中文姓名聲調探析

Abstract

中文姓名相關文獻以探討命名之動機、方式、語義、姓名用字之性別差異等議題居多，聲調組合的選用不僅少有著墨，罕見大規模量化研究法，並且缺乏聲學資料佐證。本研究聚焦聲調，以20,200筆姓名聲調組合的量化分析以及聲調感知實驗探討中文三字姓名。主要目的在於探究兩大問題：（一）姓氏聲調是否影響雙名聲調組合的選擇？（二）姓名聲調組合的選用是否因被命名者的性別而異？

姓名聲調組合量化統計的語料蒐集自2009年台灣地區高中職入學榜單，男女名各10,100筆，分析結果彙整如下：（一）尾字聲調選用比例依序為二聲＞一聲＞四聲＞三聲，且二聲的選用率超過50%，大幅領先其他調類，男女皆然。（二）尾字字調若按平仄分類，平聲（一、二聲）和仄聲（三、四聲）的比例為3:1，且女名尾字多平聲，男名尾字多仄聲。（三）中字聲調：選用比例依序為四聲＞二聲＞一聲＞三聲，男女皆然。（四）雙名聲調的16種聲調組合中，男女名均以4-2組合居冠，3-3組合殿後，且前五名尾字多為平聲。（五）雙名聲調偏好「仄-平」組合，最不喜「仄-仄」組合。若以性別觀之，「平-平」組合多為女名採用，「仄-仄」組合則多見於男名。（六）全名聲調組合以兩兩抑揚相間的比例居高，顯示中文命名聲調偏好抑揚起伏。（七）雙名聲調組合的選用不僅因性別而異，且不受姓氏聲調影響。

感知實驗刺激項為一虛擬的三字名 [lu ma na]，發音人以2（二聲、四聲）×4×4共計32種聲調組合唸讀。為避免受試者利用音段線索作為判斷依據，刺激項以低通濾波技術（截斷頻率為350赫茲）處理，實驗依刺激項之姓氏聲調分為兩組，各組再按被命名者之性別區分實驗情境，透過成對比較的方式，請受試者選擇適合該姓氏聲調的名字。實驗結果顯示：（一）受試者選擇適合的雙名聲調組合時，並不考慮搭配姓氏聲調，而以被命名者之性別作為依據，其判準主要表現於尾字聲調。男名尾字聲調四聲最多，三聲、二聲次之，一聲殿後，而女名尾字聲調一聲、二聲居多，三聲、四聲次之，男女名尾字聲調偏好互為鏡像。（二）男名之雙名聲調多為三聲、四聲之組合，一聲、二聲之組合則少獲青睞，但女名之實驗結果恰恰相反，顯示雙名聲調偏好的性別互補。

姓名聲調量化統計及感知實驗結果均顯示（一）姓氏聲調與雙名聲調的選用無關，（二）聲調組合的選用具性別差異，（三）三、四聲不適用於女名。

本論文的兩個貢獻是：（一）量化研究統計結果顯示，不論姓氏聲調，台灣地區雙名偏好時長「短-長」、音強「強-弱」、音高「低-高」之對比組合，藉由節奏變化凸顯喚稱。漢語雖非重音語言，命名仍可透過聲調的語音特質對比，展現節奏效果，從而呼應音節間的韻律關係。（二）感知實驗結果除了凸顯中文命名尾字聲調選用的重要性，因受試者依據性別命名，亦呼應文獻對於尾字調的探討。

Many previous studies have investigated Chinese personal names, from the perspectives of motivation, strategy, meaning, and word choice preference for different genders and so forth. Little has focused on tone combination selection in naming. This thesis departs from the literature in doing a quantitative study and a perception experiment on the latter issue. The quantitative study deals with tone combinations of 20,200 names. The perception experiment is conducted to examine results of the quantitative analysis and of the literature. Two primary research questions to be addressed are (a) for three-character names, whether tone of the family name affects tone combination in the given name, and (b) whether the gender of the name owner is a determinant in naming. It is hoped that my attempt to answer these questions will lead to a better understanding of the relationship between tone arrangement and Chinese naming.

To ensure homogeneity in the quantitative study, 10,100 names for either gender are collected from the lists of 2009 senior high and vocational school Joint Entrance Examination in Taiwan as the name corpus. The results of tone combination analysis are: (a) For the second syllable in the given name (SSGN), the occurrence rate in decreasing order is tone 2, tone 1, tone 4, and tone 3. More than half of the SSGN pool bears tone 2, without gender difference. (b) The even/oblique tone categorization reveals a proportion of three to one in the SSGN pool. (c) For the first syllable in the given name, the occurrence rate in decreasing order is tone 4, tone 2, tone 1, and tone 3 for both genders. (d) Among 16 possible tone combinations for the disyllabic given names, the 4-2 combination tops the scale, and the 3-3 combination brings up the rear. The two ends of the scale hold true in both male and female names. (e) Most full names feature a register distinction between any two adjacent syllables. (f) Tone combination of the given name is affected by gender, but not by tone of the family name.

The perception experiment is based on 2 (tones 2 and 4) ×4×4 tone combinations of a nonsense segmental makeup [lu ma na]. To reduce segmental influence, the test materials are manipulated by the low-pass filter technique. Subjects are asked to decide, by pairwise comparison, the more suitable tone sequence for the family name. Important findings of this experiment include: (a) Subjects make their judgments based on the gender of the name owner and the tone of the last syllable, whose preference scales for male and female names turn out to be respectively tone 4 > tone 3, tone 2 > and tone 1 and tone 1, tone 2 > tone 3, tone 4, in descending order. (b) Tone 3 and tone 4 are favored choices for male given names, and tone 1 and tone 2 are disfavored. Female given names exhibit the opposite picture.

The quantitative study and the perception experiment both reveal that (a) the gender of the name owner rather than tone of the family name affects tone combination selection for the given name, and (b) tone 3 and tone 4 are disfavored options for female given names.

The present study has two contributions: (a) The quantitative study shows no matter what tone sequence the first name takes, the preferred combinations are characteristic of “short-long” contrast in duration, “strong-weak” contrast in intensity, and “low-high” contrast in pitch to highlight the function of name-calling. As a non-stress language, Taiwan Mandarin also employs two main phonetic cues of stress, namely pitch and duration, to present the rhythmic effect and thus the relative metrical prominence between syllables. (b) The perception experiment further indicates the importance of the last tone in naming. The choice is basically made to echo the gender of the name owner.