Homer Multitext Project


Neel Smith

1§1 In this paper, we present some results of recent study of manuscripts of the Homer Multitext project. While individual sections of this paper were primarily the responsibility of one or two co-authors, we selected the progression of topics discussed here to illustrate three kinds of new discoveries that undergraduate contributors to the Homer Multitext project are making. We begin with a discussion of the physical production of three manuscripts in Venice and the Escorial Monastery that have not previously been analyzed in this way. The second section looks at how scribes combined Iliadic text and notes in a single layout. This section begins by looking at the page-by-page organization of the Venetus A manuscript; this is followed by a comparison of the organization of material in bifolio spreads in a manuscript in the Escorial Monastery. The third section analyzes the textual content of the Venetus A’s scholarly notes to suggest what kinds of sources the scribe might have consulted, and some ways he adapted them to the Venetus A manuscript.

How were the physical pages of a manuscript produced and bound?

Identifying quire divisions in three unpublished manuscripts

Melissa Browne and Stephanie Lindeborg

2§1 Careful examination of a manuscript can tell us a great deal about the way in which they were constructed and bound. Manuscripts were typically written in sections of several sheets that were stacked and folded in half. Each full sheet sheet is a bifolio: two physical pages or folios, each with a recto (front) and verso (back) side, so each bifolio sheet contains four pages of the manuscript. The groups of sheets are called quires. Quires that contain three sheets are called ternions, and four sheets, quaternions. Some quires may contain an extra half sheet.

2§2 When composing a manuscript, a scribe would work quire by quire, numbering each stack of sheets so that they would eventually be bound in the correct order. This was necessary because the scribe who wrote the manuscript was not necessarily the person who bound the manuscript together. The numbering could also be useful during the process of writing the manuscript, because it would allow the scribe to move back and forth among quires and add additional features without losing track of the correct order of the manuscript. In Byzantine Greek, numbers were written using the “Milesian” notational system, in which letters of the alphabet represent numeric values (Alpha equals one, Beta equals two and so forth). Numbers are typically written with bars or hooks above them, to distinguish them clearly from alphabetic letters. We can identify quire arrangements by finding the numbers that the scribes wrote in for their own use, typically in the bottom corners of the first or last folio in a quire.

Escorial Upsilon 1.1

2§3 The Escorial Manuscript Upsilon 1.1 is an eleventh-century Byzantine manuscript of the Iliad that is housed in the Escorial monastery just outside of Madrid. The quire arrangement of this manuscript has never been published. Last summer, while composing an edition of the scholia (scholarly commentaries written in the margins of the text) of Iliad 2, our team of researchers decided to look into the quire arrangement of the manuscript. After a sweep through the entire manuscript, we discovered both Greek numerals from the original composition of the manuscript and Arabic numerals written in a later hand. The manuscript contains a total of 328 folios, which are organized in forty quires. These quires, however, are of various lengths. Most of them are regular four-sheet quaternions that contain a total of eight folios each. There are quite a few three-sheet ternions that contain six folios each. There are also quite a few seven-folio quires, composed of three and a half sheets. Most of the quires are numbered in the original hand both on the top and bottom of the folded stack. Some are marked only on the top, and a few are numbered only on the bottom. In most cases, where numbers might seem to be missing, this can be explained by either the later insertion of replacement sheets in the manuscript where the originals have fallen out, by fading of the original ink, or by trimming of the folio when the manuscript was rebound (as the numbers tend to appear in the inner margin and this margin shrank when the manuscript was cut and rebound). An example of the Upsilon 1.1 quire numbers can be see at the bottom left of the folio linked here: Quire numbers in Upsilon 1.1.

Venetus B

2§4 The Escorial Upsilon 1.1 manuscript is often said to be a twin of another eleventh-century Byzantine manuscript, the Venetus B, located in the Biblioteca Marciana in Venice. Because of this connection, especially when we have observed oddities, it has often been our practice to compare the two manuscripts. Therefore after finding the quire arrangement of the Upsilon 1.1, we looked into the quire arrangement of the Venetus B. These findings yielded much less variation than the Upsilon 1.1. This manuscript was marked only with Greek numerals. It too contained forty quires, but a total of 338 folios, ten more than the Upsilon 1.1. All of the Venetus B’s quires are quaternions. This manuscript starts off marking quire numbers only on the top of the quire, but after the fourteenth quire, marks both top and bottom, with a few interruptions to that pattern through the rest of the manuscript. An example of the Venetus B quire numbers can be found at the bottom left of the folio linked here: Quire numbers in Venetus B. One of the eventual goals of our researchers is to further examine the relationship between the Venetus B and the Upsilon 1.1 to determine why we have such a stark contrast in the quire arrangement of two manuscripts that are reportedly twins.

