Notes on "the Maori-Polynesian Comparative Dictionary."^*

It has been said of some books, that difficult as they may have been to write they were as difficult to review. But this is a dangerous plea for the amateur reviewer with which to cover his own shortcomings, since it might well be replied, a priori, that a master of his subject may write even easily, what an outsider with a reasonable suspicion of his own ignorance, would find great difficulty, and without such a suspicion great peril, in criticising And there is moreover, in the case of Mr Tregear's Dictionary, the practical answer, that several critics, all presumably more or less learned, did in fact review it, and were able to ascertain and declare its merits within a very short time of its publication. Even more than this—sufficient as we must suppose their investigation to have been, most of them did not apparently find, in a book of some 700 pages, any defect to which it was worth calling the serious attention of those they were addressing.

Nevertheless, cost what it may, I must confess that I have felt great difficulty, indeed many difficulties, in dealing with this book, although, instead of an adequate review for which I am not qualified, I have undertaken only to comment upon certain parts and aspects of it. That I have left other parts and aspects unnoticed, or but little noticed, is according to the conditions on which I undertook this paper.

Mr. Tregear presents his work to us as two-fold.^† It is intended to serve as a Maori dictionary in the ordinary sense, but with many additional words, and with certain new features—notably the inclusion of mythological persons and places, a manifest improvement: and it is also as he puts it somewhat

^* The Maori Polynesian Comparative Dictionary. By Edward Tregear, F.R.G.S., F.R.H.S., &c. Wellington—Lyon & Blair.

^† See circular advertisement issued with the book,

page 2 generally, "philological"; but this is explained to mean that "for the first time the student of Maori will be enabled to ascertain the original value of a native word by comparing it with its sister words in the other dialects of the Polynesian language; a reader may now readily ascertain whether the New Zealander has kept the primitive sense of a word, or whether it has received a local warp." That is a good deal, the more especially as the author's help to the reader is merely the collocation of so many contemporary words. But more than that—with only the same help—"the European student will by means of this lexicon be in a position to compare the Oceanic tongues with the languages of the great continents, as the grouping together of vital words permits the most uncorrupted form to be discriminated and adopted."

The book, then, if it fulfils the high aims of its author, may be called two works in one, a Maori dictionary and a comparative dictionary of Maori and other Oceanic languages. To do either of these as it should be done, might well occupy several competent men for several years. If Mr. Tregear, single-handed and as a bye-work, in the leisure of a few years, has done both and done them well, it will be agreed that he has made not only a big book, but under the circumstances, a great one. The crucial question is as to its quality; and to this I will shortly return.

Of the difficulties he met with in the Maori part of his dictionary Mr. Tregear does not say much, but his clerical labours alone must have been severe if he had as a beginning to copy out "the 7000 words, with their meanings, which have hitherto been included in the authentic lists": the last curious phrase being, I presume, a delicate euphemism for Archdeacon William's Dictionary, though including no doubt the two or three hundred valuable words given in Dr. Shortland's works. Hardships of this kind seldom, luckily, fall to the lot of even the most hapless author. Yet this was a small part of the merely mechanical labour involved, to say nothing of the higher kinds. In a paper (Trans. N.Z.I, xxiii. 532) published apparently about the time of the completion of the work, Mr. Tregear has occasion to mention his own "untiring industry," and the candid reader will justify the compliment.

As to the difficulty of the 'comparative' part of his task, Mr. Tregear speaks with entire frankness of its stupendous character. "Several attempts," he says (p. ix.) "have been made to produce a Comparative Polynesian Dictionary, but so gigantic was the labour, so enormous the mass of material, that the compilers have shrunk back appalled in the initiatory stages of the work, and all that remains of their efforts has "been a few imperfect and unreliable pages of vocabulary page 3 "scattered here and there through books treating of the Malayan "and Pacific Islands."

It may be appropriate here to interject that if the foregoing is a fair account of the quality of the work of Mr. Tregear's predecessors, our regret at the smallness of its quantity will be itself as small. But one is tempted to ask, who were these early compilers, as pusillanimous as they were rash, who on the very threshold of a great work shrank back appalled into their native obscurity, terrified, as it seems, into silence at the mere sight of what they had undertaken? Mr. Tregear is not of their kind, but, from a feeling of delicacy no doubt, does not name them; and I only refer to them from fear lest they should be identified with others of a very diffierent kind whose works indeed are the only ones I happen to possess containing some short comparative vocabularies of Oceanic languages, travellers I mean, and missionaries, such as Capt. Cook (and his assistants), J. Craufurd, the Rev. G. Turner, A. E. Wallace, and last but not least, Dr. Codrington; men who, though having other objects in view, have in fact acted as pioneers of learning, going where alone particular forms of knowledge were to be had at firsthand, and bringing back, amongst much else of value, what they could of this kind; thus supplying to the comparative philologist, a more or less substantial part of his material.

But, passing that by, and willingly admitting what is indeed too obvious for discussion—the great labour involved in a work of this kind—I will return to the preliminary question I proposed, a question all important to the New Zealand student, and to the foreign scholar, who would use Mr. Tregear's Dictionary as a work of reference: can it be taken as trustworthy, and authoritative? Does it, as far as it goes, present to us genuine well-arranged knowledge, and nothing else? and does it fairly represent the work and views of the highest Maori, Polynesian, and Oceanic scholarship?

This evidently implies and suggests the further question:—Is Mr. Tregear duly qualified for the high work he has undertaken?

The question may seem unnecessary—I hope not invidious—to those who are acquainted with Mr. Tregear, especially to those who, like the members of our Metropolitan Scientific Society, may be said to have already answered it in the affirmative, and—may I say?—with acclamation, bestowing upon him their highest honour about the time of the publication of his Dictionary, and at the end of his year of office putting on record a special commendation of his book. But though myself only a student of Maori, not a Maori scholar, still less a Polynesian scholar, I claim a strong interest in the progress of page 4 Maori and Polynesian scholarship, and in that interest, I think, the question should be asked and fully considered.

Now, whatever other qualifications besides industry Mr. Tregear may have or may want, he is certainly an enthusiast. If he has not hitherto been known in connection with questions of exact scholarship in Maori or other Oceanic languages, he has long been known as keenly interested in certain Polynesian questions which he has prosecuted—in his own way no doubt, and with varying success—but always with enthusiasm. And this enthusiastic temper of mind by its very nature induces him to take a large and generous view of the importance of the work he is upon, and, incidentally, of his own share in it. Remembering this, the reader will perhaps be able to see clearly, and to justify, certain points in the book which might otherwise seem obscure, or even open to animadversion.

Indeed, the note of enthusiasm, as I may call it, is struck in the very first word in the book if we may take it as beginning with the title-page. It is common, I believe, among lexicographers to use the indefinite article in the title of their dictionaries; to take a New Zealand instance, Bishop and Archdeacon Williams call theirs "A Dictionary of the New Zealand Language"; treating it as one of a class, although at the time of its publication, the rest of the class may have had only a potential existence. But this practice, derived merely perhaps from a blind literary instinct, does not hinder Mr. Tregear from taking an independent view of his own work, as one by itself, requiring therefore the definite article: it is called accordingly "The Maori-Polynesian Comparative Dictionary," and when fully examined, I believe, it will be found to have characteristics entitling it to be called unique.

And in the same way as Mr. Tregear's work is not a dictionary, but the dictionary; so Mr. Tregear is not an author of it—one of three or more—one who has built on the foundation, or enlarged the building of others, but the author, who alone in any true sense can claim, and therefore whose name alone receives, a place on the title-page.

It must not, however, be inferred from this that Mr. Tregear is under no obligation to others, or that he makes no acknowledgment of his obligations. In his Preface (p. xii.) he says:—" I have to thank certain authors for the advantages I have received from their works, and without which I should have been unable to present my dictionary in so complete a form. First of these valuable books is Williams' New Zealand Dictionary. Although I have made considerable additions to the stock of Maori words, the work of Bishop and Archdeacon Williams (father and son) has been the basis of my structure, as it has been for many years the authority and reference for all Maori and English translators. Its fidelity and usefulness page 5 "is so widely recognised that no word of praise from me would "raise it in public estimation."

He then names a large number of other authors and friends to whom thanks are due, and concludes with a merited compliment to his publisher, Mr. J. R. Blair. "To his enterprise," he says, "I owe the fact that I am able to present a technical book bristling with typographical difficulties in a manner the accuracy of which is a credit to the publishing "firm and to this young colony." The modest list of corrigenda issued with the book supports this view, since it shows only ten words wrong in 699 closely printed pages; and though that list will, as I believe, have to be multiplied many times in the Maori part alone, the compliment is well deserved; the errors which are fairly numerous throughout the book, seem to be rarely those of its printer.

I will pause here a moment to say that, if I understand-him rightly, it is easy to confirm the truth of Mr. Tregear's statement just quoted, that without the advantages derived from the works of the authors whom he thanks he would have been "unable to to present his Dictionary in so complete a form." Seeing that without these he would have wanted" practically all his Maori words, by far the greater part of all his other Polynesian and his Melanesian words, practically all his Polynesian and Melanesian mythology, not to mention other things, say perhaps 95 per cent. of his book, it is evident, unless I have seriously mistaken the position, that in saying without all this his dictionary would not have been so complete, he was well within the truth.

The acknowledgment, however, of obligation to other authors mentioned above, is subject to a most material qualification, which puzzled me greatly, but which, no doubt, is referrible to the same enthusiastic view of the whole matter of which I have spoken. In an earlier part of the Preface than that quoted Mr. Tregear, in a few forcible words defines the nature if not the extent of his indebtedness to others. He says (p. x.; the italics are mine):—" Farther on I have thanked those authors and those friends from whom I have received assistance. This refers to the raw material only. In collecting the vocabularies, in searching for comparisons, in making quotations for examples, in the compilation, in the whole of the philological and literary work I have been unassisted."

Evidently then, those who have contributed only the raw material, whatever the mere quantity of this may have been, could not reasonably expect to rank with the one man who has done all the skilled work in the book, all the philological and literary work it contains; whose skill and learning, that is, have converted the crude mass into the artistic article we have in the Dictionary; nor, therefore, can they or any of them complain that their names do not appear with his on the title-page. This page 6 view has, it seems, been tacitly countenanced if not consciously acquiesed in by those learned writers and speakers who in the Press and before scientific societies have given most laudatory notices of Mr. Tregear's book without so much as a reference to that of Bishop and Archdeacon Williams. Otherwise they could hardly have avoided mention of a work which has supplied, and commonly by the simplest of all literary processes, the great bulk of all the Maori words and meanings Mr. Tregear has given us in the corresponding part of his work. Or is it possible that these learned critics, or some of them, had not yet gone so far in their study of the Maori language as to know of the existence of the only Maori dictionary?^†

One word more upon the title-page. When I found as it seemed to me, partly from Mr. Tregear's admission, partly from comparison of the two books in many places, that he had in his enthusiasm borrowed and reprinted the whole of Archdeacon Williams' Dictionary, which, as I suppose, seemed suitable, all the words, very nearly all the meanings, and some of the examples, I confess I was surprised; the more especially as I had heard it reported, some two or three years before, that Archdeacon Williams was himself preparing a new edition. However, the thought occurred, that Mr. Tregear was perhaps inaugurating a new era in the literature of learning, anticipating somewhat that general social millennium which is no doubt coming, though not yet here, when a man's rights will be measured by his needs, and his duties by the needs of others, when he who has will give, and he who wants will take; and that as he had freely borrowed, so he would as freely lend; that his work, in short, would be at the service of any subsequent lexicographer who, desiring to make his own dictionary a little more complete, should have the wish and the industry to copy out Mr. Tregear's as a beginning. But then, looking at the foot of the title page, I was saddened to find the ominous formula guarding the copyright, and warning the hasty that the literary millennium at all events had not yet arrived, and, in New Zealand at least, and so far as concerned Mr. Tregear's work, was not, for the present, to be anticipated. I trust, however, this may not prevent Archdeacon Williams from re-borrowing his own, if, as I hope, he is in fact preparing another edition.

To return now to the last question proposed. Industry and enthusiasm Mr. Tregear certainly has: but these are of very general value and application; has he the special qualifications—not easily found—which are needed in the writer of a Maori dictionary which is also comparative? First, or among the first, of these is evidently a thorough knowledge of at least that one of the compared languages which is treated as the standard of comparison—in this case Maori. And this knowledge should, I presume, include both the practical knowledge of it, such as an educated Maori, one who knows the lore of his own people, has of it; and such a scientific knowledge of it, as a student of the science of language would gain who studied it in its relation to other languages of the same family, and if possible in the light derived from the habit of similar studies. By study of this kind I do not mean the looking through two, or more vocalularies for words spelt more or less alike, putting them down beside each other, and there stopping. This simple form of comparative philology in the case of languages, or dialects of one language, as closely akin as all the members of the Eastern Polynesian group are to each other, is, I can easily believe, a useful if not a necessary first step in the matter, but a first step only, analogous to the carter's or hodman's share in a building. What I mean by study here includes not only the first step towards a comparison but the effective comparison itself; it includes in particular an investigation of the structure of the language, an analysis of it so as to ascertain what may be called its fundamental facts, and a classification of these facts showing them, at least provisionally, to be subject to law, that is within the domain of science. Evidently also, one who would assume the place of a foremost authority and teacher of the language, is bound to know, as included in the foregoing, the best that has been written about it by those who have known it best, both practically and theoretically.

Has Mr. Tregear such a knowledge as this of the Maori language?

In the case of some authors—Archdeacon Maunsell for instance, or Dr. Shortland—their record would be a sufficient answer to such a question. If either of these venerable Maori scholars had published a Maori dictionary, no one, I am sure, would doubt that it would be invaluable. But I do not think I shall be contradicted when I say that, previously to the publication of his dictionary, Mr. Tregear had no such record. Not that he was by any means unknown as a writer on Maori and Polynesian subjects, and his writings showed an extensive use of vocabularies and much speculative philology, of which the predominant features were simplicity and boldness. But so far as I could judge, the discussion or exposition, on its linguistic side, was conducted hardly, if at all, upon the plane page 8 of scholarship—the test being, as it seemed to me, that learning, though it might give greater facility, was not essential for carrying on the comparative philological process or for judging of its results; the appeal was not to the learning of the reader, but to his eye, and his uninstructed imagination.

In 1885, some five or six years before the completion of his dictionary, Mr. Tregear published his two earliest works on these subjects: The Aryan Maori and The Maori in Asia. They at once made him famous among those who concurred in his views of the aims and methods of philology. Nevertheless I should not have referred to them, I should have thought it hardly humane to do so, but for the fact that the first and principal one is cited in this dictionary as an authority, and has a special "abbreviation" of its own for readier reference—an honor not conceded to such works as Archdeacon Maunsell's Grammar or Dr. Codrington's Melanesian Languages. No one I believe would be able to infer from these earliest writings or those which followed, that Mr. Tregear's knowledge of Maori was then either extensive or exact; grammatical questions were not dealt with, and—subtract Williams' Dictionary from them—there would be practically no vocabulary left. It may indeed be queried from the evidence those writings afford whether of what he had then learnt he had not a good deal to unlearn. If any Maori scholar or student doubts this I will ask him to look at them for himself. I shall here only indicate the nature of the proof by two facts;—First (Ar. M. p. 9.) Mr Tregear tells those "wholly unacquainted with the Maori language," that "the vowels are to be pronounced as in French, thus: mere like the English Mary: Kati as if written Kah-tee; and the u like oo, as patua, like pa-too-ah; * * * haere (the ae like English eye)," &c. On this I would say in passing that I understand some of the French vowels vary much according to accent and other circumstances, and it seems hardly fair to a poor person wholly unacquainted with Maori, even if knowing a little French, to turn him loose among; the several sounds to choose for himself without a hint to guide him. Long ago Archdeacon Maunsell compared the Maori e to to a French e, but that was the e of café. As to mere being like "Mary," I should say that the two e's of the former are exactly alike barring a slight stress on the first, while in "Mary" the a is certainly not like the y, nor is either like the short Maori e of mere. Again, if Maori u is the same as French u and English oo (of too), the French referred to must surely be of "the school of Stratford atte Bow," Lastly, if in Maori you know the sound of each vowel, you know the sound (apart of course from length and stress) of every possible combination of them; but supposing a pupil of Mr. Tregear's of the pre-dictionary days were to give a Maori recitation in which the vowels and their combinations were given (in some sort) their French sounds, would a Maori page 9 listener know he was hearing his own language? Secondly, among the proverbs of which he gives amended translations to make them support the Aryan theory of the book, is this one:—Me he toroa ngungunu, which Mr. Colenso, from whom it was taken, translates "Like an albatros folding its wings up neatly," Mr. Tregear's amended version (At: M. 77) is:—" Lest the bull bite you." Now, ignoring the unusual mode of attack suggested on the part of the bull, and allowing the author's subjective views of the form and meaning of Maori words 4000 or 5000 years ago, to stand for the objective facts he thought they might be, allowing also for the grave perturbation necessarily caused by the presence in an enthusiastic mind of a theory then about to revolutionize philology, I am still unable to conceive how Mr. Tregear's meaning could be got from these four Maori words. Possibly others may have more success. If not, the difficulty probably arises from the author's relying upon pre-Vedic Maori syntax as much as on pre-Vedic vocabulary: he gives us some glimpses of the latter; upon the former he is wholly silent.

Whether at the time of the publication of these earliest works he was acquainted with more than a few words of the other Polynesian languages does not appear; his later papers shew a more extended acquaintance, but of the same kind. His method also remained in substance unchanged.

The conclusion then seems to me clear that the proof of Mr. Tregear's qualifications for the great work he has undertaken—whether as to his knowledge of the languages, or of the elementary scientific principles involved in their investigation and comparison—is not to be found in his previous writings, but must be sought for in the dictionary itself. It will add greatly to the wonder of this last, if it appears, that, starting with the modest equipment disclosed in his earliest writings, he has in the leisure of five or six years, and while actually writing his dictionary, acquired such a mastery of the Maori language, and such a sufficient knowledge of the other languages and of the technicalities of his work, as were absolutely needed to enable him to do that work competently.

I propose now, in the first place, to offer some general criticisms upon the book, looking at it more especially as a Maori dictionary, and then to examine parts of it in some detail.

But first I will say a few words on a part of it not hitherto included in a Maori dictionary, but very rightly added by Mr. Tregear—the mythological. This seems to me the best part of book; in it the author has made a very useful beginning of what is essential to understanding not Maori history only but the ways of thought of the people and a large part of their language. But its value I think is seriously lessened by the fact that Mr. Tregear does not sufficiently sift his authorities and the materials they supply. The rule seems to be: A legend is a legend, and page 10 entitled to rank as genuine if it is printed in the works from which the author draws his information. This may he generally true, but it is inevitable that these stories should be of very varying degrees of value as evidence of the real and complete tradition and belief of the Maori people on the matters related, according to the persons from whom they came, and the times at which and the persons by whom they were written down: at least some effort should have been made to classify and sift them, and if this could not be done the reader should have been sufficiently warned. The class of most-learned among the Maoris was, I suppose, always a small one, and is now probably all but extinct; while even the half-learned are being rapidly displaced by the all-but-ignorant. Moreover, it is impossible that an intelligent and imaginative people like the Maoris should live for many years in what I may call the neighbourhood of the Bible (which they have long had in their own language), and beside, and in part amongst, an English community, without acquiring and assimilating a large number of new facts and notions, even new ways of thought, which would in part at least, soon seem old; and without learning and mixing with their old vanishing legends, events, and characters from histories, foreign of course in incident, but still more so in purpose and conception. Sir George Grey in his invaluable collection of Maori poems (p. 13) has preserved one in which the Saviour and Tumatauenga, the Maori god of war, hold the same position in successive verses. In 1862, at the raising of the King's flag at Mataitawa,^* the attendant ceremonies comprised some of their old karakia and part of the Church Service. The Pai Marire religion afterwards called Hauhau, first promulgated by Te Ua in the same year, had in the beginning its Trinity: "Atua Matua Pai Marire; Atua Tamaiti Pai Marire; Atua Wairua Tapu, Pai Marire; rire, rire, hau" Its prayers were an absurd jumble of Maori and English ^†; but were meant and taken with the utmost seriousness: I lost sight personally of its late developments. The history of any of the Maori "prophets," Te Ua, Te Kooti, Te Whiti, and others, would give abundant illustrations both of the conscious and the unconscious adoption of the foreign element. Some noteworthy examples are to be found even in that great storehouse of Maori tradition, White's Ancient History of the Maori.

