Introduction

The research presented here has its origins in the philological controversy carried on for the last hundred years concerning the poem generally known as the "first Partheneion" by Alcman; following tradition, the fragment is the first of the poems attributed to Alcman in the Poetae Melici Graeci edited by D.L. Page (Oxford 1962), to which edition I shall refer throughout. Even if we are now sure that the beginning of the first poem of the first book of Partheneia in the Alexandrian edition of Alcman corresponds to the short new fragment S1, the edition of Page has not been superseded by vol. I of Poetarum Melicorum Graecorum Fragmenta edited by M. Davies (Oxford 1991). Since even the title of the partheneion transmitted to us by the "Papyrus Mariette" (P. Louvre E 3320) is uncertain, scholarly interpretation of these hundred or so lines in the dialect of Laconia has progressed in a circular manner; the most important interpretations will be set out briefly in the first part of this introduction. My own attempt to interpret the fragments may seem presumptuous in such a context; however, I thought it would be fruitful to follow a less travelled path and to examine the main contested points of interpretation in light of the society and the culture in which the poem was born and which furnished the occasion for its composition. In this volume, the reader will find the (updated) results of the preliminary research on the anthropological aspects of the poem's occasion. The reading and the interpretation of the poem in light of this broad context are given in the French version of the second volume of this research (Les choeurs de jeunes filles en Grèce archaïque II, Alcman. Roma 1977) and in my subsequent edition of the fragments of Alcman (Roma 1983); to those two volumes, I shall refer as Choeurs II and Alcman respectively. The justification of this method applied to the anthropology of literature will be found in the second part of the introduction.

With the appearance in 1957 of volume 24 of The Oxyrhynchus Papyri, a second important fragment of Alcman's was made available; these sixty odd lines were also put under the heading Partheneia. [1] The length of the fragment, third in Page's edition of the Poetae Melici Graeci, has attracted my attention in the same way as did the "first Partheneion." The poem was composed likewise for a chorus of young Laconian women, and its interpretation is based for the most {1|2} part on that of the first fragment and poses similar problems of comprehension. I therefore intend to give the material for the interpretation of both.

I shall here briefly enumerate the principal arguments of philological hermeneutics in an attempt to give a coherent interpretation of the two poems in question. Because these remarks are introductory, they will inevitably tend to be less specific.

Other poetic fragments attributed to Alcman were also doubtless composed for young women's choruses, [2] but because of their brevity I shall use them only as evidence to enrich the commentary. These fragmentary lines will also be the primary source for re-creating the historical, religious, and cultural context in which the two most important fragments were composed and performed.

A second source is to be found in passages by classical authors describing choral performances by women or girls. These passages are very diverse, ranging anywhere from Homeric poetry to history and philosophy. Included in this category are fragments of poems written for adolescent choruses by authors other than Alcman, particularly, for example, the Partheneia of Pindar, which will be examined as to denomination and occasion. [3]

Finally, evidence from authors on the fringes of the Alexandrian tradition, with its encyclopedic and critical tendency, will constitute the third source of information about the religion and society of Alcman's time. We know how the political and cultural history of Sparta has been idealized from antiquity on, so I shall attempt to take into account as far as possible the influence of ideology on these very varied sources. The studies that have been published on this subject will help me to maintain a critical distance from the information given by Archaic authors. [4]

In the editio princeps of the Mariette Papyrus published in 1863, Egger concluded from the opening myth in fragment 1 that the poem could belong to the Hymn to the Dioskouroi mentioned by several authors in connection with Alcman. [5] But the following year ten Brink related the non-mythical part of the poem to the παρθένεια ᾄσματα, songs for young women attributed by two {2|3} sources to the Spartan poet, and four years later Ahrens identified the whole poem as a partheneion. Philologists have since agreed on this, although there is no evidence directly relating to the poem to confirm it. [6] The situation is similar for fragment 3; Lobel, the first editor of this poem, states simply that everything in the lines we have shows that it is a Partheneion by Alcman. [7] This has also not been questioned.