Marciana 841

2§5 A thirteenth-century manuscript of the Iliad with scholia, now in the Biblioteca Marciana in Venice, Marciana 841 was originally the second of a two-volume set. Volume one contained Iliad books 1-12, and is no longer extant. The extant manuscript originally included books 13-24, but the first section has been lost. In its present form, it begins halfway into book 14 (line 14.419), and continues uninterrupted through the end of Book 24.

2§6 Three sets of numbers were added to the pages at different points in the manuscript’s history. The latest phase in the numeration of the manuscript is a set of Arabic numbers from 1 recto up to 194 recto, indicating folio numbers. These numbers, located in the upper right-hand corner of each recto side, are the most recent addition (a relative term) to the manuscript. This numbering must have been done when the manuscript was in its present, damaged state—that is, missing its first volume, as well as book 13 and half of book 14 from the second volume, since the first folio number is ‘1’ on the first recto page of the manuscript. The Marciana’s folio numbers often blend with the next-oldest, “phase” of numeration in Marciana 841: ascending Arabic numerals, increasing in consecutive order from 14 to 24. These appear in the upper right-hand corner of nearly every folio side (recto and verso) of the extant codex, and indicate the book number contained on the folio side. These book numbers are not as well-preserved as either the folio numbers or the Greek quire marks, most likely because of their location in the oft-thumbed far-right of the upper right-hand corner.

2§7 The third set of numbers is contemporary with the original composition of the manuscript, and indicates the quire number. As in the Escorial manuscript, these numbers are written in normal Greek “Milesian” notation; and occur consistently in the same location in each quire, on either the recto or verso side of a folio on the bottom left or bottom middle. The folios of the manuscript were bound in a regular pattern; they are marked off in quaternions, sets of eight folios, or sixteen pages. There are 24 labeled quires in the extant part of the manuscript, quires 32-55. The end quire 55 is double the length of the average quire, encompassing 16 folios.

2§8 From the number of quires in Marciana 841 (24), as well as the number of folio sides contained within each quire (8), we can reconstruct just how many folios are missing from the beginning of the manuscript. The first quire number, a nearly illegible ‘32,’ appears on 1r. This means we are missing 31 quires from the codex. If we multiple 31 quires by 8 folios per quire, we have a total of around 248 folios missing from Marciana 841: a staggering number.

2§9 The quire numbers in Marciana 841 also shed light on production of the manuscript. When recording the location of quire marks in Marciana 841, I found a troublesome gap of 15 folios without any quire mark. From folio 1 recto to 57 recto, the scribe wrote eight quire marks on the front (recto) page of each new quire. The gap of 15 folios without a quire mark occurs from 57 recto (quire 39), until one finally appears on 72 verso (quire 40)—on the back page of the next quire. Not only did the quire mark switch from recto to verso, but it took on a new appearance; the quire marks recorded on the recto side do not have bars or hooks written over them, but the quire mark on 72 verso does. These marks continue on the back side of the last page of the quire until 140 verso—where the scribe switches back to recording on the recto side of the quire, on 141 recto. After these two facing quire marks, the marks continue to appear on the recto side of the first page of each quire until the end of the manuscript: quire 50 to the final quire, 55.

2§10 This clearly reflects the work of two different scribes. The first scribe numbered all quires—in regular quaternions of eight folios per quire—until quire 40. Here, someone else took over. This scribe numbered quires differently than the previous scribe: on the last page of the quire rather than on the first page and flagging the numbers with a bar-and-hook mark. At quire 50, perhaps the first scribe came back and saw the work which had been done—and gently reminded, I imagine, the second scribe which method he preferred. Thus, although from quire 50 until end-quire 55, the quire marks appear on the recto side, they retain the bar-and-hook symbol above each mark.

2§11 It is also possible that the switch from marking on recto sides to marking on verso sides could reflect a secondary level of organization or stacking of the quires, for the marks seem to switch from the front to the back page every ten quires. Quires 32-40 are on recto, 40-49 on verso, and 50-55 on recto again.

2§12 Both of these theories may apply to the quire markings in Marciana 841. Owing to the distinct appearance of the bar-and-hook symbol on quire marks 40 onward, it is likely that two scribes worked on the manuscript’s physical layout. However, the jump from recto, to verso, and back again to recto, might reflect a general practice in the marking of quires in a codex. An inquiry into the quire divisions in other Byzantine manuscripts from a similar time period might reveal more definitively whether this pattern is attested elsewhere.

