***

Let us consider a scholion in the Venetus A on Iliad 3.100. At this point in the narrative, Hektor has proposed on behalf of his brother Alexander a duel between the two men who claim Helen, Alexander and Menelaos. In his reply, Menelaos agrees that he should fight Alexander alone because the original dispute was between the two of them. The marginal note in the Venetus A here records a multiform of this reply.

The main text of the Venetus A, as well as the texts of the Venetus B and the two Escorial manuscripts and most modern editions, reads:

The scholion, however, preserves Zenodotus’ alternative reading of ἄτης in place of ἀρχῆς, which then means “because of Alexander’s error” (though the word atē is obviously hard to translate and encompasses much more than a simple mistake). The note itself reads:

Even in this one brief note in the margins of the Venetus A, we catch a glimpse of the multiformity of the Iliad. The differing readings give us evidence that multiple versions of the Iliad were available to the Alexandrian scholars. Why should we not consider Zenodotus’ reading a true multiform or, we might also say, a performance variation, a potential end to a line in a performance when the Iliad was an oral poem? One piece of evidence in support of doing so is that the phrase Ἀλεξάνδρου ἕνεκ’ ἄτης appears in this same position, after the weak caesura and completing the line, in at least two other lines:

Here is verse 24.28 in context:

At the Iliad 24 line, an intermarginal scholion in the Venetus A records ἕνεκ’ ἀρχῆς as an alternate reading. The line at Iliad 6.356 likewise reads ἀρχῆς in several papyri and some manuscripts as either the main reading or an alternate one. So in each of these three lines, we have evidence for both multiforms, and that evidence tells us something about how oral composition in performance works, when a singer either by training or under the circumstances of performance might choose one or the other phrase in any of these places. Either Alexander started the whole Trojan War (with his choice of Aphrodite in the Judgment, and by implication Helen) or, in taking a woman, he fell victim, like Agamemnon (Iliad 19.91), to atē.

***

I want to move from this example to make a larger point about editorial practice when it comes to an oral poem like the Iliad. How do we deal with the multiformity of the Iliad when we approach the text as an editor? The choices we make affect how students and scholars of the Iliad will understand the poem.

***

In antiquity, both Aristarchus and Aristophanes seem to have athetized verse 19.327 (not the whole passage, as West does, but just that line) because Skyros is not so far from Troy that Achilles would not have heard news of him and because they deemed theoeides to be inappropriate. The following note is found on folio 258r of the Venetus A:

Let me first note that once again athetesis is employed partly in response to the existence of more than one version of the text. Once again West has bracketed a passage that has not only universal manuscript support but additional support in the fact that more than one version of it existed in antiquity. For him, as for the Alexandrian editors, the existence of multiple versions of the passage makes it suspect. A Parry-Lord approach to the history of the text expects multiformity, and so can admit both as authentically generated verses.

But there is more to say about the logic of the scholion from a Parry-Lord perspective. Why is θεοειδής used ἀκαίρως? I’m really not sure why Aristophanes and/or Aristarchus felt this to be the case. Richard Martin has written two articles that show that this word, which is also used of Telemakhos the first time we ever see him in the Odyssey, evokes a traditional theme in Homeric poetry, that of the beautiful young man who must grow up and prove himself. Martin argues that the word encompasses the entirety of the story arc, from the moment when the young man is still very much untested and lacking in maturity to the time when he does in fact develop a character that matches his looks. I quote Martin here:

I don’t want to reargue Martin’s arguments, but I would suggest that a thorough analysis grounded in the way formulaic language operates is the right approach here, and could lead us to a greater understanding of the full impact of calling Neoptolemos θεοειδής. It might lead us to a fuller appreciation of the heartbreak in Achilles’ words—unlike Odysseus, he will not see his son grown up and matured. Might we find here in the Iliad, as so often in the Odyssey, an implicit contrast between the fates of the two central heroes of Homeric epic? Perhaps that is a stretch, but certainly we would need to marshal all available evidence to find out.