A study of the history of the Partheneia as a literary genre presents itself, therefore, as the first step in the analysis of the poems of this category. If the Partheneia as genre has precise, distinctive features, identifying them could have useful implications for the two poems studied here, which have traditionally been attributed to this literary class.

Such an analysis shows that the category of the Partheneia was probably not defined before the Alexandrian period, and only when confronted, for editorial reasons, with the necessity of dividing the poems of the Archaic lyricists among several books, did scholars begin to speak of the partheneia, meaning a literary genre. Prior to this, a few fugitive references in Aristophanes and Aristoxenos show that they were aware of the distinctive character of the poems sung by choruses of young women, without being more precise about the contents. It is therefore not possible to speak of a true literary genre before the Archaic period, and even when the Alexandrians used the term partheneion, its definition remained very general; the partheneion is nothing but a poem sung by a chorus of adolescent girls for adolescent girls, as I tried to show in the last chapter of Choeurs II.

The negative results of this inquiry have led me to use the term partheneion for the two poems studied here only as an exception, since it was probably not pertinent for the period, and to use instead the number of the fragments as listed in Page's edition. As a final consequence, I have been led to examine the descriptions of performances of women's choruses to see whether the Greeks divided them into categories with definite characteristics. This study will be found at the end of the morphology of the female lyric chorus, the first of the studies presented here (chapter 2), and is based on the indications given by fragments 1 and 3 by Alcman. {3|4}

If we leave aside the external designation of these two fragments, and the possibility of classifying them, and pass to an analysis of their content, we notice that the essential feature common to both is the opposition of an I/we to one or two women who are spoken about in the third person. Philologists generally agree that this I/we represents the chorus singing the poem and that the chorus is made up of παρθένοι, young women; [8] in fragment 1, some incomplete lines (line 98f.) suggest that these young singers numbered ten or eleven. In addition to the chorus members, fragment 3 has one young woman, Astymelousa, and fragment 1 mentions two, Hagesichora and Agido; these three are all objects of praise for the chorus members. Fragment 1 also names a third woman, Ainesimbrota, who takes no part in the choral performance.

Interpreters of fragment 1 have generally attributed to Hagesichora the role of choregos mentioned in line 44, and it is likely that Astymelousa fulfills the same function in fragment 3. On the other hand, opinions on Agido's role are divided and have her variously as assistant to the choregos, a priestess, or the choregos of a rival group. [9]

The questions posed by the relations between the protagonists of the two groups and their respective status —number, qualities and modalities of the activity of the chorus members, function, characteristics and modalities of the leader's role —lead me to develop in my second chapter a sort of morphology of the female lyric chorus. With all the extant references to performances of women's choruses incorporated in this morphology, it should be possible to give a more precise definition of the function of each protagonist. In addition, the section on the modalities of the role played by the choregos and by the chorus members will be the occasion for another look at the problem of the partheneia as literary genre, and for an examination—based on an analysis of the forms of the songs sung by a chorus of young women—of the poetic genres that correspond to this Alexandrian category in the Archaic period.

In fragment 3, as in fragment 1, certain expressions suggest that the chorus members, or at any rate the young women to whom they address their praises, are engaged in certain activities in addition to their musical duties; these are for example δραμήται (line 59), μάχονται (line 63), θωστήρια (line 81), ἀγῶνα (3, 8), διέβα (3, 70), etc. In the same fragment, with the exception of the two generic expressions σιῶν (line 36) and σιοί/[σι]ῶν (line 82f.), two epicleses of {4|5} divinities (Ὀρθρίαι, line 61, Ἀῶτις line 87) are mentioned in the description of acts of a probable ritual character (φᾶρος φεροίσαις line 61, ϝανδάνην ἐρῶ line 88). And in fragment 3, the expression πυλεῶν’ ἔχοισα (line 65) could also be an allusion to the performance of a ritual act honoring a deity. In any case, these allusions prove that fragments 1 and 3 were composed for a particular occasion. They therefore raise the question as to whether the poems were dedicated to a divinity and of the possible religious nature of the festival at which they were performed.

