Ammianus Marcellinus and the Representation of Historical Reality by Timothy D. Barnes (review)

HUMANITIES 173 Philo and Antiochus - the former, down to the time of his Roman Books, a sceptic, the latter tending towards dogmatism. T. Dorandi, R.J. Hankinson, and G. Striker explore this from different angles, with Striker, for instance, seeking to elicit from the Antiochean arguments of Lucullus the reasoning behind Philo's altered stance. Academici libri comes out ofa tradition ofphilosophical arguments about knowledge that were already well rehearsed by the time of its composition. For this reason and because of the loss of so much of the text many of the arguments are elusive. M. Bumyeat however, fleshes them out with admirable patience and skill. There are also worthwhile papers by K.A. Algra on Cicero's use of the ethicaldivisiones in Academicilibri,J. Barnes on logical problems in the work, and J. Allen on the representation of Carneadean argu,nent here. I am not sure, however, that J. Glucker's paper 'Socrates in the Academic Books and Other Ciceronian Works' really belongs in this volume: it takes up a problem for which Academici libri provides some material but not one that is important to understanding that work and its contribution in the same sense as the others. (ANDREW R. DYCK) Timothy D. Barnes. Ammianus Marcellinus and the Representation ofHistorical Reality Cornell University Press. xiv, 290. us $45.00 The sUIvlvmg portion of Ammianus's narrative provides our most complete account of the political and military history of the later Roman Empire from 353 to 378. While successive scholars from Edward Gibbon on have based their understanding of this period on the Res Gestae, Barnes argues that Ammianus 'failed in his obligation as a historian to strive to transcend personal bias.' I would like tQ touch on three of the book's especially striking arguments in this brief review. The first deals with the structure of the Res Gestae. Only books 14 through 31 (treating the 25 years 353 to 378) are extant. Because Ammianus . tells the reader that he started his narrative with the reign of Nerva (96-98), the lost first thirteen books of the Res Gestae would have dealt with around 257 years, demonstrating an uneven scale of narrative. Barnes points out that thirty-one would have been an anomalous total number of books by the traditions of ancient composition. Arguing that Ammianus composed his work in hexads (groups of six books), Barnes extrapolates that Ammianus really wrote his work in thirty-six books and that what we have surviving are actually books 19-36 (which were misnumbered in the transmission of the text to 14-31). Barnes realizes that no manuscript preserves anything but the traditional 14-31 numeration scheme, and suggests that there may have been a scribal correction from 'xviiii' to 'xiiii' to jibe with a citation by Priscian. Following this argument, the lost books would have 174 LETTERS IN CANADA 1998 been longer, and thus the scale of narrative would have not been so different from the surviving eighteen books. Possible, and hard to prove or disprove, this argument is the most radical attempt to deal with the lost books since the 18805. Another important part of Barnes's book treats the origin ofAmmianus. Until about ten years ago, scholars believed that Ammianus hailed from Antioch because of his praise of that city and also because a contemporary letter from Libanius to a certain 'Marceilinus' has' survived praising his fellow Antiochene for his literary success in Rome. However, Barnes (following work by G.W. Bowersock and Charles Fornara) points out that forms of sungraphos used by Libanius to describe the work of 'Marcellinus' could signify forms of literary composition aside from history. Barnes questions whether Libanius would have been as demeaning to Ammianus and his work as the letter appears to be towards 'Marce11inus.J As opposed to some recent theories, Barnes favours an origin for Ammianus somewhere in either Syria or Phoenicia because of the historian's knowledge of some Syriac. The region is doubtless correct but it does not rule out Antioch. Ammianus does praise that city more than others. If Antioch is a possible home town for Ammianus, could not the letter from Libanius still be to the...