Artemis: The Goddess Ephesus Had Before Her Name

Oldest mention of goddess Artemis: the Pylos tablets

The oldest mention of goddess Artemis is usually identified in Linear B records from the Mycenaean palace at Pylos, in the late Bronze Age. The two forms are a-te-mi-to and a-ti-mi-te. They are generally read as forms of her name, although an administrative entry cannot establish when people first began worshipping her. These tablets precede the Greek poetry and Ephesian statues discussed here.

Marie Louise Nosch prints both records in Approaches to Artemis in Bronze Age Greece, pages 24 to 26. PY identifies the findspot, Pylos; Es 650 and Un 219 identify different documents, rather than successive episodes of a myth.

The grammatical labels describe how the name works in the entry: the genitive means “of Artemis,” while the dative means “to” or “for Artemis.” Nosch’s printed transcription preserves these short fragments: PY Es 650, line 5, a3-ki-wa-ro , a-te-mi-to , do-e-ro (page 24); PY Un 219, line 5, a-ti-mi-te , O 1 (page 25). These are excerpts, not complete lines. The sign O is left as printed, without assigning it a certain commodity.

The word translated as “servant” or “slave” needs a separate caution. Nosch notes on page 25 that people described as belonging to a divinity commonly held plots of land; the label may identify an economic or religious affiliation rather than low social status. That reading does not justify projecting a modern category of slavery onto Aikiwaro.

Dartmouth's survey of Mycenaean religion lists A-TI-MI-TE as Artemis among the Pylos divine names. The same survey separates textual evidence from religious pictures whose figures cannot always be named. A goddess on a seal and a name on a tablet are different kinds of evidence; joining them requires an argument.

A possible third Pylos reference is much less secure. Nosch discusses PY Fn 837, line 5, where the word for priest is followed by the broken beginning a-ti[. Restoring the missing signs as Artemis is one option, but they could belong to a personal name instead. Counting that fragment as a third certain mention would conceal the damage to the record.

The script dates also need care. The Scuola Normale Superiore introduction to Linear B dates the documented script to the fourteenth and thirteenth centuries BCE and explains its use for an early form of Greek. That introductory range does not give either Artemis entry an exact calendar year. Michael Ventris deciphered the script in 1952; that is the date of the modern reading, not the date of the goddess.

For the earliest surviving written name, the Pylos records are the useful starting point. For the older Anatolian cult at Ephesus, the discussion below addresses a separate question about local worship and its later identification with Artemis. Neither kind of evidence supplies a date on which a goddess was invented.

A goddess Ephesus already had before the Greeks showed up

At Ephesus, the Greek goddess was identified with a local Anatolian cult. Historian Jona Lendering describes an older fertility goddess at the site in his account of Ephesian Artemis. Artimus is a Lydian form of the name, but the spelling alone cannot establish which cult or name came first. The Ephesian statue differs markedly from the bow-carrying virgin huntress of Greek poetry.

By the time the Lydian king Croesus funded a rebuilding of her temple around 560 to 550 BCE, the sanctuary was already old. The historian Herodotus recorded that at Ephesus there were "the oxen of gold and the greater part of the pillars" that Croesus dedicated, and one of those pillars survives today: a marble column drum, now in the British Museum, inscribed in archaic Greek with the words "King Croesus dedicated" the column. The temple he helped fund eventually earned a spot among the Seven Wonders of the Ancient World, a list that includes the Statue of Zeus at Olympia and the Hanging Gardens of Babylon, but almost nothing about the goddess worshipped inside it matched the Artemis described in Greek poetry.

The chest ornaments and competing interpretations

The most distinctive feature of the Ephesian cult statue is a row of smooth, egg-shaped protrusions covering the lower half of her chest, arranged in tight horizontal bands. For a long time the popular assumption was that these were breasts, a literal image of a many-breasted mother goddess. Scholars studying the surviving copies point to a problem with that reading: many copies show smooth ovals with little anatomical detail, although LiDonnici notes a few examples with nipples.

