connexion) is a ..b..c? Again, the work is in great part planned with careful regard to logical relation. The Buddhists had not elaborated the intellectual vehicle of genus and species, as the Greeks did, hence they had not the convenience of a logic of Definition. There is scarcely an answer in any of these Niddesas but may perhaps be judged to suffer in precision and lucidity from lack of it. They substitute for definition proper what J. S. Mill might have called predication of æquipollent terms-in other words, the method of the dictionary. In this way precision of meaning is not to be expected, since nearly all so-called synonyms do but mutually overlap in meaning without coinciding; and hence the only way to ensure no part of the connotation being left out is to lump together a number \of approximate equivalents, and gather that the term in question is defined by such properties as the aggregate possesses in common. If this is the rationale of the Buddhist method, the inclusion, in the answer, of the very term which is to be defined becomes no longer the fallacy it is in Western logic. Indeed, where there is no pursuit of exact science, nor of sciences involving 'physical division,' but only a system of research into the intangible products and processes of mind and character, involving aspects and phases, i.e., logical division, I am not sure that a good case might not be made out for Buddhist method. It is less rigid, and lends itself better, perhaps, to a field of thought where a difference in aspects is a difference in things."

However that may be, the absence of a development of the relation of Particular and Universal, of One and All, is met by a great attention to degree of Plurality. Number plays a great part in Buddhist classes and categories.2 Whether this was inherited from a more ancient lore, such as Pythagoras is said to have drawn from, or whether this feature was artificially developed for mnemonic purposes, I do not know. Probably there is truth in both alternatives.

1 Professor J. Ward, Ency. Brit., 9th ed., 'Psychology.' 2 Cf. especially, not only Book II. of this work, but also the whole of the Anguttara.

But of all numbers none plays so great a part in aiding methodological coherency and logical consistency as that of duality. I refer of course especially to its application in the case of the correlatives, Positive and Negative.

Throughout most of Book II. the learner is greatly aided by being questioned on positive terms and their opposites, taken simply and also in combination with other similarly dichotomized pairs. The opposite is not always a contradictory. Room is then left in the universe of discourse' for a third class, which in its turn comes into question. Thus the whole of Book I. is a development of the triplet of questions with which Book III. begins (a-kusalam being really the Contrary of kusalam, though formally its Contradictory): What is A? What is B? What is (ab), i.e., non-A and non-B? In Book III. there is no obvious ground of logic or method for the serial order or limits observed in the Clusters' or Groups, and the interpolated sets of 'Pairs' of miscellaneous questions. Nevertheless a uniform method of catechizing characterizes the former.

Finally, there is, in the way of mnemonic and intellectual aid, the simplifying and unifying effect attained by causing all the questions (exclusive of sub-inquiries) to refer to the one category of dhamma.

There is, it is true, a whole Book of questions referring to rupam, but this constitutes a very much elaborated sub-inquiry on 'form' as one sub-species of a species of dhamma-rūpino dhamma, as distinguished from all the rest, which are a-rupino dhamma. This will appear more clearly if the argument of the work is very concisely stated.

Those who can consult the text will see that the Matika, or table of subjects of all the questions (which I have not held it useful to reproduce), refers exclusively to Book III. Book III. in fact contains the entire work considered as an inquiry (not necessarily exhaustive) into the concrete, or, as one might say, the applied ethics of Buddhism. In it many if not all fundamental concepts

are taken as already defined and granted. Hence Books I. and II. are introductory and, as it were, of the nature of inquiry into data. Book II. is psycho-physical; Book I. is psychological. Together they constitute a very elaborate development, and again a sub-development, of the first triplet of questions in Book III., viz.: dhamma which are good, i.e., make good karma, those which are bad, and those which make no karma (the indeterminates). Now, of these last some are simply and solely results1 of good or bad dhammā, and some are not so, but are states of mind and expressions of mind entailing no moral result (on the agent). Some again, while making no karma, are of neither of these two species, but are dhamma which might be called either unmoral (rupa m),3 or else supermoral (uncompounded element or Nirvana). These are held to constitute a third and fourth species of the third class of dhamma called indeterminate. But the former of the two alone receives detailed and systematic treatment.

Hence the whole manual is shown to be, as it professes to be, a compendium, or, more literally, a co-enumeration of dhamma.

