This being so, errors of introspection, supposing such to be found, will in the main be sufficiently distinguished from those of perception. Even an hallucination of sense, whether setting out from a subjective sensation or not, always contains the semblance of a sense-impression, and so would not be correctly classed with errors of introspection.
Just as introspection must be marked off from perception, so must it be distinguished from memory. It may be contended that, strictly speaking, all introspection is retrospection, since even in attending to a present feeling the mind is reflectively representing to itself the immediately preceding momentary experience of that feeling. Yet the adoption of this view does not hinder us from drawing a broad distinction between acts of introspection and acts of memory. Introspection must be regarded as confined to the knowledge of immediately antecedent mental states with reference to which, no error of memory can be supposed to arise.
It follows from this that an illusion of introspection could only be found in connection with the apprehension of present or immediately antecedent mental states. On the other hand, any illusions connected with the consciousness of personal continuity and identity would fall rather under the class of mnemonic than that of introspective error.
Once more, introspection must be carefully distinguished from what I have called belief. Some of our beliefs may be found to grow out of and be compounded of a number of introspections. Thus, my conception of my own character, or my psychological conception of mind as a whole, may be seen to arise by a combination of the results of a number of acts of introspection. Yet, supposing this to be so, we must still distinguish between the single presentative act of introspection and the representative belief growing out of it.
It follows from this that, though an error of the latter sort might conceivably have its origin in one of the former; though, for example, a man’s illusory opinion of himself might be found to involve errors of introspection, yet the two kinds of illusion would be sufficiently unlike. The latter would be a simple presentative error, the former a compound representative error.
Finally, in order to complete this preliminary demarcation of our subject-matter, it is necessary to distinguish between an introspection (apparent or real) of a feeling or idea, and a process of inference based on this feeling. The term introspective knowledge must, it is plain, be confined to what is or appears to be in the mind at the moment of inspection.