Chapter 15. The Deipnosophists as a Text: Genesis, Uses

In this strange path that associates banquets, a library, lexica, and insatiable scholars, we now need to ask what was the aim of the game. Is Athenaeus perhaps the victim of the most severe Alexandrine syndrome, that of a compilation mechanism destined to go indefinitely round in circles? Are the dialogue and its setup then nothing more than a miserable attempt to give literary coating to thousands of lexicographical entries and reading notes, or, put briefly, for the “cards” of an obsessive reader, which should never have left the shelves of his private library to enter the space of public circulation (ekdosis)? To count on the coherence of the text presupposes a reflection on Athenaeus’ project and on the form of his work; this is precisely what we shall attempt to do now, after having brought to conclusion our “ethnographic” observation of Larensius’ circle, investigated in its playful conviviality and its intellectual exercises.
The third book, of which we only have an abbreviated version, begins abruptly: “Since the grammarian Callimachus said that a great book is equal to a great evil” (3.72a). Is this the sigh of a learned copyist, or of the epitomizer himself, at the beginning of what appears like a grueling task? Or the caveat that an author addresses to himself, having a foreboding of the amplitude of his work? I would like to see in it an invitation to reflect on the Deipnosophists as a book, and also as a mega biblion, as a “great book”.
How do things stand as to the nature of this book, its structure, its genesis, its uses, and the ways in which it was supposed to be read? What is the status of a book that reuses fragments of thousands of other books? To consider the Deipnosophists in the context of its genesis leads us once again to analyze the practices of the literati, which are the center of attention in Athenaeus’ text itself, through the interaction of the members of Larensius’ circle. Like Democritus (8.336d), Athenaeus is an expert on comic theater: his monograph on Archippus testifies to his taste for that literary genre. And, once again like Democritus, Athenaeus uses the technique of the extract. One initial key to reading the Deipnosophists would be to consider the work as the collection of its author’s reading notes, the map of his wanderings among books, whose linearity has been blown to pieces in favor of a different logic which is no longer bibliographical but thematic. [1] The reuse, quotation, paraphrase, and juxtaposition of fragments of texts of varying size are an essential component of Athenaeus’ writing. The author (as we have seen) sometimes reuses long extracts from books in his work; such is the case of the quotation of the On Alexandria by Callixeinus of Rhodes, Lynceus’ Dinner-Party Letters, and the treatise by Dioscorides of Tarsus that Athenaeus quotes or paraphrases in the course of his long digression on the way of life of Homeric heroes (1.8e ff.).
On the whole, however, Athenaeus’ extracts are much shorter fragments, a few lines, or sometimes a single word. Taking notes during one’s own readings can serve two different purposes: either one wants to extract from a book all the content and all the words that are noteworthy, all the facts worth being remembered for their surprising or enigmatic character; or one can have in mind the filtering of the extracted material according to a given question or writing project. That filtering operation can also be applied to a collection of extracts that was put together previously, or even to a lexicon. In that operation, which is characteristic of the literary activity of the Hellenistic and the imperial periods, and even beyond, the book is considered as a source of information and raw material that linear reading has the function of bringing together. Extraction enables the segmentation of those materials and their archiving on a level different from the original text. It certainly answers a preoccupation of ergonomic order, since extraction is selection, but also subdivision of a continuous text into elements that are independent, objectivized, mobile, and such that they lend themselves to any sort of new combination. The procedure also has intellectual effects, since decontextualization cannot happen without a form of recontextualization, which produces specific meanings: alphabetical order, thematic groupings, or the juxtaposition of various solutions brought to the same problem are so many forms of reorganization that produce knowledge and create poles of local coherence around an object or a question.
Gathering extracts from a text and keeping them in the order of succession they had in the original text means to epitomize, albeit in elliptical mode. Besides, that is the operation undertaken in Athenaeus’ text itself. The form of compilation expresses the lack of satisfaction of a reader (or a group of readers), and in a way achieves a transformation of the reading protocol, a transformation that dissociates content from form and the information from the letter of the text. The author of the epitome of the Deipnosophists has filtered the informative content of the text eliminating facts that were useless in his eyes, in this case a great number of bibliographic clarifications and the digressions linked to the dialogue of the deipnosophists. On the other hand, gathering extracts from a text, distributing them according to a different order and sometimes mixing them with extracts from other texts, results in the creation of a new text. The juxtaposition of heterogeneous statements, their reuse within a discursive line like Athenaeus’ dialogue, produces specific meanings and constructs a new knowledge. The Deipnosophists produces a global knowledge that none of the reused quotations could have produced by itself as such. Such is the effect of the ordering of information in series, following a subject or a problem.
