no previous

I: Introductory

  ― 1 ―
view facsimile

  ― 1 ―

Lecture 1

Hume and His Predecessors—Rationalism and Empiricism—dependent and independent existence

Hume is to be treated in relation to the other modern philosophers, from Descartes to Hegel.note

  • l. The Continental Rationalists:
    • Descartes 1576-1650
    • Spinoza 1632-1677
    • Leibniz 1646-1716
  • 2. The English Empiricists:
    • Locke 1632-1704
    • Berkeley 1685-1753
    • Hume 1711-1776
  • 3. The German Idealistsnote
    • Kant 1724-1804
    • Hegel 1770-1831

  ― 2 ―
First of all there is the question of the description of the first three as rationalists and the next three as empiricists; and the view has been taken that we have here two opposing philosophies each of which nevertheless has something valuable in it and that Kant in his critical philosophy combines or makes use of both rationalist and empiricist trends and thus sets philosophy on its feet. Then it is considered by idealist historians that this development was completed by Hegel.

Now I shall try to argue in this course that no reconciliation is possible between rationalism and empiricism and that the thinking of the so-called English empiricists is predominantly rationalistic. Of course, it is impossible for any thinker to be purely rationalistic. It is impossible to avoid embodying a certain amount of empirical material in any doctrine, because otherwise there would be nothing to talk about. And again it is the case that these British thinkers did assert certain empirical facts in opposition to the main Cartesian theory. That is to say, they do not, like their predecessors, consider that it is necessary to start from principles or rational truths which are somehow held to be superior to fact and from which subordinate truths can somehow be elicited or unfolded. They professed, on the contrary, to start from what was self-evident.

Now this rejection of something ultimate or essential, from which knowledge can be unfolded is certainly an empiricist procedure, but nevertheless Locke, Berkeley and Hume made so many rationalistic assumptions as to take most of the value from their empirical assertions; and the very notion they have of what was

  ― 3 ―
given, or of data, is a rationalistic notion because in rationalist theory we have the fundamental distinction between “that which is in itself” and “that which is through something else”; we have the distinction between the “self-subsistent” and the “dependent” or “relatively existent”.

Now in the first place, even to take the view that there are relatively or dependently existing things is to be rationalistic because this subordinate reality can only be conceived in contrast to a higher reality and both of these are to be contrasted with actual existence, that is, with what we can empirically recognise.

In speaking of actual existents that we know, we can call them, in a certain sense, dependent and, in a certain sense, independent; they are dependent in the sense that they are conditioned, that they exist in situations and are affected by their surroundings and, in general, have a history. And the conception of philosophy as the search for the unconditioned or, as the idealists would have it, for the self-conditioned, is a rationalistic one. That is, it is opposed to the recognition of the complexity and historical character of things; it is opposed to the recognition of things existing in propositional form; and, instead of that, we have the assertion of that which exists in and through itself as contrasted with that which exists in and through something else. And in the former notion we have an example of the rationalistic denial of relations and an illogical assertion of self relations, as in the notion of the “self-conditioned”, the “self-subsistent”, the

  ― 4 ―
“self-conscious”, etc.

In the second place, they are, in a sense, independent in that they are just those actual things and, however they may have been conditioned, do now occur; so that we have also to reject the notion of dependent existence or relative existence as in the use of the term “idea” by Locke and Berkeley for instance, and to assert that whatever a thing depends on, however it may come to be, it must exist absolutely or in an unambiguous sense and have characters of its own, if it is to be a discussable thing at all. And secondly, such notions as that of the “idea” in these theories are rationalistic because they imply some inner nature of a thing, something which makes it what it is, as contrasted with what we find it to be.

Now in the theory of Descartes, and likewise in the theory of Berkeley, we have this notion of something which exists by its own nature or which sustains itself, and it is the assertion of such entities that forms the basis of all ontological arguments.

The fundamental argument of that type is found in Descartes' argument for the existence of a perfect being, namely, that that being exists because it is its nature to exist; and if we said it did not exist, then we should be attributing imperfection to a being that we had admitted to be perfect and so contradicting ourselves.

