The Theophilus legend in its multiple text versions and pictorial adaptations was immensely popular in the Middle Ages, primarily in the thirteenth and fourteenth centuries. Root sees interest in the tale and its vast corpus of illustrations emerging from twelfth-century scholasticism and presents a close reading of text and picture within the framework of Hugh of St Victor's doctrine of exemplarism, drawing in particular on the work of Boyd Taylor Coolman and on the many studies of seeing and reading, visions and visionary, likeness and presence, and the creation of man in the image of God and man's active imitation of Christ with the Virgin Mary as both mediatrix and speculatrix. Theophilus moves from dissemblance exemplified by his contract with the devil to resemblance through the salvific power of the Virign Mary but he must himself exemplify the process of manducatio (leading upwards, cf. Abbot Suger of Saint-Denis, writing shortly before the mid-twelfth century) and manifest his resemblance through public display and communion, through which God's light illuminates his face. All this corresponds, Root suggests, to anxiety of identity in thirteenth-century culture. More relevant in my view, and not discussed by Root, is the affective impact of the Fourth Lateran Council's requirement of annual communion preceded by confession, which resulted in an explosion in the production of devotional books – and their illustration – for the lay public from 1215 onwards. The Theophilus legend exemplifies the effectiveness of prayer and culminates in communion and in the transmission to others of the lived experience. Perhaps more could have been made of the textual differences between the text versions of Paul the Deacon, writing in Latin, William Adgar in Anglo-Norman and Gautier de Coinci in French.

Leaving aside depictions of Theophilus in monumental sculpture, Root bases his interpretation on fifty-four manuscripts, one German, one Spanish and the rest English or French. Not all the manuscripts illustrate all the stages of the processes acted out by Theophilus; twenty-seven manuscripts, most of them (fifteen) psalters, books of hours, a breviary and a Bible, have only a single scene, and not all the same one, while the fullest pictorial account is given in the Besançon copy of Gautier de Coinci's Miracles de Nostre-Dame which devotes an exceptional twenty-seven of its 186 illustrations to Theophilus -- more than to any of the other miracles of the Virgin. The Smithfield Decretals is next, with twenty-three illustrations but treated as small marginal scenes; after that comes the Matutinalbuch with thirteen, the Carew-Poyntz Hours with twelve, and the De Brailes Hours with ten. The rest have between two and eight illustrations, while the Psalter of Troyes and miscellany, BnF ms lat. 238 (omitted here), has on fo. 78v a single miniature in two registers, depicting Theophilus praying before a statue of the Virgin and Child and below the Virgin handing a scroll inscribed THEOPHILE and touching him on the back as he kneels (in death?) before an altar with a diagonally placed cross on it. This is one of several full-page miniatures preceding major Psalms, this one preceding Psalm lxviii (‘Salvum me fac’). The date range is from about 1210 to about 1380 with a few outliers made in the fifteenth century. Especially noteworthy are three examples of Theophilus images in the context of chronicles, one in ms M.751, Abrégé des histoires divines in French, interrupting a list of popes, following John iii and preceding Vigilius. It depicts the devil holding a scroll and touching Theophilus' mouth with his pen (fig. 15), an interesting depiction of the fusion of writing and speech and of coercion, perhaps reminiscent of a medieval punishment for bearing false witness where stakes were driven through the tongue of the offender (as in the Costuma of Agen, BM, Agen, ms 42); and in two copies of the Speculum historiale of Vincent of Beauvais in the translation of Jean de Vignay, preceding the miracles of St Maur and followed by those of St Lomer in BnF, ms fr. 51 (dated 1463). A useful appendix charts which episodes are found in which manuscripts, and the book is copiouly illustrated with sixty figures, six in colour.

Root treats the subjects in a broad framework corresponding to what he sees as the four major parts of the story, beginning with the social commitments and responsibilities of homage, with contract and seal, in chapter i, leading to the burgeoning practice of self-examination and penance (which corresponds closely in my view to the 1215 requirement of confession) in chapter ii. This downward trajectory is most fully represented in the Besançon manuscript, while the presence of a Jewish intermediary, caricaturised, appears in the Lambeth Apocalypse and the Smithfield Decretals. Thus Root presents chapters i and ii as the broad image-based category of dissemblance, where sin transforms the resemblance man has to God (based on Genesis i.26) into a negative dissemblance. Chapter iii charts Theophilus' return to resemblance through the intervention of the Virgin, the most popular aspect of the story and the one most often selected for illustration. The Virgin fulfils the function of mediation, shifting the emphasis of story and picture to spiritual preoccupations and responsibilities. Root examines in detail the nature of the interaction between the Virgin herself, her image and Theophilus's perception of her and his growing awareness of his own self-reflection through a reorientation of seeing – with the eyes of the heart. Attendant details are whether Theophilus is asleep or awake, how the Virgin retrieves the charter, whether or not she chastises the devil, the juxtapositions and interrelations of the Virgin and her image. An interesting section is Root's analysis of Gautier de Coinci's language of images. Many of the illustrative sequences stop with the Virgin's retrieving of the contract, but the story continues in a further nine manuscripts to explore visually the new identity that the Virgin's retrieval and return of the contract make possible. What is now required of Theophilus is positive activity, his return to the church with the bishop, his open and public confession, and communion. The burning of the contract is part of this, but it is depicted only in the De Brailes Hours and in the Besançon Gautier de Coinci manuscript. Most interesting are the images where the patron or owner is represented, further underlining the affective aspects of the depiction of Theophilus as exemplum. The Madrid Cantigas manuscript includes King Alfonso x (fig. 59), and the patrons of Stowe 17 and Fitzwilliam 288 are inserted into Theophilus scenes, while KB 71 A 24 inserts an author portrait of Gautier de Coinci writing, and William de Brailes included his own portrait (fig. 60) between the burning of the contract and the soul of Theophilus borne to heaven. All in all this is an insightful study with broad implications for understanding the function of images and the ways in which perceptions of them were explored and exploited in medieval art and thought.