2§13 It is also interesting to note that the extant manuscript begins with a labeled quire, 32. As the quire contains a full quaternion of 8 folios, is this an indication that the manuscript was separated from its no longer extant contents along a pre-existent seam? Or, was this an instance of pure chance? Further examination of the manuscript will tell.

Example of What Quire Arrangements Can Reveal

2§14 There are many things that quire arrangements can tell us about scribal practices and their sources. During the summer of 2011, a team of three researchers, working primarily on Iliad 1, also examined large numerals in the margins of a manuscript called the Venetus A. The  Venetus A is a tenth-century Byzantine manuscript also located in the Biblioteca Marciana and is our oldest complete source for the Iliad. These numerals in the margins of the Venetus A numbered what are known as “epic similes” in the Homeric poetry, an extended simile that goes on for several lines. As our researchers recorded the visual evidence of these numerals, using urn citation to make them uniquely identifiable, it became apparent that 35 out of 193 similes were unnumbered. One of the largest gaps present extends from simile 13 to simile 38. The last number present is 12 on folio 33v Simile number 12 and it returns again with 39 on folio 60r Simile number 39. This gap in numbers can be explained by understanding where the quires falls in the manuscript. The missing numbers would have been contained in quires 4 through 6. No simile numerals appear in any of these three quires. The lack of numerals shed light on how the scribe composed his manuscript. If the scribe wrote this manuscript in several passes, it becomes perfectly possible that he could have accidentally skipped three quires as he was numbering the epic similes. The gap also reveals new insights into the scribe’s sources. If the scribe was numbering similes of his own initiative, the numbers would likely pick up and not skip over 13 through 38. The fact that the numbers do not continue sequentially means that the scribe must have been copying from an external source, perhaps a lost work on Homeric similes. Without an understanding of the quire arrangement, we would not be able to draw fascinating conclusions such as these.

How were text and commentary organized and laid out?

Examples from the Venetus A manuscript

Nik Churik

3§1 Marilena Maniaci has studied how the scribe of the Venetus A manuscript organizes the scholia in relation to the folio page and the body text (see “Words within Words”). Maniaci’s work focuses largely upon how space is used and managed on folios dense with scholia. She proposes metrics for assessing how the scribe balanced the competing needs of including large quantities of commentary and placing the commentary within close visual proximity of the text it comments on. While the “hidden rules” she proposes generally apply to folios of the Venetus A, by examining less heavily annotated folio pages, we can see that they need to be modified to account for the range of practice we can observe across the whole manuscript.

3§2 We propose that the arrangement of scholia in relation to the text body follows a regular pattern. Scholia are placed in one of three zones: the first six or so lines of Iliadic text are commented upon in the top border region of page; the final six lines or so of text are commented upon in the bottom border region; the middle thirteen or so lines are commented upon in the exterior margin. In this middle region, the first scholion will usually be adjacent to the line on which it comments, and then the subsequent scholia will follow in order.

3§3 If we look, for example, at folio 95 recto, there are only two main scholia: one for line two on the page, the other for line twenty-one, and with the exception of two intermarginal scholia in the middle of the page, no other text constrains the scribe’s choice of placement. The scribe could have placed the scholia directly adjacent to their respective lines. Yet he does not do so, and here, that is clearly his choice, not a necessity imposed by lack of space. This suggests that the division of the commentary into three zones is a standard practice expected by scribe and reader: the placement of scholia follows a sort of protocol for situating the scholia in relation to the text.

3§4 Within the largest region, the exterior, middle zone, the placement of individual scholia is sometimes adjusted to accommodate a large quantity of material. While scholia in this zone usually begin adjacent to the first line commented on, when the scholia would exceed the space available if this were followed, the scribe moves scholia up far enough to fit all of the scholia as close as possible to the lines they refer to. This is illustrated on 100 recto: the first scholion begins four lines above the line to which it refers, and then the rest continue, more or less immediately following each other. The scholia referring to line 18 begins the scholia in the bottom region.

3§5 A second variation from the standard is seen on 97 recto, where all of the scholia for that page, for lines 2, 4, 5, and 7, are in the top region. This exceeds the usual limit of placing commentary only on the first six lines in the top zone. Perhaps this is done because the seventh line is still relatively near to the top and this would have been the only scholion in the exterior zone: since the reader has already been moving from main text to top zone, placing this scholion in the top zone lets the reader complete this section of commentary without having to shift zones. The protocol for placement provides readability for the reader; that is to say, as much as possible, it seems the scribe tries to make the progression of scholia obvious, accessible, and continual.

