The Subject of Consciousness
© C.O. Evans

Chapter 4

Unprojected Consciousness - 4.4.13
[table of contents]  [previous]  [next]

4. The Logical Dependence of Mental Images and Thoughts on Bodily Activity

[13] By having offered a defence of Ribot's claim that a necessary condition of imaging is the inhibition of appropriate bodily activity, I have supported the thesis that reflection depends on such inhibition in the case of one of its species. It remains to be argued that the thesis is true also of the species of reflection still outstanding - namely, thinking. I mean to refer here to the type of thinking that involves having thoughts.

It would seem difficult to argue - as I did in the case of imaging that thinking cannot occur simultaneously with certain bodily activities, and in consequence that a necessary condition of thinking is the suspension of competing bodily activity. The connection between bodily activity and thinking will have to be argued for in a more involved manner. The course Ribot chose was to adopt an image theory of thinking. This would neatly solve the problem since if we think 'in' images, no thinking could take place without the muscular inhibition already shown to be necessary to the occurrence of images. This course is no longer open to us because the image theory of thinking has since been shown to be fallacious. 102 In fairness to Ribot, however, it should be pointed out that he was far from being unsophisticated about the image theory of thinking. For many philosophers the image theory of thinking is equated with thinking 'in' visual images. This Ribot would have rejected out of hand: he endorsed Galton's work on image thinking, and stressed the fact that many people think exclusively 'in' non-visual images. Some indeed are visual imagists, but some are auditory imagists, and others are kinaesthetic imagists, and so on.

The course I shall take will be somewhat different from Ribot's, but it will lead to the same result: it will show the dependence of our thinking on our bodily activities. I shall not employ an image theory of thinking, but I shall argue that some of our thinking is dependent on imagery, and when it is not it is dependent on perception. Thus in either case it will still be true that thinking depends on bodily activity. I shall argue against any third alternative, and I shall defend my position by arguing against a philosopher

-page 139-

who proposes just such a third alternative. In his article entitled 'Thoughts', 103 Ginnane argues that Ryle had, in The Concept of 11ind, overlooked the existence of thoughts. He means by 'thoughts' such having of thoughts as are described in such standard descriptions as 'it occurred to me', 'it crossed my mind', and 'it dawned on me that . .' The having of a thought is, he argues, episodic. It is meaningful to ask when and where it occurred. Now with all this I am in agreement. It is the following contention that I dispute:

How can we reconcile the fact that thoughts are occurrences with the fact that they do not involve the alteration of any stuff at all, not even shadow-stuff such as mental imagery? The answer is vexing but inevitable: we just have to learn to live with the mystery: thoughts are sui generis. Thoughts just cannot be "explained" by equating them with something else of a more familiar kind something we can get our teeth into - and that is all there is to it. 104

My objection to this view is based on the argument (a) that all the facts mentioned by Ginnane can be adequately explained without its being true that thoughts are sui generis in his sense; and (b) that the admittance of sui generis thoughts into consciousness offends against the Law of Parsimony. It is instructive to compare this passage of Ginnane's with the following one from Price's Thinking and Experience.

In the last two chapters the term "symbol" has been used in a very wide sense, to mean roughly "whatever we think with". It would seem that there is no such thing as pure or naked thinking; or if conceivably there could be, it is beyond the reach of human frailty, even though superhuman intelligences may be capable of it. The human mind, it seems, must always have sensible or quasi-sensible media "in" which we think.' 105

Now it seems to me that this passage is intended to rule out precisely the sort of view put forward by Ginnane. Thoughts which are sui generis are 'naked' thoughts which only a

-page 140-

super-human intelligence might be capable of. Price goes on to identify the media in which we think:

In free thinking we think "in" all sorts of sensible and quasi-sensible particulars, and indeed in principle there is no limit to their variety. We think in words, in images, in gestures or incipient gestures, in pantomimic actions, with models or sketches or other sensible replicas. 106

The symbols 'with' which we think are instantiated either as individual tokens that happen to exist and which we just make use of, or as tokens produced by ourselves. In both cases the awareness of the token may be called an occurrence, and it is this fact, I think, that accounts for the belief that thoughts are occurrences. When Ginnane claims that thoughts are datable as to the time and place of their occurrence, what we really date is the time and place of the awareness of the token by means of which the thought is symbolized; whether it be word, image, or gesture.

It must be admitted that after considering this alternative, Ginnane rejects it as 'patently false'. In fact, however, he does the image theory scant justice, and he would I am sure have taken it far more seriously had he taken into account Price's analysis of the Imagist Theory of Thinking. Ginanne's argument amounts to this: since thoughts are not identical with images, the occurrence of thoughts has nothing to do with the occurrence of images. Thus, referring to the image theory of thinking, he says:

This view, though easily refuted, still has its adherents. It is patently false because there is no self-contradiction involved in someone saying that it occurred to him at a particular time that such-and-such, whilst at the same time steadfastly denying that he had any mental images whatsoever at the time in question. In fact not only could such a claim be made without self-contradiction, it could very often be made quite truly. In any case, no collection of images, however complicated, could ever fully correspond to a thought. No images could, of themselves, amount to, or constitute, the thought that I report when I say, for example, "It occurred to me that Peter might drop in today for a drink" The images are equivocal in a way in which the thought is quite definite. Images can never be anything more than illustrations of my thoughts, just

-page 141-

as pictures in a book can never be more than illustrations of the text.' 107

In this argument the premisses are true, but the conclusion is false. It is a non-sequitur from beginning to end. Price has shown quite conclusively that a visual image is usually very much more than an 'illustration' of a thought. It is one of the more successful symbols of thoughts. The flaw in Ginnane's argument is his failure to deal with the possibility that on those occasions on which he has a thought without having 'any mental images whatsoever at the time', his thought can only occur because it is expressed in some other form of symbol. It cannot be argued that a mental image is never a symbol because it is not always a symbol; but that seems to be the assumption behind Ginnane's conclusion that 'images can never be anything more than illustrations of . . . thoughts'. As I shall show this error vitiates his entire position.

