This is a book about a book. Its subject, the brainchild of the classical scholar and editor Jodocus Badius Ascensius (Josse Bade of Assche, 1461/2-1535), is an edition of Terence produced in 1493 by the Lyon printer Johannes Trechsel. It was by any measure a remarkable work, the first illustrated edition of a classical Latin text and striking testimony to the rapidly developing skills of the earliest printers. A quarto volume of substantial size (319 folios, i.e. 638 pages), it has at its core a text of Terence in an elegant Roman font surrounded in smaller type by a commentary adapted by Badius from the earlier work of Guido Juvenalis (Guy Jouenneaux, ca. 1450-1507) and illustrated with 161 original woodcuts. Two of these are full page pictures (reproduced here as fig. 1-2), one of Guy (or is it Terence?) working at his desk and the other of a Roman stage complete with aediles, actors and audience in contemporary dress and, in the arches below the stage structure, prostitutes plying their trade. The other woodcuts are half-page representations of Terence’s stage action that are rich in detail, with a more fully realized theatricality than the Carolingian miniatures that may have inspired them. The book itself must have been a significant investment—just setting the text may have taken Trechsel’s workers as many as 160 days—and was evidently a prestige volume rather than a money-maker, destined to be the proud possession of aristocrats and clerics. It must have been expensive: in 1499, a deacon in Reval (modern Tallinn) noted in the flyleaf of his copy that he had paid over a gold florin for it.
So complex an undertaking will naturally have had a correspondingly complex genesis, which the authors take great care to untangle. The results of their investigation are presented in what is itself a handsomely produced book, well laid out, scrupulously annotated, and very effectively illustrated. Its opening chapters place Terence in the early history of printing (1-2), trace the career of Badius, his three editions of Terence, and his adaptation of Guy’s work (3-4), and discuss the woodcuts and their putative relation to both the MSS illustrations and Renaissance conventions of gesture (5). These are followed by chapters on the representation of theatrical conventions in the woodcuts (6) and the legacy of the Lyon Terence in sixteenth-century book production (7). The authors’ close attention to detail combines with their extensive knowledge of both fifteenth-century literary culture and the developing book trade to yield not just appreciation of this extraordinary book but a fuller understanding of the world that produced it. All this is no small achievement, and the growing interest in reception studies makes their investigation especially timely. Classicists, though, are likely to be intrigued as much by the questions it raises as by the answers it provides. Those who study the production of books have an expertise somewhat different from those of us interested primarily in their content, and seeing things as these authors do is refreshing as well as instructive. Admittedly, it can also be unsettling. The richness of detail provided in exploring the world of fifteenth-century books does not extend to the world of Roman-era plays, so that a Latinist may at times feel like an eavesdropper on the conversation of strangers. Here topics as diverse as the settling of German printers in Lyon, the possible dates of Badius’ time in Ferrara and what he may have studied there, and the representation of male and female attire in early book illustration are richly documented from primary sources and accompanied by ample citations of relevant scholarship, but references to Terence have a more secondhand feel, with readers directed largely to handbook entries and survey articles.
Every so often, that can be a problem. The authors have an excellent discussion, for example, of how act divisions found their way into Renaissance editions of Terence (pp. 109-14, though without mention of Plautus), but they are less well informed about what those divisions represent. Badius indicated them with running headers, producing divisions the authors say correspond to those “now used in modern editions” (p. 112), which is presumably what they then mean when they credit him with moving an act break in Hecyra “to its proper position” (p. 126). These divisions are important, they tell us, “for understanding the dramaturgy of Terence’s plays” (p. 223). That is not quite the case, which is why “modern editions” now routinely eliminate the breaks imposed by these early editors. They were responding to occasional traces in the Latin texts of the underlying Greek models, which did employ the five-act structure that inspired Horace’s famous dictum (Ars 189-90), but since Roman plays were written for continuous performance, dividing them into acts creates only a false sense of their dramaturgy.
More often, though, Latinists may simply want to know a little more about something a little different. Because the authors are more focused on what Badius did than on why he did it, discussion of his working method rarely goes into our kind of detail. Yet that method is curious. In matters of text, for example, his usual procedure was to accept the authority of Donatus, whose commentary sometimes records variant readings and expresses textual preferences. That is a reasonable procedure, but Badius apparently made one textual emendation on his own authority. It came at Phormio 1016. Convinced that the sense required objective genitives where the manuscripts all read neglegentia tua neque odio…tuo, he printed tui and noted that no less an authority than Lorenzo Valla had proposed a similar correction for the MSS’s desiderio tuo at Heauton timorumenos 307. Badius was well read, and this could be thought a reasonable correction. An editor explicating Terence systematically ex Terentio could cite as parallels Hecyra 219 audiui cepisse odium tui Philumenam and Hecyra 580 ut caperet odium illam mei. T.’s usage varied, however, and so the correction is unnecessary. Badius himself lets stand without comment sine tuo magno malo at Andria 179 and does not follow Valla at Hau. 307, where he prints desiderio tuo. How, then, was he thinking about the text? Was he simply inconsistent or following some other principle?
The very layout of his book, itself no small technical achievement, raises even more interesting questions: what kind of reading strategy does that layout presuppose, and for what kind of reader? The manuscript tradition, whether presenting the plays in verse, as the ancient Bembinus does, or in prose, as many later manuscripts do, offers a text that invites continuous reading. Many early printed editions do the same. Here, roughly contemporary with the Lyon volume, is the first edition of Terence (and the first classical Latin text) printed in England, which (its Gothic font not withstanding), offers modern readers a familiar experience.
