CHINESE AND ENGLISH PEAR STORIES

The Chinese stories are transcribed into characters by native speakers, following as closely as possible the original sounds on the tapes. Phonetic transcriptions are used where characters for the words do not exist. The Cantonese transcripts include many Cantonese characters. Fonts for these can be downloaded. Characters are used for the transcripts made in Taiwan and Hong Kong; the mainland transcripts use simplified characters.

The Mandarin tapes also include full romanization in standard Hanyu Pinyin. The Cantonese tapes are also romanized using the Yale system, which has the best support in reference books. The Hakka transcripts are romanized in Dr. Lau Chun-fat's 1997 transcription. The other transcripts remain unromanized, because so many competing romanizations exist without a standard. Romanization tables for Cantonese and Hakka are included.

Mandarin stories were collected in Taipei in 1976 at National Taiwan University by Mary Erbaugh. Most participants were native speakers of Mandarin, children of immigrants from Shandong and Jiangsu. (Fuller language background details are attached to each transcript.) Most held bachelor or master's degrees in Chinese literature or history, and were working as teachers of Mandarin. The non-teachers were current university students. All had studied English for at least 10 years; many had very fluent English.

Ms. Chiang Tz'u (O), a fellow teacher and friend, interviewed the 19 participants. Mr. Francisco Huang Yun-wen () did the original hand-written transcriptions, which were later typed, edited and numbered by Ms. Carine Yiu Yuk Man (ɱ). Ms. Selena Choi Wai Yuk () added the pinyin romanization. The Mandarin Pears have stimulated number of studies of comparative narrative, clause structure, word order and anaphora, noun classifiers, time and aspect, (Chen 1986 a, b, 1987, Christensen 1994, 2000, Chui 1994, Cumming 1984, Erbaugh 1986, 1987, 1990, 2002, Polio 1994, Smith and Erbaugh 2001; Yang in progress).

Cantonese stories were collected in Hong Kong, 1996, by Mary Erbaugh, with generous support from the City University of Hong Kong. The Research Grants Council of the Hong Kong Special Administrative Region, China (CERG grant 9040294-590) funded the expansion to other dialects. Thirty speakers were recorded, the minimum number for many statistical tests. Business and science undergraduates at City University were recruited, to avoid pulling the sample toward a heavily literary or Mandarinized style. All spoke fluent English, but only a few spoke Mandarin.

Interviewers and transcribers were Ms. Patricia Man Yuk-hing (ɭ), and Ms. Germaime Li (Ŵf). Original recordings are on audio cassette. Additional editing and analysis was done by Mr. Lawrence Cheung (iܨ}) and Carine Yiu (ɱ).

An additional set of Pears Stories from 7-year-old Cantonese children in Hong Kong, to be available on the internet via the Childes language data exchange, at childes@mail.talkbank.org.

Hakka speakers were recorded in Hong Kong, primarily in villages in the Sai Kong Peninsula. They are speakers of Bao An, rather than Meixian, Hakka. Many also speak fluent Cantonese. The rapid rise of Cantonese in Hong Kong schools and public makes it almost impossible to find a group of fluent young speakers. Hong Kong Census records show the percentage of residents using Hakka a 'usual language' fell from 15% in 1911 to 2% in 1991. Most fluent speakers are over 60 years old, and relatively less educated.

Fortunately, Mr. Jonathan Wong Wai Kuen (v) a recent graduate of Chinese University, grew up in the village and is a fluent speaker, as is Dr. Lau Chun-fat (Bo) of the Hong Kong Polytechnic University. Mr. Wong pulled his best connections among village relatives and fellow church members to record these interviews on digital video and mini-disk from largely middle aged and elderly speakers in their homes. Many participants watched the film, then said 'there's not much to say'. Only the more extensive descriptions are included. Dr. Lau transcribed the interviews in characters, and in romanization, as in his 1997 dictionary.

Shanghai Wu. Dr. Tao Huan () of Fudan University in Shanghai organized the recording and transcription for Shanghai Wu dialect, Xiamen Min dialect, and Nanchang Xiang dialect during the summer of 2000. All the mainland students also speak fluent Mandarin, and have studied English for at least 6 years. Dr. Tao personally supervised the recording and analysis of the Shanghai native speakers, all students at Fudan University. Original recordings are on audio cassette.

Xiamen Min. Dr. Qian Zunxiang () of Xiamen University collected the Southern Min sample at Xiamen University in 2000. Original recordings are on mini-audio cassette.

Changsha Xiang. Professor Li Lan () of the Chinese Academy of Social Sciences collected and analysed the Xiang sample at Hunan Normal University. Original recordings are on audio cassette.

Nanchang Gan. Dr. Yan Sen (C) collected the Gan sample at Jiangxi Normal University. Original recordings are on audio cassette.

English stories were recorded from undergraduates at the University of California 1975. Few of these students would have been fluent in a foreign language. Transcripts come from appendix to Professor Chafe's 1980 volume, pp. 310-19. With his permission, they have been numbered and re-typed for ease in processing.

Production of the Website. Mr. Denis Yip (h) and Ms. Kennis Ng (dg) edited the tapes and transcripts at the Department of Chinese, Translation and Linguistics at the City University of Hong Kong. Mr. Yip also digitalized the audiotapes and designed the website.

Some warnings. The scientific demand for a uniform and accessible set of native speakers means that almost every woman who tells the Pear Stories is a highly educated multilingual. They should not be used as a standard for normal but ordinary speech. For ethnographic purposes, we would have liked to record monolingual speakers. As China grows richer and better educated, such 'pure' dialect speakers are increasingly elderly, confined to remote rural districts, and dauntingly difficult to find and record.

Although describing the film produces pleasingly enthusiastic and relaxed speech, the speakers are still in an artificial setting. The interviewer, also, makes many fewer comments than an ordinary conversation partner. And, of course, the limited content of the stories is obvious.

Although we tried our best for a uniform procedure for interview, recording, transcription and editing, variations do creep in for a project of this scope and geographic range. In particular, outside of Hong Kong, we were not able to videotape. And some of the sound recordings are not as good as we hoped. We have filtered out as much background noise as possible, but recordings which are hard to listen to are marked as (-) in the table of contents. All transcriptions, however, are accurate, edited, and checked by other native speakers.