There is no doubt that causal relations hold between events in the mind and events outside it. Sometimes these relations are fairly simple. The striking of a clock is the cause of our thinking of its striking. In such a case the external thing is linked with the thought ‘of’ it in a fairly direct fashion, and the view here taken is that to be a thought ‘of’ the striking is to be merely a thought caused in this fashion by the striking. A thought of the striking is nothing else and nothing more than a thought caused by it.

But most thoughts are ‘of’ things which are not present and not producing direct effects in the mind. This is so when we read. What is directly affecting the mind is words on paper, but the thoughts aroused are not thoughts ‘of’ the words, but of other things which the words stand for. How, then, can a causal theory of thinking explain the relation between these remote things and the thoughts which are ‘of’ them? To answer this we must look at the way in which we learn what words stand for. Without a process of learning we should only think of the words.

The process of learning to use words is not difficult to analyse. On a number of occasions the word is heard in connection with objects of a certain kind. Later the word is heard in the absence of any such object. In accordance with one of the few fundamental laws known about mental process, something then happens in the mind which is like what would happen if such an object were actually present and engaging the attention. The word has become a sign of an object of that kind. The word which formerly was a part of the cause of a certain effect in the mind is now followed by a similar effect in the absence of the rest of the previous cause, namely, an object of the kind in question. This kind of causation appears to be peculiar to living tissue. The relation now between the thought and what it is ‘of’ is more indirect, the thought is ‘of’ something which formerly was part cause, together with the sign, of similar thoughts. It is ‘of’ the missing part of the sign, or more strictly ‘of’ anything which would complete the sign as a cause.

Thoughts by this account are general, they are of anything like such and such things, except when the object thought of and the thought are connected by direct causal relations, as, for instance, when we think of a word we are hearing. Only when these direct relations hold can we succeed in thinking simply of ‘That’. We have to think instead of ‘something of a kind’. By various means, however, we can contrive that there shall only be one thing of the kind, and so the need for particularity in our thoughts is satisfied. The commonest way in which we do this is by thoughts which make the kind spatial and temporal. A thought of ‘mosquito’ becomes a thought of ‘mosquito there now’ by combining a thought of ‘thing of mosquito kind’ with a thought of ‘thing of there kind’ and a thought of ‘thing of now kind’. The awkwardness of these phrases, it may be mentioned, is irrelevant. Combined thoughts of this sort, we may notice, are capable of truth and falsity, whereas a simple thought—of ‘whatever is now’ for instance—can only be true. Whether a thought is true or false depends simply upon whether there is anything of the kind referred to, and there must be something now. It is by no means certain that there must be anything there always. And most probably no mosquito is where we thought it was then.

The natural generality and vagueness of all reference which is not made specific by the aid of space and time is of great importance for the understanding of the senses in which poetry may be said to be true. (Cf. Chapter XXXV.)

In the reading of poetry the thought due simply to the words, their sense it may be called, comes first; but other thoughts are not of less importance. These may be due to the auditory verbal imagery, and we have onomatopœia,[*] but this is rarely independent of the sense. More important are the further thoughts caused by the sense, the network of interpretation and conjecture which arises therefrom, with its opportunities for aberrations and misunderstanding. Poems, however, differ fundamentally in the extent to which such further interpretation is necessary. The mere sense without any further reflection is very often sufficient thought, in Swinburne, for instance, for the full response—

There glowing ghosts of flowers

Draw down, draw nigh;

And wings of swift spent hours

Take flight and fly;