In 1979, the archaeologist Gerhard Seiterle proposed a different answer in the journal Antike Welt, after presenting the idea a year earlier at the International Congress for Classical Archaeology in London. He argued the ornaments were bull testicles, hung on the statue as fertility offerings, and built a physical model of the statue with real bull scrota attached to make his case. The theory caught on. Classicist Walter Burkert and archaeologist Brita Alroth both accepted versions of it, and it has been repeated in popular accounts of the goddess ever since. Not every specialist agreed: archaeologist Robert Fleischer rejected Seiterle's reconstruction, and other scholars have proposed the objects were gourds or dates instead, plant-based fertility symbols known elsewhere in the ancient Mediterranean.

A scholarly reconsideration appeared in 1992, when Lynn LiDonnici published "The Images of Artemis Ephesia and Greco-Roman Worship: A Reconsideration" in the Harvard Theological Review. LiDonnici argued that Seiterle's bull-testicle theory rested on thinner evidence than its popularity suggested, and that the objects were likely not originally intended to represent any single, literal thing. Her research turned up a separate wrinkle: by the Roman period, some Roman-period writers described the ornaments as breasts, whatever the original sculptors intended centuries earlier. Those written descriptions do not establish what the original sculptors intended, and LiDonnici's paper documents why the popular identification requires caution.

Ephesus told Rome she wasn't even born on Delos

Birth traditions varied. Homeric Hymn 9 places Artemis in Ortygia and Apollo in Delos; Tacitus reports a familiar belief that both were born on Delos. Ephesus challenged the Delian account, and made its case in the most consequential venue available: the Roman Senate. In 22 CE, the emperor Tiberius ordered cities across the Greek world that claimed asylum rights for their temples, meaning fugitives and debtors who reached the sanctuary could not be touched, to send envoys to Rome and justify those privileges before the Senate. The historian Tacitus recorded the proceedings in his Annals, and Ephesus went first.

According to Tacitus, the Ephesian envoys "declared that Diana and Apollo were not born at Delos, as was the vulgar belief. They had in their own country a river Cenchrius, a grove Ortygia, where Latona, as she leaned in the pangs of labour on an olive still standing, gave birth to those two deities." It was not an idle claim. Asylum rights were valuable and contested, and Ephesus was rewriting one of the best-known birth stories in Greek mythology, directly contradicting Delos's own version, in order to win a legal argument in front of Roman senators nearly eight centuries after Homer.

The temple whose sanctity Ephesus was defending that day did not survive antiquity. An earlier version burned down in 356 BCE, its rebuilt successor was sacked in a Gothic raid in 262 CE, and today the site holds a single reconstructed column standing over a marshy field of scattered stone, a startling comedown for a Wonder of the Ancient World whose birthplace claim Roman senators had once taken seriously enough to hear formally argued.

The huntress underneath the layers

Set the Ephesian cult aside and the Artemis of mainstream Greek myth is a more familiar figure: daughter of Zeus and the Titaness Leto, twin sister of Apollo, and goddess of the hunt, the wilderness, and wild animals. Ancient sources describe her as one of the virgin goddesses of Olympus, sworn never to marry, and closely tied to childbirth and the protection of young girls even while remaining unmarried herself. She appears consistently as one of the Twelve Olympians in a way some other seats on that roster did not, a rare point of ancient agreement about a goddess whose regional cults otherwise diverged as sharply as the one at Ephesus.

To the Romans she was Diana, a name that carried the same core attributes across the empire: hunting, the moon, and the protection of wild places. It is Diana, not Artemis, who appears in Tacitus's account of the Ephesian Senate hearing. The goddess Rome heard argued over in 22 CE and the goddess whose statue still puzzles archaeologists at Ephesus were, by that point, treated as the same figure by everyone involved, even though her Anatolian origins and her Olympian mythology never fully merged into one consistent story.

For a literary example of a plant carrying different meanings in different settings, the willow symbolism guide compares named texts and farewell customs.