However, to deepen the mystery surrounding fragment 1, fate has denied us any knowledge of Orthria or Aotis. It happens that a scholium on line 61 in fragment 1 proposes ὀρθίαι in place of ὀρθρίαι. [10] This conjecture would bring us into contact with the name of one of the most widely celebrated divinities in Sparta, Artemis Orthia, a goddess whose attributes would very likely presume the presence of a chorus of young women in a ritual observance. However, a metrical impediment prevents acceptance of this, and there have been other interpretations of the word; some scholars deny any reference to a god. [11] Since an allusion to Artemis Orthia is rejected, and if we exclude the interpretation of Aotis as a very hypothetical goddess of the Dawn sometimes identified with Artemis or with Aphrodite, the names of Medea, Helen, the Leukippides and Eileithyia have been suggested to fill the void of the dedication of fragment 1. [12] In fragment 3, although there is no direct reference to a divinity, the probable ritual character of the πυλεών that Astymelousa carries might connect it with the Spartan cult of Aphrodite-Hera. [13]

The problem of the dedication of the two poems reappears when we attempt to identify the actions of the young women who sang them. Among the expressions possibly relating to ritual actions, the verb δραμήται in fragment 1. 59 could refer to a footrace between Hagesichora (if she is indeed meant!) and Agido, while the φᾶρος φεροίσαις (1. 61) might refer to an offering given by the chorus. But just as some see only a metaphor in the reference to a race, doubt has also been thrown on the object carried by the chorus members; it could be a veil or a plough depending on the meaning given to φᾶρος. [14] Also, the ἀγών {5|6} mentioned by the chorus in fragment 3.8 has perhaps some connection with the μάχονται of 1.63. But in fragment 1, the identification of those engaged in this "battle," the meaning of which is as doubtful as that of the race just mentioned, depends on the interpretation given to Πεληάδες in line 60. And this word could represent, given all the interpretations suggested, a metaphorical description of Hagesichora and Agido, or an astronomical reference to the constellation of the Pleiades, thus indicating the season in which the feast was celebrated, or again, the name of another chorus competing against the one singing fragment 1. [15] Thus, however certain might be the dedication of fragments 1 and 3 to a divinity and their connection with the ritual practices of a cult, no definitive conclusion has yet been reached.

As an extension to the above reflections, I propose to examine all the cults in which a singing chorus of young girls or women would have appeared in Greece. This would have a double purpose: firstly, to determine with some precision the cults in which part of the ritual was performed by a female chorus; secondly, to determine to which gods these choral rituals were dedicated. It is surely not by chance that the divinities and the rites suggested as a solution to the problem of the dedication of fragments 1 and 3 all mark turning points in a woman's life: puberty, marriage, and childbirth.

The first part of the chapter on cults (chapter 3) contains a study of the purpose and function of those Greek cults in which a female chorus performs part of the ritual, and an analysis of the deities for whom the cults were performed. After attempting to classify these religious events, I shall then examine, in a second part, those cults specific to Laconia, focusing attention on their role in the rites of passage of women. This analysis will also include a list of the ritual practices carried out in Sparta by girls and women involving choral dances, and it will furnish the necessary arguments for establishing a {6|7} correspondence among the allusions to cults in fragments 1 and 3 and to a Spartan cult in particular.

