This collection assembles seventeen articles, six new, with most of the remainder having appeared in diverse collected volumes over the last thirty-six years. Matthews decided not to reprint journal articles which can now be easily accessed through JSTOR, thereby excluding some important items — though, as is true for most established scholars, the increasing demands of friends, editors and conferences leave little time for journal-writing. The one exception (17) is the essay on ‘Macsen, Maximus, Constantine’ from The Welsh History Review (1983) which is yet to achieve electronic dissemination. Much of the collection clusters around the central themes of M.'s scholarly activity: the world of Symmachus (10–11), Ammianus (9), the Theodosian Code and its post-Roman off-shoots (13–16, a particularly useful assemblage), travel (7), and Gibbon (1). Interest in historiography directs attention to Ronald Syme (3), and then Syme, in combination with Gibbon's belated recognition that his story of the decline of Rome might have started with the first imperial transition in a.d. 14, naturally leads to Tacitus. As a result, four of the six new items involve Tacitus or the world of the first century a.d. (2, 4–6), the others being on the Bordeaux Itinerary (8) and Augustine's Confessions (12).
Two chapters demonstrate M.'s ability to contribute new ideas to the interpretation of familiar texts. Augustine's Confessions is much-discussed, no passage more so than that in which Augustine described the psychological experience in the garden of his Milan house which underpinned his conversion. The description combines realistic detail and symbolic references, though where divisions are to be drawn between the elements is disputed. The child in the next-door garden whose chant of tolle lege Augustine interpreted as an instruction to pick up his book and read, leading him to open the text of Paul's Epistles and read the injunction of Romans 13.13–14.1, has often been interpreted as part of the symbolic structure: an angelic voice whose words have no obvious connection with children's games. However, by suggesting that it is not possible to know the actual word order of the sing-song chant and, hence, that to recover the child's perspective the words might be inverted to lege tolle, M. plausibly posits a game in which the participants are exhorted to select an item, perhaps a knucklebone, which is to be picked up or removed. The chant may well be part of the real context, whereas Augustine's fevered hearing transformed it into a symbolic injunction (ch. 12). Even more discussed is Tacitus' account of the Senate debate in a.d. 14 about the accession of Tiberius. Through close attention to how the tense exchanges might have been recorded in the senatorial acta, a plausible source for Tacitus, M. is able to separate out what material originated in the acta and what was contributed by Tacitus' ironical analysis of senatorial self-abasement; in turn this allows him, through intelligent attention to the wording of Velleius Paterculus, to identify the basis for the main discussion not as an offer to Tiberius of any particular powers, which would have been superfluous, but as a consular request that Tiberius should assume his adopted father's statio. Having clarified the basis of the debate, M. then engages with the chronological difficulty posed by the apparent presence of Drusus in the Senate on 17 September before his departure to deal with mutinies in Pannonia where his arrival can be dated by a lunar eclipse on 27 September. Again, M.'s careful consideration of how Tacitus' sources might have presented information, combined with accurate knowledge of official procedures, permits him to identify that the wording of Tacitus' account actually indicates that Drusus was no longer in Rome on the 17th, presumably because he had already been dispatched to Pannonia.
Of the other new items, ch. 5 surveys six first- and early second-century equestrians, with a rare correction of Syme regarding Cornelius Fuscus (102). Individually one might see little gain from such minutiae, but collectively they help to create the intimate understanding of social contexts and relationships which underpins M.'s mastery of the fourth century. With regard to the first century this sensitivity to social systems permits M. to advance an elegant solution (ch. 6) to the conundrum of the unique epigraphic example of a substantially complete Roman will, often known as the ‘Testament of Dasumius’ (CIL 6.10229). By attending to the diverse links revealed in the will and the language used to refer to different individuals, M. is able to propose that the testator, far from being a man of consular status, was probably a freedman, still someone of great wealth and impressive contacts but also a person who felt the need to recognize certain benefactors and superiors in his will as well as to perpetuate mundane hostilities. This also helps to account for this epigraphic rarity: testators of safer status had no need to advertise their contacts in this way, whereas this man was closer to the anxieties of Trimalchio. Ch. 8 on the Bordeaux pilgrim demonstrates the importance both of understanding the practicalities of the Roman Empire and of attending to what a text does not say, if wider inferences are to be drawn from a particular work. Under M.'s scrutiny the Itinerary emerges not as a reflection of a radically new and integrated Christian empire but as testimony to a lack of engagement with wider issues in comparison to personal interests. The last of the new chapters (ch. 2) is a philosophical-historiographical discourse on Thucydides, Hobbes and Tacitus. This is connected to other chapters in terms of topic through Hobbes' commentary on Tacitus' account of Tiberius' accession, and of approach through its close attention to the issues of language and the operation of power.
M. cites Gibbon's observation in the Advertisement to the Notes in the first volume of Decline and Fall that ‘Diligence and accuracy are the only merits which any historical writer may ascribe to himself’ (7). M. himself displays these in abundance, and much more besides: a sensitive feel for the operation of the ancient world; an alertness to the cultural perspectives of diverse texts; a clarity of exposition from the earliest to the latest papers — not for M. the progressive opacity of Syme! This collection is significantly greater than the sum of its parts, with the importance of textual context providing a unifying theme. It is testimony to the consistent excellence of M.'s scholarship over the past generation, and a reminder of the wide-ranging and humane inspiration with which for two decades M. made Oxford an inspiring as well as supportive place in which to pursue late Roman studies.
