Between parchment, paper, and theories. Some examples from the Biblioteca Nacional.
In looking at manuscripts these past days, it has become clear that my previous theories about materiality, the urge to reconstruct text, and the concept I have tried to develop called use are generally worse than I thought. I have come to the conclusion that in terms of textual fragments, the pressing question is not whether the use of typologies is appropriate or not to begin or to serve as the frame for a study about fragments, but rather: How is it that we and the manuscripts themselves construct and deconstruct these typologies?
I began some months ago with a very elementary fragment definition and typology. The typology was similar to the following:
Fragments not intended to be fragments
Pieces of texts collected in what seem to be thematic anthologies
Parts of a given work copied in 1 volume in which it is difficult to identify why, exactly, they are present in one volume.
References to a text in another work
Fragments intended to look like fragments, but are actually indeed whole. Etcétera.
It was largely based on Gumbrecht (Powers of Philology, “Eat your Fragments!”), reinventions of typologies and definitions presented in several essays on classical fragments from a volume called Collecting fragments-Fragmente sammeln (Göttingen: Vandenhoeck & Ruprecht, 1997) of which “Eat your Fragments!” is a part, and several observations about metonymy, particularly the notion that a piece of something whole suggests both an inability to access the past and our only way to come in contact with it. I developed the fragment definition by looking into some spatial theories (Bachelard, theories of ruins in the Romantic period and the works of romantic authors (Goethe and Schiller) and decided that most attempts to classify fragments rely on the idea that a fragment is an incomplete text. I contrasted these textual fragment typologies with the images that the word fragment can evoke, bits of broken off parchment, glass, clay, stone, or fabric. In an attempt to have manuscript text retain its more physical, parchment, paper and ink character, I began to propose the concept of material text, which insists on the inseparability of text from its material support. As a result of more manuscript experiences, I now have come to regard material text as a useful means by which I might understand the inclination and in other cases, ardent desire, to put documents back together and to restore their text. My present conclusion is that at the root of this impetus is not always an obsession with deciphering and a want to assign meaning. Several days ago, while looking at a Fernán Pérez de Guzmán anthology incomplete at both beginning and end like so many manuscripts, I studied the reconstruction of the last stanza of the incomplete copy of the Setecientas that is extant in the manuscript. The last folio is torn on the diagonal, but has been restored, with new paper attached to the torn. In modern pencil a reader has mimicked the gothic letter of the scribe and completed the missing half of the verses. Here, the need to decipher is not urgent; with great certainty, we can complete the verse and propose a meaning, either from memory or from the consultation of another of the manuscripts or printed editions. At least the textual reconstruction, if not in part the addition of new paper, was motivated by aesthetics, not hermeneutics. The space, for the reconstructor's eye and those with whom he consulted, was to be filled with text, and not just text in any letter, but by one that matched, in so far as he was able or was interested, that of the scribe.
Each manuscript, in and as a result of our use of it, constantly creates and destroys its own classification. BN Mss. 9264, which contains the first book with several alterations of chapters of Martín Pérez’s Libro de las confesiones (ca. 1330), serves as an interesting example. The first book addresses sins common to all classes and condition of people, the second those particular to determined groups, primarily clerics, and the third discusses the sacraments.
This testimony is a complete version of the first part of the Confesiones, but a fragment of the entire work, as the author intended according to the prologue present in other manuscripts,--which, along with the tablas that were to precede each section, are not included in BN 9264.
In his description of the codex that accompanies his edition, Antonio García writes the following, which I cite here as a means to reflect further on the definition of fragment:
Este códice de la Biblioteca Nacional no contiene el prólogo ni las tablas, pero incluye íntegra la primera parte y es una buena copia. Lo que sucede es que en el folio 32v el amanuense saltó del c.33 al 41 y copia el final del códice, en los fol. 134r-139v, los capítulos 34-40 que había omitido en el lugar correspondiente. No se trata, pues, de un fragmento, sino de una copia completa, aunque con las anomalías indicadas. Faltan también los fol.2rv, 25rv, 95r-97v, que están con hojas en blanco. Este códice es el más cercano al de la versión portuguesa de la primera parte.
Martín Pérez, Libro de las confesiones, Una radiografía de la sociedad medieval española, Antonio García y García, Bernardo Alfonso Rodríguez, y Francisco Cantelar Rodríguez eds. (Madrid: Biblioteca de autores cristianos, 2002) XV.
A fragment is thus defined in part by how it compares in completeness to other testimonies. Since there is no extant codex that contains a complete version of the Libro de las confesiones, each part, for Antonio García, represents a complete. What he calls fragments are those that are more incomplete, in this case, much more incomplete: two folios of the first book copied in the 14th century at Montserrat, Biblioteca del Monasterio, MS 1076, and a 15th century guard sheet at the Biblioteca de Calbildo, Toledo, containing chapters 14-16 of the first chapter of the first book.
At the same time, however, without the prologue and the tablas, this work, at least the person who decided to use it in this way, stands alone. More interesting is the strange coincidence that the alternation of chapters provides quite a definitive end to the text. The last sentence essentially theorizes, albeit probably entirely accidentally, that this manuscript is not fragmentary, but whole, at least in terms of the use of the text that it proposes:
¶ Estas cosas son aqui puestas por que façen al confesor mester ¶ E si algunas cosas aqui fallesçen, demandelas a los letrados o enbie alla sus confesados o al obispo, ca non ca non se pueden todas las cosas escriuir (139v).
Assuming that fragmentation is always a violent act and defining fragment in terms of the discrepancy between how an author intended a work to end and how it appears in a given manuscript avoids too many of the complexities of fragmentation. From the evidence we possess, it is clear that at least in some manuscript traditions and maybe in many more, the Libro de las confesiones did not circulate in its complete form. The work and what appears to have been the original intent has been changed according to its use, or did not prove most useful for those who produced and were managing the text in its material form.
There is one manuscript that I have asked for on 4 separate occasions. It is a compilation of wisdom texts, the first of which consists of a paragraph introducing some supposed Aristotelan advice:
[Q]uando aristotiles en greçia fue casado que non pudo yr con su criado alexandre en las huestes nin en los logares por do el yua fazia le muy gran mengua et dapnno et enbiole alexandre su carta en que le enbio rrogar por ser mas sano. E otrosi que le enbiase commo podiesse conosçer las naturlaezas de los omes por que les naturalezas conosçiese a cada vno si fuesse bueno o malo. E aristotiles enbio gelo por escripto en esta manera que se sigue.
Immediately following the que se sigue is the tabla for another work that has been identified. The codex also contains what H.O. Bizzarri has identified as the only extant copy of a particular version of the Libro de los buenos proverbios. Other texts seem to be spliced together to form new “complete works” or to at least appear that way, maybe even in part, for aesthetic reasons. The first text (transcribed above) might be continued on 32v, but it may be a different collection of Aristotelan or pseudo Aristotlean advice altogether. All of this is yet, or never to be determined with any certainty. But I will keep asking.
The question in the end is not simply, What is a fragment? or What type of typology do we need to talk about fragments?, or Can we really make a fragment typology?, but rather how does each use of a material book, in a given moment, remake, destroy or force a reevaltuation of what a work is, what a version is, and what rewriting is, that elusive word that seems to serve as a way of avoiding coming to terms with a process that is never the same. It is possible that this is not about fragments at all.