3§6 The scribe’s consciousness of space is very clearly illustrated by comparison between 94 recto and 96 verso. On 94 recto, the first scholion is at the top of the page and comments on the third line. The subsequent three scholia, all commenting on the sixth line, are also on the top. On 96 verso, the first two scholia comment on the sixth line, yet the first scholion begins adjacent to line six and the subsequent scholia continue from there. The placement keeps the three scholia together and thereby minimizes the distance between the scholia and the lines they comment. This, again, demonstrates the scribe’s understanding of ease of reading for the reader.

3§7 An unusually placed scholion begins 99 recto. The scholion for the second line lies at the second, not at the top of the page. Reasons for this are not apparent; the content, which mentions the exchange of ἀντι for ἀποκρίνονται. This, however, cannot be said definitively because of the lack of comparisons.

How were text and commentary organized and laid out?

Escorial Upsilon 1.1

Rebecca Musgrave and Brian Clark

4§1 The scribe of the Escorial manuscript Upsilon 1.1 chose to lay out the scholia in a different way than the scribe of the Ventus A. In contrast to the Venetus A, Upsilon 1.1 only has main scholia written in the hand of the original scribe. The interlinear, intermarginal, and exterior scholia which is found on the Upsilon 1.1 is written in another hand, and the current focus of the teams at Holy Cross is the scholia which is written in the hand of the original scribe. The main scholia on the Upsilon 1.1 wraps around the main text of the Iliad, similar to how the main scholia does in the Venetus A. The scribe of the Upsilon 1.1 linked the scholia to the text which the scholia comment on using numerical scholia markers, which function similarly to footnotes. The scholia are numbered sequentially from the verso page to the recto page. This makes sense, since the manuscript was meant to be open in front of the reader, and he would be able to see both pages simultaneously. The photographs of the manuscript were taken folio by folio, so one of the first projects of the teams at Holy Cross was to compile bifolio images of the Upsilon 1.1 so the manuscript can be viewed as it is meant to be, like an open book in front of the reader. Here is an example of a typical bifolio image of the Upsilon 1.1. The scholia wraps around the main text of the Iliad, and it ends on the bottom of the verso page with the number 12 and continues sequentially on the recto page with the number 13While compiling these bifolio images, our teams came across several irregularities, and examining these on their own and comparing the folio on which they are found to the corresponding folio in the “twin” manuscript of the Upsilon 1.1, the Venetus B, have given us insight into the process of how scribe of the Upsilon 1.1 laid out the manuscript and worked with the source material available to him.

4§2 Occasionally the scribe would make an error while numbering the scholia on one folio, but this mistake would not carry on to the following folio. One example of this appears on folio 10 verso – folio 11 recto. The scribe places two 11s on 10v. The first scholion numbered 11 has an end of scholion marker, and the second scholion numbered 11 also has an end of scholion marker. These two 11s are followed by a scholion numbered 12, but on the following page, the scholia begin with a scholion numbered 14. This mistake suggests that the scribe of Υ.1.1 would organize the scholia in single folios so that if he did make a mistake, as he did in this case, the mistake would not carry over to the following folio.

4§3 On occasion “double numbers” appear in the main scholia. One example of this can be seen on folio 66 recto. The double numbers appear at the top of the page, where the scribe wrote two 13s and two 14s. The first 13 appears in the interior margin of the manuscript, while the second 13 appears within the text of the main scholia. This happens again with the number fourteen: the first 14 appears in the interior margin, while the second 14 appears within the text. One possible theory to explain this was that the scribe accidentally put in two scholia with the same number, but this was easily disproven by counting the end of scholia markers and determining that indeed two numbers exist for a single scholion. The second theory is that the scribe would occasionally place extra numbers in the interior margin to make the manuscript more accessible to the reader. This theory indicates that the Υ.1.1 was not a luxury manuscript meant to be put on display; it was meant to be used, and the scribe sacrificed some aesthetic qualities to make the scholia more easy to be read by placing extra numbers within the margin of the folio.

4§4 Irregularities on another folio in Υ.1.1 gave further insight into how the scribe went about writing each page of the manuscript. On folio 88 recto the numbering within the main scholia is very irregular; however, the corresponding scholia markers within the text are numbered sequentially in the correct order. The numbering within the main scholia goes 15, 16, 18, 12, 17, and 20. The numbering continues on regularly after this. The numbering within the text of the Iliad is normal, going sequentially from 15, 16, 17, 18, 19, to 20. This indicates that the scribe would write the main scholia separately from when he would fill in the corresponding numerical scholia markers, since if he had done them simultaneously, this mistake would have carried over from the numbering of the main shcolia into the numerical markers within the text.