The Maoris, I believe, have genuine traditions of probably several Moods. But in A.H.M., I. Chap. xii. there is another flood-

^* At which, I believe, Mr. Parris, Native Commissioner, and I, on his invitation, were the only Europeans present.

^† I had a copy of one given me by a believer beginning: "Porini Hoia Tewhera Teihana!" These are the English words (a little disguised), Fall-in Soldiers, Devils, attention I The quotation in the text is also from a Maori MS. of the time in my possession.

page 11 story, besides other stories of wholesale and purposeless destruction, and a number of two or three-line biographies of pious or wicked Maoris, all evidently by the same hand, and all evidently composed or completely recast in the light—if light it may be called—of half understood and dimply remembered Bible teaching and phrases; the names are Maori, the scenes and point of view sometimes Maori, sometimes pseudo-Christian, often a confusion of the two. In this Maori version of the Noachian flood when the human race is about to be destroyed, a few virtuous persons make themselves a great raft of totara and other trees tied together with supplejacks and creepers, and having on it a wooden house stored with fern-root, kumara and dogs for food—then they pray for rain upon their neighbours, and it is sent in such quantities that all men but themselves are drowned; they drift about on the waters for seven months when the flood begins to subside; during the eighth they land, offer sacrifices to all the gods, and lastly, see the rainbow in the sky, which concludes the matter. And what are the reasons for the flood? Such surely as no sane Maori could have conceived a century ago:—Men were wicked; they were in the habit of fighting; and they would not listen to the preachers of the doctrines of Tane! The "doctrines of Tane," as I understand, being the well-known story that he had lifted up the sky from off the earth on which it was lying, either with his shoulders, or by standing on his head and kicking it up with his feet! Whereas these sinful sceptics openly declared that the sky and earth were as they always had been; that Tane had actually done nothing; and the preachers might eat their own sermons for food—worse still, the very worst of all according to the historian, they might even eat the heads of their sermons—a great curse it seems, for though to an Englishman the heads of a sermon, however numerous, would seem a less serious matter than the sermon itself, to a Maori any hostile or contemptuous reference to a head in which he was interested, might be highly dangerous.

Mr. Tregear (p. 558) says of this story: "The most consecutive and valuable account of a deluge relates that evil being everywhere triumphant in the world, Parawhenuamea and Tupunuiauta preached to wicked mortals in vain, and that the holy doctrines of Tane and the teaching as to the separation of Rangi (Heaven) and Papa (the earth) were derided." On this I will say that Mr. Tregear must have been in his most enthusiastic mood when, improving on his original he wrote of the holy doctrines of Tane, whether these are to be taken as meaning narrative or incantations. According to the Maori view Tane is as much concerned with holiness, as dirt is like sunshine.

But Mr. Tregear's scepticism—I am afraid even his critical page 12 judgment—is not easily aroused. Fornander, bis great predecessor in the Aryan-Maori theory, and to whom in the matter of Polynesian mythology and otherwise, he is more indebted perhaps than the reader can easily gather from his pages, boldly takes fidelity to the Chaldean or Hebrew original as one test of merit in, among others, a certain Marquesan flood-story—Mr. Tregear apparently takes the same view though he does not obtrude it. He speaks of this same story as "by far the best preserved evidence of the possession of an antique belief in a great flood." Here again, as in Tane's case, it will, I think, be found that his enthusiasm has improved on the original, and Fornander's translation of it, by entitling the Lord Ocean (i.e., the sea whose overflow caused the flood) the "Divine Being."

Now, I will ask you to contrast with the sketch just given of what may be called Noah's flood in New Zealand, and especially with the reasons for it, a short sketch of what I believe in its main outlines to be a genuinely Maori flood-story, known as the Tide or Flood of Ruatapu. It is from the Ancient History, Vol. III. This Ruatapu was a mythical hero or demi-god, younger son of the more widely known Uenuku, but his mother was a captive and so had lost her rank. He had seriously offended Uenuku by using the latter's sacred comb on his own inferior head, and Uenuku had abused him for it in a most degrading way, calling him a man of no birth, a nobody, and so on. Of course according to Maori notions of honour and morality, such an insult justified if it did not demand some signal vengeance. But there was not the least need that this should fall directly upon the offender. It was certainly not necessary, and would not, I think, have been strictly proper, for Ruatapu to kill his father with his own hand; it was quite enough, and quite right, if he could punish him indirectly, at whatever cost to others who had not offended. He therefore built a large canoe, invited the principal young chiefs (to the number of 140) to go with him on its first voyage, and having got them far out to sea, he contrived to drown or spear all of them but one, who swam ashore, and whom he charged to caution the people there that he should be with them (to drown them too) in a certain month; and accordingly in that month he appeared as a great wave, which drowned all then on the land except those who, acting on his message, had ascended a certain mountain he had named. It need not surprise us that among those saved was Uenuku, the occasion of all the destruction.

Comparing the last story with the first it will be seen, that, different as they are in incident, there is a much deeper difference still. The motive, the formative idea of the one, comes from a moral world radically distinct from that of the other—and it is to a difference of this kind that I should attach most weight in deciding a question of common or diverse origin.

But though, as I think, Mr. Tregear is too ready to accept as genuine Maori tradition what on the present evidence I must hold to be mainly reminiscences of the Bible narrative of Noah's flood, it is only fair to say that he is not willing to accept Noah himself as an ancestor of the Maori people, though the name of the patriarch in fact appears in one of the Maori genealogies which he prints at the end of his Dictionary! The author's note on this is no doubt just. He says (p. 667.): "The introduction of the name of Noa (Noah) instead of Kaitangata as the father of Hema is only a foolish perversion of missionary teaching, confounding Hema with Shem!" This as I said is just. The fact shows that the "foolish" Maori had been arguing with himself (just as if he had been a young New Zealand philologist) that, as the name Hema was so like the name Shem, they must be the same name and stand for the same man; and (changing to the Maori line of argument) as the whitemen knew more than the Maoris about Shem, and they said Shem's father was Noah, his father must have been Noah, and not Kaitangata as the Maoris used to say when they knew no better. It shows also how much Mr. Tregear's readers may gain when he allows his critical faculty free play. And lastly, the fact that Noah's name actually appears in so solemn a thing as a Maori genealogy (than which to the Maori mind nothing required greater accuracy) proves the reality of the danger I spoke of—that of the intermixture of Maori and foreign lore.

One criticism which applies, to the book generally, is that the author has given us a very great deal of the work of others with far too little specific acknowledgment. I am looking at it now as a practical matter. The general acknowledgement he has made with its curious qualification, may or may not satisfy the canons of literary taste, his readers must judge when they have ascertained the extent of his obligations, and with these compared the ackowledgement. But the student's difficulty is practical; he is concerned to know as to each word and meaning, or generally as to each material statement, whose authority he has to rely on. He is told that Williams' Dictionary has been the basis of the work, yet I believe he will not find a dozen specific references to it; so far I have found five only. If ho has not the earlier work he will have no means of discriminating. If he has both, and compares them at all carefully, he will find that though commonly the transfer from one book to the other has been in the simplest form, or with only trifling alterations, many other alterations have been made, and of these he may think a considerable proportion are not improvements. In any case if he has a substantial knowledge of Maori and Maori matters he may estimate at quite different values the authority of Bishop or Archdeacon Williams and that of Mr. Tregear. The former he will know as Maori scholars; the latter, in this respect page 14 and for the present, he may well have a difficulty in classing.

It will be noticed that Mr. Tregear in his Introduction does not discuss, nor, I think, even mention the structure of the Maori language, or of the other languages he is comparing with it: and for this the student has good ground of complaint. He does not even say to which of the recognised families, or other great divisions of human speech these Oceanic languages in his view belong or arc related; or whether, as seems the better opinion, they, for the present at least, must be put in a group by themselves; he even makes it very doubtful whether about one half of the languages he deals with are in fact related to the other half. He tells us the Maori speech is a dialect of the great Polynesian language, but what the characteristics and affinities of this language are he does not say. Yet if he has so studied the Maori language as to have gained some considerable insight into its structure, why does he not give his readers the very great help it would be to them, at least to all of them of the student class?—especially those who come to his book wholly ignorant of the languages it treats of. If on the other hand he has not gained this insight into the Maori language, how can he reasonably compare it with other languages, or even its component parts with each other? As it seems to me, he would not know the units of comparison; he would bo measuring without a rule. But is he here quite candid with his readers?

Not many years ago if he had not gained much insight into the structure of Maori he had arrived at a very decided opinion as to its affinities. In the Aryan Maori published as I have said in 1885, treating as one the two distinct questions of race and language, he left no doubt whatever as to how both the Maori and his language were to be classed. In the Introduction to the work he says:—" I now proceed to assert, positively, 1. That the Maori is an Aryan. * * * 3. That his language has preserved, in an almost inconceivable purity, the speech of "his Aryan forefathers, and compared with which the Greek and Latin tongues are mere corruptions." * * * These expressions though strong could not be called hasty or ill considered, except in a sense applicable to the conception and execution of the whole book; they fairly represent its purpose and spirit; he adds:—" To prova these bold assertions is my task in the following chapters." The opinion thus forcibly expressed was as tenaciously held. At the end of the work he says:—This book contains doubtless some slight errors of detail, yet, I feel proud to have written it. Not yet have I seen one shadow of disproof as I went on; every step has confirmed and strengthened the one proceeding, until I feel so assured of the truth of my view of the origin of the Maori race, that if not one man in New Zealand agreed with me, I could wait with calm confidence for the verdict of the European page 15 "scholars." And he adds with the same frankness and force:—"I have boon the first to apply the scientific method to the "Maori language, and to prove the fellow-ship of the Polynesian to the races of Europe." But there must evidently have been something wanting. My own guess was, that his scientific method—(the method of vocabularies, the method, as it seemed to me, of looking into two given languages without regard to structure or grammar for words as much alike in form and as little different in meaning as possible, and treating these words as therefore related)—seemed to the European scholars too much like the royal road to knowledge. It was at least philology without grammar or historical research, if not rather philology without learning. At all events the verdict, though long as well as confidently waited for, never apparently came. Indeed, such slight expressions of opinion as I saw from known European scholars were, to put it kindly, not sympathetic. Perhaps it was looking on them as more than human to expect them to welcome a proposal which in reality involved a fundamental revision, not to say reversal, of the results attained by the labours for nearly a century of hosts of illustrious scholars. And why were they to do this? Because a gentleman in New Zealand, not then known to fame, had discovered the Maori dictionary, and how to use it.

Nevertheless, for three years or so after the publication of the Aryan Maori, up to at least October, 1888, Mr. Tregear maintained, with considerable insistence and in the face of some adverse criticism, the Aryan character of the Maori in race and language. Whether that belief was subsequently modified, and if so when and how far, I cannot say.^* Now during those three years a great part of his dictionary must have been written; it was finished some two years later. But must not an energetic belief of this kind even when less intense than at first, bias, if not determine his judgment on many vital questions in dealing with these languages? If so, the student should have been warned; in any case, knowing Mr. Tregear's former belief and its intensity, he should not have been left in doubt.

I will give an illustration; and in so doing will not speak of the structure of the Maori language, but, taking it as representing the Polynesian languages, will ask what are some of its leading characteristics or fundamental facts, as we know it? Here, in my opinion, are two of them, and I hardly know which to put first. One is, the very great, I would say, superior value of the vowels; as shewn (1) in their much greater frequency of occurence compared with the consonants; (2) in the extreme nicety required in their perception and pronunciation;

^* See an obscure passage in the Dict. p. ix.; and Aust. A.A.S., Vol, III, p. 851.

page 16 (3) in their great stability, especially the unaccented; and (4) in the essential character of the open syllable, every syllable being either a single vowel, or a single vowel preceded by a consonant. The other is, the prevailingly disyllabic character of the language: the monosyllables are necessarily trifling in number; and if we follow the far-reaching if obvious suggestion of Archdeacon Williams, that the long vowel is the equivalent of two short ones, the greater part of all other words, not themsevles disyllables, can be most satisfactorily resolved into disyllabic elements. It will be understood I am here only summarizing. Now assuming these propositions to have as yet only so much evidence in their favour as to give them a claim to be discussed, can Mr. Tregear discuss or weigh them fairly, if he has already decided that the language is of the Aryan family and therefore that he must look inter alia for monosyllabic, not disyllabic roots, and these by no means necessarily or even commonly open syllables, but often effectually closed, many genuine Aryan roots it seems showing two, or three, consonants to one vowel.^* And are we not driven to assume some such powerful disturbing cause as this, to account for the singular fact that the author of a comparative dictionary has passed over in silence, such obvious, such necessary, and I will add, such tempting questions?

A kindred and not less important matter, that of grammar, on which so many things might and so much ought, to have been said, is treated worse, I think, than by merely being passed over in silence. The student, looking at the Introduction, and noticing the attractive heading "Grammar," will at first sight be disappointed at the trifling space devoted to so great a subject; on a close inspection his disappointment will be still greater at the way that little space is filled. Mr. Tregear begins by protesting strongly, though to me I confess unintelligbly, against 'forcing rules of grammar upon these languages, and—excepting a singular remedy he has tried and some noticeable words upon accent and pronunciation, of which I will speak separately,—with that protest he ends. "I have," he says, "carefully avoided the use of letters to mark the native words as substantive, adjective, verb, &c. It is an unwise if not a mischievous, effort to make if wo endeavour to force the rules of grammar which fit (more or less) the modern stage of the English tongue upon a language belonging to the utterly unequal grammar-period in which the Polynesian speech is now found. I

^* See the list of Aryan roots given in Skeat's Etymological Dictionary; and compare the list of Sanscrit roots discussed in Max Müiler's Science of Thought. But see Sayce Principles of Com, Phil 104, &c. It is not to be assumed that the disyllabic Polynesian radicals may not ultimately be reducible to monosyllabic roots; but I say it would not be reasonable to begin the investigation by ignoring the one fact which is perhaps the most obvious, and the most frequently recurring of all that have to be dealt with in Eastern Polynesian languages.

page 17 use these expressions with consideration, because I believe that there is a constant progress or decay in all languages, affecting their character and rendering their forms unsuitable." He adds that this is certainly so with regard to English, but probably less so with Polynesian, and concludes with a passage I will cite a little farther on. Now so far as I understand this it means that it is unwise, if not mischievous, to force unsuitable rules of grammar upon the Maori language, and that one principal way of doing this is to mark its words, with letters, as being what in English are called the several parts of speech; and the practical remedy he applies is the simple one, surely unique, of carefully leaving these obnoxious letters, and the words they represent, altogether out of his dictionary. Whether the mischief would be as acute, or on the other hand would fee lessened, or removed if the marking were done by words at length instead of by single letters he gives us no hint; judging from his practice, however, I should infer that the danger would be somewhat less if words at length were used, perhaps a little less still with a periphrasis; hence we find in this dictionary "i, an interjection," "ma, the conjunction," and "te, singular definite article," "nga, the plural article," and a few others; but more frequently the safer periphrastic formula, "a word used as a verb," &c., e.g., "he, a, an: a word used as an indefinite article," "kai, a prefix to words used as transitive verbs," &c., But putting these all together their number is inconsiderable, and therefore I conclude, not without hesitation, that though marking by letters is most to be avoided, the protest againt any overt attempt to distinguish in a Maori dictionary between the different parts of speech is intended to be general.

The great offenders in the matter of marking native words in this way are of course our grammarians and dictionary-makers, and especially Bishop and Archdeacon Williams, who have taken the trouble to mark every word in their dictionary as noun, verb, &c., and in the most objectionable way, by letters, as a. for adjective, n. for noun, often even giving one word, two or three such marks where it has commonly two or three grammatical functions; worse still, perhaps, Archdeacon Williams has gone farther and has not only marked the verbs as verbs, but refining still more has actually distinguished the transitive from the intransitive, using again the objectionable symbols v.t. and v.i.

I will not stop to discuss this matter, beyond saying a word upon a compensatory loss the student may often suffer for being relieved from the undefined if not undefinable danger which Mr. Tregear so much fears. It is upon the last named distinction between neuter and active verbs. Now it seems to me that the writer of a dictionary who is relieved of the obligation to mark such distinctions, must be very apt, especially if working against time, to relieve himself of the obligation to notice that page 18 any such distinctions exist. This two-fold relief would make it much easier, I suppose, to write a dictionary, but very much harder to use it with advantage. Take one of the commonest words in the language, korero. Arch. Williams gives this as verb transitive, to tell, say, address; and intransitive, to speak, to talk. Mr. Tregear having this as his 'raw material,' but having as matter of conscience relieved himself of the need to mark the distinction between the two sets of meanings, goes one step farther and ignores the distinction itself, and so omits the very common intransitive meaning, speak, talk, altogether; yet his next word, korerorero, the frequentative form of this one, shows only the intransitive meaning given it by Arch. Williams, and in his words, but omitting of course the dangerous v.i. It reappears also in his edition of Arch. Williams' Engilsh-Maori part, but there it is "too obvious to be overlooked. This is not at all an isolated case.

Now, I could quite understand a scholar in his very early years longing, if not publicly advocating, that many troublesome grammatical distinctions should be wiped out of his grammars and dictionaries; but he would soon find the reform if carried out, practically useless, if he could not carry it one step farther and abolish the distinctions themselves; or at least convince his masters that they did not exist. In Mr. Tregear's case comparing his words with his practice, it seems impossible to give a statement of his views not self-contradictory; I cannot say whether he does or does not deny the existence in the Polynesian languages of the grammatical categories of noun, verb, adjective and so on—or rather whether he does not both affirm and deny.

So as to syntax; his position seems to be that it may be learnt but cannot be taught; that no useful rules can be enunciated, but nevertheless may be acquired. "The effort," he says (p. xiii), "to adapt Maori words to rules of English grammar is evaded by the complex simplicity (if I may use such an expression) of the native language, where one word may serve either as verb, noun, or adjective, according to its context, and wherein particles whose use only practice can render familiar, are able to link words into sentences capable of rendering very subtle and sensitive expression. If we attempt to retain these particles in the net of English grammar, we shall be in the unpleasant situation of having to lay down "rules with more exceptions than examples." We may all of us, though not Maori scholars, agree that practice alone can make one famililiar with the use of the Maori particles, if only as a particular case of the universal rule embodied in our excellent copy-book apophthegm, "practice makes perfect." But, as I understand the passage, its main argument is that Maori, being an analytic language, is particularly unsuitable for page 19 the application of English rules of grammar. Probably, however, so far as the "argument is relevant, the contrary would be nearer the truth, since English in its present state is said to be one of the most analytic languages known.

