¹ Leonard Bloomfield, Language 396 (New York, 1933).

An earlier version of this paper was read at the 1963 annual meeting of the Linguistic Society of America.

² Yuen Ren Chao and Lien Sheng Yang, Concise dictionary of Spoken Chinese (Cambridge, Mass., 1947); Uang Yi and others, Gwoyeu tsyrdean (Shanghai, 1943, preface dated 1937). Chinese names, book titles and other references not given as linguistic citations are cited in the Gwoyeu Romatzyh (as found in the Concise dictionary). Linguistic citations are in my own transcription; the corresponding Chinese graphs, or ‘characters’ as they are sometimes called, are given only when absolutely necessary.

³ Luh Faa-yan and others, Chieh yunn (preface dated 601). This riming dictionary was lost until fragments were discovered among manuscripts from Tunhuang around the turn of this century. Its readings are well preserved in the later expanded version: Chern Perngnian, Chiou Iong, and others, Goang yunn (preface dated 1011). A modern reprint of the Goang yunn is edited by Jou Tzuu-mo, ‘Goang yunn’ jiaw-been (Peking, 1960). Sheen Jianshyh, ‘Goang yunn’ sheng-shih (Peiping, 1945) is a rearranged but complete version of the Goang yunn and adds a convenient character index.

⁴ The earliest extant displays of this phonology date from fragments of tables by Shoouuen, a Buddhist bhiksu who lived in the tenth century. These fragments were discovered among the manuscripts from Tunhuang and have been reproduced in Liou Fuh, Duenhwang dwo-suoo, Academia Sinica, Institute of History and Philology (Peking, 1925–35). More complete is the Yunn jinq, of unknown authorship; extant versions carry a preface (by Jang Lin-jy) dated 1161. There seems to be no clear evidence for dating the Yunn jinq earlier than the middle of the tenth century.

⁵ See, for example, Joseph Edkins, China's place in philology: An attempt to show that the languages of Europe and Asia have a common origin (London, 1871). Though Edkins's method is shaky, assuming cognate relationship after assembling look-alikes in various languages of the world, he knows something, anyway, about European linguistic techniques (or at least the results), and his work is important as an early attempt to assign phonetic values to the phonological structure implied by rime-books of the Chieh yunn tradition and already discovered by Chinese scholars. Though unsystematic, he is the first to arrive at some phonetic values through extensive comparison with various modern Chinese dialects. Earlier attempts by Westerners to deal with traditional Chinese ‘spelling’ systems, such as Joshua Marshman, Elements of Chinese grammar ... 90–118 (Serampore, 1814), were intuitive—largely guesses on the analogy of the Sanskrit alphabet. Z. Volpicelli, Chinese phonology ... (Shanghai, 1896), reviews previous work on reconstruction, and makes a concerted attempt to use modern dialect material. Again, though his methods are not up to modern standards of rigor, he arrives at plausible reconstructions, at least for initials.

⁶ The most important of Karlgren's works on MC phonology are four. (1) Études sur la phonologie chinoise (Leiden, 1915–26), a detailed comparison of modern Chinese and Sino-Xenic phonetics, and an evaluation of earlier Chinese studies of Chieh yunn phonology, embodying careful decisions as to the supposed phonetic realization of MC phonology on the basis of modern comparisons. Included is a ‘dictionnaire’ of common words representing virtually every possible MC syllable, giving the main modern reflexes (Sino-Korean ; Sino-Japanese: Kan-on and Go-on; Sino-Vietnamese; Cantonese; Hakka; Swatow, Amoy, and Foochow; Wenchow and Shanghai; Peking and fifteen other Mandarin dialects). Jaw Yuanrenn (Y. R. Chao), Lii Fangguey (F. K. Li), and Luo Charngpeir, Jonggwo inyunnshyue yanjiow (Shanghai, 1937), is a critical translation of Karlgren's Études, embodying in both the text and the ‘dictionnaire’ many revisions and corrections, and converting Karlgren's romanizations (based on Lundell's phonetic alphabet) to a more familiar notation based on the International Phonetic Alphabet.

