Resolving a Century-Old Problem of a Scholion’s Lemma

Athena wearing Zeus's aegis, one of the topics of this scholion

This post comes from the team of editors creating the HMT editions of Iliad 15 and Iliad 18 in the Venetus A manuscript during the Holy Cross Summer Research program in the Humanities, Arts, and Social Sciences: Brian Clark '15, Claude Hanley '18, Stephanie Neville '17, Charlie Schufreider '17, Alex Simrell '16, and Melody Wauke '17. Their perceptive solution to the problem of this particular scholion and its lemma demonstrates their masterful familiarity with the Venetus A manuscript and the practices of its scribe. — Mary Ebbott

Among the many potentially befuddling characteristics of the Venetus A manuscript, one thing that is usually fairly straightforward is the connection made to the Iliad text by a scholion’s lemma, an excerpted word or phrase from the Iliad text at the beginning of a scholion which cues the reader as to what lines the scholion will be commenting on. While some scholia have no explicit lemma, these scholia usually indicate clearly what Iliad lines are being commented on based on their content. When a scholion’s lemma has no clear connection to the Iliad text, however, editors are thrown for a loop.


We find such a case at the beginning of Iliad 15, after a newly awakened Zeus sends Iris to order Poseidon to stop his assault on the Trojans. Chafing against the assumed supremacy of his brother, Poseidon angrily reminds Iris that, being a son of Kronos, he has equal authority over the actions of the battlefield as Zeus. Still, Poseidon yields and Zeus then orders Apollo to rouse Hector to arms.

The lemma in question begins the third main scholion on folio 194 verso of the Venetus A.

Detail of Venetus A 194v: see zoomable version here

Like other lemmata, it is written in the same semi-uncial lettering different from the rest of the scholion. Additionally it is set apart from the rest of the scholion by some sort of punctuation, in this case a colon. It reads:

Κρόνος χρησμὸν λαβὼν:

This string of words does not appear anywhere on the page. Discrepancy between a lemma and text does happen elsewhere in the Venetus A. Often the differing words in the lemmata actually suggest an apparent multiformity. However, for a lemma to differ so greatly that the discrepancy goes beyond spelling differences or word order is certainly unusual. The Holy Cross team examined the content of the scholion to see if any explanation of multiformity existed, but no such discussion followed:

ὅτι ὀ ΐδιος αὐτὸν τῆς βασιλείας μεταστήσει υἱὸς, τὰ γεννώμενα κατέπινεν Ῥέα δὲ τεκοῦσα Δία, Κρόνῳ μὲν αὐτοῦ λίθον σπαργανώσασα ἔδωκε καταπιεῖν· τὸ δὲ παιδίον εἰς Κρήτην διακομίσασα, ἔδωκε τρέφειν Θέμοδι καὶ Ἀμαλθίᾳ ἡ ἢν αἴξ, ταύτην οἱ Τιτᾶνες ὁποτ ἂν ἐθεάσαντο ἐφοβοῦντο· αὕτη δὲ τοὺς αὑτῆς μαζοὺς ὑπέχουσα ἔτρεφε τὸ παιδίον. αὐξηθεὶς δὲ ὁ Ζεὺς μετέστησε τῆς βασιλείας τὸν πατέρα. πολεμούντων δὲ αὐτὸν τῶν Τιτάνων, Θέμις συνεβούλευσε, τῷ τῆς Ἀμαλθίας δέρματι σκεπαστηρίῳ χρήσασθαι εἶναι γὰρ αὐτὴν ἀεί φόβητρον, πεισθεὶς δὲ ὁ Ζεῦς ἐποίησεν καὶ τοὺς Τιτᾶνας ἐνίκησεν· ἐντεῦθεν αὐτὸν φησὶν αἰγήοχον προσαγορευθῆναι

