Ha Guangtian (SOAS, University of London)
According to one theory, the sharp division between Classical Arabic and varieties of colloquial Arabic, rather than being exemplary of what has been called “diglossia” in socio-linguistics, was in fact the consequence of Islam’s historical conquest and expansion. From the nomads in the Arabian Peninsula to cities in the Mesopotamia, the spread of Islam also carried the Arabic language to urban groups of heterogeneous linguistic repertoire. The making of “Classical Arabic,” its “elevation” above the “colloquial,” was as much an effort at “preserving” and “purifying” Arabic of its urban “corruption” – hence a highly political act – as it was an endeavour to conserve the linguistic rigidity and sanctity of the Holy Qur’an. As the elite Arab scholars tried to push back against Arabic’s inevitable phonetic, lexicographical, and grammatical shifts, even the recitation of the Holy Qur’an, where the most elevated register of “Classical Arabic” is supposed to reign supreme, had been “corrupted” by readings whose sounds threw into sharp relief the extent of linguistic transformation that governs the recitation of even the holiest of all holy texts among Muslims.
The issue is considerably more complicated than paying attention to the histories of different Arabic “dialects,” for dialects are rarely allowed – at least according to what we know and particularly as we know it since the early twentieth century – into marked Islamic rituals such as the five daily prayers. However, among the Jahriyya Sufis and many other Hui Muslims, Sufis and non-Sufis alike, a blatant contradiction can be observed: on the one hand, the kind of Arabic they use for daily and other rituals differs significantly from any other Arabic “dialect.” The sounds of this Hui Arabic are barely recognizable to a speaker of Arabic, whatever the dialect and wherever she comes from in the vast Arabic-speaking region. On the other hand, almost like a calculated transgression, this “deformed” Arabic has been used only and exclusively in a religious context. It has never become a spoken language, though the Chinese transliteration and transcription of certain key words (such as īmān, transcribed as yimani, 伊瑪尼) have acquired wide circulation in the everyday language of the Hui. The Hui Arabic, in other words, is not even a “dialect,” both sociologically and linguistically, and yet there it is, used, cherished, and revered, crowned as the only legitimate language when a Hui offers her prayer to the all-mighty god.
Tracing the subtle and often unnoticeable evolution of this Hui Arabic is a daunting task, not the least because it is scarcely acknowledged, let alone openly defended, among the Hui themselves. One ventures into a deeply ambivalent field in bringing to light this issue and attempting to justify, by means of historico-linguistic studies, the existence of Hui Arabic not as a “deformation” but as a consequence of concrete and often unanticipated interpenetration of languages and scripts, a consequence, in other words, of the emergence, expansion, and collapse of empires. The Hui Arabic points to what has been wiped out, buried, and left to rot under the thin layer of soil on which is built the modern Chinese nation-state and the modern partition of Central Asia and the Middle East along national and ethno-linguistic lines.
The early twentieth century constituted a particularly critical period for the sound of Hui Islam and Jahriyya Sufism. A key transformation was silently but palpably working its way through: while transcription used to be the dominant mode of bridging Arabic, Persian, Turkic and Mongol languages on the one hand, and Chinese on the other in the imperial period, hence lending priority to listening and the reproduction of sound by means of script, from the early twentieth century on, particularly since the 1920s, transliteration began to supersede transcription as the authoritative mode of inter-lingual practice. The concern was no longer how to reproduce sound as much as it was how to establish a uniform standard and construct a thesaurus of translation that facilitates an ideally smooth migration of words, meanings, and potentially the divine blessing, from Arabic to Chinese, from the sacred Qu’ran to its Chinese interpretation, to the exclusion of all other languages. History was silenced, myriad transcriptions disciplined and reduced, sounds transformed into texts, and the relation between speech and script where the ear took precedence gave way to an inter-textual relationship where the scanning eye of the spectator and the meditating mind of the reader prevail.
Thus, we witnessed two intertwined changes in the early twentieth century: 1) transcribed sounds in historical records were converted and “standardized” into uniform textual segments that established an exclusive one-to-one correlation with the “original”; 2) this conversion was accomplished at the expense of neglecting the mediating role of a variety of Eurasian languages, dialects, and scripts, and reinforced the imaginary re-connection of Hui Muslims to the “centre” of Islam in the Middle East and North Africa, hence eclipsing the vast area of Eurasia that lies in between. When writing becomes the rule, history sinks into oblivion and amnesia asserts domination.
