Yet despite this understanding of the raving Pythia as a consequence of cultural mistranslation and subsequent elaboration, and despite the fact that no source mentions it before the first century BC, the picture of the Pythia breathing in vapors from a chasm below her tripod has always been the dominant model for understanding how the oracle at Delphi functioned. To such an extent that finding the mechanism of the vapors was originally regarded as the litmus test for successful archaeological investigation at Delphi. The original excavators of the site were extremely disappointed not to find a chasm below the temple—they felt almost cheated by the “deception” of the literary sources.36 The stakes were understandably high: at the time of Delphi’s excavation in the 1890s, interest in the oracle, and in psychic research more generally, could not have been stronger. In 1891 the burlesque opera Apollo, or The Oracle at Delphi played to great acclaim on Broadway. In the same year, John Collier painted his famous Priestess of Delphi in which a sensual priestess breathes in vapors from her tripod over a chasm (see plate 4), and the Society of Psychical Research was started by Cambridge academics and published its first volume examining the oracle at Delphi. In the wake of the disappointing excavations, thus, there was a feeling that the ancient sources had lied. The scholar A. P. Oppé in 1904 in the Journal of Hellenic Studies argued that the entire practice at Delphi was a farce, a sham, put on by the priests of Apollo, tricking the ancient world.37 Others sought different explanations for the Pythia’s madness: they focused on the laurel leaves, and suggested the Pythia had been high from eating laurel. One German scholar, Professor Oesterreich, even ate laurel leaves to test the theory, remarking disappointedly that he felt no different.38 Others opined that the answer relied not in some form of drug, but in psychology. Herbert Parke and Donald Wormell argued in the 1950s that the Pythia, in the heat of the moment after so much preparation on the particular day of consultation, and after so many years perhaps involved with the temple as one of the women guarding the sacred flame, would have found herself in an emotionally intense relationship with the god, and could easily have fallen victim to self-induced hypnosis.39 More recently, scholars have employed a series of anthropological approaches to understand belief in spirit possession, and applied these to how the Pythia may have functioned.40
The chasm idea, however, was hard to forget. The Rev. T. Dempsey in the early twentieth century argued that perhaps, just as Plutarch had suggested the oracle worked less in his day thanks to less pneuma, the chasm had, in modern times, completely closed up.41 Others sought even more ingenious explanations for how the vapors had been created without a chasm, including one in which the Pythia herself descended to a room below her tripod to light a fire that produced the smoke (possibly from the hemp plant) she then breathed in, as if it were vapors from the god.42 This explanation, coupled with the analysis of a particular stone block filled with mysterious holes and grooves, thought to be that on which the tripod and omphalos were positioned and through which the vapors arose (still on view at Delphi but now recognized as a stone later recut as an olive press), crystallized the sense that the ancients had “bought” a hoax for more than a thousand years at Delphi.43
More recently, the debate over the presence of inspiring vapors at Delphi has re-emerged, thanks to a reassessment of Delphi’s geology.44 Analysis by the geologist Jelle De Boer and the archaeologist John Hale through the 1980s and 1990s led, in the first decade of the twenty-first century, to their publishing evidence for two major geological fault lines crossing at Delphi (one running east-west, the other north-south) directly underneath the temple of Apollo (see plates 1, 2; figs. 0.1, 0.2). At the same time, they argued that the bedrock beneath the temple was fissured, which would allow for small amounts of gas to rise up through the rock, despite the absence of a chasm. This gas originated from the bituminous (full of hydrocarbons) limestone naturally occurring in this area, which would have been stimulated to release its gas by shifts in the active fault lines beneath. Testing both the travertine (itself a product found only in active fault areas) and the water beneath the temple at Delphi, they found ethane, methane, and ethylene, which had been used as an anaesthesia in the 1920s, thanks to its ability to produce a pleasant, disembodied, trancelike state. They postulated that the geology of Delphi could thus have produced enough of these potent gases to, within an enclosed space like the adyton, put the Pythia into a trancelike state.45 The ancients may not have been lying after all.
This research created huge excitement in public and academic circles, but in reality, while fascinating, it still did not really solve the problem. Even if intoxicating gases were produced in the temple at Delphi, and these gases did “inspire” the Pythia (despite that none of the sources before the first century BC point to this as the method of inspiration), how did the answers she gave, even if massaged and shaped by the priests of Apollo, remain suitable, useful, right enough for the oracle to continue as a valid institution for over one thousand years? Or as Simon Price, a scholar with a reputation for pushing straight to the heart of a problem, put it: “Why was it that the sane, rational Greeks wanted to hear the rantings of an old woman up in the hills of central Greece?”46 To understand this, we must put the process of oracular consultation at Delphi in its wider religious context, and think more carefully about the way in which the oracle was perceived.
