Scholia Reviews ns 11 (2002) 23.

Paul Tombeur (ed.), Bibliotheca Teubneriana Latina 1. Munich: K. G. Saur and Brussels, Brepols Publishers, 1999. CD ROM and User's Guide in English, German, French and Italian. Pp. 208. ISBN 2-503- 50921-5 (Brepols); 3-519-01999-X (Teubner). EURO 820.00.[[1]]

John Hilton
University of Natal, Durban

Is it true to say that in the last forty years computer technology has transformed the way Classicists carry out their research and teaching? The road has certainly not always been an easy one; often it has been a disheartening case of two steps forward, one back, as market forces have disrupted previously comfortable working patterns. Some of the more grandiose claims made for technology in the context of the humanities have failed to win acceptance and what progress has been made has been uneven -- the internet and communications have scored signal successes while statistical analyses of discourse and narrative have fallen away. Despite the investment of huge amounts of funding and time, areas of research that statistical techniques promised to solve finally, such as authorship studies, have proved more problematic than first thought.[[2]] Looked at from another perspective, however -- that of access to information and the construction of knowledge -- advances have been so great that it is hard to imagine working without the tools that help to filter the data and shape the production of research and teaching materials today; concordances and search engines have proliferated and the accuracy of the instruments available for measuring linguistic, stylistic, metrical, and textual problems has improved greatly. The appearance of the present textbase and its associated software makes an important addition to the resources available to scholars of the ancient languages and will be welcomed above all by the new generation of classical philologists who have grown accustomed to working in a digital world.

In 1999 the Bibliotheca Teubneriana marked its 150th anniversary by issuing this CD ROM of Latin texts using software developed for the Cetedoc Library of Christian Latin Texts under the leadership of Paul Tombeur from the Catholic University of Louvain-la-neuve. A second edition (BTL2) has since appeared. The project therefore builds on a relatively long-established base.[[3]] The aim of the work is 'to publish a database comprising all the Latin texts which have appeared in the Bibliotheca Teubneriana since its foundation' (User's Guide p. 50). The earlier version of the project contains the Classical Latin authors; the newer release, which I have not seen, was to add 'fragments, other Late Antique authors and those patristic writers published by Teubner, as well as some medieval and neo-Latin works' (ibid.).

This review will focus on the two main aspects of this product: the data and the software.

As far as former is concerned, comparison with the Latin texts on the Packard Humanities Institute CD ROM is inevitable but problematic. BTL1 certainly contains texts not available on the PHI/CCAT version: Macrobius, for instance, and Ammianus Marcellinus. At the same time, I miss the admittedly quirky inclusion of the Vulgate Latin Bible and the King James and Revised Standard Version on the earlier CD. However, the difference between the two collections can best be seen in the case of a particular author. The PHI/CCAT disk takes the text of Helm (1959) for Apuleius' Apologia and Florida, Robertson and Valette (1940-1946) for the Metamorphoses, Beaujeu (1973) for the De Deo Socratis, De Platone, De Mundo and the philosophical fragments, and Morel (1927) for the poetic fragments. BTL1 on the other hand has Helm for the Metamorphoses (1955), Florida (1959), and Apologia (1963), Moreschini (1991) for the De Deo Socratis, De Platone, De Mundo, Peri Hermeneias, and Asclepius, and Riese (1906) for the Anechomenos. In this instance, the two disks complement each other reasonably well. However, since neither compilation includes a textual apparatus, which the scrupulous scholar will want to consult, the choice of texts is important but not crucial. BTL1 also offers notes on the texts used. In the case of Apuleius, these cover questions such as the authenticity of the works attributed to this author and the date at which they appeared, but these complex issues naturally require much fuller discussion in the specialist literature.[[4]] Thus the note on the Florida reads rather confusingly: 'In addition to the 23 extracts that make up the traditional corpus of the 'Florida', the five fragments gathered under the title 'De deo Socratis. Prologus' were probably parts of the same work. So these five fragments have been treated as separate.' In fact, these fragments are included in the text of the De deo Socratis. Nevertheless, judged purely on the texts incorporated in this CD, the present product indubitably provides scholars of Latin with an important resource that extends the range of texts previously available electronically.

