Intuition and Construction in Berkeley's Account of Visual Space

Intuition and Construction in Berkeley's Account of Visual Space LORNE FALKENSTEIN BERKELEY IS P.F~OWNED as one of the major proponents of a constructivist program, in the theory of space cognition. According to this program, we do not directly or immediately see space. Rather, our experience of visual space is the product of association or unconscious inference. But while Berkeley originated this program, he only partially and ambivalently adhered to it. Unlike its nineteenth-century successors, Berkeley's theory does not reduce everything that pertains to visual space to a construction on more primitive data. In the end, it takes a significant and fi~ndamental aspect, two-dimensional layout, to be directly and immediately seen--or so I will show in what follows. This intuitionism about two-dimensional visual layout is an aspect of Berkeley 's theory of vision which he rarely acknowledges--on the contrary, he obfuscates it. As a result, his position has been an occasion for some divergence of interpretation. Historians of the theory of visual perception have tended to read Berkeley as denying that any sort of space, however primitive, is directly or immediately seen.' Philosophers commenting on Berkeley have ' In strictness I should say an associat/omstprogram, sincethe notion that certain features or relations of visual space (such as depth) need to be added by the perceiver antedates Berkeley (one finds it, for instance, in Descartesand Malebranche). Berkeley was the first to claim that the construction does not rest on necessaryconnection, but only on learned association. , Edwin G. Boring, Sensationand Perceptionin the Histo~ ofExperimentalPsy/cholog~(NewYork: Appleton-Century-Crofts, Inc., 194~), ~8; Nicholas Pastore, SelectiveHistoryof Theoriesof Visual Perception: z65o-z95o (New York: Oxford, 1971), 179; MichaelJ. Morgan, Molyneux'sQuestion: Vision, Touchand thePhilosophyofPerception(Cambridge: Cambridge UniversityPress, 1977),6i. A major exception to this tradition is Gary Hatfieid, whose recent The Natural and the Normativr Theoriesof Spatial Perceptionfrom Kant to Helmholtz (Cambridge, Mass.: MIT Press, 199o) breaks new ground in taking the theory of two-dimensional layout to be one of the most significant aspects of the history of theories of visual perception (see for example 131-3z). In this book [6M 64 JOURNAL OF THE HISTORY OF PHILOSOPHY 32" 1 JANUARY 199 4 taken him to be just as obviously convinced that two-dimensional layout is originally present in vision.s While I think the philosophers are correct, I also think that the question is far from being clear-cut, and that all sides to the dispute have hitherto paid too litde attention to just how grudging and oblique Berkeley's concession to intuitionism is, and have not given careful enough consideration to the problem of how Berkeley could advocate intuitionism about two-dimensional layout while still remaining true to central tenets of his. theory of vision.4 It is the latter problem which I propose particularly to canvass in this paper. Before turning to it, however, I will first demonstrate that Berkeley's theory of vision is indeed intuitionist about twodimensional visual layout. Though this has not been questioned in certain circles, it has been recently denie d , and even those who do not question Berkeley's intuitionism should appreciate that the question is not as straightforward or as easily answered as they may hitherto have supposed. 1. THE CONTROVERSY OVER BERKELEY'S INTUITIONISM Berkeley published his major work on visual perception, the New TheoTy of V'ts/on, in I7O9, but it was over a hundred years later that the program he had defined caught the fancy of philosophers and psychologists interested in the problems of perception. From then on it rapidly gained ascendancy.s Hatlield does not explicidydraw the conclusion that Berkeley'sconstrnctivismwas mitigated, but see Hatfield and Willianj3Epstein, "The Sensory Core and the Medieval Foundations of Early Modern Perceptual Theory,"Is/s70 (1979): 363-84, esp. 380 and n. 67. sD. M. Armsmong, Ber~L,,y'sTheorj of Vision(Melbourne: Melbourne University Press, 196o), 55; George Pitcher, Berkeby(London: Roudedge & Kegan Paul, t977), 53-54; Alan Donagan, "Berkeley's Theory of the Immediate Objects of Vision," in Peter K. Machamer and Robert G. Turnbull, eds., Studiesin Pc.rcepfion(Columbus, Ohio: Ohio State UniversityPress, 1978),3x~-35, esp. 318-~o. For previous...