As A.D. Smith suggests in his essay "Of Primary and Secondary Qualities", the philosophical implications that extend from any system which proposes to define these two qualities of material objects - and, therefore, any theoretical distinction between them - are quite extensive. Since the question of primary and secondary qualities cuts directly into the essential problems of sensory experience and its subsequent human interpretation, the problem is hardly a minute, academic footnote. Rather, any inquiry into the question of primary vs. secondary material qualities is bound to inherit no less a philosophical problem than "the nature of the relation between the senses and intellect and the physical world" (Smith, p. 223) in its entirety - no small issue, indeed.

It therefore comes as little surprise that all inquiry into John Locke's particular understanding of and distinction between possible primary and secondary qualities must find origin in something as fundamental as Locke's own conception and classification of ideas themselves. Indeed, it is within such framework, as regards external, physical objects and their relationship to the mind's concepts of them as such, that the question of primary-secondary distinction arises at all. Insight into the Lockean primary-secondary distinction must therefore begin with an understanding into Locke's philosophy of ideas itself, and therefore, into Locke's concepts of how material objects become represented within the faculties for understanding within the human mind.

Locke first sets his general definition of an 'idea' early on within his Essay, establishing it as "the object of the understanding when a man thinks" (Essay, I.I. §8). Although, as J.L. Mackie notes, Locke will eventually make efficient use of this word not only
to refer to the elements that form the content of a sensory perception while we are having it, but also to remembered images, imaginary constructions, and concepts, apparently without realizing the difficulty of making a single sort of item do all these jobs (Mackie, p. 11),
the term is chiefly and most generally used to describe any item which is found within the working intellect.

Such items, Locke goes on to argue, all derive their existence "from experience," (Essay, II.I. §2), or rather, from external sensory impressions. "In that {i.e. experience}," he continues
all knowledge is founded;…our observation employed either about external, sensible objects; or about the internal operations of our minds…is that, which supplies our understandings with all the materials of thinking (ibid, emphases mine).
Thus the distinction initially arises between those ideas which derive from direct sensory experience with objects and those which are instead actively produced by the human imagination or intellect. This distinction, which perhaps partially mollifies Mackie's earlier objection to vague usage of terms, will lead Locke to classify all ideas as belonging to one of two subsets, simple or complex.

As previously implied, Locke will go on to initially define simple ideas as those which "enter by the senses, simple and unmixed" (II.II.§1), although he will later expand upon this definition. What is important to note, however, is that such simple ideas are each created entirely within the intellect, as Michael Ayers puts it, "not because the perceived qualities are really distinct entities in the object…but because of the variety of ways in which the object acts on us through the senses" (Ayers, p. 9). In other words, it is not necessarily the case that because one might have a simple idea of an object's colour or taste, for example, that these simple ideas must correspond directly to physical reality. Our senses, as it were, are imperfect tools, and although they do provide us with simple ideas, not all of these arrive in a form that is perfect, whole and direct.

Locke also allows that simple ideas do not always arrive from the operations of a single independent sense. Multiple senses at work simultaneously, as when one watches the contour of an object at the same time that they are touching, can produce the same simple idea, that of shape; an object's "Space, Extension, Figure, Rest and Motion" (Essay, II.V) are each listed by Locke as simultaneously verifiable by the sight and touch. Meditation, reflection, and remembering, also, are permitted as sources of simple ideas, so long as the mind which engages in such actions does so passively and without "volition" or use of the "will". Pleasure, pain, power, existence, and unity (and their inverses, where appropriate), for instance, are all listed by Locke as simple ideas that the mind may produce. Most generally, however, Locke argues "that whatsoever is…able, by affecting our senses, to cause any perception in the Mind, doth produce in the understanding a simple idea" (Essay, II.VIII.§1).

Complex ideas, on the other hand, are entirely the domain of the active mind. "When the understanding is once stored with these simple ideas," Locke argues, "it has the power to repeat, compare, and unite them to an almost infinite variety, and so can make at pleasure new complex ideas" (II.II.§2). Simple ideas, as previously distinguished, are therefore the raw building blocks, as Locke sees it, to the various complex structures that the human intellect can create from them. Locke lists the possibilities of complex ideas in some detail, which include unity and abstraction among their ranks, but for the purposes of this essay, it is enough to say that they must arise from simple ideas.

