Ideas and words
Like Hobbes, Locke attached his empiricist account of the origin of ideas to a theory of meaning. He was motivated by a belief that scholastic and Cartesian philosophy achieve their interesting results largely by assuming that certain key terms have a meaning and that the meaning is understood. On examination, however, these terms are often found to have a meaning other than the one intended, or sometimes no meaning at all.
Words have meaning, according to Locke, because they are the ‘signs’ of, or ‘stand for’, ideas. (Not much of a theory, of course, since ‘sign’ and ‘stand for’ are precisely the terms that need to be explained by a theory of meaning.) Communication is the process whereby words, which are attached to ideas in my mind, issue from my mouth and impinge on your ear, so causing the same ideas to arise in your mind.
The theory is open to serious criticism. In particular, it confuses the relation of meaning, which is governed by rules and conventions, with the natural relation between a word and the ideas that are aroused by it. The word ‘cow’ conventionally signifies a certain kind of animal; but it arouses in many people the ideas of milk, farmyards and pasture. Laurence Sterne put the criticism in a nice piece of satire:
—My young master in London is dead! said Obadiah—
—A green satin nightgown of my mother's, which had been twice scoured, was the first idea which Obadiah's exclamation brought into Suzannah's head.—Well might Locke write a chapter on the imperfection of words.—Then, quoth Suzannah, we must all go into mourning.—But, note a second time: the word mourning, notwithstanding Suzannah made use of it herself, failed also of doing its office; it excited not one single idea, tinged either with grey or black,—all was green.—The green satin nightgown hung there still. (Tristram Shandy, Book 5, chapter 7).
One of the achievements of modern philosophy, an achievement which is owed largely to Wittgenstein, is that it has taken the point of such satire seriously. It has given proper foundation to the view of language as a practical skill, governed by conventions which need make no reference to such accidental occurrences as Locke’s mental ‘ideas’. It could be further objected to Locke that, on his own account of what an idea is, I could never know that you mean the same by a word as I do. In particular, the idea that I associate with the word ‘pain’ might be associated by you with the word ‘pleasure’; this difference between us lying as it were undisclosed beneath the mask of our common usage. Such a theory, which removes from meaning its essential ‘publicity’, would for this reason now be almost universally rejected.
The physical world
It remains now to state briefly the view of the world and of scientific enquiry that Locke derived from his theory of knowledge. In many respects this view reflected an improved theory of the nature of science; some aspects of it have indeed been restored to favour in recent years as scientists have come to understand their utility. Locke derived from his friend Robert Boyle and ultimately from Pierre Gassendi (1592-1655) an interest in the distinction between primary and secondary qualities. He also enquired—in a wholly novel and illuminating way—into the concepts of essence and substance, endeavouring both to reinstate them as fundamental scientific notions, and at the same time to free them from the metaphysical confusion introduced by rationalist ways of describing them. In this he made a philosophical step the significance of which was unappreciated for over two centuries.
Among complex ideas Locke distinguished those of modes, substances and relations. These correspond to the grammatical categories of predicate, subject and relation. As he sometimes seemed to recognise, however, it is not right to say that we have an idea of the individual substance. Part of the obscurity in the theory of abstract ideas comes about because all ideas seem to be inherently general: that is, they represent properties, of which it would make sense to say that more than one object possesses them (just as more than one person may exactly correspond to the image in a painting). How then do we arrive at a conception of the individual thing which is the subject of predication? Locke was anxious to avoid the paradoxes of Spinozism, and to preserve a notion of substance that allowed for the existence of many—possibly infinitely many—substantial things. So he could not take refuge in the Cartesian idea of substance.
It is first necessary, Locke thought, to distinguish ideas from qualities; qualities being the powers of objects to produce ideas in us. Primary qualities are supposedly both inseparable from the objects in which they inhere, and also generative of simple ideas. They are the qualities of extension, motion, mass and so on, and are the true subjects of scientific investigation. Secondary qualities are nothing but certain powers to produce sensations (the power of sugar to produce a sweet taste, of red things to produce certain characteristic visual impressions, and so on).
It is difficult to be precise about this distinction (which could be drawn differently for different purposes). But one assertion that Locke makes about it is certainly of crucial significance, both historically and philosophically. Whereas primary qualities resemble the ideas that are produced by them, secondary qualities do not. And this enables us to say that there is a sense in which primary qualities are really in the objects which possess them, whereas secondary qualities are not. Berkeley objected to this, saying that it is absurd to suppose that any quality of a material substance can resemble an idea, since ideas are mental entities, belonging to a wholly different realm, and it is prima facie absurd to suppose that ideas can resemble things which are not ideas.
