Keith Allen

Locke Defines Knowledge at the beginning of Book IV of the Essay concerning Human Understanding as “the perception of the connexion and agreement, or disagreement and repugnancy of any of our Ideas” (E IV.i.2).1 So defined, knowledge varies along two dimensions. On the one hand, there are four “sorts” of knowledge: of identity or diversity; relation; co-existence or necessary connection; and real existence. On the other hand, there are three “degrees” of knowledge: intuitive knowledge, which consists in the “immediate” perception of agreement or disagreement between ideas (E IV.ii.1); demonstrative knowledge, which consists in the perception of agreement or disagreement by way of intermediate (or intervening)..

This paper presents an ‘over-representational’ account of blurred visual experiences. The basic idea is that blurred experiences provide too much, inconsistent, information about objects’ spatial boundaries, by representing them as simultaneously located at multiple locations. This account attempts to avoid problems with alternative accounts of blurred experience, according to which blur is a property of a visual field, a way of perceiving, a form of mis-representation, and a form of under-representation.

Objects appear different as the illumination under which they are perceived varies. This fact is sometimes thought to pose a problem for the view that colours are mind-independent properties: if a coloured object appears different under different illuminations, then under which illumination does the object appear the colour it really is? I argue that given the nature of natural daylight, and certain plausible assumptions about the nature of the colours it illuminates, there is a non-arbitrary reason to suppose that it is under natural daylight that we are able to perceive the real colours of objects.

Jonathan Cohen has produced a powerful argument for Colour Relationalism: the metaphysical thesis that colours are relational properties of a certain sort—relational with respect to perceivers and circumstances. Cohen makes two important assumptions: one is that Colour Relationalism and Colour Irrealism (which include Colour Eliminativism, Fictionalism and other “error theories”) are rivals; the second is that “error theories” are theories of last resort. In this paper, I challenge both assumptions. In particular, I argue that there is good reason to think that Colour Relationalism needs to be supplemented by some version ofan Error theory. In so doing, I examine, in detail, the issues raised in a paper by Janet Levin in which she engages in an instructive debate with Colin McGinn. Levin’s paper is important since Cohen places heavy reliance on her arguments, both in defending his relationalist theory against crucial objections, and in his dismissal of error theories.

The view that the mind-dependence of colour is implicit in our ordinary thinking has a distinguished history. With its origins in Berkeley, the view has proved especially popular amongst so-called ‘Oxford’ philosophers, proponents including Cook Wilson (1904: 773-4), Pritchard (1909: 86-7), Ryle (1949: 209), Kneale (1950: 123) and McDowell (1985: 112). Gareth Evans’s discussion of secondary qualities in “Things Without the Mind” is representative of this tradition. It is his version of the view that I consider in this paper.

What, according to Locke, are ideas? I argue that Locke does not give an account of the nature of ideas. In the Essay, the question is simply set to one side, as recommended by the “Historical, plain Method” that Locke employs. This is exemplified by his characterization of ‘ideas’ in E I.i.8, and the discussion of the inverted spectrum hypothesis in E II.xxxii. In this respect, Locke's attitude towards the nature of ideas in the Essay is reminiscent of Boyle's diffident attitude towards the nature of matter. In posthumously published work, however, Locke suggests that the enquiry into the nature of ideas is one of the things that the enquiry into the extent of human knowledge undertaken in the Essay actually shows to lie beyond the “compass of human understanding”. In this respect, Locke's attitude towards the nature of ideas is reminiscent of Sydenham's attitude towards the nature of diseases.