Li Tanfei draws on her own experience and writes about the origin and history of the phenomonon of r-coloring in Madarin Chinese.
Scroll down to find versions in Simplified and Traditional Chinese.
Just recently, my roommates were struggling with their recording homework for Mandarin Chinese, so I decided to help them with the tricky pronunciations. I looked at their textbook and noticed the notation for rhotacisation, which means a sound change that converts a consonant to a rhotic consonant by adding a r-like sound, also known as r-coloring. This is noted in pin-yin (the romanised system of phonetic notation for Chinese) by adding an ‘r’ at the end of a character’s syllable, like ‘yīhuìr’ (一会儿), which means ‘a while’, rather than the standardised pronunciation ‘yī’ ‘huì’ （一会）that one will find in dictionaries for each character. In the logographic writing, the character ‘儿’, read as ‘er’ singularly, is added to indicate a r-coloring of a character without changing its meaning. I took a trip down memory lane and tried to recall what I did in my Chinese classes back in my hometown in Baoding, Hebei province, in China, and I could not remember being taught to read and write the rhotic ‘儿’. Apparently, this habit of rhotacisation was instilled in me organically as I was from the dialectic region of Mandarin where this pronunciation is common. However, when taken out of the context in which this habit of speaking exists, this notation becomes very confusing because one does not know how to pronounce it, to which characters it can or cannot be added, and whether it changes the meaning of certain characters.
It is even more intriguing to think about how we could trace the origin of such habits of pronunciation because of the lack of audio recordings until only less than 200 years ago. My online research proved that the answer to this question is more complicated than I thought. The word Mandarin in English denotes both the northern dialect that is now taken as the standard pronunciation of the official Chinese language, and any official of the top 9 ranks in the Chinese imperial civil service system. This implies that the Mandarin dialect was adapted from the speaking habits of the powerful back in the days. The rhotacisation exists in both the northern dialects of Beijing, Hebei, the North-eastern Provinces (which are fairly close to Mandarin today), and the southern dialects in Nanjing, Sichuan, and Chongqing. As the Beijing dialect is taken as the basis of Mandarin pronunciation, its conventions of r-coloring is also standardised in textbooks. Throughout history, the standard pronunciation in each dynasty and period is often based on how the habitants and officials of the dynastic capitals spoke, but dialectic habits of speaking were far from unified and standardised even today. So, how did the Beijing dialect come about and why is it formalised as the official pronunciation?
One of the earliest records of the existence of a common pronunciation system for the Han Chinese language can be found in Confucius Analects, a collection of Confucius’ wisest words (551 BCE – 479 BCE) by his disciples. One of the entries testifies that the recitation of literary classics and reading of verses in rituals were done in ‘ya yan’ (雅言), which literally translates into ‘elegant language’. Here, the character ‘雅’ (ya) is partially homonymous to ‘夏’ (xia), which suggests that ‘ya yan’ can be traced back to the dialect of Luohe (Henan province today), the capital tongue of Xia dynasty (夏朝) around 2070 BCE. Several dynasties that came after Xia also had their capitals located close to the region around Luohe, including Chang’an (Xi’an today) and Luoyang, thus making ‘ya yan’ the progenitor of Mandarin. After years of unrest and disunity between the Han ethnic group and the other ethnicities starting from the Rebellion of the Eight Princes at the end of West Jin dynasty, the high-class of the Jin migrated to the south, and the ‘Luohe ya yin’ of central China was molded and shaped by the dialect used in the new capital, Jinling (today’s Nanjing, where the name of the city literally means the southern capital, a reference to its history). The original southern dialect, wu yin, was spread to Japan during the Northern and Southern dynasty (420 CE to 589 CE) when bilateral trade flourished. Many traces of wu yin can still be found in Japanese pronunciation today. In the Sui dynasty that followed, Fanyan Lu and several other literatis compiled a 5-part index of Chinese characters based on phonetic rhymes and alliterations, titled ‘Qie Yun’ (‘切韵’, where ‘yun – 韵’ means rhyme, or musical sounds). It is a hybrid of both southern and northern pronunciations. The ‘Tang Yun’ of Tang dynasty (618 CE – 907 CE) and the ‘Guang Yun’ of Song dynasty (960 CE – 1279 CE) further categorised the Chinese characters based on the pronunciations in ‘Qie Yun’, which served as rhyme books for generations of poets and essayists from then on.
