Linking poetry and scholia in medieval Homeric manuscripts

In our rationale for a digital edition of the Homeric epics, we have observed (e.g., in “Digital Criticism”) that the layout of a print edition of the Iliad or Odyssey affects the reader’s perception of the text. As Casey Dué and I say in that article: “A standard print edition will present a main text, and then record alternative readings in an apparatus (generally printed at the bottom of the page in smaller-sized font), giving the impression that there is the text — and then there is everything else. Compounding this problem and further obscuring the situation for nonspecialists, the apparatus as developed and practiced in classical textual criticism uses conventions and abbreviations that can only be deciphered by those who have received special training in these practices.” In other words, the layout of the page both assumes and projects a particular view of the text and its transmission, one that is at odds with the historical reality of the composition and transmission of the Homeric epics. Not only the conventional layout of print critical editions, but also the very limitations of print as a medium, are fundamental reasons why a digital edition is necessary to fully realize a critical edition of the Homeric epics.

With that pivotal significance of page layout in mind, we can then consider further questions about how the medieval manuscripts that we have digitally photographed for the HMT present a page layout and connect multiple texts on the same page. For just how the layout of the page in our medieval manuscripts presents, organizes, and links the multiple texts on each page is crucial for understanding the relationship between these texts. And, of course, a digital edition that incorporates and references the digital images of these pages does not lose that embedded information the way print editions (and electronic editions made directly from print editions) have. The aspect of layout that I will consider here in further detail is how the multiple texts on one page of these manuscripts are linked together for the reader.

In a conference presentation I gave along with HMT Associate Editor Leonard Muellner at the University of Lausanne, Switzerland last August, and now in a forthcoming article, I began examining this topic with what seems like it should be a simple question: how does a reader read a page of the Venetus A manuscript? It is clear that each page is carefully set up to accommodate these multiple sets of texts (see also Maniaci 2006). How, then, should a page of the Venetus A be read? Where does my eye begin reading, and how does it move through all the texts on a page? Just where to start is not obvious. The expected approach of reading top to bottom, left to right, quickly proves not correct, since I then start with commentary, and lines of poetry would be interspersed with comments. In fact, such a method of reading the page would be impossible, because the comments along the right side of the page require reading down through a comment before returning to the left side of the page, where the next line of the poetry starts. If I start by reading the lines of the Iliad do I stop each time I expect a comment, and go to look for it? Do I read all 25 lines, and then start at the top of the page and read the scholia in order from top to bottom, and left to right? Or do I read all the main scholia, then go back and read each of the other sets, the intermarginal, the interior, the exterior, and then the interlinear—and in turn, or in a single pass? Should multiple comments on the same line but belonging to different sets of scholia be read together?

The seemingly simple questions of how to read the page reveal the complexities of the relationships between the texts on the page. A codex of this size and grandeur was obviously not meant to be picked up and read like a paperback novel, but the basic questions of what the scribe’s expectations of his readers were and how readers might use this codex still require further investigation: we have not yet figured out all of the reading strategies the manuscript makes possible. Now that the manuscript is available for repeated reading by means of the digital images, we (and other scholars) can begin to conduct that kind of investigation. Yet as readers of all kinds of texts and page layouts, we know that we are not bound by the author’s (or scribe’s or typesetter’s) expectations. Thus, the Venetus A’s format for multitextual reading is an elaborate case of the possibilities for multitextual reading—different orders, different combinations—that are inherent in a page:

“The page is thus that physical aspect of the book that most persistently invites our eyes to move in directions other than the forward one, that potentially asserts, visually, the synchronic (and recursive) aspects of a narrative, over and against the diachronic ones. Roll or codex, the page is a block of text that realizes, in miniature, what is true of the entire book: all of these words are here together, at the same time” (Butler 2011: 9).

The design of the complex pages of the Venetus A is self-contained, yet nevertheless offers multiple ways of reading what they contain: our eyes can (and sometimes must) move in directions other than forward. A digital edition allows us to retain these possibilities in ways that a print edition cannot. Our digital edition of the manuscript encodes the location on the page of each scholion, for example, and the digital texts are linked to the images, so that a reader can easily go back and forth between them. Thus the structured markup of the scholia, including its location on the page, helps to restore the spatial information that the manuscript assumes the reader has by virtue of looking at the page. The page layout creates possibilities of and even the need for reading strategies. So what are the ways in which the layout and other features of the page indicate the connections between the multiple sets of texts? As we study the first five manuscripts for which we have digital photography, we are starting to see a number of different strategies that the scribes used for linking texts on the page.