ὅτι [Because] his own son will remove him [Kronos] from his dominion, he [Kronos] gulped down his begotten children. But Rhea, after giving birth to Zeus, wrapped a stone in swaddling clothes and then gave it to Kronos to devour. As for the child, she sent him to Crete to be raised by Themis and Amalthia, who was a goat and one whom, whenever the Titans laid eyes on her, was feared. Nursing him at her breast Amalthia raised the child, and once he had grown Zeus stripped his father of his kingdom. But when the Titans were making war with him, Themis advised him to make use of Amalthia’s hide as a shield. For she advised that Amalthia was always terrifying. Persuaded Zeus did so and conquered the Titans. For this reason he says that he was addressed as “aegis-bearing.”

So not only does the scholion concerning Zeus’s upbringing shed no light on the multiformity of the text, but the scholion’s first word, ὅτι, only confuses matters further. ὅτι is typically used at the beginning of scholia to correlate with Aristarchean critical marks, such that they mean, “[the critical mark was placed there] because.” So not only does the lemma not exist, but the scholion supposedly corresponds with some Aristarchean critical mark that also does not exist, and the content is not typical of the kind of editorial comments Aristarchus makes, either.

Our team at Holy Cross was not the first editorial team to struggle with this scholion. Both Erbse and Dindorf recognized the peculiarity of the scholion, and having analyzed the content, concluded that the scribe had made a mistake in placing this scholion here and decided that he meant to comment on line 229 of book 15:

ἀλλὰ σύ γ᾽ ἐν χείρεσσι λάβ᾽ αἰγίδα θυσσανόεσσαν, (15.229)

On the one hand this conclusion makes some sense. That line contains mention of the aegis, it contains some form of the word λαμβάνω, a word from the lemma, and the name Kronos, another word in the lemma, appears just four lines above. While it would still be a stretch, it is perhaps understandable how there might exist some multiform which included our given lemma in the around this part of the text. The problem with this interpretation is that line 15.229 appears on folio 195v, two folios later from where the scholion actually appears. Such a “mistake” seems extremely unlikely for the scribe of the Venetus A who, on every other account, scrupulously connects scholia with their textual counterparts on the same physical page.

Assuming that the scribe did not make that egregious error, some other explanation is required. Our Holy Cross team decided to look at the surrounding scholia to look for positioning clues. The main scholia are ordered sequentially with the line they comment on. So a scholion on line 2 will succeed a scholion on line 1 while immediately preceding a scholion on line 3. In this case, our scholion of interest is sandwiched by two grammatical scholia on line 15.187. Since a single line can have multiple scholia, it is only logical that our scholion in question must also comment on the line. An examination of the line 15.187 reveals that such a conclusion is not too far-fetched:

τρεῖς γάρ τ᾽ ἐκ Κρόνου εἰμὲν ἀδελφεοὶ οὓς τέκετο Ῥέα (15.187)

While the text and lemma do not match, the poetic line is connected to the scholion’s content: namely, it is about one child of Kronos and Rhea. It seems safe to say that the scholion is simply providing an expanded mythological background to the story of Kronos’s children.

How then must we take this ghost lemma? It has every indication of being a lemma in that it is written in the same lettering that is used for lemmata and is set apart from the rest of the scholion by a colon. The solution then is to concede that the scribe did make a mistake or, at least, that some scribe at some point in the manuscript tradition made a mistake. One must concede that Κρόνος χρησμὸν λαβὼν is not a lemma after all, but merely the beginning of the scholion mistakenly written as a lemma. If one removes the colon after λαβὼν, first sentence of the scholion reads:

Κρόνος χρησμὸν λαβὼν ὅτι ὀ ΐδιος αὐτὸν τῆς βασιλείας μεταστήσει υἱὸς, τὰ γεννώμενα κατέπινεν

Kronos, having received an oracle that his own son will remove him from his dominion, gulped down his begotten children.

