II: Hume's Treatise and the Theory of Ideas

Lecture 6

Impressions and Ideas

In Part I, Section I of the Treatise Hume is discussing the origin of our ideas. He speaks of the “perceptions of the human mind” and says that they are divisible into two kinds: “impressions” and “ideas”, and he says that:

“The difference betwixt these consists in the degrees of force and liveliness with which they strike upon the mind, and make their way into our thought or consciousness. Those perceptions which enter with most force and violence, we may name impressions; and under this name I comprehend all our sensations, passions and emotions, as they make their first appearance in the soul. By ideas, I mean the faint images of these in thinking and reasoning; such as, for instance, are all

  ― 8 ―
view facsimile

the perceptions excited by the present discourse, excepting only those which arise from the sight and touch, and excepting the immediate pleasure or uneasiness it may occasion.” (T., p. 1.)

We see that this distinction is similar to Berkeley's distinction between sensations and images, the latter being reproductions or reflections of the former, though Hume does not go on, as Berkeley does, to distinguish between the sources of these occurrences as to whether they are occasioned by our own mind, or by an outside mind, because Hume does not think that we are in a position to give any account of mind and its operations apart from the perceptions it has. Nevertheless, this very passage we have considered implies a distinction between a mind and something which enters it, and if we

  ― 30 ―
are to make that distinction, we must know both of these things and we must know them both as actual or independent existences.

Hume recognises that we cannot, on the basis of the doctrine of ideas, that is, of the relative existence of what is known, really account for knowledge of what has ideas, but nevertheless he still retains the Cartesian conception of mind knowing things as dependent on itself, of a certain self-knowledge or a certain characteristic of mentality which goes along with all the knowledge we can have. Hume does not see, in fact, that the doctrine of ideas can be maintained neither along with nor apart from a theory of the knowledge of mind, the doctrine of relative existence breaking down however we may attempt to take it.

That is one point, that even in laying down this theory of the elements of knowledge, Hume implies a great deal of prior knowledge which can only be knowledge of independent existence and, in particular, of actual minds interacting with other actual things. If we do not have this knowledge, then terms like “perception” and “impression” are meaningless.

And the other main point is that there is no question of individual impressions, of units either of knowledge or of existence but only of complexity or interaction and that even in talking of his impressions and ideas, Hume implies that it is a question of the existence of situations and of the knowledge of situations, any such knowledge being also a situation.

We can criticise Hume, just as we can criticise Berkeley, on the ground that to say that some perceptions are vivid or forceful, and that others, which may be more or less

  ― 31 ―
similar to the former, are weak and faint, implies the complexity of these so-called perceptions, and indeed, implies that we are concerned not with individual ideas but with propositions; and we can see more and more clearly, as the argument advances, the failure of Hume to maintain any doctrine of simplicity, and his implicit admission of the fact that the question is of independent things which can be known.

We find also that Hume distinguishes impressions from ideas not merely on the ground of their force but also in terms of the order of their appearance in mind. Now certainly the distinction should be made on one or other of these grounds, but not on both. But actually we find, as the argument proceeds, that Hume is unable to maintain either distinction. He has to admit at the very beginning, that the impressions may be faint, and that the ideas may be vivid, as, for example, they are vivid in certain dreams and hallucinations. But if he says that, then he is admitting that vividness is not the basis of distinction, that he can know in some other way that a perception, even though it is faint, is an impression; or a perception, even though vivid, is an idea—implying some prior and unstated theory of the distinction between impressions and ideas, the point being that the explicit statement of that theory would require the abandonment of the whole doctrine of ideas.

The real distinction is one between what we know of a thing when it is in front of us when we are observing it, and what we know of a thing when, as we put it, it isn't there; for example, something that happened yesterday or something that is outside the

  ― 32 ―
range of our observation at the moment. To admit that that was the distinction would be to admit that the question was of the existence of independent things and of the direct knowledge of that existence and the main point of the criticism of this part of Hume's theory is to establish that there is no other way in which the necessary distinction can be made, just as in the case of Locke we saw that the only way in which he could make his distinction between ideas of primary qualities and ideas of secondary qualities was by assuming a direct knowledge of external things and of a process of division being carried out on these things.

We find, then, that Hume is quite inconsistent in saying that, while in a few instances an idea may be so vivid as to be mistaken for an impression and vice versa, the two are in general so very different that the division must be made. If the division can be made, as Hume makes it in spite of these exceptional cases, it is because the real basis of division is not the criterion of vividness.

We find, again, that Hume later admits, as regards the order of appearance of impressions and ideas, that we can have an idea, even one of his own simple ideas so-called, of which we have had no previous impression. He takes the case of a person who has got to know, through his impressions, every shade of blue except one. Then it is supposed that all the shades he is acquainted with are placed in order, from the deepest to the lightest, and Hume says that this person will not merely perceive a blank in the series but be able, from his own imagination, to supply the deficiency (T., p. 6). Now it might be said that Hume could quite consistently have

  ― 33 ―
refrained from making this admission, that he only weakens his theory by doing so and that he has no reason for doing it. Now that may be so, but whether Hume makes this admission or not he can discuss the relations between impressions and ideas only by implying the possibility of knowing and manipulating actual things.

Again, even although he makes the distinction between simple and complex ideas, he would have to admit that we have

  ― 9 ―
view facsimile

hallucinations, etc., just as much about those ideas he calls simple as about those he admits to be complex. We can, in any event, think something that is not the case, and it is this possibility that leads theorists like Locke, Berkeley and Hume to make their distinction between original and reproduced perceptions and also to make the distinction between simple and complex perceptions. But at the same time, so long as the theory of ideas is retained, it is quite impossible to show how we can think what is not the case. In fact, it is impossible to show how we can think at all; and the very reference to what is or is not the case implies something beyond the range of the doctrine of ideas. It is, in any case, quite impossible to solve the problem of error, of cognition in which we can be mistaken, by reference to a supposed kind of cognition of simple ideas, of first principles or of any other rational entity in which we cannot be mistaken.

While noting this, we should equally note the constant implication in all these discussions of a knowledge of actual fact; and even if as regards the distinction between impressions and ideas, an attempt was made to retain the criterion of order of

  ― 34 ―
appearance, if, for example, Hume had not made the admission about shades of blue, we are still referring here to definite events which can be definitely dated, and such that it can be said quite unambiguously that one takes place earlier than another. Without doing so, we cannot here distinguish between the idea we have, and the event we could call the entry of that idea into our mind, because we could only know that entry also as an idea or set of ideas; and, of course, we imply in speaking of this relation of “entering into”, both the knowledge and the independent existence of the two terms related.