Venetus A

The 10th-century Venetus A (Marcianus Graecus Z. 454 [=822]) uses a combination of spatial arrangement and lemmata, excerpted words from the poetic line(s) being commented on. In the Venetus A, the main block of the poetic text is apparent in the middle of the page—the blacker ink and larger script make it visually prominent. The commentary brackets the text in an open frame, and does so in multiple groupings, appearing also in the interior margin and even in the spaces between the lines of poetry. Even on less full, more typical pages, such as 43v or 46r, these multiple sets of writing on a single page are evident. The sets of scholia, which are distinguished by their location on the page, include: (1) the main scholia, which are written above, outside, and below the lines of the epic, in a bracketing shape (yellow in the image of 46r below); (2) the intermarginal scholia, located between the poetry and the main scholia and written in a different kind of script (green); (3) the interior scholia, written in the gutters (that is, toward the bound edge of the page) (purple); (4) the exterior scholia, written in the outer margin, toward the edge of the page beyond the main scholia (orange); and (5) interlinear scholia, written between the lines of the poetry (pink).

Scholia inventory and mapping by Melissa Browne of Venetus A, 46r

The placement of the comment in one of the five sets is already conveying some information about them, presumably about the source(s) of the comment, although there is still much to be investigated about that question. The placement of the text on the page also has a spatial relationship to the line it comments on: the scholia generally follow the order of the poetic lines. In addition, as we can see on these pages 43v and 46r, if the main scholia did not fill the page, the scribe left blank space in the main scholia area to move the later comments down the page to be closer to the line(s) they comment on.

Aiding the reader in the visual coordination of text and comment in the main scholia are the lemmata, a word or words taken from the poetic line to indicate which line is commented on. It provides the eye a means of moving more easily between text and comment. We readers can look at a line of poetry and then find the same (or similar: that will have to be the subject of another post) words introducing a comment on the line. The correspondence (or near correspondence) of words from the poetry are what the reader looks for in connecting the two texts.

The intermarginal, interior, exterior, and interlinear scholia in the Venetus A, however, rely almost entirely on location on the page to indicate their relationship with other texts. (The exceptions are the longer exterior scholia, such as those seen on 12r and 12v, which rely on content alone to link to particular lines.) But the briefer scholia (sometimes only one word) in these sets are written adjacent to, parallel to (in the case of exterior scholia), or above the line they comment on, and that spatial arrangement allows the reader to see what is being commented on. For example, the interior scholion on 43v reads simply, διὰ τοῦ α τὸ πεπασθε Ἀρίσταρχος, which means “with an ‘a’ [alpha], “πέπασθε”, according to Aristarchus.” When it is seen on the page written right next to πέποσθε at the end of the line (Iliad 3.99), its meaning, that Aristarchus read πέπασθε in place of πέποσθε on that line, is easily grasped.

Detail shot of 43v of the Venetus A The scribe of course assumed that anyone reading any particular text on the page had ready access to all the other texts on the page and also had the spatial information conveyed by their layout (such as what word πέπασθε is meant to substitute for). Because modern print editions of the scholia have seen fit to separate these texts, that easy and intuitive understanding of their relationship has been lost: when removed from the page, it is not clear why the difference of the spelling with an alpha is being noted, and the reader has to hunting through another printed text to figure it out.