The particularity of Athenaeus lies in his not having taken compilation to its extreme stage, which would have consisted in suppressing from the quotation all the indicators of its initial context and author, and objectivizing the contents by paraphrasing or summarizing them. The indication of the bibliographic source has authenticating value, and anchors every quotation within the space of the library. It also poses, in a provocative way, the question of the relation with that library.
Modern commentators have evidently wondered about the amplitude of Athenaeus’ readings: has he read all the texts he cites? Or, more precisely, are the quotations the result of his activity of eklogē practiced directly on the works? Or has Athenaeus used intermediate sources that themselves had already gathered quotations on the most disparate subjects? The search for Athenaeus’ sources is a subject of zētēsis that would probably not have been scorned by Larensius’ circle, and it is a subject that will continue to provide thesis subjects in universities for several generations still. However, using Athenaeus in order to reconstruct previous compilations that he may have cannibalized in silence, and of which practically nothing has come down to us, is something that belongs to the field of highly conjectural activity, and I shall not endeavor to proceed in that direction. [2] I shall, on the contrary, attempt to reformulate the problem. What implicitly underlies the Quellenforschung is a depreciative vision both of the author and of the discursive form and the intellectual project within which he inscribes himself. It is rather a perspective of accusation, since the techniques of philology bring to light the cannibalization of the works of others; a theft with aggravating circumstances, moreover, since apart from everything else, that way of proceeding shows remarkable intellectual laziness, the author limiting himself to drawing from lexica and compilations, and then redistributing the materials and making them his own. Such an accusation, however, presupposes a mode of book circulation where the status of author is recognized and protected; an intellectual community with its rules and principles; and a specific status of the work as a personal intellectual creation.
Now how do things stand as to those reference works: lexica, collections of quotations, and thematic monographs bringing together the written sources concerning a given subject? They give access to textual materials that may have escaped the reader of the works in question, and a compiler like Athenaeus can, in theory, for the same authors, use his own reading notes and integrate them with this or that quotation provided by a lexicon, without necessarily mentioning the intermediate source. What is essential is the information thus gathered.
The Deipnosophists are a deconstructed library, cut into pieces that are then redistributed according to the thread of the meal and the sumposion. Those fragments go from the isolated word to the quotation of many dozens of verses. The lexica and compilations, which from the third century BC had also undertaken to sift through the library according to multiple grids, were Athenaeus’ natural working instruments, and are, besides, often quoted as such. They individualize, select, and combine those quotations of varied origins; they read the classical authors at the first, second, and third degree; they create poles of local and thematic coherence by making the fragments interact and by making them instrumental to a project of cultural archaeology: compilation hides an activity that is properly literary and is a form of rewriting that can be traced both in the framing of the quotations (how do they start? and how do they end?) and in the modifications made to the letter and to the meaning of the texts. Those modifications can, on the one hand, be the result of the imprecision of memory, or of the mistakes introduced in the progressive stages of transcription of reading notes, from the tablets to the hupomnēmata and to the finished work. However, a careful reading also reveals a subtle and deliberate game by virtue of which Athenaeus modifies or changes the text of the fragments so as to better insert them within a thematic treatment. [3] That game reveals a real activity on the texts, both on the level of the selection of materials and on that of transcription and assemblage. The very project of Athenaeus would be meaningless if it were based in large part on one or more earlier collections—not to mention that an enterprise such as that would be insulting to Larensius, his bibliophile protector (assuming that we do not relegate Larensius to fiction).
A close reading of Athenaeus shows that in reality he very frequently mentions the indirect sources from which he gained access to a quotation of an ancient author. It is by multiplying the filters and sieves that he can put together his materials, and his practice of compilation may reflect the working techniques of his characters. An indirect quotation also presents an intellectual, almost aesthetic interest: it adds a diachronic dimension to the vast synchronic network of connections between fragments of text. It inscribes them in a tradition, in chains of transmission, and it inscribes itself in the continuity of an intellectual practice that is attested by the lexicographers, the scholars, and the grammarians of the past, but also by all the great readers, be they philosophers, historians, or exegetes. It doubly proves the culture of the compiler, who not only extracts materials from the texts he has at his disposal, but knows where to look for quotations of texts to which he has no direct access, since he knows the “basic instruments” of literary activity: “Those iambic verses have been quoted by Didymus and by Pamphilus” (11.487c). And finally, the indirect quotation constitutes an interesting object in itself, and offers suggestive material to the contemporary reader for the study of text transmission. What does it mean to read Empedocles through a quotation by Theophrastus (10.423f)? Or Anacreon through Simmias (11.472e)? What does it mean to read an iambic composition by Hipponax through Lysanias and his work The Iambic Poets (7.304a–b), or through Pamphilus’ lexicon (2.69d), or Hermippus’ monograph Hipponax (7.327b)? Or Asius through Duris (12.525e)? Or to come to an entry of Theodorus’ Attic Glossary thanks to Pamphilus’ lexicon (15.677b)? Even when books had become impossible to find, the indirect tradition still conveyed precious fragments; Athenaeus himself participates in that historical process.