Meanwhile, since the Tang dynasty, the ethnic minorities residing in the north, like Qidan and Nǚzhen, developed a dialect called ‘you zhou hua’ （幽州话，幽州 – you zhou – is in today’s Beijing). When the Yuan dynasty chose Dadu (today’s Beijing) as the capital, the dialect of Dadu, which was influenced by ‘you zhou hua’, was taken as the new version of Mandarin. One of the records of the prevailing pronunciations include ‘zhong yuan yin yun’, which compiled the rhymes often used in the lyrical poems particular to Yuan. Many characters were indexed to a pronunciation that is close to its modern rendition in the Beijing dialect. In comparison, tonal accent special to ‘ya yin’ was lost, and thus the rhymes in the poems from previous dynasties are sometimes off when they are read in Yuan and modern mandarin. When the monarchy was reclaimed by the Han Chinese from the Mongols of Yuan, Ming dynasty was founded, the capital was changed to Nanjing then back to Beijing, which re-introduced the southern dialects to the northern Mandarin. This gave rise to many polyphonic characters, like ‘色’ (which is read as ‘se’ when referring to colours and ‘shai’ when referring to dices) and ‘拨’ (which means to peel when read as ‘bo’ and to pull when read as ‘ba’). When Ming was replaced by the Qing dynasty by the ethnic Manchus, Mandarin started to incorporate ethnic dialects of the Manchu, the Mongol, and the Khui. The famous Chinese classical novel ‘Hong lou meng’ included words whose roots were found in ethnic dialects or languages, like ‘ba bu de’ comes from the Manchurian word ‘bɑhɑci’, which means to long for, and ‘hu tong’ from the Mongolese word ᠬᠤᠳᠳᠤᠭ（hvdvm), which refers to the quaint residential alleys in Beijing [1].
Under the rein of Emperor Yongzheng in Qing dynasty, the government set up schools to standardise pronunciation in Fujian and Guangdong province, which provided training to potential civil servants who entered the Kejǚ examination for an official post. Although the results were far from satisfactory, such schools established the Manchurised Mandarin as a national language, which replaced the influence of southern accents and dialects. Later, under Emperor Guangxu, the pedagogical director of the Imperial University of Beijing, Rulun Wu, visited Japan and was inspired by the standardisation of Japanese based on the Tokyo dialect, thus started the process of institutionalising and popularising a national standardised language in Qing China as well. Right after the Republic of China was established, literary purists advocated for a classical Mandarin that included ancient and obsolete pronunciations, making it too hard to learn. Later on, the May 4th anti-imperialist movement kickstarted the wave of vernacularism, which sought to simplify written Chinese to make it comprehensible to the average citizens just like spoken Chinese is in order to accelerate the spread of knowledge and facilitate social reform. After the People’s Republic of China was founded, the state department defined Mandarin Chinese as a common language for the Han ethnic group that is based on the pronunciation and habit of expressions of the Beijing dialect on February 6th, 1956.
It can thus be understood that the inclusion of r-colouring in written expressions stems from the vernacularisation of written Chinese under the influence of the May 4th movement. Some of the rhotacisations are collocated, like ‘皮影儿’(piying-r, which means shadow puppet) and never ‘皮影’ (‘piying’), while others expresses intimacy and affection, like ‘小孩儿’ (xiaohai-r, or ‘kids’) and ‘脸蛋儿’ (liandan-r, or cheeks). In other cases, it denotes something of a small quantity or size, like ‘米粒儿’ (mili-r, which means a rice grain) and ‘门缝儿’ (menfeng-r, which means the door crack). Sometimes adding the r-colouring is added to a character’s phononym, like ‘馅儿’ (xian-r) in ‘馅儿饼’ (xian-r bing, or stuffed pie) refers to the filling, but ‘馅’ (‘xian’) in ‘陷阱’ (xian jing, which means a trap) cannot be rhotarised even though it has the same pronunciation as the former ‘xian’ so as not to be misconstrued. The r-colouring also exists in southern dialects, but the characters that we add the ‘r’ sounds to are sometimes different – it all depends on the local habit of speaking. In short, whether to add the ‘r’ is based on the meaning of the character, the tone of the expression, and one’s dialect. When it is added, the vowel or compound vowel sound is slightly changed to include an inseparable ‘er’ sound, which is almost like how the Americans read the word ‘cheer’ compared to the British way. The native speakers might feel that the rhotacisation cuts one some slacks in enunciation, but this is hardly appreciated by the poor little darlings, or the ‘xiao ke lian-r’ (小可怜儿), who are learning Mandarin Chinese without the language environment like my roommates. If you are fortunate enough to have this difficult language as your mother tongue, please remember to ‘bang bang mang-r’ (give a hand).