But who wants to force English grammar upon the Maori language? Mr. Tregear compels us to ask:—Is a Maori grammar possible? And what does such a question mean? Surely it is equivalent to asking, whether Maori is really a human language, or merely a confused babel of sounds? Have the Maoris settled names for things, actions, qualities and relations? Can they frame and qualify propositions so as to convey to one mind the thought or feeling that is in another? Arc the facts of the language capable or not of statement and classification? and do the recurrent facts of the language—of the language in action—recur not at random but according to law? If yes, then the classified statement of these facts and laws is, I presume, Maori grammar.

No one would say that our existing grammars are beyond improvement—that the last word, for instance, has been said upon the Maori verb—and it was open to Mr. Tregear if not incumbent on one in the position he has assumed, to give us fresh light, whether in the form of new facts or new views, on this or any other grammatical point he found wrongly, incompletely or obscurely stated; and such facts or views would have been most welcome. But it is surely idle or worse to represent Maori grammar as practically impossible, because one word may at different times discharge different grammatical functions, or because grammatical relations are mainly expressed by shifting particles and prepositions and by the order of the words—that is, because the language is not inflectional and is analytic.

It is in curious contrast to his opinion expressed and implied on Maori grammar as it exists, or rather 1 suppose I ought to say on Maori grammars as they exist, to note a point in his practice. In a few places, when he gets into what may be called a grammatical corner, he gets out of it in a way which seems to me as näive as it is effectual. In his article on ai, the relative particle, he begins:—" A particle having no "English equivalent, and only to be understood by reference to, a Maori grammar"; that on ano begins still more directly "ano [See Maori grammar]; "while in that on ko the same formula comes in later. It will be observed, first, that these are three of those very "particles" to which Mr. Tregear's protest particularly referred; secondly, that he shows no preference for one grammar over another; as the condemnation was general, so too, to this extent, is the rehabilitation.

Would it not here be pertinent to ask: What light does Mr. Tregear's Comparative Dictionary throw on the peculiarities page 20 of the Maori language? How many of its idioms has he elucidated? I certainly will not say, not one; but" I could not name one.

There is perhaps nothing more unsatisfactory in the author's Introduction than his treatment of the sounds of the Maori language, a matter of course of the very first importance. I mentioned that formerly he said the vowels were to be pronounced as in French. In his Dictionary (p. xiv) he gives another view: "The pronunciation of the vowels as printed in Maori and in all Polynesian writings is nearly that used by the Italians." But, in the first place, if wo are to assume that he knows accurately both sets of sounds, the Maori and the Italian, why not tell us in what respects they differ, especially in reference to his uncancelled direction to pronounce the Maori vowels as in French? Again, he says, "a short, almost like the English u in smut"; but the more valuable information would be, what it is quite like. No doubt to an English ear of average dulness a very short Maori a in some situations sounds like a 'neutral vowel,' but careful examination will, I believe, convince others as it did me, that every Maori (and Polynesian?) a is of the quality of a in English father, though as to length they would have to be divided into three classes at least. There is, I believe, no neutral vowel known to the Maori language, except through the medium of English speakers of it, who have learnt it late in life, or at least not in childhood; and excepting, of course, those who though they may use words and phrases with more or less fluency, cannot, in any proper sense, be said so have learnt it at all. The short form of what may be called the fundamental and universal a seems to be peculiarly distasteful to modern literary English, but it is the dominant sound in Maori, and no doubt in all other Polynesian languages.

The same rule may, I believe, be stated generally of all the Maori vowels, that each has one sound only, though of different lengths.—(Williams, First Lessons, 3.)^*

We have seen that formerly Mr Tregear likened the short e to both a and y in "Mary"; he now likens it to e in bent, and gives long e as "resembling the a of Mary.' It is difficult to fix a limit to 'resemblance,' but certainly, to my ear, neither short nor long Maori e has the same sound as the a in that well-known name as usually pronounced. Another serious mistake of Mr Tregear's, as I hold it, is as to the short o of Maori which he equates to the o of English lock. This is precisely the error against which Archdeacon Maunsell warned his readers 50 years ago, saying that the sound did not exist in Maori. It has, indeed,

^* See now also the similar view of;another high authority, Mr. T. H. Smith, for a long time the permanent chief of the Native Office, and afterwards Judge of the Native Land Court: On Maori Nomenclature, a valuable paper lately read at Auckland.

page 21 been said to be an almost peculiarly English sound, and is only heard in Maori, I believe, from the English speakers of it mentioned above. Tried by the rule just proposed, it will be found that it is not Maori because when prolonged it does not make or come near Maori ō, which is the same as the first vowel sound in the English no or go; that is, the pure o, without the glide of u commonly following it in English.

Of the Maori consonants, he says that "they have nearly the same power as in English." Here, again, if he knows that they, or some of them differ, why will he not tell us which of them, and how? An approximation is much better than nothing, but from one who has assumed the position, and with it the duties of a master, the student is entitled to know the sounds exactly, or to know why they cannot be given. Mr. Tregear mentions only three or four of the consonants. Ng, he says, is like ng in flinging, but he leaves unnoticed the beginners only difficulty over it—the main difficulty of which he is conscious in Maori pronunciation—where ng is initial. Of r and p the author says. "It is probable that formerly in some localities the r varied into l and d, the p into b etc., but the efforts to educate the Maori children in their own language have resulted in the production of a classic form in which the r and p are distinctly r and p." This is a conjecture, turned by tacit assumption into a fact, and the present non-existence of this assumed fact accounted for by an inadequate assertion. It is, I think, common knowledge among those who lived in the North Island 80 or 40 years ago, that many English speakers of Maori, especially the less educated, often used g, d, b, for k, t, p, when the Maoris, whom they thought they were imitating, used the latter only. Compare what Mr Tregear tells us about the Maori pronunciation of r, consisting solely of his conjecture how it was pronounced in some places not named, and at some time not named, with what Archdeacon Maunsell wrote about it in 1842. It has, the latter says (Gr. p. 8), two sounds, one rough; the other "is more soft, and is formed by a gentle jar of the tongue against the palate; so gentle, indeed, is the vibration that most foreigners pronounce it like d or l.' Have we not here an anticipation and answer to this particular conjecture of our author's half a century before it was made? There were then, as there are now, two distinct r's—and of the one which seemed to vary, the variation, it seems, was rather in the ear of the foreign listener, than on the tongue of the native speaker.

There are two Maori consonants, h and t, which, though not noticed by Mr. Tregear, do differ, the one in some places or in some words, the other always, from the same letters in English. By far the most common sound of Maori h is that of English h. But early visitors to the North of New Zealand often represented Maori h in some words by sh. Archdeacon Maunsell page 22 while protesting against this, speaks of the sound as a gentle sibilation preceding the h. Mr. S. Percy Smith again, suggests (Trans. xxii. 1)0) that the sound is better represented by something between s and y, which would, perhaps, be difficult to define. It has long seemed to me that the nearest representation of' this peculiar sound is a very short i or y following the h. The common cry of welcome for instance, "tiara mai, Hara mai, Hara mai!" I have often thought sounded as if it should bo spelt hiara mai, the i being only just perceptible, and not preventing the accent from falling on the first a. It might, perhaps, be called an aspirated y. The sound is much commoner in the North of New Zealand, but is not, I believe, entirely confined to it. The other point to be noted about h is that the Taranaki people commonly drop it, substituting a catch, which is here effected, I think, by a momentary cessation of the breath, thus, I suppose, giving the soft breathing for the rough,

The Maori t is, I believe, distinctly and always different from English t, the point of the tongue in pronouncing it being brought more forward, so that the sound while still t is on the way towards th. (See M. gr. 9, and William's First Lessons, 2,)

The inference I draw from Mr. Tregear's treatment of the sounds of this language, is that he has neither listened with careful attention to the pronunciation of the Maoris themselves, nor read the best that has been written upon it.

The laws of accentuation in Maori are not yet anywhere fully stated, nor has there been, I believe, any considerable discussion of the principles upon which the statement of these laws should be made. One thing, however, seems obvious, they ought to have direct reference to the disyllabic character of the language. From the author of a comparative dictionary, who has surveyed so many languages capable of throwing light on each other, we are entitled to expect something of value on all unsettled vital points, if merely a summary of how things stand; but Mr Tregear contributes nothing to this discussion, even as to Maori alone, to say nothing of all the others. I do not think he has yet realised either the importance or the complexity of the problem. What he has to say is comprised in five lines (p. xiv), to the effect that in his dictionary he has used a certain accent to "denote a lengthened stress upon the vowel so marked" (e. g. màra): that through inadvertence in a few cases this accent has been turned the wrong way, (not a serious matter, seeing that it has no other meaning, and that no accent whatever is used in ordinary Maori printing and writing), and he concludes: "Some writers of Maori prefer a double letter as maara, &c., but this is misleading, as the sound is not that of two distinct "vowels. In all cases where accents arc not used, the first syllable is more strongly marked than the others, although not with the lengthened vowel sound."

That is all. It tolls us curiously little on such a subject, and it assumes two things: First, that the author in his dictionary has marked all the long vowels that occur and no others; and secondly, that, at least in all unmarked words, there is only one accent which concerns the reader. Unfortunately, neither is true. It is hardly an exaggeration to say that Mr. Tregear depends entirely upon Archdeacon Williams' edition of the dictionary for the length of the vowels. Yet in the first 50 pages of his own dictionary he has, according to my counting, left, in principal words, more than 80 long vowels unmarked which are marked long by Archdeacon Williams, and has, moreover, in the same space, introduced a considerable number of new words, of which hardly any are marked, though certainly some—I think a good many—should be. I have not attempted to count the total number in the whole book of errors of this class—omissions to mark the long vowels (as promised), or marking short ones as long—but there arc certainly some hundreds. Yet to alter the length of a Maori vowel so as to make a long one short, or a short one long, is to alter seriously the character, and often the meaning of the word; it is equivalent to the serious mis-spelling of a word in English.

Take one case out of the multitude. Archdeacon Williams gives tiratū as "mast of a canoe," and the vowel lengths indicate at least an obvious etymology: tira, stick or pole, tū, upstanding. Mr. Tregear's version makes the first vowel long instead of short, and the last one short instead of long, thus also shifting the principal accent from the end to the beginning of the word. Then, undeterred by the change he has made, he proposes the above etymology, which, quite consistent with Archdeacon Williams' marking of the word, is as obviously inconsistent with his own. Then next he suggests a comparison with tīrau, which would, so far as form goes, be quite reasonable if he is prepared to stand by his own form of tiratu and abandon his proposed etymology, but, unfortunately, under tirau he abandons instead his own form of tiratu, and adopts that of Archdeacon Williams. The reader must decide how much of this, here and elsewhere, is due to misadventure, how much to carelessness or want of time to be careful, and how much to want of belief that care as to vowel length is essential. I venture to say that if not interested enough to investigate the general question, he should for the sake of his readers have investigated the origin and value of the long vowel, a subject as interesting theoretically as it is practically important. He would have found some useful information in Archdeacon Williams' Introduction.

With regard to there being, apart from the long vowels, only one accent of importance in a Maori word and that on the first syllable, it would, I think, be nearer the truth to say that in every Maori polysyllable there is at least one accent to every two page 24 syllables. But the closer and more reasonable approximation, I submit, would be to begin with the disyllable, and to say that as a general rule in Maori every normal disyllable (of two short syllables) has an accent on the first, and keeps it there however much the disyllable itself may enter into composition, whether in the form of partial or complete reduplication, or with other radicals; that is, each element of the compound keeps its own accent. I say nothing now of the exceptions to the rule—some of much interest—nor of its other qualifications.

But only to ascertain the real facts of accentuation, and all the other facts of pronunciation, we want a combination of Maori scholars, each with a keen ear for small differences, working in different districts, but on a concerted plan. Their work would be of the greatest value.

Ought we not, however, as soon as ever it is possible to take the broad practical hint which Edison has given to all interested in languages or language throughout the world, and employ the phonograph in recording living specimens of all these languages? We should then have a really scientific basis to start from.^†

One word more on accent. The European student was promised that by means of this dictionary he would be "in a position to compare the Oceanic tongues with the languages of the great continents." But supposing him to remember the part which, we are told, accent has played in the history of some of those continental languages, would his curiosity as to the laws of accent in all the Oceanic tongues dealt with in this book be satisfied by knowing that in one of them, Maori, Mr. Tregear has marked the long vowels—or a good many of them—and that in unmarked words the first syllable is paid to be more strongly accented than the others? Or would he want to know more on this point, as on some others, before he enters on this new and extensive field of comparison?

Mr Tregear gives a good many examples to illustrate the meanings of the Maori words in the dictionary, and this has involved great labour on his part; but then there are a great many words to be illustrated; and that many have suitable examples does not compensate us for the many left without, especially those new ones which, from being imperfectly defined or otherwise ambiguous, or of doubtful form or authenticity, have most need of them. Moreover, of the examples which are given, a large proportion are comparatively valueless from not being translated; more especially where as frequently happens they are attached to the wrong word. The author gives a good many proverbs as examples, most of them apparently from a paper in Trans, vol. xii by Mr. Colenso, who there translates and

^† Is this not worth the attention of our newly formed Polynesian Society?—a society which has a future before it of much promise, if only it be supported from without and from within as it ought to be.

page 25 comments on them all; and where these are both apt and intelligible, nothing could be better. The Maoris, as is well known, are great makers and users of proverbs and proverbial sayings; but from this very fact and the isolation of their old life, many of these are local, and are unintelligible, or liable to be taken in a wrong sense, without local or special knowledge, or unless at least some note of explanation is added. I do not say that every example needs translation, or ought to have it. Some are too simple, some may be better left without it or put into Latin (indeed, some now to be found ought, in my opinion, to be expunged),^* but in a large majority of cases if they are properly chosen, good idiomatic bits, the translations, or necessary helps for translation, would form a valuable addition to the book merely as a dictionary. Whether or not the sentence be the real unit of speech it is certainly true that many words are best explained, if not defined as parts of a sentence. Moreover, the practice of never translating an example is not only a negative, but a positive evil; and in this particular case, I believe, an evil of magnitude. It allows and fosters any tendency there may be in an author to crude and hasty work, since it indefinitely lessens the conscious need for precision of thought and expression; and takes away the readiest means of finding himself out in an initial error.

I would ask what does this book do towards promoting the comparison and study of the Maori dialects? In 1842, only two years after the foundation of the Colony, and when learning of all kinds in it might be thought to be in its infancy, the Rev. R. Maunsell (now Archdeacon of Waitemata) published at Auckland his Grammar of the New Zealand Language, which though open I dare say to criticism on some points, is still as a whole unrivalled and invaluable. In his Introduction he enumerates seven leading dialects spoken in the Northern Island "independently of minute and numerous sub-divisions "; and he notes, with true scientific insight, the importance of preserving a full record of the dialectic differences; because, being even then clearly of opinion that this island had not been colonized by a single Maori migration, he thought that by means of the dialects, the sources of the several migrations might perhaps be traced; instancing the dropping of the h in Taranaki and Rarotonga, and the substitution in the Bay of Plenty of n for the ordinary ng: he does not mention the Hawaiians, but their vocabulary shows the same substitution,

There is another fact, as I think it, which goes to support the same line of research. Some words in Maori, no doubt a

^* The "realism" of genuine science is as suitable for publication as anything else scientific. In a dictionary it must often be exceedingly difficult to draw the line, and all allowance must be made for difference of judgment. But I think a minimum test is, does a passage otherwise objectionable advance the matter in hand, our knowledge of the language or the people?

page 26 good many, seem better explicable by direct reference to the language of some other Polynesian island than by anything in Maori itself. I will only give two illustrations. Compare taha-tai and ta-tahi both meaning "sea-side." In the first, both elements taha "side,"' and tai "sea" (qualifying the other) keep their ordinary Maori meaning: in the second while ta (= taa the vowel being long) may be reasonably taken as = taha of which both vowels are short, tahi is not used for sea in Maori, but is the Tongan form of Maori tai, "sea." Again take tae-kai, "an abandoned cultivation."^* In Maori tae is not used in any sense which would explain this word, but in Tongan it appears as a privative particle or negative prefix= English un before a word, or -less after it, so that tae-kai our "abandoned cultivation," or "worn out soil," may with the help of Tongan be translated "no-food," "food-less," which may be held good till a better appears. No doubt the common Maori negative te (the vowel being long=tee)=(in origin) the Tongan tae, though its use is not quite the same, and it is in Maori conventionally treated as a separate word. Now, Tongan, I suppose, is not so nearly related to Maori as some others are, Tahitian, for instance, not to mention Rarotongan; I only took these two words as first occurring to me. But if we could sift out a considerable number of such words, and could as it were localise them among the Maori tribes, we should be getting another clue which might be of value in two or three ways,

Another considerable advantage of having the peculiarities of each dialect fully recorded is that they might then be brought into one field of view, and so compared inter se. The result would be of considerable interest phonetically and otherwise.

What then has been done to record these dialectic peculiarities? Bishop Williams made a good beginning; in his dictionary (2nd ed., 1852), he marked a large number of words, distinguishing six dialects. Archdeacon Williams' edition (1871), omits a great part of these distinctions in the first part of his dictionary, though still marking there some of the most peculiar, those of the Rarawa dialect, and giving many others in the second part. Mr Tregear's contribution to the work has been merely negative; he has, as to the dialects of the North Island (i.e. nearly all the Maori dialects), wiped out even those distinctions which Archdeacon Williams had preserved. He does not say he is doing this, nor indeed does he say a word about it, that I have found, except in the last two lines of his five or six line essay on the pronunciation of the Maori

^* Mr. Tregear gives a substantially similar meaning, "worn out soil. Ho does not name his authority, but it is probably Mr. G. H. Davies, of the Native Office, from whom 1 received this meaning of the word, and to whose kindness I am greatly indebted for many others, as well as for the loan of valuable Mss.

page 27 consonants, which perhaps indicate his reasons. He there says (Introduction p. xiv), "The pronunciation varies slightly with locality, thus tangata is in some places tanata, but these irregularities of the sub-dialects are very fluctuating and unfixed." This, for the information it contains, is about as close an approximation to nothing as is possible, while if offered as all that need be said on these dialects ("sub-dialects," as he not unreasonably calls them), it leaves the (ignorant) reader much worse off than if he had merely nothing: he is, I venture to say, gravely misled. If Mr. Tregear has studied all these dialects so well and so long as to be qualified to say what he has said, how could he find it in his philologic conscience to say so little?

This treatment of the Maori dialects would be paralleled on a larger scale if we were given a general Polynesian dictionary which did not show from which islands the several words came. But there is one dialect which he does in part distinguish, and which he calls the South Island dialect. It is that of the Ngaitahu or Kaitahu tribe, forming the majority of the Natives in this South Island. Its great peculiarity is in using k for both ng and k of the North. Though included in the sweeping two-line condemnation I have quoted as to the dialectic "irregularities" being "fluctuating and unfixed," he gives many words from it, and often marks them. Yet some, and these the most peculiar, represent, I venture to say, only a misapprehension of the facts. According to those who know them well, the Kaitahu in speaking the pure dialect always, as I have said, use k for the northern k and ng. But since all printed books and all North Island correspondence, including the official, use both letters, the Kaitahu in writing also use both, but not having their ear to guide them, use the ng at random. Hence occur such monstrosities as tingaka for tikanga, and hundreds of others. Some of these Mr. Tregear takes seriously, giving them a place in his dictionary; while the great majority are, without a word of explanation, capriciously as it seems, ignored, Mr. J. White in his Ancient History, thought it his duty to print his Kaitahu manuscripts as he found them, and not as he thought they should have been—a course having great advantages, but necessitating some form of commentary. I spoke with him about it, and pointed out the obvious danger there was of persons ignorant of the facts being seriously misled. He did not deny it, and would, I believe, if he had lived to complete the supplementary volume he intended, have added a cautionary note.