Lecture 7

Impressions and Ideas continued

The distinction between impressions and ideas which we first made in terms of vividness is later given a temporal basis, an impression being a perception as it makes its first appearance in our minds, and an idea being the perception on its later appearances. This, of course, is similar to Berkeley's distinction between sensations forced upon our minds and images, that is, reflections of the sensations, reproduced by us.note

Now Hume, in avoiding any reference to mode of production by an external or internal cause, no doubt wishes to avoid going beyond the sphere of ideas themselves so as to consider agencies operating on them, the point being, as we noted in relation to Berkeley, that the content of the proposition “This idea is forced on my mind” must be regarded as an ideal content, as made up of a number of ideas, and consequently can give no real description of the production of ideas by something that is not an idea. But Hume himself, in making his temporal distinction, is falling into a similar difficulty; he is laying down the proposition: “This perception is later than that” which if true at all, must be taken as an independent fact and not as a cluster of ideas, and which is, incidentally, inconsistent with Hume's own account in Part II of his book of the ideas of space and time.

  ― 36 ―
Moreover, it does not permit of the distinctions that we actually do make, such as that between remembering a thing that we have previously seen and seeing the same thing again. In the latter case, the reappearance of the perception in the mind would not entitle us to call it an idea as distinct from an impression; and if it is argued that in such a case we really have a different perception, that it is a matter of two distinct impressions, then the answer is that it is equally so with what Hume calls an impression and an idea; they are two distinct perceptions and the fact that, assuming this to be possible on the doctrine of simple ideas, one has characteristics which the other does not have or that they exist at different times, does not justify us in saying that they are different kinds of things, that they can be distinguished as impressions and ideas. But the fact that we can contribute any characteristic at all to them and an actual time of occurrence, shows that it is a question of propositions which are found to be the case. That is, that we are working not with ideas but with independent things. In fact, Hume's distinction depends upon an unavowed reference to external things and to ourselves as being sometimes in their presence and sometimes not in their presence, although we can also think of them in the latter case.

Now a similar point applied to Berkeley's theory of significance. One idea is supposed to portend another, to announce the second one's appearance, but until the second idea has been forced upon our minds all that we have is an image, and the sensation when it does appear, not being the same idea as the image which we

  ― 37 ―
had, cannot be said to verify our expectations. The possibility of our having expectations, then, depends not in the least on a distinction between sensations and images but on the possibility of our having different physical relations at different times to the same things and of our knowing these things under either relation.

Now Hume also has to admit that we have expectations, that we can think of things which we have not yet perceived,

  ― 10 ―
view facsimile

and he endeavours to explain this by making his distinction between simple and complex impressions and simple and complex ideas. We must always, he contends, apart from that peculiar case of the shades of blue already considered, have had a simple impression before we have the corresponding simple idea, but we can have a complex idea of which we have had no complex impression.

It is to be noted, of course, that the very reference to complexity brings in the proposition which cannot be constructed out of separate terms, that is, that we do not simply have Idea A and Idea B, although even then the relation “and” would have to be recognised as indicating an actual situation. What we have is “A being B” or the same thing being both A and B; and this cannot be accounted for in terms of simple ideas even if we say with Locke that the idea of existence accompanies all other ideas, because that would simply give us the ideas A, B and C and not any complexity; the position here being the same as in other rationalist theories such as the Pythagorean, namely, that if we start with so many simple

  ― 38 ―
units, then we end with those simple units and never reach complexity. If, then, there can be no such distinction as Hume draws, no simple units of knowledge, then the fact of our having expectations or suppositions shows that we do not need to have had what is called an impression in order to know anything; we are aware of a complex fact which we never observed before and anything that we did observe before is a complex fact of a similar character.

And when Hume treats of the matter in further detail, we find that he similarly requires a direct knowledge of facts, as when he contends that it is only in respect of our simple impressions and ideas that we can lay down the general rule that ideas and impressions resemble one another. Many of our complex ideas, he says, never had impressions that corresponded to them and many of our complex impressions never are exactly copied in ideas.

“I can imagine to myself such a city as the New Jerusalem, whose pavement is gold, and walls are rubies, tho' I never saw any such. I have seen Paris; but shall I affirm I can form such an idea of that city, as will perfectly represent all its streets and houses in their real and just proportions?” (T., p. 3.)

Now the only reason why Hume would not assert this is that he has had experience of going back to an actual place and finding it different from what he had remembered it as; and when he speaks of having an idea of Paris he can only mean that he is thinking of that actual city which he has visited and may revisit and is considering the possibility of attributing to it characters which it does not possess.

There is implied here, then, a direct knowledge of

  ― 39 ―
actual things, and it may be noted that this is implied in any theory of resembling or copying, that is, in any representational theory. All that we can mean, when we say that an idea we have resembles the thing of which it is the idea, is that we have before our minds certain actual characteristics of the thing, and so with the correspondence theory of truth. When it is said that our ideas correspond or may correspond to reality, what can be properly meant is not that there are two similar things, the idea and the reality, but that what we think is actually the case. The mere fact that an idea we had resembled a reality would have nothing to do with knowledge of that reality; the question, on the contrary, would be of knowledge of that idea, knowledge of its real characters, knowledge of it, in fact, as a real thing, which therefore did not require to be contrasted with any reality which it could be said to be of.

Lecture 8

Impressions and Ideas continued—Theories of truth—truth as correspondence—truth as coherence—states of affairs and the definite issue

We have, as leading doctrines of truth, the correspondence theory and the coherence theory; and one main objection to these theories, similar to that which holds against the theory of ideas, is that in its very statement it implies a different theory.


If we take correspondence, then, apart from the question of how we can know it, there is the assertion that “Idea A corresponds to reality B”. Now if that is so then it is true that “Idea A corresponds to reality B”,

  ― 40 ―
but when we say that that is so, we do not mean that all that corresponds to a reality which we can call C; in other words, we do not mean that the truth of that fact of correspondence lies in its correspondence with a further fact, but we mean that there just is that correspondence, that that correspondence takes place. The assertion, then, that there is that correspondence, is not an assertion of the correspondence of that fact to a further fact, and its truth does not depend on that supposed further fact, and so consistently we ought not to say in the first place that the truth of the idea, assuming it to have truth, lies in its correspondence with a certain reality; and this, of course, is connected with the general criticism of the theory of representative knowledge, namely that our knowledge of the representations is supposed to be direct, and we make assertions as to what holds of these representations, assertions which are not to be understood in a representative sense.

Of course, if we adhere strictly to a doctrine of simple ideas, we could not make any assertions at all; but actually we assert that an idea is vivid or has some other character, that one idea follows another or has some other relation to it, and all this we assert to be true in the realm of ideas. But in making these assertions, we are asserting that something or other is the case, or is an independent fact, and so we are not in the realm of ideas in the sense required by the theory of ideas, the realm, namely, of relative existence.

All this shows, then, that a theory of truth other than the correspondence theory is required to make the doctrine intelligible;

  ― 11 ―
view facsimile

and, of course, once we have recognised that what we

  ― 41 ―
assert are independent facts or what we take to be independent facts, there is no question of any distinction of realms or any distinction between ideas and real things, but whatever we know, unless we are mistaken, are real things. And, of course, we could not know any real things which we could get at merely in a representative fashion or by correspondence. In fact, as soon as we make the distinction, we imply, in the first place, that we have an acquaintance with the things we are said to know only representatively and, in the second place, if this were not so we could gain no knowledge of the represented things and could not say, in any intelligible sense, that they are represented.