Now here we have an example of the rationalistic assumption that certain propositions are self-contradictory and accordingly that their opposite establishes itself. But as against that we may bring the argument that if a thing were self-contradictory

  ― 5 ―
then its opposite would be itself and would likewise be self-contradictory. We can say, in fact, that there is no proposition that is self-contradictory, that a proposition can be disproved only by disproving a fact, and correspondingly that there is no proposition which is self-evident or self-proving because that implies that in setting out to prove it we already have it, and so this proposition in its unproved form is supposed to establish itself in its proved form.

And again there is nothing of which we can say that it is its nature to exist, because in talking of it as having a nature, we should already be recognising it as existing; and so to say that, if a perfect being did not exist it would be imperfect, is incorrect because if it did not exist, it would not be anything. And so, as against this argument of Descartes regarding that whose nature it is to exist, we have the contention of Hume that to think of a thing and to think of it as existing are precisely the same, and the contention of Kant that existence is not a predicate.

  ― 2 ―
view facsimile

Similarly, then, to the rationalistic assumption of that whose nature it is to exist, as exemplified in the “perfect being” of Descartes or the “units” of the Pythagoreans or the “One” of the Eleatics, we have in these rationalistic theories such conceptions as that whose nature it is to know or “consciousness”, that whose nature it is to be known or “idea”, that whose nature it is to be given or the “datum” and, in rationalistic ethics, that whose nature it is to be pursued or the “End.”

  ― 6 ―

Lecture 2

Hume and His Predecessors continued—Simple Impressions and Complex States of Affairs—Berkeley

As against the notion of simple entities we have the proposition or complex state of affairs.note We have, accordingly, to point out that the alleged simples of rationalist theory are really complex and that rationalists continually introduce complexity in an indirect way or in an obscure fashion and particularly that they have, in order to expound their theories, continually to employ propositions and finally that they cannot show what is the relation between the simple and the complex, between the elementary idea and the proposition. That is one fundamental point to notice, namely, that the empiricist maintains against the rationalist that nothing is pure and nothing is simple, that things exist as, and in, complex situations. And incidentally, or as part of their polemic, they have to oppose the doctrine of whole natures or essences whether the reference be to that whose nature it is to think or to be thought or anything else of that kind. They have to show that even those who uphold the doctrine of relative existence are unwittingly upholding, at the same time, a doctrine of independent or actual existence.

That applies to the whole theory of ideas, and we can see its application in the case of the theory of Berkeley who maintains that it is the nature of an idea to be known but who, in doing so, has to maintain, or he at least implies, that something actually is known and that something actually has that nature. He implies, then, that these things are being regarded as actual or independent facts and not as relative facts, whatever that might mean.

  ― 7 ―
That is, even in saying that there is something which is dependent on our knowing it or on someone's knowing it, Berkeley is asserting that independent fact, that is, that fact which is not dependent on our knowing it or on anyone else's knowing it.

In asserting these propositions, then, Berkeley is going beyond his theory of ideas and there is no longer any ground for admitting that the objects of thought first considered are dependent or relative in a sense in which the later facts considered are not; in other words, if something independent of us can be known to us, then “being known to us” is not a sign of dependence, and thus we are led to think of “being known” as a relation between independent things and in general to adopt the theory of relations between independent things and occurring throughout the whole range of existence.

A secondary way of putting the matter is to say that if ideas are in the mind, then that is where they are and their being in the mind must be recognised to be an absolute fact. And, of course, there is the further point as to the real sense in which the term “dependent” can be used. To say that a thing is dependent on something else is to say that it would not have existed but for that other thing, or that, apart from that condition, it would not have had some of the characters that it has. But to say that that is the case is to say that the thing definitely does exist and does have these characters, no matter what its conditions have been. In other words, even to say that one thing conditions another or is dependent on another is to imply that, in a certain sense, things

  ― 8 ―
are independent, that is, that we have two related things, each actually existing and each having certain characters of its own.

If in view of the propositions we put forward it is impossible to uphold the doctrine of relative existence and similarly of unconditional existence, since the proposition itself implies interrelation, then Berkeley's theory cannot be saved by introducing any other condition on which things are said to depend. For example, if he admits that certain things which he calls laws of nature do not depend on our minds and yet can be known by us, it does not avail him to say that these things depend upon an infinite mind, the mind of God, and have their existence in his knowing them. The question is of the logical possibility of relative existence and not of dependence on this or that condition. And, of course, if we are to say that the laws of nature are known by God, we still imply two things, God and a law of nature, between which this relation has to hold; or again when we say that God knows them, that is, these laws, we imply that there are these laws to be known.

