Jackie Elliott's eagerly awaited monograph, based on her 2005 Columbia doctoral thesis, represents a fundamentally important contribution to scholarship which will — and should — influence any future work on Ennius’ Annales. Chiming in with recent movements in the presentation of authors preserved solely or primarily in citation fragments (notably Tim Cornell et al. (eds), The Fragments of the Roman Historians (2013)), E. makes a detailed case for the importance of taking into account the ways in which the preoccupations of the citing sources, and indeed later editors, can fundamentally shape our view of how the ‘complete’ version of a lost text may have looked.

The monograph falls into five chapters backed up by extensive and meticulous appendices. Ch. 1, ‘Ennius and the Annalistic Tradition at Rome’, interrogates traditional assumptions about the nature of the Annales. Taking inspiration from the observation by Ingo Gildenhard that Ennius may well have written ‘annales’ before the existence of historical ‘annales’ in Rome as we have come to know them, E. starts by unpacking the biases associated with assumptions about the poem's so-called ‘annalistic’ presentation drawn from its title. Looking again at the distribution of the fragments, E. argues that the shape and pace of the poem would have been rather different from the arid year-by-year accounts of consular records with which Ennius’ epic has traditionally been associated. Ennius would have been much more innovative than previously thought in the use of time and divine machinery, which can be shown to play a part in the poem's action beyond the early so-called ‘mythological’ books. Ch. 2, ‘The Vergiliocentric Sources and the Question of the Evidence: Ennius and the Epic Tradition of Greece and Rome’, and ch. 3, ‘The Pre-Vergilian Sources’, each move to look at a different set of quoting sources for the epic, showing how the various preoccupations of these authors might distort our perceptions. In ch. 2, E. argues that in quoting passages primarily for their use of shared Homeric and other language, formulae and imagery, the ‘Vergiliocentrics’ are in danger of making us over-privilege the poem's rôle in the epic tradition, whereas earlier audiences might have found a more generically fluid entity. Although E. is consciously taking a ‘reading’ (79) in the reception history of the poem, the implicit division between ‘epic’ and ‘history’, important to E.’s argument in the following chapters, can suggest a rather one-dimensional view of the post-Ennian epic tradition at Rome. While the immediate citation environment of these sources is narrowly literary or linguistic in focus, ‘epic’ and ‘history’ are not easily separable entities in Virgil (just as they are not in Ennius), nor were they necessarily presented as such in the Servian commentary if taken as a whole. Turning to the pre-Virgilian sources, Cicero, as we learn in ch. 3, reads the Annales from a multitude of complex dialogic perspectives (to which E. does justice) as ‘essentially historiographical’ (195), while Varro exemplifies a grammatical tradition making up 38 per cent of citation sources (144), which had ‘no interest in the text as a work of literature’ (144): primarily interested in language, these sources, though they might make the poem seem more linguistically quirky, can be seen as relatively free from distortions of content.

The final chapters move towards gauging what the poem itself may have looked like. Ch. 4, ‘The Annales as Historiography: Ennius and the Invention of the Roman Past’, attempts to synthesize the Homeric and historiographical aspects of the Annales, providing an important corrective to the traditional division of the poem that either relegates its Homerizing features to the early books, or dismisses them altogether. The closing chapter, ‘Imperium sine fine: the Annales as Universal History’, is more speculative in nature, offering a reading of the poem that attempts to account for its impact on the ‘immediate audience’ (233), ‘the collective Roman psyche’ (233) and ‘literary history’ (245). With its much less sceptical attitude, this chapter gels uneasily with the rest of the book, and its argument for the Annales as universal history seems somewhat strained. As E. points out, though, Ennius is an acute version of the problem of dealing with all ancient evidence: there is perhaps always a risk that we will find what we are looking for (296).

In all this, E.’s thorough discussion of source bias will be a crucial guide to readers of the text. For that reason, it is a pity that this learned discussion ‘has nothing at all to add in terms of textual criticism’ (1), relying instead on Skutsch's (often interventionist) edition, with a glance to the recent revision by Flores et al. (Quinto Ennio. Annali (2000–2009)). Textual criticism, too, is a form of reception — one where individual bias can be just as, if not more, distortive than citation context. In a project that ‘views all our access to the Annales as essentially reception’ (6), it would have been good to see how the micro-level of textual transmission has shaped ‘what we think we know’ (295) about the Annales, too.

Ennius and the Architecture of the Annales was, as E. candidly points out, originally conceived vis-à-vis a thorough engagement — and what might best be characterized as respectful dissatisfaction — with Otto Skutsch's edition. As E. notes, Skutsch's edition, monumental and learned though it is, was, in important ways, ‘born old’ (5). E.’s book has now made clear the need for a new English-language commentary to complement or replace Skutsch in line with evolving attitudes to fragmentary evidence and in light of the evidence she presents here.