¹ BSOAS, xxi, 3, 1958, 553–77Google Scholar. I have since been able to look at the concept again and to gather further material on a study leave spent in Taipei from February to August 1972. I should like to take this opportunity to express my gratitude to the University of Melbourne and to the Myer Foundation for their support and to the National Taiwan University for the hospitality and help I received during my stay in Taipei.

In my ‘Verb complex’ I acknowledged my indebtedness to G. B. Milner for suggesting the term ‘moot’, which is used very much in the sense in which it is defined in the Oxford English dictionary. I also drew attention to Halliday's, M. A. K. ‘context of mention’in his ‘Systematic description and comparison in grammatical analysis’, in Studies in linguistic analysis. Special volume of the Philological Society, Oxford, Basil Blackwell, 1957, 61Google Scholar.

My indebtedness to Professor Chao's, Y. R. masterly work, ‘A grammar of spoken Chinese’, Berkeley, University of Oalifomia Press, 1968Google Scholar, will be evident throughout this paper. He has put us all very much in his debt.

² I only had an opportunity in the last two weeks before writing this paper of looking at Hashimoto's, Anne Y. most interesting pre-publication paper ‘Mandarin syntactic structures’, to be published in William Wang, S.-Y. (ed.), Readings in Chinese linguistics, University of Chicago PressGoogle Scholar. If I understand her correctly I find some support for my remarks on subject and object in her transformations and a very different view on the ‘passive’. But it should be stressed that a M structure is not a ‘transformation’.

³

⁴

⁵ A number of dictionaries of cherngyeu are available in Chinese. A recent Chinese-English dictionary is Yen-k'ai, Huang, A dictionary of Chinese idiomatic phrases, Taichung, Tunghai Ch'u-pan-she, 1969Google Scholar.

⁶ It was customary for elderly ladies to wind long bandage-like cotton strips around their legs puttee-wise.

⁷ I am indebted to my good friend Admiral Chao Chih-lin for these examples. A rich collection of shyehowyeu has been assembled in Tzu-sbih, Ch'en, Peip'ing hsie-hou-yu tz'u-tien, Taipei, Ta Chungkuo T'u-shu Kung-szu, 1971Google Scholar. Cf. also Schmitt, E., ‘Fünfzig Hsieh-hou-yü aus T'ai-yüanfu’, Asia Major, ix, 1933, 568–79Google Scholar, and ‘Pekinger Hsieh-hou-yü’, Sammlung Archiv für Ostasien, I, 1948, 15–19Google Scholar.

⁸ One-word sentences which do not contain a verb and indirect objects in preverbal position must be presumedly be M structures but they have not been examined here. One-word sentences containing a verb complex are sentences without a sabject and are covered by the definition in § 4.1, p. 392. One-word sentences not containing a verb are described in Chao, , Grammar, 60–7Google Scholar.

⁹ cf. Simon, H. F., ‘Two substantival complexes in standard Chinese’, BSOAS, xv, 2, 1953, 340Google Scholar, where a de-segment, there abbreviated Deseg, is described as the words preceding de when the de-segment is in colligation with a noun in the colligations [ds d c n] or [d c ds n], e.g. been hao-de i been shu or i been heen hao-de shu ‘a very good book’. Note that I now prefer the term classifier, abbreviated ‘c’, to determinator (Dor), the class determinative is now abbreviated as ‘d’ instead of the ‘Dave’in the earlier text, and Deseg becomes ds.

¹⁰ Chao, Y. R., Readings in sayable Chinese, San Francisco, Asian Language Publications, Inc., 1969, I, 76–97Google Scholar. It is to be noted that ‘Fragments of an autobiography’ is paginated 1–200 in the second part of Vol. I.

Professor Chao does not say how he achieved his ‘sayable’ style, but the autobiography is similar in style to the ‘polished’ versions of continuous recordings I have made recently in Taiwan. To achieve a ‘polished version’ the informant was first asked to talk freely and at length about some event or series of events or to present an argument on some issue. He then transcribed the recordings exactly as they came off the tape. This became text A, the ‘unpolished’ text.

He was then asked to reconsider the transcript and to reword sentences or larger piece at leisure, so that the text remained a spoken text but lost what he might consider to be false starts, inelegancies, or needless repetitions.

Mr. Chou K'e Ch'iang, who was kind enough to act as informant for me in Taiwan, has gone over the text with me in some detail, and has altered it when it contained words or idioms or sentences which he would not use or expect to have heard in Peking. These alterations were very few—a tribute to the authenticity of the text. Where they are made in the citations they have been specially noted. The references throughout are to the character text. The GR versions differ in minor conventions.

¹¹ For a description of the noun complex (N), the determinate complex (A) (abbreviated as Date in the original article), and the verb complex (V), see my ‘Substantival complexes ’and my ‘Verb complex’. In summary a complex is a structure mid-way between word and clause and is established by the colligations of what might be termed the head word of the complex, in Chinese the noun (n), the determinate (a), and the verb (v). The capital letters N, A, V denote any form of the complex, either with or without the head word.

¹⁸ of. Chao, , Grammar, 692Google Scholar.

¹³ ‘Substantival complexes’, 338, in conjunction with p. 384, n. 9, above. The class classifier is described as occurring in the colligation [determinative classifier noun] [d c n] and does not, therefore, correspond to Chao's ‘measure’, which includes what I would distinguish as determinates. Determinatives differ only in minor features from Chao's comprehensive description of the class, cf. his Grammar, 564–84.

⁴ dihi-duenn, under our definition would be described as an indirect object. Cf. § 3.1, p. 388, below, and p. 384, n. 8, above.

¹⁵ ‘Substantival complexes’, p. 333, II (ii). Chao would describe shie as a measure rather than as a determinative. Note that the complex form is, in fact, [d-ns].