Now similar difficulties come out in the detailed development of Berkeley's theory. He has to account for the admitted fact of error, the fact that we make mistakes about things, that we wrongly anticipate what is going to occur, for example, even although in this respect he takes up the same position as Descartes that there are certain things about which we cannot be mistaken, namely, about ideas, because what we know about ideas is just what we have in our minds and what we have in our minds are just the ideas themselves.

Now Berkeley is no more successful than Descartes or

  ― 9 ―
than any other thinker who makes similar assumptions, as we see in the Theaetetus for example, of finding the knowledge which is bound to be correct and finding the cognition that can err. In fact he is no more successful than other thinkers are in showing that there is a kind of knowledge that cannot be mistaken because, and this is a point made by Kemp Smith, assuming that we have once got those ideas into our minds, there is still the problem of how we are to get to know them; and if we have difficulty in knowing so-called external things because they are external or because we are confronted by them, then we should have equal difficulty in knowing the ideas which confront us in our minds.

Of course, this problem is passed over because it is assumed, as by Berkeley, that the being of these ideas and their being known by us are the same thing. But this would at the best imply that those ideas are elementary and unhistorical entities, it would conflict with the fact that they are to be found not simply by themselves but in situations and it would conflict with the fact, that even as Berkeley admits, we can discover relations between one idea and another. Thus the only solution is to treat the idea as something that we can know, although we do not

  ― 3 ―
view facsimile

know all that can be said about it, to treat it as occurring in historical situations and thus to treat it as an independent thing.

It is a matter of no moment, then, whether we call it an external thing or not so long as we understand that it is something that we can know and something that is independent of our knowing it. Whether, then, it is something outside us or something inside us is not a point of logical importance so long as it is

  ― 10 ―
admitted that we may know it in either case and know it in the same way, that is, that in any case, knowledge is a relation between two things, a knower and a known.

Now Berkeley, as was suggested, gets into special difficulties because he puts forward propositions about his ideas, and implies not merely various relations between one idea and another, relations that can be discovered after the ideas are known, but also various qualities of the ideas, so that ideas are recognised as complex. But in either case he states propositions regarding the ideas, and these propositions are to be taken as simple facts, as when he says of all ideas that they are known to us, and of some ideas, those which he calls sensations, that they are not under our control; or of the others, which he calls images, that they are under our control, and again, that some sensations are signs of others, and that sensations are more vivid and orderly than images. All these assertions imply that ideas are historical things, and are not elements out of which situations could be built up.

Now precisely the same difficulties appear in the doctrine of Hume when he makes his distinction between Impressions and Ideas, corresponding to Berkeley's distinction between sensations and images, on the basis of vividness. It is implied, if this distinction is to be made, that certain things are vivid, and that certain other things are not vivid; that is, it is implied that these are facts, just as it is implied, in order that Descartes may use his criterion of truth, namely, clearness and distinctness, that certain things are clear and distinct, and that certain other things are obscure and confused; and accordingly we can give no

  ― 11 ―
reason for preferring the clear or for saying that there is any more truth in it than in the other facts referred to, namely, the facts of obscurity and confusion.

Lecture 3

Hume and His Predecessors continued—Knowledge of Ideas and Knowledge of Things—Locke on substance

Now one of the most typically rationalist theories of Locke is his theory of Ideas, that is, elements or units of knowledge out of which science or any coherent body of knowledge has to be built up.note We also, of course, have rationalism in his theory of substance as that which supports qualities, this being an underlying something which is supposed to be known not in itself but by the function which it performs. But if we do not know it in itself, then we cannot know that that sort of thing would perform that function, and equally, of course, unless we have had experience of a substance supporting qualities, we should put the substance on the same level as the qualities. In fact, we should not even know that the qualities required to be supported. And again there is no need to postulate a substance in which various qualities inhere in order that they may cohere with one another, because, apart from our not knowing what inhering means, if the substance does, or if something does make these qualities cohere, then we find their cohering or, as Berkeley puts it, their accompanying one another; and there is no problem of how that is possible, that is, of seeking an agency of coherence, the fact of coherence being already presented to us; and,

  ― 12 ―
of course, that fact is already a sufficient indication of the complexity and not the simplicity of the things confronting us.