So how do things stand with the Deipnosophists as a text? Is it a disorderly accumulation of reading notes, roughly grouped by theme: vegetables, breads, fish, wine cups, crowns, and so on? Certainly not. Athenaeus’ characters play an important part in the construction of the text and in its thematic structuring. After all, they are the ones who quote and explore their mental libraries, bringing innumerable quotations out of them. Their dialogue rules, controls the flow of quotations, and defines their relevance, their appropriateness. That which from time to time reveals itself as appropriate in the course of the conversation, kairos, determines the distribution of the extracts. [4] Athenaeus’ text is constructed and articulated: that is demonstrated by the many internal cross-references which establish anaphoric (“we have seen that...”) or cataphoric (“we will see that...”) connections within a book or from one book to another. The mention of those connections is meaningful both within the dialogue of the deipnosophists, who are in that way required to master within their memory everything that has been said (their mental libraries become enriched not only with what they have read and extracted from books, but also what they have heard), and in relation to the reading of the work itself. Timocrates, who listens to Athenaeus’ account, is here a substitute for the reader, and all those references delineate the way in which to use the Deipnosophists as a text, its ergonomy.
Indeed, those connections testify to a double preoccupation: that of an author and of the characters he sets up, desiring to put into order a considerable mass of discontinuous, heterogeneous, and decontextualized textual materials, and that of the reader, who will have to trace his itineraries within the work. It is in that sense that Athenaeus is presented, from the very beginning of the work, as the steward (oikonomos) of the text (1.1b). For the author, as for the reader, memory is the determining factor, because the internal anaphoric and cataphoric references are not cross-references like the ones we encounter in our books with numbered pages but are points of reference that are relative and refer backwards or forwards from the passage being read. It is not expressed precisely whether that will involve a move backwards or forwards, a move of some few lines or of several rolls. [5]
Athenaeus’ composition technique cannot be dissociated from the ways in which his text is appropriated. To which readers is it addressed? And for what use? Was it intended for Athenaeus’ private use? Was it meant to circulate outside Larensius’ circle? [6] And what was its link with Larensius? Was the work perhaps built in the image of his library of ancient Greek books, condensed, exploited, run through from every direction?
A careful look at the composition of the text itself reveals a certain formal heterogeneity: indeed, Athenaeus uses different discursive genres, different regimes of textuality, such as comic or Platonic dialogue, dissertation, lexical lemmata, lists, or the simple juxtaposition of extracts. One can see one of the characteristic traits of a text marked by diversity and variety (poikilia), both in theme and in form; and perhaps one can also see the traces of the genesis of the work, having been pieced together little-by-little from the materials found while reading, and traces of the first editorial processing of specific subjects. The Deipnosophists would then be not only a library, but also a scholar’s small study, where all the stages of his work are preserved. That heterogeneity, however, also adapts itself to the nature of the materials used and to the subjects tackled, some of which lend themselves to a continuous discussion, others to a catalogue bound to be perpetually increased, others still to polemics, and to the necessary formulation of different points of view. For the reader, this results in different approaches to the text, from continuous reading to ad hoc consultation.
For example, the first book, only preserved in the epitome, contains a whole treatise built into it, with its own title, On the Way of Life of Homer’s Heroes (1.8e and 24b). Does Athenaeus reuse the treatise by Dioscorides of Tarsus, as the Suda suggests? Or is it a summary and a set of reading notes? Either way, we are here dealing with a treatise that is thematically coherent and autonomous, which could have existed independently on a roll of papyrus. Some extended speeches of the deipnosophists are also close to a hupomnēma. Plutarch, for example, treats of the question of parasites (6.234c–248c); he is followed by Democritus with a long talk on flatterers (6.248c–262b). In both cases we are dealing with a continuous exposition, a real oral dissertation, where the orator discusses the materials he has collected in his memory. Those accounts can also constitute independent treatises, and it is probable that they were written as such before they were reused within the dialogue. [7]
For the lists, Athenaeus sometimes but not always adopts the principle of alphabetical order. Those lists can be minimalistic, like the catalogue of breads, put together from various glossographers, where the name is accompanied by the bibliographic sources and by a brief definition (3.114b). At times, they allow him to stock up a large quantity of other information. The text then becomes a glossary: every entry opens with a word, then brings together all the information concerning it. The list of fish offers a particularly interesting example, because the narrator chooses to reorganize the conversations of the deipnosophists within an alphabetical list, so that Timocrates can memorize them more easily (7.277b–c). In the logic of the narration, that choice recalls once again those performances of combinatorial memory that can redistribute into an alphabetical list the materials memorized during a conversation, filtering out a certain number of facts, such as the name of the interlocutors. [8] This series of cards in alphabetical order, however, corresponds most probably to the original editorial phase, the phase, that is, when Athenaeus collected his materials, finding support in lexica and in his extracts; that was a way of proceeding that lent itself to the progressive enrichment of his database, from discovery to discovery and from reading to reading, until it found itself embedded in Book 7 of the Deipnosophists. In Book 2 we find a lexicographical sequence concerning all the dishes eaten during the initial phase of the banquet (2.51b–71f), and in Book 14 (643e ff.) a list of sweets. Neither one nor the other are in alphabetical order. Such lists can be used for the ordering of the most disparate forms of information: etymology, synonyms, morphology, usage, quotations, grammatical annotation, and so forth, juxtaposed without having been articulated. The entries create poles of local coherence, collecting the highest possible number of facts on a particular theme or topic. On that same model Athenaeus also proposes a list of courtesans (13.583d–e ff.).