It is, of course, the case that we can get knowledge by inference but only of things with which we can be acquainted, as in the case of inferring the existence of a man's parents, even if we are not acquainted with them in particular, from the existence of the man himself; and this is the case no matter what form the distinction is put in. If it is said that we believe that A is B and that the thing that is capable of existing in reality is A's being B, then if these two are distinct, in the first place we require to be acquainted with both in order to distinguish them and in the second place, we fail to show why the knowledge that A is B should be considered to have anything to do with A's being B, why it should be considered to be indirect knowledge of the latter or to correspond to it. The fact is, then, that there is not such a distinction, and when it is said that something is truly represented to our minds, all that that can really mean is that we know the thing itself.


  ― 42 ―
Now similar considerations apply to coherence. If we say that the truth of a proposition resides in its coherence with other propositions, then we are asserting that that coherence takes place and not that it coheres with other propositions; and, again, even when we are speaking of coherence we mean coherence with other true propositions, that is, with other propositions whose truth has to be understood in a different sense. If we took anything as true only because it cohered with other true propositions, then we could never start taking anything as true. The position, in fact, is similar to that which holds in the matter of proof. If it is said that we should believe a thing only when we have evidence for it, then we should require to have evidence for the evidence and so on indefinitely.

The fact is, then, that we must work with what is evident to us, that is, with what we find to be the case, and when we examine any proposition it can only be by means of other propositions which we have found to be the case; and that means that there is a certain propriety in speaking of coherence as the test of truth even if we do not take it as the meaning of truth, because the test, if we do test a proposition, can be applied only by bringing forward propositions which have not themselves been tested in that way, or if any of them has, we are at least brought back in the end to untested propositions; but that does not mean propositions that we have no right to believe, since, obviously, we have as much right to believe what we observe to be the case as to believe something that is in accordance with what we have observed to be the case; indeed, there is no question of right because

  ― 43 ―
observing something to be the case is believing.

We cannot say that any proposition that we believe requires testing because that would be to require an infinite process, so that we are not entitled to talk about the test of truth. And, of course, that is independent of the objection to taking coherence as the meaning of truth which, as we have seen, requires us to say quite definitely that certain things do cohere with one another and not that that coherence coheres with something else.

We can, of course, say that truths do actually cohere, that is, that no truth can be inconsistent with any other truth. But that, of course, gives us no solution of our problem when we are confronted with two inconsistent propositions or with a group of propositions which cannot all be true; it does not show us which of these we should reject; in fact, that is determined not by the distinction between coherence and incoherence but by insistence on what we find to be the case; that is, we do definitely accept certain propositions and therefore we are bound to reject anything that is inconsistent with them. On the other hand, to say that a proposition is consistent with a number of true propositions is not a reason for accepting it. A false proposition can be consistent with true propositions although, of course, there will be one true proposition which it contradicts; and what is required for the acceptance of a proposition by reference to other propositions is not consistency but implication.

This is connected with the objection to the view that formal logic is the science of mere consistency, a view put forward

  ― 44 ―
by Keynes, for instance.note It is not that in a syllogistic argument the conclusion is consistent with the premises but that it is implied by them. We can, of course, say that the contradictory of the conclusion is inconsistent with the premises, that is, with their being both true, but it is only by means of implication that we discover that inconsistency. Thus even the test that we apply to certain propositions, namely, that of reference to other propositions we believe to be true, is improperly described as a coherence test; it is a test by implication. But whatever test we apply, the fact remains that we can apply it only by means of propositions that we definitely believe to be true.

We find, then, that truth cannot reside in correspondence between the believed proposition and something else we call reality, nor in relations between propositions like that of coherence, but simply in their being the case or taking place; and, of course, with this is connected the fact that propositions are not forms of words. We do use words to convey our beliefs to other people and to invite their acceptance, but what we are putting forward and what we expect to be accepted is not the form of words but the state of affairs that they mean. And if we could not put this state of affairs before someone else and similarly if we could not think of it ourselves then we could not make that distinction between words and what they mean, we could not know what the words do mean. So if I say “The chalk is white”, it is precisely the chalk's being white that I wish you to think of

  ― 45 ―
and it

  ― 12 ―
view facsimile

would be precisely that that you would be thinking of if you said that that is true.

Now as regards the supposed distinction between knowing and believing, what has been said implies that there is always some definite issue; and when we say that we know something, we mean that what we find to be the case is the case, but that does not add anything to our finding it to be the case; and the fact that what we thus are said to know is an informative proposition means that it could be significantly denied, or conceived to be false; and thus we have the formal possibility of error whether what we believe is true or not; so that the only distinction we can have is between true belief and false belief or between knowledge and error.

Actual mistakes are, of course, not accounted for by this formal possibility but only by the wishes and strivings of the believer; and when we speak of believing as something which someone does, we imply the existence of such strivings and these would have be taken into account in the discussion of belief on the psychological side. It is the fact that we have such wishes that accounts for the actual occurrence of error, but, of course, it also accounts for the actual occurrence of knowledge, for our getting hold of the facts; and just as our complex activity accounts for right or wrong, so the complex activity of things, in other words, their occurrence, accounts for the formal possibility of error, for the possibility of being right or wrong about them.

  ― 46 ―

Lecture 9

Impressions and Ideas continued—Inconsistencies in Hume's Position

Hume contends that there is a great resemblance between impressions and ideas in every respect except their force or vivacity. This, of course, implies the assertion of propositions, of the fact that certain things have certain characters, and does not give any reason for thinking that what is less forceful or vivacious exists in any other way than what is more forceful and vivacious; and in order to show that there is a difference, in order to show that there is a class of impressions which are primary and a class of ideas which are secondary or, as he puts it, which are reflections of the impressions, Hume has to introduce a knowledge not merely of ideas but of actual things.

Of course, there is the further point, already noted, that if the term “idea” did have any meaning, it would be an actually or independently existing thing, but the point is that Hume, in supposing these relatively existing entities has still to introduce a reference to existences that do not exist relatively. For example, he speaks of shutting our eyes and then thinking of something that we have had an impression of before and finding that its representation is exact in everything but force; and again he says that:

“That idea of red, which we form in the dark, and that impression which strikes our eyes in sunshine differ only in degree, not in nature.” (T., p. 3.)

Now in saying that we can shut our eyes Hume is referring to actual spatio-temporal operations and not to ideas; and if an attempt is made to explain the matter by saying that “in the dark” and “in the sunshine” merely refer to other perceptions which accompany the given one so that a certain perception will be forceful

  ― 47 ―
with one set of accompaniments and faint with another set, this will not give any ground for breaking up the perception of the two classes.

And, again, we cannot say that two things differ in intensity and not in nature because there must be a real qualitative difference in the two cases. We cannot say that intensity is a characteristic which somehow floats over the other characteristics of a thing and does not enter into the thing itself so that it can be altered without altering the thing itself, because in that case we should have no ground for saying that this intensity was the intensity of that thing.

