the end. Interrogative particles are, however, placed after the verb.

4. Plural terminations, case signs, and prepositions are placed after the noun. Plural terminations precede other suffixes.

5. The direct object of a verb is placed immediately before it. 6. A noun governed by a preposition precedes the direct object of the verb.

7. Conjunctions are placed after the word or clause to which they belong.

8. Dependent clauses precede principal clauses.

These rules are in reality deducible from the principle placed at the head of Rule 1, viz. that qualifying words or phrases precede the word or phrase which they qualify. Their effect is to make the order of a sentence precisely the same in both Japanese and Korean, so that in translating from one language into the other, no inversion of the construction is required, except in the case of an occasional divergence of idiom. The following sentence will serve as an illustration.

The usual English order is, 'What good will come of regretting a thing after losing it?' but it can be varied, which is impossible in the Japanese or Korean versions. In some cases, however, a limited degree of freedom is allowed.

SUMMARY.

It may be convenient to give a brief summary of the conclusions arrived at in this paper.

I.

The phonetic systems of Japanese and Korean differ considerably. Japanese has five vowel sounds, Korean nine. Korean has a class of aspirated consonants which do not exist in Japanese. A syllable may end with a consonant

in Korean, but not in Japanese. On the other hand, neither language is monosyllabic, and in neither have we the rule of the harmony of vowels. Both languages have only one consonant for r and 7, and this letter cannot begin a word. The forms of the same Chinese words in Japanese and Korean afford a means of discovering their letter-correspondences. The most noticeable are, that a Japanese h or fis in Korean p, a Korean h in Japanese g, a Korean 7 final in Japanese tsu, chi, shi, or su. A considerable number of roots are identical in the two languages.

II.

The Japanese and Korean nations are characterized by an impersonality of conception which may be traced in every production of their national genius. In the grammar it is shown by their neglect of person, gender and number, the want of a verb 'to have,' the imperfect development of pronouns and numerals, and of a passive voice. The strong points of their grammar are the regularity with which the distinction of noun, adverb, adjective, and adverb is marked in the same word, and the copiousness of what may be called modal terminations added to verbs.

III.

Japanese and Korean depend almost exclusively on suffixes and position as grammatical appliances, to the exclusion of prefixes, augment, ablaut, and reduplication. The suffixes are usually easily distinguishable from the root and from each other, but sometimes are so closely welded as to defy grammatical analysis. Resemblances are traceable in the terminations used to distinguish the same word when used as different parts of speech.

The grammar of the noun agrees in character in Japanese and Korean, the chief differences being due to the differences in the phonetic systems, already pointed out.

In the pronouns less similarity is noticeable than might have been expected. In both languages the pronouns seem connected with such verbs as 'come,' 'go,' 'dwell,' 'be,' etc.

The numerals have some resemblance in the character of their development, but almost none in form. Only one or two agree, and these not as numerals but as words belonging to the general vocabulary. This feature is of less importance here than it would have been in the case of languages of the Aryan type.

The rules which govern the position of words in a sentence are identical in Japanese and Korean. These rules are less rigid than in Chinese, and more so than in Aryan languages.

There can be no doubt that a genuine relationship exists between Japanese and Korean, but it is by no means easy to estimate its degree. The principles applicable in the case of Aryan languages are of little use to us here. According to our experience of them, two languages with no common numerals could hardly be classed together at all, while on the other hand, the agreement of Japanese and Korean in the elaborate rules for the position of words in a sentence suggests a very close affinity indeed. Everything considered, we may perhaps regard them as equally closely allied with the most remotely connected members of the Aryan family.

It should not be too hastily inferred that because the Japanese and Korean languages differ so widely, an equal distance separates the Japanese and Koreans as nations. The questions of linguistic and ethnological affinity are distinct, though they have no doubt a bearing on each other. And with the instability of vocabulary and weakness of grammatical development which characterizes nations in the earlier stages of their progress, a considerable degree of divergence in their languages is compatible with a much closer affinity of race than would be possible with more civilized races. That the Japanese and Koreans are a case in point is rendered probable by geographical, historical, and physiological considerations.

In preparing this paper, my principal source of information respecting the Korean language has been some manuscript manuals prepared by the Japanese Interpreters resident at Fusankai, the Japanese settlement in Korea. For the

grammar, the only authority is a sketch in Dallet's "Histoire de l'Eglise de Corée," but it is unfortunately in many respects incomplete, and I have been compelled to depend upon such a knowledge of it as could be extracted from the Japanese manuals just mentioned. This may be some excuse for errors which future inquirers will doubtless discover. The late Mr. W. F. Mayers, Chinese Secretary of our Legation at Pekin, was for some time before his death engaged on a Korean Grammar, of which much was expected; but it is to be feared that it was not in a sufficiently advanced state to admit of publication. Some of his manuscripts, which I have had an opportunity of inspecting, show that he had acquired a knowledge of Korean far superior to that possessed by any other European scholar, and it is deeply to be regretted that an untimely death prevented him from giving to the world the fruits of his researches.

365

ART. XIV.-Dialects of Colloquial Arabic.-By E. T. ROGERS.

THE Arabic language is commonly spoken throughout a very large area of the old hemisphere. It is the language of the whole of North Africa, which includes Morocco, Algiers, Tunis, Tripoli, and Egypt. It is also spoken down the

Eastern coast, and in a not inconsiderable portion of the interior of that vast continent. Its home is the peninsula of Arabia, whence it spread also to Palestine, Syria, and Mesopotamia.

Arabic is also known as a written language wherever the religion of Islâm exists, namely, in Turkey, Asia Minor, Circassia, Persia, India, Tatary, etc. In these latter countries, where it is only the language of the religious books, and co-existent with colloquial languages, it has retained its original purity, whilst much of it has been incorporated into the local languages spoken by Muslims,such as Turkish, Persian, Hindustâni, etc.

But where Arabic is the language of every-day life, spoken both by the learned and the vulgar, it has naturally been influenced by a variety of local circumstances which have gradually divided it into a number of separate dialects.

The language is so rich in itself, that it can well afford to be broken up and to supply the foundation of several dialects, each correct in itself. Thus words in common use in Morocco which are unintelligible to the uneducated in Syria, and vice versa, may each have an equally pure origin. But, on the other hand, it must be admitted that we often meet with words whose meanings have been quite distorted, and with others whose origin it is impossible to trace to a classical Arabic source.