Finally, this also poses the question of the uses of the text, of its reading protocols. Was it a work that was to be read continuously, following the thread of the banquet and the symposium, and the unfolding of the meeting from the beginning to end? Was it a dialogue, between a comedy and a conversation of wise men, which made it possible to follow the interaction of characters with strong personalities, their mood changes and their witticisms? Or were the Deipnosophists a vast repertory of materials meant to be reconfigured and extracted, so as to be integrated into new hupomnēmata? [9] Or was it, on the contrary, an immense collection of reading notes intended to be memorized and reused orally in other literary banquets? Or a summa docta, designed for ad hoc consultation by whoever mastered its structure and its thematic progression, in search of a word, an explanation, the treatment of a problem? [10]
Those are the many forms of appropriation of this oceanic work, which now lead us to consider navigation as a structural key and as the essential principle in reading it.


[ back ] 1. One could compare the Deipnosophists with the treatise On the Tranquillity of the Soul 464F and 465A, which Plutarch states he composed on the basis of his collection of personal notes. See Dorandi 2000:27ff.
[ back ] 2. At the end of the nineteenth century the German philologist F. Rudolph saw in the Universal History of Favorinus of Arelate the main source of both Athenaeus and Aelian. His reconstruction is now considered obsolete.
[ back ] 3. See the stimulating remarks by Pelling 2000, who invites us to re-examine the techniques and stylistic principles of writing and compilation.
[ back ] 4. See 3.110a; 4.134c, 162a; 8.348d; 13.564f–565a, 579d–e; 14.622d.
[ back ] 5. For example, 9.401f–402a refers to 4.128a; 10.415d refers to 4.144f; 10.453c–e to 7.276a; 11.496a to 3.125f; 12.543b to 4.168d; 14.615a to 4.261c. There are also many anaphoric and cataphoric references within a single book.
[ back ] 6. The absence of papyrological attestations, the rareness of quotations from Athenaeus in later authors, and, finally, the manuscript tradition itself, which is based on the Codex Marcianus 447 and on its epitome, suggests that Athenaeus’ work did not circulate widely.
[ back ] 7. See also, at 11.487f–494b, the long discussion concerning Nestor’s cup, which also takes the form of a dissertation and which moreover is largely dependent, in this case, on the treatise by Asclepiades of Myrlea.
[ back ] 8. See also 14.616e, where Athenaeus, as he is speaking to Timocrates, decides to bring together in a single account everything that has been said by the deipnosophists on the pastimes of the banquet, without respecting the order of succession of the conversations or the names of the interlocutors.
[ back ] 9. In the fourth century, the sophist Sopatrus of Apamea, a disciple of Iamblichus, collected in twelve books all the extracts drawn from his readings. In the first book, in particular, he had collected the notes extracted from Athenaeus’ Deipnosophists. The books that follow testify to the use of other such collections: the Epitomai by Pamphila, the Universal History by Favorinus, the Miscellaneous Notes by Aristoxenus, the Lives of the Philosophers by Diogenes Laertius, and so on. Photius has preserved a detailed summary of Sopatrus’ collection (Bibliotheca, cod. 161).
[ back ] 10. The answer to those questions is not obvious, and would require a general inquiry into the modes of circulation of texts written in erudite and literary circles. Pliny the Elder was offered four hundred thousand sesterces for his hundred and sixty rolls of reading notes (Pliny the Younger, Letters 3.5.17). See also Aulus Gellius, Attic Nights 14.6.