In order to reply to this objection, we must attempt once more to free Locke’s insight from the dead theories which enclose it. We must recognise that, in speaking of a resemblance between ideas and qualities, he was misdirecting his thoughts in a way encouraged by his theory of meaning. In some sense, as Locke saw, certain scientifically determinable and measurable qualities are basic to the reality of a thing in a way that other qualities are not. The secondary qualities seem to stand in need of a perceiver, the primary qualities only in need of an object. One way of putting the point is this: if you know all the primary qualities of an object, and the nature of the man who perceives it, then this alone will enable you to explain how that object appears to him. There is no need to refer to the secondary qualities of the object in order to explain how it is perceived. The primary qualities can be said to resemble our perceptions of them in the sense that they themselves must be invoked in explaining that perception. To say this is to deny not the reality of secondary qualities, but only their centrality in any scientific view of the nature of the object that possesses them.
Real and nominal essence
Seen in this way, the distinction between primary and secondary qualities relates to another of Locke’s distinctions, that between real and nominal essence. Locke makes this new distinction in the course of exploring the nature of material things, and in subjecting the scholastic ideas of ‘substance’ and ‘essence’ to critical examination. If we construe ‘substances’ to be individual things, the bearers of qualities, then we can have no positive conception of them. They are the ineffable substrata which ‘support’ those qualities through which any object is known. Any positive conception of the individual is the idea of a quality and therefore not of the substratum itself.
Let us leave aside the (for Locke) extremely difficult question how we might then come to have such an idea as that of substance. Locke, in common with many philosophers, influenced directly or indirectly by Aristotle, recognised that such a negative conception leaves us with the task of defining the nature of an individual. An individual cannot be identified as a particular substance (even if it is identical with such a substance) since of substances, considered in isolation from their qualities, nothing can be said. As the scholastics put it, ‘individuum est ineffabile, (‘the individual is ineffable’), a doctrine which Locke in the end is driven to support. It is therefore necessary to separate among the properties of a thing those which define its essence from those in respect of which it mig
ht change without changing its nature. This is the closest we can get to the idea of an individual.
But what is this essence? In fact, Locke now speaks not of individuals but of kinds. The scholastic idea of an individual essence seemed to him to be incoherent. He regarded all problems of individuality as exhausted by enquiries on the one hand into the fundamental kind to which an individual belongs, and on the other hand into the conditions of its identity. Except for the general idea of a ‘substratum’ there was nothing to be said by way of characterising the nature of a thing. And it is possible to doubt that Locke’s empiricist theory of meaning could give him grounds for the assumption even of this ‘general’ idea of substratum. It seems absurd to suggest that we arrive at this general idea by abstraction, since abstraction would have to go so far in such a case as to leave us, so to speak, with no remainder.
As I implied, Locke’s purpose in exploring the concept of essence is partly polemical. He wished to attack the Aristotelian science which had erected itself upon a system of rigid classifications. These classifications seemed to be conceived a priori and without reference to the actual constitution of the objects which fall under them. For Locke, the only significant idea of essence must be one of constitution. The constitution of an object cannot be determined by fiat, but only by exploring the reality of the thing itself. Hence it cannot be determined a priori. Locke therefore introduced the idea of a real essence, to be distinguished from the nominal essence bestowed on an object by the arbitrary classification under which we subsume it.
Consider the classification ‘bachelor’. This defines a nominal essence, which is to say, a set of properties which we consider to be the qualifying attributes of the class of bachelors. The classification is arbitrary; we could have defined the word differently. But in so far as it exists it enables us to speak of a certain ‘essence’. We can say, for example, that it is an essential feature of a bachelor that he is unmarried, meaning that, qua bachelor, he is of necessity unmarried. But it is not an essential feature of John, who is a bachelor, that he is unmarried: on the contrary, he might choose to marry tomorrow, in which case, in ceasing to be unmarried, he ceases also to be a bachelor. Nominal essences are therefore accidents of classification; they reflect constraints embedded in our language, but these constraints do not operate on the things themselves. They hold, as the medieval logician would have put it, not de re but de dicto. Locke thought that it is only nominal essences that could be known a priori, and this is only because such knowledge would be the empty reflection of our own linguistic habits, not knowledge of the things themselves.
Now consider the classification ‘gold’. This is associated, according to Locke, with a nominal essence—gold is a yellow, metallic substance, etc. But gold has a real essence as well, in respect of which it could not change without ceasing to be the kind of stuff that it is. This real essence is not (unless by some extraordinary accident) given by the nominal essence. It has to be discovered by scientific investigation. The nominal essence guides us in that investigation only to be overthrown by it. As a matter of fact, Locke was inclined to think that real essences are unknowable. This was partly because he thought that the underlying reality of material substances must remain hidden from observation. Since his day we have found reason to reject that belief. We might come to the conclusion that what really matters to something’s being gold is, for example, its atomic weight, and not those properties in which we first based our classification. Hence empirical enquiry can decide the real essence of gold: the matter, however, could never be settled by convention.