Leaving aside the relationship between chorus and a given deity, a relationship that gives ritual significance to the actions of the chorus, we may return to the relationship between the protagonists themselves, and observe that the internal connections are marked in various ways. In fragment 1, some interpreters have suggested that the "cousinship" (line 52) relating the chorus members to their leader suggests a group of adolescents belonging to a society similar to one of the agelai in which the ephebes of Lacedaemon were enrolled. [16] And in the same way that the Spartan agelai have a justifiable function in the typically Spartan education system of the agoge because of their pedagogical role, similarly in fragment 1 certain metaphors such as the horse pulling the carriage (line 92) or the captain whose orders must be respected (line 94) seem to suggest the existence of a pedagogical relationship between the chorus members and their leader. [17] And finally it appears that these relationships also exist on the sexual level, in both fragments. Most interpreters agree that expressions such as Ἀσταφίς μοι γένοιτο (line 74), ποτιγλέποι Φίλυλλα (line 75), λυσιμελεῖ πόσωι (3.61), ποτιδέρκεται (3.62),etc., reflect the homoerotic sentiments of the chorus members for their leader. But opinion sways between Diels, who thinks that these literary expressions show actual relationships, and Page, who sees only the reflection of a certain "atmosphere of emotional intimacy" between young chorus members of the same sex. [18]

Questioning the importance of the educational and homoerotic connotations coloring the relationships between the protagonists raises the problem of the institutional character of these relationships and the part played in a specific social system by the female chorus containing them. I attempt an answer in the fourth chapter by bringing together all the (rare) references that we have to female societies and comparing them with what we know of the educational system and the homoerotic relationships that determined the form and function of the male agelai. Similarly among women, the "circle" of Sappho at Lesbos, to which several interpreters have compared Alcman's choruses of young girls, will be used as a term of comparison. [19] By defining the social rather than the religious function of the female chorus, I shall resolve the contradiction {7|8} represented by the collective expression of highly individual amorous sentiments by a group of chorus members, and, above all, determine the relationship of the chorus members with the poet—in spite of their strong links with the choregos—and, finally, with the Spartan community itself. [20]

Thus, if the perspective embraced here necessarily remains androcentric because of my own training and by reason of the cultural context in which I write, I have tried in my preliminary research for the commentary of Alcman's Partheneia to elucidate the varieties of song undertaken by choral groups in the Archaic period, paying particular attention to the different statuses of their participants, male and female, and to the relations that they maintain with one another. Concentrating fundamentally on the discourse of women, the survey of the varieties of song which are reserved for choral groups leads to the cults in which they occur and to the configuration of the divinities, outside of Lacedaemon and then in Lacedaemon itself, to whom they dedicate their song. It is the fact that the same representation of these different divinities exists (in terms of their field of action and of their manner of involvement) which, upon reflection, brings me to inquire as to the function of the choral associations and to take an anthropological approach to the amorous and more specifically homoerotic relations which appear in the collective songs understood as acts of cult. Independently of the interpretation of Alcman's fragments 1 and 3, this research should offer on its own a pragmatic approach to poetic speech and to the women's lyric chorus in their social, cultural, and affective aspects, while at the same time fully defining an essential feature of female gender and of its creation in young women in the Archaic period. But it is evident on the other hand that by not presenting in this English version volume II, which is devoted specifically to Alcman's fragments 1 and 3, I have also relinquished the light that those poems throw in turn on the problems described in volume I! {8|9}

In order to avoid the vicious circles that have swallowed up the interpretation of Alcman's Partheneia, I have purposely used the text only as a fulcrum for the ritual, cultural, social, and psycho-sexual circumstances surrounding the performance of these poems. Marxist criticism does nothing less, but in different forms. By stating that "the bonds that link the 'meanings' of a text to the socio-historical conditions of the text are not secondary, but are the basis of the meanings themselves," [21] Marxist literary analysis invites the interpreter to delve into the "conditions of production" of the work being studied. The literary work, a product of a given society, bears the mark of that society and can only be understood by studying the social and cultural conditions that brought it forth.

But if the literary work can be considered in this way, it must also be recognized that this same work itself fulfills a specific function in the socio-cultural entity. The analysis of the conditions of production can also serve to determine its raison d'être. It may have been created for a specific social reason, and this is generally the case with archaic Greek poetry. Defined as a poetry of occasion, [22] in contrast to modern poetry, it assumes a definite social function and can only be understood by reference to the circumstances of its creation. Archaic "literature" is never gratuitous, nor does it have the critical dimension of Alexandrian or modern poetry; it is always subject to the demands of the civic community for which it exists; it has to be understood as a social act.