Achilles’ lament for Patroklos, including the verses about Neoptolemos, is typical as well. Many laments contain a narrative of both the past and the future—Briseis’ lament for Patroklos earlier in book 19 does this, as Pucci (1993) has shown. Peleus and Neoptolemos are Achilles’ past and future. Understanding Achilles’ words within the context of women’s songs of lament for the dead opens up a host of poetic possibilities. By likening Achilles’ grief to that of Briseis and the soon-to-be captive women of Troy, the Iliad depicts Achilles not only as having the greatest κλέος of all, but also profound ἄχος (and the two, in fact, go hand in hand). Achilles himself makes the connection in Iliad 18.121–125: “But now may I win good kleos, and may I cause some one of the deep-girdled Trojan and Dardanian women to wipe the tears from their delicate cheeks with both hands and lament unceasingly. And they may know that too long I have held back from battle.” Andromache’s song of sorrow is thus Achilles’ song of glory. But Achilles’ song is also a lament for Patroklos—and for himself:

In chapter 1 I suggested that the Iliad that survives for us exhibits a tension between two competing traditional themes: Achilles’ μῆνις and his ἄχος. Where West sees interpolated “Cyclic” material that is extraneous to the Iliad, I find evidence for the wider poetic traditions from which our Iliad emerged and with which it interacted.

Our two approaches are not entirely incompatible: in some respects these are merely two different ways of labeling and conceptualizing the same phenomenon. The distinction lies in what we do with the text that survives. While West, like Aristarchus, separates out the portions of the poem he believes are attributable to a master poet and distinguishes (via bracketing and in some cases omission) others as not Homeric, the editors of the Homer Multitext have made the decision to treat each historical instantiation of the Iliad as just that: a whole unto itself, worthy of study for its own sake as well as for what it teaches us about the larger Homeric tradition.

This fundamental editorial decision—that is, to represent each histori-cal instantiation of the Iliad as a whole worthy of study (in comparison with other historical instantiations)—has been the subject of criticism in Massimo Magnani’s recent (2018) analysis of the strengths and weaknesses of the Homer Multitext. Magnani is careful to point out that the unusual choices we have made as editors are grounded in the Iliad’s special status as a poem composed in performance, and he notes that we rely on Albert Lord’s formulation that for poetry composed in performance, there is no “original” that can be reconstructed. But he laments the absence of a single text to read (the establishment of which he calls “the final objective of every philological operation”) as well as the absence of a traditional apparatus criticus. Given that we have chosen to study the multiformity that survives in the Homeric tradition as natural and expected reflexes of oral composition in performance, Magnani feels that we have failed to adequately make distinctions between what he calls “aedic” and “rhapsodic variations.” Magnani concludes, “There will certainly be a moment when the critical editor will be supported or even replaced by the AI, but the task of choosing the best possible text in a methodologically correct manner cannot but remain the essential purpose of the philological activity for most of the manuscript traditions handed down by a plurality of manuscripts” (2018:100).

I will have more to say below about the question of distinguishing between “aedic” and “rhapsodic” variations—we do indeed actively seek to blur such distinctions. But as for the absence of an apparatus criticus, Mary Ebbott and I have argued elsewhere about the problems inherent in a typical apparatus (see, e.g., Dué and Ebbott 2009, 2010:153–159, and 2017, and above, p. 13). Only the most specialized consulters of an apparatus criticus can decode what the apparatus is attempting to convey, and even they will be often at a loss. As I have explored in Chapter 4, there are types of information that an apparatus simply can’t convey, especially when meaning is conveyed visually within a manuscript, as is so often the case in the Venetus A. (A glance at even a single folio will confirm this assertion.) But Magnani’s point is well taken. How can a user of the Homer Multitext make comparisons between readings without an apparatus criticus? In fact, during this phase of the project we have chosen not to worry excessively about how precisely our data will be used to make such comparisons. Our energy thus far has been focused on first and foremost simply creating the archival data that we hope others will use in dynamic and unforeseen ways. Even after twenty years of work we have only fully edited and published one manuscript of the Iliad and select papyri; there are lifetimes of work still to do. The Homer Multitext will only begin to be realized when all of the oldest manuscripts with scholia have been completely edited and published, and all of the extent papyri, and all of the works that quote the Homeric epics.