Venetus B and E3

I am grouping these two 11th-century manuscripts (Venetus B is Marcianus Graecus Z. 453 [=821] and E3 is Allen’s designation for Escorialensis 291 [Υ i.1]) together because they share their main method of linking scholia to text: a numbering system. Greek numbers (that is, letters with a keraia written after to distinguish them from letters as letters) are used to link comments to lines, and even particular parts of the lines by being written both above the line being commented on and then before the corresponding comment. These two manuscripts have long been considered closely related to one another, and their similar layout and use of this numbering system contributes to that impression. The numbering system as employed in these manuscripts is also native to the codex format. (The codex is basically our normal “book” format: rectangular pages bound between covers. It differs from the earlier “roll” format: for more on this change in format, see Ebbott 2009 in Recapturing a Homeric Legacy, available in PDF for free). The lemmata system used in the main scholia of the Venetus A was useful for the earlier “roll” format. From what we know of Aristarchus’ editorial practices, he had the text in one roll, and his commentary (hupomnēmata) in another. The critical signs we see in the Venetus A (which no longer have a direct linking function in that codex, since most of the corresponding scholia do not contain the sign as well) and the lemmata helped the reader move back and forth between rolls, which could be opened simultaneously to the matching portions of text and commentary. (For more on the critical signs in the Venetus A, see Bird 2009.) But the numbering in both the Venetus B and E3 start at 1 (α’) at the top of the verso (left-hand) page, and continue sequentially down the page and then across to the recto (right-hand) page. When you turn the page, the numbering starts again at 1. Thus the layout and system of text coordination assumes a codex format, with a two-page spread visible when the book is open. One thing to note about modern print editions of the scholia from these manuscripts: when a “lemma” is cited for a scholion, it is likely the word in the text that the number is written over. Lemmata as we see them in manuscripts like the Venetus A are not used in these two manuscripts for the purposes of linking line and comment.

In addition to this numbered scholia, both the B and E3 manuscripts have additional notes written in different, later handwriting. In E3 these notes are placed in interlinear, intermarginal, interior, or exterior positions, in a much smaller quantity than those we see in the Venetus A, and, it appears, in a less systematic way (we have only just begun to investigate these scholia, which have not previously been published). These scholia seem also to depend on spatial proximity to provide connection between the poetry and the comment. In B the second set of scholia instead is connected in a way similar to the numbered scholia, but it uses symbols to make the connection: again, written both over the line and before the comment. A team of undergraduate researchers at Brandeis University, under the direction of Lenny Muellner, will be collecting and cataloguing the symbols used, and then we will be able to investigate whether the individual symbols have a particular meaning or are used in some systematic way.


The manuscript that Allen called E4 (Escorialensis 509 [Ω i.12]) is also from the 11th century. As we begin to study this manuscript in earnest, we are finding that it has many unusual features, so my remarks here are only preliminary. (See also the previous posts by Casey Dué about this manuscript.) E4 uses a combination of several linking methods: numbers, symbols, lemmata, and color.

Looking at the bifolio spread of 187v–188r that Casey discussed in her “Dog of Orion” post, I see that some of the marginal scholia are linked by Greek numbers, in a manner similar to B and E3: the number is written both over the line of poetry and before the corresponding note. The numbering starts at 1 on the verso (left-hand) page and continues sequentially on the recto (right-hand) page, again like B and E3. When I turn the page to 188v, I see that the numbering starts at 1 again. The major difference on these pages, which start Book 22 and (as Casey noted) must be read together, is that the first three numbered scholia appear on 187v, and coordinate to text on the facing page, 188r, where the corresponding numbers are written about the lines of poetry. In B and E3, the corresponding text and comment are, with only a few exceptions seen so far, on the same page, even as the numbering system itself continues across the bifolio spread of the open codex.

E4 also uses symbols in manner parallel to its use of numbers, with a symbol written above the line commented on and in front of the corresponding note. Some important differences from the use of symbols to link scholia seen in B should be noted, however. In B, the scholia using symbol are in a different hand and seem to have been added to the manuscript later. They appear as though they are placed according to what space was available around the numbered scholia. In E4, the symbol scholia are written in the same handwriting as the numbered scholia, and are written within the same block of scholia as those that are numbered. The numbered scholia and those linked with symbols are not separated spatially: the scribe intermingled them according to some principle we have yet to discern. Another question that we have begun to explore is whether the numbers and symbols indicate what source the scribe was taking the comment from (if he was, as seems likely, constructing this set of scholia himself), and whether his sources already had numbers and symbols associated with the scholia.

E4 also shows the use of lemmata, similar to the linking system we saw in A. As Casey noted in her “Dog of Orion” post, a lemma is used to link a comment on 187v to a text on 188r (and, as she noted, the lemma is itself a different reading of this line from what we see in the text on 188r). On 90v, which I looked at earlier for its scholion about Rhesos, the marginal comment on 10.437, which appears on the same page as the line it comments on, uses both a symbol and a lemma to link the comment to the text. There the lemma is, in fact, the entire line. In each of these cases, the lemma, or the first part of it, is written in a red ink. Indeed, with both the lemmata and the symbols, the scribe has made use of colored ink as part of his methods of linking texts on the page. Both a purplish-red and an orangish-red are used for some symbols and some lemmata. The color certainly helps my eye to pick up the linked texts more quickly. Whether the different shades of red mean anything in terms of the source or the content of the scholia so linked is yet another question we will have to investigate further.