However, this ground-shifting change was not without its discontents. “You cannot alter the text at will and still claim it is from the original,” Chen Yuan, a prominent historian of the period, objected vehemently in one of the earliest and foundational articles written on the history of Islam in China. “Each generation [i.e. historical period] has its own transcription. You can easily tell which period they came from just by looking at such transcriptions as 摩訶末 and 謨罕募德.” Chen Yuan’s objection stemmed from a basic fact he was not able to fully grasp at the time: the increasing reduction of historically different transcriptions of Muhammad to 穆罕默德 at the expense of all others, hence eliminating the mark of time in the course of frequent quotation and re-publication, was an indication of a complete change in the mode of relationship the Hui had with sound and text. While it was the sound that mediated the Hui’s relationship to text in the imperial times, the twentieth century was characterized by the effectuation – never fully complete and hegemonic – of a reversal: that the Hui’s relation to the sound of recitation is increasingly mediated by the canonized and standardized text.
Chen Yuan’s List of Transcriptions of Muhammad
暮門 [mù mén] Tang Dynasty (618-907 CE)
摩訶末 [mó hē mò] Tang Dynasty
麻霞勿 [má xià wù] Song Dynasty (960-1279 CE)
馬合麻 [mă hé má] Yuan Dynasty (1271-1368 CE)
謨罕蓦德 [mò hăn mò dé] Late Yuan to Early Ming Period (1368-1644 CE)
穆罕默德 [mù hăn mò dé] Contemporary Times
This is not the only list that Chen Yuan devised. It is not even the one he was most famous for. He drew up a chain of transcriptions that showed the shifting sounds of the word “Hui” throughout history, its multiple phonemecization and change of tone (relying on the character chosen) being a result, perhaps not surprisingly, of its mediated migration between different languages and scripts (figure 1, audio file 1). The point became even more compelling when we turn to his triple chain that revealed the hair-splitting sonic history of the word “Muslim” in Chinese historical records (figure 2, audio file 2).
Figure 1. The Chain of “Hui” (a partial rendition using pinyin): huí hé – huí hú – wài wŭ – wěi wù – wèi wú ér – wèi wù ér – húi hui. From Chen Yuan, “A Brief History of Islam in China”, 1928
Figure 2. Three Chains of “Muslim” (a partial rendition using pinyin): 1. mó sī lăn – mò sù lŭ mán – móu sù lŭ mán – mù sù ér mán – mù sù mán – mù shì lín; 2. dà shí mă – dá shī mán – dá shí mán; 3. pŭ sù wán – pú sù wò – pù sù măn. From Chen Yuan, “A Brief History of Islam in China”, 1928
What emerged and seems to have prevailed since the early twentieth century is a new linguistic ideology. It no longer allows the existence of a “middle ground” where sounds and phonemes traverse linguistic boundaries and cohere, however temporarily, into new units and create new syntactical and lexical constructions. Languages need to be fixed, and inter-lingual – rather than translingual – activities necessarily have to presuppose a priori the existence of distinct and discrete languages in the first place. A sound must belong whether to Arabic, Persian (only occasionally involved), or Chinese – the three “major” languages that are now considered to be the cornerstone in the formation of Hui Islam, to the exclusion all other linguistic, cultural, and religious influences. The displacement of transcription by transliteration, and the “standardization” of the “Chinese-Islamic terms,” has over the entire course of the twentieth century consistently re-structured the way in which contemporary Hui Muslims deal with the history that made them and the sound that lingers above the text.
Classical Arabic and Jahriyya Pronunciation. Adapted from Table 2.1 in Janet Watson, The Phonology and Morphology of Arabic, p. 13. Jahriyya pronunciation in red. IPA used only when the sound of Jahriyya pronunciation differs considerably and consistently from the Classical Arabic pronunciation, thus justifying its separate categorization. The transliteration of Arabic alphabets follows the conventional rules of the International Journal of Middle East Studies. Red solid lines link sounds that bear the same phonetic value (at least to my ears), while those that tend to pass into each other are linked by red dotted lines.
Two ways of pronouncing the Arabic alphabet
The first is the Jahriyya pronunciation, and the second is the “standard” pronunciation, as recorded by a Jahriyya.