Skip to main content

For lovers of classical Latin texts it might be said, after Dickens, that it is the best of times, it is the worst of times. Thanks to the wonders of the internet, our texts have never been so readily accessible: collections like, Perseus (, and the Packard Humanities Institute database (—to name only the most widely used—make available multiple versions of the most commonly read authors and texts—along with many not so commonly read—with a few clicks of a mouse or strokes on a keyboard. These resources are almost universally regarded as a boon, and I certainly so regard them.

And yet it is also probably true that at no time since the fifteenth century have so many people read so many classical Latin texts with so little idea of where the text they happen to be reading came from or what it actually represents. Often the texts are posted without indication of provenance or only the haziest sort of attribution (“posted by so-and-so from the University of Such-and-Such”). Often they are derived from out-of-date editions that are used simply because they are out of copyright. And even in the best case—that of the Packard Humanities Institute database, which comprises texts from the OCT or Budé series or the Bibliotheca Teubneriana that were, when the database was assembled, among the best available—the texts are presented without a critical apparatus. As a result, readers can know why the text they are reading is the text they are reading—what its documentary basis is, what choices its editor made, and why—only by referring to the printed edition, which rather defeats the purpose of the electronic version.

The Library of Digital Latin Texts (LDLT) will be the first such collection to include a critical apparatus with every edition it publishes. Editors who contribute to the series will be fully present to their readers: they will have the opportunity to lay their cards on the table and say, in effect, “Here is my evidence: don’t you agree with my understanding of it?”—and readers will have the opportunity to say, “As a matter of fact, no.” Of course that relation between editors and readers is implied also in print editions that are furnished with an apparatus. But editors who contribute to the LDLT will also be able to include, if they wish, an amount of information that exceeds the capacities of even the most generous apparatus in a printed edition. More important, editors will be able to include, if they wish, kinds of information that the apparatus of a printed edition is fundamentally incapable of displaying. In the second half of my talk I describe one form of information that meets this description and that will enable editors to communicate their understanding of their texts’ documentary basis with unprecedented clarity: a typology of variants developed by Sam Huskey and myself that will allow readings to be tagged in such a way as to reveal not just what readings the various sources for the text offer but also how the editor believes they came to be and what they signify, for the history of the text’s transmission and for its constitution. After explaining what this typology is, I will show how it will be applied, using a passage from Servius’ commentary on Aeneid 9-12 that will be among the first texts that the LDLT publishes.