Now in regard to Locke's ideas, there is a question whether they are to be considered as objects of knowledge or simply as elements which make knowledge possible because in the fourth book treating of what he calls knowledge, Locke is concerned with the question of the relation between ideas and of a possible correspondence between such a set of ideas and some external fact. But we cannot say that the elements out of which knowledge is built up are themselves unknown because then the relations between them would also be unknown, and indeed, Locke says in the beginning of the first book that he means by idea “whatsoever is the Object of the Understanding when a Man thinks”.note

But if that is the meaning, then what he later treats of as relations between ideas should themselves be called ideas and so should any supposed correspondence or agreement between them and outside things and similarly the outside things and substances themselves if we think about them. And, of course, in so far as we think of or speak about ourselves, we also should be described as ideas.

But allowing for these criticisms, we have the doctrine of simple elements of knowledge, and it has to be pointed out that any building up on the basis of these elements is impossible for the idea, being just what is known, we cannot know any more about it; if the relations between ideas are given, then they are just simple ideas like the others and not factors in a complex situation; if they are not given, we know nothing about them.

The second point is, as Locke's argument soon shows,

  ― 13 ―
that what he calls his ideas, anything he can take as an idea, is not really simple but complex. We have ideas of substance and existence, etc., which go along with any single idea that we receive but this is no solution because existence and the like, no matter how many other ideas they accompany, must also be single ideas, and to say that we have an idea of red and an idea of existence at the same time, even assuming that on Locke's theory that could be said, would be no reason for saying that red exists. The point again is that unless we start with complexes and not with simples, with propositions and not with unitary ideas, we cannot have any knowledge whatever; and there is the further point that even when we say that we are starting with unitary ideas, we are really asserting propositions all the time.

  ― 4 ―
view facsimile

And thirdly, in distinguishing between knowledge of ideas and knowledge of other things, in saying that it is through having ideas that we know natural things, for example, Locke is constantly implying that we do not have this indirect and representative knowledge of matter, even assuming that to be possible, but that we have a direct knowledge of external things. For example, in making his distinction between primary and secondary qualities, and between the different sets of ideas given to us by these qualities (a distinction that he has to make in order to make out that we know anything about the external things at all, but a distinction that must be a forced one because it is a distinction among things we know, which on his view should be ideas, and not between things that we know and things that we do not know), Locke uses division as a criterion; the qualities which really

  ― 14 ―
belong to a thing being those which remain no matter how much the thing may be divided. And, of course, the qualities which he finds in that way to be primary are simply those which are implied in divisibility. But not only can no reason be given for taking division as the criterion but also Locke speaks as if we had a direct knowledge of the breaking up of bodies and not a mere division in ideas or “ideal” division. Thus, however inconsistent it may be with his doctrine of ideas, Locke implies a direct acquaintance with external things, and if he did not do so, if he remained consistent with the doctrine of ideas, then, as Berkeley points out, he could make no distinction between different sets of ideas so as to speak of primary and secondary qualities.

Locke means by “quality” a power of a substance to produce an idea in the mind and by a “primary quality” a power of a substance to produce in our minds ideas of extension, motion and figure. A “secondary quality” is a power in the substance to produce in our minds ideas of colour, sound and the like; and the distinction between primary qualities and secondary qualities is that the primary quality resembles the idea produced in our minds whereas the secondary quality has no resemblance to the idea produced in our mind; and thus Locke says not that colour and sound and the like are secondary qualities of bodies but that they are not qualities of bodies at all but ideas in our mind, brought about there by the action of the bodies according to their primary qualities, that is, according to their extensions, figures and motions of their parts.

Now that means that the bodies have extension, figure

  ― 15 ―
and motion as qualities in the ordinary sense, that is, that the bodies are extended, have shape and do move. But even then Locke does not say that the size, shape and motion of the bodies themselves are exactly the same as the size, shape and motion of our ideas. The position is, then, that bodies having size, shape and motion produce in our minds ideas of shape, size and motion, that is, ideas which are like the bodies themselves, and also ideas of colour, sound and so on, that is, ideas which are not like the bodies themselves. But how, without comparing the bodies and the ideas, thus making the bodies also ideas, we are able to say that the bodies are like ideas in some respects and unlike them in other respects, Locke, as Berkeley points out, cannot show.