Oracles were an essential, and respected, part of the Greek world. They were also everywhere you looked. Scholarship has demonstrated the vast array of oracular sanctuaries on offer, which varied from the Pythian priestess at Delphi to the consultation of the rustling of leaves of Zeus’s sacred tree at Dodona in northern Greece, to consultation with spirits of the dead, like at Heracleia Pontice on the Black Sea (see map 1).47 Sometimes even the same god could have very different forms of oracular consultation at his different sanctuaries: so while Apollo Pythios (as Apollo was worshiped at Delphi) had the Pythia at Delphi, at the sanctuary of Apollo Pythaios in Argos, his priestess took part in nocturnal sacrifices and drank the blood of the sacrificial victims as part of her inspiration to prophesize.48 But this form of divination (putting a question to a god through a priestly representative) was also just one of the forms of divination available in ancient Greece. Another was the reading of signs from particular natural events and actions and interpreting them in relation to a particular question. Just about everything could be read: the flight of birds (although not the movement of fish), patterns of words, sneezing, entrails, fire, vegetables, ripples on water, reflections in mirrors, trees, atmospheric phenomena, stars, as well as randomized dice, beans and other forms of “lot” oracle. In addition, there was a host of wandering chresmologoi or manteis (“oracle-tellers” or “seers”) who could be engaged on the street in any major Greek city for consultations, which could be conducted in a variety of ways, from the reading of appropriate oracle responses in books of oracles to connecting with dead spirits.49
Key here is that the Greek world was filled with a “constant hum” of divine communication.50 It was a system used by all levels of Greek society, and as well, it was a system in which everyone had their “preferred” form of communication, which could alter depending on the type and importance of the question to be asked. The Athenian general Nicias in the fifth century BC had his own personal seer as did many other military commanders. In the sixth century BC, Peisistratus, the Athenian tyrant, never consulted Delphi, but liked using chresmologoi. Alexander the Great in the fourth century BC liked his manteis to come from Asia Minor. Such seers could be incredibly well respected: Lampon was a seer in the fifth century BC but also a friend of the famous general and statesman Pericles and responsible for the foundation of Thurii in South Italy. Nicias’s chief seer, Stilbides, was also one of his top soldiers.
The importance of divination does not mean, however, that the oracular system was never mocked in Greek culture. The consultation of oracles was lampooned in Greek comedy: in Aristophanes’ Knights and Birds, for example, oracle sellers are figures
of fun. The strength of their connection with the divine too could be questioned. Euripides, in a fragment of an otherwise lost play (Frag. 973N), wrote “the best seer is the one who guessed right.” Sometimes too their usefulness could be questioned. Xenophon, in the fourth century BC, argued that divination became useful only when human capacity ended.51 We shall see in the coming chapters instances wherein even the oracle of Apollo at Delphi was said to have been bribed and to have become biased, or was treated with circumspection by even its most loyal consultants. But all these instances represent an aberration from the norm, an aberration that did not in the long term shake belief in the system as a whole, a system that continued to speak of divination as a useful and real connection to the gods.
It is difficult for us to understand this kind of mindset today. In the 1930s, the anthropologist Sir Edward Evans-Pritchard tried to understand the Greeks’ acceptance of oracles by observing the Azande community in Central Africa, who used a form of poison-chicken oracle to resolve disputes and make difficult community, as well as individual, decisions. His research showed that within the framework of a particular culture and life in which everyone agreed that the poison-chicken oracle was a proper and respected way to choose a solution, it was as good a means as any to help run a community.52 But where does this leave us with the Pythia? On the one hand, we have to imagine ourselves into a society in which oracular connection with the divine was a commonly accepted cultural activity into a world that was believed to be controlled by the gods. Those gods could be for you or against you, and so it made sense to make every effort not only to appease them with offerings, but to use a variety of methods of divination to find out what they had in mind for the future. At the same time, the gathering of tradition about the power and importance of an oracle like the Pythia at Delphi, coupled with the prolonged rituals encountered before the consultation, would have helped to ensure belief in the process and its outcome. This does not mean people had to know exactly how the Pythia connected with the god: even Plutarch, himself (as indicated above) a priest at Delphi in the first century AD, was content to relate the argument and debate among his friends, who each had their own ideas about how the inspiration took place and why it happened less frequently in their time. Key is that—however it happened—there was a belief in a connection between the divine and human world through the Pythia.
At the same time, a number of other factors must be taken into consideration to understand how the Pythia retained her reputation for over one thousand years. The first is the kind of information sought from the Pythia. The questioning of the Pythia in the sixth century BC by King Croesus of Lydia in Asia Minor is often thought, because it is so well known, to be typical of how consultants put their questions to the oracle. But one of the things Herodotus, our major source for this encounter, is likely indicating is that Croesus’s encounter with Delphi shows how little this “non-Greek” understood about Greek culture. As we shall see in later chapters, Croesus’s direct question to oracles all around the Mediterranean in order to find out which was the “best” (the “what am I doing at this moment?” question) was, in fact, a very unusual type of question to ask an oracle. Not because it was a test of present knowledge, but because it involved such a direct request for information. Very rarely, it seems, did consultants ask the oracle direct questions about the future (so Croesus’s second question about whether he would win the war against King Cyrus of Persia was again an odd form of question).53 Instead, most questions put to the oracle seem to have been in the form of “would it be better and more profitable for me to do X or Y?” or else, “to which god shall I pray before I do X?”54 This is to say, consultants presented problems to the Pythia in the form of options, or rather sought guidance for how their goals might come about, rather than asking directly what would happen in the future.