I have, later on, pointed out another source from which, as it seems to me, Mr. Tregear has drawn some of his new words—not a great number perhaps, but still too many, and some of his most notable. To distinguish it from the Kai-tahu I have ventured to call it the Kai-ta or Printers dialect, since all words page 28 of this class class owe their origin, or at least their perfected form, to the Press, and the highly original views it often takes of the Maori language. It will, I think, be found that some of the most surprising of Mr Tregear's new words, involving new departures in the language, are to be credited to that source.

I shall not say much on the comparative part of the work, though the subject is of first class importance, and though much might be said as to the author's treatment of it—what he has done, and still more, perhaps, what he has not done. Indeed, I think the most general criticism to which it is open is that it has been published before it was ready—at least one stage too soon.

"Whether to arrive at the older meanings of individual words, or to get a scientiffic insight into a language itself, the incalculable value, not to say the imperative need of comparison with related languages, is too obvious, if not too well known, to need another slenderly qualified advocate. But it was not always so in New Zealand. Five or six years ago, at a time when Mr. Tregear thought that his claim to rank the Maori language as an elder sister of the Aryan family could be established upon an examination of Maori alone of the Polynesian group, such a comparison with closely related languages, and between its own dialects, was suggested to him,^† as a necessary preliminary to the very much wider comparison he was attempting. It was suggested also that the laws of phonetic interchange between the compared languages must first be ascertained; and that a systematic examination of the structure of Maori words was needed to distinguish their radical and non-radical parts. But these suggestions were only from the outside. If they came to Mr. Tregear's attention, their doubtful source may well have obscured any little merit they had in being true. He kept on his own course. Later, however, it seems, though still "some years" before 1890 (see Trans, xxiii, 531), when, "wishing to compare certain Maori words" with those of languages "used on the Continent of Asia," he was assured by "one of the masters of modern philology that the Maori tongue was not in a position for comparison." Apparently, what may be called the family comparison had first to be made. Moreover, "the European scholar "—accustomed to very different things—" needed a long and special preparation before he could grasp the mode or comprehend the genius of apparently simple tongues." It remained therefore for some Polynesian scholar to arrange and put into a form easy to manipulate, and to be comprehended at ft glance, the various related words used in the different island groups." This was authoritative. It pointed out clearly the work to be done, and, presumably, though not quite so clearly, the hand to do it. One thing only must have been omitted—but that

^† Trans, N,Z.1., vol. xix. pp. 554-6.

page 29 was a great omission—how it was to be done. The Polynesian scholar was to arrange the related words, but how was he to know which were related? This question was among the suggestions made to Mr. Tregear in 1886 (Trans. xix, 554), but he put it aside then as, no doubt, to him too obvious for discission, and it has not, so far as his practice shows, troubled him since. He might, then at least, fairly say that the eye which, without light from historical or grammatical research, could see the relationship of a Maori word to a Greek or a Gothic word, might certainly be trusted to see whether or not two Polynesian words were related. Hence, although I should have thought this question, in some form, a principal question, necessarily and fully to be discussed in the introduction to a comparative Polynesian dictionary, Mr. Tregear does not think it even needs mention.

Hence, also, I think two things follow: On the one hand, although the author has been working with great industry in a perfectly legitimate and useful field, his results are not of one half the value they ought to be; on the other hand, as I have said, his work has been published at least one stage too soon. He has brought together a great mass of materials for comparison of all degrees of value, but these have surely been sent to the printer before being submitted to the comparative philologist; the comparison itself has yet to be made. He puts words side by side without remark, and says "compare them"; and in 99 cases out of 100 that is all; but that by itself, I apprehend, is no more "comparison," than it would be if a comparative anatomist were, without comment, or direction, or reference to previous instructions, to tell his pupils to compare two distinct animal forms which he put before them, and of which the degree and particulars, or even the fact, of relationship had yet to be determined. The author's contribution to the comparison is the bringing together of so many words; but on what principle are these words selected? He does not tell us. Each reader, therefore, must guess for himself. Is it any other principle than that I supposed: to find words as much alike in appearance as possible, and not too different in meaning, and to treat them as therefore related? Even this, in languages as closely akin as some of these are, will often bring related words together. Classification, no doubt, is founded on likeness, but it is likeness in essentials; and beginners at all events must bear in mind the truism that only things comparable must be compared. Suppose, as is constantly happening, the student has to compare simple with compound words, or compound words with each other; how is he to analyze these compounds, or even to know that they are compounds? Ho must know in every case, or he will be but blindly guessing; and that means he must first study the structure of the language—how the words are built up. What has page 30 guided Mr. Tregear in all the work of this kind he has done? He does not tell us; as in his first work, so in this, anyone interested to know must, if he can, infer from the author's practice the principles on which he works. In practice he quite commonly—inevitably, as I should say—though never, I believe, expressly, recognises the disyllabic structure of the Polynesian languages: in making his "comparisons" he quite commonly ignores it. At one and the same time, in a sentence of a dozen words he will recognise and ignore the composite character of a word. And at the same time he will ignore the wholesome rule that the etymologist has to dispose satisfactorily of the whole of the word he is dealing with. To take one instance out of many. He compares tāiri with iri and mōiri, on the unavowed, but in my opinion true, theory that iri is the disyllabic radical, ta and mo being prefixes, in these words obsolescent as to meaning, if not yet obsolete. That is a distinct, if unconscious, recognition of the radical of two syllables, and of the principle of analysing Maori words on that basis. Then in the remainder of the same short sentence he ignores this principle which he has just recognised, treats tairi as one homogeneous word, forgets mōiri, forgets that the "of tairi is long, and that, moreover, the word is of three syllables, and so compares it with tare, which, having only two short syllables, has probably as good a right to the character of radical as any other word in the language; so that, therefore, if it was to be compared at all, it should surely have been with iri, its fellow-radical, and not with the composite ta-iri. But then iri and taw are not much alike; while if you did not spell ta-iri, taairi (as you must to give a Maori its true sound independently of his previous knowledge of the word), and stand far enough off, not to be inconvenienced by the superfluous middle i, then, applying the dangerous, it convenient, maxim, so often, I believe, exemplified in this book (though not yet recognised in our standard philological works), de minimis non curat philologus, "the philologist does not concern himself with trifles," it will be seen that tairi and tare are quite presentably alike. A little later tare is compared with tarewa; yet of this, rewa is treated, though by no means named, as the radical, and ta, with a long vowel (=taa) is again the prefix; so that, according to my view, and what I may call the greater part of Mr. Tregear's practice, the comparison, instead of being between the composite tāiri, tā-rewa, and the simple tare, should have been (if at all) between the three radicals iri, tare, and rewa.

But though, in his comparative work, the author relies, almost always, on the simple juxtaposition of words, or this introduced by the symbol "cf"—which, liberally enough, he translates "compare (confero)"; thus, as it seems giving his readers a choice as to mood and person—in a few rare cases Mr. Tregear goes beyond it, and gives the student something bolder. Thus, page 31 under Itani, water, is this:—" Note: Unlikely, as at first sight appears, the Maori word hani, water, is a compound of ringi, to pour out; r changes with n often in Polynesian dialects, as Tongan nima, five, with Maori rima five. Thus the Hawaiian nini, to pour out=the Maori ringi, to pour out; and hanini= haringi. The Maori word ngongi, water (ngo-ngi) may also be a compound of ringi to pour out." Assuming hani to he a Maori word for water, as I am prepared to believe on any sufficient authority, the argument seems to be this: because Hawaiian nini=Maori ringi, pour out (as no doubt it does), therefore [Hawaiian] hanini (which exists) = [Maori] hariagi (which does not exist); and therefore Maori hani=a compound of Maori ringi. I confess that I am unable to follow this; though I, by no means, refuse a cheerful, if qualified, assent to the maxim on which it seems to bo ultimately founded—(I mean the American version of an older form)—"things which are equal to the same are equal to everything else."

Mr. Tregear's treatment of this word is a fair example of his method whenever, leaving the simplicity and safety of "compare (confero)," he takes a bolder course, His "philology," as of old, seems to be always pro hâc vice. He certainly does not, as I should have thought any one certainly would who had any sympathy for methodical inquiry in a new field, view and review his facts until he could educe some apparent laws of general application under which he could range them, then state these laws clearly—not dogmatically as being true, but for the purpose of testing their truth—so that he and all interested could test them by all relevant facts, and, until he could so state them, keeping silence. Instead of this, in practice, he does as he did in the beginning, deals with each case singly; he follows on another line the ancient law-givers, who sat in the gate and made the law they administered for each case as it arose. How else could he avoid discussing the necessary preliminary question as to the form and nature of Polynesian radicals? or ignore the primâ facie disyllabic character of haul, the word in question, and of Hawaiian nini, with its reduplicate forms ni-nini and nini-nini, and its compounds ha-nini and ma-nini, and of the Maori ringi with its reduplicates, and its compounds ma-ringi, ringi-hanga, &c.? And on what general principle, or special facts does he assume that the Hawaiian nini, with its two consonants the same, is more primitive than the stronger Maori ringi, supported, as it is, by most, if not all, the other Polynesian dialects? Or that the Maoris borrowed from the Hawaiians—not the word hand, meaning water (for Mr. Tregear does not even conjecture that the Hawaiians used it in that sense),—but two-thirds of a compound word meaning "pour out," in order to use it for water? Lastly, on what principle does he decide, either that hani, ringi, and ngongi have no other relatives in page 32 Polynesia, or if they have that none of the latter need be taken into the account before making an unqualified dogmatic statement as to the structure and some of the affinities of one of the former?

If this mode of treating the language be passed as lawful, are we not less likely to reach any orderly account of the facts of the language, than a state of chaos worse than that from which perhaps the language itself first arose?

I will give another instance in which Mr. Tregear follows in part the same, in part another mode, not, however, it seems to me, with complete success. He says (Introd. xxiii.):—" As an example how deceptively the letter changes may cloak a real affinity, I will present the Malagasy vorondolo, an owl, as equivalent to Maori runt, an owl. Volo is used as an equivalent for "feathers," the Polynesian huruhuru: the v (* * *)=h and o=u. The Malagasy, however, use vorona as a general name for birds (probably i.e. "the feathered creatures"), as vorombola a "peacock," &c. "The nd of ndolo may be considered as equivalent to the Fijian, in which every d is nd; and as d is merely a form of r and l (dikydiky=likyliky; roa= Malay, dua, &c.) and o=u, therefore dolo is a form of ruru. Thus voro ndolo means "bird-ruru"; and unlikely, as at first sight appears the relationship, it is probable."

On this I will ask, in passing, is not an unconditioned universal proposition, "that d is merely a form of r and l (in reference to languages some of which have only one of these letters, some all three), a dangerous instrument in the hands of a young and ardent philologist?—one with which he is more likely to cut his own fingers than to do any useful work?

Now, if voron-dolo is spelt as, I presume, it is pronounced, vurun-dulu (the Malagasy using the symbol o for the sound u), it is easy to believe that the first element, from its likeness to Maori huru, feathers, &c., may mean bird, and that the second, from its likeness to rum, may mean owl. But the author's method of analysing the word ought to be closely observed, the more especially as this is almost the only occasion, in the whole book, where he offers the student an etymological model. He tells us that roro=feathers, and rorona=feathered creatures=birds. But then, straightway, he lets this particular vorona, bird fly right away, or otherwise escape, leaving only voro, feathers, behind; and, thereafter, it is to this voro he confines our attention as the first element of voro-ndolo, the final a being completely gone, and the n having, in some way, become part of the second element, ndolo. To account for this nd, he does not resort to the phonetic laws of the Malagasy language, though they seem to explain it well, but goes—as philologists, I presume, often must, and as amateur philologists, I dare say, often do when they need not—a long way off, in this case right across the Indian page 33 and into the Pacific Ocean, to Fiji, where, as he says truly, every d=nd; the argument apparently being, that if the Fijians always make d=nd, the Malagasy may well be supposed to do so sometimes. He does not explain why the n of vorona becomes m in vorombola, nor why the Malagasy form of Maori ruru should (transliterating the vowels) be dulu or ndulu, instead of lulu, as might seem more likely; nor why he first tells us that vara means feathers, and vorona birds; and at last, without notice, makes roro alone mean bird.

Now, taking his materials, is it posssible to make a less dangerous or more economical use of them, trying first the resources of the Malagasy language itself before going to Fiji? I speak with great diffidence, because I can hardly claim to be even a student of the Malagasy language; but 1 can claim to have done what apparently Mr. Tregear has not yet found time to do—read a short Malagasy Grammar.^* There, at the beginning, I find what should make a student very cautious in undertaking Malagasy etymologies—hence you will kindly take the little I have to say as mere suggestion. There is in that language a complicated set of phonetic laws, especially as to letter-changes involved in the making of compound words. One of these is that, unlike Polynesian, a syllable may be closed by either of two consonants n or m; hence, roron-dolo would not be an impossible division of our word. Another is that n is admissible immediately before d, but not before b or p; while m is admissible before It or P, but not before d, and a final n before an initial b of the second word becomes m. This is illustrated by voron-dolo on the one hand, and vorom-bola on the other. Another is, that the final syllable na, as in vorona (one of Mr. Van der Tuuk's three "dumb syllables"), is not only itself liable to change, or disappear wholly or partially, but is also, even if it disappears, the cause of change in the first consonant of the second word in the compound. Thus, liana and vady combined make tian-bady; and fonosina and lo make fonosin-do. If, therefore, we suppose the two elements of our word to be vorona and lolo (= lulu), they would, it seems, combine to make that word in the exact form given, voron-dolo, which is yet consistent with voronm-bola. Conversely, we can argue back (though with less assurance) that dolo of the compound may represent as its separate form lolo; that is, that the Malagasy form of the Maori ruru, owl, is not dulu, but like the Samoan, lulu.

Now, it is quite reasonable to suggest, and may be as easy to prove that I, as a novice in the language, have misunderstood or misapplied these phonetic rules; but two things are indisputable—that important and complicated phonetic rules exist; and that if any one would treat of Malagasy etymology,

^* Parker's, in Trübner's Collection of Simplified Grammars. E

page 34 or of how many words in that language are, or are not, related to words in other Oceanic languages, it is his first duty to make himself acquainted with the existence of these rules, and then to master and apply their details.

Scholars who have studied the structure and Grammar of the Oceanic languages, class the Malagasy as certainly one of them, that is, as related to Polynesian, Melanesian, Malay, and others. But Mr. Tregear, who, in the comparison of languages, looks mainly, if not only, to their vocabularies, has, to say the least, serious doubts. "I have not," he says, "been able hitherto to trace even a possible affinity between Malagasy and Maori in more than one hundred words out of ten thousand "[?]" in each language."' Yet it is perhaps more sad than surprising that the "method of vocabularies" which had enabled Mr Tregear to discover such numberless Maori affinities in so many of the Aryan languages of Europe and Asia,^* should have failed its master so signally here. But its failure does not suggest to him any doubt of its sufficiency. If, what other people think the facts, and his view of them, do not agree, why should he assume that the fault is his?

The above statement that our author had only hitherto been able to trace a possible affinity in 100 words out of 10,000 in each language, would be misleading if not taken to imply an intercomparison of some sort between, substantially, all the 10,000 words on one side, with, substantially, all the 10,000 on the other, and in that case the number of individual comparisons would manifestly have been very great indeed; and I would say that the mere mechanism of comparison when done on this scale would be of much interest, if Mr. Tregear would but disclose it. But here again, like the mathematicians of a bygone age, he is prodigal of results, but chary over his methods, and it is easier to see how some of his results could not have been arrived at, than how they could. To compare each of 10,000 words with each of 10,000 other words, though it might not overtax Mr. Tregear's industry, would have made a heavy demand upon his time. Apparently, if he had managed to devote twelve hours a day for 365 days in the year, and allowed only 10 seconds for each comparison—and 1 suppose even an intuitional philologist could not do much of a philologic act in less than 10 seconds—he would have spent over 60 years on this one very small branch of his subject alone. But as presumably the whole dictionary did not occupy him a tenth part of that time, he must obviously have had a very much more expeditious way of dealing with the 20,000 words in question. And, perhaps, even, it

^* We have his own authority (Trans. xviii. p. 14) for saying that one Maori word (Ariki) he had traced in every Aryan tongue; a great feat, as was remarked at the time.

page 35 may be found that Mr. Tregear's investigation has been as little exhaustive in character as in extent.

It cannot be said that the author has passed over without notice the fundamental question of letter interchange in and between these languages, since more than three-fourths of his introduction is occupied with long lists of alleged interchanges. They are, unfortunately, only of a few of the most nearly related languages, those of the Eastern Polynesian group, some eight or nine out of the 40 or 50 dealt with, more or less, in the book. But, even with these, he has made no attempt to carry his investigation beyond the preliminary stage, where, if left, his results are, I am afraid, as likely to be misleading as to be useful.

In the whole of these lists, we have, so far as appears, only the author's conjectures put into the dogmatic form. He assumes throughout what he has to prove, and he gives the reader no hint that he is doing so. How does ho know that a and c interchange in Maori? Because he finds among others such forms as tutai and tutei, where the only difference is in those two vowels. But, to make them evidence, the two words must be taken as two forms of the same word; that is, the assumption must be made that a and e take each other's places in the two words, the point to be proved. The argument, in any single case, is circular, and is admissible only if the assumed truth is admitted to be assumed, and valid to the extent to which other facts support the assumption. But here, as before, Mr. Tregear is prodigal of results, and gives no hint how they were obtained. Yet the question is fundamental:—How many instances has he in support of each statement of these letter-changes? Some are as rare as others are common.

In the case of tutai and tutei, I have no doubt he is right, because I know that similar cases are very numerous; that in an indefinitely large number of words, "and e seem very easily to interchange, especially when acented and before i, even although a consonant intervene. But when he puts the interchange of e with i, or o with u, on the same level, I entirely disagree, because really satisfactory instances of these changes are hard to find; while, to confound e with i, or o with u, especially when unaccented, seems to come naturally to the English ear and tongue, and is, I believe, only too common among those English speakers of Maori who did not learn the language in their early years. I believe a good many of the new words in this dictionary (and I suspect in some other Polynesian dictionaries) owe their existence to this confusion. In this matter, I can claim to speak as, in a somewhat sinister sense, an expert. I have for more than 30 years been, as chance offered, a collector of Maori words, and I must own to having recorded a good many such novelties myself, especially in the earlier part of the time. But I have not published them.

Speaking of the Maori, he says, "the vowels sometimes interchange with each other," and he then gives examples showing that every vowel interchanges with every other. He does not say under what conditions these changes occur; nor whether there are any conditions under which any of them must, or cannot occur; nor whether any one is more or less frequent than any other. Thus, by an affirmative rule—universal, as including all classes of change as possible, indefinite as to the number of individual cases included in each class—he sets the whole matter at large—anything may be anything else—and there leaves it.

He does the same with the consonants, though to a less extent. There is nothing, indeed, to show that every Maori consonant may not interchange with every other, but it is not so stated. Still Mr. Tregear shows that of the ten consonants three interchange with each of seven others; three others interchange each with five others; and so on in a descending proportion—and, as with the vowels, all these interchanges are unconditioned.