It is clear then that, regardless of current fashion, it is necessary to find the meaning of Archaic poetry by studying the ritual, social, political or military occasions for it. This should not prevent us from realizing that there is no such thing as an unprejudiced reading. [23] However, it is not my intention to examine the cultural and psychological motivations we are all subject to as interpreters and which make us feel the need, as if it were an imperative, to explain its function in a given social situation. {9|10}

This concept of the socio-historical framework as an essential element of the literary work will help me not to overlook, in a study that often has recourse to synchronic methods of analysis, the historical dimension of the object being studied. Far from contradicting each other, synchronic and diachronic perspectives are complementary, and a synchronic analysis should always be balanced by historical considerations, principally in terms of the relevance of the results obtained from synchronic considerations within a given historical period, in this case Sparta at the end of the seventh century when Alcman was active.

The diachronic perspective will often be based on works written by historians. But since Spartan history has been so idealized and deformed, as is evident even in modern writings, another danger threatens us: historians have often used Alcman's poetry to substantiate their view of the aristocratic character of the society and culture of Sparta. [24] An uncritical use of these studies in appraising the social and cultural reality would lead to a vicious circle. I shall try to avoid this.

But before history comes synchrony.

Ethnological and anthropological research offers the philologist a very precious instrument to interpret and round out the often incomplete information of our sources concerning the social and religious institutions of antiquity. I have tried to show elsewhere that, with regard to methodology, structural analysis as defined by Lévi-Strauss is one of the surest means of obtaining a rigorous, but differentiated, comparison of the institutions of a society distant from us in time, as is ancient Greek society, and of contemporary societies called "primitive." [25] Not that we should in any way reduce or assimilate the Greek institutions to those of societies with a tribal structure; a diachronic analysis should prevent us from doing this. However, comparison shows that in the Archaic period, where rituals are concerned, Greek institutions present striking analogies, both structural and functional, with what is wrongly called "primitive."

This is particularly true for the so-called "rites of tribal initiation": several historians of religion have shown that this social institution, common to almost {10|11} all populations living in tribal patterns, existed also in the Archaic period in several regions of Greece and particularly in Sparta, and not on a "subsistence" level but with a completely functioning politics and religion. [26] In light of these studies it seems that a certain number of Spartan cults for adolescent girls and boys may mark the religious consecration of the various stages of initiation. Since I intend to make use of this institution, proper to tribal societies, in my comparative structural analysis, I shall give here its generic pattern, to which I shall refer as needed during my interpretation of the Greek rites and cults. This is by no means original; it is a summary of a number of studies made of the form and content of this almost universal institution. [27]

To begin with, what do we understand by tribal initiation? This particular type of initiation (rite de passage) aims to confer on the individual, by a more or less lengthy series of rites, full-fledged membership in the community formed by the tribal society. It integrates mainly adolescents, since this type of rite is usually addressed to adolescents, male and female, into the system of institutions and norms that govern the political, social, cultural, and religious life of the adult community.

So the rituals of tribal initiation concern the totality of the body politic, and are therefore official and public. Adolescents who have reached the age when they {11|12} can be granted adult status are obliged to take part; they are initiated as a group, thereby conferring on the tribal rites a collective character, different from the rites of puberty. Consecrating a particular moment in the physiological cycle of life peculiar to the individual, such as the menarche for young girls, the rite of puberty has a different and more private character than tribal initiation; the differences have often not been distinguished with sufficient rigor. [28]

Tribal rites of initiation belong to the large category of rites of passage now also called "rites of institution." While assuming a distinct semantic value and a specific function, they follow the formal schema as it has been defined since the beginning of the century in Van Gennep's decisive work. [29] According to this system, any rite of passage has three basic moments: first, separation (from the old state), next, a marginal phase during which the individual's status hangs between the old and new, finally, a period of admission (to the new status) and of reintegration. It is thus a simple sequence of leaving an old order and joining a new, with a neutral period in between.