(2) Analytic dictionary, Chinese and Sino-Japanese (Paris, 1923) is a predecessor of the following work. Chinese graphs are grouped according to their phonetic elements, then arranged according to the Pk. reading of the bare phonetic. Pk., Cantonese, MC, and Sino-Japanese readings are given for each entry, and brief English definitions.

(3) Grammata serica = BMFEA 12.1–471 (1940) and Grammata serica recensa = BMFEA 29.1–332 (1957). The second of these book-length articles revises the first, but leaves out the informative introduction. Again, graphs are grouped as in the Analytic dictionary, but are arranged according to the spelling of the Old Chinese phonology that Karlgren reconstructed on the basis of shapes of graphs and riming conventions in the Booh of Odes (about 1000–500 b.c.), and which he calls Archaic Chinese. Old Chinese, MC, and Pk. readings are given (only Grammata serica recensa gives tonal indications), and the location of the earliest classical sources are supplied with corresponding English meanings.

(4) Compendium of phonetics in Ancient [MC] and Archaic Chinese = BMFEA 26.211–367 (1954) explains the general process and the special problems of reconstruction and gives patterns of sound change from Old Chinese through to Pk.

Important and accessible revisions of Karlgren's data are to be found in three works. (1) Y. R. Chao, ‘Distinctions within Ancient Chinese’, HJAS 5.217–23 (1940), combines some of Karlgren's MC initials and some of his medials and represents probably the first attempt to apply phonemic principles to MC. (2) Samuel E. Martin, The phonemes of Ancient Chinese, Supplement to JAOS 16 (1953), extends the phonemicization of MC along lines suggested by Chao. (3) E. G. Pulleyblank, ‘The consonantal system of Old Chinese’, Asia Major NS 9.58–144, 206–65 (1962–3), includes as part of the introductory material (58–85) a valuable review of the research on MC since Karlgren. See Pulleyblank for further bibliography on MC.

⁷ Hugh M. Stimson, ‘Ancient Chinese -p, -t, -k endings in the Peking dialect’, Lg. 38.376–84 (1962).

⁸ Jang Tzyhlieh (Liau Wen-ing, ed.), Jenqtzyh tong, yiinjyi shanq 57b (1685; Jang's preface dated 1670), under the appropriate graph gives the reading pil and the definition ‘female genital orifice’.

⁹ Maan Sychian et al., English-Hakka dictionary 581 (Goong-cbii Chubaansheh, Taiwan, 1958), under ‘vagina’, gives 'tjē-pâi (vulg.)'. Yang Shyr-ferng, Tair-uan Tauryuan Kehjia fangyan 393, Academia Sinica, Institute of History and Philology Monographs A.22 (Taipei, 1957) gives 't∫i pai ... neutzyy in [female genitalia]'.

¹⁰ W. Campbell, A dictionary of the Amoy vernacular ... 536 (Tainan, 1913; 7th edition, Jiawhuey Gongbawsheh, Tainan, 1958) gives ‘Pi ... [female genitalia]; chi-bai’; 50 gives ‘chi ...: chi-bai [that is, female genitalia]‘.

¹¹ Bernard F. Meyer and Theodore F. Wempe, The student's Cantonese-English dictionary 483–4 (Hong Kong, 1935; 3rd edition, New York, 1947) gives ‘pei ... vagina’.

¹² Y. R. Chao, private correspondence. I have not found this word attested in any printed source.

¹³ Haigh Roop, private communication. M. J. Cuaz, Essai de dictionnaire français-siamois 976 (Bangkok, 1903), gives 'Hí' after the entry 'vagin, lèvres de'.

¹⁴ Wang Lih, Hann-yeu shyliuh shyue 44, 332–4 (Shanghai, 1958).

¹⁵ Stimson 384.