[1] Editor’s note: This is one of the interpretations of the origin of “Hu Tong”. The other one states that “Hu Tong” comes from the historical evolution of the Chinese language itself.
简体中文版本
最近与我合租的巴政室友都很头疼他们的中文录音作业，所以有一天晚上我让他们把课本拿来和我一起读，来对付各种刁钻的读音。我在他们的课本上看到许多带有儿化音的词，而且这个儿化音还被标注在了拼音里，比如“一会儿”（‘yīhuìr’)而非“一会”（‘yīhuì’）- 前者更接近口语发音习惯，而后者是这个词最书面的模式，拼音也只注明了这两个字的标准发音而非包括儿化音“r”。我不记得我在河北保定老家的小学课本上见过‘yīhuìr’这种写法，也或许是我因为身处语言环境当中，没有意识到这很奇怪。然而这给我的母语非中文的同学产生了很大的困扰，因为他们很难掌握发音技巧，更难理解哪些词“应该”加儿化音，哪些词不用 – 比如，我自己都没办法解释清楚为什么“饭馆儿“相较”饭馆“对我这个北方人来说更舒适，但“下馆儿子”引起极度不适。
这让我开始查找儿化音的来由和使用规律，结果发现其中的道道儿更加拗口。所谓儿化音，是现代普通话和一些官话方言中给一些特定词末的韵母音加卷舌音的现象。儿化的运用常见于北京，河北，东北三省，陕西，陕西等北方地区，但在南京话，四川话和重庆话中同样存在。其中，北京话作为普通话的标准音，将儿化音带进了汉语课本里。然而最初的汉语读音与如今的京腔相差甚远。每个朝代都城居民和官员的读音习惯常常被定为标准音，所以直到元朝1271年忽必烈改京师中都为大都（今北京市）前，所谓的标准读音经历了漫长的发展过程。
首先，《论语》中记载道，“子所雅言，诗、书、执礼，皆雅言也。” 也就是说，在诵读《诗经》、《尚书》和主持仪式典礼的时候，古人通常使用“雅言”。这是有史载的最古老的民族共同语之一，其中“雅”与“夏”相通，所以“雅言”可追溯到夏朝和商朝时期（约2070 BCE – 1046 BCE）的都城洛河（黄河、洛水一带，今河南）古语。之后各个王朝的首都基本集中于黄河洛水的长安、洛阳一带，尤其是东周、东汉、西晋三朝都定都于洛阳，使得“雅言”逐渐成形为最早的“官话”。在西晋八王之乱、南北朝五胡乱华后，晋室官宦显贵衣冠南迁，中原雅音随之南移，形成“金陵雅音”，也就是洛阳读书音融合金陵（亦建康，今南京）当地吴音的产物。同时，吴音在南北朝时代传入日本，是日本汉字读音中占比较高的读音之一。隋朝时的《切韵》（公元601）融合金陵雅音和洛阳-邺下（邺城，今河北临漳邺镇）古音，以此统一书面声韵，为后世沿袭。唐代的《唐韵》和宋代的《广韵》整合了韵声韵律，根据读音将汉字归类，为诗词歌赋的撰写作参考。
同时，自唐朝，北方契丹、女真等少数民族活跃的地带发展出了“幽州话”。元朝定都大都（北京）以后，“大都语”由幽州话发展而来，成为北京话的初身。其中，《中原音韵》根据元朝盛行的杂剧散曲的用韵编写，基本归纳了大都语的声韵调系统，同今天的北京话相当接近，但失去了之前雅音中的入声，以至于许多唐诗宋词在新的读音中变得不再押韵。到了汉人掌权的明代，明成祖朱棣将首都从明太祖定的南京迁到北京，同时也将南方官话体系再次带到了北京话中。现代汉语中的多音字的存在 – 如“色se/shai”“拨bo/bao” – 多是因为同一个字在南方和北方官话中的不同读音都被保留了下来。清朝建立后，北京话开始吸收满族、蒙古族和回族的语言要素。清代著成的《红楼梦》中也有汉化的满语和蒙古语的影子，如借满语的“巴不得”、“跟前”，和由蒙古语演变来的“胡同” [1]。