No longer does one have to infer from context who the father is whose dominion is being stripped away, nor does one have to supply a subject for κατέπινεν. Both cases are elucidated by the clearly nominative form Κρόνος. Most importantly the ὅτι which likely threw off Erbse and Dindorf given its usual formulaic structure in the scholia serves instead here as just a marker of indirect speech, “that,” rather than an indicator of critical marks. And with that conclusion our diplomatic edition was able to not only keep true to the manuscript’s layout, unlike Erbse and Dindorf, but also make sense of something that had baffled some of the brightest Homeric scholars.

Dingbats and Doohickeys in the Venetus A

This post was written by Brian Clark (Holy Cross '15) and Alex Simrell (Holy Cross '16). In it they observe the practices of the Venetus A scribe when he has too much material for his usual layout of certain types of scholia on the same page, and they draw some preliminary conclusions from those observations. Their work was accomplished during the Holy Cross Summer Research program in the Humanities, Arts, and Social Sciences and was supported by the Center for Hellenic Studies. — Mary Ebbott

During our work on Iliad 18 this summer, our team found evidence that supports the theory that the scribe of the Venetus A intentionally wrote certain types of comments into specific predetermined regions on the folio. Certain folios still bear the marks that divide up the page into these different areas. Generally, a folio has the text of the poem, surrounded by five categories of scholia: main, intermarginal, interior, interlinear, and exterior. We do not yet fully understand the function of each different group, but we now know that the placement of these groups matter. Perhaps the position on the folio indicates something about the source material for the comment.

Sometimes, when dealing with a very dense page, the scribe was forced to break his rules about the placement of scholia. For example, folio 248v, which covers Iliad 18.480–18.504, is highly packed with comments about the astrological bodies found on the shield of Achilles.

Folio 248v of the Venetus A manuscript: view it in detail in the Homer Mulitext manuscript browser

In the exterior margin, there are three scholia which are not written in the usual hand of the exterior scholia (you can see a typical exterior scholion above these three). Additionally, these scholia have distinctive connecting signs that connect the scholia to the interior margin.

Exterior margin detail of 248v: see zoomable version here

The presence of these connecting signs—dingbats or doohickeys, if you will—are common in other manuscripts, such as the Venetus B, and are similar to the numbered footnotes in the Upsilon 1.1 [see this earlier post for more on how the Venertus B and Upsilon 1.1. link their scholia to the poetry]. 248v is not the first instance of these connecting signs in the Venetus A, but it is just now that we are able to draw conclusions based on our observations over the years.

Detail of interior margin of 248v: see zoomable version here

The use of these signs supports the claim that the scribe intentionally laid out this manuscript with a desire to place certain scholia in specific regions of the folio. By adding these signs, the scribe is guiding the reader not to take these three scholia as exteriors, but rather to read them as part of the interior scholia. On this crowded folio, there is not enough room in the interior margin for the scribe to write all of the interior scholia where they belong. As a result, he was forced to write these three scholia outside of their intended location.

Detail of 248v showing both exterior and interior margins of 248v: see zoomable version here

The first two connecting signs are clearly in the interior margin, and you can see how filling that space with those two comments would have made the margin far too crowded. The last one, however, is written in the interlinear position, above the word ἀρωγοί. Still, we feel that this last scholion is meant to be an interior scholion. The space where the scribe would have placed this connecting sign is taken up by another scholion, thereby forcing him to move the sign to the interlinear position. One could theorize that he trusts his reader to recognize this scholion as an interior, rather than as an interlinear, due to the length and content of the comment.

Another argument for these seemingly exterior scholia to be taken as interior scholia is the nature of their comments. In addition to the different scribal hand used for the exterior scholia, these comments generally lack any introductory or explanatory material. Typical exteriors are comprised of just a few words, while these three scholia offer a more complete explanation of the comment.