In the case of tribal initiation rites, the old order is represented by the community of childhood, the new order by the socio-cultural system of the adult community. The first phase therefore marks the separation of young girls and boys of initiation age from childhood, as represented by life within the family. This moment is often accompanied by an act of violence, symbolizing the uprooting of the adolescents from their old way of life characterized by ignorance, irresponsibility, and the marked absence of sexuality in the child. [30]

This phase is followed by a marginal period of a certain length, anywhere from a few weeks to several months. It is expressed as a movement towards segregation and reversal in relation to the adult collective life. Segregation is at first spatial in that the initiates pass a period of time outside the home, isolated in the bush: in short, they have left the world of "culture" for that of "nature." Fed with special food, speaking another language, sometimes taking on a different name, sleeping in the rough, dressed in special clothes or entirely nude, perhaps disguised, they lead a life that inverts the values of the social life of the {12|13} community. Outside the authority of the law, they can undertake acts of depredation and violence.

But their activity can also be defined in a positive manner. They must submit to a series of trials varying from fasting to whippings, destined to harden the youths before they confront the difficulties of the struggle for survival that their entrance to adult life represents. Above all, though, the initiates receive an education on various levels. First, on the religious plane, boys and girls are initiated into the traditions of narrative and myth on which their society is based; through dance and song they learn the traditional stories of the community and certain of the ritual practices associated with them. On the plane of communal living, they are informed of the rules of conduct and the laws on which the cohesion of the tribe depends. Finally, they are introduced, and this is an important point, to the customs and norms of adult sexuality. This may be marked by a real physiological intervention, such as circumcision for boys or clitoridectomy for girls. In addition, by practicing the rite of tribal initiation, itself a religious act, adolescents make contact with the cult and sacred life of their community. The marginal phase imposed upon boys and girls is generally a period during which their death is symbolized and dramatized in a variety of ways. The act of circumcision may express it; treating the initiate as a dead body, even to the point of burial, is another. [31]

Initiate death is always followed by rebirth, a rebirth ritualized in the third phase of initiation dealing with acceptance into the new order. The rites of exit from initiation are generally public. The return of the adolescents to the bosom of the tribe is often represented by a long procession during which the newly initiated are presented to the community. By dances and songs, the neo-initiates show how the education they received makes them worthy of being admitted as full members of the adult society. The rite of entrance itself is often marked by a collective banquet that consecrates the union of the initiates and the members of the tribe.

By integrating new members into the social body, tribal initiation rites tend to assure the continuity of the community not only physically, but above all in the behavior of the individuals composing it and in the social and religious norms that assure its cohesion. Their aim is to form and educate future members of the tribe so that, once integrated, they will repeat the political, cultural, and sacred gestures and words of their predecessors by fully assuming the consequences. [32] To summarize, tribal initiation functions as a continuing {13|14} reproduction of society, undertaken in European societies for quite some time by way of bourgeois education in the West, and socialist education in the East. [33]

Since the social function and the gender status of a woman and those of a man are different within the community, the tribal initiation of girls is different from that of boys. Ethnologists have often thought that female tribal initiation ceremonies were secondary to those of adolescent boys; this seems not to be the case. [34] If it is true that (individual) puberty rites are more frequent for girls because of the singular event of menarche, female tribal initiation rites are nevertheless widespread; in form, they follow exactly the system described above for all rites of passage, and functionally they have the same goals.

On the other hand, as regards their content, the import of the mysteries revealed during the liminal period is weighted on the side of sexuality, marriage, and maternity in pointing out the gender role of the future adult woman. This does not prevent the girls, like the boys, from being initiated into the mythological, religious, and ethical traditions of the tribe. Moreover, it seems that in ceremonies for girls, the presentation of the neo-initiates to the community, enhanced by singing and dancing, often acquires a dimension that it does not have for boys.

