2. Latin from A to P: The TLL in the 20th Century

After more than a century, the basic look of the TLL, with its two eighty-four line columns per page, remains reassuringly familiar. Yet this superficial continuity masks real changes, not least in the basic corpus of material. Obviously the Thesaurus reflects the discovery of new texts. Some of these, like the new Gallus, are familiar to almost any classicist. Others may be more familiar to specialists, especially students of late antiquity: new letters and sermons of Augustine, new passages of Rutilius Namatianus's De reditu suo and Tiberius Claudius Donatus's commentary on the Aeneid, portions of a gospel epic by one Severus, the "Appendix Maximiana."¹ And to these literary texts must be added the constant stream of newly published inscriptions, bringing with them new senses, new constructions, new spellings.

On the flip side, the TLL's corpus has also grown smaller. Users of the 1990 Index are familiar with the annotations non iam affertur and its even more disapproving cousin, nunc spernimus. Most often the work has been dropped because a reattribution or redating pushes it beyond the TLL's terminus of 600 c.e. But other cases are more curious (for additional instances see Krömer 2003). The 1990 Index still includes a text it refers to as "Epist. Vinisii." This is a brief text published by Nicholson in 1904, a letter from a fourth-century bishop in Roman Britain warning a parishioner against an Arian rival, a certain Biliconus. Or so it was believed. A little over ten years ago, Roger Tomlin showed, in perhaps the wittiest article ever published in ZPE, that this text is in fact an ordinary curse tablet, one of scores or hundreds that survive from the sacred spring in Bath. The saintly "Bishop Vinisius," his correspondent "Nigra," and the dastardly "Biliconus" existed only in the first editor's fantasy; it did not help that he was trying to read the tablet upside down.

Less obtrusive—but far more important in the aggregate—are incremental improvements in familiar texts. New commentaries and other works have improved the handling of specific passages and terms. New editions, built on a sounder manuscript basis, often clear up difficulties or eliminate bogus readings. Yet there is still much work to be done here, especially among later and vulgar texts. To cite one example, TLL VI.1.1172.59 includes a brief article for fortiusculus, a word attested only in Fulgentius the Mythographer. The Thesaurus slips are based on the only critical edition of Fulgentius, the 1898 Teubner edition by Rudolf Helm, who printed fortiuscula on the strength of two ninth-century manuscripts, R and H (only one of which he had himself collated). In the course of preparing a new edition of Fulgentius, I have recollated Helm's manuscripts as well as a number of others, and I can report that R and H, like all the others, have fortiuncula. Fortiusculus is a ghost-word, which should be removed both from the text of Fulgentius and from the TLL. The Thesaurus, like the rest of us, is at the mercy of the available editions, and it is chastening to realize that a Thesaurist today, armed with the latest in digital technology, would still make exactly the same error.

If the material has undergone changes, so too (and to a far greater degree) has the Thesaurus's handling of it. Many characteristic TLL features took time to reach their current form: the introductory "Kopf," with data on etymology, orthography, etc.; the hierarchical subdivisions; the complex systems of parentheses and brackets; the laconic break-offs al. and et saepe. Even the twin organizing principles of chronological development and "exclusive opposition" were not present from the outset. Here the differences between the opening volumes (A–B) and those that follow are perhaps the most marked. At first glance, the earliest articles are little more than vessels into which citations have been poured; finer distinctions are sometimes made, but rarely signaled explicitly. In the volumes for C and D one can observe a development toward the precision and articulation of classic articles...