And similarly, we cannot say, on the theory of ideas, that the two perceptions are perceptions of the same thing, and consequently, we could not call one a reproduction of the other no matter how similar they were; that is, we could not say “this idea is a less vivid perception of that which we previously had in impression” unless there is a thing of which these are said to be idea and impression respectively.

Here, then, Hume implies a knowledge of things which is contrary to his theory of ideas, and the same applies, as we have already noted, to his theory of complex ideas as contrasted with impressions in the case of a complex idea of the city of Paris. In that connection he says that we can have a complex idea for which we never had a complex impression, but this complex idea is made up of simple ideas connected to each of which we have had a

  ― 48 ―
simple impression. But, as we have noticed, the point is either that we have so many simple ideas, in which case the question of complexity does not arise at all and Hume's distinction is unfounded, or complexity means something more than being made up of a number of elements. It means being together: it means interrelation among things, and unless we are going to take relations as additional simple ideas, in which case they fail to relate their terms, we have to admit that the complex relations that we recognise are not reducible to simple elements and therefore that it is false to say that all our ideas are simple ideas or collections of simple ideas.

A simple idea should mean that which, if it is to be known at all, must be known absolutely by itself, and, of course, such ideas could never be linked or connected. And the conclusion is that there could be no simple elements even as contrasted with complex situations because the two could never be brought into any sort of relation.

These are some of the difficulties affecting Hume's theory of impressions and their reproduction in ideas. And the main point Hume has wished to make is that there must be this general resemblance between impressions and ideas; he says that:

“if any one should deny this universal resemblance, I know no way of convincing him, but by desiring him to shew a simple impression, that has not a correspondent idea, or a simple idea, that has not a correspondent impression,” (T., p. 4.)

the point being that in order to refer to a particular case the challenged would have to employ a word with a general meaning, that is, he would have to admit the existence of the idea of the object in question and in order to refer to this object, the

  ― 49 ―
person would have to have had experience of it,

  ― 13 ―
view facsimile

would have to have had an impression. But if we are able to overthrow the distinction between the simple and the complex, if the things that Hume calls simple turn out, on examination, to be really complex, then Hume has himself admitted, in saying that we can have complex ideas which correspond to no complex impression, that there are cases in which this resemblance does not hold, that there is no general rule such as he attempts to set up, that we can imagine things which we have never observed. And, of course, the distinction between impressions and ideas does really imply some such distinction as that between observation and imagination, a distinction, that is, between different ways in which we react to situations or different attitudes that we take up, but not a distinction on the objective side, not a distinction between types of situations to which we can take up an attitude.

The conclusion to which Hume comes, then, a conclusion which is of very great importance for his theory of causality, namely, that all our simple ideas in their first appearance are derived from simple impressions which are correspondent to them and which they exactly represent, must be regarded as unsound. It is unsound because of the false theory of simplicity, because it implies a reference to actual events other than those conveyed by the terms “impressions” and “ideas”, and again, as we have seen, because of this confused notion of representation, of a thing being like another thing in every respect but intensity and of knowledge of a given thing being somehow in an indirect way knowledge of another thing.

  ― 50 ―
Another example which Hume gives of the precedence of ideas by impressions is as follows:

“To give a child an idea of scarlet or orange, of sweet or bitter, I present the objects, or in other words, convey to him these impressions; but proceed not so absurdly, as to endeavour to produce the impressions by exciting the ideas.” (T., p. 5.)

Now, of course, presenting the objects can only mean putting actual things in front of the child, and even although Hume translates this into conveying the impressions, he has still made reference to many things which are not accounted for in terms of the theory of impressions and ideas. He speaks of the child, he speaks of ourselves as taking up a certain relation to the child, and he speaks of this relation of “conveying” whereby the child is led to have certain impressions and afterwards, ideas.

Now none of these things can be accounted for in terms of impressions alone, but all that Hume could say was that presenting and conveying consist of so many impressions, so that there could be no question of the conveying of impressions, that the child consists of so many impressions, so that there is no question of conveying impressions to him and, of course, that we ourselves are not anything distinct from impressions such that we can have them and operate about them. In fact, here as elsewhere, Hume is forced to express himself in a realist manner in attempting to convey an anti-realist theory.

The general position is, then, that anything that we can know consists of propositions, propositions which convey actual existence, and that in order to speak about knowledge in particular

  ― 51 ―
we must make assertions or state propositions concerning the knower and the known and the relation between them; and the fact that Hume is unable to maintain his various distinctions, that he admits vivid ideas and faint impressions and that he admits the possibility, one found in rare instances, of having an idea of that of which we have no impression, these facts show that even Hume himself was unable to make a consistent statement of his position; and, of course, to assert that something holds in general, although it does not hold in every case, that we can continue to use rules to which exceptions have been found, is to take a quite illogical view.

There is, of course, a certain pragmatist character in Hume's position. He wishes to work out a practical way of dealing with certain problems and to show what kind of knowledge we can adhere to in practice even if it has theoretical difficulties. But the fact that there are these inconsistencies shows that fundamentally the theory is unsound and that it does not hold in practice, the point being that even if certain conclusions that Hume draws from the theory of impressions and ideas were sound, they are not made any more so by deriving them from an unsound theory.

In considering the general views of Hume in the Treatise, then, we have to reject the terminology that he constantly employs of ideas and impressions, and we can see in the later part of the argument, as in this earlier part where he is stating his basis, that he still implies the existence of independent things and our knowledge of independent things. This comes out in his distinction between relations of ideas and matters

  ― 52 ―
of fact, and also in his theory of the necessary and of causality, although in the second part of this book he tries to work out a theory of space and time in terms of ideas, in other parts of the Treatise he takes a different and realistic view of space and time.

Lecture 10

Memory and Imagination—Association of Ideas—Substance—Abstraction—Complexity in Space and Time

(Concerning A Treatise of Human Nature, Part I, Section III, Of the ideas of the memory and imagination; Section IV, Of the connexion or association of ideas; Section V, Of relations; Section VI, Of modes and substances; Section VII, Of abstract ideas.)

Hume has to try to show, in terms of his theory of ideas, how it is possible to make the actual distinctions that we do make, how it is possible to develop scientific theory as we do;

  ― 14 ―
view facsimile

and we have seen that he attempts to do so by distinguishing between the simple and the complex and between impressions and ideas, complexity being necessary if we are to have any science, any connected account of things, and the distinction between impressions and ideas being necessary as a type of representationalism which is bound to arise on any doctrine of ideas, particularly because of the necessity of distinguishing between knowledge and error. We have seen that Hume, consistently with the doctrine of ideas and with Berkeley's criticism of Locke, admits that we have no direct knowledge of outward things; but impressions are practically outward things as compared with ideas which are their representatives just as, in Berkeley's theory, the order of sensations is supposed to show us something that is true independent of us; and, of course, as we have seen, Hume, in making the distinction, implies a knowledge of outward things and at the same time is leading up to his theory of belief, that is, his attempt to account for belief solely in terms of ideal existence.

