Let A represent the idea in the latent content and C the corresponding
"symbol" in the manifest content. Suppose that in a number of cases a
correlation is observed between A, the antecedent latent idea, and C, its
consequent or sequential manifestation in the dream-consciousness.
Thereafter, the observer comes to interpret the re-appearance of C in a
dream narrative as a sign of the presence of the affiliated idea A, in the
latent content. And, as Thomas Hobbes phrased the matter in 1651, the
oftener they have been observed in like connection, the less uncertain is
the sign.[15] Now this is precisely the way we come to recognize the verbal
signs of our mother-tongue. And our confidence that a given speech C' is
significant of a meaning A', in the speaker's intent, is arrived at by
relying upon, if not consciously formulating, just such a causal connection.
Where an existing language is concerned, this is a perfectly legitimate
tooling of thought. But in applying such inferences to a supposititious
language of dreams, psycho-analysts are begging the question, as well as
running into other kinds of fallacy as to the powers of the Unconscious.
The meanings and significations of dream-items are not so simply made out as
in language. For one cannot readily make sure that the relationship or
affiliation between A and C has been observed in its purity; there is an
uncertainty coming from the possible interposition of a variable factor,
which may have vitiated the observation, as Alfred Sidgwick points out in
his "Application of Logic."[10] So let us well consider the basis of any
inference of meaning in dreams, and how far the language-analogy applies.