In the unfinished third book of Claudian's De raptu Prosperinae, a bereaved Ceres discovers a spider labouring to fill ‘the space that remained’ in her daughter's unfinished tapestry (Rapt. 3.154–8). In this slender, ambitious book Aaron Pelttari argues that Latin poets of the ‘long fourth century’ left space within their texts for newly empowered readers, who were invited to ‘make sense out of the fragments of his text’ through active and iterative interpretation (8). P.'s project is thought-provoking, his reading of late antique poetry usually perspicacious and not rarely invigorating. The Space that Remains will be fundamental to future discussions of Latin textuality, compositional practices and the horizons of readers’ expectations in Late Antiquity.

P.'s libellus contains a concise introduction, which includes a useful précis of recent scholarly approaches to late antique poetry, four chapters and a brief conclusion. In the first chapter, ‘Text, Interpretation and Authority’, P. sketches the interpretative practices of Jerome, Augustine and Macrobius (among others) to glean a bird's eye view of how (some) influential late antique readers forged meaning in Christian scripture and Vergil. P. views Jerome as ‘the ideal author-as-reader’ (17), who founded his own authority on interpreting texts. The interpretation of Vergil's poems — endowed with allegorical significance by Macrobius, Servius and Donatus — demonstrates how ‘the creativity of late antiquity is concealed under the guise of simple exposition’ (33). P. is most persuasive on Macrobius and plants an important seed: ‘rather than as an agon between opponents, literary history is imagined as the re-presentation of earlier auctores’ (30).

Prefaces were one device through which late antique authors could ‘foster the active involvement in the text’ (53). Building on Gennette's theory of the paratext as a textual space shared by author and reader, P. represents this prefatory impulse as a distinctly ‘post-classical’ move that acknowledges and encourages the centrality of the reader (49). Classical authors may pen overtures or overtly metapoetic reflections within the same literary frame as their other works; but post-classical authors telegraph their secondariness and so ‘use prefaces to situate their works within a particular, ephemeral context’ (49). P.'s work here complements L. Jansen's The Roman Paratext (which unfortunately appeared too late for P. to engage directly). Claudian and Prudentius use their allegorical prefaces to declare themselves as authors and ‘provide coherent readings of poems’ (61); Ausonius ‘draw[s] attention to the distance between himself and the text’ (64), creating an opening for the reader to exercise interpretative control over the text. (Here we see the connection with ch. 1.) But by offering a first reading, authors not only fix the text ‘in its present situation’ (72); they reduce the gap between canonical and new texts, which have been injected into the literary tradition by the author's own act of interpretation. Exploring this dynamic would have provided a valuable connection to the interpretive theories in ch. 1.

Umberto Eco's Opera aperta (1961) lights the way for ch. 3, ‘Open Texts and Layers of Meaning’. P. deftly juxtaposes disparate genres to squeeze new insights from Eco's familiar theory. P. styles Optatian's dizzyingly intricate figural poetry, Prudentius’ allegorical Psychomachia and patchwork Vergilian centos as related examples of how late antique poets consistently destabilized the ‘univocal text’ (73) to allow their strong readers to identify the text's semantic layers and to share in the work of composition. Here P. gives an important twist on Michael Robert's articulation of the late antique interest in fragmented narratives and tessellated descriptions. Repetitive ornamentalization of late antique texts is not mere variatio for aesthetic pleasure, but a powerful marker of the openness of the whole text (113–14).

In ch. 4, ‘The Presence of the Reader’, P. unveils his theory of post-classical modes of allusion. While not merely ornamental, these imitations nevertheless ‘resist the impulse to integrate old and new’ (160). P. challenges ‘received notions of the dynamics of appropriation’ that are founded on rivalrous aemulatio (160; note the conscious echo of S. Hinds (1998)), offering a tripartite taxonomy for textual reuse: (a) ‘nonreferential allusions’, which allude without expecting the reader necessarily to read with or against the hypotext (although these, to be sure, were also deployed by earlier poets); (b) ‘juxtaposed allusions’, a quasi-centonic mode that lays bare a text's composite nature by collocating disjointed fragments of a classical text; and (c) ‘apposed allusion’, in which (near) verbatim quotations exist in tension with their new, ventriloquial context. These late antique poets then routinely invert Russell's principles of classical imitation: an allusion's context is not always worth imitating; the letter of the allusion often takes precedence over its spirit; borrowings are often explicitly acknowledged, resist full integration and engage in a non-competitive relationship with their source. Thus P. stakes a claim that the features of late antique poetry oft dismissed as frigid formalities are important ‘marker[s] of the text's ability to communicate a sense of presence to its contemporary audience’ (162). How, one wonders, do these modes operate alongside referential allusions, which P. recognizes are also present in these texts? Or in biblical epic? Or classicizing poems of later centuries?

It is rare to finish an academic work and wish for more. But P.'s inaugural foray has left many matters unplumbed. It is testament to P.'s lucid style, verve and insight that this book makes his readers better able to undertake the difficult work of explicating late antique poetry and this reader hopeful that more is to come from P.