The method of treatment or procedure termed Abhidhamma (for Abhidhamma is treatment rather than matter) is, according to the Mātikā, held to end at the end of the chapter entitled Pitthi-dukam or Supplementary Set of Pairs. The last thirty-seven pairs of questions and answers, on the other hand, are entitled Suttantikadukam. They are of a miscellaneous character, and are in many cases not logically opposed. Buddhaghosa has nothing to say by way of explaining their inclusion, nor the principle determining their choice or number. Nor is it easy to deduce any explanation from the nature or the treatment of them. The name Suttantika may mean that they are pairs of terms met with in the Dialogues, or

1 Book I., Part III., ch. i.
3 Book II.

4

5 §§ 1296-1366.

2 Ibid., ch. ii. Appendix II.

in all the four Nikayas. This is true and verifiable. But I for one cannot venture to predicate anything further respecting them.

V.

On the Chief Subject of Inquiry-Dhammā.

If I have called Buddhist ethics psychological, especially as the subject is treated in this work, it is much in the same way in which I should call Plato's psychology ethical. Neither the founders of Buddhism nor of Platonic Socratism had elaborated any organic system of psychology or of ethics respectively. Yet it is hardly overstating the case for either school of thought to say that whereas the latter psychologized from an ethical standpoint, the former built their ethical doctrine on a basis of psychological principles. For whatever the far-reaching term dham mo may in our manual have precisely signified to the early Buddhists, it invariably elicits, throughout Book I., a reply in terms of subjective consciousness. The discussion in the Commentary, which I have reproduced below, p. 2, note 3, on dhammarammanam, leaves it practically beyond doubt that dhammo, when thus related to mano, is as a visual object to visual perception-is, namely, mental object in general. It thus is shown to be equivalent to Herbart's Vorstellung, to Locke's idea-' whatsoever is the immediate object of perception, thought or understanding' -and to Professor Ward's 'presentation."1

The dham mà in question always prove to be, whatever their ethical value, factors of citta m used evidently in its widest sense, i.e., concrete mental process or state. Again, the analysis of rupam in Book II., as a species of indeterminate' dhamma, is almost wholly a study in the phenomena of sensation and of the human organism as sentient. Finally, in Book III. the questions on various dhamma are for the most part answered in terms of the four mental skandhas, of the cittani dealt with in Book I., and of the springs of action as shown in their

1 Ency. Brit.,' 9th ed., art. 'Psychology.'

effect on will. Thus the whole inquiry in its most generalized expression comes practically to this: Given man as a moral being, what do we find to be the content of his consciousness?

Now this term dham mo is, as readers are already aware, susceptible of more than one interpretation. Even when used for the body of ethical doctrine it was applied with varying extension, i.e., either to the whole doctrine, or to the Suttantas as opposed to Vinaya and Abhidhamma, or to the doctrine of the Four Truths only. But whatever in this connexion is the denotation, the connotation is easy to fix. That this is not the case where the term has, so to speak, a secular or 'profane' meaning is seen in the various renderings and discussions of it. The late H. C. Warren in particular has described the difficulties, first of determining what the word, in this or that connexion, was intended to convey, and then of discovering any word or words adequate to serve as equivalent to it. One step towards a solution may be made if we can get at a Buddhist survey of the meanings of d hammo from the Buddhists' own philosophical point of view. And this we are now enabled to do in consequence of the editing of the Atthasālinī. In it we read Buddhaghosa's analysis of the term, the various meanings it conveyed to Buddhists of the fifth century A.D., and his judgment, which would be held as authoritative, of the special significance it possessed in the questions of the Dhamma - sangani. 'The word dhammo,' runs the passage (p. 38), 'is met with [as meaning] doctrine (pariyatti), condition or cause (hetu), virtue or good quality (gu ņ o), absence of essence or of living soul (nissatta - nijjivatā),' etc. Illustrative texts are then given of each meaning, those referring to the last being the beginning of the answer in our Manual

Cf., e.g., Oldenberg, 'Buddha,' etc., 3rd ed., p. 290; Warren, 'Buddhism in Translations,' pp. 116, 364; Kern, 'Indian Buddhism,' p. 51, n. 3; Neumann, 'Reden des Gotamo,' pp. 13, 23, 91; Gogerly, 'Ceylon Friend,' 1874, p. 21.

с