Dancing Places and Doves: Place-names and their Epithets in the Catalogue of Ships

[495] and Arkesilaos and Prothoenor and Klonios

[496] who inhabited Hyria and rocky Aulis

[497] and Skhoinos and Skolos and many-peaked Eteonos

[498] and wondrous Graia and Mykalessos with its broad dancing places

[499] and those who inhabited Harma and Eilesios and Erythrai

[500] and those who held Eleon and Hyle and Peteon

[501] Okalea and Medeon the well-built citadel,

[502] Kopai and Eutresis and Thisbe of the many doves

In my previous post I explored the notion of "walk-on characters" in the Iliad, those characters who appear in the narrative seemingly only to die or who fade from view after only a brief mention. In that post I argued that such characters are as integrated into the traditional system in which the Iliad and Odyssey were composed as more prominent characters, and that the Catalogue of Ships in particular functions much like an index to the totality of the epic tradition. I conclude by quoting Mabel Lang (In Carter and Morris, The Ages of Homer [1995], 161): "Speculative in the extreme? Yes, but sensible if ones sees the Catalogue of Ships not as a survey of actual political geography, but as a poetic attempt to list as many famous heroes as might possibly have fought in the Trojan War, although in the Iliad at least, several have little or no part. These heroes must have been known to the bards, complete with epithets and epitheted place-names, from their local exploits."

In this post I want to explore the much vexed question of the "epitheted place-names" to which Lang refers. Was there a real place named Mykalessos and did it actually have broad dancing places? Why is Medeon a "well built citadel" but Okalea is not described at all? Did Thisbe really have many doves, and if so how do we know? How did such places and epithets make it into the epic tradition to begin with? I don't expect to be able to thoroughly answer all of these questions in one blog post, but I would like to at least make a start by exploring further the epithets used of Mykalessos and Thisbe. In so doing I will build on previous work in which I have discussed the poetics of various other kinds of noun-epithet combinations, and here as always I am much indebted to the work of Milman Parry, whose first doctoral thesis was entitled L’Épithète traditionelle dans Homère; Essaie sur un problème de style homérique (= The Traditional Epithet in Homer). (See especially Dué and Ebbott 2010 ad 10.3 and 10.283 for definitions and the history of scholarship on epithets.)

Lang's work suggests that the walk-on characters in the Iliad are in fact local hero heroes whose deeds in the Trojan War and/or other epic narratives would have been sung in particular places. At some point what had previously been local songs came to be performed more widely and by other, non-local singers, or at the very least, their heroes came to be incorporated into a wider narrative tradition that eventually resulted in our Iliad. By making it into the Iliad, these local heroes became part of a Panhellenic poetic tradition at some distance removed from the local songs from which they originated. If this conceptualization is correct, we have to understand that different singers and different audiences may have known more or less specific information about these heroes and the towns from which they hailed, depending on their familiarity with the local traditions that formed these characters' back-stories. And yet, as a notional totality at least, the full biography of these more local heroes was at some point known to the epic tradition as a whole.

Likewise, this notional totality that I am invoking had a broad knowledge of the geography of Greece, though it is clear that much as the dialect of the Homeric epic evolved to incorporate Aeolic and Ionic and even Attic elements, so too did new places come into the system, sometimes (no doubt) at the expense of other places which fell out of circulation. I have argued with reference to the character of Briseis that some audiences (e.g., archaic or earlier, Aeolic) would have had a clear understanding of her own unique life story, while other, later, and more Panhellenic audiences would have understood her story only paradigmatically (Briseis as a typical captive woman from a town sacked by Achilles, the daughter or wife of the local king). (See Dué 2002.)

A place like Mykalessos or Thisbe then may have had syntagmatic associations, that is details that were true and specific to the actual towns, as well as paradigmatic associations (characteristics that they share with other epic places, details that may or may not have had anything to do with their "real" geographical features). Each place may have at one time had a set of particular epithets and formulas that were used of that place in particular by bards familiar with that place. Some or all of that formulaic language may have at one point become a part of the larger, more Panhellenic epic tradition, but not all of what came into the system stayed in the system. As the tradition and its formulaic diction evolved so too did the poets' and audiences' understanding of those places evolve.