This is surely very like adopting in practice as a serious and exact truth the well-known epigram of Voltaire, who, according, to Max Müller (Lect. II, 238), "defined etymology as a science in which vowels signify nothing at all, and consonants very little." The Maori etymologist, if he would connect any word with any other, may sometimes have to go beyond Mr. Tregear's phonetic rules, but need not break them, since they are all affirmative—all enabling, none restrictive. Indeed, a main feature, as from the first it has seemed to me, of our author's philology, is his persistent refusal—not in words, but in practice—to submit himself to any recognised restraints in his dealing with language For illustrations see his dictionary passim.

I will not dwell further upon this beyond citing a saying of Bacon's I lately met with. Speaking of the memory, he urges, as one condition of successful recollection, the need of an abscissio infiniti, a cutting oil of the infinite, the need of having some limits within which to search. It is easier to hunt deer in a park than in a forest. He is dealing with the knowledge we, in a sense, already have, but which, hidden away in the mind, is not easily recalled. But the rule applies as well to the search for fresh knowledge. It is practically hopeless in finite time to search in an infinite field; and we know, as implied in the very nature of language, that it must have its limits in this sense, its limiting laws; just as every other subject must which is capable of scientific investigation. These laws when known, while they restrain us in the sense of keeping us within due limits, are, at the same time, our surest guides. Until the inquirer knows that such laws exist he is hardly on the way to learn; until he knows something about them, he is surely not qualified to teach.

Anyone desiring to find the laws of letter-interchange in Maori page 37 would need, I presume, among other things, to ascertain, in some sort, the relative and actual frequency of the several changes. According to my own investigation of the vowels, it will be found that, dividing them into two groups (1) a, e, o, and (2) i, u, the interchanges within each group are relatively and actually very frequent (as well in Maori as between Maori and other Polynesian languages); while those between the groups are comparatively so infrequent as to justify their being treated, provisionally, as abnormal.

The disappearance of letters from Maori words, of which Mr. Tregear gives some examples, is properly to be considered together with letter-interchanges, though it is an important fact, not noticed by him, that, excepting the Taranaki h, a Maori consonant dropping out of a word, leaves no trace of its former presence.^* But I believe it a mistake to say that this ever happens in case of the vowels; and the distinction helps materially, I would suggest, to mark the relative value of consonant and vowel in the structure of the language. An apparent exception is in partial reduplication. But there the omission is of the whole second syllable, however composed, and is ex hypothesi done with a purpose.

Mr. Tregear says that a is lost in ngoki, as compared with ngaoki; and, no doubt, to a hasty view that seems on the face of it clearly so. But here he is trusting too much to the eye; as in so many words in his dictionary he ignores difference in length of the vowels, no doubt from not having carried his general investigation far enough to recognise its essential character. The a and 0 of ngaoki are both short; the o of ngōki is long; that is, =00, and is sometimes, and more properly so written by the Maoris. Therefore, the true equation is ngaoki=ngooki, and the case is not one of a lost a, but of an interchange of a and o.

Again he says u is lost in haware, hokeke, and tokeke, as compared with hauware, houkeke, and toukeke. But here, also, the u is disappearing to the eye, leaves its effect for the ear, and so is in some sense transformed rather than lost, since in each case the remaining vowel of the disyllable is doubled in length. Whether this is to be called an interchange or a 'compensatory lengthening' must be settled on a balance of considerations—the fact remains that hau,=(in the sense of having become) haa (not ha), hou=hoo, and tou=too. Compare, also, the first elements in kou-tou and ko-rua (=koo-rua), and the last word with its Samoan equivalent, 'ou-lua, and Maori ta-tou (taa-tou) with Tongan tau-tolu, and so on. If, as seems also to happen, the

^* If the view before proposed of the Taranaki h is correct, it is not an exception to the above rule, but another case of consonant interchange—the soft breathing for the rough. As to the consonantal character of the former of these see Lepsius, Standard Alphabet, p. 68, and Brugmann Corny. Grammar (English Ed.) p. 21.

page 38 unaccented i (i.e., when the second vowel in a disyllable) is similarly liable to be transformed, we may probably identify paa-hau and pai-hau, maa-hee and mai-hea, and a good many others; and the same considerations may help to establish a community of origin between such pairs as ra (=raa) and rangi, and po (=poo), and pongi.

Among the subjects which, I suppose, must necessarily be considered in any real intercomparison of these languages, and which should, therefore, receive at least some notice in the introduction to a comparative dictionary, but have none in this, are the following:—Re-duplication—partial and complete—of the radicals; and quaere whether 'analogy' has not, in a few cases, introduced a spurious or abnormal reduplication, so as to include in it what is probably a prefix. Metathesis—the interchange of places of the two consonants, or of the two syllables of a radical; and probably connected with this, according to Professor A. H. Koane, is the interesting question of Infixes—and with the last are obviously connected Prefixes and Suffixes, two branches of the subject of wide interest and importance. Assimilation and Dissimilation also need notice; but they do not seem at present to play much part in Eastern Polynesia, though they do, I understand, in some other Oceanic languages; and there is an interesting practice of vowel dissimilation shown in the Tongan vocabulary, in the reduplication of certain classes of radicals. And, lastly, I would ask, is there, in any of these languages—Maori for instance, as that is most accesible to us—any evidence of a tendency to organic change?

I will now pass on to a more detailed examination of the work. In this I have compared a good many articles in it with the work which is acknowledged to have been its basis—Arch. Williams' Dictionary; and where, as often happened, Mr. Tregcar's examples seemed to me wrong or inappropriate, and there were references, I looked these up, and noted the result; in these cases the corrections are in the imperative; in other cases the emendations usually, at least, take the form of query or suggestion. Some of these corrections may seem small; but it should be remembered:—In the first place, that in Maori a very small difference to look at is often a very large one in fact; secondly, that where the errata, even if small, are numerous—and I have not set down nearly all I found, and my search, of course, made no attempt to be exhaustive—they are some index of the quality of the work, at least in its last stage, possibly in all its stages; and thirdly, even the really small corrections may be of practical use to beginners, or even those a little more advanced, who have not the necessary works for reference.

To save space I have taken the liberty of referring to Mr. Tregear as T., and to Archdeacon Williams as W.; but these letters will sometimes stand for their respective dictionaries. page 39 Archdeacon Maunsell's Grammar is indicated as previously by "M. Gr." or Gr. Ex. will often stand for "example"; in other cases I have used Mr. Tregear's abbreviations.

The marginal numbers refer to the pages of the dictionary; the following Maori word shows the article commented on.

1.—A.—T. gives as the meaning of this word, "God, the Deity (one auth.)." Farther on he gives as the first meaning of Atua, "God"; and of Io "(myth.) God, the Supreme Being." No hint is given, as in other cases, that these meanings are modern; presumably, therefore, this is not intended. Are we then to understand that A, Io, and Atua meant to the Maori people, or some of them, before their intercourse with the Europeans, what the words God, the Deity, the Supreme Being, mean in English?.

If any one will reflect on what these English words denote and connote, on the Being they designate and his attributes, and then on what is known of the Maoris and their conception of their so-called "religion," its scope, and methods, before they had heard of Christianity, he will recognise the tremendous call T. is making on his credulity, and the painful insufficiency of the supporting evidence; and he will be still less inclined to concede what is asked, when he notes the evident unconsciousness of the author that what is asked is in the least out of the common.

1.—A.—" A prefix to proper names, pronouns," &c. This is the whole of T.'s account of this important word, the personal prefix. W. defines it as a prefix to proper names and personal pronouns (except ahau), and occasionally to common nouns; and he gives three classes, and three sub-classes of cases in which it occurs, with six appropriate examples. This being the "raw material" supplied by W. alone, besides much else by other eminent writers, not to mention what is heard whenever the language is spoken, T. applies his appropriate method (philologic or literary—the latter, I think), with the result that all of this information, which is supposed to concern his reader, is compressed into the six words above quoted, "a prefix to proper names, pronouns, &c." W.'s significant statement that it is not used before ahau [though it is before au] drops out with the rest, and therefore needs no troublesome notice.^* But T. applies also the comparative method, and adduces "comparatives "from two other languages, both Polynesian, but from none other. Yet Dr. Codrington (whose book stands near the beginning of T.'s very respectable list of "Works Consulted," whatever that somewhat ambiguous phrase may mean) points out that "the use of a 'personal article,' a remarkable feature in a language, is found

^* Yet it is also most interesting. There seems little doubt that Dr. Shortland is right when he says (How to Learn Maori, p. 13) that the first a of ahau is the personal prefix.

page 40 certainly to prevail in Melanesia, in Polynesia, in Madagascar, and almost certainly in the Malay Archipelago. The meaning and use is identical "(Mel. Lang, 110). One of the forms of the Melanesian personal article is a, as in Maori; the others are e and i, and I do not feel sure that one, or possibly both of those may not in some way be found in Maori also. But are not these, and the like, exactly the points on which the student might look to a comparative dictionary for help?

1.—A. (prep.)—W. says of this characteristic word, "a, prep. 1, of, belonginy to; used in speaking of actions of any kind, food, children, slaves, &c., but not in speaking of the parts of a whole, names, qualities, houses, lands, water for drinking, clothes, &c."—for which latter, as appears later, 0 is used. The one has long been called "the active a," and the other "the passive o" and this, whether they are used alone or in combination to make certain pronominal or possessive forms, as aku, aku; malm, moku; mau, mou, and so on, thirty or forty in all, without counting the double compound forms, a ratou, na ratou, &c., conventionally written as two words, yet no doubt showing the composition of the others, each of which is conventionally written as one word.

Of the use and importance of this distinction between the active a and the passive o, Archdeacon Maunsell and Bishop Williams give some forcible illustrations:—He potto maku, a striking for me (to suffer); he patu maku, an instrument for me to strike with. Naku Una whare, I built that house; noku tena whare, I dwell in it (it is mine). He hangi mau, an oven with which you may cook food; he hangi mau, an oven in which you are to be cooked—a most offensive curse (M. Gr., 188). Te patunga tenei a Ngatiporou, this is the place where Ngatiporou killed certain persons; te patunga tenei o Ngatiporou, this is the place where Ngatiporou were killed. Archdeacon Maunsell's Grammar has been before the New Zealand public for fifty years, and in that time it has reached T.'s book-shelves, and been "consulted," but I fear we must conclude that he has not yet found time to read it. Yet the vital distinction between a and o it points out is forced upon the attention of the mere tyro in Maori. His sense of its importance grows with his knowledge, and his observance of it is one good test of his progress. How then does T. treat it? His definition of a in this sense consists of the first three words of W.'s article, "Of, belonging to;" turning to o we find the very same words repeated: "o, of, belonging to," and no indication of the difference; in each case the essential facts given by W. and others, and ready at hand, are conspicuously absent, and T. gives the student nothing in their place. Of the thirty or forty forms I mentioned in which a or o appear, I have found two only, ma and na, in which T., copying from W., notices the active use of a forms. Even then page 41 he does not contrast their passive correlatives mo and no. Thus, though giving an isolated scrap of borrowed information, still missing or ignoring its main significance.

But he adds something of his own on another line, the comparative, showing that a means "of, belonging to" in six or seven other Polynesian languages. His comparative method, however, as I have said, does not include, unless accidentally, any critical examination of the words compared, their relations inter se, or with others. He brings two or more words together, and the reader must do the rest. In this case, indeed, it was hardly to be expected that he would notice, in these other languages, a distinction, useful, elegant, and characteristic as it is, which he himself had either not perceived in Maori, or had looked on as too trifling to mention. Yet this same distinction is well known and treated as important, in at least several parts of Polynesia, if not universally. See as to Samoan and Hawaiian., the Rev. S. Whitmee's edition of Pratt's Samoan Dictionary, 1878, pp. 5-6. See also his practice in the dictionary itself in noting as to many nouns, whether they take the a or o form of the possessive. The distinction is well recognised also in Tahiti, and from the existence of similar double forms in other Polynesian islands and settlements, as in Tonga and Uvea, I should suppose the recognition general.

2.—Aha.—For he aha ana read e aha ana. How did the Cockney h get there?

5.—At,——A relative particle, a word of great interest and constant use, principally as supplying to a large extent the want of relative pronouns in Maori; familiarity, therefore, with its chief functions is essential for even an elementary knowledge of the language. W., being a Maori grammarian, that is to say, having an intimate and idiomatic knowledge of the Maori language, treats it at considerable length, setting out seven or more cases in which it is used, with numerous illustrative examples all translated, forming together a valuable compendium. T. with this, amongst other, as his material, works it up to this effect:—"Ai, a particle having no English equivalent, and only to be understood by reference to a Maori grammar." He adds, however, a four-line summary of its various uses with three untranslated examples; and concludes with some "comparatives." The student, I think, will infer that so far as a knowledge of Maori is concerned, T. has reached his result by the application this time of what may be called the negative philologic method, since he cancels in the finished product most of the "philologic" facts of which the raw material was full.

10.—Ana,—T.'s account of this verbal particle, one of the most frequently used words in the language, is taken imperfectly from W., whose article might well be added to but not shortened. T. confines its use to action, ignoring state, condition, quality, page 42 &c., apart from action. He ignores, also, the obvious fact that e-ana takes its time from the rest of the sentence when any time is indicated, and will often, therefore, mean a past, future, or conditional, instead of a present and actual state, or action. And even if his "rapid action" expresses the truth conveyed by W.'s "rapid succession of actions," T. might have ascertained by reference to Maunsell's Grammar that ana is often used without e to give emphasis to the verb in animated sentences, and to express continuance of action, whether rapid or not. T.'s last statement about its denoting "finality of action" (of which, as often where it is most wanted, he gives no example), I confess I could make nothing of until by referring to W., and the example he gives, I saw how, historically, and by the application, as I suppose, of the literary method, it had been arrived at.

W., I may add, gives five examples, T. one.

12.—Ano.—T., as before stated, begins his short article on this useful, but not particularly difficult adverb, with the näive direction to the reader: "See Maori Grammar." But surely it is barely civil while taking from another man's dictionary substantially, if not quite word for word, all the meanings you intend to publish in your own, and still leaving valuable material behind, to leave him unnamed, and refer the reader for the rest to "Maori Grammar" in general. The student, also, if he is to be sent elsewhere, and into such dangerous company, might well ask a more specific reference.

Compare W.'s fourteen examples with T.'s five. In the first of the latter, for kouto read koutou, and in the last for am e wai read ano he wai.

14.—Ao.—T. gives as a 6th meaning "mankind (met)," but this meaning is not to be found in his ex., "ko tenei tangata no roto i te whenua, e hara i tenei ao (i e., he was not a man but a supernatural being)," The explanation misses the real antithesis which is between this upper world (tenei ao), where men, among other beings, live, and the underground world from which Tumutumu-whenua came. In the very next sentence of the original we are told that his wife, Te Repo, was a tahurangi or fairy [and therefore not human], but was of this world (te ao nei) living on the lofty mountains, although not visible to ordinary men.

15.—"Aokai, the Pleiades (see Matariki)"—But Aokai is not there mentioned. T. again unfortunately gives either ex. nor reference. Is it an epithet expressing an attribute of Matariki, and possibly meant for ā-kai, as in the saying Ko Matariki a-kai ki-uta?' Matariki driving food a-shore,' i.e., driving fish shoreward, where it may be caught.

27.—Ata.—In "kei rongo mat aku hoa i pata au nei," insert the omitted i before au; it was the speaker himself who had been beaten.

27.—Atakite.—T. says, "to behold dimly; obscure;" citing Te atakitea atu te whetu o te rangi. On this I should propose for dimly to read clearly, and to strike out obscure, which, as a meaning of atakite, does not seem to have even a misunderstanding to support it. Dr. Shortland, from whom the passage is taken, translates, "the stars of the heavens are obscurely seen," the three last words translating te ata kitea; lit, not clearly seen=obscurely seen. Could T.'s note have been written before he had become acquainted with te as a common negative? There is, of course, no more reason for printing atakite as one word than any other pair made up of verb and adverb. This very same ata kitea is printed separately in Sir George Grey's version of the same poem (G.P., 240).

29.—Ati.——W. gives this as used only in the names of tribes: Te ati-awa, &c., meaning off-spring, descendants; and suggests (as Dr. Shortland did, S D., 307) that Ngāti is a contraction of Nga ati, as in Ngatimaru, &c. T. re-words this, but has not apparently allowed himself time to notice that W. marks the a of Ngati long; nor to notice what ought to be as obvious that if Ngati is really Nga ati the a must be long; he therefore leaves it short. On the other hand, he tacitly overrules W. (who confines the meaning of ati as "descendant" to the tribal prefix), and gives this as an independent meaning. Unfortunately, the two examples he cites in support of his view are of another ati altogether, not a noun, but an adversative conjunction, "but, then" (c.f., oti), as will be seen in the full text of the passages cited.

30.—Atu.—A long and instructive article might, and ought to have been written on this characteristic and most frequently used directive adverb, the principal meaning of which, as pointed out by Dr. Shortland,^* is 'forward from the front of the speaker,' as its equally indispensable correlative mai means principally 'towards him as he is facing, or supposed to be facing.' The reader should compare T.'s meagre article giving three meanings only, and one of these not very exactly, and his three examples, with Mr. Colenso's 30 meanings and 80 or 90 examples (G.—2, 1882). If T. had mastered the meaning of all these examples be might have classed them under less than 30 headings, but he certainly would not have got them under three. Some of the notions and distinctions involved are rather subtle, but it is just as to these that we want a mind capable of real dictionary work, and with sufficient time at command, to help us; a good palpable difference most of us can detect unaided.

36—E.——T.'s treatment of this important verbal particle shews the same negative sympathy as is shown for all the others—indeed, for every word having many grammatical functions.

^* How, &c., 48.

page 44 He says, imperfectly following W., whose own account is too brief:—E, a particle generally preceding a word used as a verb to express the future tense. When e is followed by ana it denotes present time, as it does when preceding numerals." Now, no doubt, e is often used for the future, though not so often nor so unrestrictedly as ha (M. Gr 148), and it should have been stated that it is often also used for the present, especially with nei, &c., te tangata e haere mai nei, the man who is coming here (close by); more often to denote a contingency or condition, e riri ia if he be angry, e hau if there be a wind (M. Gr. 140); and very often for the imperative, e waha get on (my) back; e ara, arise; e noho, sit down, stay, and so (from one going) farewell; and (in a compound sentence) it is often used of the past. Te mea i kore ai e nehua, the reason (they) were not buried. Again, it is true that in a simple sentence e-ana may commonly be translated by an English present indicative, but as is shown by Dr. Maunsell, and indeed is obvious, it may, in a compound sentence, represent past, present, future, or conditional. Hence the view that it asserts nothing of time is probably the best; the time must be inferred from its surroundings.

41.—Haeatanga, an opening, &c.—Taken by T. from W., but with a syllable left out. Read Haeatatanga; so also in "Key."

49.—Hapuka.—Is this name for the familiar fish hapuku, known to the Maoris otherwise than through English fish-buyers, and this dictionary?

49.—Hara.—For ka rarue, which is meaningless, read ha rarua. A passive ending in e would be a novelty.

50.—Haramai.—W. says:—" v.i (contraction of haere mai) come towards the speaker, often used as an expression of welcome. Pass. haramaitia, be come for, Te korero i haramaitia e ia; the report on account of which he came.

T. varies this, assuming to correct W., but only by putting the specialized before the general meaning, the cart before the horse. He says, "Haramai, an expression of welcome, meaning to come towards the speaker; a contraction of haere mai; passive haramaitia, to be come for." He gives no example, and adds nothing of his own but the disarrangement of the order of the meanings. The word was certainly used widely in the general sense "come hither," for which see G.P. passim. With regard to the statement that hara is a contraction of haere, I would ask, how is this known or even made probable? The cases, I think, must be very few, if any exist, where two Maori vowels are contracted into one short vowel. Moreover, to those who believe in the importance, not to say the essential character of the disyllable in these languages, forms like Maori hara and Hawaian hele, will, primā facie, appear more primitive than trisyllables such as haere and haels.