Now although he brings out some of the difficulties of Locke's theory, Berkeley falls into the same main difficulties, in particular, the difficulty of accounting for a knowledge of propositions in terms of elementary ideas; and that difficulty is aggravated by Berkeley's theory of notions whereby we are enabled to know the mind or active entity on which passive ideas depend. He is unable to show how we could know that passive ideas depend upon the active mind since the passive ideas are known by way of idea and the active mind by way of notion; and of course he is unable to show how we could know the relation of dependence as one actually connecting the active and the passive or how we could know any alleged relation even among ideas, a relation, for example, of accompanying or of signifying, except as an additional idea.

Hume seizes upon the difficulty as far as notions are concerned, but he has exactly the same difficulty as regards

  ― 16 ―
relations between ideas. He cannot account for these relations in terms of the original ideas and like both Locke and Berkeley he continually implies that we have begun with, and that we have, a direct knowledge not of ideas but of complex states of affairs.

Referring, however, simply to Berkeley's view in the meantime, we can say that he asserts various propositions:

  • Ideas are known by me
  • Some ideas called sensations are not controlled by me
  • Some ideas called images are controlled by me
  • Some ideas are vivid
  • Some groups of ideas are coherent

and so on. All these are propositions; all these propositions, if they are true, must be regarded as independently existing facts, facts not dependent upon my knowing them, and existing in the same sense as I exist. It may be, of course, that all these propositions are known but that is a further proposition and an additional fact; and thus the general position is that in all our theory we are concerned with what exists, and we are concerned with what is complex.

Now Berkeley's theory of simple natures comes out particularly clearly in his distinction between mind and ideas, the active and the passive. The active is so very active that it cannot possibly be passive or be acted upon while the passive is so passive that it cannot possibly act. Here we have the doctrine of active and passive natures and the denial of complexity whereby the same thing can be both active and passive, as is indicated, for example, in Plato's Sophist; but if activity is the whole

  ― 17 ―
nature of anything, then we cannot distinguish one active being from another, we cannot describe it, or regard it as having a history; and in the same way we could not distinguish one passive being or idea from another, and in the end we could not even distinguish the active from the passive since to do so we should require to have an active and passive thing and not pure activity and pure passivity.

Lecture 4

Hume and His Predecessors continued—Berkeley's Theory of Abstraction—causality—other minds

Berkeley's theory of abstraction is based on the contention that we cannot think separately of what cannot exist separately, and thus to know anything at all is to know all about it; and the consequence of that view is that we cannot admit complexity because in order to recognise complexity we have to recognise as independent facts a thing's having one character and its having another character.

Now it is not the case that we cannot think separately of what cannot exist separately. To take the example of man's mortality: we can know men without knowing them to be mortal even though they cannot be without being mortal. Again, we can know a triangle without knowing a great number of the properties

  ― 5 ―
view facsimile

that a triangle does actually possess, properties which, in the course of geometrical investigation, we find them to possess.

Now Berkeley's position has only this as a basis, that when we say a thing is active, we are not thereby, or at the same time saying that it is passive, but nevertheless we may be implying that it is passive. It may be that all things both act

  ― 18 ―
and are acted on; and we see, in considering Berkeley's own admissions, that in order to develop his theory he has to acknowledge that that which is active can also be acted upon, which implies that that which he began by saying was acted upon is not thereby proved to be incapable of acting.

One important point to be made is that if we try to think of the purely active, that whose whole nature it is to act, then we cannot distinguish between different forms of activity, and yet Berkeley does distinguish between the activities of knowing and willing while assuming that these are acts of the same being; and, of course, if they are, they will naturally affect one another or be acted on as well as act.

Starting, then, from the contention that knowing is a form of acting, a contention which, in its propositional form, is inconsistent with Berkeley's views on abstraction, Berkeley maintains that the knower is an agent and thus, not being passive, cannot be acted on by the object which it knows, and cannot even be known in the same way. But he has to admit that it can be known in some way and even if he says that notions are the peculiar means whereby we know active beings he is, in saying so, assigning a passive part to those active beings. And further, he has to admit that minds can be acted on in receiving sensations, in having ideas forced upon them, and consequently he has admitted that activity and passivity may be characteristic of the same thing; and therefore he has failed to prove that the objects of knowledge, which he calls ideas, cannot act upon our minds and cause us to know them.

