53020120

.4.17

Thanks are tendered to Dr. Roland G. Kent, Professor of Comparative Philology in the University of Pennsylvania, for the assistance which he has given in the preparation of this dissertation.

344192

PREFACE

To the student of philology, semasiology offers an interesting field of work. It does not pin one down to the dry-asdust study of forms, but goes back to the beginning and searches for an intelligent understanding of the original purpose and meaning of these forms. It requires as much clearness of thought as a mathematical problem, but it permits also of a free use of the imagination to supply the links which are missing in the recorded development of language.

Many interesting articles have been written on the subject of suffixes. The theories suggested for the development of suffixes are so numerous and so varied, and yet they all seem so plausible, that one is reduced to the state of mind of wondering which writer on the subject is the best guesser. The suffix permits of such unlimited use of the imagination because it is so broken down in form that its original character is merely a matter of conjecture. When we leave this field and go over into the subject of compounds, we find less room for guesswork, because the members usually retain enough of their original form to permit of identification. But here we find another difficulty. Although the meaning and the form of the individual members may be clear, the link connecting the two members is frequently a matter of doubt. This dissertation is written for the purpose of showing some phases of this problem, of studying the semantics of compounds, of classifying some of the possible meanings which may connect the two members.

In most nominal compounds, the initial member is a stemform containing no case ending or other sign to denote its relation to the final member. Hence, in order to determine upon its meaning, it is necessary to consider all the possible relations which could connect the two ideas represented in the compound and then decide which is found in the actual usage of the word. Even after the meaning is determined upon, another difficulty presents itself that of classification. The meanings included in each class vary in the different languages.

A relation which is properly classified under Source in one language may belong under Location in another. Even in the same language, a word may fit equally well under Possession and Source. It is not possible to distinguish between direct and indirect object, because the distinction cannot in all cases be drawn with definiteness and certainty. For example, horsedealer and horse-seller are both placed in the Objective Class, although the English idiom requires to deal in horses and to sell horses.

In order to decide upon the meanings and classifications accurately, it is necessary to be well acquainted with the idioms of the languages involved and to think clearly in accordance with these idioms. The English idioms employed in the expression of the thoughts represented by the words of other languages must not be allowed to bias the judgment in determining upon the meaning. But because of these difficulties, there may well be differences of opinion as to the correct classification. The writer begs reviewers and critics to bear this in mind. E. F.