That said, now that we do have a complete data set, tools are coming into existence that will soon make the need for an apparatus criticus obsolete. Many different types of searching, text analysis, and topic modeling are already in common use by humanists, and those employing such methods now have the complete Venetus A manuscript of the Iliad to work with. There is no need to build software or online tools specific to the Homer Multitext, though we have in fact made some tools available on our project website and will continue to do so. But because our data has been created and encoded in generic, archival, and open source formats, we can trust that as the tools of our discipline evolve, our data will be compatible with them.

On a more conceptual level, the editors of the Homer Multitext have concluded that making the evidence of the historical sources available to every individual reader, thereby allowing them the tools and access to make editorial judgments for themselves, outweighs the practical advantages of an apparatus criticus, as useful as it may have been for scholars of earlier eras. The fact is, print editions will continue to exist in physical libraries for those desiring to do the kind of philology that Magnani describes. The editors of the Homer Multitext have never desired to replicate or facilitate nineteenth- or twentieth-century philology. Rather, we feel that it is time, and now possible, to ask new questions.

***

My aim in this concluding chapter is not to single out for censure the late Martin West, whom I have greatly admired since my earliest days as a student of Greek poetry and from whose scholarship I have learned much of what I know about textual criticism. Rather my aim has been to show how a multitextual approach solves problems for Homeric criticism and opens up possibilities for the next generation of Homeric scholarship that conventional textual criticism cannot.

As I listened to that episode the first time I heard it, I could not help but think of the paradigm shift in Homeric studies initiated by the fieldwork of Milman Parry and Albert Lord in the former Yugoslavia. Parry’s 1928 doctoral dissertation on the traditional epithet in Homer is a brilliant demonstration of the economy and traditionality of Homeric diction, but, as I noted in chapter 1, even Parry himself did not grasp the implications of this work initially:

It was only when Parry went to Yugoslavia to observe the still-flourishing South Slavic oral epic song tradition that he came to understand that Homeric poetry was not only traditional but oral—that is, composed anew every time in performance, by means of a sophisticated system of traditional phraseology and diction. For Parry, witnessing the workings of a living oral epic song tradition was a paradigm shift. Suddenly, by analogy with the South Slavic tradition, the workings of the Homeric system of composition became clear to him.

Just as Ćor Huso embodied for Parry the Yugoslav traditional singer, Avdo was for Lord on a practical level a living, breathing example of a supremely talented oral poet to whom Homer could be compared. Lord’s Singer of Tales is remarkable for its straightforward exposition of the practical workings of the traditional system in which poets like Avdo composed their songs; it is no surprise therefore that he found a great deal of power in the concrete example that Avdo provided. Avdo dictated songs, was recorded on disk, and was even captured on a very early form of video called “kinescope.” Their initial encounter was in the 1930s, and Lord found him and recorded him again in the 1950s. He was in many ways the test case for Lord’s theories about the South Slavic (and by extension the Homeric) poetic system.

The photograph of Avdo featured on the cover of The Singer of Tales was one that Lord had taken on his first trip to Yugoslavia and was to accompany his unpublished essay. The caption reads: “Avdo Medjedovitch, peasant farmer, is the finest singer the expedition encountered. His poems reached as many as fifteen thousand lines. A veritable Yugoslav Homer!” Here is Lord’s fuller description of Avdo in the essay:

In these excerpts I think we can see how important Avdo was for Lord’s earliest conception of Homer as an oral poet. Whereas Parry’s never-completed articles comparing the South Slavic and Homeric traditions focused on the hazy figure of Ćor Huso, Lord, when invited to give a lecture on La poesia epica e la sua formazione, entitled his talk “Tradition and the Oral Poet: Homer, Huso, and Avdo Medjedović”(see Lord 1970). As early as his 1948 article, “Homer, Parry, and Huso,” Lord links Avdo directly with Parry’s Huso: “During the summer of 1935, while collecting at Bijelo Polje, Parry came across a singer named Avdo Međedović, one of those who had heard Ćor Huso in his youth, whose powers of invention and story-telling were far above the ordinary.”