He admits that we can know what exists independently

  ― 19 ―
of us in having notions of other minds; he admits also that what he calls “laws of nature”, or the regular order of our sensations, are independent of our knowing them, and thus he has no ground for saying that these laws must be kept in existence by some other mind, the infinite mind, which causes us to know them. We may just as well take the common sense view that it is the situations themselves existing independently that affect us, as we affect them.

And here we have an objection to Berkeley's division of ideas into sensations and images, those which are under our control and those which are not, because in volitional acts we do not merely imagine things, but we actually produce changes among those objects which Berkeley calls objects of sensation; and this would imply, if we took the view that these sensations are in God's mind, that we can interfere with God's ideas, that even the presumed infinite mind is not purely active, but is capable of being acted upon. In fact, we cannot think at all of the purely active or introduce it in any way into a connected theory; and thus although we may say that knowing is an activity and being known a passivity, we have no ground for saying that knower and known exist in various ways or are known in different ways or are incapable of both acting and being acted upon.

Now granted that there are these fundamental objections to Berkeley's theory, there are further points to be considered in connection with his doctrine of causality. It is sometimes said that even on his assumptions it would be just as reasonable to say we have a notion of matter as the cause of our ideas as to say that we have a notion of mind as the cause of them. But the point

  ― 20 ―
is that on Berkeley's view, we have no knowledge of matter either as causing or as anything else, and although Locke may say that matter supports extension, figure and motion, we know these only as ideas, as supported by the mind, and there is no other support that we could think of them as having or requiring. But it is Berkeley's contention that we do have an experience of mind causing our ideas as when our own mind causes our ideas of imagination or reproduces ideas. Thus we have knowledge of mind causing ideas, and we have no knowledge of any other cause of ideas, and therefore in the case of ideas that we do not cause, it would be, to use Moore's expression, a reasonable inference to say that it was also a mind which caused these other ideas.

That being Berkeley's position, the question of a notion of matter provides no valid criticism of him; but the real objection to his view has to be taken along with his theory of the recognition of mind as having ideas and in certain cases causing them, the recognition of mind as having some relation to the things it knows, even although on his view mind is something additional to ideas, that is, cannot be known in the same way.

Now that implies that we can never know the fact that mind has ideas, but the assumption that Berkeley makes at the beginning of the Principles is the Cartesian assumption that whenever we have any knowledge we know that we have it; that is, we have all the time a certain implicit knowledge of ourselves in relation to the objects we contemplate, we know ourselves simply by being ourselves and neither can nor require to have ourselves as objects.

  ― 21 ―
Now Hume rightly departs from this view and concerns himself solely with the objects and not with something which is implied by them as having them; but in calling the objects “impressions” and “ideas”, he is still holding to the consequences of the Cartesian doctrine of a peculiar and fundamental self-knowledge; that is, he makes the things we know relative without giving them anything to be relative to, since we could not know it without making it also relative, this being more consistent than Berkeley's view that, although we could not know it as idea without making it relative, we could know it in the notional manner without affecting its independence.

Nevertheless, in starting with ideas Hume is still adhering to the Cartesian doctrine, cogito ergo sum, and this is borne out by the fact that he goes on to talk about us and our minds as if we did have independent knowledge of these things.

A subsidiary argument against Berkeley is concerned with our knowledge of other minds. In the first place, we cannot on his view admit the existence of other human minds because all the sensations we get are supposed to be caused by the infinite mind and to be arranged according to the laws of nature that that mind has laid down; and we cannot among those sensations or sensible objects distinguish between our own bodies and other people's bodies because no set of our sensations is more intimately related to our minds than any other set. In the second place, we cannot even know anything about an infinite mind because even if we say this idea is controlled by me and that idea is not controlled by me,

  ― 6 ―
view facsimile

then, apart from the assumption of complexity in the assertion of

  ― 22 ―
these propositions, this idea controlled by me and that idea uncontrolled by me, these are just ideas of mine like any others; and so are the ideas of vividness and coherence; so that we cannot get anything in the realm of ideas that points to some mind outside it. This does not mean, in the third place, that if Berkeley were consistent he would be a solipsist because he cannot even derive a knowledge of his own mind as an underlying agency from the ideas he has. Solipsism is not a coherent or consistent position because we can find, on that doctrine, no relation between mind and its ideas; and if we say there is any relation, we are implying that both mind and ideas have independent existence and thus abandoning the doctrine that there is a single existing thing, namely, oneself.

