A full repertory of conjectures on Catullus is presented here in a critical apparatus alongside his poems. It is published in the first instance online in order to make it accessible for the widest possible public, and to render future updates possible. An edition on paper with a detailed introduction is also being prepared.
Every effort has been made to include every conjecture on Catullus that has been made so far, with certain limitations that are set out below. Whoever publishes a new conjecture, or discovers one that has been omitted or wrongly ascribed, is asked to contact the editor.
The repertory aims to include all conjectures on Catullus that have been published in print since the editio princeps of 1472. Hitherto unpublished conjectures by scholars who are no longer alive, and readings of interest that arose not as conjectures but for example as misprints, are included at the editor’s discretion. The repertory is not intended to serve as the first place of publication for new conjectures; those who would like to publish one are directed towards specialist journals such as the Classical Quarterly, Exemplaria Classica, and Paideia.
The conjectures and variants that first appear in Catullus’ 120-odd manuscripts except TOGR, known collectively as the codices recentiores, call for special comment. As of January 2013, there exists no comprehensive study of these manuscripts; our knowledge of their origins and relationships remains limited, and a full stemma codicum of Catullus still has to be drawn up. As a result, many questions remain open that have a bearing on the work of the editor. It is still debated whether the recentiores offer any independent evidence for the text, that is to say, whether they present readings that are derived not from TOGR, but from an older manuscript that no longer survives. This editor knows of no convincing evidence for any such readings. On the other hand, the text of most recentiores stems clearly from OGR; so any attractive reading in the recentiores is far more likely to be a humanistic conjecture than a relic that has been preserved through a lost manuscript that was independent from OGR. It is assumed in this edition that all innovations in Catullus’ renaissance manuscripts are due to conjecture, or the usual processes of textual corruption.
Humanistic conjectures present further problems. Many of them first appear in the 100 or so manuscripts of the poems of Catullus that were copied in Italy in the 15th century. It has already been noted that the stemma codicum of these manuscripts has not been drawn up yet; nor will it be easy to do so, given the extent of contamination (the mixing of readings with different origins) at this stage of the manuscript tradition. As a result, in the case of most fifteenth-century conjectures on Catullus we cannot tell who first made them or where. And given the fact that many manuscripts have not been dated securely, we cannot put our sources for the text into a chronological sequence, so we cannot even tell when a conjecture appears for the first time. This, and the fact that we are dealing with over 100 manuscripts, makes quoting the humanistic conjectures on Catullus rather more complicated than one would wish. If we do our best to assign each recent conjecture faithfully to its author, we may want to apply the same standard to conjectures from the Renaissance; but if we fall short, it may not only be due to our own weaknesses, but also to the difficulty of the material that is at stake.
An elegant solution to this problem was proposed by R.A.B. Mynors in his 1958 Oxford Classical Text: he used the Greek letters αβγδεζηθ to distinguish eight layers of humanistic corrections in Catullus’ manuscript tradition. But on closer inspection his solution turns out to be unsound for a variety of reasons. For one thing, these letters stand for very different entities: α and β each for one single manuscript, γ and δ for groups of codices that are only distantly related at best, ε and ζ for what appear to be genuinely related groups of manuscripts, and η and θ each for one single manuscript and what appear to be its descendants. In the case of every manuscript, and in this case most notably of α and β, there is a fundamental difference between the conjectures that appear in their text, those that were added by the first hand as variants or corrections, and those that were added by later hands: the first group may well pre-date the manuscript; the second group may stem from the exemplar, or from another manuscript, or they may be the own creations of the scribe; and the third group need not have anything to do with the first two, and it may be much later. Furthermore, many conjectures that Mynors attributes to γδεζηθ already appear in earlier manuscripts, especially in undated ones. All this makes Mynors’ use of Greek letters to indicate layers of contamination in the tradition thoroughly problematic. His system has been abandoned here, and the editor suggests that it should be used with great caution, if at all, unless a more detailed study should reveal that it does have some merits.
Here a purely chronological approach has been used: the editor has tried to quote the earliest known attestation or attestations of each conjecture. This has been a challenge not only due to the amount of collating involved, but also because the date of many of the codices recentiores is uncertain. Future research may well cast more light on Catullus’ humanistic manuscripts, and on the conjectures that first appear in them.
* * *
Catullus’ manuscripts are quoted in three ways. TOGRm are quoted by their siglum alone. The other surviving manuscripts are quoted according to their place in the List of manuscripts, followed by the year or period in which they were copied. Thus MS. 78 a. 1423 refers to Paris, Bibliothèque nationale de France, Parisinus lat. 7989, to which Catullus was added in 1423. The list is based on that of D.F.S. Thomson (1997: 72-92), with additions by Kiss (2012). Thomson’s list has been corrected or revised in the light of recent research in a number of places. Finally, on a few occasions manuscripts are described in ways that will be self-explanatory
Books, articles, and other works listed in the Bibliography are quoted in the apparatus by the name of the author, the year of the work, and page number(s), in the format Froehlich 1849 232-3 and Herrmann 1957a 672, where the latter work should be distinguished from Herrmann 1957. In the case of editions, commentaries, and other easily searchable works a page number is normally omitted: thus one may find D.F.S. Thomson 1997 or Muretus 1554 in contextu. Wherever the name of the author of a conjecture is followed by a year, and no indications are made to the contrary, this indicates that the editor has seen the work in question himself.
The apparatus has been built up chronologically. This principle has been extended to the point of adding the year of death of the author of a work that was published posthumously. However, if a work was published several years after it was written, but still during the life of its author, it has not been possible to take into account this delay. (One notable example of the latter phenomenon is Vossius’ commentary of 1684, which had composed some forty years before it was published according to the preface of its printer.)
The following abbreviations are used often:
a. = annum, anni, anno
ca. = circa
cfr. = confer
coni. = coniecit
corr. = correxit
damn. = damnauit
def. = defendit
del. = deleuit
dub. = dubitanter
e.g. = exempli gratia
fort. = fortasse
ibid. = ibidem
i.e. = id est
in comm. = in commentariis
MS. = (codex) manu scriptus
MSS. = (codices) manu scripti
om. = omisit
prob. = probauit
s. = saeculi, saeculo
saec. = saeculi, saeculo
scr. = scripsit
suppl. = suppleuit
ut uid. = ut uidetur