4§5 Often when irregularities are encountered in the manuscript of the Υ.1.1, the corresponding folio of the Venetus B is consulted to compare how the manuscripts are either similar or different. The Venetus B manuscript has often been considered the “twin” of the Υ.1.1 because of its similar layout; however, our research with these two manuscripts has revealed that the relationship between these two manuscripts is more distant than that of a twin, and perhaps calling it a “cousin” would be more appropriate. One example of the differences between these two manuscripts can be seen by comparing folio 88 recto of Υ.1.1 to folio 91 recto of Venetus B. The numbering of the main scholia in the Venetus B is also irregular, but not as irregular as in the Υ.1.1. In the Venetus B, the numbering goes 15, 16, 18, 19, 17, and 20. The numbering of scholia markers within the text of the Iliad is normal just as it is in the Υ.1.1. The numbering of the Venetus B makes the mistake in more clear. In the Venetus B, only one number is out of order, the 17. And if the numbering of the Venetus B is compared with that of the Υ.1.1, the only difference is that in the Υ.1.1 the scribe wrote a 12 whereas in the Venetus B the scribe wrote a 19. Since the letters θ and β look fairly similar, especially in the scribe’s handwriting, the scribe of Υ.1.1 may have misread the 19 for a 12. This could indicate that both the Υ.1.1 and the Venetus B were working from the same source which had the irregular numbering. The other possibility is that both manuscripts may have been working from separate source material which had copied from the same source, so the mistake may not have been done directly by the scribe of the Υ.1.1 but by the scribe of the source material from which he was copying.

4§6 Through our research, we have acquired a deeper understanding of what we call “scribal choice.” Previously, these scribes have been thought of as mindless copy machines that were merely reproducing exactly what they had in front of them, rather than active participants in the construction of their own texts. In this example, found on 111v of the Upsilon 1.1, you can see that this folio is absent the usual numerical footnotes at the start of each scholion. Upon closer inspection, you will see that only those footnotes that should appear in the outer margin of the page are missing, as 2 (β’) and then eleven through fourteen (ια’ – ιδ’) are present within the block of scholia text. The scholia that are without footnotes are, however, set off in a different manner. The first letter of each scholion is out-dented from the rest of the scholia text.

4§7 It is likely that whatever source from which the scribe (or some scribe before him) was copying did not have the numerical footnotes on this particular folio. When these ancient manuscripts were rebound, the edges of the pages were often re-cropped. Occasionally, this process would cut off the text that appeared at the far edges of the page. A likely scenario is that the source from which the Upsilon 1.1 was copied had a scholia layout similar to the Venetus B, where the numerical footnotes are slightly removed from the text of the scholia. When the source was rebound, it is possible that the footnotes were cut off. The scribe of the Upsilon 1.1 most likely decided to out-dent the initial letter of each scholion that began in the outer margin so that the reader would still be able to distinguish each scholion from the next. This explains both the irregular layout, as well as the appearance of the numerical footnotes with the scholia that do not begin in the outer margin. This observation refutes the claim that the scribes were not engaged in the creation of their edition of the Iliad. These irregular occurrences suggest that the scribe was an active agent in the process and worked to both stay faithful to his sources by copying exactly what he saw, as well as making it easier for the reader by adding his own method with which to differentiate the individual scholia.

4§8 A further topic of interest that our research has shed a great deal of light on is the relation between the Upsilon 1.1 and the Venetus B. Scholars have long believed these two manuscripts to be exact twins, based largely on the fact that they share an identical bi-folio layout with numbered scholia linking to the text. This example, found on 59v of the Upsilon 1.1, supports this claim and demonstrates that both manuscripts are a compilation of different sources. You can see that this folio begins with the usual numerical footnotes, but then has two additional “dingbat scholia makers,” or a symbols that act as footnote markers. After these two dingbats, the numerical footnotes continue in proper sequence. The dingbats interrupt the numerical footnotes, but do not take the place of any one number. Additionally, the dingbats function exactly like the numerical footnotes, as they appear intra-linearly in the Iliadic text and connect the scholion to a specific part of the text. The occurrence of the different type of scholia within the typical scholia suggests that the scribe was working from multiple sources in order to compile his edition of the Iliad.