By the 1490s, however, the proliferation of commentaries was encouraging editors to include considerable amounts of paratextual material, which they then squeezed before, after, and around the target text. The Lyon Terence takes this trend to considerable lengths: its pages often present little (sometimes very little) Terence but always a good deal of commentary. Where do readers look when confronted with such a format, and with what effect? And what do the illustrations contribute to their experience? The authors’ figure 5.5 provides a striking example of the result. The page opens the scene in Eunuchus where Parmeno presents Phaedria’s gifts, a slave girl and the (false) eunuch, to Thais.
The elegance of the layout is at once apparent, and so is the challenge it presents to readers coming to the text with modern expectations. We look first at the woodcut, where Parmeno in the foreground directs attention to the kneeling slave girl, conspicuous for her black hands and face. That detail is not explicit in the text. Parmeno there says only ex Aethiopiast usque haec (“This one comes all the way from Aethiopia,” 471), leading Donatus to explain that her exoticism, and thus her value as a gift, derives from the remoteness of her origin (ad 167, 471). Only on the next page does the commentary specify her race (Haec maura est ex aethiopia), in effect glossing not the text but the illustration. Much of that commentary simply expands and recasts Terence’s dialogue, tucking observations and explanations into its extended paraphrases. We might expect that sort of exegesis in a commentary originally designed, as this one apparently was, for student use (pp. 122-3), and Badius made no appreciable change to that notional focus. Thus he warns young readers against the easy-going Micio’s acceptance of his son’s whoring and drinking: propterea, iuvenis lector, non sequaris haec comica dicta (ad Ad. 101-2, quoted p. 132). Yet schoolboys were not the only readers Badius evidently had in mind: his preface calls attention to the woodcuts, which he claims make it possible for even the illiterate to follow the plots of the comedies (p. 1). A target readership, then, of students and functional illiterates? How many of either group in the fifteenth-century had the means or the desire to purchase a book on the scale of the Lyon Terence? That deacon in Reval was no illiterate and surely did not spend his gold florin and three albi for a schoolbook. Was this perhaps a book more to be admired than read, or was it at heart a particularly fine crib, an elegant retelling with the original text included to give the exercise a veneer of legitimacy? One way or another, the apparent disjunctions between content and form, advertisement and reality are very intriguing.
As the authors make clear in their final chapter, the Lyon Terence inspired copycats and imitators well into the next century. If not itself a commercial success, it certainly showed the way for other illustrated editions that were, a prime example being the Italian edition of Lazzaro de’ Soardi that went through four versions between 1499 and 1512 (pp. 196-206). It is easy to understand their appeal. Looking back over the history of scholarship, however, a student of Terence would probably say that the most influential sixteenth-century edition was not derived from Badius or any of his imitators but was the work of Marc-Antoine Muret (1526-1585), first printed by Paolo Manuzio in 1555 and then frequently copied. It did not reproduce the layout or reflect the priorities of the Lyon Terence. For Muret and the many editors who followed his example, Anne Dacier in the seventeenth century and George Colman in the eighteenth prominent among them, commentaries and paratexts were designed to enrich, not dominate the act of reading. The change seems important for the history of reception as well as the history of printing, and now that so many old books are accessible in digital formats, the course of that development might well be worth tracing. Such an inquiry would of course take us well beyond the remit of the present study, but like so many solid and useful works, this one makes us not only grateful for what it teaches but eager to know still more.
 The book is no.75 in the catalogue of H. W. Lawton, Térence en France au XVIe Siècle (Paris 1926), no. 89 in G. Cupaiuolo, Bibliografia terenziana (1470-1983) (Naples 1984). Thirty-eight copies (and the leaf of a thirty-ninth) survive. Of digital copies available, that of the Bayerische Staatsbibliothek bsb11303143 is particularly user-friendly.
 The Roman elements, including the prostitutes, seem to derive from Isid. Orig. 18.42.1-2, though the architecture looks to medieval precedents (pp. 175-7).
 Evidently a high price, though the authors provide no comparanda (p. 20). Neither the original price nor the print run is known. Trechsel made no second printing, but Lawton lists eight derivative reproductions between 1493 and 1501.
 The Blackwell Companion to Terence (2013) is frequently cited, as are the essays in the authors’ previous edited volume, Terence Between Late Antiquity and the Age of Printing: Illustration, Commentary and Performance (Leiden 2015).
 To be fair, since Badius, like most of these early editors, presented the text as continuous prose, such running headers, together with the scene illustrations, are the only effective means of finding one’s way through a play.
 As reported on pp. 117-18, the emendation, along with its rationale, was carried over from Badius’ first, 1491 edition of Terence. The Lyon edition does not highlight textual matters. Donatus, who had no difficulty with the MSS reading, glosses nec neglegentia circa te nec odio, inquit, factum est.
 The six plays were printed separately by R. Pynson between 1495 and 1497 but apparently designed to be bound together (87 Lawton/97 Cupaiuolo). The (imperfect) British Library copy is available at ProQuest: Early English Books Online.
 Eu. 454=506, i.e. III.2 in the traditional reckoning, this page includes only lines 545-9. For the growth of paratexts in early editions of Terence and the market forces that encouraged them, see P. F. Gehl, “Selling Terence in Renaissance Italy,” in C. S. Kraus and C. Stray, eds. Classical Commentaries (Oxford 2016) 254-6.
 Muret’s edition is no. 319 in Lawton (no. 439 in Cupaiuolo), with fifteen reproductions by 1594 and many more well into the seventeenth century. The title page emphasizes Muret’s priorities: “Pub. Terentii Afri Comoediæ sex, ex M. Antonii Mureti exemplari accuratissimè emendatæ. Additis ex P. Bembi vetustissimo codice varijs lectionibus ac breuib. annotationib. partim im margine, partim post scenas singulas adscriptis, quibus loci obscuriores explicantur…” A digital version is available at ProQuest: Early English Books Online.