Where did I get my name?
I was named after a goddess whose temple you passed on the way to the oracle. I am not an oracle. I am the road a question travels — and the room where it is discussed.
Pausanias arrives at Delphi
Some 1,850 years ago the Greek travel writer Pausanias climbed the slope of Mount Parnassus to Delphi, home of the most famous oracle of the ancient world, and wrote down everything he saw. He was a careful man: he always kept apart what he had seen himself from what he had been told. When he writes “it is said that”, he means exactly that.
Near the sanctuary Pausanias saw the temples in a row, and one of them was dedicated to Athena Pronoia — Athena Forethought. The name was read in two ways in antiquity. Pronaia meant “standing before the temple”: the goddess stood ahead of the main temple, the traveller’s first stopping place. Pronoia meant foresight, providence and forethought. I took my name in both senses. I am a place you pass through before understanding is born — and I do not hand out answers ready-made, but arrange the journey on which they are weighed together.
The traveller’s road to the oracle
No one at Delphi walked straight up to the oracle. The road had several stages, and each had its task.
The first stop was the spring of Kastalia by the entrance to the sanctuary. Pausanias praises its water as sweet to drink and pleasant to bathe in. Everyone purified themselves there — the enquirers, the priesthood and the Pythia herself. Kastalia foretold nothing. It made sure that nothing unclear was carried into the temple.
Carved into the forecourt of the temple were the maxims of the sages, two of which have outlived the rest: Know thyself and Nothing in excess. The first reminded the enquirer of the limits of their own views; the second, that a good question is precisely bounded.
In the innermost chamber, the adyton, sat the Pythia on her tripod. Her words were not her own but the god’s, and so her name was not recorded — the message carried the weight, not the one who spoke it. Pausanias tells that a second spring, Kassotis, ran underground into the chamber, and that its water made the women there seers. The spring was named after a nymph of Parnassus.
And then came the decisive stage of the story, the one so often forgotten: the god gave the signs, but a human made the meaning. The Pythia’s words were written down by priests, and the enquirer took them home to be interpreted. Herodotus tells how, on the eve of the Persian war, the Athenians received an answer in which a “wooden wall” would save the city. The assembly argued over whether it meant the palisade of the Acropolis or the fleet. Themistocles argued for the fleet — and the victory at Salamis grew out of that interpretation, not out of the oracle’s words. Delphi even had a room of its own for such talk: the Lesche, the “place of conversation”, where people gathered to speak of matters of importance.
The same name, another age
In the early 1950s researchers at the RAND Corporation in the United States, Olaf Helmer and Norman Dalkey, developed a method in which a group of experts was asked the same question about the future several times over: each answered independently first, then saw the others’ answers anonymously and could revise their own position. It is said that the name Delphi was coined by the philosopher Abraham Kaplan — and that Dalkey later thought it unfortunate, because it hinted at something oracular and faintly occult. The name stayed nonetheless, and under it the method spread from technology forecasting to health care, education and policy work.
In Finland, Delphi has taken a form of its own. Here the emphasis shifted early from seeking consensus to examining reasons: in argumentative Delphi the interesting thing is not what the experts agree on but why they disagree. When the method moved online, the rounds became conversation and the panel a community that reads each other’s reasoning and answers it. I am a continuation of that tradition.
Ancient Delphi and Delphi today
The parallel is more than a pretty name. Every stage of the Delphic process has a counterpart in what Delphi foresight does today — and at every point there is also a turn where the modern method does otherwise.
At Delphi one Pythia spoke. In a Delphi panel a group of experts unknown to one another speaks, and the answer is a distribution, not a sentence. Unanimity is not the aim; the aim is that those who disagree know why they disagree.
The Pythia’s name was not recorded, because the message was the god’s. In a panel the panelist’s name is not shown, because an argument must work through its content, not through the standing of the one who makes it. Anonymity protects the same thing in both: the separation of the message from the authority of its speaker.
The oracle’s answer was often deliberately open to interpretation, and the burden of interpreting it fell on the enquirer. Modern Delphi does the opposite: it requires every position to be justified and makes disagreement legible. Delphi never said when; modern Delphi asks precisely that.
The oracle was asked once. A Delphi panel returns to the same question round after round, and between rounds each participant reads the others’ reasoning and may change their mind. Know thyself still holds as guidance here: revising one’s position is the core of the method, not a weakness of it.
In antiquity one washed at the spring before entering the temple. At my threshold the panelist is met by the AI helper Kastalia, who talks with them before the first round: opening up the questions and helping put a thought into words. Kastalia does not rank views and claims nothing on anyone’s behalf. Nothing in excess is her rule: every thesis sharp, none of them leading.
The spring in the innermost chamber fed seeing, out of sight. In the analysis phase a second helper, Kassotis, is at work, opening the panel’s talk layer by layer: arguments, the moves of the dialogue, tensions, weak signals. But Kassotis is running water, not the oracle’s voice. It utters no prophecies. Synthesis and conclusions belong to the researcher — just as at Delphi the god gave the signs, but a human made the meaning.
The Athenians’ wooden wall was settled in debate, not in the chamber. My rounds are that Lesche: the room where reasons meet one another and interpretation is made together.
The loveliest detail in the story is that Kastalia and Kassotis draw their water from the same underground stream of Mount Parnassus. One body of knowledge, two springs: one rises at the gate for all to see, the other flows within, in support of the researcher. The same rule binds both: the AI carries the water; the seeing is human work.
Why the name is a good one
Dalkey feared that the name Delphi promised prophecies. My name promises none. Pronoia was not the oracle but the temple before the oracle — the place where the traveller prepared to ask well. That is a more honest promise for futures research than a seer’s voice: I do not see the future, but I help a group of experts see further together than any of them could alone.
Welcome to the spring.
The historical details rest on Pausanias, Periēgēsis Hellados (Description of Greece), book 10: the temple of Athena Pronoia (10.8.6), Kastalia (10.8.9–10), the maxims of the temple (10.24.1), Kassotis (10.24.7) and the Lesche (10.25.1). The story of the wooden wall is from book seven of Herodotus’ Histories (7.141–143). On the origins of the Delphi method at RAND, see Dalkey & Helmer (1963) and Dalkey’s own assessment of the name (1968).