Are Japanese Turu and Ito Related to Korean Turumi and Sil Respectively

Abstract

For a long time now deep interest has been felt in the affinity problem of Japanese. This situation has resulted in various suggestions being put forward. But if the theories expressed hitherto are broadly divided into two, the two types may be called the 'Northern hypothesis ' and the ' Southern hypothesis.' However, as far as concerns a matter of genetic relationship-setting aside the question of borrowing-, it cannot be considered that a single language can be simultaneously traced directly back to two separate parent languages. Consequently the ' Northern hypothesis' and the 'Southern hypothesis' are incompatible, and in fact they have competed with one another. But it is di~cult to come to a final conclusion as to which of the two is actually correct. Be that as it may, the two hypotheses have a point of difference in their character. To use an ana10gy, it may be described as follows. The ' Southern hypothesis ' might be acceptable if it did not deceive itself by making such errors as ' Nifhe divided by three gives tree.' The example is, perhaps, ridiculous, but it seems to be upon the basis of such ridiculous etymologies that the ' Southern hypothesis ' is built (.see below, note 25). The ' Northern hypothesis ' on the other hand, to judge it in the light of the orthodox dogma of comparative philology, can definitely not be regarded as something proved, but there is less fear in it of confusing ' th,ree ' with ' tree ' by their accidental resemblance. The ' Northern hypothesis' i~> more likely to say ' Te'n divided by three gives th,re~.' This is the error inherent in the use of lar*"e and rough