalai Lama, who was only twenty-four years old at the time, turned to the Tibetan State Oracle for guidance:

Things deteriorated over the next few days and the Dalai Lama consulted the oracle a second time; but the advice was the same. When it became clear that the Chinese were indeed planning to attack the crowd and shell the Norbulingka palace, the Dalai Lama sent conciliatory messages to the Chinese general in charge and to the general's chief Tibetan collaborator in an attempt to buy time. He continues:

I have quoted this account at length because no paraphrase could do it justice. Like the Athenians, Spartans, and Argives who consulted Delphi on the eve of the Persian invasion of 480, the head of the Tibetan state sought the advice of an oracle in an extremely difficult and dangerous situation, a situation in which a wrong decision could have had fatal results. But the Dalai Lama has an advantage in that his most authoritative oracle travels with him and can be consulted at any time, whereas Greek communities had of necessity to send sacred envoys to oracular sanctuaries. There were, of course, local oracular shrines in or near many Greeks cities (such as those of Amphiaraus and Trophonius in Boeotia). But the authority of the oracles at the Panhellenic sanctuaries of Delphi and Dodona was much greater than that of other oracles, as well as of the numerous itinerant seers (manteis in Greek), whose primary techniques of divination were augury and extispicy. Although the Dalai Lama is recording these events long after the fact and his account is not impartial (given that the story is being told from the point of view of victims of imperialist aggression), his account is consistent with his lifelong reliance on divination.

Finally, at the time when I first explored Tibetan oracles, the only descriptions were written accounts, and those were mostly by outsiders. That has changed. In 2005 the director David Cherniack was granted permission to film a consultation by the Dalai Lama of Thupten Ngodrup, the thirteenth Nechung Medium and State Oracle of Tibet. Two other oracles were consulted at the same time: the Gadong oracle, who was as yet unable to speak (perhaps owing to contracting tuberculosis on moving to India), and the Temma oracle. In the documentary, which was released in 2010, the Dalai Lama is shown consulting these three oracles. It should be completely clear to any viewer that these spirit mediums are not ‘faking it’; they have entered an altered state of consciousness, regardless of whether the cause is a self-induced ‘psychosis’ or possession by ancient Tibetan spirits.Footnote ³³

After the consultation, the Dalai Lama gave a summary of what transpired: the Nechung oracle requested that the Dalai Lama live a long life and predicted that he would have successful negotiations with the Chinese in that year, as well as successful trips abroad. The Temma oracle, who is a woman in her twenties, also spoke of him having a long life, and even uttered poems on that topic. A few aspects of this consultation are noteworthy. The Dalai Lama does not use an intermediary, but places his head very close to that of the oracle and speaks to the oracle directly; the few attendants who are present merely support the oracle's trembling body; and at the very moment when the oracles are being spoken, the Dalai Lama himself writes with pen and paper. Although the texts of the oracles are not circulated, the Dalai Lama is apparently writing them down, or at least making notes, for his own use.Footnote ³⁴

What, if anything, can this tell us about Delphic practice? Analogy, of course, does not constitute proof.Footnote ³⁵ What the Tibetan example does prove, however, is that a human being, without the use of mind-altering stimulants, is able to enter a deep state of trance, and while in that state to utter intelligible, and sometimes enigmatic, prophecies in verse, and even, while in a trance state, to write these answers down.Footnote ³⁶ Moreover, in both cultures the oracle is a medium possessed by a supernatural power who speaks through him or her, as opposed to a shaman whose soul leaves the body in order to bring back messages from the land of spirits.Footnote ³⁷

A test case: Sparta consults Delphi in 432 BC

Let us now put these findings to the test. As I mentioned at the start, few topics are more controversial, in the study of Greek culture generally, than the nature of Delphic oracles. So the following example has been chosen carefully – yet even so it will not be unproblematic. During the fifty years that followed the unsuccessful Persian invasion of Greece in 480–479, the Greek world became divided between the leadership of Athens and that of Sparta. This bipolar structure was fragile, and a series of disputes ultimately led to open war between the two hegemonies. In 432 bc the Spartan assembly voted that the Athenians were in violation of the Thirty Years’ Peace that had been concluded in 446 bc (Thuc. 1.87–8). Despite modern assertions to the contrary, this was not a declaration of war per se, but merely a statement of record.Footnote ³⁸ They then summoned their allies to Sparta for a league congress where this resolution would again be put to the vote (Thuc. 1.119, 125).

During the interlude between the preliminary meeting of the Spartan assembly and the subsequent convening of their allies, the Spartans consulted the oracle of Apollo at Delphi. Thucydides, who is generally considered to be the most precise of all ancient historians, describes this consultation as follows:Footnote ³⁹

Some modern scholars have doubted the genuineness of this response because Thucydides introduces it with the words ‘it is said’.Footnote ⁴⁰ For our purposes, however, what is important is that within the world of Thucydides’ text the oracle's authenticity is simply assumed. Both the Corinthian ambassadors who address the meeting of Sparta's allies in 432 (Thuc. 1.123) and certain unnamed Athenians (2.54.4–5) in 430 refer to this oracular response as if it were a fact.Footnote ⁴¹ In any case, the ‘authenticity’ of the oracle is irrelevant to its obvious impact on Hellenic opinion. Even if the Spartans somehow fabricated, or elaborated, this answer to serve as propaganda for the coming war, we are still left with the distinct impression that most Greeks would have believed the Delphic priestess to be capable of such a response. In sum, although we cannot know whether this oracle is an utterance of the Pythia in the form that Thucydides reports it, we can say that Thucydides depicts characters in his history as accepting its authenticity at face value, and there are even indications in the narrative (see below) that Thucydides himself did so as well.

