After two dozen pages of dense discussion, in which the conventional evidence for Virgil's life is subjected to close analysis, an expert recently concluded: ‘It may now be apparent that very little external information indeed may legitimately be used in the understanding of Virgil and his work.’ In fact the reader of that discussion may be forgiven for thinking that the only items which survive detailed scrutiny are the dates of Virgil's birth (15 October 70 bc) and death (21 September 19 bc). Virgil's friend Horace, by contrast, seems to have been so profligate with the personal information inserted into his poetry that we appear scarcely to need external evidence at all. Horace himself mentions the year of his birth three times (Epod. 13.6, Carm. 3.21.1, Epist. 1.20.26–8), and this, together with other passages, presents us with – or allows us to construct – the figure of the poet with which we are all familiar.

Yet this very familiarity is deceptive and elusive. In the first place, all readers have their own ‘Horace’. As L. P. Wilkinson wrote (evidently during his time as a code-breaker at Bletchley Park during the Second World War), ‘We English think of him as one of us; but it appears that no less to the French he is one of them, and even to the Germans one of them.’ Secondly, there is the ever-present problem of the relationship between ‘literature’ and ‘life’.