54.—Haumaruru.—For ka ko te tapapa read ko to te, &c.

59.—He.—W. says, "3. In difficulty or trouble, ka he toku manawa, I am out of breath," or "I am out of heart." T. says, "2. A difficulty, trouble; to be in trouble. 8. To be acquainted with. 4. Suffocated (I he te wanawa)." Here T., by implication, assumes to correct W. for associating troubles of the mind with troubles of the lungs, under the phrase, which expresses them both, and so separates them by another meaning sufficiently foreign, such as it is; and, further, corrects him by substituting "suffocated" for "out of breath," that is, the extreme case only, for all the lesser degrees. As to the third meaning, I should propose to reverse it, for "acquainted" to read "unacquainted." We have at least Dr. Maunsell's authority for the latter (Gr., 54—64), not to mention the general character of the word he.

59.—Whakahe—In me whakatika ata nga whakahe, for ata read atu. In his ex. whakahe has not the meaning it is cited to support, mistake, error—but means accusation. See J. White's translation and post, whakatika.

60.—Hei—An extraordinary omission should hero he noted:—Hei, the preposition future and verbal particle, a very useful and interesting word—is not given by T.

68.—Hikutoto.—For ko tauatia ki te taua hikutoto, which is unintelligible, read as in orig, kia tauatia, &c.

76.—Ho (to give).—W. says, "used only in the compound forms, hoake, homai, hoatu," This, I thought, was agreed, but T. amends it. He says, "Ho, a word expressive of the action of giving, presenting, &c. It is very rarely used except in composition, as homai, give (hither), hoatu, give (away from the speaker); mehemea ka kaiponu koe i ho kai, kaore i hoatu e koe, Mss." This example, I presume, is intended to show not only the common use of ho with atu, as in the second clause, but also one of the "very rare" uses of it without ake. atu, or mai, as in the first clause. T.'s all but unbroken rule is never to translate a Maori example, and he follows it here; and if the observance of the rule often, as here, leaves the student in difficulties, he must find his compensation in the thought that it must often relievo T. of difficulties not less great. Of the present example I will only say that if the first ho is to be taken as ho, to give, or in any way seriously, h and all, the clause in which it occurs is simply unintelligible. If T. has any other support for the doubt he raises as to W.'s statement, why does he not produce it?

77.—Hoehoe.—For nana i hoehoea te moana read nana i hoehoe, &c. Was it copyist or printer who put the verb into the passive, and who read the proofs?

77.—Hoatu.—In ka hoatu he ia te wai, for he ia read e ia, Here, again, how did the h get in?

79.—Hoki.—W. gives as one use of this very useful adverb, "To give emphasis to an assent or affirmation, &c."—adding appropriate examples. T. alters this, and in so doing lets its meaning escape. He says, "to give emphasis, to assent." Thus, by inserting the comma, and omitting the article, he makes a new verb of an old adverb; and, by giving no example, he deprives his reader of the means of testing the reality and use of the transformation.

81.—Hongi.—For ki hongi hi nga wahine read ka hongi, &c.

89.—Whaka-hua.—For i ratou ingoa read i o ratou ingoa.

90.—Huanui.—For the unintelligible me moe mana read me moe maua; and in ref. for 12 read 2.

92.—Hurarere.—For this read Hukarere, and in ex. for kia read kei, reversing the meaning.

93.—Hume:—T. says, "2. A coward"; but for this read "drawn in, tucked in." His example shows that it is he whiore hume, a tail tucked in (between the legs), which represents the soward as with us.

96.—Huripoki.—W. says, "Huripoki, v.t., turn upside down. Huripokia te kohue. turn over the soil with a spade." T. puts it in this way, "Huripoki, to turn upside down; c.f., hurt, to turn, poki, to place with the concave side downwards. Huripokia te kohue, to turn over the ground with a spade." T., to the real loss of his readers, seldom takes an ex. of W.'s; seldomer still does he favour them with a translation of any Maori example at all. The present case will perhaps make the younger of them wish he had done so oftener.

It will be observed in W.'s article (1) that huripokia te kohue is an example of the first meaning, turn (something) upside down, the thing selected being a kohue (kohua) or cooking-pot; the verb being in the imperative passive; (2) that his last sentence gives a second meaning, or rather an amplification of the first. Unfortunately, T. takes the second meaning as translating the example of the first, with the curious result that the three words which mean, "Let the cooking-pot be turned upside down," are rendered, "To turn over the soil with a spade." It is an instance of the unusually successful application, as it seems to me, of the negative philologic method.

99.—I, prep.—A casual consulter of this dictionary might well be struck at first sight by the number and variety of the meanings T. gives to this important word (the smallest, I suppose, of all possible words, but in Maori one of the most useful). Even a comparison with W. would show that while the latter's last meaning is numbered 15, T.'s last is 28. A closer comparison, however, will show that the larger number has been arrived at by a re-division of W.'s material, not by adding to it. T.'s is a remarkably near approach to a word for word copy of W.'s in everything but the numbering and the examples. Of page 47 the latter, the student will find to his loss that while W. gives 84 or 35, T. gives only 20.

99.—Ia.—W. says, "pron., 3rd person, singular, he, she, it"; and he gives a view theoretically interesting, and information practically valuable of the substitution in certain cases of other pronominal forms for ia: adding an important observation, that ia is not ordinarily used in speaking of inanimate things.

T.'s article is at least concise: "Ia, he, she, or it "; and, having given one example, he goes on, without another word, to his comparatives. Yet his comparatives themselves should have shown him, what we must suppose he himself had not learnt in learning the Maori language, the importance of W.'s observation. And if he had been enough interested in the Melanesian languages to look at their grammars as well as their vocabularies, he would have seen some cases of their dislike to use the third personal pronoun for inanimate things (principally the plural, but also the singular; Mel. Lang., 120, 266); and if, further, he had acted on the advice he sometimes wisely, if a little inconsistently, gives his readers, and consulted a Maori Grammar (e.g., Dr. Maunsell's, p. 34) he would have found it stated "that there is no word in Maori to denote the word it with its dual and plural. Their place is generally supplied by some artifice of the construction."

101.—Ihu.—For tukua read kukua T. gives two common meanings to ihu; 1. The nose; 2. The bow of a canoe; and adds another not to be found in W., 3. The foresail. Here, unless I am mistaken, by again applying the negative method, he is able to find three new words or meanings in one short passage of ten words only. He quotes as from P.M., 72, Maranga to te ihu, te waenga, me te kei. The student will probably think there is something wrong about this. Reference to the original will show that me te kei should read me to te kei, a great improvement, but he will still find te waenga. If, as is probable, he should, nevertheless, suspect that this te is a misprint for to, and should look to the same story in G.P., lix., he will find his suspicions justified, as there printed it is to waenga. The passage describes the starting of the Arawa canoe from Hawaiki for New Zealand. The arch-thief Tamatekapua is by stratagem kidnapping the celebrated priest, Ngatoro and his wife, to perform the ceremonies necessary for a prosperous voyage, and is in a great hurry to get away. The moment they are on board he orders the anchor to be hauled up, and the sails set. Accordingly, "Maranga to te ihu, to waenga, me to te kei."—" (Then) are raised the [sail] of the bow, that of midships, and that of the stern," and away goes the vessel. T. had a slightly corrupt text to deal with, and he, or his printer, further corrupted it, but even as it is in his dictionary, the to of to te ihu ought to have put and kept him on the right road, as no doubt it would if he had not ignored it.

108.—Ikanuiatahua. "Muri ihote karakia to te Ikanuiotahua.'—Insert ko after iho, and for to read ho,

104.—Tnaina.—In te ra i whiti nei read, as might be expected, e whiti nei.

106.—Io.—T. says, "To (myth). God, the Supreme Being." [See A, ante, but here, I will ask, is it not at best a grave solecism on the author's part, while declaring this To to be mythological, to give as its English equivalents the names he has chosen, and that without qualification?]

T.'s ex. may be translated "To is the chief god; he made the earth and the heaven." Compare this obvious Maori reflection of Bible teaching with the words of a real authority on genuine Maori belief, the famous Te Heuheu. The Rev. R. Taylor says (Ika, 108) "Speaking to Te Heuheu, the powerful chief of Taupo, of God as being the creator of all things, he ridiculed the idea and said: 'Is there one maker of all things amongst you Europeans? Is not one a carpenter, another a blacksmith, another a shipbuilder, and another a housebuilder? And so it was in the beginning; one made this, another that: Tone made trees, Ru mountains, Tanqaroa fish, and so forth. Your religion is of to-day, ours from remote antiquity. Do not think then to destroy our ancient faith with your fresh-born religion.'" This, I believe, is consonant with all genuine Maori tradition.

110.—Ka.—T.'s whole account of this most useful verbal particle is taken imperfectly from W., whose account was if I may say so, itself too brief. T. says, "Ka, an inceptive particle. It is used to denote one action changing to another, or the commencement of another occurrence: [ex.]. At the beginning of a sentence. When, as soon as: [ex.]." It will be observed that this takes account of actions and occurrences only, ignoring state, condition and quality. No doubt in one sense ka is inceptive; in another, it might be called "completive," as with the numerals, ka toru this is the third, this makes up the number three (M. Gr., 140). Its use, also, in the imperative should have been mentioned: tana ka hoki, let us return; a, ka kore ia v whakao mat, ka moimoi; if he shall not answer, you must (call) moimoi. The first ka in the last example shows another of its common uses, that in contingent statements. The student's attention should also have been called to Dr. Shortland's statement (Bow, &c., p. 38), "Ka is independent of time, and may be used with past, present, or future. It merely gives to the word to which it is prefixed the force of a verb." And a good deal more of value might have been added on this as on all the other verbal particles.

111.—Kaha.—For whaka-hangia (which, I suppose, might mean "baked,") read whaka-kahangia, "strengthened."

112.—Kahawai.,—For ngak read ngako.

116.—Kaia.—Ko te tangata nana i te timata te kaia, omit the unintelligible te before timata,

116.—Kaiahika, wounded.—It is repeated in the "Key," but is it a Maori word? Kaiakiko, for wounded, was common in the wartime 80 years ago.

117.—Kaihau. "The priest (tohunga) who eats the hau or portion set apart for the atua, or deity (see whangaihau)."—This is too general. I understand that the hau in the whangaihau was often a lock of hair of the slain, a scalplock; and after a battle these might be numerous. The Maori was a hardy eater, but the priest would hardly be expected to eat these.

117.—Kaioraora. "Ki te nui e te kaioraora a nga tuakana nona." Here copyist or 'reader,' or both, must have been thinking of other tilings. For e read o, and for nona read mona,

117.—Kaiponu.—For ngaro read ngare,

122.—Kani.—In "a tangohia ratou etahi wahine, &c., the verbal particle is omitted, and by the further omission of the preposition e, it is left an open question who takes whom, whether the women carry off the men, or the men the women. It should read a tangohia ana e ratou, &c. The ref., also, is wrong-to Ecclesiastes instead of Judges.

126.—Kapu.—In na wai a mehua nga wai (which ought to have arrested somebody's attention), for a read i; and why does T. admit into his examples hybrid words (as mehua, Eng. "measure"), which he 'rigidly' excludes from interpretation? (Introd. p. xxiv,) I have met with several others.

130.—Karewarewa, karearea, and kaeaea.—These three names (or as I should say three forms of one name), are, as Sir W. Buller says, applied to both his species of the smaller hawk, that is, (1) the quail-hawk (Harpa, N Z.), and (2) the sparrow hawk or bush-hawk (H. ferox); certainly, I should say, all three names are used of the latter. T., however, assumes without describing a distinction, on the one hand, between a karearea and a karewarewa and on the other, between a bush-hawk and a sparrow-hawk. He does not, however, propose a third species, nor give his reasons or authorities. We were assured (Circ. advt.) that the 'scientific nomenclature of native birds, plants, fishes, &c., had received careful attention, and might be relied on as accurate in description, and almost complete in detail.' I do not know what exactly the latter part of this means, but the reader must not, as a rule, expect any description whatever of the birds, plants, and fishes themselves, however accurately described their nomenclature may be, and however nearly complete in detail. It is, however, to be noted that Mr. Tregear (Dict., p. x.) modestly admits that this branch of his subject is not "absolutely perfect," and this ought to modify, if it does not quite disarm, criticism.

141.—keke.—For ko te whaia read ko to whaea, &c.

142.—Kei.—As to T.'s converting the stern of a canoe into a rnizzen sail, on the strength of a misprint and a little misunderstanding, see ihu, ante. But note that here in re-quoting the same ex, he corrects the misprint previously added to it, yet repeats the same curious view of its meaning.

145.—Whaka-kiki.—W. gives this as to 'instigate,' with an appropriate ex. T. unaccountably makes it into 'investigate,' citing a part of the same ex.; the part showing the force of the word, the object of the instigation being cut off.

152.—Ko.—A word of equal use and interest. W. says, "a particle used before proper nouns and personal pronouns, and also before common nouns when they are preceded by te or any of its compounds, to, toku, ta matou, &c., which elsewhere he calls definitives (First Lessons, pp. 15, 16): 1. To give emphasis, and hence frequently to denote the predicate. 2. To direct attention to the subject about which something is about to be said. 8. To specify particularly what has been already alluded to in a more general way; and so also to indicate or enumerate the individuals signified by a dual or plural pronoun."

These rules, if not complete, are with their seven examples of very great help to the student.

T., having this before him (not to mention what other authorities have written, and what every competent Maori speaker knows), gives us the following: "Ko, a particle used when the predicate is either a proper name, a personal pronoun, a local noun, or the interrogatives wai or hea; also before a common noun with any of the definitives except he (see Maori Grammar)." "Whether the student is to use it on every such occasion, or if not, when? and its purpose and effect if he does use it, he is naively told to dicover for himself. "When he gets among the grammarians, be may ask, might he not profitably exchange T.'s whole article for six words from Dr. Maunsell? who (Gr., 124 and 152) describes ko as "the article of specification and emphasis."

153.—Koa.—"W. says "ad. intensive: 1. indeed; 2. in entreaty." T. says, "an intensive: 1. indeed; 2. in entirety." His conversion of 'in entreaty' into 'in entirety' (as, a little before, instigate into investigate), if it amuses some readers, may puzzle others who have not W. If T. had given an ex. and translated it, and then looked again at his definition, he would no doubt have recalled the novelty which the latter contains.

155.—Kohatu.—For Hihi ona, &c., read Hihi ana, &c., the verbal particle for the possessive.

155.—Kohikohi, " (myth), the name of the aborigines of New Zealand when discovered by the Polynesians (Maori)."—This is too positive, and too general. Why does not T. give his authority? How many other names are there for supposed N.Z. aborigines (i.e., people said to have been found here on the arrival page 51 of sonic of the well-known canoes), and how many Maori tribes agree in any one name? And how is it known or guessed that they were not Polynesians? T. himself gives several other names (see Kahuitoka, Toi, Upokotoea, and Hiti, and there are several others). Under Hiti it will be seen to be a debateable question whether there were any inhabitants hero before the Maoris. The evidence of tradition is both weak and conflicting.

156.—Kohore, "no, not."—This, to say the least, unusual form of kahore is given by T. as from the 'South Island' dialect. But do we not want more evidence before accepting it? I would suggest that, as with some other of T.'s most distinctively new words, it comes not from the 'South Island' but the printer's dialect. The paper containing it (Wohler's, Trans, vii.), valuable as it is, is greatly disfigured by misprints; yet in nine other places in the same short story the usual kahore appears.

160.—Kororohimako, an obvious misprint for kokorohimako.

164.—Kopere-tane.—T. says, "an exclamatory phrase uttered by the leader of a party (usually a war party) as the signal for immediate action." A phrase of similar meaning, and to a careless eye of similar form (though essentially different grammatically), kopere taua (or kopere tatou), "advance," or "let us go," is well known. (Te Rou, 196, 840; A. H. M. iv., 135.) Is Kopere-tane a miscopy of this, or really another genuine form? It surely deserved a word of explanation.

172.—Koromatua.—For kotahi i toetoe read kotahl i toe, How came the copyist to reduplicate this verb with the subject in the singular?

172.—Oromiko-taranga.—To this prefix K; as also before owhaitau and owhakararo on page 178—evident mishaps.

179.—Kua.—W. says, "a verbal particle denoting that the action is completed at the time indicated." T. re-words without having apparently quite understood this useful and concise definition, and without having drawn on his own practical knowledge of the language; and so he further shortens what was already too short, by leaving out a vital part of it—the last four words. He says, "a verbal particle denoting the completion of past action." But kua commonly marks the perfect whether past, present, or future; the completion of the action at the time indicated, whatever that time may be. Kua karanga ia, he had, has, or will have called (W., p. xiv.) the time of the sentence shewing which is meant. So Arch. Maunsell "Akuanei ko Hone kua tae, presently it will be John, who (emphatically) has got there, i.e., John will have got there first (Gr., 152). Kua mate ahau, e ora ana ano nga rakau nei, "these trees will live longer than I"; lit. "I died, these trees arc still alive" (Gr. 168). Moreover, kua is not concerned with action only, but also with state, condition, quality, &c. Kua po, it is night; kua koroheke, page 52 (he) has become an old man; kua kore, (it) has disappeared, (more lit.)—has become not.

201.—Mamae.—W. by accident, no doubt, omits what is, I suppose, the commonest use of this word (i.e., as a noun meaning 'pain,' whether of body or mind), and so begins with the adjectival or verbal use of it. By a coincidence T. must have met with exactly the same accident, since he also omits this commoner use and begins with the other; and a further coincidence is also to be noted, that in the second part of each book, W.'s English-Maori part and T's "Key," which is founded on it, mamae appears as the equivalent of the English noun "pain."

W. gives two meanings to the causative form whaka-mamae, one transitive, the other intransitive, which T. copies faithfully enough; but as he, on principle, omits this too-refined distinction in verbs, it is not surprising that his one example should stray from the intransitive division, where it belongs, to the transitive, where it is quite out of place: Rangiuru, the subject of the passage quoted, was not causing pain to any one, but was for a long time in pain.

203.—Mamutu.—A new word of T.'s, meaning, he says, "1. clean; 2. power, authority (as mana)" If the first meaning is right, should not the word be written as two, taking ma as the common word for clean, and mutu as an intensive? As to the second which is found in the printed version of a letter of Potatau's, may it not there be a misprint for ua-mutu, the nape of the neck (perhaps strictly the uppermost cervical vertebra)? I do not know the combination, but both ua and mutu have the former sense, and mutu, at least, has both (compare also mo-ua, tuta, and tutanga). T.'s ex. could then be translated, "It is not right that one man should come without excuse to put his foot upon the neck of another"; and mana, as a secondary meaning, would come naturally, in certain senses.

