DODONA, in Epirus, the seat of the most ancient and venerable of all Hellenic sanctuaries. Its ruins are at Dramisos, near Tsacharovista. In later times the Greeks of the south looked on the inhabitants of Epirus as barbarians; nevertheless for Dodona they always preserved a certain reverence, and the temple there was the object of frequent missions from them. This temple was dedicated to Zeus, and connected with the temple was an oracle which enjoyed more reputation in Greece than any other save that at Delphi, and which would seem to date from earlier times than the worship of Zeus; for the normal method of gathering the responses of the oracle was by listening to the rustling of an old oak tree, which was supposed to be the seat of the deity. We seem here to have a remnant of the very ancient and widely diffused tree-worship. Sometimes, however, auguries were taken in other manners, being drawn from the moaning of doves in the branches, the murmur of a fountain which rose close by, or the resounding of the wind in the brazen caldrons which formed a circle all round the temple. Croesus proposed to the oracle his well-known question; Lysander sought to obtain from it a sanction for his ambitious views; the Athenians frequently appealed to its authority during the Peloponnesian War. But the most frequent votaries were the neighbouring tribes of the Acarnanians and Aetolians, together with the Boeotians, who claimed a special connexion with the district.

Dodona is not unfrequently mentioned by ancient writers. It is spoken of in the Iliad as the stormy abode of Selli who sleep on the ground and wash not their feet, and in the Odyssey an imaginary visit of Odysseus to the oracle is referred to. A Hesiodic fragment gives a complete description of the Dodonaea or Hellopia, which is called a district full of corn-fields, of herds and flocks and of shepherds, where is built on an extremity (ἐπ᾽ ἐσχατίῃ) Dodona, where Zeus dwells in the stem of an oak (φηγός). The priestesses were called doves (πέλειαι) and Herodotus tells a story which he learned at Egyptian Thebes, that the oracle of Dodona was founded by an Egyptian priestess who was carried away by the Phoenicians, but says that the local legend substitutes for this priestess a black dove, a substitution in which he tries to find a rational meaning. From inscriptions and later writers we learn that in historical times there was worshipped, together with Zeus, a consort named Dione (see further Zeus; Oracle; Dione).

The ruins, consisting of a theatre, the walls of a town, and some other buildings, had been conjectured to be those of Dodona by Wordsworth in 1832, but the conjecture was changed into ascertained fact by the excavations of Constantin Carapanos. In 1875 he made some preliminary investigations; soon after, an extensive discovery of antiquities was made by peasants, digging without authority; and after this M. Carapanos made a systematic excavation of the whole site to a considerable depth. The topographical and architectural results are disappointing, and show either that the site always retained its primitive simplicity, or else that whatever buildings once existed have been very completely destroyed.

To the south of the hill, on which are the walls of the town, and to the east of the theatre, is a plateau about 200 yds. long and 50 yds. wide. Towards the eastern end of this terrace are the scanty remains of a building which can hardly be anything but the temple of Zeus; it appears to have consisted of pronaos, naos or cella, and opisthodomus, and some of the lower drums of the internal columns of the cella were still resting on their foundations. No trace of any external colonnade was found. The temple was about 130 ft. by 80 ft. It had been converted into a Christian church, and hardly anything of its architecture seems to have survived. In it and around it were found the most interesting products of excavation—statuettes and decorative bronzes, many of them bearing dedications to Zeus Naïus and Dione, and inscriptions, including many small tablets of lead which contained the questions put to the oracle. Farther to the west, on the same terrace, were two rectangular buildings, which M. Carapanos conjectures to have been connected with the oracle, but which show no distinguishing features.

Below the terrace was a precinct, surrounded by walls and flanked with porticoes and other buildings; it is over 100 yds. in length and breadth, and of irregular shape. One of the buildings on the south-western side contained a pedestal or altar, and is identified by M. Carapanos as a temple of Aphrodite, on the insufficient evidence of a single dedicated object; it does not seem to have any of the characteristics of a temple. In front of the porticoes are rows of pedestals, which once bore statues and other dedications. At the southern corner of the precinct is a kind of gate or propylaeum, flanked with two towers, between which are placed two coarse limestone drums. If these are in situ and belong to the original gateway, it must have been of a very rough character; it does not seem probable that they carried, as M. Carapanos suggests, the statuette and bronze bowl by which divinations were carried on.

The chief interest of the excavation centres in the smaller antiquities discovered, which have now been transferred from M. Carapanos’s collection to the National Museum in Athens. Among the dedications, the most interesting historically are a set of weapons dedicated by King Pyrrhus from the spoils of the Romans, including characteristic specimens of the pilum. The leaden tablets of the oracle contain no certain example of a response, though there are many questions, varying from matters of public policy or private enterprise to inquiries after stolen goods.

The temple of Dodona was destroyed by the Aetolians in 219 B.C., but the oracle survived to the times of Pausanias and even of the emperor Julian.

See C. Wordsworth, Greece (1839), p. 247; Constantin Carapanos, Dodone et ses ruines (Paris, 1878). For the oracle inscriptions, see E. S. Roberts in Journal of Hellenic Studies, vol. i. p. 228.  (E. Gr.)