  ― 53 ―
Now in this connection he has to make further distinctions, in particular, this distinction of ideas of memory and ideas of imagination, in order to show how error may arise; but, of course, error cannot be accounted for in terms of ideas even if we do recognise kinds of ideas. Actually Hume fails to show that there are kinds of ideas; as we have noted, he fails to show that ideas and impressions are really different. The distinction between ideas of memory and ideas of imagination is again made in terms of force or vivacity; the former, although they have not the force of impressions, have a greater force than images. Again the distinction is one of degree, a distinction of which we have already seen the difficulties, and one which conflicts with Hume's admission of exceptions, that admission showing that the real basis of distinction is quite different. In fact, we could not make the distinction between memory and imagination without a definite reference to events, and to events which are past, this implying their occurrence in actual time; and, in fact, our knowledge of occurrences in actual time implies a recognition of objective truth and correspondingly of falsity.

When we say that we remember, we imply that we know correctly what took place and thus that we are referring to an actual event in the past; that is, merely to have a peculiar sort of idea would not be enough to constitute memory, and, again, even if we were entitled to call what was remembered an impression, what we should be aware of would be the past occurrence of that impression, so that even if this past occurrence had been something in our minds, it would be quite different from what is in our minds

  ― 54 ―
now when we are aware of it. To say, then, that I am aware now that something happened in the past, or that something happened to me in the past, is not to say that I have a certain kind of idea now. Solely in terms of what is happening now, we could not call the present idea an idea of the memory.

In the case of the imagination, it may be suggested that what we imagine is largely dependent on what we remember, or at least, that our imagination obtains its material from previous experience; that is, apart altogether from the assumption of elementary constituents of knowledge. The main point is that, when we are said to imagine, we imagine not simple, isolated ideas but complex situations, that is, situations in space and time. We may not assign a precise place or date to the objects imagined, yet we cannot imagine them unplaced or out of time; that is, we are not imagining anything non-spatial or non-temporal.

Now the possibility of our imagining is connected with the possibility of our having false beliefs, and it is connected with our powers of anticipation, of our expecting situations to arise in places and at times to which we have not yet had direct access. Now all this is possible if we recognise that knowledge arises in our interaction with things and that, on account of our wishes, of the ways in which we try to affect situations, there is the permanent possibility of error. Our expectations, then, might quite well be connected with what we have experienced in the past, and they may be rearrangements of

  ― 55 ―
materials, but rearrangements in such a way and by reference to such places and times that they cannot yet be shown to be false, that is, not to occur, though they may not turn out to be true.

The question is, then, to specify as exactly as possible, in relation to other places and times, the places and times involved in what we imagine. This, for example, is what the psychoanalyst endeavours to do with the material of dreams which would, in general, be said to be the work of the imagination; and when in the course of analysis the places and times have been more exactly specified, we generally find that what has been imagined has been the reversal of some situation in our past, that is, a falsification of history or a repudiation of memory on account of its unpleasant features, that is, features not in accordance with our wishes.

The fact, then, that space and time are always involved and that more precise placing and dating is always possible shows that it is not a question of classifying the objects of knowledge in the way suggested as memories, images and so forth, but of situations to which we may react in various ways and, in particular, of truth and falsity in regard to these situations, that is, whether they occurred or not. It is only by reference to actual occurrences and our reactions to these occurrences that we can account for knowledge and error and for the various proceedings called remembering, imagining and so forth.

Now in Section IV Hume goes on to discuss the association of ideas. Now the main point here is that this

  ― 56 ―
association implies that the ideas are together in some way; and this state of “being together” is not to be explained in terms of the ideas themselves, that is, of individual, isolated entities. What is implied is that from the beginning we know things not as really distinct from one another but as interrelated, that is, we know situations; and if a thing itself is not a

  ― 15 ―
view facsimile

situation, then it would be impossible to construct situations out of any number of simple things. In order that things may be in situations, may be related to one another in any of the ways Hume mentions, resemblance, contiguity, causality, they must exist in a common medium which we may regard as that of space and time and each must be itself complex; it cannot be a unit.

This means, then, that there is no real distinction between things and situations, any more than there is between terms and propositions. The content of a proposition may be expressed as a term and made the subject of another proposition as when we say, for example, that “The grass is green”, and then that “The greenness of the grass pleases the eye”, here the subject in the second proposition having the same significance as the whole of the first proposition. Of course, although this implies that we cannot separate things from situations or uphold the theory of simple and complex, it does not imply that there is no distinction. There is a difference between a term and a proposition just as there is between a subject and a predicate,

  ― 57 ―
although every term may be either subject or predicate, or between cause and effect, even if all causes are effects and all effects, causes. We can deal with the very same event in two different relations and in one of these we call it a cause and in the other an effect, and so generally we can deal with two different functions of the same thing without implying the existence of two different things.

The main point has been that Hume's theory of association, if the least we can deal with is a complex situation, gives him no way in which he can talk of the association of ideas as if ideas could be separate and then together; and similar considerations apply to his discussion of relations and to his discussion of modes and substances.

In regard to substance, Hume points out that we have no separate and distinct idea of substance or, as he puts it, we have no idea of it except as a collection of particular qualities. But Hume does not explain with what justification he uses the term “quality” which seems to be a relative term and to imply something of which it is the quality. He cannot explain this without indicating that what is in question is a proposition and not a separate idea; he allows himself to make use of the language that implies propositional knowledge without making this perfectly explicit. But even so the term “collection” requires explanation because if there is no separate idea of substance, there is equally no separate idea of a collection, there are simply the particular ideas which are said to be collected, so Hume has really given no

  ― 58 ―
account even of his own use of the term “substance”, he has given no justification for speaking of a collection, he has not shown how a number of ideas can be together and, apart from a reference to space and time, it is impossible to show this.

Again, following Berkeley, Hume rejects abstraction as he is bound to do if he is to retain his theory of simple ideas because from what is simple nothing can be abstracted. But in rejecting substance and in rejecting abstraction, that is, in rejecting Locke's confused attempt to account for subjects, on the one hand, and predicates, on the other, or in general to account for complex situations on the basis of unitary perceptions, Berkeley and Hume do not themselves succeed in giving any account of a proposition. They are reduced to making statements about order among simple ideas, but how this order is to apply to these ideas and whether it is itself a simple idea or not are problems for which Berkeley and Hume can supply no solution. To admit that this order and the various relations specified imply space and time would be to admit the untenability of the theory of simple ideas, a theory which they rationalistically assume to be above criticism.

Accordingly, we find that they treat space and time, in so far as we have knowledge of them, as themselves consisting of so many simple ideas and thus as incapable of accounting for any sort of relation among ideas. Hume goes into the matter in greater detail than Berkeley did and tries to account for our knowledge of space in terms of simple ideas of extension and

  ― 59 ―
similarly with time; and, of course, he falls into confusions in doing so and makes implicit assumptions contrary to his own theory.

Lecture 11

Space and Time

(Concerning A Treatise of Human Nature, Part II, Of the Ideas of Space and Time.)