Something first needs to be said about the alternative sorts of token-individuals by means of which thoughts can be symbolized. Undue confusion is caused by an ambiguity in the use of the word 'image'. As Price says,

The Imagists, we have seen, draw a sharp distinction between image thinking and verbal thinking. But is there not a sense in which some verbal thinking is itself image thinking? Certainly we do often think in or with verbal images - visual, or auditory, or kinaesthetic.

But these are not the sort of images with which the Imagist is concerned, and this sort of thinking is not what he means by "Image thinking". 108

But the Image Theorists are wrong to think that non-verbal visual imagery is the only true source of the symbols 'in' which we think. We can just as well use as symbols the auditory images of the sounds of words, or the kinaesthetic images associated with the articulation of words. In fact we need not use images as symbols at all.

To placate the Anti-Image philosophers, we now turn to an important point which the Imagists have completely overlooked.

-page 142-

Although mental images are quasi-instantiative particulars, they are not the only ones. Perfectly good perceptible objects, denizens of the public material world in which the Anti-Imagists feel so much at home, may have this quasi-instantiative function, and may cash our words in absence, or approximate to cashing them, in very much the same way as mental images do. 109

It can now be appreciated that Ginanne's thesis that thoughts are sui generis, is not established by his denial that mental imagery is a vehicle of thought. He should have attempted to refute the view that the vehicle of thought must either be a visual image, or a non-visual image, or a perceptible object, or a gesture. This could only be done if it could be shown that a thought could occur that was not instantiated in any of these vehicles. In other words, the refutation must consist in a denial of all such disjuncts: a denial of just one is ineffective. It cannot be said that Girmane even begins to satisfy this condition.

Ginnane's attempt to refute the view opposing his own is so far from being successful that it actually undermines the very premiss on which is based his claim that thoughts are sui generis. If we accept Price's account, according to which the symbols 'in' which a thought is carried are drawn either from mental images or from perceptible objects, we have an argument for denying that thoughts are sui generis. The occurrence of a thought entails the existence of its symbolic vehicle, and this in turn entails awareness of the token-individual of the symbol. It follows that whenever a thought occurs there is necessarily associated with it a token-individual the awareness of which is itself a datable occurrence. But if we have one such occurrence whenever we have a thought, we have all we need in order to account for the datability of the thought, and we do not need to look for something additional which can be dated when a thought occurs. A fortiori there is no need to postulate an ostensibly purely spiritual element of consciousness as the occurrence in question.

If my argument against Girmane is successful, it supports Ribot's contention that not even in the case of reflection do we find a form of attention that is independent of the condition he has laid down for all attention; namely, the inhibition of movement. The argument has in fact carried us even further. We can now see that the phantom approach can be rejected in the case of the occurrence

-page 143-

of thoughts. Not even our thoughts are 'without definite seat'. On the contrary their seat (origin) will be determined by the species of the token-individual belonging to the symbolic vehicle of the thought. Thus the sentient approach is vindicated even in the case of that mental phenomenon which would appear to be the chief hope of the phantom approach.

My discussion of the different ways in which unprojected consciousness is involved in two types of attention, interrogative attention and executive attention, also enables us to clear up a difficulty presented by Ribot's theory of attention. Ribot is not as clear as he might be about the connection between the cognitive aspect of attention and its basis in muscular activity. He describes attention as a 'predominant intellectual state', but it is by no means obvious how it can be at once an intellectual state and a muscular activity. The answer is suggested by the analysis I have offered of the two ways in which unprojected consciousness may relate to an object of attention. If the relation is effected through a relevancy system in which a master-idea is guiding the attention bestowed on an object, then the attention is an intellectual state. It is to be noted that even here the master-idea can itself only exist because of control over certain bodily activities (as specified by the sentient approach). The attention so described is interrogative attention. If, on the other hand, the relation is effected through a relevancy system in which kinaesthetic sensation is guiding the attention bestowed on an object, then attention is muscular activity. The attention so described corresponds very roughly with executive attention.

The discussion of the last three chapters forms the groundwork for the direct enquiry into the nature of the experiential self which follows. The theory of the self I shall advance is an interpretation of the conclusions already reached, and is based on the distinctions for which I have argued. No substantive propositions of major significance are introduced for the first time in the chapters that follow. From this point of view we have reached a natural turning point in the enquiry, and I now proceed to consider the implications of what has gone before, for our understanding of the self.

[table of contents]  [previous]  [next]


102. See below, p. 140.
back to text

103. W.J. Ginnane, 'Thoughts', Mind, LXIX, 275 (1960).
back to text

104. Ibid. , p. 388.
back to text

105. H.H. Price, Thinking and Experience (London, 1953), pp. 237-8.
back to text

106. Loc. cit..
back to text

107. Ginnane, op. cit. , p. 387.
back to text

108. Op. cit.., pp. 243-4.
back to text

109. Ibid,, p. 256.
back to text