_Their discrepancies—Reduced to three types: (1) Classification of pleasures and pains—(2) Classification of emotions: two forms, empiric and analytico-comparative—(3) Classification of representations, intellectualist form—Critical remarks—Impossibility of any classification._ The confusion of that department of psychology which deals with the feelings and the vagueness of its terminology appear in all their fulness with the problem of classification. Although, for reasons which will be given at the end of the chapter, a complete and satisfactory classification seems to me impossible, there has been no lack of attempts, and it must be admitted that they are not illegitimate, at any rate as approximative efforts towards a provisional order. Within the last fifty years, in spite of the very moderate amount of zeal with which psychologists have studied the feelings, we find about twenty treatises signed by well-known names, not to mention minor variants.[90] They are far from being in agreement, except on some few points, and when they are compared, in order, if possible, to reconcile them, the first impression is one of inextricable confusion and hopeless divergence. If we examine them a little more carefully, we begin to see light. We see that the differences are only in the objects classed and the methods followed; in one word, it is possible to attempt a classification of these classifications; and when this is done, we find, if I am not mistaken, that they can all be reduced to three types. (1) Some writers, virtually, classify only pleasures and pains, tracing back the whole of affective life to their modalities. (2) Others classify the emotions properly so called; and here we must distinguish two groups, according as the method employed is purely empiric, and founded on current observation, or has recourse to analysis and genetic research, after the manner of the so-called natural classifications. (3) Lastly, others classify intellectual states pure and simple, and, conversely, the affective states which accompany them: this is the intellectualist method. Footnote 90: For a historical summary of these classifications, consult especially Sully, _The Human Mind_, vol. ii., Appendix F, p. 357, and Bain, _Emotions_, Appendix B. In order to justify our distinction, we shall successively examine these three types. This excursion on an ungrateful soil will not be altogether without, at least, a negative utility. I. As many writers show a common tendency to reduce the whole of the affective life to pleasures and pains, considered as essential and fundamental phenomena, it is natural that one category of classifications should have been based upon these.