Now Berkeley takes up the question of distance in connection with his theory of vision where he contends that the notion of distance is not something that we can be given in vision but is just the variation in ideas; visual ideas suggest tactual and muscular ideas, that is, they suggest ideas of muscular sensation, walking, for instance, that would be involved in getting ourselves into a position to have those tactual ideas which we call “touching the object”.

Now, of course, this theory is defective in that it does not in the least show how the connection between these different sets of ideas could be translated into the notion of distance, a distance which is of the same sort as distance in a lateral direction or what we may describe as the plane of vision. And, in the second place, this plane of vision itself implies independent space whether we talk about third dimension or not; it implies spatial continuity which is thought to be disposed of by the succession of muscular and tactual ideas; and this succession implies temporal continuity, in other words, the succession cannot be regarded as the succession of our ideas, and so Berkeley fails to develop a theory of space and time on the basis of individual ideas and their supposed relation.

Now we have a minor example of his inconsistency in

  ― 60 ―
talking about being in or out of a room. This could be explained by saying that the difference is between having or not having a certain group of sensations, although even so, when he says that the table is in his room, he cannot explain that simply by saying that the table occurs along with or after certain other sensations because then we might as well say that the room was in the table; and speaking of “along with” or “together with” already implies a reference to space.

  ― 16 ―
view facsimile

Similarly, when he explains the theory of the rotation of the earth in terms of the way in which various ideas signify others, he has to say that we mean by asserting the rotation of the earth that if we were in a certain position we should have certain sensations. Again the attempt might be made to say that “being in a certain position” means having certain sensations, but all this implies a complicated theory of sensations that we do not actually have, and of relations with which we are simply not acquainted, just as in the case of the alleged third dimension or distance from the eyes.

Incidentally, the plane of vision would have to be a certain distance from the eye, so that it is quite evident, even in Berkeley's own language, that we do not mean, by the revolution of the earth, this interrelation of sensations but an actual movement of a body in relation to another one; and of course this movement involves temporal as well as spatial continuity.

Now Hume begins his discussion of the ideas of space and time by contesting the theory of the infinite divisibility of space and time or our ideas of

  ― 61 ―
space and time,

“'Tis also obvious that, whatever is capable of being divided in infinitum, must consist of an infinite number of parts, and that 'tis therefore impossible to set any bounds to the number of parts, without setting bounds at the same time to the division.” (T., pp. 26–27.)

This was very much what Berkeley had contended in saying that whatever was infinitely divisible must be infinite so that, as he argued, it would have neither shape nor form; and thus the assumptions of physical scientists, of Newton and his followers, involved contradiction.

From the view that the infinitely divisible must have an infinite number of parts and from the further contention that the capacity of mind is limited, Hume contends that any idea we form of space cannot be infinitely divisible; that is, we cannot know space and so time to be infinitely divisible. But even if we accepted the statement about the capacity of the mind, we could not accept the fact that the infinitely divisible must have an infinite number of parts. What we are bound to assert is that no limit can be set to the number of parts; we cannot say that a thing has just so many parts and no more; but it does not follow that we could ever take the parts arrived at, as the result of a process of division, and say that their number is infinite. What is implied is that, however far division is carried, it can still go further; but at any point we like to take and however we go in arriving at that point, the number of parts into which the thing has been divided is finite.

The assumption, in fact, is that a thing has a certain

  ― 62 ―
number of parts and that that number is either infinite, in which case the thing itself will be infinite, or it is finite, in which case we could say just how many parts the thing had. But, of course, this is a quite unfounded assumption; we cannot say that there is a certain number of the parts of a thing; we cannot say that there is anything which is an ultimate part or which is an elementary part; anything is a part of a thing which is contained in that thing and there is no limit to such a number of contents that we can find.

Now Hume considers that any sensible magnitude has a finite number of parts, that it is reducible to so many unit parts and that it is from these units which are indivisible that magnitude is built up. But as against this we can say that either this is building up magnitude out of that which has not magnitude or, if the unit has magnitude, then it must be recognised to be divisible, that is, we must recognise that it has two sides with a certain distance between them, and consequently, that a line could be drawn between those two sides. If that were not possible, then the supposed two sides would coalesce and the thing would have no magnitude, and it would be absurd to say that out of such parts any actual magnitude could be built up.

We see, then, that Hume's theory of the elementary ideas of space, and similarly of time, is just as incoherent as Zeno shows the Pythagorean theory of units to be.

Now actually, in contemplating an extension, we do not require to think of it as divided into parts; we can always, if

  ― 63 ―
necessary, distinguish one part from another, but we still apprehend the extended and continuous; we do not actually or mentally make out divisions in an observed surface, and yet we should immediately recognise that it has divisions and has magnitude. The point is that if we are to be able to know extension or spatial magnitude, we must begin with continuous space, that is, with that which is both continuous and divisible and not with unit parts, and in the same way we must begin with situations and not with unit ideas in developing a knowledge of things in general.

Now in supporting his theory of unit ideas of extension, Hume argues in this fashion:

“Put a spot of ink upon paper, fix your eye upon that spot, and retire to such a distance that at last you lose sight of it; 'tis plain that at the moment before it vanished, the image or impression, was perfectly indivisible. 'Tis not for want of rays of light striking on our eyes, that the minute parts of distant bodies convey not any sensible impression; but because they are remov'd beyond that distance at which their impressions were reduc'd to a minimum and were incapable of any further diminution.” (T., p. 27.)

Hume is supporting Berkeley's doctrine of “minima divisibila”; but it is to be noticed that in order even to state his theory, Hume has to express himself in terms not of impressions but of actual things

  ― 17 ―
view facsimile

and the notions of actual things in space and time. Thus when he speaks of our retiring to a certain distance he is speaking of a process of moving continuously through space; and at the same time he thinks he can have a discontinuous theory of time, a theory of time as composed of indivisible moments, and thus we could have

  ― 64 ―
two successive moments, the last at which we could see the spot and the first at which we could not see it; but even here Hume admits that though we can no longer see the spot, there may still be rays of light coming to the eye; that is, he is referring to physical bodies and movement in space and time.

He proceeds in this way:

“A microscope or telescope, which renders them visible, produces not any new rays of light, but only spreads those which always flowed from them; and by that means both gives parts to impressions, which to the naked eye appear simple and uncompounded, and advances to a minimum what was formerly imperceptible.” (T., p. 28.)

Now here Hume is admitting two things, firstly, that there can be an impression that can appear in different ways at different times whereas in terms of his theory he would be compelled to say that we have an entirely new impression, so that what he is here calling an impression is really a complex thing; and secondly, that there can be something which has a real extent or magnitude although this is less than the minimum perceptible magnitude; and yet he goes on to maintain that the very same argument which proves the minimum impression of ideas of space and time also proves the minimum elements of space and time themselves. But we see in this particular illustration that Hume is contrasting the continuity and infinite divisibility of space and time themselves with the alleged discontinuity and finitude of our elementary ideas of space and time.

It would be necessary to say, then, in combating his

  ― 65 ―
view that if, when we were in a certain position, we were able to see a particular spot of ink, we should have to be able to see it as having magnitude and thus as having parts and that if in another position, no matter how near to the former, we could not see it as having magnitude, then we could not see it at all.

