Sangarawa, Lakunusera, and several other works. And this most probably arose from the decline of the Singhalese as a language after the general destruction of literary records in the reigns of several Kings, and also from a frequent reference to,and close imitation of, the paraphrases and commentaries —the principal prose remnants of an ancient date, which, ex-necessitate, adopt a redundant style; one ill adapted to other species of composition.

These examples may perhaps suffice to shew, that as regards construction, there does not exist any the slightest difference between the ancient and the modern Singhalese; whereas in the construction of the English, even the socalled Johnsonian style, of a comparatively modern date, is now generally set aside for another of a different and more recent order. If however, further illustration upon this subject be necessary, the reader will find it in the subjoined paragraph of an inscription of remote date (A. D. 262), and a modern version thereof which follows it:

සිරිබර කැතකුල කොත්මකාවස් රජ පර පුරෙන් බට කැන උස බි අබහයි සලමෙවන් මහරජහට ළමේකුලෙන් සමජැයි දෙවු ගො බී සෙවු රැජනකුස ඉපද ඈපාමභයා සිරි විඳ පිලිවෙලසෙයි රජ වැතුමා සිරින් ලක්දිවූ පහයමින් සිට සිරිසඟ බෝයි අබහයි මහ රජ හුතු මාසත් ලැඟුසොලොස්වන හවුරුදුයෙහි වප්සඳදැන්මස්හිද සපක්දවස් සෙයි ගිරි වෙහෙර ඉසො අබහයි ගිරි වෙහෙර හි ඉසා වසන මහබික්සඟ හිමියන් මහ සෙන්වාකරයිතුමා බෑවත් හිමිය නන්සෙයිගිරි වෙහොම්හි පෙරෙතුබූ සිරිත්නජ අඃහයිගිරි වෙහෙර් හි සීරීනිජරුස්වා ගෙනැමෙ වෙහෙරට මේ සිරිත්තුඩුව ව ටීනි සිය හා සසාද මෙ වෙහෙර වසන මහ බික්සඟහිමියනට ඉසා කැමියනට ඉසා දස්නට ඉසා කටයුතු ඉසා ලබනු දිය යුතුයේ ඉසාවි වරුනෙන් එසේ කොට මේ සිරිත්තබනුලදී

Modernised thus (A. D. 1830)—

උත්තම ප්‍රියාවෙන් භාර වූක්ෂත්‍රිය කුලයට කොත වැනිමක් කාක වශරාජපරම්පරාවෙන් පැවත ආවා වූ උතුම් අබයසේලන මැති මෙබඳුම වු වන ඇති මේඝවන මහරජානන්ට එමමේඝ වර්ත කුලයෙන් සමානජා තිවු දෙවු ගොන් යයි කියන ලද අභිසේ

ක කරණ ලද රාජ මහේසිකාවගේ කුස ඉපිද පලමුවෙනි කටපත් ව හිඳ අනුක්‍රමයෙන් රජකමටපත්ව තමාගේ සෝභාවෙන් ලං කාද්වීපය ප්‍රසන්න කරමින් සිට ප්‍රිසඟබෝධි අබය මහරජ තෙම ත පිරජකමට පත්වූ සොලොස්වන අවුරුද්දෙහි ව මසපුර දස වක්දවස්හි චේතියගිරි විහාරයෙහිද අබසගිරි විහාරයෙහිද යන මේ විහාර දෙකේ වසන මහා බික්ෂු සංහයා වහන්සේලා රැස්කර වාතමාගේ භාත්‍රැ වූ ගුරුන්වහන්සේ විසින් චේතියගිරි විහාර යෙහි ද ස්වකිය අබයගිරි විහාරයෙහිද පෙර තුබුවාරි ත්‍රයන් තමන් . ටරුචිවගෙණ මේ මාගේ විහාරයට ද මෙ කිවාරිත්‍රයන්තු බුව හො ත් වටනේයයිඊට නිසි සි හිනුවන ඇති අමාත්‍යයන්හා සමග සන් සන්දනය කොට මෙම විහාරයෙහි වාසය කරන්නා වූ සංඝ

යා වහන්සේට ද කර්මාන්තකාරයින්ටද දාසයන්ට ද ක්‍රත්‍ය වල ටද ලාභදානාදියට ද නිසිව්‍යවස්තා නියම කිරීමෙන් සමකොට මේ මතු ප්‍රකාස වන වාරිත්‍ර විධිය ව්‍යවස්තා ස්ථාපනය කරණ ලද් දේය.

"The great king Sree Sangabo Abaya (born unto the great and illustrious King Abayasla, a descendant of the dynasty of Okāka, which is a pinnacle of the very illustrious royal race of Keth, and born in the womb of the installed Queen-Consort Dewugon of the same illustrious race) having risen to the first offices of the state, and having in the usual course succeeded to the regal office, and illumined the island of Lanka by the effulgence of his Majesty, hath on [this] 10th day of the growing Moon in the Month of Wak [Octr.-Novr.] and in the sixteenth year of his reign, summoned the Clergy of both the temples Sègira and Abayagiri; and being desirous of instituting with reference to his [new] temple the same rules which were prescribed · by his elder brother Sovereign Lord, respecting the Temples Sègira and Abayagiri-having also consulted competent persons in that behalf-and having further ordained that the said rules should govern the priests of this temple, its labourers, slaves, affairs, receipts, and disbursements, &c. and having thus assimilated the rules [in respect of all

the three temples] hath perpetuated the hereinafter mentioned rules of practice.