4§9 When we find irregularities in the Upsilon 1.1, we compare them to the corresponding folio in the Venetus B, the supposed twin. In this instance, the matching Venetus B folio, 60v, is laid out identically to the Upsilon 1.1. The dingbat scholia appear in between the same numerical footnotes, and are above the same words in the Iliadic text (ὀρυγμαδὸς and ῥέε – 8.63 and 8.65). This demonstrates that the two scribes were working from identical source material, as they both included the same information in the same manner.

4§10 However, other aspects of our research have altered the strength of the assertion that these two manuscripts are indeed exact twins. A revealing difference can be found on folio 97r in the Upsilon 1.1 and 101r in the Venetus B. Both manuscripts typically have twenty-five lines per folio. On 97r of the Upsilon 1.1, the last five lines have hooked marks at the beginning. These five lines, which would appear to be 7.413-7.7.417, do not appear at this location in our modern text of the Iliad. However, the lines do appear at 7.430-7.434 and are repeated in their rightful (by our modern edition) location on the next folio. We believe that the hooked marks indicate that the authenticity of these lines at their location on 97r is questionable. The scribe made an active choice to include the lines, while still differentiating them from the commonly accepted text.

4§11 The corresponding folio in the Venetus B, 101r, provides a remarkable comparison. Instead of the standard 25 lines, this folio has only nineteen. There is no reference to the five extra lines that, if the scribe had wanted, could have appeared at 7.413. Yet, the scribe must have at least been aware of the existence of these lines, as he has chosen to write five less lines on this folio. This comparison shows that both scribes were working from a source that had these five extra lines, yet they both dealt with the same information differently. The scribe of the Upsilon 1.1 included the lines, yet set them off as different from the main text, while the scribe of the Venetus B chose not to include them at all, but skipped lines in his folio so as to ensure that his edition would line up with his source.

4§12 Our research has taught us a great deal about the workings of a medieval scribe. Through comparisons between the Upsilon 1.1 and the Venetus B, we have come to view the scribes as active participants in the text, instead of passive copiers. Further, we have altered the view that these two manuscripts are in fact identical twins. In light of the differences that we have found, we believe that these manuscripts should be viewed more as cousins. They share a common layout and identical source material, yet the two scribes repeatedly took individual paths when presented with the same information. We look forward to continuing our research and learning all we can from these ancient manuscripts.

How were scholarly notes, or scholia, edited?

Sources and working methods of the Venetus A scholiast

Matthew Angiollilo

5§1 One of the benefits of having the digital photography of a number of Homeric manuscripts is the ability to quickly make comparisons between multiple texts, seeing exactly what is on each folio. This development is not only more convenient but allows the examination of material that previously could not be viewed simultaneously and can yield results as to the types of sources and the editing process of the scribes who composed these manuscripts, in this case the Venetus A, and the Upsilon 1.1. To illustrate this point, there is a vast body of examples, of which I have selected a couple. One occurs in scholia in both the Venetus A and the Upsilon 1.1 referring to line 53 of Iliad, Book 2. At this stage of the narrative, Agamemnon is calling together an assembly of the Greek kings and recounts the description of the dream he had the night before, proclaiming that now the Greeks would be able to destroy Troy and bring the war to an end. Line 53 begins with the word βουλὴν, the council Agamemnon is calling and the word the scholion will focus on Venetus A, folio 25 recto. The Venetus A has two scholia on this line, whereas the Upsilon 1.1. has one, a larger note, which combines the information contained in the two Venetus A scholia Upsilon 1.1, bifolio spread 19v-20.r. It is also notable the sources that are mentioned in each of these notes.

5§2 The first scholion for line 53 in the Venetus A, mentions the editions of all three notable Alexandrian commentators, those of Aristarchus, Zenodotus and Aristophanes, who were scholars at the Library of Alexandria whose work is preserved largely in these manuscripts and discusses their approach to a textual problem in this line Venetus A, folio 25 recto. This is a rare occurrence in the scholia of the Venetus A, as often one of these editors will be mentioned as a source for a reading or interpretation of a text, and occasionally two of their views will be contrasted. The naming of all three editor’s editions in the same scholion is unusual. In the case of this scholion the point of disagreement is a minor one, namely whether the “Ν (nu)” on βουλὴν should be removed or remain in the line, with Aristophanes and Aristrachus removing it and Zenodotus allowing the letter to remain. In addition to the work of these three editors the scholiast also mentions other versions of the Iliad that he has access to and is able to employ for comparison to arrive at a correct reading. The scribe includes the opinions of αἱ πλείους “the majority of editions”, and the χαριέσταται “the most elegant” versions, which were those edited previously by scholars. The scholiast also indicates the reading of ταῖς κοιναῖς “the common” editions, which was a version that was viewed as authoritative but relatively uncorrected text as to how to handle this question, with the common edition leaving in the “Ν (nu)” and the majority and more elegant editions leaving it out. Although the textual issue is a minor one, this scholion nicely displays the wide variety of sources on hand for the scribe to consider and from which to draw conclusions, and reveals that he, in the course of his work, is choosing which opinions to include, ignoring some and condensing others.