Tribal initiation rites therefore aim to make adolescents into adults capable of playing their particular social role in the community and of assuming its responsibilities. [35] If the social function of men in communities with a tribal structure that has almost no division of labor is generally associated with the "exterior" survival of the tribe through activities such as hunting, raising stock, agriculture, or war, the woman's role corresponds to "internal" survival and mainly concerns the reproduction of the community through marriage, adult sexuality, and maternity. Tribal initiation thus creates a complete system of reproduction of the community, particularly in the domain of the social relationships of sex.

It must also be pointed out that the end of initiation is not always the same as the beginning of integration, and that the process of integration of neo-initiates can embrace several ceremonies over a period of time. This is particularly the case for girls, for whom a certain distance in time may separate the end of initiation from the marriage ceremony that marks their true entry into the adult world. We shall see that this distinction is especially noticeable in the case of ancient Sparta, where the young initiates enjoy an intermediary status {14|15} between the conclusion of their initiation and their enrollment in the army or their marriage. Each of these events is, of course, marked by specific festivities.

But in several of the Greek cults analyzed, the meaning of the "ritual," often minimally described by the sources, can be partially reconstructed by means of the "myth" which is presented by the narrative as an explanation of the cultural practice. This type of myth of "the first time" will be called a foundation myth, or, to use a Greek word, aition. I have to stress here that I use the words ritual and myth as referring to modern categories, coined in the field of social and cultural anthropology. Even if these words have been borrowed from the Latin and from the Greek languages respectively, the notions they designate do not correspond exactly to those of the ancients. [36] For this reason, I will use alternatively the less marked terms of narrative or story and of cultic practice. But using the narrative structure of certain myths to define the function of certain rites to the extent that both are the cultural products of the same symbolic process leads us to the semiotic approach proposed here.

To speak of semiotics in a work that merely utilizes on different levels the Saussurian distinction between signifier and signified may seem pretentious. Nevertheless, aspects of the semiotic approach seemed useful throughout the research.

There are three points concerning semiotics I should like to make:

First, in the morphology of the lyric chorus worked out in chapter 2, I have made frequent use of the dichotomy signifier/signified. This distinction separates out from among the variety of signifiers referring to the different protagonists of the choral group and their respective activities a small number of signifieds that I have called semantic features. [37] By means of this reduction of the signifier to the homogeneous level of the signified, the features describe the different aspects of the roles of the lyric chorus and its performances. A morphology can only be created by defining distinctive homogeneous characteristics which can be used as a criterion for classification. It is with this condition alone that I was able to respond to the demand to define the figure and function of each protagonist in the chorus and establish a classification of the different types of choral performance within their scope. {15|16}

This condition is also responsible for the inclusion in my body of evidence of the numerous iconographical documents of the Archaic period representing female choruses. [38] The rudimentary semantic analysis proposed will allow me to reduce the obvious heterogeneity of both the discursive and iconic evidence.

Determining characteristics of entities that one intends to compare and contrast raises the problem of the criteria used to limit the corpus being analyzed. I have mentioned my intention of using all available evidence of female or mixed choral performances for the morphology of the lyric chorus. The most obvious criterion for selection consists in incorporating into the corpus all evidence in which the activity signified by the term χορός and its derivatives is attributed to women (occasionally also associated with men). Even without taking into account iconographical documents, there is still a whole series of characteristic choral situations in which χορός and its derivatives are never used, but where the activity is described by a combination of signifiers such as ὕμνος + ὀρχηθμός or μέλπω + κιθαρίζω. Moreover, the term χορός itself, along with the signifieds 'dance and song performed by a group of singers' and 'group of singers,' also has the meaning of 'dance area' which is not immediately pertinent to my research. Consequently, homogeneous elements of the signified and not of the signifier should serve as a criterion for restricting the corpus to pertinent evidence. In the case of my morphology, it is the constellation 'group of girls or women (or a mixed group) performing a dance' that has determined the contents of the text to be analyzed. [39]

Because of the lack of a direct correspondence between the areas of signifier and signified, my selection cannot be exhaustive. Also, since the documentation on women's choruses is very meager, in certain cases I shall have recourse to male choruses for examples. I nevertheless hope to obtain results that are statistically satisfying.