清朝雍正年间，为了统一官话，朝廷在福建和广东设立了“正音书院”，并由此招收科举考生来学习推广北京话。这虽然收效甚微，但是奠定了满族式汉语作为国家共同语的地位，并逐渐取代了南音的影响。到了光绪年间，京师大学堂总教习吴汝纶到日本考察，受到日本统一使用东京话的启发，回国后推行以北京话统一全国语言。民国初期，学者推行以北京话为基础但同时包括古汉语入声和尖团音的“老国音”，却成了发音难学的“死语言”。1919年五四运动后，白话文兴起，旨在简化书面语使其与口语一样明白易懂，以便加快信息与知识的传播。同时，以北京语音为准的“新国语”也被广泛推广。新中国成立后的1956年2月6日，国务院发表指示推广普通话，并将普通话正式定义为“以北京语音为标准音，以北方话为基础方言，以典范的现代白话文著作为语法规范的现代汉民族共同语。”
回头来说儿化音。书面语中包括儿化音的现象也是书面趋同口语的体现。儿化的词有一些是固定词汇，如“皮影儿”，其他的表示喜爱、亲切，如“小孩儿”、“脸蛋儿”，或表示小巧、少量，如“米粒儿”、“门缝儿”。根据读音加儿化有时候容易产生歧义，比如“馅儿饼”中的“馅儿”表示的是剁碎了的食料；“陷阱”中的“陷”与“馅”读音相同，但前者断不能加儿化音。儿化的现象也不只局限于北方，只是南北儿化习惯不同，比如四川的“老板儿”和洛阳的“鸡娃儿”。因此，儿化音是根据字意、语气和地方口语习惯等形成的变音，在一定程度上会将“儿”音融合进被儿化的字成一个音，从而改变原字的韵母。这样的发音对与习惯儿化的人来说可能使口语更随意顺口，但对于刚刚学习普通话的朋友们实在不太友好。如果读者遇到任何一个被汉语读音困扰的小可怜儿，记得多帮帮忙儿。
[1] 编者按：来源蒙古语是其中一种假说。另一种说法为胡同来自汉语语音的音变。
繁體中文版本
最近與我合租的巴政室友都很頭疼他們的中文錄音作業，所以有一天晚上我讓他們把課本拿來和我一起讀，來對付各種刁鑽的讀音。 我在他們的課本上看到許多帶有兒化音的詞，而且這個兒化音還被標註在了拼音裡，比如”一會兒”（’yīhuìr’）而非”一會”（’yīhuì’）- 前者更接近口語發音習慣，而後者是這個詞最書面的模式，拼音也只註明瞭這兩個字的標準發音而非包括兒化音”r”。 我不記得我在河北保定老家的小學課本上見過『yīhuìr』這種寫法，也或許是我因為身處語言環境當中，沒有意識到這很奇怪。 然而這給我的母語非中文的同學產生了很大的困擾，因為他們很難掌握發音技巧，更難理解哪些詞”應該”加兒化音，哪些詞不用 – 比如，我自己都沒辦法解釋清楚為什麼”飯館兒”相較”飯館”對我這個北方人來說更舒適，但”下館兒子”引起極度不適。
這讓我開始查找兒化音的來由和使用規律，結果發現其中的道道兒更加拗口。 所謂兒化音，是現代普通話和一些官話方言中給一些特定詞末的韻母音加捲舌音的現象。 兒化的運用常見於北京，河北，東北三省，陝西，陝西等北方地區，但在南京話，四川話和重慶話中同樣存在。 其中，北京話作為普通話的標準音，將兒化音帶進了漢語課本裡。 然而最初的漢語讀音與如今的京腔相差甚遠。 每個朝代都城居民和官員的讀音習慣常常被定為標準音，所以直到元朝1271年忽必烈改京師中都為大都（今北京市）前，所謂的標準讀音經歷了漫長的發展過程。