5§3 If we turn to the Upsilon 1.1 and the larger scholion written in that manuscript we see how the scholiast treated the information found in the Venetus A scholia again stating that “the majority” and the “more elegant” editions redact the “Ν (nu)”, along with Aristarchus, whereas Zenodotus keeps it in Upsilon 1.1, bifolio spread 19v-20r. However, the Upsilon 1.1. contains, unlike the Venetus A, (and you can see this is a larger note) a lengthy description of the leaders being assembled and their nationalities, which the Venetus A editor does not include. A comparison of the content of the scholia from these two manuscripts reveals that the editors of the Venetus A and Upsilon 1.1 had similar sources on hand, that relayed like information regarding a certain textual problem in the Iliad. These examples help to illustrate the working method of the scribe and reveal that his task was not that of a mere copyist but one where the amount and variety of material at his disposal necessitated that he be selective and exercise judgement as to the content of is notes.

5§4 Another example that indicates the sources that were used by these scribes and their practice in editing scholia in there respective manuscripts is later in Iliad, Book 2 in scholia at line 258 in both the Venetus A and the Upsilon 1.1 pointing to a variety of different readings for this verse. At this stage of the narrative, Odysseus is scolding Thersites for a speech delivered against Agamemnon. The scholion centers around different ways to begin this line, which begins as we can see from the lemma reads “εἴ κ᾽ ἔτι σ᾽ ἀφραίνοντα” Venetus A, folio 29 recto. Before delving into the information recorded in the scholia, it is interesting to note that another type of version of the Iliad were the “city editions” or *politikai*, which were edited editions of the *Iliad*, that were associated with certain locations throughout the Greek world, such as those from Chios, Argos, Cyprus and Massalia. In the scholion to this line in the Upsilon 1.1 the writer lists various opinions on the correct way to begin the verse Upsilon 1.1, bifolio spread 24v-25r. Aristarchus prefers “εἰ δ’ ἐτί σε” whereas Zenodotus has “εἴ κ’ ἔτι.” The attention then shifts to the end of the line and the scholiast indicates a number of different variations of how different editions complete the line. The city edition of Sinope ends the line “ὡς τὸ πάρος περ”, the edition from Massalia ends the line with “ὕστερον αὖτις.” Finally the last option given is from the edition of Philomen who ends the line with “ἐν Δαναοῖσιν.”

5§5 Turning to the note on this line in the Venetus A Venetus A, folio 29 recto, the scholiast interestingly does not identify content as being from Aristarchus or Zenodotus, nor does he mention the city editions and the variations they provide for the ending of line 258 by name. However, the Venetus A scholiast does indicate that he has knowledge of the editions of the Alexandrian scholars as well as those of the city editions as the Venetus A scholion appears to be reacting toward those versions. The Venetus A note begins “ὅτι περισσὸς ὁ κε” “because the “κε” is superfluous.” Remember the version of Zenodotus began the line as “εἴ κ’ ἔτι” and the scholiast calls this reading into question dubbing it as unnecessary. This finding will also support the claim of Tom Arralde in the next segment that “oti” scholia seem to designate Aristarchan content as the scholion from the Upsilon 1.1 indicates clearly that Aristarchus’ reading for the beginning of this line was “εἰ δ’ ἐτί σε” a beginning without the “κε”, which would seem to indicate that Aristarchus was rejecting the Zenodotean reading.

5§6 The scholion from the Venetus A then states the alternative line ending “ὕστερον αὔτης” Venetus A, folio 29 recto. This was the ending for the line given in the Upsilon 1.1. scholion as belonging to the city edition of Massalia, which the scholiast of the Venetus A does not mention by name in the text but seems to acknowledge as an alternate reading for this line. The Venetus A also contains the reading of Philomen and it becomes clear that the editors of both of these manuscripts had a number of different versions of information at their disposal, but also that in regard to some scholia, the information they select to relay is similar, which may suggest analogous sources. Also in comparing a variety of manuscripts we are able to better identify the sources of Iliadic commentary. By looking to a scholion in the Upsilon 1.1, it is possible with greater likelihood of identifying the origin of certain versions, in this case locating references to Aristarchan and Zenodotean commentary, along with the text of the city edition of Massalia in the Venetus A, even though the sources for this text are not identified by name.