Lord’s comments about Avdo, especially in these earliest descriptions of him, focus on his excellence as a composer (despite the weakness of his voice), his superiority to other poets, and the length of his songs. It is not insignificant that in his unpublished essay Lord misestimates the length of Avdo’s song at fifteen–sixteen thousand verses, the approximate length of the Iliad, whereas in fact the longest song that Avdo recorded was 13,331 verses long. By 1948 Lord was careful to report the accurate total of Avdo’s verses, but he made certain to point out how extraordinary the length of Avdo’s songs were in comparison with his fellow singers, whose songs averaged only a few hundred lines. Clearly it was Lord’s first impression that Avdo provided the answer to the still hotly debated Homeric Question.

It would be easy to criticize Lord’s youthful essay, and few people would find it necessary to do so. And even if we jump forward, decades later, it seems obvious that Lord initially conceived of the paradigm of a dictating oral poet Homer because he was imagining him in Avdo’s image. The technology used to record Avdo was cutting edge at that time, and Lord would never have been so anachronistic as to suggest that Homer was recorded on audio disk, but to assume the technologies required for writing (pen, ink, loose or bound sheets of readily available “paper,” skilled scribes, etc.) for “Homer’s time” is an equally anachronistic projection. As much as Lord’s work is responsible for setting in motion the paradigm shift in Homeric studies that has allowed many scholars to abandon the “Homer as original genius” genre of criticism, he himself had his blind spots on this crucial point. Lord could have his Homer and his oral tradition too. But it is also true that Lord never speculated about the historical circumstances under which the Iliad and the Odyssey might have been dictated. For Lord, the question of the text fixation of the Homeric poems was not essential; rather he was concerned with the dynamic process, that is to say their ongoing recomposition in performance.

Parry, unlike Lord, did not have the opportunity to rethink his earlier work, or to conduct further fieldwork, or to spend decades studying the the South Slavic tradition and the Homeric poems as Lord did. His early writings on the economy of Homeric diction are a brilliant first step towards an entirely new way of conceiving of the composition of the Homeric poems, but they are only the beginning. Like Kuhn or Darwin, Parry’s work has been assailed by many as mistaken in this or that particular, or not sufficiently thorough to have worked out all aspects of the system it seeks to describe.

As Mary Ebbott and I discuss in our 2010 book, Iliad 10 and the Poetics of Ambush, much scholarship has been devoted to refining Parry’s initial findings about the economy of Homeric diction and the nature of the Homeric formula. There is strong resistance among those who feel that Parry’s work somehow minimizes the artistry of the poems or that the principles he outlined restrict the creativity of poets composing in this medium. Thus even those who accept Parry’s findings often seek to amend significant aspects of his arguments. We feel that the scope of Parry’s and Lord’s insights has been ignored, misread or misrepresented, or dismissed too quickly. Some (though certainly not all) efforts to revise Parry and Lord are built on a misunderstanding of the principles they documented in their fieldwork and a lack of awareness of, or at least appreciation for, the kind of meaning made possible by an oral poetic tradition.