203.—Manahuna.—T. says, "eels which wriggle into dark holes." Is this new name intended to denote a new species of the common eel discovered by T.? If so, does the habit described sufficiently distinguish this eel from its fellows? Presumably any eel would wriggle anywhere, and always by preference into a dark hole, when any human being was near enough to take an observation. Valuable, therefore, as habits may be in classification—and it must be remembered that they are often more enduring than form—it is evident that a habit common to the whole genus is not enough on which to found even a young naturalist's variety. Could T. tell us of any structural difference marking the manahma '? A Maori friend suggests that the name is not that of the eel, but of the dark hole or place in which it lives

204.—Manako.—T. says, "the constellation of Magellan's Cloud (one auth.)." Who is the one authority, and under what page 53 circumstances does he make this curious statement, or the statement on which this curious statement is founded? What is the constellation of Magellan's Cloud? And which of them is meant? Astronomers agree with the many in recognising two Mcgallanic Clouds, but do not, commonly, class either of them singly, or both together, among the constellations. The Maoris also recognise two, and have at least two names for each; while the short description I have heard of both, in answer to the question what they were—" he purehu ra "—would fairly translate their astronomical name, nubecula, Mr. J. White, A. H. M., 1, 52 (trans.) speaks of Manako-tea (white Magellan Cloud), and Manako-uri (black Magellan Cloud), and Te-ika-o-te-raki (which he also calls Mango-roa), the Big Magellan Cloud. But he is in part interpreting a doubtful diagram, and the passage seems to want revision. Is he not including as clouds of Magellan both the Coalsack and the Milky Way?

207.—Mānu.—Besides the known intransitive meaning of this word, to float, T. gives a transitive one, "to launch, to cause to float." But this, I venture to say, is not supported by the only examples he quotes: "Ka manu ia te waka" [ka haere a Whakatau, &c.] [Now the canoe floats, and Whakatau goes on his voyage.] T.'s mistake, I presume, arose from taking ia for the personal pronoun instead of the conjunction. But in that case the sentence would not construe (there is no transitive particle, and Whakatau should have appeared in the first clause); while, in the other case, giving ia a common meaning (which, by the way, I believe, it has in eight or nine other places in the same one-page story) and giving mānu its common meaning, there is not the least difficulty.

211.—Maori.—This word, the most widely-known, I suppose, in the language, has received singularly little attention. T., for instance, has practically nothing to tell us of its meaning beyond what Bishop Williams wrote 40 or 50 years ago, and that was only one word. The latter gave then as its meaning "native." T. says now, "native, indigenous." Yet, I believe, anyone who takes the only course now open, and collects as many cases as possible of its use in the genuine language—the older of course the better—cannot fail of getting some additional insight. If I may judge from a small collection I have myself made, he will find: First, that it is not used as a race name, nor as a noun, but as a qualifying word, an adjective or an adverb. A Maori, as he is now called, was not he Maori, but he tangata maori. Secondly: That its most general or fundamental meaning seems to be 'common, ordinary,' sometimes with an implication of praise, sometimes of disparagement. A thing may be common, mere, unimproved, as contrasted with what is uncommon or superior—hence Irawaru's bird-spear was called maori, because it was plain, and would not hold the birds while page 54 Maui's was barbed, and would: Bo of a face not ornamented with the tattoo. Mr. J. White in this connection translates maori, untattooed, i.e., plain, common. So, also, the lesser stars are called whetu maori, Maori or common stars in contrast to Rehua, and the other great ones; so (in a Maori letter in 18G8) a steamer is called a steamer (tima), while an (English) sailing vessel is called a Kaipuke maori, or Maori ship, though it was maori only in being an ordinary ship, when contrasted with the steamer. On the other hand, a thing may be common in a good sense, normal, real, genuine, as opposed to the uncommon; abnormal, fictitious, having the appearance but not the reality; wai maori is fresh water, common, genuine water, in contrast to wai-tai or sea-water; so a man was a tangata maori, a common or real man, as opposed to other beings (sometimes even called tangata, men), patupaearehe (fairies), tupua (goblins, or monsters), or atua (gods)—human in appearance but not in reality, and particularly in later times as opposed to Europeans, who at first universally, I believe, throughout Oceania were looked on as gods, or monsters or ghosts, and not real men. Of the Melanesians, Dr. Codrington says, "When white men first appear to Melanesians they are taken for ghosts, dead men come back; when white men ask the natives what they are, they proclaim themselves to be men, and not ghosts "(Mel. Lang, 82 and 467). This is the same antithesis, which in Polynesia was expressed by the word maori, and possibly in Melanesia may be found traces of the word, as well as the notion, and when I sec among T.'s 'comparatives' the Mangarewan maori, said to mean Polynesian and Oceanic, I should like to hear more before accepting them as representing the ancient usage of the language.

215.—Marenganui.—Should not this be maringanui? I have long known the latter with the meaning "It is well" (and see G. P., 2G7 and 888).

216.—Marewa.—For ka marama ki runga read ka marewa, &c.

220.—Maruaroa.—W. says "the name of the third month." T. copies this, but adds, "a season answering to our June (early winter)," which, as an addition to the other, and not in substitution for it, is to me confusing. The authorities, I think, are agreed that the first Maori month corresponded roughly to our June. And, accordingly (in A. H. M. III. 80, and trans. 51) these two are taken as equivalent to each other, and also to Maruaroa. The Maori year, I understand, began with the new moon following the first appearance in the morning of Matariki (the Pleiades), or, according to some, of Puanga (Rigel); but the two rise here about the same time, the difference in Right Ascension being approximately compensated by that in Declination. Last year, 1891, the first Maori month, according to Hoani Nahi's Almanac, began on June sixth.

224.—Whaka-matuku.—For this read Whaka-mataku, obviously a misprint. But it helps to show the imperative need in Maori printing that copyist and proof-reader should have a keen eye, and a strong sense of the value of small differences. They, at all events, must not put into practice on their own account T.'s unqualified rule of the occasional interchange of every Maori vowel with every other. In the present case, the error—putting an u for an a—is not so serious as in many others; yet it might produce confusion, since, instead of "to terrify," which was aimed at, it substitutes "to turn into a little blue heron."

226.—Matariki.—T. says "the Pleiades, a constellation, the sign of the first month." This is true, but how is it a sign of the first month? T. does not say. But in the 'comparatives' the Tahitian Matarii the Pleiades is said to be "(b.) a year or season reckoned by the appearance of the Pleiades." But how reckoned? Here again T. does not say. The reader is left uninformed, and would be apt to suppose the Maori year and the Tahitian year began together. Yet, on the contrary, the two appear to have begun some six months apart, if we take the only Tahitian year mentioned by Mr. Ellis as regulated by the Pleiades. That celebrated constellation, indeed, by its appearance above the horizon, marked the beginning of both, but here when that happened in the morning, there when it happened in the evening. The Tahitians, according to Mr. Ellis (Pol. Res., 1, 87), besides other divisions of the year, divided it into two parts: the first, "Matarii above," when that constellation could be seen in the evening; the second, "Matarii below," when it could not be so seen. Something corresponding to this Tahitian practice seems to have been observed at Hawaii, on the other side of the Equator.

In the first ex., for te whitu o te tau, I should read te whetu, &c.; it is printed whitu in orig. (G. P., 254), but see the longer version at p. 308, and see also what is meant.

231.—Matou.—In the ex. "kohore ia matou, ko tona hakoro ia," as to kohore see ante under that word; and before matou insert the necessary i. For matua read matau.

232.—Mau, "9. to know, to recognise." This meaning, not in W. and new to me, T., I presume, found in the passage of which he quotes the first few words. I have added the rest that the reader may see, not the word only, but the sense in which it is used. Katahi ka mauria te taane e te wahine ra [ki roto ki te whare whatu kakahu ra tahutahu ai]. "Thereupon the husband was carried (mauria) by the wife into that weaving-house, that she might there tend him." T.'s mistake, no doubt, arose from an incautious reading of Mr. White's English, which is translation plus explanation.

239.—Me, with.—For pai rawa nga takitaki, &c., read pae, page 56 &c. Here, by the substitution of i for e, pai for pae, the meaning is reversed; a good instance of the need for accuracy in dealing with Maori vowels. It is the destruction of the fences, not their well-being, that is predicated.

239.—Mea.—T. gives as one meaning "a thing of no consequence; it does not matter," &c. But surely that is not mea " thing," but me a?=me aha? lit. 'What must be done? What does it matter? It does not matter.' Compare the common he a? What? for he aha?

246.—Moe.—For moenamo read moenanu.

250.—Mokotokupu.—For this (twice) read mokotukupu

268.—Niu.—T. says "a means of divination by throwing small sticks; the sticks so thrown." But also, though not mentioned, the upright stick or sticks thrown at. See for an account of the ceremony by one who had seen it, Te Rou, 54—5. In the Paimarire or Hauhau 'religion' the worshippers-post round which the believers trotted in a circle, was called niu, from the upright stick in the old niu ceremony. The first set up of these modern niu was the lower flagstaff at Kaitake, Taranaki,^* The Polynesian niu is the cocoa-nut tree, as well as its fruit.

271.—Nukarau.—For kei nukurautia koe read kei nukarautia tatou.

273.—Ngaeo.—For te kukume-toka read te kuku-moe-toka.

280—Ngatiuhatua.—T. says, "a name of the patupaearehe (fairies)." But this should be struck out, unless he has in reserve some evidence beyond what I would venture to call the misunderstanding he adduces in its support. It is arrived at by taking "that tribe" (the fairies) in one sentence to mean the same as "this tribe the Ngatiwhatua" in the next.

281.—Ngawari.—For mauna atu read mauria atu, and for vii. 5 read vii. 51.

284.—Ngoikore.—For he puta nga uri read ka puta,&c.

293.—Ori.—For ka oria i te hau read ka oria e te hau.

295.—Owha.—T. says, "to warn; warning, alarming," and supports this unexpected meaning by an example in which I am afraid it is not to be found. It is a passage from a farewell address to Sir G. Grey in 1853: ko a matou kupu owha enei ki a koe, which C. 0. Davis translates, "these are our words of love to you "(where 'words of love '=kupu owha); and they go on to quote the fifth commandment as to the obligation to love their parents, &c, As there is nothing of warning or alarm here, it will be safer for the student to accept T.'s meaning in a somewhat contrary sense, and translate owha farewell, the expression or feeling of love and respect, especially at parting.

296.—Pa.—For i kaika ana read e kaika ana, and in i runga ake te ate insert i after ake.

302.—Pai.—In he tino nui pai toku pai atu ki a koe, for the first pai read pu; and under whakapai, for rawatai read rawatia.

305.—Pakawha.—For to pakawha read te pakawha; insert "?" at the end, and in ref. for Rew. read Hopa.

306.—Pakete.—T. gives two words of this form: 1. "to be forced out; to shove out, to expel "; [both active and passive?] and 2. "a bow of the archer." The latter he speaks of as "a doubtful word: Murihiku dialect," I would suggest for further enquiry that they may both represent the English word bucket, introduced possibly in the South by the old whalers. As to the first, Dr. Shortland (Southern Districts of N.Z. p. 811) gives pakete-tia as "shoved out, done up, &c.," with which compare the meanings of 'bucket' in the New English Dictionary, e.g., ' to give the bucket'='to give the sack'; so 'bucketed '=' sacked' in the workman's sense? Again, 'bucketed 'as a horse by hard riding; 'pumped,' done up. As to the second, compare 'bucket,' the bent piece of wood on which a dead pig was hung, and which is suggested as the origin of the slang phrase "to kick the bucket." The provincial saying "as wrong, or as bent as a bucket" would apply to the archer's bow.

313.—Paoho.—Under this T. says "to bark at a pig is paoho, to bark at a man is tau." Is not this a considerable misunderstanding? Probably tau, to bark, is imitative, and regards mainly the sound; while paoho, I believe, regards rather the purpose or effect of the barking, to give the alarm, call attention, &c., by barking. Compare whaka-oho, to arouse, startle. Indeed, I would suggest that in pā-oho, the prefix pā is a causative=whaka in force, if not in origin, as in some other Maori words; and compare tā both in Maori and Tahitian.

314.—Papa.—For nana e takatakahi, &c., read nana i, &c.

323.—Pareramaumu—unable to swim.—This is taken by T. from W., but is it not in the latter a misprint for parera maunu? I know it in the latter form, and also with the elements transposed maunu-parera, as one who cannot swim. See also T.'s mounu with the same meaning.

327.—Patu.—T. quotes "e haere ana ahau ki te wahi e patua ma nga weru, and it is so in Trans. vii., 51; but as it stands is not e patua ma nga weru obviously printer's Maori for e patua ai nga weru? See it so printed in A. H. M. II., 131.

329.—Pawera.—For i mauri ai read i mariri ai,

332.—Peke.—T.'s first three meanings of this word, leap over, jump up, and leap down, may be conveniently and properly reduced to one, i.e., to jump, the rest of the sentence determining the direction. His sixth meaning, "to conceal," is new to me, but is not borne out by his example: ka peke a Pungarehu raua ko tana hoa ki runga ki te matao [noho ai]. "Pungarehu and his page 58 friend jumped up to the window, and there sat." J. White translates "sat at" the window. There was no concealment; quite the contrary. They were planning to kill the poua-kai, or man-eating bird, of this island, and were the bait in their own trap. The great bird made for them from a distance, and, when near enough, struck at one of them, and thus, having come within reach of their stone axes, it was killed.

332.—Pekepeke.—W. says, "quick"; T. says, "quick, swift, speedy," Is it superior knowledge, or the literary faculty which gives as three meanings for one, but instead of one example, none?

332.—Pena.—W. says, "like that which is near, or has some reference to the person spoken to." T. quite unaccountably alters this into "like something near or referring to the person spoken of." Thus, by the change of preposition, depriving the word of the whole peculiarity of its meaning.

340.—Pipiri.—For fro raua ko Papa read ko Rangi raua ho Papa.

340.—Piringi, a shelter.—This, I would venture to call one of T.'s most distinctly new words. He supports it by a quotation from the Maori Bible (of 1868?), where it certainly occurs (Psalms lxi., 3): Hei piringi hoki koe moku, &c., "For thou has been a shelter for me." Nevertheless, and in spite of T.'s adoption of it, I venture to think it is not a Maori word. T., himself, connects it with piri, but how? The connection would be of the greatest interest, but it remains his secret. I would add that piringa, the word one would expect, is the regular verbal noun of piri, piripiri, to stick to, keep close to, hide, shelter behind, and so that which shelters, cover in a military sense. Moreover, this piringa, and not piringi, is actually found in three older editions of the Psalms which I have examined, and also in this 1868 edition, in the very next verse to the one quoted. But T. is well enough satisfied with it to repeat it in his "Key," though there, indeed, under "sheltered," not "shelter," thus introducing another difficulty. Yet though piringi, as a verbal noun ending in i instead of a, would. I suppose, be unique in the language, T. does not think it worth while calling the reader's attention to it.

342.—Po.—T.says, "4. Eternity: kua mate ki te po; passed into eternity." These two phrases are equivalent so far as each means that some one has died, and no farther; take away mate from the one and eternity will go with it. The Maori mate is ambiguous; it may mean til or it may mean dead. If a Maori wants to make it clear he means dead he adds an intensive word or phrase, as mate rawa, or mate atu ki te. Po, or ki te Reinga, or to be very emphatic mate rawa atu ki te Po, and so on. And evidently several other words have as good a claim to the translated eternity, e.g., Te Reinga, Pae-rau, Waro. But page 59 supposing T. had to put into Maori a proposition which, upon reflection, he will, no doubt, admit to be true, that anyone who is now in time, is as much in eternity as be ever can be-would he make his meaning clear if for eternity he used the word he gives as its equivalent?

350—Popokonuihaura (Clematis).—I have heard this as Popokonui-a-Hura, which, I would suggest, as on the face of it a more probable form, since popoko=upoko.

352.—Pongi.—For whaono ana read whaona ana; whaono would involve a new departure in the Maori passive.

365.—Puanga.—For Tautoro read Tautoru.

369.—Pukana.—For he putanga read hei putanga, and for ka pu read ka pukana.

371.—Pukupuku.—In the ex. for he read hei, and in ref. for 9 read 29.

372.—Pukuka, a glutton.—For this read the ordinary puku-kai. The latter is given as a comparative under puku, but pukuka re-appears in the "Key."

386.—Raho.—For kohera read kohera, and for ko po read ka po.

395—Rangona, "(a passive form of Rongo), to hear," &c.—The parenthesis should be removed, as it gives to the passive the meaning of the active. But this change of passive to active will not surprise the student so much as the statement that ranqona is a passive of rongo, coupled with the subsequent statement (p. 423), "Bongo (passive rongona)," as if the latter not only existed in this sense, but were the ordinary form. If so, there has been 'a conspiracy of silence' about it, or worse, among our grammarians for half a century. Rango is, I believe, unique in the language in changing, as it seems, the first vowel of its radical in forming its passive; and one of perhaps two only which show apparent change in either vowel, the other, or one other, being mea.

403.—Raukataura.—For this read Raukatauri. The latter name is also given to the long tapering leathery case of a well-known moth (Liothula omnivora?) This case is said to bo the flute on which the goddess played; perhaps strictly te pu a Raukataura.

414.—Ririki.—For kotoa read katoa.

414.—Rikiriki.—T. copies from W., "In small portions or sections; in fragments"; but prefixes of his own, "in particles." His only example, however, is of the common general meaning, "small" (which he has put into a separate article, and to which it should be transferred). It is "Upoko rikiriki e!" This, as might be expected, J. White translates "Little heads." If T. translates it according to the meaning he cites it to support, "heads in particles, sections, or fragments," he encounters the difficulty that it is an address to living men, page 60 and misses the obviously intended antithesis of mata-nui, " big face," in the next clause.

423.—Rona,—Near foot of first column, for wairoa read waiora. Although Browning has given a mode t immortality to the river Wairoa,^* it must not be allowed to usurp the place of the Wai-ora-a-Tane, "the living water of Tane," which ought to be still more celebrated, and will be whenever it finds its sacred poet, since it was itself capabable of conferring immortality (of a kind) on those who bathed in it.

423.—Rongo, (passive Rongona), to hear.—As to passive, see Rangona, ante. T. gives (from W.J as one meaning, "tidings, report fame"; also, but not from W.. "sound, noise." He supports the latter with an example of the former, na, ka toe te rongo ki ona teina [taina, in orig.], which Sir G. Grey translates, as the words and the facts require, "At last the news," &c.

428.—Rou.—In ka tako kei raro, for tako read taka.

440.—Taepa, "a goblin, a spectre."—I have heard of a creature with a name sounding a good deal like this, for nearly 40 years, but always from Englishmen who did not talk Maori, or from Maoris who thought they were talking English. Is it genuine? The nearest approach to it I have hitherto seen in print is in the Maori Messenger for July, 1868 (No. 6, p. 10), where the English article, evidently the original, has "instigated by some typo," &c., but the Maori translation evades it by a circumlocution. Presumably the English writer thought he was introducing an evil spirit well-known to the natives, whose name would give point to the translation, but the translator did not see it.

441—Tahanga.—In katahi ka marama ake, for marama read maranga.

443.—Tahi.—In "korua pea ko te Arahore i haere tahi mai?" For Arahore read Arahori. The proverb is against those telling 'travellers tales,' &c Mr. Colenso (Trans. xii., 124) translates "Perhaps thou and False-road came here together?" so that the change of i to e—of hori, false, to hori, not—would just take away the point.

446.—Tai.—T., copying W., says, "an exclamation of address to a married woman. E tai!" Add, however, "or to a man," since it is used to both sexes. This, also, T. might have learnt from Maunsell's Grammar, or one acquainted with Waikato usage—not to mention Maori letters and newspapers.

447.—Taina.—T.'s distinctions between the two dialectic variants taina and teina have the advantages and disadvantages

^* See The Guardian Angel, last line but one. The "Alfred, dear friend" of that poem was the late Alfred Domett, for many years a Nelson resident, himself no mean poet, and one of the most original of the minds which helped to found this Colony. For Browning's view of what he might have become, see "Waring."

page 61 of novelty: I should advise the beginner to investigate the matter for himself before accepting them. So far as I have been able to test them since seeing this book, I do not find them founded in reality.