Now all this implies actual positions and connections in space and yet we find Hume arguing at the beginning of Section IV of this part that:

“The capacity of the mind is not infinite; and consequently no idea of extension or duration consists of an infinite number of parts or inferior ideas, but of a finite number, and these simple and indivisible: 'Tis therefore possible for space and time to exist conformable to this idea: And if it be possible, 'tis certain they actually do exist conformable to it, since their infinite divisibility is utterly impossible and contradictory.” (T., p. 39.)

In this pretended argument we find Hume not only assuming what he professes to prove, namely, the difference between space and time and our ideas thereof, a distinction which he proceeds to do away with, arguing that they must be as they are known to be and they must be known as they are in idea, yet while he thus does away with the distinction in the earlier argument where he tried to prove the existence of minimum ideas he could proceed only by assuming the existence of an actual space and time different from these ideas.

Lecture 12

Ideas of Existence and of External Existence

(Concerning A Treatise of Human Nature, Part II, Section VI, “Of the Ideas of Existence, and of External Existence” .)

Owing to the view Hume takes up of space and time as

  ― 66 ―
made up of so many unit ideas, he is unable to think of existence as occupation of space and time. In speaking of the unit spaces and times he has contended that it is impossible to think of these except as occupied by some quality, some concrete thing over and above the spaces and times themselves. But if we press this view to its logical conclusion it involves the admission of complex objects of knowledge not built up out of previously given constituents. In other words, if we took ideas of space and time as simply so many simple ideas among others, then we could not speak of any thing or idea as being in space or time and we could not speak of the extension of anything. The most we could say, although even that, as we have seen, has its difficulties, is that an idea of space accompanies certain other ideas or that an idea of time accompanies another idea, so Hume is inconsistent in saying that we cannot think of unit spaces and times except as accompanied by some quality.

Now if the theory of unit ideas is to be maintained and if we are not to think of existence as occupation of space and time or as occurrence, then the only way we could distinguish between truth and falsity would be by contrasting the existent with the non-existent; but even this cannot be done because we cannot attach the idea of existence as a predicate to some of our ideas and the idea of non-existence as a predicate to others. On the contrary, every idea we have is supposed to be given simply and completely, and so, as Hume says, to think of such an idea and

  ― 67 ―
to think of it as existing are two ways of expressing exactly the same thing. In Hume's own words,

“The idea of existence, then, is the very same with the idea of what we conceive to be existent.” (T., p. 66.)

but if that is so, then the difficulty is to determine how the term “existence” ever came to be used and thought to have a meaning.

The position is similar to that in which Berkeley says that in the case of sensible things their being and their being perceived are the same, in which case the expression “perceived” would seem to add nothing to the meaning and thus to be itself meaningless; so if the idea of yellow and the idea of an existing yellow are precisely the same idea, then the expression “existing” would not seem to convey anything.

In fact, Hume can give no account of the distinction we make between existence and non-existence, a distinction which still has to be accounted for even if we cannot say that there are non-existing things. In order to account for it, we have to pass from the simple to the complex, from a particular simplicity to a universal complexity, from the theory of the isolated idea to the theory of the proposition; and then, in distinguishing true from false propositions, a distinction made possible by the distinction between affirmative and negative propositions, we can see what is meant by existence and non-existence.

If we believe a proposition which as a matter of fact is false, the fact that we believe it means that we think of the

  ― 68 ―
content of the proposition as having existence in precisely the same way as the content of a proposition which we also believe and which happens to be true. But we do make the distinction between truth and falsity and, of course, a false belief is such precisely because we wrongly believe it to be true.

  ― 18 ―
view facsimile

We are capable of disbelieving something we believed before; that is, what we previously believed to be true or to exist we now believe not to be true or not to exist; where we previously thought A was B we now think that A is not B, or that A's being B does not exist, that there is no such actual situation.

The possibility of our recognising this, then, depends upon the complexity of what we know; it depends upon the subject and predicate being distinguished as well as connected in the proposition. In this case of knowledge and error, then, or of true and false belief, as in the case of memory and imagination and of similar distinctions, it is not a matter of having different kinds of ideas or objects of knowledge but of different relations we have to the things we come into contact with, these things in every case being complex. If we had simple ideas, it would be impossible to find in some of them the quality of existence and in others the quality of non-existence, but it would be equally impossible to find existence in all of them.

The same point arises here as in the criticism of

  ― 69 ―
Locke's theory that the idea of existence is one of those which comes along with all other ideas, namely, that if idea A and the idea of existence come together that is no reason for saying that A exists. Of course, if we could say that true and false ideas were different sorts of ideas, we should still be confronted with complexity, with certain propositions of the kind “These ideas are of the kind A”, and “Those ideas are of the kind B”.

The distinction, then, does not depend on the kind of things contemplated but on a distinction between our different attitudes and relations to things and especially on the recognition of the fact that we have anticipations or expectations in which we are sometimes led astray. Even if what we anticipate does not actually occur, still in anticipating it we think of it as occurring, and then if our anticipation is corrected by experience we think of it as not occurring, not in the sense that we would say that among the objects of knowledge there are the two classes, occurrences and non-occurrences, but in the sense that the opposite or contradictory is recognised by us as occurring; and that, of course, is made possible by the distinction between the subject A and the predicate B, so that in anticipating that A will proceed in the manner that we call B, when we come to have A presented to us, we find it not proceeding in that manner. But, of course, in order that that should be possible it must be the very same A and likewise the very same activity B that are in question.

  ― 70 ―
As far as the theory of simple ideas is concerned these facts of making mistakes and of correcting them cannot be accounted for; nor can they be accounted for on the basis of the qualitative character of the terms concerned. What we should have on that view in anticipation would be an idea A and at the same time an idea B. Then subsequently we should have a third idea C which might be very like B or might not be like it at all; but there would be nothing at all to describe as a verification or a falsification of our ideas. To account for our expectations being fulfilled we should have to say that the object of our anticipation and the object later presented to us were exactly the same, namely, the proposition “A is B”; to account for the non-fulfilment of our expectations we should have to say that we expected A to be B and later discovered A not to be B, still thinking of the very same A and the very same kind of activity in the two cases.

The distinction, then, depends on the distinction between the negative and affirmative propositions or on the distinction between A being B and A not being B, that is, on the distinction between the affirmative and negative copula “is” and “is not”; but the distinction could not be made in respect of the terms A and B themselves, though, of course, to recognise this is not to imply that these terms themselves are simple.

It is not, then, that in distinguishing true and false propositions, existence and non-existence, in this way, we think of

  ― 71 ―
the copula, affirmative or negative, as a separate object. Hume is quite correct in thinking that if a separate or isolated idea of existence were required, it would not be forthcoming; but what he cannot explain, on his view, is how there could be such a term as “existence” at all, or how anyone could have thought that he understood what it meant; in fact, that explanation is impossible unless we drop the theory of ideas and take the theory of propositions.