Ayers has some problem with this concept of simple versus complex ideas, and he notes that, for instance, "a pink patch does not appear as a duality of shape and colour, however natural it may be to abstract one from the other" (Ayers, 9). True enough, but this is not a problem per se within the simple/complex dichotomy of ideas, since the observation of a pink patch (or of any object, really) is entirely within the domain of simple ideas. Ayers conclusion, then, that "there is…something problematic about Locke's conception of given simple ideas… and the compositional simple-complex model" is not necessarily true, although it does raise some interesting questions about the process of sensory experience.

What Ayers essentially assumes is that the concept of a simple idea implies that a mind can only receive a single piece of sensory information at any given time, and that just as the texture or smell of an object provides (or at least seems to provide) just one idea at a time, so must all simple ideas received from vision be restricted to one genre of information at any given moment. Yet, Locke clearly allows that simple ideas can be received from two senses simultaneously (cf. II.V of his Essay), and therefore allows for multiple simultaneous bits of sensory experience to form one single, simple idea. Furthermore, he includes 'abstraction' within his list of complex ideas. One assumes, therefore, that his answer to Ayers would run thusly: that, indeed, the viewing of a pink patch does appear immediately to the mind as an unmitigated hybrid of shape and color, but it is in the process of stripping the patch down to these abstract, bare qualities that one starts to actively create complex ideas. The mind receives sensations of shape and of color simultaneously, in other words - as it might at once receive shape and contour instead - but it forms independent abstract concepts of these two qualities afterwards; rather, Locke would argue, one's mind does not receive color and shape independently and piece these together abstractly to form a pink patch.

This question, however, raises important issues with regards to Locke's concept of an object's "qualities" and how these relate to a mind's simple idea of that object. Locke defines the "qualities" of an object as "the power to produce any idea in our mind" (Essay, II.VIII.§8), and makes a clear distinction between these qualities at work within the object and the simple ideas which they produce in an observant mind. In other words, pink and patch-shaped are independent "qualities" of a pink patch (at least one would hope that they were), but both come together to act upon a mind simultaneously and produce one single, simple idea.

Locke's goes on to distinguish further, however, between such qualities inherent within a given object. "Qualities thus considered in bodies are," he argues, "first such as are utterly inseparable from the body," (Essay, II.VIII.§9) or physical manifestation of a given object. He continues:
Such qualities, which in truth 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 as colours, sounds, tasts, etc. These I call secondary qualities. (Essay, II.VIII.§10)
He also adds that there are such tertiary qualities "which are allowed to be barely powers", which do not operate directly upon the intellect itself, but rather, upon other objects.

What Locke refers to as 'primary qualities' are therefore linked initially to those attributes which are 'utterly inseparable' from an object's physical existence, whereas the 'secondary qualities' of an object are those which are attributed to the 'motion of its insensible parts.' Thus is the distinction between primary and secondary initially drawn along lines of sensibility - primary qualities are those which are immediately sensible, whereas secondary qualities are mere reflections of the 'insensible' mechanisms at work within an object, "produced…by the operation of insensible particles on our senses" (II.VIII.§13).

Locke goes on to create other methods of distinction between the primary and secondary qualities of an object, however. In §14 of the same book and chapter, he suggests that secondary qualities "depend on those primary qualities," thereby introducing the idea of a relationship between the two categories. He also states, in §15, that "the ideas of primary qualities are resemblances of" the qualities themselves, whereas "the ideas, produced in us by these secondary qualities, have no resemblance of them at all." This he reiterates in §22, noting, with regard to secondary qualities, the "difference between the qualities in body and the ideas produced by them in mind."

These are, however, four or five different portions of the same concept: that the imperceptible secondary qualities of an object, by virtue of their imperceptibility, can only be understood indirectly through the senses, not directly like the primary qualities. The secondary qualities do 'depend' on the primary qualities, in the sense that they too have "bulk, figure, texture, and motion" (Essay, II.VIII.§14), and, because of their imperceptibility, they do not equate to the ideas that we have of them in the same way that primary ideas such as solidity (which I gloss as meaning the same thing as 'mass'), extension, figure, or mobility, directly correspond to our ideas of them as such.

