Among the metaphors we live by (to borrow from Lakoff and Johnson), visual metaphors figure prominently in our common discourse about cognition. We are well aware, of course, that in claiming to "see" meanings or "view" ideas in certain lights, we are speaking figuratively. Yet the analogy between thinking and seeing is more than a linguistic artifact. Until the seventeenth century, Western thinkers took it seriously, none more seriously perhaps than the thirteenth-century scholastic community. The upshot was an epistemology based on two cardinal assumptions: first, that the physical world manifests itself to us through species passing from external objects into the soul, and second, that visual perception is providentially designed to yield a faithful conceptual picture of physical reality and its ordering structure.
This latter assumption found strong support in Perspectivist optics, which took mature form in the 1260s as an amalgam of Alhacenian visual analysis and species-theory. Hence, by the later thirteenth century it was widely believed not only that vision is essentially cognitive but also that cognition is essentially visual. During the first half of the fourteenth century, this idea was subjected to trenchant criticism and thereby rejected out of hand by several thinkers, among whom Nicolaus of Autrecourt is a particular subject of emphasis in this study.
It has become something of a truism lately that theories are neither accepted nor rejected on logical grounds alone. This one, Denery claims, is no exception since there were compelling theological grounds for its adoption. Foremost among these was a new emphasis on self-presentation in the thirteenth century. That emphasis manifested itself at two levels. On the one hand, the clerical orders were encouraged to give the right appearance and to suit the appearance to specific settings. This was especially important among the new preaching orders, whose members had to put on so many public faces that they came to adopt what Denery calls "heuristics of adaptation." On the other hand, new modes of confession and penance required deep introspection on the penitent's part, so that he might "see" himself for the complex of sinful motives and actions he is, as well as experience the contrition-yielding horror such a sight arouses. In both cases, the appearances should reveal the underlying reality: the preacher's spiritual commitment reflected in his mien, the penitent's contrition in the recognition of his vitiated nature.
But how can we be sure that what we see in these instances is true? This question is particularly vexing in the case of religious introspection, for, as St. Paul reminds us, our moral vision is distorted by the mirror that darkens it. Enter the analogy between physical vision and cognition. Visual illusions, such as those arising from reflection, were of paramount concern to Perspectivist theorists, who devoted considerable effort to rationalizing them according to unmet normative conditions, such as adequate illumination, appropriate distance, or a healthy eye. Reflected vision is subject to transgressions of these conditions, but adding insult to their injury is that, when seen in mirrors, objects invariably appear displaced and often look distorted in size, shape, and orientation. Fortunately, reason enables us to rectify such illusions according to specific rules that govern visual perception in general and reflection in particular.
If the analogy between physical and moral vision holds true, then reason ought to be the corrective in both cases. At first blush, this seems to be Peter of Limoges's position in the Tractatus moralis de oculo (c. 1280), but, as Denery insists, Peter's real point is that our inner, moral vision is so clouded we can rarely tell either how or even that it is distorted. The hoped-for normative circumstances and principles enabling us to rectify it are far too vague and protean to yield certainty from introspection. Given the tight conceptual linkage between vision and cognition, moreover, to question the reliability of introspection is almost automatically to...