Lecture 5

Hume and His Predecessors continued—Kinds of Existence

There is the question of (1) a knowledge of other (finite) minds; (2) a knowledge of an infinite mind; (3) a knowledge of our own mind. Now we cannot, consistently with Berkeley's assumptions admit the possibility of an inferential knowledge of other human minds; that is, we cannot infer the existence of another person from any set of ideas that we receive.

The usual argument from analogy, the fact that our own minds are associated with our own bodies and certain peculiar kinds of bodily behaviour, leading us to suppose that similar bodies and similar behaviour that we are acquainted with and that we know not to be our own, are associated with a mind which we have to consider not to be our own mind; that analogical argument will not hold for

  ― 23 ―
Berkeley because there is no set of ideas or passive objects of our perception which we can call ours in any more intimate sense than any of our other ideas. That is, what we call our bodies will be, on Berkeley's view, just so many of our ideas existing in our mind, and what we call other people's bodies will be so many other ideas also existing in our mind. In any case, the analogical argument would be unsound, apart from the formal weakness of all analogy, unless we had been acquainted with an actual association of mind and idea, which implies a direct acquaintance with at least one mind on the same footing with the thing the mind knows. And, of course, if we can be directly acquainted with one mind, there is no reason why we should not be directly acquainted with another.

Alexander contends that the analogical argument must break down because analogy could not give us the very special conception of another mind, and therefore Alexander says that we must be aware of other minds in some special way, a way which he calls “assurance”, this taking place through cooperative action, or through the assistance which one mind can give to another, as contrasted with the resistance which matter offers to our minds; and on the lines of this view, when other people resist us we do not think of them as behaving in a specially mental way but rather in obstructing us, behaving mechanically or irrationally.note

But the defects of this theory of Alexander are, in the first place, that it suggests that the conception of another mind is more difficult than the conception of our own mind, whereas the problem is the same in either case, namely, of knowing some

  ― 24 ―
specific thing with specific characters; and on Alexander's own showing, it is through our acquaintance with other people that we come to have an articulated knowledge of ourselves. And secondly, assurance, as a peculiar way of knowing, cannot be linked up with other ways of knowing, and creates a difficulty in Alexander's theory additional to that created by his distinction between “enjoyment” as a way of knowing our own minds, the mind knowing itself by being itself, and “contemplation” as a way of knowing non-mental things; so that just as in Berkeley's theory of notions and ideas, we have no way of knowing a relation between a mind and a non-mental thing such as that involved when we say the mind contemplates a thing; and with the introduction of assurance, we have no way of knowing that another mind has a certain relation to us so that, for example, we can be “assured” of it, or to any other thing.

It is implied, then, that any knowledge we have of minds must proceed from a direct acquaintance with minds, that otherwise we should have nothing to infer from a knowledge of what was non-mental and that no priority can be given to knowledge of our own minds. As a matter of fact, we find that while in some respects a mind knows itself better than other people do, in other respects it knows itself worse, has illusions about itself that other people do not possess.

But as far as Berkeley is concerned at any rate, there can be no basis for analogy because in either case we simply have so many of our ideas; and if the distinction is made between the

  ― 25 ―
ideas that we produce or reproduce, and the ideas that are forced on us, then we have no means of distinguishing among the latter, between those forced on us from one source and those which have another source. If, instead of God being the source of all our sensations, other minds were the source of some of them, Berkeley's theory of laws of nature and his argument in support of the existence of the infinite mind would break down. We should rather have a number of competing minds all attempting to legislate for us and there would be no set laws of nature. And, of course, we ourselves would also be competing because what we mean by an act of will is not an act which makes us imagine something different but one which makes us sense something different, one which makes an alteration in those objects which we believe to be subject to the laws of nature, and not simply one that capriciously rearranges ideas.