That is not to say that our approach and the interpretations offered in our book on Iliad 10 have not also greatly benefited from the work of scholars who have sought to better understand such essential concepts as the Homeric formula and the complex relationship between orality and literacy in ancient Greece. There is, however, a significant difference between scholarship that expands the central insights of Parry and Lord’s work, even while modifying certain notions or definitions, and scholarship that sets out to “prove” Parry (more often than Lord) “wrong” in order to conclude, usually with no further justification, that Homer wrote, or somehow “broke free” of the oral tradition of these epics. These criticisms, like those cited by Lienhard against Kuhn and Darwin, seem to me to react to Parry as if he had “written the last word on the subject, not the first.” As Lienhard concludes at the end of the episode:

Building on Parry’s beginnings and Lord’s subsequent decades of scholarship, as well as the findings of those who have been inspired by Lord’s work—such as John Foley, Richard Martin, Leonard Muellner, and Gregory Nagy (all of whom approach Homeric poetry as a system rather than a man)—Mary Ebbott and I propose that it is now possible to enact the paradigm shift in Homeric studies that Parry and Lord set in motion. The paradigm shift we envision is borne out in our many years of work as coeditors of the Homer Multitext, and indeed it has only been made fully possible by that work. In our 2010 book, rather than looking to the intention or skill of a particular composer in order to explain the poetry, we attempted to ground our readings in the meanings made possible by an oral tradition. In de-emphasizing authorship we did not deny the possibility that some form of authorship, in terms of the poet as a creative artist composing in performance, could exist in this oral tradition. But we asserted that when the search for Homer’s genius is abandoned, many more illuminating possibilities present themselves.

The implications of such a shift are as true for the whole of Homeric poetry as they are for Iliad 10:

It strikes me as I revisit these words that I would now happily substitute “attested multiforms” wherever it currently says “Doloneia” or “Iliad 10” in that paragraph.

I submit that we should also cease to draw sharp boundaries between material deemed to be pristine, Homeric, or “by Homer” and that of other Archaic epic traditions. I do not mean to suggest that each song tradition did not have, to a certain extent, its own poetics, traditional themes, characters, accompanying formulaic language, and structure, but I believe that in the early phases of these traditions at least, the boundaries were far more fluid than we typically admit. By approaching the Iliad this way, we lose “Homer,” the individual genius, and we lose a fixed and monolithic entity that we can analyze as a self-contained unit. But we gain infinitely more Iliads to inform our appreciation of the one we know, and a greatly expanded poetic universe. We may lose “intertextuality,” but we gain “resonance” (Lord 1960:65; Graziosi and Haubold 2005), “traditional referentiality” (Foley 1999), “interformularity” (Bakker 2013:157–169), and “interaction” among speech and musical genres both past and present (Martin, forthcoming). These methodologies, all of which are grounded in an understanding of oral poetics and Homeric poetry as a system, open our eyes to the world of epic song in which each new performance was composed and received by its audience.

***

The creation of a Homer Multitext was first contemplated in the late 1990s in the midst of a different kind of paradigm shift within academia and the broader culture. The idea of a digital edition of the Homeric epics was without a doubt a response to the widespread adoption of the Internet as a means for scholarly communication, and previous editors of the Homeric epics simply could not have conceived of such an edition.

My collaborators and I were able to envision a multitext edition of the Iliad that was not constrained by the boundaries of the printed page, the cost of paper, the logistics of binding, the expense of color ink, or the need for specialized typesetting. Not only was it suddenly possible to imagine a freely accessible digital edition, but in those early days it was already possible for pioneers in this kind of work to imagine—as Ross Scaife, Christopher Blackwell, and Neel Smith did—a digital edition encoded for posterity, one that did not rely on proprietary software or soon-to-be obsolete tools or applications. A true scholarly edition, as they envisioned it, would not only provide free access to data, but would also openly license that data for scholarly use (and re-use) and verification by others. In bringing that vision to reality we have theorized and experimented and gone back to the drawing board many times. We are certainly in the midst of the “delicious flawed beginnings” of a new way of doing things. The digital tools and methods available to our students when they conduct research using our edition evolve from year to year, far beyond our early imaginings. But because our gaze has always been to the future, and is guided by an appreciation of the long past of Homeric scholarship, I am hopeful that we have built a foundation upon which many others will collaboratively construct a new paradigm.