S.'s volume seeks to consider ‘the dreams of elegy with special attention to their visual qualities' (p. 3). In her introduction, she draws attention to the conceptualisation of dreams as phenomena that are primarily perceived in visual terms, citing examples from the epigraphic record (where the collocation ex uisu or ex uiso is not infrequently found in dedicatory epigrams referring to dreams) and from literature: she points to the use of participial uisus to refer to figures appearing in dreams (e.g. Hector at Aen. 2.270–1 or Cynthia at Prop. 4.7.3). This leads to a consideration of the dynamics of viewing apparently shared by textual representations of dreams and visual art: ‘Rome's elegiac poets responded to the challenge of conveying the experience of dreaming in a textual narrative’, she argues, ‘by characterizing the dream as an image and emphasizing the visual component of the dream experience’ (p. 16). As with ecphrasis, S. suggests, dream-narration functions as ‘a mode of description and communication’, a narrative ‘tool’ (p. 12), and ‘is capable of revealing as much about the visual experience of the viewer-narrator as it may [about] what was viewed’ (p. 17).

These ideas are expanded in S.’s first chapter, and grounded in a number of Roman authors' descriptions of dreaming as a visual experience, notably Ennius; the lexicon and rhetorical devices used in dream descriptions are explored. Particular emphasis is given to the ambiguities in meaning of uideri, focusing on Lucretius, DRN 4 and Cicero's Academica: the subtleties of the middle and passive forms of this verb are important for much of the discussion of focalisation that follows in the book, but unfortunately the material is not worked as hard as it could have been (for a much fuller treatment, see T. Reinhardt, ‘To See and to Be Seen: on Vision and Perception in Lucretius and Cicero’, in G.D. Williams and K. Volk [edd.], Roman Reflections: Studies in Latin Philosophy [2015], pp. 63–91). S.’s consideration of the use of similar language in ecphrastic descriptions in the Aeneid provides an interesting discussion of Virgil's portrayal of perception and focalisation, but one that is all too cursory. (She treats, for example, Aeneas' catching sight of Dido as he views the murals on the temple of Juno articulately [1.453–97], but omits any mention of the Diana-simile [1.498–502].)

The most striking omission from this introductory section, and from the book as a whole, however, is any mention of Hellenistic treatments of dreams as pertinent background to a discussion of Roman elegiac dreams. Given the emphasis on the visual aspects of dreams and their potential comparison with (ecphrases of) artworks in S.’s study, Medea's dream in the Argonautica (3.616–32) may not seem of much importance, since it is not particularly concerned with sight (but cf. 3.691: λεύσσω ὀνείρατα λυγρά). The description of the dream which opens Moschus' Europa, however, is steeped in the language of the visual and functions as a complement to the ecphrasis of Europa's basket (43–62), which, like the dream, presages the heroine's rape. The basket is described as an object to be seen (θηητόν, μέγα θαῦμα [38]), and the dream is described as something that she saw (τὸ γὰρ ὡς ὕπαρ εἶδεν ὄνειρον [17]), an image that she held before her eyes when she had awoken (ἀμϕοτέρας δέ | εἰσέτι πεπταμένοισιν ἐν ὄμμασιν εἶχε γυναῖκας [18–19]). The juxtaposition of dream-narration and ecphrasis in this poem, of which the elegists were aware (cf. Ov. Met. 2.836–3.2), provides a pertinent exploration by a Hellenistic predecessor of many of the themes treated in S.’s book, and its absence from her discussion is unfortunate.

The four substantive chapters of the book offer close readings of elegiac dream-narrations, which are followed by a consideration of comparable artworks and a discussion of how the authors may have ‘exploited techniques and … dynamics [of viewing] familiar from artistic representations' of similar scenes to those described in the texts (p. 200) in order to ‘engage the reader of the dream description as though he … were a viewer of the dream itself’ (p. 219). This is perhaps most effectively achieved in S.’s exploration of Rhea Silvia's dream in the proem of Fasti 3, alongside a number of artistic representations of her concomitant rape. There is a considerable play on focalisation and perception in 3.1–40, which encourages the reader to consider their own position in relation to what is being placed before their mind's eye. The passage is steeped in the language of the visual, and we are offered a number of perspectives on the same event: Ovid's narration of Mars' rape of Rhea Silvia (21–2), and her own subjective recounting of a dream about the meaning of which she remains perplexed (27–38). Our focalisation of her dream is also influenced by her own uncertainty about its status – an somno clarius illud erat? (28) – expressed, we might note, as a question reminiscent of the ambivalence felt by Europa in Moschus' poem (see line 17 quoted above). This refiguring of focalisation in Ovid's poem is usefully explored alongside plastic representations of the same scene in several media and allows S. persuasively to demonstrate how Ovid ‘acts like a visual artist creating a work to engage potential viewers' (p. 215).

S.'s discussion of Propertius 2.26a and Propertius 3.3 are also convincing. Her comparison of Propertius' varied mythological imagery in 2.26a to a series of pendant images contributes to a cohesively vibrant reading of the poem which opens up hermeneutic possibilities rather than delimiting them. The mythological comparanda are not necessarily linked by any sort of narrative, but, when juxtaposed, encourage the consideration of ‘thematic differences … and formal similarities' (p. 130), and thus challenge and reshape one's interpretation of any and all of the scenes depicted or described.

The treatment of focalisation in the dream-narration of Propertius 3.3 is of particular interest, as the epiphanic formula uisus erat is transformed into the first person, so that the elegist seems to dream himself (Visus eram [1]); this is later challenged as it becomes clear that he is also seen by Apollo (cum me Castalia speculans ex arbore Phoebus [13]): Propertius comes to seem to himself to be seen. The appearance of Apollo is felicitously compared to the ecphrastic statuary of 2.31 (Propertius' description of Palatine Apollo's temple), though it is a shame that the god's appearance in 4.6 is not brought into dialogue with these passages, as the elegist's response to Virgil's ecphrastic description of Apollo on Aeneas' shield (Aen. 8.704) would make an interesting complement to S.’s discussion. The juxtaposition of nature and artifice in the Muses' cave is usefully compared not only to contemporary nymphaea and grottoes, but also to the cave of Silenus in Eclogue 6, providing a salutary reminder of the importance of the Eclogues to Propertius' self-fashioning as a poet. It is a shame that the ecphrases of caves in the first book of Longus' Daphnis and Chloe are not adduced here, concerned as they are with questions of ars and natura, perception and sight: θέαμα εἶδον, claims the author in his preface, κάλλιστον ὧν εἶδον, εἰκόνος γραϕήν.

S.'s exploration of Tibullus 1.5 focuses on the ϕαντασία of the rural life that he desires with Delia (21–34). This is germane to the subject of her book, S. claims, as descriptions of dreams, like those of ϕαντασίαι, ‘deliberately [conjure] in the mind images of things not present to the eye’ (p. 57). Whilst this may be so, her assertion that the passage in question has ‘a visual focus' (p. 59) seems not to be founded on much; Tibullus is not cast as an ‘observer’ of Delia's imagined activities at any point, as S. would have it (p. 63).

This volume engages our readings of Roman elegy with a diverse range of artistic media. It has some useful observations to make on the texts themselves and on the construction of readers as viewers and vice versa, but disappoints in its unwillingness to engage with important points of contact with Greek literature. A good number of images furnish its pages, but their production in monochrome is unhelpful.