Further, the signs do not link the scholia to a specific word in the Iliad line. For example the second scholion, which comments on Iliad 18.499 (ἀνδρὸς ἀποφθιμένου· ὃ μὲν εὔχετο πάντ᾽ ἀποδοῦναι), reads

παρα Ζηνοδότῳ "αποκταμενου" καὶ ἐν ταῖς πλείσταις καὶ ἔστιν οὐκ απιθανος ἡ γραφή ⁑

Zenodotus writes the word "αποκταμενου" [instead of the word "ἀποφθιμένου"] and this is the reading in most editions. This is not an untrustworthy reading

As you can see, this comment is not about the word directly next to the connecting sign (that is, ἀποδοῦναι), but instead it provides a multiform for the second word of the line, ἀποφθιμένου.

Not only does a folio like this help us better understand the practices of a medieval scribe, but it also is another example of the benefits of a diplomatic digital edition that is linked to citable evidence. A printed edition can say that these scholia are “out of place,” but cannot accurately show the function of these connecting signs. Our editions preserve the original placement of these scholia while assigning them intelligent labels based on the evidence of the scribe’s normal practices.

What's in a Name (of a Manuscript)?

Along with the exciting new additions (with more to come) and the reorganization of the Homer Multitext website, you may have noticed that the primary names used for our Iliad manuscripts have been changing as well.

Here are the “shorthand” designations we will use henceforth, and the other possible designations or past names we have used for our five current manuscripts:

HMT designations       [names in brackets are other library catalog designations and modern
                                                                                                 edition sigla]

Venetus A or Marciana 822    [=Marcianus Graecus Z. 454, Allen’s A, West’s A]    

Venetus B or Marciana 821    [=Marcianus Graecus Z. 453, Allen’s B, West’s B]
   
Marciana 841       [=Marciana Graecus Z. 458, Allen’s U4; West does not include]
   
Escorial Υ.1.1      [=Escorialensis 294, Escorialensis 291, Allen’s E3, West’s E]

Escorial Ω.1.12    [=Escorialensis 513, Escorialensis 509, Allen’s E4, West’s F]

Why are there so many names for each manuscript? And how did we settle on the names that we now use?

One reason why there are multiple ways to refer to each of our manuscripts is that names or other designations come from two different sources: the libraries which house the manuscripts, and modern editors’ designations or “sigla.” Within each of those sources, in turn, there are additional reasons that names or designations change over time.

Modern editors’ names
When a modern editor collates multiple manuscripts to create a modern edition, s/he might use sigla to designate the manuscripts so that s/he can use those one or two letters and/or numbers in the cramped spaces of a traditional print apparatus criticus. Even the first modern publication of the Venetian manuscripts created such sigla: Villoison in his 1788 edition of Homer’s Iliad with the Scholia called the two manuscripts “A” and “B” for ease of reference in tight print spaces when he compiled the scholia. (You can see digital photographs of Villoison's edition here.) That is the origin of the names “Venetus A” and “Venetus B” (“Venetus” is the Latinate version of “Venice”).

When we look at how two twentieth-century editors have created sigla, we see that each created his own system, resulting in further designations. Allen calls the two manuscripts from the Escorial “E3” and “E4”: he actually collated four manuscripts from the Escorial and designated them 1–4 in the same order as the Escorial’s catalog numbered them, so although these two are the oldest of the four, they end up as “3” and “4.” West, however, uses far fewer manuscripts in his edition, and chose to call these two “E” (=“E3”) and “F” (=“E4”). Allen gave to the Venetian manuscript Marciana 841 the siglum “U4”: in addition to the Venetus A and Venetus B, Allen consulted 13 other manuscripts in Venice at the Marciana, and gave them the sigla U1–13, using a Latinate consonantal “U” for “Venice” since he uses “V” for the manuscripts in the Vatican library. West does not include this manuscript, and so did not create a siglum for it.

We have written before about the problems with the traditional apparatus criticus, both in terms of the difficulties in deciphering from it what a particular manuscript actually has, and in the way it presents barriers to readers, obscuring what the editor has seen in manuscripts in a secret code, as it were. Since we will not have the constraints of a printed page, and because we want to avoid reinforcing the “outsiders not welcome” feel of using such abbreviations, we will not follow in the footsteps of these print editions by using their sigla or by creating our own for the manuscripts.

