Exegit monumentum. With the publication of W.’s Annals 4, the coping stone is placed on the Cambridge Classical Commentaries of Annals 1–6, begun in 1972 by F.R.D. Goodyear, who commented on the first two books, and continued in 1996 when W. collaborated with R.H. Martin on the third book. That volume was dedicated to Goodyear, and W. in the prefaces to the later commentaries (Ann. 5–6 in 2017, and now Ann. 4) pays tribute to both Goodyear and Martin; his work has long shown the influence of these scholars, but now in their absence we might say postremo suo tantum ingenio utebatur. This final commentary shows us what is distinctive about W.’s approach to Tacitus, such as we have also encountered in his commentaries on Ann. 5–6 and Agricola.

The most prominent of these is W.’s insistence on marking sound effects in Tacitean prose: amicitia Germanici perniciosa (Ann. 4.18.1) is rendered in such a way as to bring out three kinds of echoing syllable – am icitia Germanici perniciosa. (This feature also draws attention to the scrupulous vigilance of the copy-editor and proofreaders.) W. has made the case energetically for reviewing Tacitus as an author given to wordplay (in Tacitus Reviewed [1998], pp. 218–29), though he did not in that essay fully articulate what difference this should make for our interpretation of specific episodes. The commentary provides an opportunity to develop this interpretative approach further, but W. is inclined simply to mark the elements of wordplay and leave it at that. It would be good to see further comment, in order to give undergraduate students a sense that it is not enough simply to register features and identify them as ‘rhetorical’, that more needs to be said. A related feature of Tacitean style, which W. was more vocal in identifying in his Agricola commentary (2014), is rhythm, where he provided a vigorous counter-narrative to those who would exclude Tacitus entirely from such considerations. It is disappointing that he has only occasionally, commenting on the Annals, continued this practice, which would supplement his stylistic observations. For instance, in his exegesis of the description of Capri (Ann. 4.67), W. is alive to the rhetorical framework of the laus locorum, citing several parallels from Menander. And again he is attentive to wordplay, marking up alliteration and syllabic repetition in caeli temperies hieme mitis obiectu montis, quo saeva ventorum arcentur. Here would be an opportunity to note also the parallel rhythm between mitis obiectu montis and saeva ventorum arcentur, which serves to emphasise the harmony between geographical features of a locus amoenus. W. is of course the trailblazer in arguing trenchantly for the need to take seriously the rhetorical nature of ancient historiography. The commentary seems the ideal place to pursue that broader argument through the detailed engagement with Tacitean style and its consequences for the history he records.

Martin and W.’s 1989 ‘green and yellow’ commentary on Annals 4 has been a consistent favourite for undergraduate Latin courses: the material is attractive, ‘the best Tacitus ever wrote’, Syme asserted, and the interpretative insights offered by the commentators are consistently excellent. The Cambridge ‘orange’ commentaries of course address a different audience, and W. assures us in the preface that ‘[t]he present book should be read on the assumption that its contents are more or less entirely new’ (p. x). Given that I find it difficult to imagine how one would improve upon the 1989 commentary, I am not interested in complaining if the exegetical content of the present publication is, in places, ‘less entirely new’. The identification of the narrative structure, for instance, remains the same, with only a few adjustments to the subtitles: 4.74–5 is now labelled ‘Sejanus and a wedding’ to match the subtitle of 4.1–3, ‘Sejanus and a seduction’ (though I have always thought there was more than one seduction in these chapters). But it would be perverse to make too many changes to an analysis of narrative structure that makes so much sense of the text. Another valuable aspect of the 1989 Annals 4, and of the 1996 commentary on Annals 3, was the occasional dialogue between the commentators, as they canvassed their diverging interpretations. Although this happened only very occasionally, it created a very distinctive voice for the Martin–Woodman works, and it is interesting to see W.’s stance on some of these points in the new commentary. For 4.3.3, neque femina omissa pudicitia alia abnuerit, W. 2018 seems to have come round more to Martin's position that this is a generalisation rather than a particular comment on Livia Drusi: ‘[i]t is very difficult to know which of these two meanings is correct’ (p. 76). Tacitus’ incongruous comment on Tiberian vice early in his life at 4.57.2 (et Rhodi secreto vitare coetus, recondere voluptates insuerat), in 1989 was taken by Martin as explanatory and by W. as concessive (a rather fine distinction in a passage marked by disjunction and parenthesis). W. 2018 produces a synthesis of the two interpretations, which makes a tighter argument from the loosely connected preceding remarks about Tiberius’ vice and physical appearance, and thereby resolves the apparent contradiction this creates with the stages of Tiberius’ life as sketched out in Tacitus’ obituary. The interpretations of the 1989 commentary and W.’s reasons for rejecting his earlier reading are also rehearsed in a footnote. It could be said, then, that the best term for this commentary is not ‘more or less entirely new’ but ‘entirely updated’, as is evident also from the selective use of recent scholarship for specific notes.

‘Thus the passage is to be read as follows.’ W.’s resolutions of the various difficulties around the Rhodes comment is illustrative of his approach as commentator and interpreter of Tacitus, well summed up and interrogated by M. Lavan in his review of the Agricola commentary (Histos 9 [2015], p. xliii): ‘the mode of reading obscures the possibility that ambiguity might be an essential and productive feature of the text. The problem becomes more urgent as one raises one's gaze from particular sentences to the larger questions raised by the work as a whole’. The larger questions of Annals, brought into focus in Book 4 through the dominance of Sejanus, have to do with the decline of senatorial politics and with the diversification and debasement of forms of political influence, which makes it difficult to judge responsibility for that decline. As Tacitus puts it, repente turbare fortuna coepit, saevire ipse aut saevientibus vires praebere. Initium et causa penes Aelium Seianum (4.1.1). W. works to parcel out different stages and roles of agency here: since Tiberius shows no sign of saevitia in this book, aut is read as corrective; since Sejanus is introduced as initium et causa, W. hesitates to include him among the saevientes. This is all entirely acceptable, indeed exemplary for the kind of attentive close reading that represents the best of our discipline. But agency in history and politics is not that tidy, as Tacitus seems to indicate at many points in his narrative (one thinks of the notorious introduction of the maiestas trials at Annals 1.72–3), and this opening to Book 4, if allowed the space for ambiguity, could be seen to pose an open question to the reader.

It might be said that the commentary form encourages an author to answer questions conclusively rather than to leave them hanging – although the opposite could also be claimed. W. therefore may not feel the need to provide a definitive over-arching interpretation of Annals 4, and readers turning to his oddly oblique introduction for such an interpretation will be disappointed. Nevertheless, there is much here to appreciate and reflect upon, as in all W.’s work. Present and future Taciteans will gain immeasurably by traversing the Annals with W. by their side.