Such a process, focused around guidance rather than revelation, underlines the kind of occasions on which people chose to consult the oracle at Delphi, particularly instances in which individuals, or a community, were having trouble reaching a consensus over which of a particular set of potential actions to take. It is this usefulness of an oracle at moments of community indecision that has been, as we shall explore in the following chapters, thought critical in turning the Delphic oracle into such a well-known and important institution in the Greek world in the late eighth and seventh centuries BC, the period that bore witness to the tectonic forces of community creation that laid the bedrock for the landscape of the classical world.
Second to be considered is the question of the Pythia’s response. Scholars have long remarked on the perfect hexameter verse responses reported in the literary sources as coming straight from the Pythia. There has been substantial doubt cast over her ability to utter directly such poetry, and instead scholars have pointed the finger at her priests as the ones who constructed the responses. At the same time, scholars have pointed to Delphi as a developing “information center,” since it was one of the few places in the ancient world people were going to from all over on a regular basis, and bringing with them information about their homelands. As a result, a picture has formed of priests who were “plugged in” to the information “hub” at Delphi and thus able to give better-informed guesses about which options were better and outcomes more likely (and have more ability to put those responses into verse). I have some sympathy with elements of this picture: no doubt Delphi was, especially in the sixth through fourth centuries BC, something of an information center. We can only imagine the exchange of information going on during the nine days of the year that consultants from all over the Greek world could turn up (not to mention the days either side they had to wait, or at the Pythian games, or while constructing monuments or while visiting the sanctuary to see it in all its splendor), and the degree to which this information fed back into the consultation system. But at the same time, this can only at best explain the success of Delphi once it had become a success. When Delphi’s oracle first began to be consulted, there was precious little more information going to Delphi than elsewhere.
More fundamental in understanding the Pythia’s success from the outset is the in-built ambiguity of her responses, in part as a direct result of the nature of the questions asked. As we saw above, questions normally came in the form of “is it better and more profitable that I do X or Y” or “which gods should I pray to before I do X.” As a result, if the Pythia replied “do X” or “it is better and more profitable to do X,” and X proved disastrous, people would still never know how bad option Y might have been in comparison. As well, even if the oracle replied telling you to pray to a particular god, this indicated only a prerequisite action to ensure a chance of success; it did not guarantee a good result: praying to a god came with the understanding that there was no guarantee the god would listen in return. The comparative form of the question, and the unequal nature of the relationship between human and divine, ensured that it was impossible for the oracle to be categorically wrong in its response.
Sometimes that ambiguity seems to have been taken a step further in the form of a more complex answer, which in turn demanded a process of further interpretation from the consultant. King Croesus’s inappropriate question about waging war, for example, received a very ambiguous Delphic answer: “Croesus, having crossed the river Halys, will destroy a great empire.” The response doesn’t make clear whether it will be Croesus’s empire or that of his enemies. In Herodotus’s narrative, Croesus took it to mean his enemies’, but it turned out to be his own (Croesus lost his kingdom as a result of losing the battle). Once again, the oracle, thanks to the ambiguity of the response could not be argued to have been wrong: it was Croesus who had chosen to misinterpret the Pythia’s response.55
Plutarch, in the first century AD, commented on this well-known ambiguity in oracular responses of the past, noting that, in his day, responses tended to be more direct, but that in olden times, ambiguous replies were necessary because they protected the Pythia from the powerful p
eople who came to consult her: “Apollo, though not prepared to conceal the truth, manifests it in roundabout ways: by clothing it in poetic form he rids it of what is harsh or offensive, as one does with brilliant light by reflecting it and thus splitting it into several rays.”56 Not for nothing was the god Apollo often known as Apollo Loxias, Apollo “the ambiguous one.”
Moreover, and crucially, an ambiguous response demanded further debate and deliberation from the consultant and his city. What often began as an issue the community could not decide on, was referred to the Delphic oracle for further enlightenment, and was thus often sent back to the community for continued deliberation about the problem, but with the fresh information/indication and momentum toward making a decision in the form of the god’s response. As classicist Sarah I. Johnston puts it, consulting the oracle at Delphi “extends [consultants’] agency; it puts new reins in their hands.”57 Consulting the Pythia thus did not always provide a quick answer to a straightforward question, but rather paved the path for a process of deliberation that allowed the community to come to its own decision.58 Indeed the very process of deciding to consult Delphi, sending representatives to ask the question, waiting for one of the rare consultation days and potentially more than one if it was a very busy time, and then returning with a response that became part of a further debate meant a decision to consult Delphi substantially slowed down the decision process and gave the community much longer to mull over the issue.