Up until now, however, we have used these sigla in our writing about the manuscripts and on the website. We have done so because they are short, and therefore handy, and because we learned to do so as part of our academic discipline. But we will avoid using most of these from now on, for the reasons just stated as well as some further considerations I will explain now.

Because “Venetus A” and “Venetus B” are both easy to say and write and widely known names, we will continue to use them as our shorthand for these manuscripts. They have become such established names (and the manuscripts are so important for our textual history of the Iliad) that both Allen in his sigla and West in his use “A” and “B” to designate these manuscripts. But in addition to the hassle of needing to include multiple equivalences any time we refer to the manuscript using the sigla of a modern editor (such as noting the Allen’s E4 = West’s F), using one modern editor’s sigla as our normal reference could imply some scholarly choices or allegiances that we are not meaning to make. So our decision is to use library names and designations rather than sigla from any one modern edition, with the exception of Venetus A and Venetus B because those are so widely known and consistently used.

Library catalogs and shelving
Deciding to use library names and designations wasn’t the end of our decision making process, however, because in these cases, too, there is more than one possible reference. We have had the habit (once again from our training in the discipline) of using Latinate forms for the library names: Marcianus and Escorialensis. But in our desire to welcome nonspecialists to the Homer Multitext, we have decided to avoid such Latinate forms, and use the actual library names, Marciana and Escorial, in our designations for their manuscripts.

Then we must locate them, as it were, within the library, since we have more than one manuscript from each. The libraries that house our manuscripts naturally catalog them, and give them catalog numbers when they do. The Marciana and the Escorial libraries have long histories as institutions, so it is not surprising that they have cataloged their manuscript collections more than once. We have at times used or included designations for the three manuscripts from Venice that were assigned in the catalog compiled by Zanetti in 1740 (http://marciana.venezia.sbn.it/fondo-antico-e-appendice). Those designations, “Marcianus Graecus Z.” plus a number (454, 453, and 458 for our three manuscripts) include the name of the library in the Latinate form, the Latinate designation “Graecus” because Zanetti cataloged the Greek manuscripts separately from the Latin and Italian manuscripts, and the “Z.” to indicate Zanetti’s catalog. So there was both a “Greek 454” (our Venetus A) and, likely, a Latin 454, requiring the specification of the language to distinguish the two. Zanetti also grouped the manuscripts within the language into genres, so the Greek poetry manuscripts all have numbers between 438 and 461.

Similarly at the Escorial, different catalogs have given the two manuscripts different catalog numbers over time: 291 or 294 for what Allen called his “E3” and 509 or 513 for what Allen called his “E4.” Because these numbers have changed over time, using one over the other complicates our designations: even if we choose to use the latest one as the most “current,” it is scholarly practice to include the older numbers so that it is clear that it is indeed the same manuscript. So, just as with using any one editor’s sigla, using one of the Escorial catalogs creates the need for citing the other (294 = 291, e.g.).

Our solution is to use the designation the library has for the shelving of the manuscripts, since that is a designation rooted in some reality about the manuscript as a physical object: namely, where they can be found. For the Escorial, those designations are a Greek letter, followed by two numbers, separated by periods: Escorial Υ.1.1 and Ω.1.12. The numbers from the Marciana of 821, 822, and 841 also refer to their physical place in the library, and are attached to the codices, as this picture of the spine of the Venetus A shows (the label toward the bottom reads “822”).

Spine of Venetus A manuscript, photo published by the Homer Multitext project

The long history of our discipline shows itself in these changing and accumulating designations. As we reinvent the critical edition, we want to use names that reflect the manuscript as a historical object and are lasting, easy to use, and easy to understand. We hope the designations we have chosen will fulfill those desires.