and so eagerly give free rein to their fancy as in the interpretation of place-names — just that branch of the science of philology which the trained investigator regards as the most difficult, and as making the severest demands upon his knowledge.
A vast number of other place-names have been corrupted by faulty hearing or faulty spelling at the time the locality was surveyed, or as a consequence of deficient local knowledge on the part of the persons from whom the topographers have chanced to seek their information with regard to the various localities they have studied. Nor have the maps been subjected to the revision of a scientific philologist before being printed; and once they are printed, they have been regarded, consciously or unconsciously, as authoritative and unalterably correct.
These reflections, suggested by the history of place-names in Sweden, may without any very great change be applied to the study of the nomenclature of several other civilised countries. Whence we see, that place-names are corrupted by those who study them even when they have themselves from childhood spoken the language to which those names belong. How much more easily therefore, and how much more frequently, will the place-names be corrupted when the language to which they belong is not the mother tongue either of the observer who records them or of the scholar who edits them, and who have either a very slight knowledge, or none at all, of the language in question. In such circumstances it is evident, that mistakes are unavoidable, and indeed they may be expected to be fairly numerous, especially when the observer is not a trained philologist nor can possibly be so.
To ensure absolute and complete accuracy in the topographical nomenclature of any region, the names ought therefore to be recorded by a skilled philologist. Nor is it sufficient that he should have had a thorough grounding in the general study of philology; it is equally indispensable that he should know the dialect spoken in the region under investigation, no matter whether it is a dialect of his own mother-tongue or of some other language. To philologists this statement is self-evident. To non-philologists it may possibly appear too exacting. It will not be superfluous therefore if I illustrate the point from my own experience as a student of the Lapp dialects. Whenever I visit a fresh dialectal region, I find invariably that the observations I gather during the first few days regularly turn out to be more or less defective and inaccurate in respect of various minutiae. The reason of this is, that for the first few days my ears are not yet accustomed to the new dialect. And this happens notwithstanding that I am familiar with investigations of this character, that I possess more than a theoretical acquaintance with phonetics, know the Lapp language, and do not possess a too insensitive ear. In the case of a less practised inquirer, the errors would not be confined to minutiae alone, but they would extend to more important matters, a thing which may indeed also happen in the case of even a very well-trained observer. For it is no easy matter to hear correctly, perfectly simple though this may seem to the layman. In place of the more or less unfamiliar sounds that are pronounced by the person you are questioning, you unconsciously substitute sounds you are familiar with that most closely resemble them, and often a pretty considerable time elapses before the difference is perceived. Very frequently it turns out, when you come to sift and arrange the materials you have collected, that the difference is one of no consequence, and in
IIIFカラー高解像度