For example, if one takes our pink patch again, and, for the sake of argument, turns it instead into a three-dimensional pink sphere (so as to create more mass), Locke would argue that the shape, mass, and motion of the sphere itself were primary and irrefutable physical qualities of the object, whereas its pinkness would be a secondary attribute. This is because the actual physical mechanisms that went into making this particular sphere pink (which involve the particular ins and outs of optical physics) are operating on a scale and speed that exceed our sensory capacities. Mackie puts this much better than I do:
It is not that we sometimes make mistakes about colours…but that even under ideal conditions, when we are as right as it is possible to be about colours, colours as we see them are totally different not only from the powers to produce such sensations, with which Locke equates the secondary qualities, but also from the ground or basis of these powers in the things we call coloured (p. 14).
The main implication, then, of the fact that the secondary qualities of an object operate outside of the range of our senses falls upon Locke's understanding of ideas within the mind. If our senses cannot 'pick up' the secondary qualities at work directly, then the obvious consequence is that there is a distortion between these qualities as they really work within an object and how that object exists within the mind itself, which can only comprehend the secondary mechanisms at work indirectly. Smith puts this much better than I do:
It is best to characterize a quality as primary or secondary by reference to a concept or (simple) "idea" to which it corresponds…We start with a battery of conceptual responses and we critically evaluate them; those that are discarded as failing faithfully to represent the world as it really is are regarded as being responses to secondary qualities…There is nothing intrinsically dubious or second-rate about secondary qualities: they are perfectly objective characteristics of the furniture of the world: it is our representations of them that are inadequate or misleading (pp. 229-30).
In other words, it's not a given object's fault that we don't comprehend it fully - it's the problem of our inherently limited senses. Smith later comments on "the desire for a given mind to delineate the intrinsic nature of physical objects"; he concludes that "concepts are secondary when they appear to do this, but actually (and, I might add, necessarily) fail" (p. 232, emphasis mine).

There are several possible questions, however, that may be raised against Locke's view on primary and secondary qualities in matter. One of the most notorious, Molyneux's problem, involves the theoretical problem of a man who has been blind since birth who regains his eyesight at an advanced age. Will he, the question runs, be able to distinguish spatial relationships the way that other men do? The implication, one might assume, if he cannot, is that all visual-spatial relationships are somehow non-primary, and that the distinction between the two genres of qualities is therefore arbitrary. This position is combated nicely, however, by Mackie:
We can infer that Locke's reason for saying that Molyneux's man would not be able, before touching the objects he was now seeing for the first time…is that this man would not have acquired automatic interpretations of various patterns of shading as indicators of three-dimensional shapes such as spherical convexity or the corner of a cube projecting towards a viewer. (p. 30)
In other words, the fact that the visual acuity of this formerly-blind man was not yet fully developed does nothing to change the objective reality of the mass, shape, or motion of the object before him.

Another attack that can be brought against Locke's primary/secondary distinction is in the supposedly universally accurate quality of primary qualities themselves. If, for instance, it were to be found that mass, shape, and motion were somehow not the most objective methods of distinguishing matter, then it would appear that primary qualities would be unseated from their place of objective dominance. Smith answers this well:
Another reason for the problematic nature of the applicability of the concept of a primary quality today is the waning of the picture of matter as a single ultimate physical nature, and, with it, the idea of Catholick i.e. universal, clear affections of matter… In view of this, my inclination is either simply to ditch the term "primary quality," or use the term to advert to the properties deemed fundamental by current science (pp. 252-3).
Finally, one last problem with Locke's primary/secondary distinction also attacks the primary qualities, on the grounds that they are too cozy with the particular interests if physicists. This is refuted beautifully by Mackie:
The physical considerations do not concern merely features which are scientifically interesting and important; they show that there is no good reason for postulating features of a certain other sort, namely thoroughly objective features which resemble our ideas of secondary qualities. (p. 18)
In other words, for this objection to be valid, one would have to imagine colors or tastes or other secondary qualities as somehow being the objective standard for all objects, which is clearly absurd.

Locke's system of ideas, and his primary/secondary distinction in particular, is therefore enormously important for an understanding not only of the external physical world, but also of how the human mind perceives that world around itself. It not only acknowledges but also makes room for the limitations of the human senses, and by doing so, clears the way for an objective scientific method. It provides additional philosophical tools for Cartesian skeptical inquiry, and suggests that the best place to begin looking for truth is in the limitations of our own sensory abilities.


Primary Sources

Locke, John. An Essay concerning Human Understanding Nidditch, Peter H., ed. Oxford University Press, Oxford, 1975.

Secondary Sources

Ayers, Michael. Locke: Ideas and Things. Phoenix Press, Great Britain, 1997.

Mackie, J.L. Problems From Locke. Clarendon Press, Oxford, 1976

Smith, A.D. "Of Primary and Secondary Qualities", pp. 221-254, (from Philosophical Review, vol. XCIX, 1990).