首先，《論語》中記載道，”子所雅言，詩、書、執禮，皆雅言也。 ” 也就是說，在誦讀《詩經》、《尚書》和主持儀式典禮的時候，古人通常使用”雅言”。 這是有史載的最古老的民族共同語之一，其中”雅”與”夏”相通，所以”雅言”可追溯到夏朝和商朝時期（約2070 BCE – 1046 BCE）的都城洛河（黃河、洛水一帶，今河南）古語。 之後各個王朝的首都基本集中於黃河洛水的長安、洛陽一帶，尤其是東周、東漢、西晉三朝都定都於洛陽，使得”雅言”逐漸成形為最早的”官話”。 在西晉八王之亂、南北朝五胡亂華後，晉室官宦顯貴衣冠南遷，中原雅音隨之南移，形成”金陵雅音”，也就是洛陽讀書音融合金陵（亦建康，今南京）當地吳音的產物。 同時，吳音在南北朝時代傳入日本，是日本漢字讀音中佔比較高的讀音之一。 隋朝時的《切韻》（西元601）融合金陵雅音和洛陽-鄴下（鄴城，今河北臨漳鄴鎮）古音，以此統一書面聲韻，為後世沿襲。 唐代的《唐韻》和宋代的《廣韻》整合了韻聲韻律，根據讀音將漢字歸類，為詩詞歌賦的撰寫作參考。
同時，自唐朝，北方契丹、女真等少數民族活躍的地帶發展出了”幽州話”。 元朝定都大都（北京）以後，”大都語”由幽州話發展而來，成為北京話的初身。 其中，《中原音韻》根據元朝盛行的雜劇散曲的用韻編寫，基本歸納了大都語的聲韻調系統，同今天的北京話相當接近，但失去了之前雅音中的入聲，以至於許多唐詩宋詞在新的讀音中變得不再押韻。 到了漢人掌權的明代，明成祖朱棣將首都從明太祖定的南京遷到北京，同時也將南方官話體系再次帶到了北京話中。 現代漢語中的多音字的存在 – 如”色se/shai””撥bo/bao” – 多是因為同一個字在南方和北方官話中的不同讀音都被保留了下來。 清朝建立后，北京話開始吸收滿族、蒙古族和回族的語言要素。 清代著成的《紅樓夢》中也有漢化的滿語和蒙古語的影子，如借滿語的”巴不得”、”跟前”，和由蒙古語演變來的”胡同”。[1]
清朝雍正年間，為了統一官話，朝廷在福建和廣東設立了”正音書院”，並由此招收科舉考生來學習推廣北京話。 這雖然收效甚微，但是奠定了滿族式漢語作為國家共同語的地位，並逐漸取代了南音的影響。 到了光緒年間，京師大學堂總教習吳汝綸到日本考察，受到日本統一使用東京話的啟發，回國后推行以北京話統一全國語言。 民國初期，學者推行以北京話為基礎但同時包括古漢語入聲和尖團音的”老國音”，卻成了發音難學的”死語言”。 1919年五四運動后，白話文興起，旨在簡化書面語使其與口語一樣明白易懂，以便加快資訊與知識的傳播。 同時，以北京語音為準的「新國語」也被廣泛推廣。 新中國成立後的1956年2月6日，國務院發表指示推廣普通話，並將普通話正式定義為”以北京語音為標準音，以北方話為基礎方言，以典範的現代白話文著作為語法規範的現代漢民族共同語。 ”
回頭來說兒化音。 書面語中包括兒化音的現象也是書面趨同口語的體現。 兒化的詞有一些是固定詞彙，如”皮影兒”，其他的表示喜愛、親切，如”小孩兒”、”臉蛋兒”，或表示小巧、少量，如”米粒兒”、”門縫兒”。 根據讀音加兒化有時候容易產生歧義，比如「餡兒餅」中的「餡兒」表示的是剁碎了的食料;”陷阱”中的”陷”與”餡”讀音相同，但前者斷不能加兒化音。 兒化的現象也不只局限於北方，只是南北兒化習慣不同，比如四川的”老闆兒”和洛陽的”雞娃兒”。 因此，兒化音是根據字意、語氣和地方口語習慣等形成的變音，在一定程度上會將”兒”音融合進被兒化的字成一個音，從而改變原字的韻母。 這樣的發音對與習慣兒化的人來說可能使口語更隨意順口，但對於剛剛學習普通話的朋友們實在不太友好。 如果讀者遇到任何一個被漢語讀音困擾的小可憐兒，記得多幫幫忙兒。
[1] 編者按：來源蒙古語是其中一種假說。 另一種說法為衚衕來自漢語語音的音變。
Images:
https://zhuanlan.zhihu.com/p/330726871
googlemaps.com
Le Dragon Déchaîné 
Merci beaucoup Tanfei ! Passionnant.
LikeLike