Locke, John (1632–1704)

4. Five sorts of idea

Book II of the Essay presents an alternative to Aristotle’s doctrine of categories, the traditional typology of entities capable of being named or predicated (see Aristotle §7). That Locke’s classification is of ideas rather than of things stresses that the categories are purely conceptual. He identifies five broad types: simple ideas, ideas of simple modes, ideas of mixed modes, ideas of substances and ideas of relations. Simple ideas come first in the Lockean order of knowledge, as substances come first in the Aristotelian order of being. Simple ideas are necessarily given in experience, whereas complex ideas can be constructed by ‘enlarging’ (’repeating’) or ‘compounding’ simple ideas. Ideas of relations result from ‘comparing’ ideas. ‘Abstracting’ is more a matter of focusing on an idea or, better, an aspect of an idea, whether given or constructed, than of creating a new one (see §3). Locke sometimes acknowledges that the overarching compositional model is problematic in its application, but it is put into doubt even by his formal introduction of the notion of a simple idea. The ideas of the sensible qualities of a body, Locke claims, though produced by the same body, in some cases by the same sense, are evidently distinct from one another, each being ‘nothing but one uniform appearance, or conception in the mind’ (Essay II.ii.1). Yet to ascribe the conceptual distinctions between, for example, a thing’s shape, its motion and its colour to a primitive articulation of appearance is to beg a crucial question.

Under the topic of simple ideas Locke expounds his famous distinction between primary and secondary qualities (Essay II.viii). Since the cause of a simple idea may be quite different in character from the idea itself, we should distinguish the idea in the mind from the corresponding quality (that is, the power to cause the idea) in bodies. Certain qualities, however, are necessary to our conception of bodies as such. These are the primary qualities, ‘solidity, extension, figure, motion, or rest, and number’, just those which figured in corpuscularian speculations. Locke’s proposal (displaying the tension, described in §3 of this entry, between two conceptions of representation) is that in the perception of a primary quality the represented cause, the basis of the power in the object, is qualitatively like the idea caused: ‘A circle and square are the same, whether in idea or existence’ (Essay II.viii.18). Only this will allow that the action of external bodies on the senses is ‘by impulse, the only way which we can conceive bodies operate in’ (Essay II.viii.12) – an appeal to the seventeenth-century commonplace that mechanical explanations are peculiarly intelligible. But then ideas of ‘colours, sounds, tastes, etc.’, Locke’s ‘secondary qualities’, must also be mechanically stimulated. Hence secondary qualities ‘are nothing in the objects themselves, but powers to produce various sensations in us by their primary qualities, i.e. by the bulk, figure, texture, and motion of their insensible parts’ (Essay II.viii.10). Ontologically they are in the same boat as the power of fire to cause pain or, indeed, its power to melt wax.

The idea of power itself Locke attributes to experience of regular patterns of change, giving rise first to expectations that ‘like changes will for the future be made in the same things, by like agents, and by the like ways’, and then to the thought that in one thing exists the possibility of being changed and in another ‘the possibility of making that change’ (Essay II.xxi.1). So we form the idea of power, active and passive: the power of fire to melt wax and the power of wax to be melted are aspects of fire and wax known and identified only through their joint effect. The idea of power is thus a place-marker for attributes which could in principle be known more directly.

The ideas or experiences of pleasure and pain are important simple ideas, since they are responsible for our ideas of good and evil, and are ‘the hinges on which our passions turn’ (Essay II.xx.3). (This hedonistic theory of motivation and value is examined in §9 of this entry.)

‘Simple modes’ constitute another problematic category. Locke starts with modes of extension, the subject-matter of geometry, with which he compares modes of duration. Here his thesis is that we acquire ideas of particular modes of extension (that is, determinate lengths and figures) or duration (that is, periods) in experience, and can then repeat (or divide) them so as to construct ideas of possible lengths, figures or periods not previously experienced. Roughly, ‘modification’ here is compounding like with like. The same model supplies Locke’s account of ideas of numbers, achieved by the repetition or addition of units, aided and ordered by the linguistic technique of counting. Yet he also recognizes qualitative simple modes, effectively conceding that ideas of different ‘shades of the same [experienced] colour’ are constructible. Even with quantitative ‘modes’, where the ‘repetition’ model has some plausibility, it is problematic what is a simple idea. The idea of determinable extension is a plausible candidate, with its determinates as ‘modes’, but the repetition model presupposes simple units. Locke impatiently responds that the smallest sensible point ‘may perhaps be the fittest to be consider’d by us as a simple idea of that kind’ (Essay (5th edn) II.xv.5n.), but he was evidently more concerned to argue that ideas of novel determinate figures are somehow constructible from what has been given, and so to subvert a Platonic-Cartesian argument for innateness, than to insist on the adequacy of a rigid compositional model.