How were scholarly notes, or scholia, edited?

Sources and working methods of the Venetus A scholiast

Thomas Arralde

6§1 To continue with Matt’s theme of identifying the sources of the Homeric scholia, I would like to look at a specific type of scholia, which Matt alluded to. The point I would like to make is two-fold: first, that we can identify material written by Aristarchus in the Venetus A based on the phrasing of the scholia, and second, that scholiasts were not blindly copying from a compilation of Alexandrian material, but instead working dynamically with their sources, which might have even been original material. The test case I would like to use to support these claims is what we have named hoti scholia in the Venetus A.

6§2 Matt mentioned Aristarchus before, but I would like to go a little more in depth about why Aristarchus is so important, why identifying his commentary can be problematic, and how his work was arranged. Aristarchus’ work is seen as a turning point in the history of the Homeric Tradition, in which the works of Homer became relatively rigid and texts were studied outside the context of performance. Because of Aristarchus’ pivotal role in the textual history of the Homeric epics, identifying his comments from what does survive is an important task for Homeric scholars. Doing so is not always straightforward, however, since scholiasts did not always write the name of the scholar whose work they were referring to, as Matt also mentioned.

6§3 As for the format of Aristarchus’ work, it is supposed to have been as you can see here: critical sign, then lemma, and then comment. Critical signs are marks that Aristarchus put next to interesting lines in the Homeric epics that he wanted to comment on, lemmata are quotations of the specific passage referred to. Aristarchus employed critical signs so that the text of the epics and their commentary, which were written on two separate scrolls, could be easily matched. With the advent of the codex, the use of critical signs next to both line and comment became superfluous, and were mostly removed from the scholia but retained next to the text. Evidence of Aristarchus’ use of these signs is the presence of the word hoti at the beginning of certain scholia, on its own meaning something like “the sign is here because” (although sometimes the name of the critical sign is written out, for example: he diple hoti, the diple because).

6§4 In our work on editing the scholia of Book 1 of the Venetus A, we noticed that the word hoti often began scholia, and that these scholia often continued with the name Zenodotus, the first publisher of an Alexandrian edition of Homer, as you can see here. Our consistent documentation of when Zenodotus appeared in the scholia allowed us to observe how frequently his name appeared in the context of hoti scholia.

6§5 What we found is that Aristarchus’ name almost never appears in these scholia, but Zenodotus’ does quite frequently. This fact made us consider the possibility that these scholia are direct quotations from Aristarchus’ lost works. It makes sense that the scribe of the Venetus A is quoting directly from Aristarchus in the hoti scholia since Aristarchus would not refer to himself in the 3rd person in his own edition, and since he would be commenting on the edition of his predecessor, Zenodotus. Later commentators referring to Aristarchus’ works would cite Aristarchus as we see in other scholia, writing simply Aristarchus, or houtos Aristarchos (thus Aristarchus). These different formulations of scholia suggest that the scribe might not have copied from only one source, but from different sources speaking in different voices.

6§6 Possible evidence of different sources in the scholia may be seen not only in the differences between the types of scholia, but within individual scholia themselves. To illustrate my point I would like to look at the Venetus A’s treatment of line 700 in Book 5 of the Iliad. The main scholion referring to this line begins houtos Aristarchos (thus Aristarchus), a beginning characteristic of post-Aristarchean commentary. Later in this same scholion, however, a new sentence starts he diple hoti (the diple because), a beginning characteristic of what we think could be Aristarchus’ own material. This single scholion, then, could have been composed from two different commentaries; the juxtaposition of two distinct scholia types makes it difficult to believe that it was copied from a single homogenous text.

6§7 An exciting find from last summer was that line 2 in book 2 of the Iliad had extremely similar commentary in both the Venetus A and the Y.1.1. This scholion, at parts word for word in the two manuscripts, begins in the Venetus A with ὅτι, and in the Y.1.1 with Ἀρίσταρχος φησι (Aristarchus says). The fact that the Venetus A scholion has more content than the Y.1.1.’s, as you can see, agrees with what we have seen elsewhere in their scholia, namely that, since the Venetus A is a luxury manuscript, it has a more expanded treatment of scholarly material when there is overlap between the two. This find, more than anything, supports the idea that ὅτι scholia are genuine Aristarchean material, handled differently by different scholiasts, and not copied unthinkingly from a single scholia compilation. If scholiasts did copy from a compilation, it was a dynamic process of selection and incorporation, but it is also possible that these choices were made from the original works of Alexandrian scholars.


Leave a Reply