¹⁸ ‘Substantival complexes’, 336–9, and especially the first paragraph of p. 338.

¹⁷ ‘Verb complex’, 573–6. See also Chao's, ‘auxiliary verbs’, Grammar, 731–48Google Scholar.

¹⁸ ‘Verb complex’, 571–3. See also Chao's, ‘pre-pivotal verb’, Grammar, 125, 706, 729Google Scholar.

¹⁹ ‘Verb complex’, 565–8, and 562,563. See also Chao's, ‘verbs in series’, especially Grammar, 335–50 and 754–69Google Scholar.

²⁰ ‘Verb complex’ 561, 562.

²¹ Chao, , Readings in sayable Chinese, p. 48, I. 4Google Scholar.

²² ‘Verb oomplex’, 577, and 561, 562.

²³ Chao, , Grammar, 692 and 729Google Scholar.

²⁴ Chao, , Grammar, 692Google Scholar.

²⁵ Chao, , Grammar, 729Google Scholar.

²⁶ Chao, , Grammar, 692Google Scholar.

²⁷ I hope to offer a more detailed description of this category at another time. It was described as ‘extension’(E) in ‘Verb complex’, § 5.3, pp. 558–9, although the present definition is somewhat different.

²⁸ cf. Chao, , Grammar, 316Google Scholar.

²⁹ An isolate which occurs with a determinative in the colligation [determinative determinate] and cannot be followed by a noun, e.g. nah-tian ‘that day’; cf. ‘Substantival complexes’, 351–5.

³⁰ Chao, , Grammar, § 5.4, pp. 301–25, and especially 308–9Google Scholar.

³¹ cf. Chao, , Grammar, 73Google Scholar.

³² I was sitting in someone's office when he said this into the telephone. Some informants have found it difficult to accept.

³³ Chao, , Grammar, § 5.4.6.1, pp. 312–15Google Scholar, ‘A cognate objeot may consist of an expression for (a) the number of times of an action, (b) its duration, (c) its extent, (d) the course of locomotion, or, less often, its destination’. Under our definition, all the examples except dahbuleau duoshao, listed under (a), (b), and (c) are determinates. The nouns listed under (d) would be considered direct objects.

³⁴ Chao, , Grammar, 124–7Google Scholar.

³⁵ Below, § 3.2.5, p. 391.

³⁶ cf. ‘Verb complex’, 557–8 and 571–3, and also the third example in § 3.2.1 below.

³⁷ Or, of course, jiang, goan, and geei; cf. ‘Verb complex’, § 8.1.3, p. 566.

³⁸ The others are air, show, jiaw, ranq, you; cf. ‘Verb complex’, § 8.1.4, p. 566.

³⁹ Chao, , Grammar, 86Google Scholar, shows that a subject may be preceded by the coverbs you and guei which gives some credence to the notion of bey as a marker of the subject. He also adds an example where bey is actually followed by an object marked by bae (or baa), Bey jeh jiiwen chyan bae jeh sheauren manguoh ‘(It was) by this little (that) fooled this small man’ (‘This small man was fooled by this little bit of cash’—my alternative translation).

⁴⁰ Chao, , Grammar, § 8.2.2(3), p. 751Google Scholar, does not differentiate two functions but regards bey as a preposition with the object omitted.

⁴¹ cf. below, § 3.2.4, p. 391, and § 4.3, p. 396.

⁴² cf. Chao, , Grammar, § 5.4.5, p. 312Google Scholar, ‘a transitive verb does not become intransitive by not having an object. In general an object to a transitive verb is omitted if it has occurred in a near context whether or not as object to the verb in question’; ibid., § 8.1.5.1, p. 701, ‘Aotually if it is implied in the sitnational context, no object needs even to be present in the linguistic context’.

⁴³ This, like the definition of ‘direct object’ above, is an ad hoc definition. I hope to offer a more detailed description another time. It will be seen, for instance, that no account has been taken of sentences which have no verb.

⁴⁴ i.e. bey, air, show, rang, and geei, the last two only when they may be replaced by bey since they also have other functions, cf. ‘Verb complex’, §§ 8.1.4, 8.1.3, 8.1.1, and p. 665, n. 1, for geei, and § 10.1.1 for ranq.

⁴⁵ An isolate which occurs with a determinative in the colligation [determinative determinate] [d a] and cannot be followed by a noon, e.g. nah-tian ‘that day’, cf. ‘Substantival complexes’ 351–5.

⁴⁶ cf. Chao, , Grammar, §§ 2.10.1–2.10.6, pp. 94–100Google Scholar.

⁴⁷ It may, perhaps, prove necessary to distinguish a further sub-class signifying ‘all of them’, e.g. chanq chanql, ual, yn shy, niann shu in example 90.29 in § 4.3 (p. 396), each of which might be regarded as ‘generalized mooting’.

⁴⁸ cf. Chao, , Grammar, 77Google Scholar, ‘thus yeou shie has become practically a compound meaning “some”’. An alternative solution is to regard yeou as a linkverb whose substantival (‘pivot’ in Chao's terms) is the object of the following clause. Cf. ‘Verb complex’, § 10.1.3, and Chao, , Grammar, 729Google Scholar, although Professor Chao would, I think, regard it as the subject of shyuebuhuey.

⁴⁹ Mr. Chou preferred shianq before bii rather than before hao niann.

⁵⁰ cf. also ‘yoked clauses’, above, § 4.1, p. 393.

⁵¹ Mr. Chou could not accept the final le in this sentence.

⁵² Note that these are not regarded as M structures.

⁵³ Mr. Chou prefers igeh ren to ireel in the original text.

⁵⁴ cf. Ghao, , Grammar, 69–78, 702–4Google Scholar, who takes a different view.