The software provided with this CD shows a very high degree of elaboration. Perhaps inevitably the complex interface of the BTL takes time to get used to. The search window uses Latin terms while file and editing operations are accessed through one of the modern European languages. Searches are conducted by a process of selection of texts (using the inquisitio tab) by means of a number of fields all of which may be linked using the Boolean operators (although the user has to select each field separately as the titles are not filtered automatically by previous selections). The fields are: auctor, titulus, clavis, aetas and formae. The first two fields are the most useful and self-explanatory. The last field is least useful, since it distinguishes only two eras -- classical and patristic. The clavis field refers to the classification number used by the Handbuch der lateinischen Literatur der Antike (LLA).[[5]] One wonders how widely available this manual is outside of Europe and how useful its codification of Latin texts will prove to be. Nevertheless the codes of the LLA do enable the user to search the data chronologically or by genre and may be further refined using generic markers such as PH for philosophical works, and RH for rhetorical works. The descriptors of the TLL, such as 'CIC. OFF. - PROSA' for the prose work of Cicero, the De Officiis, may also be used. The full clavis for this last work is "268.PH - DE OFFICIIS - CICERO (MARCUS TULLIUS CICERO)- TLL CIC. OFF. - PROSA". In addition to this, each of the fields may be searched by the entire entry (the full name of the author, title, code, or era) or by any form (word, code, or descriptor) of which the field is composed. The software therefore makes a great deal of power available to the user; for example, one may search all prose, verse or mixed prose and verse compositions. It is also possible (more or less) to search texts grouped according to seven chronological periods (early works to Cicero, Cicero to Augustus, Tiberius till 117, 117-284, 284-374, 374-430, 430- 568), though some extraneous works appear when wildcards such as "2??" are used for this purpose. The usefulness of these rather schematic categories is open to question, however, and the degree to which generic searches are possible is often more apparent than real. There seems to be no way of conducting a search on all philosophical works, for example, other than by constructing a list. In any case, I suspect that users will want to specify works to be searched themselves explicitly rather than depending on preconstructed groupings.

Once author, title, clavis or era have been selected, the work or works may be examined on a sentence-by-sentence basis (sententiae) or as a whole textus. Notes on the text may be viewed using the memento tab. The actual forms to be searched for are specified in the formae field and may include wildcards and Boolean operators (AND, OR, NOT). The search process would have been clearer if the formae field occurred first, since the other fields (author, title, LLA code, era) act as filters delimiting the scope of the search.

Output from a search may be saved as a text file. However, the maximum number of consecutive sententiae that may be copied to a file is limited to thirty.

Once users have become accustomed to using the complex search engine of the BTL, they will undoubtedly be impressed by its speed and power. The availability of the full range of Teubner texts on one CD is indisputably an enormous advantage. For scholars working in Europe or the United States the price is excellent value, despite the absence of a full apparatus criticus, especially for Classics programmes that have not managed to acquire the entire collection before. For those working in developing countries the cost of keeping up with worthwhile initiatives like the BTL is becoming prohibitive. Nevertheless, the BTL CD will provide any Classics programme with a valuable technical resource for research and teaching.

NOTES

[[1]] BTL1 is no longer being marketed. BTL2 is available for EURO 860.00, US$834.00. A network price is also available: see further http://www.maierphil.de/ClasPhil/TEUBPRET.HTM.

[[2]] I. Marriott, 'The Authorship of the Historia Augusta: Two Computer Studies', JRS 69 (1979) 65-77, whose methodology was subsequently questioned by David Sansone, 'The Computer and the Historia Augusta: A Note on Marriott', JRS 80 (1990) 174-77. For a review of the ground covered in the last forty years, see the bibliographical surveys in Stephen Waite's review, Calculi and its successor, Joseph Tebben's Computing and the Classics.

[[3]] See P. Tombeur, Base de données pour la tradition occidentale latine, guide de l'utilisateur (Database for the Western Latin Tradition, User's Guide), incl. Cetedoc Library of Christian Latin Texts CD (Turnhout 1991). CETEDOC is an acronym for the Centre de Traitement Electronique des Documents at the Catholic University at Louvain-la- neuve. The database is discussed by P. Tombeur, 'Base de données pour l'ensemble des textes patristiques et médiévaux latins: la CETEDOC Library of Christian Latin Texts', Revue informatique et statistique dans les sciences humaines (previously known as Revue: International Organization for Ancient Languages Analysis by Computer) 27 (1991) 199-211. Tombeur's involvement in the CETEDOC project dates back to 1977: cf. P. Tombeur, 'Vox Latina: Belgian Initiatives in Data Processing the Intellectual Language of Europe A.D. 197-1965', Computers and the Humanities 12.1-2 (1978) 13-18.

[[4]] See, for example, S. J. Harrison, Apuleius: A Latin Sophist (Oxford 2000).

[[5]] R. Herzog and P. L. Schmidt (edd.), Handbuch der lateinischen Literatur der Antike (Munich 1989-).