We have thus seen that the Elu is no other than the Singhalese language: but the question still remains" Is it a dialect of the Sanscrit?" †

Let us first inquire, What is a dialect? It is defined by several lexicographers thus: "DIALECT, dialectique F., dialectica L., DIALECTIKE G., is a manner of speech peculiar to some part of a country, and differing from the manner used in other parts, yet all using the same radical language as to the substance of it." Now those who maintain that the Singhalese is a dialect of the Sanscrit, do so upon the ground that many of its words are of Sanscrit origin. But this is no more correct than that the Portuguese, which abounds with Latin terms, and the English with French, are respectively dialects of the language from whence such terms are derived. And, if the premises whence the conclusion under consideration is inferred be correct, we may as reasonably affirm that the Singhalese is also a dialect of the Maghadi or Pali; since words derived from the Pali into Singhalese are as numerous as those from the Sanscrit. Indeed the

author of the Sidath' Sangarawa says:

"Words may be divided into three classes: 1st, o

purely native Elu words; 2ndly,

Pali, Sanscrit, &c.; and 3rdly,

words common to Elu,

e, words derived from

Since translating the above passage, we have found an English version of the entire Inscription by Mr. Armour: which see in the Ceylon Calendar for 1834. The words within brackets [] in the above, as in other translations elsewhere have been supplied by us.

"The language of Ceylon is distinct and unique, though like most of the Indian languages, it is supposed to be a derivative from the Sanserit, "-Pridham, I. p. 272.

"The Elu has undoubtedly given birth to the vernacular language of this country. It appears to claim great antiquity; and being derived from the Sanscrit, a great proportion of the words may be traced to that source."Clough's Preface to the Singhalese Dictionary.

the Pali, Sanscrit, &c. but slightly different from the original by their adoption into the Elu." p. 4.

Upon the above process of reasoning, therefore, we must conclude that the Singhalese is a dialect of both the Sanscrit and Pali. But this, upon a view of the definition with which we have set out, is absurd. For, since the Elu has roots and words of its own, and words too, which (though bearing some affinity to) are not derived from, the Sanscrit; they cannot be pronounced to be the same radical language as to "the substance of it.

Again, a language and the dialect of that language are one and "the same radical language." e. g. The Attic, the Ionic, the Doric, and the Æolic, are dialects of the same radical language, the Greek; and agree with each other in the gene ral principles of declensions, conjugation, &c.; but, I believe, differ from one another in spelling or pronunciation, or both; variations, which in the words of our definition, affect merely "the manner of speech," and "the manner used." Now the Singhalese must be considered devoid of this alleged relationship, if some at least of its principal grammatical forms are different from those in the Sanscrit. To this test we shall submit the Singhalese, in order to ascertain if it be derived from the Sanscrit. And how do we find them? Even more different from each other upon substantial points, than is Pali from Sanscrit. For, a great portion of the Singhalese language is not derived from the Sanscrit; the Singhalese has but two genders, whereas the Sanscrit has three (see Grammar § 24); in the former the verbs are not conjugated as in the latter, nor are the roots the same in both; the changes which words undergo in the Singhalese are altogether upon a process different from, and less certain than, that in the Sanscrit; the declensions are also different in the Singhalese from those in the Sanscrit, the dual being unknown to the former; &c.

If moreover, it be true that the roots of words in a parent language, as well as those in that which is supposed to be its

dialect, are identically the same, the following examples will shew that this identity is wanting between the Sanscrit and Singhalese roots; although at the same time, it is clear, there is such an affinity between them as to confirm our theory, that they are both the offspring of one common parent. e. g. óró 'do,' Elu, 6 Sanscrit; 'see,' Elu, Sanscrit; D'eat' Elu, Sanscrit. Without multiplying examples, we may here refer the reader to § 57 of the Grammar, where he will find a number of verbal roots, with which he may easily compare the roots of like signification in the Sanscrit.

The adjectives have not any degrees of comparison, nor indeed is the relative pronoun used in the Singhalese. (See Appendix C.)

The formation of cases presents no less a peculiarity in the Singhalese. In the first place, the vocative suffix in Sanscrit is formed from, or rather, is a modification of, the termination proper to the nominative. Professor Bopp says in his Comparative Grammar, § 204, "The vocative, in the Sanscrit family of languages, has either no case-sign at all, or is identical with the nominative; the former is the principle, the latter the practical corruption." In the Singhalese no two case-signs are alike in the nominative and the vocative, except ; and the use of this very vowel, it is remarkable, presents a peculiarity distinguishable from the Sanscrit, which in general disclaims long vowels in the vocative singular. The fact, however, that there are nine other case-signs in the vocative, totally distinct and different from those proper to the nominative (compare §§ 26 and 37), clearly establishes the want of that particular relationship which is ascribed to the Singhalese and Sanscrit. In the next place, the instrumental and the auxiliary present no difference of case terminations in the Sanscrit and Pali, whereas in the Singhalese, no two signs in those cases are identical, according to the Grammar (compare §§ 28, 29.) Lastly, it is also remarkable, that in the