Secondly, I have said that I intend to pass from the internal analysis of choral performance to an external study that will integrate the different aspects of the cults containing these performances. Here again I find myself face to face with the need for classification. The numerous cults that contain performances by female choruses do not always have the same religious function, since they celebrate different moments of a woman's life-cycle.

In ancient Greece, the function of a cult is essentially defined by the qualities of the god who is being worshipped. As I have observed, by using Dumézil's distinction, revised by Detienne, between a god's sphere of influence and the ways in which he/she acts, it is possible to define a semantic field for every {16|17} divinity. [40] This semantic field is composed of the collection of qualities that characterize the divinity and his/her modes of action in a particular sphere of influence; it limits for each cult the configuration of ritual practices performed. Very simply, the influence of Artemis, virgin goddess, is over female adolescents, and therefore the rites are performed by girls; whereas Hera, the legitimate wife of Zeus, reigns over legal marriage, and the rites are performed by young married women. Artemis' function is to resolve the physiological and social contradictions brought on by female puberty; Hera protects the rites of passage of marriage while guaranteeing its legality. Any god's field of influence is generally in harmony with the semantic figure of the god. However, it can happen that two different gods are active in the same area, but in their own ways, and they can intervene at the same moment, but in different domains, as for instance Hera and Aphrodite in the field of marriage. Identical ritual practices, such as choral performances, can thus have different functions according to the semantic form of the divinity they celebrate.

A definition arrived at by contrasting the semantic fields of various divinities celebrated by female choruses will serve as the first element of a classification of the cults examined.

Thirdly, the role of the cults is not solely defined by the semantic form of the divinity, but also by reference to the foundation myths, the aitia upon which they are based. The aetiological narrative of the myth gives an explanation of the ritual, justifies its existence, and generally finds its logical end in the establishment of the ritual concerned. The recounting of the myth adds a syntactic dimension to the semantic problems. Indeed, the ritual itself, far from being static, is generally the dramatization of a transition, a change of status, or the resolution of a crisis; its syntactic dimension is therefore essential. But because of the gaps in documentation in the case of Greece, it is often only in the aition of the ritual that we can understand which moment is being acted out.

It would of course be wrong to say that the story of the legend reflects the ritual word for word. But in analyzing the details, I shall have occasion to state that syntactically as well as semantically the aition often lends meaning to the events and function of the ritual it underlies. And the foundation myth generally reproduces in its own vocabulary and syntax the structures of the ritual with which it is connected. This is particularly true of the aitia at the source of the rituals of tribal initiation, and in more than one case it is possible to find in these myths the semantic values and the syntactic development characteristic of the schema of tribal initiation as defined in the preceding section. Using the structural instrument offered us by social anthropology, comparisons will be {17|18} made on the level of the ritual as well as of the myth, without forgetting that these two levels are distinct and structured according to a logic specific to each.

Thus the syntax and the semantics of the foundation myths will constitute the second element for defining the function and creating a classification of the cults examined.

The insertion of the semantic values and syntactical structures of these rituals and myths into the more general structures of the anthropological model will give them a more complete meaning and thus explain their function. Indeed, in a Saussurian perspective, it seems that the function of an element can be defined by its place in a larger system and by the parameters assigned to it by the system. In any case, I shall define the lyric chorus in this way when I try to determine its situation and role in the social system of ancient Sparta, above and beyond the semantic analysis of the cults that serve as its framework. Far from being incompatible, semantic and functional considerations complement each other in an explanation of the meaning of the object analyzed. The former break it down into its component parts, while the latter target its relationship with the system or systems above it in the hierarchy on which it depends; both have their origins in the semiotic working of a symbolic system. They will be used alternately in this study in an attempt to give the (female) lyric chorus the most complete definition possible. {18|19}