This book is an original contribution to the debate on the meaning and role of divine illumination in Augustine's thought and in the later tradition. It has a clear narrative involving three ‘discoveries’: (i) that Augustine provided an intrinsicist rather than extrinsicist account of divine illumination (chapter 1); (ii) that Bonaventure was not the great medieval exponent of Augustine that historians have thought him to be (chapter 4), for that honour falls to Aquinas (chapter 5); and (iii) that the later Franciscan (Scotus et al.) rejection of Bonaventure's teaching on divine illumination did not mark a journey into the Aristotelian fold, but lay the basis for the modern distinction between reason and faith as ‘mutually exclusive extremes’ (chapter 6). In chapter 3, Schumacher shows how originating with Alexander of Hales, the Franciscan tradition was influenced by Avicenna's metaphysics, which possesses a notion of abstraction quite distinct from that of Aristotle and Thomas, the latter being ‘more compatible’ with Augustine's view of the ‘mind's cognitive work’ (p. 99). These chapters, where Schumacher addresses the ‘utterly distinctive Franciscan intellectual tradition’ (p. 237) with its roots in the metaphysics of Avicenna, are particularly successful. In the seventh and final chapter, ‘The Future of Augustine's Theory of Knowledge’, Schumacher attempts to provide the outline of a new theological theory of knowledge, involving ‘a recovery of Augustine's understanding of knowledge by illumination’ (p. 218).

The other major character in the book is Anselm (chapter 2). The chapter on Anselm omits any reference to a pertinent section of Anselm's Monologion for a theory of knowledge (chh. 65–67). In M66, Anselm writes that it is only through the rational mind itself that the rational mind ‘is most able to progress in the discovery’ of the supreme essence. The mind is the mirror and image of God, in which it sees Him, whom it cannot see directly (cf. M67). In an Anselmian-Augustinian theory of knowledge there is in the human knower a ‘lux rationis’ (M6), a ‘lucerna veritatis’ (M19), that is the mirror and image of the divine light. The divine light is not something external to the human knower, but, rather, like the divine image is constitutive of what it is to be human. This would appear to support Schumacher's thesis concerning the intrinsicist nature of illumination in the Augustinian tradition.

It is not the case that the only authority Anselm names is Augustine, nor that Augustine is named only in the Monologion and Proslogion (pp. 71f.). Aristotle is named on eleven occasions (ten in De Grammatico and one in Cur Deus Homo). Augustine is not named in the Proslogion, but is named in the Monologion, De Incarnatione Verbi and letters. Schumacher's treatment of the Proslogion argument (pp. 76–82) needs to take account of the fact that God is not the differentiated species of a genus and therefore cannot be defined in the dialectical tradition in which Anselm operates. The middle term of Anselm's argument is not ‘Supreme Being’, but ‘that than which nothing greater can be thought’. Substitute the former for the latter and his argument does not work.

Schumacher is keen to deny to the Augustinian intellectual tradition what she sees as the Franciscan rationalist use of the a priori. She believes the term a priori to mean unrelated to experience, and to carry with it the notion of direct intuition or innate knowledge apart from experience (cf. pp. 145, 149, 214). But a transcendental argument, for example, is one that concerns the (non-intuitive) a priori conditions governing the possibility of experience. Augustine's teaching on divine illumination can be seen in such terms. Furthermore, I would argue, and think that this argument strengthens Schumacher's position, that Anselm's Proslogion argument, when viewed in the context of the Monologion and the wider Augustinian tradition (which is how Schumacher wishes us to view it, cf. p. 72), is to be understood as a transcendental deduction of the image of God in the human knower. It is not necessary to adhere to the Kantian terminology, but the point here is that this would make it an a priori argument. Although the author explicitly denies a Kantian ‘turn to the subject’ in Augustine's theory (p. 57), in the penultimate paragraph of the book she comes close to recognizing a transcendental turn, referring to ‘the precedent that Anselm and Aquinas set when they themselves updated Augustine's claim that divine illumination is the condition of possibility of all human knowledge’ (p. 239).

Although Schumacher makes only one explicit reference to him (p. 222), hovering in the background of this book is the figure of Michael Polanyi, who also set out to restore the balance between belief and reason along Augustinian lines. For Polanyi, Augustine was the first ‘post-critical’ philosopher, who taught that knowledge was guided by antecedent belief (cf. Personal Knowledge, p. 266.) But Polanyi's ‘fiduciary programme’ is not identical to that of Augustine. ‘Antecedent belief’ is primarily a ‘secular’ concept, employed in an attempt to wean modernity away from a ‘scientific rationalism that would permit us to believe only explicit statements based on tangible data and derived from these by a formal inference, open to repeated testing’ (The Tacit Dimension, p. 62). It may leave room for religious faith, but it is not identical with religious faith. Schumacher's account of faith appears to shift between a weaker (Polanyian) notion of antecedent belief and a stronger (Augustinian) notion of faith, without fully clarifying the boundaries between the two. Thus, from the premise that reason ‘involves elements resembling faith’, she draws the conclusion that faith is therefore justified ‘in virtue of the role it plays in rendering reason functional and sound’ (p. 218). However, the conclusion does not follow from the premise, unless she means here by faith the weaker ‘antecedent beliefs’ that we bring to knowledge, which are not self-evidently theological. But her argument requires faith to have the stronger theological meaning. Schumacher is aware of the ‘lingering question’ of the identity of these weaker and stronger claims (p. 226), and in the final few pages of the final chapter she provides an outline of how she would go about establishing the stronger claim, that belief in the Christian God is rational, ‘inasmuch as it bears the burden of rendering reason functional and sound’ (p. 232). It will be interesting to see how a distinctively theological theory of knowledge could be developed along the lines she suggests and I look forward with much anticipation to her making ‘a full-blown effort to do this in future’ (p. 223). Why is there no bibliography?