Now on the matter of external existence, Hume definitely falls back on Berkeley's view. He says:

“that 'tis universally allow'd by philosophers, and is besides pretty obvious of itself, that nothing is ever really present with the mind but its perceptions or impressions and ideas, and that external objects become known to us only by those perceptions they occasion.” (T., p. 67.)

Now he goes on:

“since nothing is ever present to the mind but perceptions, and since all ideas are deriv'd from something antecedently present to the mind; it follows that 'tis impossible for us so much as to conceive or form an idea of anything specifically different from ideas and impressions,” (T., p. 67.)

and he concludes that:

“The farthest we can go towards a conception of external objects, when suppos'd specifically different from our perceptions, is to form a relative idea of them without pretending to comprehend the related objects.” (T., p. 68.)

Lecture 13

Ideas of Existence and of External Existence continued

In the first place, we see that in this argument,

  ― 72 ―
Hume begins by referring to the mind, by talking about what is “present with the mind”, although there is nothing in impressions and ideas which enables him to make this reference, and though his previous argument regarding existence implies the rejection of the notion of existence in the mind. That is, if to think of an idea and to think of it as existing are the same thing, then it will still be the same thing to think of it as existing in the mind; that will not really add anything to the idea itself.

In speaking of external objects as occasioning our perceptions, Hume, of course, is only taking the prevailing view, the view of Locke in particular, and what he goes on to say about our only having a relative idea of what occasions our perceptions implies a criticism of Locke's view.

Now if by our perceptions is simply meant what we know or are directly aware of, then Hume is quite correct in saying that we cannot get from that any specific knowledge of something other than those perceptions, something that can be said to occasion them, because if we did have that knowledge, then these known things would be included among our perceptions. But this does not entitle Hume to say that what we know is in the mind, that what is external to the mind must be something quite different from what we know. It has not been proved at all that the greater part of what we know is not external to us, and the very use of the term “external”

  ― 19 ―
view facsimile

would imply a knowledge of two things, one of which is outside the other.

That difficulty always arises when people speak about the external world; that is, it is assumed that we have some special

  ― 73 ―
knowledge of an internal world or internal existence and that we can then contrast that with external existence as an object of actual or possible knowledge. But, of course, unless we have the same kind of knowledge of the two things we cannot know that one is external to the other, and if we do have knowledge of the two things, so that we can say, for example, that a chair is external to our minds, then the position is exactly the same as when we say that a chair is external to a table; that is, if external existence means the existence of things external to a certain person, then we are not entitled to regard it as a special type of existence, one which the person himself would not know, because if the things are external to him, he is external to them. But if external existence means the existence of a thing in such a way that it is external to, or outside, other things, then again that is a characteristic of existence in general and not a special kind of existence; it is simply existence in space.

The question is, then, of what exists independently or absolutely, and that, of course, means anything that exists at all. If a person exists, he exists independently; if chairs and tables exist, they exist independently, and if a person's perceptions exist then they also exist independently, so that even if Hume says that knowledge of external things is relative and problematical, when he admits the existence of us and our perceptions or simply the existence of our perceptions, he is admitting the existence of various independent things. And if, as seems to be implied, when he speaks of our minds, he admits that we do exist, then it is

  ― 74 ―
possible for us to know some of these independent things without making them ours, without making them other than independent of, or distinct from, us.

Now if there were a thing which could be roughly described as a perception of ours, then it is quite true that the question could be raised: “what is the occasion or cause of that perception?”, and it is quite possible that the cause of that perception should be something we are unacquainted with. But, of course, we could not have the idea of causality merely as a relative idea; our knowledge of causality could not be in every case problematical or else we simply should not know what problem was being raised; that is, unless we were acquainted with something causing something else, that is, with the whole situation of effect and cause and the relation between them, we could not raise the problem of what is the cause in another case where we did not happen to know it.note

The points here raised are connected with a difficulty in Hume's theory of causality and one similar to that which applies to Berkeley's theory of signs, and a knowledge of the laws of nature as a means of regulating our conduct. According to Hume, we recognise causal connection only in those cases in which there is a sequence or customary sequence in our ideas.

Now this would imply that if we had been accustomed to finding an idea B succeed an idea A and if in a particular case we did not observe that succession, we should conclude that there

  ― 75 ―
was no regulating connection and give up expecting B when A was given. That would be the only possible conclusion if we were restricted to ideas; in other words, if, in the working out of a scientific theory, we were concerned only with the order in which we perceived things (and incidentally restriction to ideas in this way would make it impossible for us to think that of two events which we know the one we learn about later actually happened earlier, yet we very often do think that,) then what we perceived first would have to be regarded as the cause of what we subsequently perceived.

But returning to the previous point, in the case considered we do not accept the fact that we have not perceived what we had previously regarded as the effect of a given cause as a sufficient reason for thinking that the thing has not taken place; on the contrary, we are quite capable of thinking that the effect has taken place only we have failed to perceive it, and in many cases we do not wait for an effect to take place and yet we go away convinced that it will take place in spite of our not perceiving it.

Now all this goes to show that the assumptions we make in everyday life and in the course of scientific investigation are in contradiction with the view of Hume that we are primarily concerned with our perceptions. It shows that we draw a perfectly definite distinction between what we perceive and what actually exists without its being implied that we cannot perceive what

  ― 76 ―
actually exists. There is no question, then, as far as these assumptions are concerned of any inference to external existence or any relative knowledge of it. On the contrary, it is implied that we recognise the existence of certain things independent of us and that we also recognise the existence of certain relations, relations of perception between us and other things, which implies that we also recognise our own existence. It is only if we recognise all that, that we can possibly say, as we do say, that things have happened although we did not perceive them, and that we can distinguish the order in which we get to know about certain things from the order of the things themselves. But this, of course, means that we are talking about definite events, independent occurrences, having spatial and temporal relations, these occurrences including ourselves.

But the recognition of such facts is implied even in Hume's theory. We have seen that, in spite of his theory of perceptions, he talks about the city of Paris as an independently existing thing and similarly about putting a spot of ink on paper and going back from it. But apart from matters like that that might be regarded as mere inconsistencies, Hume has to admit an actual temporal order among perceptions themselves, and that means that he is treating these perceptions as independent occurrences or, if we like to put it so, as external existences; and this incidentally implies that he cannot consistently treat time, and similarly with space, as made up of a number of individual ideas because these individual ideas and the others they are said to be

  ― 77 ―
connected with, have to be taken as having spatial and temporal relations to one another.

We find, then, that Hume has failed to maintain his theory of space and time or the ideas of space and time; and that is connected with his failure to maintain his theory of perceptions, that is, a theory which does not begin with the recognition of independent existence or of spatio-temporal occurrences. And in that connection also, as we have seen, he fails to maintain the theory of the simple and the elementary as that out of which complexity has to be built up. The alternative

  ― 20 ―
view facsimile

seems to be, then, that theory is possible only if we recognise throughout independent existence or spatio-temporal existence or, if we care to put it so, existence in the propositional form, the assertion of existence being just what we mean when we say that a certain proposition is true.