L'ensemble annoncé dans le titre comporte les mots suivants qui peuvent remplir, selon le contexte, les fonctions de nom, verbe ou adjectif: srid ou srid-pa et sri; sprul ou sprul-pa et 'phrul; so, sos, gso-ba, 'cho-ba, bços. Bien entendu, les divers elements de cet ensemble relevent aussi d'autres ensembles séinantiques qui coupent le premier. Leur groupement peut paraître arbitraire; il est le fait de l'auteur de cette note qui a cru y déceler une certaine homogéneité de conception. Chacun de ces mots appartient à une famille ou à, un arbre généalogique dont les ramifications peuvent former avec d'autres un réseau compliqué. Le regroupement proposé vise à mettre á jour des connotations inconnues ou insuffisamment définies par les dictionnaires. L'auteur a eu pour principe de se laisser guider par les associations trouvées dans les textes mêmes, les contextes, plutôt que d'élaborer des définitions basées sur l'usage exclusif des dictionnaires. II est pleinement conscient du fait que les auteurs de ces textes appartiennent à une époque relativement tardive de l'histoire de la langue. Mais il pense aussi que, si ces auteurs ne pouvaient sûrement pas avoir une connaissance consciente de l'étymologie des mots qu'ils employaient, ils on dû être guidés inconsciemment, à. leur insu, par le sens de leur propre langue que le savant européen ne peut avoir qu' à, un faible degré. Certes les rapprochements faits par les auteurs indigénes, les connotations ou associations d'idées qu'ils ont retenues ne reflètent pas nécessairement l'étymologie réelle (remontant historiquement à la formation du tibètain en tant que membre de la grande famille des langues tibéto-birmanes).

Madame C. Caillat in a magistral article in Journal Asiatique, CCXLVIII, 1, 1960, 41–64, followed by a note in JA, CCXLIX, 4, 1961, 497–502, has in establishing the meaning of Pa.1 phāsu- confirmed its connexion with sprś. I question Mme. Caillat's exposition on one point only, namely the exact form of the original derivative of sprś lying behind Pa. phāsu-. Like the previous scholars whom she quotes—Kern, Hoernle, Pischel, and Schwarzwald—she assumes the original form to have been *sparśu-, the resultant *phassu- becoming phāsu-, although recognizing that examples of such a change in Pali are rare; and for most of these an alternative explanation will be offered below. Nevertheless in Prakrit and, according to Pischel (Gr. Pk., § 62), with special frequency in Ardhamāgadhi the first geminate to be shortened with lengthening of a previous vowel was -ss- (Turner, BSOAS, XXXIII, 1,1970,171). The Jaina texts in Ardhamāgadhī have both phāsaī ∽ Pa. phassati < *sparśati and phāsuya- ∽ Pa. phāsu(ka)-; and it is tempting to see the saffle origin for ā in both Ardhamāgadhī words phāsai and phāsuya-. But the existence of Pa. phāsu(ka)- invites the assumption of another origin than spass- both for it and Pk.amg. phāsuya-, for according to Mme. Caillat (JA, 1960,47) the antiquity of Pa. phāsu- is assured by the occurrence of aphāsu- in the Pātimokkha as well as by phāsu- in the Aśokan Calcutta-Bairāt Minor Rock Edict.

One of the most striking developments during the T'ang period is the gradual improvement of the position of merchants and artisans, traditionally held in low esteem, and the parallel change in the attitude of government and the ruling élite towards them. These changes, which I have described in some detail elsewhere, led in the last century of the T'ang and the following period to the gradual acceptance of the desirability, and indeed the necessity, of levying taxation from the merchants and from the urban population. Th is in turn implied the tacit abandonment of the long-established ‘physiocratic’ doctrine that the state revenue system should be based exclusively upon taxes and labour services levied upon the peasantry. The document which forms the subject of this paper is perhaps the fullest and most closely argued statement by a T'ang statesman of the traditional position on this matter.

Among the many ballads and popular songs of Canton appears one Southern Air called ‘Apart from Number Nine’. Several stories are centred about the composition of this air. Around the 1820's and 1830's, the most sought after singing-girl in Canton seemed to be one named Number Nine. In a recent article by Jean Yow-wen (Kan Yau-man) there is a version of this song (pp. 25–6), attributed to Jrip Mrerng-shanqx with its background history. A few lines of another version of the same song were mentioned earlier by Sirn Jruk-chenqx This version, attributed to Zhiw Zir-jrunq (Jau Tzyy-yong) appeared side by side with an apparently impromptu composition by the girl herself, unmistakably addressed to Zhiw. There is yet another version which has been taped for me from a broadcast by Eadio Hong Kong. It differs greatly from the longer version given by Mr. Kan in his article. For the discussion of authorship it is necessary to include these pieces for comparison.