450.—Takahanga (for takahihanga).—So T. puts it. Yet he does not give us takahihanga (presumably in his view the right form) in its proper place. Is this because he has not found it actually existing, and only mentions it as the ideally perfect form? Yet the contrary view (which I hold) that takahanga is not only the actual, but the normal form of the verbal noun from takahi, is, whether right or wrong, at least suggestive, because it challenges the radical character of the last syllable of takahi.

453.—Takeke—Bishop Williams (ed. 1852) says "Takeke, to be acquired; Takeke noa nga tini kupu Maori i a koutou; all the native words have been acquired by you." In the last edition (1871) Archdeacon Williams changes this a little, "Takeke, a. altogether acquired," giving the same example, but, to the peril of the incautious, untranslated. T., omitting this example, and giving none of his own, now says, "Takeke, altogether acquired; not an original possession." Of this, the first half is borrowed; how was the second half arrived at? It seems to me, not by the 'philologic,' but by the 'literary' method; not by a re-interpretation of the Maori, for that it would not fit, but by a misinterpretation of the two English words constituting the first half.

458.—Tamaka.—For "fine strands" read "five strands."

460.—Whakatana—T. cites "ko Whakatau, potiki ahau, e whakatane ia ia." Here the omission of a letter confuses the whole sentence; for ahau, I, read ahaku, my. The same error occurs on p. 330, under peha.

472.—Tapoko,—W. says. "v.i., 2. Sink in mire"; meaning, I presume, by the difference in type that sink was the general meaning of which "in mire" was a particular case. T. copies the words but in one type "2. to sink in the mire "; thus wholly and erroneously specializing the meaning. And his example is peculiarly unfortunate, for, though at first sight it seems to fit, this has only been arrived at by cutting off the end of it which specifies the material sunk into, that being not mire but stone. "A, e ta poko ua [for ua read na] ano te taunga o nqa waewae o Hotumauea." [The feet of Hotumauea in alighting sank in.] If the quotation had been carried three words farther, so as to include i te kowhatu (as it is, curiously enough, on p. 484), it would have appeared that Hotumauea (who was something of a giant) had taken so prodigious a leap that his feet sank into the very rock itself, and the story goes on to tell that the marks are to be seen to this day. But T., unkindly, I submit, to all concerned, takes away the rock, and the wonder, and leaves the hero sticking in the mud.

475.—Tarahanga.—For i ringa read i runga.

477.—Taraheke.—For ka rangi road ka ranga.

480.—Tarie.—T. gives this as equivalent to taria, passive of tatari, to wait. Hitherto, as I understand, the authorities were agreed that every suffix forming a passive was or ended in a. T. now gives us one ending in e. If I am right he is thus taking an entirely new departure in the Maori language. He does it without a word of comment; and on what authority? That of the type-setter of vol. vii. of the New Zealand Transactions! and that particular part of the text is worse than most of it. See the same incident much better printed in A. H. M., II., 70, where taria appears, not taria.

481.—Taringa-here, "(myth) a fairy or elf with a face resembling a cat."—T. tells us this, but does not tell us how, in the long interval between their last sight of a cat before coming to New Zealand and their first sight of one here, the Maoris kept alive the memory of the cat-like face of the Taringa-here? I am assuming that it was not a parvenu of this present century, but a respectable aboriginal fairy of immemorial antiquity. Otherwise, I submit, T. is exciting our curiosity, and our wonder upon false pretences.

483.—Tatei—This, I presume, is meant for tātea (of which the first a is long, taa=tae). It is spelt tatea three times in A. H. M. III., and at p. 84 has the secondary meaning, not given by T., of 'off-spring.'

483.—Tatou.—In "matou, we, including person or persons addressed," for including read excluding.

490.—'launaha.—In katahi ka ratou ka taunahanaha, &c., omit the unintelligible ka before ratou,

491.—Taupuhipuhi.—T., says "to lean one on another," and this is right as far as it goes. His example is "Taupuhipuhi atu ra korua nei ki te hoa "; but the original has as the sense requires—"korua nei ko te hoa." Tau here is the prefix of mutuality or reciprocal action: "go you and your companion mutually embracing." The words were addressed to Sir George Grey in 1853, and referred to his leaving New Zealand in company with Bishop Selwyn. The image is that of two lovers going off arm in arm, or with arms about each other. Substituting, as in T.'s text, ki for ko would, I suppose, introduce a third person for the two others to support—quite another thing.

494.—Tautoru, "the constellation of Orion." Is this a true identification '? If so, it implies a great deal; if not, it is correspondingly dangerous. I will not ask how far the notions of any two ordinary Englishmen out of a hundred would agree as to the boundaries, or even the form of Orion if put on paper, but do the Maoris figure Tautoru as a huntsman or warrior, or at least as a human being? and if so, is he, when on the meridian, page 63 and most conspicuous, standing on bis head or his heels? As I understand the word Tautoru, it means "the Belt of Orion," or rather for even that is ambiguous the three bright stars in the belt. But even if we assign, as I should propose to do, a personal meaning to tau, the font makes a plural of it.

It is deplorable how little of Maori star-lore is on record, and of that little how much is doubtful, or worse. (Take, for instance, the Rev. R. Taylor's impossible account of what he calls the chief Maori constellation 'the canoe of Tamarereti,' of which the stern is the Belt of Orion (on the equator), the sail the Hyades, and the bow the Pleiades, conveniently, if curiously, placed on the far side of the South Pole, near the Southern Cross, which serves for an anchor. T. appears to take this seriously, at least as to the remarkable length of the canoe.) But a good deal might even yet be done if our Maori scholars in different parts of the country, where any learned, or half-learned Maoris are yet left, would cross-examine them in sight of the stars, and with a star atlas for reference.

498.—Tawharu, to bend &c. For wharau read wharua.

498.—Tawharu.—T. says "(South Island Dialect) Eight; the eighth: Hei tawhitu hei tawharu ka haere mat ia." I think T. here again attributes to the dialect of the South Island what really belongs to the Printer's dialect. I should say, therefore, for hei tawhitu read hei te whitu, and far hei tawharu head hei te waru. T. ought, I submit, to have seen, in spite of printer or scribe, that hei here required the definite article after it, and before the numeral, that therefore, ta having no other conceivable function where it is, was, in some way what was otherwise missing; and he should either have given it us as a new form of the article, or if he thought the evidence too slight for such a novelty, he should have treated it as te in disguise. He might also have arrived at the same result in another way, if he had looked at the same story, told in the same dialect, in a volume he has quoted from, more perhaps than any other—the first volume of White's Ancient History, where at pp. G2 and 66 exactly the same expressions occur, only spelt as usual hei te whitu and hei te want.

500.—Tawhitu, " Seven, seventh.'—This as shown above should be omitted. A further point not there mentioned is that while T. treats the ta of tawhitu and tawharu as negligible when these words have a cardinal value, he allows it the full function of the article when they are regarded as ordinals.

503.—Tekau ten.—One Maori word T. compares with this is kau."to swim"; he does not say why, and I certainly cannot guess. The main idea which he seeks to establish by his other comparatives, that kau represents an original Polynesian word meaning 'collection, assemblage' has a good deal of evidence page 64 in its favour, and the suggestion is quite reasonable that this is its meaning in tekau. But the discussion must include many more facts, and take generally a much wider range, and will involve a much more critical inquiry, before this can be deemed even provisionally settled. Probably kau was once of very general meaning, and has assumed, or rather is one of, several varying forms. And there is, I think, a strong case for the suggestion that the kau of tekau may be the same as that in rakau—as to the composition of which latter word see Dr. Codrington's instructive observations (Mel. Lan., p. 95.); also, on the general question, his chapter on Melanesian Numeration. In the latter the Maori word rau meaning '100,' but also meaning 'leaf,' may find explanation since some Melanesians use a cycas frond having many leaflets with which they count in several cases, by turning down a leaflet as a tally for every 10, so that in the examples he gives the whole leaf= 100 (p. 249). Speaking of tally suggests the observation that while it is our duty to go abroad for every relevant fact we can find, it does not become us to ignore important facts at home. If T.'s unaccountable aversion to Maori grammar had not kept him from making himself familiar with Archdeacon Maunsell's admirable work, and the very useful grammar prefixed to Bishop Williams's Dictionary of 1852, he would have seen that these two of our oldest and best authorities concurred in saying that with some of the Natives ngahuru meant ten and tekau eleven; Bishop Williams saying that they counted by elevens, the eleventh being a tally, and he compares our "baker's dozen." There is a proof also that the record of some of the most important of their counting was kept by a "tally," in, I suppose, the etymological sense of the word, a notched (stick), that is, their rakau whakapapa, or stick on which the successive generations were marked by notches.

509.—Whaka-tika.—T. says, "3. to correct, to put right." His ex., however, appears to be of a nearly opposite meaning, "to admit an accusation"; me whaka-tika atu nga whakahe" [mai a to matua teina a Te-tauri], which J. White translates, "And we will own the truth, and now admit the error which Te-tauri charged us with," T. takes from W. his first and second meanings of whakatika, strighten, straighten oneself, stand upright; but without apparent reason omits the last "rise up; start on a journey '; when he ought rather to have added "and so go or come, as atu or mai is used."

520.—To, to drag.—For na wai e to? a combination which ought to have excited somebody's curiosity, read ma wai e to?

521.—To, "up to, as high as."—T. takes this from W., but omits his example, and gives no other. Yet we specially want as many cases as possible in which this word occurs, for it seems an open question whether it is a whole word or half a word. Is it ever found without nga after it? If not, is this nga page 65 really the plural article or an integral part of the same word, thus making it tonga? (see M. Gr. 84). There are difficulties on either view, and the first step is to bring together all possible examples.

524.—Tohunga.—T. says, "3. The soul or intelligent spirit of a human being." The example in support, however, is insufficient, being ka hutia te ki runga ki a Rona (to which the reference is only "C. O D.," but should be M. M, 167). It is not, I venture to say, really founded on this passage, but on a mistaken view of C. O. Davis' free translation of it," "the spirit of the chieftain is taking its flight to Rona." The poetical license of translating tohunga, 'chieftain' instead of priest,' and in representing that his spirit, and not his body, went up to Bona (the man in the moon) was not very great, but it was enough apparently, to mislead T. Compare J. White's more literal translation of the same line A. H. M. II., 20, "The high priest is lifted up to Rona."

525.—Toi.—T. copies W. in saying, "to trot, to move briskly"; but seems again very unfortunate in his example:—Kei te toi poto, a, i te ata kei te toi roa. This, taking it as illustrating the only meaning given of toi, must, I presume, be translated: "at the short trot, and in the morning at the long trot," which is not lucid, whether the fault is T.'s or mine. The facts are these:—The great priest Ngatoro had gone back to Hawaiki to destroy the 'multitude of Manaia." He asks his sister, who is married to Manaia, as to their habits. The quotation above is her answer to the question, where are they in the evening? "At the short trot, and in the morning at the long trot." Evidently something different or something more is wanted in the meaning to be given to toi. The proverb waiho i te toipoto, if it means "keep together, be united," suggests the secondary meaning T. was wanting.

541.—Tua.—W. says, "a form of address used by the Ngatiwhatua tribe "; of which name it will be observed it forms the last part—the Maoris often familiarily using the first or last part of a name as an address instead of the whole—but, of course, it is only appropriate where the full name is appropriate. T. defines it as "a word of address to a man"; thus improperly suggesting its generality, and concealing its origin.

543.—Tuapae,—T. has copied his example faithfully—too faithfully, I think. In te tuapae o utu, is not utu an obvious misprint for utu?

546.—Tui.—For tuia tu tatou waka read to tatou, &c.

554.—Tungou. "To nod, to beckon." 2. "To nod the head as a sign of dissent."—If T. has authority for this second statement he ought to have given it, since he is reversing the ordinary meaning of the word. A Maori nod is a motion of the head upwards or backwards, and means, on meeting a slight page 66 salutation, and in conversation, assent; raising the eyebrows has the same name and the same double meaning. Dr. Shortland (S.R., 84) quoting a saying of Thenga's Te rakau e takoto nei, tungou, tungou, "O tree lying there raise your head, raise your head," says, tungou = , a sign of dissent with the Greeks, but the common sign of assent with the Maori."

559.—In the first column near the bottom, for Tuhirangi read Tahurangi, and for Tahurangi read Kawhia. Two statements also here require correction when compared with the authority cited to support them (G. 8.—16.) T. says: "Neither of them" [Tumutumuwhenua and his wife Repo] "were of the people of this world; they were of the Tuhirangi" [read Tahurangi] "fairy people." Each clause is incorrect. Tumutumu was not of the people of this world, nor was he therefore a Tahurangi, since they were of this world. Repo, on the other hand, was a Tahurangi, and was therefore of the people of this world, though not of the human race.

564.—Turu—In this case T. takes word, meaning, example, and translation from W. His own contribution is to take the passive turua, separate it from its active (or rather its simple form) as if a new word, omit to state it is the passive or a passive at all, and omit W.'s ex. without giving another.

568.—Tuwha, to distribute.—This is from W., but another form tuachawha is coupled with it. The latter is not supported by authority or example, nor I would say by analogy. Both vowels of tuwha (of which another common form is tuha) are short, and in that case it is, to say the least, very rare, and presumably abnormal, to find the second syllable reduplicated; and a real instance of it would deserve comment.

In the second ex, for paka read papa, and for XIV. read XII.

579.—Whaka-uru.—T. says "2. To fasten together." If that is so, it is not supported by the example he cites for the purpose: "Ka whakaurua ma ratou i ami taura," &c. In this, for ma ratou read ma roto. When the famous canoe Tainui arrived in New Zealand, her people found a stranded whale, round which, unfortunately, those before them had tied flax ropes as a sign of ownership. But the Tamui people were not to be beaten. They also made flax ropes, and dried them at the fire, thus making them appear older. These ropes were then tied under (ma raro i) all the others, they were inserted within (whaka-urua ma roto i) those other ropes, thus making further evidence of having been tied on first. And they got their whale, though not by 'fastening their ropes together.'

581.—Uraroa.—In kei mate Tarakihi, though so printed in Trans. XII. p. 139, I should propose to insert a before Tarakihi; and for kia matenga ururoa though so printed in W. I should propose to read kia mate a ururoa, treating matenga as an obvious page 67 misprint in W. and making the Maori agree with W.'s translation which T. had before him, as well as the common form of the proverb from Mr. Colenso's paper.

584.—Waenga. "4. The mainsail of a canoe.—For the reasons for striking this out see, ihu, ante.

585.—Waea.—W. says, v.i., 'be weary,' and gives an example. T. says, 'to be tired; weary '; and also gives an example: "Ka waea te kanohi, kei te tirohanga atu; G. P., 62." But the original has waia, as it evidently requires, T.'s ex., therefore, should be corrected, and put under waia on p. 589.

585.—Waero, "4. enemies, inimical, hostile."—Is this, in this sense, a genuine Maori word? In the Taranaki war the hostile natives were often, though I think in a half jocular sense, called waero, but was not the extended name waero-mene, otherwise wild-men?

589.—Wai.—T. says, "mi, who?" adding a remark not to be found in W., that "wai is generally preceded by ko." He had better have said that it is generally preceded by some particle: by ko when that is appropriate; oftener, probably by a, since that occurs with most of the oblique uses of wai, and often when it is in the nominative. But it is also, of course, often preceded by na, ma, ta, to, o, no or mo, according to the sense to be expressed. Should not a dictionary-writer refrain from remarks which, while they cannot benefit those who have an elementary knowledge of the language, may easily confuse the mere beginner?

591.—Wairo (for waero), a tail, &c.—Has T. any authority for this beyond some unfortunate misprint? He says, see waero, but there it is not even mentioned; while analogy and all his comparatives are against it.

594—Wānanga—For 'Laura read Tauira.

598.—Wee, water.—A startling novelty given without example, authority, or remark.

607.—Whakariki, a "war cry."—And we are gravely told to compare "whakaariki a war-party of the enemy," as if they were two words instead of two ways of writing one word which, in each case, has the same pronunciation. The difference, I apprehend, is not so great as if in English "enemy" was called a war-cry," and "en-e-my" a hostile war-party. As I understand it, the cry ko te whakaariki! (lit. the enemy!) is not, in the common acceptation, a war-cry, but was a cry of alarm when the enemy was suddenly discovered close at hand.

623.—Whiowhio.—Whiowhio. ventris read eructatio ventris; a previous and worse case of eccentricity in the Latin language, I preferred to let alone.

In addition to the foregoing I have noted a good many other misprints, mis-copyings, and other eorrigenda. That is without counting the very numerous errors in relation to marking page 68 the length of the vowels; errors, which, as I have said, are equivalent to the serious mis-spelling of English words, a grave fault in a dictionary where they are professedly marked.

Mr. Tregear gives a second part to his dictionary, calling it "Key to the Maori Words." It consists chiefly, though not avowedly, of the second or English-Maori part of Archdeacon "Williams' Dictionary, with some additions, some trifling alterations, and a new name. 1 do not propose to discuss it.

It will have been seen that my own answer to the question 1 proposed as to the author's fitness for the arduous work he undertook, and as to the trustworthiness of the book he has given us, is, on some important points, distinctly in the negative. I think he began the book long before he was ready, began his building before he had nearly laid his foundation, and, having begun it, allowed himself far too little time in which to do it, even as he himself might have done it, if he had not felt it a matter of such urgency to get it published. It shows, as I think, throughout that part which I have particularly examined, constantly recurring evidence of insufficient knowledge and hasty work. And if that is so as to Maori, it is not to be presumed he has escaped error in dealing with the many other languages with which he is, to say the least, less familiar.

In spite, therefore, of the author's manifest enthusiasm, and the large amount of useful material which his industry has brought together, I fear that he has done almost as much to hinder as to promote the cause he would serve; that until his work has undergone a complete revision, and been to a large extent recast, it can only with very large and indefinite reservations be accepted as trustworthy; and that to those who consult it, instead of being a uniformly safe guide, it will often prove a source of serious danger, and in direct proportion to their need for its help.

One other word in conclusion. The great English Dictionary now being published at Oxford is, we are told, founded mainly on materials collected during more than 25 years by the Philological Society. Up to 1884, when the first part appeared, 1300 persons had lent their help in collecting three and a half millions of illustrative quotations—and, according to present appearances the preparation of the collected materials for the press seems likely to occupy another 20 years or so. Now, if it is permitted to compare small things with great, I would say that in this great example, all who are interested in such a Maori dictionary as we might have, may find a twofold lesson—not to delay the beginning of the book, and not to hurry the end.

Will not then the Polynesian Society organise the collection of material for a really comprehensive Maori dictionary, which shall page 69 be illustrated throughout from the other Oceanic languages after an adequate discussion of the principles determining the relationship of words, and other necessary points in structure and idiom; collection and discussion going on together to their mutual profit?

One other lesson we must draw from the same example: the imperative need for co-öperation in such a work. This is especially true of a language a large part of which has still to be sought from the widely-scattered speakers of it themselves, who, moreover, as has often been said, and still more often regretted' are rapidly losing their old knowledge of it—that is, I suppose! not less than half of it in quantity, and how much more in value?

But there is no need to urge the importance of the work to be done, nor the need of co-öperation in doing it, on a Society whose existence itself is the clearest recognition of both.

There are, no doubt, a good many in New Zealand, and outside, who would willingly respond to the Society's appeal for help. And such a dictionary would evidently not be merely of local value, but would be of inestimable advantage to all students of the Oceanic languages. It would go a long way towards giving an insight into a large number, if not all, of these languages, and towards solving some of the most interesting problems concerning the several races which speak them.