It is this evolution that explains at least in part why the political geography of the Catalogue of Ships cannot be tied to one particular era. Some towns mentioned in the Catalogue (e.g., Eutresis in 2.502) were uninhabited after the end of the Bronze Age (though Eutresis was reoccupied beginning in the 6th Century BCE; see Simpson and Lazenby, The Catalogue of Ships in Homer's Iliad [1970], 27), suggesting a Bronze Age date for the Catalogue (as argued by Simpson and Lazenby), but others, such as Sparta, were not particularly important in the Bronze Age and flourished only in later times. (Sparta seems to have displaced the Bronze Age Therapne completely; see O. Dickinson, "Catalogue of Ships," in The Homer Encyclopedia [ed. M. Finkelberg, 2011].) Athens, so important a city from archaic times onwards, was a relatively minor fortified citadel in the Bronze Age, which might explain why Athens is barely featured in our Iliad and might be another indication of a Bronze Age date for the Catalogue. And yet the geographical evidence preserved in the Bronze Age Linear B tablets often does not match up with that of the Homeric texts. (An example would be Pylos, whose territory as revealed by the Linear B tablets does not match what is described in the Catalogue; see again Dickinson 2011.)

The work of E. S. Sherratt noted in my last post suggests that we should not be looking for a single political reality reflected in the Catalogue of Ships, and that it would be fruitless to attempt to separate Bronze Age geographical details from later ones. Elements from more than one reality entered the system of formulaic diction and were seamlessly integrated over time. At the same time, as this process was on-going, the reality of any particular location (its particular natural features, precise geographical location, etc) faded in importance, and instead its poetic/epic identity superseded it. A place like Mykalessos was understood within the tradition to have broad dancing places, and it may well have had them at one time, but poets of later eras need not have known whether or not this was this was the case. Within the poetic tradition, Mykalessos had broad dancing places.

Mykalessos was not the only place to have broad dancing places, however; Sparta and two other cities as well are described this way in the Odyssey (Elis 4.635, Thebe 11.265, Sparta 13.414 and 15.1). Archaeological remains are not sufficient to tell us whether these places actually had broad dancing places. A place that certainly did have them was Knossos on Minoan Crete, which is intriguingly remembered on the shield of Achilles as having a dancing place made by Daidalos for Ariadne (Iliad 19.590-592):

ἐν δὲ χορὸν ποίκιλλε περικλυτὸς ἀμφιγυήεις,

τῷ ἴκελον οἷόν ποτ᾽ ἐνὶ Κνωσῷ εὐρείῃ

Δαίδαλος ἤσκησεν καλλιπλοκάμῳ Ἀριάδνῃ.

And on it a dancing place was wrought by the very famous god who was lame in both legs,

like the one which once in broad Knossos

Daidalos made for Ariadne of the beautifully braided hair.

Here the dancing place is not said to be broad; instead Knossos itself is described as broad (εὐρείῃ).

I see two possibilities here. One is that Mykalessos really did have a spacious dancing place, more so than other places, and this detail about the town has been preserved in the epic diction. Another is that the memory of great cities of the Bronze Age past such as Knossos (whose preserved frescoes and seals from the Mycenaean period depict what appears to be choral dancing) has resulted in the creation within the poetic diction of a generic and ornamental epithet analogous to "good at the war shout" (βοὴν ἀγαθὸς), which Mary Ebbott and I have discussed extensively in connection with Iliad 10.283. (See Dué and Ebbott 2010 ad loc.) As we write there:

But let us notice first that Parry does not say here that this epithet has no meaning at all; he says only that it does not specify one hero in a way that it specifies no other hero. In other words, the heroes designated βοὴν ἀγαθός are, indeed, good at the battle shout. The fact that more than one hero is so designated suggests that such a skill would have been considered a good and useful one for a warrior, just as the formula itself is good and useful for the singer who is composing in performance.

Just as being good at the war shout was considered a good quality for the epic hero to have, so too it seems that having broad dancing places was a quality associated with ancient cities.

