p The range of moot problems associated with the study of phantasy extends far beyond the problems of experimental methods, materials and objects of investigation. Above all, we have to point out that the experiments, the result of which can easily be predicted a priori, are notably inconsistent methodologically. In Chapter II, in connection with the discussion of analysis and synthesis, we asserted that whenever the material of a task is integral, its solution, naturally, presupposes but a single act—analysis; if the material has been divided by an experimenter, a subject is expected to synthesize its components. When Welch (273) advised his subjects to contrive specific household objects out of an assigned set of geometrical figures, and Newell, Shaw and Simon (221) instructed their subjects to find a way out of a maze, we can, in foreshadowing the results, predict that in the first experiment the subjects would be performing combinatory acts, while in the second, full or partial scanning of variants. Consequently, the results become known long before these experiments have been performed, which questions the very valuableness of such experiments.

p Insufficient informativeness of the results of experiments aimed at studying intellectual activity, obviously caused a great number of psychologists to give preference to essentially different methods of obtaining information, such as observation and, especially, introspection (self-observation). Way back at the onset of the experimental study of thinking Ribot and Binet qualified introspection as the only way to adequate cognition of the inner world (126, 9). At present, Donald M. Johnson who has never underestimated the importance of authenticity, reliability and accuracy in the experimental results, points out that "the method may be too blunt to detect the subtleties of the 82 thought processes" (192, 42). However, even though Johnson admits the equality and usefulness of both experimental and introspective methods, believing that they supplement each other, a number of other psychologists, such as K. Heymann, attribute the main role to introspection (183).

p The role and significance of introspection has still not been fully determined. In our opinion, the truth has been most closely approached by those psychologists who consider it useful to obtain information from all sources, including introspection. Yet, it would be necessary to emphasize that the data obtained through introspection has to be regarded as primary raw material which would have to undergo a serious transformation. In other words,- the driving mechanisms of phantasy, its reasons and motivation have to be assessed by an experimenter rather than by his subject who only may serve as the source of initial information. This opinion is expressed by N. Kostyleff (196, 15-251), A. Burloud, Van de Geer (157, 12), etc. Consequently, the introspection data may be regarded as a means of description but not of explanation.

p A question arises: what does an explanation of psychic phenomena have to imply? One of the most widely circulating answers to this question is contained in the conception of reductionism which reduces the notions of a particular science to those of other sciences. In summarizing the methodological orientations of a number of psychologists, John Gaito and Dwight Leonard maintained that certain psychologists revealed a pronounced trend to interpret an explanation as a reduction of psychic phenomena to underlying physiological and biochemical correlates (155, 69). This interpretation agrees with the methodological principle of isomorphism which endows all psychic phenomena with definite neurophysiological correlates. The hypothesis of isomorphism is a contemporary form of psychophysical parallelism. Besides, as we have proved in considering the non-psychological conceptions of phantasy (see Chapter II), a simple substitution of the terms of one science by the terms of another adds but little to our general knowledge. A number of foreign authors, however, have realized the futility of reductionism  [82•*  since it leads to disregard for the qualitative originality and specific features of each science [e.g., Gregory (167, 150-151), and others].

p If we, in contrast to the adherents of reductionism, admit not only the reality of psychic processes, but also the reality of specific laws governing them, we may, quite reasonably, find it possible to explain psychic facts by the means of a particular psychological science. Naturally, this allows us to use the data provided by conformable sciences for interpreting the phenomena of other levels associated with phantasy (e. g., numerous essential peculiarities of the motor components of creativity must be explained by physiological laws).

p At the same time, the ends and the character of a psychological explanation are interpreted differently by different authors. There are investigators who regard a psychological explanation as dovetailing observed facts with well-known and unquestionable truths. Charles Spearman writes on this account: "the creativity should be traced down not to processes or laws of any novel and wonderful kind, but on the contrary to those which are most familiar and most self-evident...

p “These ... laws should be ultimate, so as to have the widest possible extent; the laws should not be deducible from ... more general" (252, 14). It would be easy to notice that Spearman’s criteria may turn out to be incompatible: quite probably, the most extensive and universal laws, which may provide an adequate explanation of all the facts, will- eventually turn out to be most "novel and surprising”.

p D. E. Berlyne gives this interpretation to the scientific explanation in psychology: "To explain an event means to reconstruct a succession of transformations that connect it, link by link, with some state of affairs that we take as our starting point" (122, 82). Such an interpretation, in turn, reveals at least two issues: first, how far do we have to advance in search of "the starting point"; second, what has to be adopted as "the starting point”?

p In connection with this, we find it necessary to dwell on the distinctive features of descriptive and explanatory concepts in psychology.  [83•*  In a number of publications we pointed out that a strict differentiation of descriptive and explanatory notions is an indication of the theoretical maturity of a science. On the 84 other hand, by neglecting the production of explanatory notions, confusing them with descriptive notions and, still worse, substituting the former by the latter we will lapse into empiricism and positivism (81, 79).

p The other reason for attributing such a great significance to the differentiation of descriptive and explanatory notions lies in the fact that this enables us to define more precisely the subject of our argument in discussing various experiments and theories and to establish the aspects which need additional elaborations: basic facts or a theoretical construction. This differentiation will also engender certain criteria for evaluating descriptive and explanatory notions. Thus, it would be advisable to assess positively a profusion of descriptive notions, reflecting the diversified character of natural phenomena. Moreover, the number of new descriptive notions introduced by a researcher attests to the volume of information he has obtained from reality as a result of his experiments and observations.

p Explanatory notions have to meet completely different demands. They should naturally materialize from the fundamentals of a particular science and organically combine disintegrated and even remote phenomena, exposing their unity against the background of their diversity. In their having a great number of common features, explanatory notions may not be numerous: on the contrary, the fewer of them are used for interpreting facts the more exact and perfect a theory seems to us (77, 140).

The elaboration of explanatory notions for the phenomena of phantasy calls, first and foremost, for a profound study of the respective experimental data which, however, (as we have already noted), must contain new information on psychic reality, i.e., not predictable a priori.