we read an author like Voltaire we feel that in his hundred volumes he said about all he had to say. He puts his thoughts before us lucidly and vigorously, but not profoundly. In the case of Goethe, on the other hand, who could think both scientifically and poetically, we often realize that in spite of the enormous extent of his works his language is not infrequently an indication of his thoughts rather than the expression of the thoughts themselves. The existence of Dante Societies, and Shakespeare Societies, and Goethe Societies has its justification in the conviction that the thoughts of these master minds can be fully understood only by the cooperation of many of inferior caliber. By thus combining and comparing the individual and partial views of a number of separate intellects, they may gain at least an approximately adequate grasp of the psyche of these prodigies.
Perhaps the most remarkable phenomenon exhibited by all the languages of the world is a process we can hardly call by any other name than deterioration. Take, for example, the Greek. Many of its oldest words are both longer and more sonorous than the later ones. But often even the earliest form shows evidence of weakening, abrasion and contraction. At least one consonant was lost in the period lying between the prehistoric and the historic. In many words two syllables are drawn together into one, or a shorter takes the place of a longer word. Sometimes the longer form existed for a time alongside the shorter, eventually to displace it entirely. Often the attenuated vowel e takes the place of the more sonorous a. It would almost seem as if when a word of two or more syllables having a traditional signification had for a time been in use it dawned upon the primitive mind that it could be shortened without losing its meaning. In Greek and Latin this process is not carried very far, but in French all the words derived from the latter language consist only of that portion that precedes and includes the accented syllable. We know that when an uneducated person tries to reproduce the pronunciation of a long foreign word or one that is foreign to him, he usually gets only a part of it and that part often incorrectly: education becomes "edication," sitting" sit'n," somewhat "sumet," the other "t'other," and so on with innumerable examples. This is the every-day process. But where and when shall we place the era of upbuilding? When were the fuller forms in use? The procedure that falls within our ken furnishes us with no answer to the question, not so much as an approximation thereto. Even those languages of which the study began only a generation or two ago exhibit the same phenomenon. The Bantu, the most widely disseminated speech of South Africa, presents many instances where two or more syllables are contracted into one, and where former syllables have left but a single letter as evidence of their one-time existence. As soon as a language is reduced to writing, or becomes a matter of study, and the rising generation is taught to pattern after its predecessors, the