(For more on Bronze Age connections to the dancing place for Ariadne see S. Lonsdale, "A Dancing Floor for Ariadne [Iliad 18.590-592]: Aspects of Ritual Movement in Homer and Minoan Religion" in Carter and Morris, The Ages of Homer [1995].)

Geoffrey Kirk, in the section of his commentary on the Iliad that introduces the Catalogue of Ships, argues that all of the epithets used to describe cities in the Catalogue, "save about eight can be divided into one or other of four general categories of meaning" (Kirk, The Iliad: A Commentary, vol. 1, p. 175). The categories are as follows: well-built town; rocky, steep, high; fertile, broad, by sea/river; lovely, holy, rich. I have some disagreements with Kirk's classification (for example, he groups πολυτρήρωνά ["of the many doves," on which see below] with adjectives meaning rocky or steep and he does not include εὐρύχορον in his groupings at all) but I can see his point. Most cities in the Catalogue are described in ways that might be considered generic and ornamental, that is to say, not particular to any real city of any particular era. They have characteristics that would be good and useful for any city. Mykalessos may have at one time been renowned for its dancing places, but later audiences more likely understood the epithet along the lines of "having broad dancing places, in the way that all good cities do" or possibly "having broad dancing places, in the way that all cities of the heroic past did."  Whatever syntagmatic meaning the epithet once had has given way to a more paradigmatic one.

Thisbe on the other hand is an example of a place with a particular natural feature that seems to have been preserved within the tradition, not unlike the way that the very ancient vestiges of the Arcado-Cypriote dialect have been preserved within formulaic diction. (See the work of Milman Parry on the Homeric dialect cited in my last post.) Modern travelers (James Frazer in his edition of Pausanias [vol. 5 p. 162], Michael Wood in his book and documentary In Search of the Trojan War) have observed that the place believed to be ancient Thisbe (as evidenced by inscriptions) is indeed inhabited by many doves, as its epithet πολυτρήρωνά suggests. (See also Strabo 9.411, who observed them near the port.) Could this be an example of an epithet with syntagmatic meaning — meaning specific to the real Thisbe — that has persisted within the system? If so it is not the only such place: Messe (in the region of Sparta) is likewise designated πολυτρήρωνά at Iliad 2.582 in the same metrical position. Messe too has been observed by modern travelers to be a home to birds: "The identification of Messe with the site at Tigani receives some support from the constant din created by the 'pigeons and seafowl' in the cliffs of Thyrides to he south, which calls to mind the Homeric epithet πολυτρήρων" (Simpson and Lazenby 1970, 77).

I have been suggesting that places, like heroes, and, at a more basic level, formulas, had to enter the system of Homeric diction at some point, and that the formulaic diction associated with those places evolved as the system evolved. Some formulas persisted and may have retained something of their original meaning for centuries, until the epics crystallized into the form in which we now have them, while most other formulas evolved to become more generic, in that they were associated with cities in general. Mykalessa and Thisbe are just two examples of places whose traditional epithets underwent this evolution. Singers from those regions or the towns themselves may have indeed known them to have particular characteristics, but later singers and later audiences from other places most likely only knew them by their poetic identities, which may or may not have maintained characteristics particular to them. The fact that only a handful of places are described as εὐρύχορον and πολυτρήρωνά in our surviving evidence is suggestive that the formulas were created and used because they were indeed true of those places, but we must be aware of the limitations of our evidence. As Mary Ebbott and I note with reference to βοὴν ἀγαθὸς, if the Iliad did not survive and we had only the Odyssey, we might think that only Menelaos was ever so described. If more epic poetry survived, and especially if more catalogue poetry survived, we might find many such cities described as having dancing places and being full of doves. Even so, as I have tried to argue here and elsewhere, the fact that multiple cities are described as εὐρύχορον and πολυτρήρωνά does not make these traditional descriptions devoid of meaning, it just gives them meaning of a different kind, a kind that is quite typical of oral poetry.

Note: For more on the identification of Mykalessa and Thisbe and other named towns with actual historical places see Simpson and Lazenby (1970), though not all identifications of places mentioned in the Catalogue are universally accepted.