In the case of modes, and of simple ideas (in other words in the case of the ideas corresponding to qualities), real and nominal essence cannot be distinguished. It is only in the case of substances that the distinction can be made. But as the example indicates, there are definite ‘kind’ terms—such as ‘gold’—which admit of the distinction. Do they therefore denote substances? Surely not—at least, not in the sense Locke intended. Gold is not an individual thing, but a stuff. In other words, it is a substance in the more familiar, common sense of the term. And now we begin to see, what neither Locke nor the rationalists were equipped to see, that real essences belong not only to individuals but also to kinds.
Personal identity
Locke’s explorations of the concept of essence did not provide a satisfactory account of the nature of individual substances. He came to realise that the concept of identity must play an important part in distinguishing individuals from kinds. He made suggestions as to the deep intrinsic connection between the individuation of a thing and its location in space and time; but his most important contribution in this area was to raise the problem of personal identity in its modern form. Locke argued that to be a human being is one thing, to be a person is another. Human beings can endure where a person ceases, and perhaps vice versa. A human being is an organism, whose identity is determined by the continuity of that organism in accordance with the real essence which it possesses. But the organism is not identical with the person; men can suffer radical changes of personality; or we can imagine a personality that, after enduring in one organism, suddenly disappears to reappear simultaneously and intact in some other, erstwhile sleeping body. Many thought-experiments can be performed which will point to the conclusion that identity of man and identity of person are separate ideas. In which case, in what does the identity of a person consist?
Locke proposed a criterion of identity, sometimes described as ‘the continuity of consciousness’. So far as my memories link me to the past and my desires and intentions project me into the future, so far am I the same person over time. Thomas Reid famously objected that such a criterion could deliver two conflicting answers to the question of identity. The old general may remember the young officer, who remembers the boy who stole the apples, even though the boy has been forgotten by the general. So the general both is and is not identical with the boy. But the objection is not lethal and suggests merely that we should amend Locke’s approach. We should define personal identity in terms of a chain of interlocking memories, linking the general to all his previous activities: the old man remembers the middle-aged man who remembers the youth who remembers the child. If the chain is unbroken, then perhaps identity is secure.
More serious is the objection made by Bishop Butler. Suppose I have the thought of standing in this room once before. What makes this thought into a memory? Surely, the fact that I identify myself as standing in this room. But how do I know that this identification is correct? I must have grounds for judging that it was once I who stood in this room. False memory claims are no grounds for identity; true memory claims (‘genuine’ memories) are grounds for identity, but only because their truth depends upon the truth of a claim about identity. The criterion, in short, is circular.
Butler’s objection is still much discussed. Locke’s criterion may have an appearance of circularity: but perhaps the circle is not vicious. It is vicious only if it presupposes what it sets out to prove; and it is by no means obvious that this is so.
The concept of cause
Throughout Locke’s lifetime the scientific revolution had proceeded unabated. The Royal Society had been founded, and Boyle (1627-1691) had written widely and sceptically of the traditional science, in a way that engaged directly with contemporary philosophical issues. Boyle followed Bacon in rejecting all research into final causes as irrelevant to science; but he was reluctant, in his search for the particular causes of observable phenomena, to take too much guidance from Descartes’ a priori method, which assumed that fundamental principles could be derived from metaphysics alone. In particular, Boyle rejected the very metaphysical-seeming law that Descartes had put at the heart of his physics: the law of the conservation of motion. This law was to be revived in a new form by Newton, and, when the Principia was finally published almost simultaneously with Locke’s Essay, philosophers were confronted with an extraordinary synthesis of a priori speculation and empirical method, in which
seemingly irrebuttable laws were held forth as governing and explaining the whole chaotic world of transient phenomena. It was not until Kant that the philosophical significance of Newton’s theories was finally encompassed. Meanwhile Leibniz vigorously combated Newton’s absolute view of space, while the empiricists occupied themselves with understanding the deep and difficult concept of causality upon which Newtonian physics had been erected.
Locke had already recognised that, in accordance with his principles, it must be possible to give an account of the experience from which the idea of causality derives its content. He had no difficulty in resolving this problem to his satisfaction. The exercise of will presents us, he thought, with an experience of causality which is immediate, indubitable and irreducible to anything more basic. In a sense Berkeley followed Locke in this doctrine: that is to say, he thought that in so far as we have an idea of true causality, it can only be one of will, the exercise of which is experienced by us both as an activity and as something suffered. When we observe nature, however, we are confronted by the regular succession of events, but not by any experience of volition. To say that there is a will to attract that draws masses together is to speak in a way that is misleading and unwarranted, since all we can observe is the confluence of masses. If we refer to a law of nature here, then that law is nothing more than the expression of the regular and seemingly immutable fashion in which this motion occurs. (Berkeley thus attacked Newton for speaking of ‘attraction’ or ‘force’ in his theory of gravity, since these terms imply the presence of something more than is strictly observable.)
A Short History of Modern Philosophy: From Descartes to Wittgenstein, Second Edition Page 11