The form of the question that the Spartans put to the god, as well as the fact that they consulted the god at all in such circumstances, is not surprising. Both the question (‘Is it good to do x?’) and the partially ambiguous reply (‘you will succeed if you try as hard as you can’) are typical features of Delphic prophecy.Footnote ⁴² Delphic ambiguity in general has been doubted, even though Heraclitus and Aristotle tell us that answers could be ambiguous in nature.Footnote ⁴³ Ambiguity is an important randomizing device, the purpose of which is to establish resistance to human manipulation.Footnote ⁴⁴ This is true even in the case of the oracle delivered to various Spartans concerning the restoration of Pleistoanax (discussed above) that is recorded by Thucydides (‘otherwise they would plough with a silver ploughshare’ is ambiguous, even if easy enough to interpret).

In the response of 432 bc, Apollo was giving good advice and he could not be wrong (if the Spartans lost, it would be their own fault for not trying quite hard enough). Moreover, the form of the question indicates that the Spartans were genuinely asking for advice rather than simply seeking validation for what they had already decided to do,Footnote ⁴⁵ even if most modern scholars, accepting Thucydides’ portrayal of Spartan eagerness for war (1.23, 88, 118.2), find this impossible to imagine. What is really surprising – even remarkably so – is the additional statement that Apollo offered to assist the Spartans in the war. That is, this is remarkable in a specifically Greek context; in the ancient Near East proclamations of divine support to kings were commonplace. Such are the oracles delivered by the ecstatic prophetesses of Istar at Arbela in Assyria to the Assyrian kings Esarhaddon (681–669 bc) and Assurbanipal (668–626 bc), of which the following is a typical example:

Apollo did not disappoint. In the summer of 430, plague broke out in Athens. This plague, which has never been successfully diagnosed, wrought unparalleled death in Athens, killing between a quarter and a third of the population.Footnote ⁴⁷ Thucydides says that those who knew about the oracle given to the Spartans conjectured that the present events were in accord with it:

It would be an error to infer from this passage that only a very few Athenians knew about this oracle, since the Spartans would have been motivated to spread the word about it as widely as possible. We can be reasonably certain that the Athenians collectively must have believed that Apollo had caused this particular plague, just as he had caused the plague that struck the Achaean camp at the beginning of the Iliad.Footnote ⁴⁸ And that explains why Pericles classified the plague as being among things ‘sent by the gods’ (ta daimonia) when he addressed the people on the current state of affairs in the spring of 430 (Thuc. 2.64.2). Thucydides does not spell out the Homeric parallel, but he does not need to. He himself gives more space and more emphasis to the debate over whether an ancient oracle (which he calls an epos) had predicted that a pestilence (loimos) or a famine (limos) would come with a Dorian war. But what surely weighed most heavily on people's minds was the connection between the plague and Delphic Apollo's explicit promise of support to Sparta.

Lisa Kallet has provocatively argued that the passage quoted above shows that even Thucydides himself accepted the connection between the oracle and the plague. Referring to the second sentence of the quotation (2.54.5), she writes, ‘These comments move beyond endorsing the authenticity of the oracle and the conjecture that it might be true; they appear to support its fulfillment.’Footnote ⁴⁹ The syntax of the sentence, with its men/de construction, seems to support her interpretation (although it does not prove it). If this is correct, and if Thucydides believed that Delphic Apollo, through the voice of the Pythia, had proclaimed the oracle that he quotes, it tells us something important about the form, function, and acceptance of oracles. In this particular case, it tells us that it was credible to Thucydides and his contemporaries that Apollo himself, even if invisibly as at the beginning of the Iliad, was participating in this war.

Unfortunately, it is unrecoverable from Thucydides’ text whether the oracle given to the Spartans in 432 was in verse or in prose.Footnote ⁵⁰ Nevertheless, my hope is that divination in Tibet, both historically and as a living tradition in the Tibetan diaspora, may shed light on practices that are otherwise unverifiable in ancient Greece. In both cultures oracles were important and indispensable aids to decision-making by individuals and communities.Footnote ⁵¹ And in both cultures the consultant needed to interpret oracles, whether in prose or verse, which could be ambiguous. In a brief, and surprisingly unsympathetic, essay on Tibetan oracles, the anthropologist Prince Peter of Greece and Denmark makes one apropos remark in his description of the local oracle (Chos-dje) at Kalimpong, on the northern border of West Bengal. He observes, ‘Questions were put to him, and he answered them somewhat indistinctly and certainly ambiguously as seems to have been the practice of all good oracles ever since the Delphic Pythia.’Footnote ⁵²