Another target in Locke’s account of simple modes is Descartes’ conceptual identification of space and matter in the thesis that the essence of matter is extension. For Locke, both the essence of matter and the nature of space are unknown. He argues that our idea of a vacuum is not contradictory, since our ordinary idea of body includes solidity as well as extension, but he declines to choose between relational and realist theories of space. Yet comparison of the Essay with earlier notes and drafts indicates that, having first held a Hobbesian relational view, Locke came gradually to favour a realism close to that of Newton (see Descartes, R. §§8, 11; Newton, I. §4).

Ideas of mixed modes arise with the combination of unlike simple ideas, as in the idea of a rainbow. But Locke’s paradigms are ideas of human actions and institutions, the materials of demonstrative moral and political theory. Like ideas of geometrical figures, ideas of mixed modes can properly be formed without regard to what exists. Ethical thought is none the worse for being about a virtue or motive or political constitution which is nowhere actually instantiated. Ideas of substances are different, for they concern the real rather than the ideal: ‘When we speak of justice, or gratitude…our thoughts terminate in the abstract ideas of those virtues, and look not further; as they do, when we speak of a horse, or iron, whose specific ideas we consider not, as barely in the mind, but as in things themselves, which afford the original patterns of those ideas’ (Essay III.v.12). Moreover, whereas ideas of substances are formed on the presumption that the complex idea represents a really or naturally united thing, the unity of mixed modes is essentially conceptual. Indeed, ‘Though…it be the mind that makes the collection, tis the name which is, as it were the knot, that ties them fast together’. Different languages slice up the field of human life and action in different ways, determined by the practices and priorities of the communities that speak them. This thesis can be extended to natural modes such as freezing, since even here it is the term tied to a striking appearance, not a natural boundary, which slices out the particular process from the general process of nature. That, Locke plausibly assumes, is not how it is with horses.

The chief thought behind Locke’s somewhat confusing account of ideas of substances is that our idea of a thing or stuff is a compound of ideas of its qualities, but the thing itself is not a compound of qualities (Essay II.xxiii). The substance–accident structure is a feature of our ideas and language, not a structure in reality. It is a feature which marks our ignorance of the underlying nature of things, since we always conceive and talk of a substance as a thing possessing certain qualities, that is, as a ‘substratum, in which [the qualities] do subsist, and from which they do result, which therefore we call substance’. The mistake of dogmatic philosophers is to think that they can form simple conceptions of substances matching their unitary natures. Aristotelians are so misled by language that, just because, ‘for quick despatch’, we employ one name, ‘gold’ or ‘swan’, they think it a ‘simple term’ corresponding to a ‘simple apprehension’. Cartesians take the simple essences of matter and spirit to be extension and thought. Yet so far are we from catching the nature of any thing in our complex idea of it that, if it is asked what the subject is of the qualities by which we define it (the colour and weight of gold, for example), the best answer we can give is ‘the solid extended parts’, that is, the mechanistic ‘corpuscularian’ hypothesis as advanced by Boyle. If it is asked in turn ‘what it is, that solidity and extension inhere in,’ we can only say, ‘we know not what’. Our idea of the substance is of ‘nothing, but the supposed, but unknown support of those qualities, we find existing, which we imagine cannot subsist, sine re substante, without something to support them’. Such an idea is ‘obscure and relative’. Ideas of specific substances are ‘nothing but several combinations of simple ideas, co-existing in such, though unknown, cause of their union as makes the whole subsist of itself’ (Essay II.xxiii.6). Locke’s point is that no theory, not even the corpuscular hypothesis, gives an account of the ultimate nature of things.

Finally come ideas of relations – father, son-in-law, enemy, young, blacker, lawful and so on (Essay II.xxv–xxviii). Like ideas of modes, ideas of relations can properly be constructed without regard to reality, in particular if they are conventional relations. Adequate ideas even of natural relations, Locke claims, are possible without adequate ideas of the things related: we can grasp the essence of fatherhood without knowing the essence of man or even the mechanisms of reproduction. His point is that the biological details are irrelevant to the rights and duties of a father – a question rationally determined in his own attack on patriarchalism in Two Treatises. From this point of view, relations are theoretically close to modes. Yet Locke does allow certain relations to have peculiar ontological significance. Causal, spatial and temporal relations are universal relations which pertain to all finite beings. Identity and diversity are so too: a thing is diverse from anything existing in a different place at the same time, ‘how like and indistinguishable soever it may be in all other respects’, and the continuity of individual substances is spatio-temporal. The last important type of relations to be picked out for special discussion is that of moral relations, or the relations of actions to some law ‘whereby good or evil is drawn on us, from the will and power of the law-maker’ (Essay II.viii.5).