If, then, the order of sensations entitles us to infer the existence of an ordering mind, it cannot at the same time enable us to infer the existence of a subordinate or finite mind, and so there is no argument whatever, on Berkeley's basis, for the existence of other human beings. But secondly, as has been pointed out, there is

  ― 7 ―
view facsimile

no argument for an infinite mind on the basis of the theory of ideas because the supposed ground of distinction between ideas and images consisting of the recognition of what is controlled by me and what is not controlled by me can only be said to consist of additional ideas, ideas which do not take us beyond the mind in which they exist. If, on the other hand, we admit a knowledge not simply of ideas but of propositions such as “X is produced by me”

  ― 26 ―
and “Y is not produced by me”, then we are recognising certain facts, things which simply are so, independently of whether anyone knows them or not, and therefore we are cutting out the inference from other things to mind and the theory of the dependent existence of the non-mental.

And then, in the third place, there is no basis for any inference from ideas to our own minds as having these ideas, there is no basis for the view that the mind is additional to the ideas or things contemplated, because on that assumption we can never know that there is any relation between mind and ideas; and the view that when we know something we are in some way knowing that we know it is simply untenable: to assert that such and such is the case and to assert that I know it is the case are two different things. (That, however, is the Cartesian assumption of Berkeley). If we do know situations of the sort “I am aware of X”, then we must be aware at the same time and in the same way and by the same judgment of that which we call “I”, of that which we call “X” and of the relation of “awareness” between them; and that means that we must take ourselves and X both to have independent existence in order that they may be related.

We know mental facts and we know non-mental facts, and we know relations between the two, and to assert that ideas have dependent existence or are dependent on mind is to assert that dependence as an independent fact, so that Berkeley's theory is really incoherent and we cannot say that to be consistent he would

  ― 27 ―
have to be a solipsist but that he cannot be consistent unless he recognises the independent facts.

Now Lenin, for example, in his Materialism and Empirio-Criticism contends that the position of Berkeley is reducible to solipsism and is to be rejected for that reason by those who believe in the existence of many things.note He implies, then, that we cannot refute Berkeley in his own terms but that the most we can do is to say that he is wrong. But actually this is not the case, because in putting forward a theory at all, Berkeley has to assume the very independence he is denying; in using propositions as he must do, he goes against the whole theory of self-subsistent minds and their ideas. It is not a matter, then, simply of choosing whether we shall be solipsists or not, of simply finding ourselves on one side or the other of the philosophical fence, but the fact is that Berkeley submits himself to the conditions governing demonstration and that under those conditions he can be refuted, his position can be shown to be inconsistent; so Hume is mistaken when he says that Berkeley's arguments admit of no answer and produce no conviction. They do admit of an answer and if this were not so, then the fact that he is unconvinced by the arguments would be of no philosophical importance.

The logical position is that we have to reject, by reference to propositions or to discourse, as Locke would say, any conception of kinds of existence, relative or ultimate, and thus we reject Berkeley's theory of ideas because they are conceived of

  ― 28 ―
as existing relatively to something which has them, this being the basis of the logical refutation of Berkeley (and not the line that is taken by Kemp Smith, for example, in his Prolegomena to an Idealist Theory of Knowledge where he objects to Berkeley's theory of ideas as the sole objects of knowledge by saying that he must always think of an idea as an idea of something, and thus that the assertion of ideas implies the assertion of external things.note This is to revert to the position of Locke which Berkeley has refuted, and that is what is done by most modern critics of whatever school, for example, Kemp Smith, Lenin or Bertrand Russell).

The important point to grasp about the philosophy of Berkeley is just that he has refuted Locke and thus has made any return to a representative theory of knowledge quite out of the line of philosophical development.

What Hume does is to show that, on Berkeley's assumptions, mind also can only have a derivative existence, and that is what it really has in Berkeley's theory of notions, the implication being that there must be a mind because ideas must have something to rest upon or to have them. But this leaves Hume simply with derivative existence and nothing from which it can be derived, that is, with the relative and no absolute, and this in itself is an incoherent theory. When Hume talks about impressions and ideas he is assuming something upon which impressions are to be made, and the only way out of the difficulty is to admit the independent existence of things, whether mental or non-mental, and a direct acquaintance with these different classes of things, and we find in Hume, as in

  ― 29 ―
Berkeley and Locke, that in spite of himself he negatively assumes this independent existence and this direct acquaintance throughout his argument.

no previous