E4 also contains some interlinear scholia, at least some of which use symbols to connect to the lines of poetry, but not all. These interlinear scholia appear between the lines of the prose paraphrase. That placement raises questions of whether the placement was simply one of available space, or whether in some cases the prose paraphrase itself needed explaining, or whether the annotator expected at least some of his readers to be reading the prose paraphrase first (or only).


For the Venetian manuscript that Allen called U4 (Marcianus Graecus Z. 458 [=841]), I will quote the description sent to me by Melissa Browne, Holy Cross Class of 2012. Browne is making a digital editio princeps (first critical edition) of this manuscript as her senior honors thesis. She has already inventoried every scholion in the manuscript, and, I would wager, knows it better than anyone else in the world.

According to Melissa: “The roughly 700 scholia of the U4 codex, as a general rule, do not follow one specific method of linking scholia to Iliad text. The scholia themselves we may divide into two distinct types: ‘graphetai’ scholia, and all other scholia.

Graphetai scholia, which consistently begin with the letters gamma and rho combined into a symbol (seen at left), usually appear directly to the left or right of the line upon which they comment, as they provide alternate readings of a line or half line which the scribe has, for some reason or another, decided not to [choose for the line itself]. Some scholia beginning with the graphetai symbol do appear to the right of the prose paraphrase passage; whether these graphetai scholia comment on the prose paraphrase or on the Iliad text will be an interesting question to consider. All other scholia in U4 we consider as one category, since there are no particular spatial patterns of placement or changes of ink/writing style/size which would distinguish certain scholia as particular ‘types’ (‘interior’ or ‘interlinear’, for example). The scribe does not use lemmata, as in Venetus A, but he does make (inconsistent) use of a system of symbols linking text and scholia, as in the Venetus B and E4. Most often, the scribe of U4 places his commentary near the horizontal scoring line of the line upon which he wishes to add notation. If there happens to be a larger number of scholia on a given recto or verso page, the scribe puts the symbol system to use. As U4 contains far fewer scholia than A, B, E3 or E4, his loose spatial system and employment of symbols, while not the most consistent, generally allows for the reader to link text and scholia without too much trouble.”

A central point I want to return to is that the scribe of each of these manuscripts constructed the page under the assumption that the reader would have the same page before his or her eyes, and therefore would have access to all of the texts on that page (or bifolio spread) simultaneously. These various systems of linking scholia were an aid to the reader’s eye, and perhaps also an aid for the scribe himself for the organization of multiple texts (perhaps from multiple sources) on the same page. As we become more familiar with the manuscripts we have had the chance to study through digital photographs, the importance of that relationship on the page becomes more and more apparent. That importance is the reason why the Homer Multitext has incorporated the photographs themselves into our digital editions though the visual inventory and scholia mapping that our undergraduate researchers have been doing, and why we insist on a complete, comprehensive accounting of the entire contents of our manuscripts. Users of the edition will be able to move from the digital text to the precise location of that text on the photograph of the page: they will thereby have easy access to the page itself, and all the important information contained its in arrangement.

Works cited

  • Bird, G. 2009. “Critical Signs―Drawing Attention to ‘Special’ Lines of Homer’s Iliad in the Manuscript Venetus A.” In Dué 2009:89–115.
  • Butler, Shane. 2011. The Matter of the Page: Essays in Search of Ancient and Medieval Authors. Madison, Wisc.: University of Wisconsin Press.
  • Dué, Casey, ed.. 2009. Recapturing a Homeric Legacy: Images and Insights from the Venetus A Manuscript of the Iliad. Cambridge, MA. [Available on-line at]
  • Dué, Casey and Mary Ebbott. 2009. “Digital Criticism: Editorial Standards for the Homer Multitext.” Digital Humanities Quarterly 3.1,
  • #Ebbott, Mary. 2009. “Text and Technologies: the Iliad and the Venetus A.” In Dué 2009:31–55.
  • Maniaci, M. 2006. “Words within Words: Layout Strategies in Some Glossed Manuscripts of the Iliad.” Manuscripta 50:241–268.

Website © 2019-2020, the Homer Multitext project. For licensing on image collections, see the Image Archive page.

Powered by Hydejack v8.1.1