Content uploaded by Qing Lai
Author content
All content in this area was uploaded by Qing Lai on Aug 07, 2019
Content may be subject to copyright.
Full Terms & Conditions of access and use can be found at
https://www.tandfonline.com/action/journalInformation?journalCode=mcsa20
Chinese Sociological Review
ISSN: 2162-0555 (Print) 2162-0563 (Online) Journal homepage: https://www.tandfonline.com/loi/mcsa20
The Making of Sino Muslim Identity: Han Kitab in
the Chinese Xidaotang
Qing Lai
To cite this article: Qing Lai (2019): The Making of Sino Muslim Identity: Han Kitab in the Chinese
Xidaotang, Chinese Sociological Review, DOI: 10.1080/21620555.2019.1636218
To link to this article: https://doi.org/10.1080/21620555.2019.1636218
Published online: 02 Aug 2019.
Submit your article to this journal
Article views: 11
View Crossmark data
The Making of Sino Muslim Identity: Han
Kitab in the Chinese Xidaotang
Qing Lai, Florida International University, USA
Abstract: The seemingly oxymoronic term Sino Muslim has a historical rele-
vance of more than 1,300 years in China. Historically, migration and inter-
marriage gave rise to a Sinophone Muslim population, upon which a body of
Islamic literature written in Chinese three hundred years ago, known as Han
Kitab, had played an instrumental role in shaping a distinct Sino Muslim iden-
tity. Despite dramatic transformations of modern Chinese society, we found
that the Han Kitab tradition continues to be alive today in China’s grassroots-
level Muslim society. This article documents, for the first time with a probabil-
ity sample, an extensive exposure to the Han Kitab literature among ordinary
followers of a major Chinese Islamic sect, Xidaotang. More importantly, such
exposure—in both private and public spheres of life—simultaneously contrib-
utes to the core Islamic religiosity as well as distinctive affinities to their fellow
Chinese people regardless of religion (in comparison to foreign Muslims).
These findings have profound implications given the increasing significance of
China as an agent of globalization.
As China rises as a major power of globalization, a sound understanding
of the Sino-Islam relations becomes increasingly important. Is there a pos-
sibility for reconciliation between the two civilizations? At the individual
level, has a hybrid identity of Sino Muslim ever existed? Conventional wis-
dom has it that both Sinic and Islamic civilizations are expansive in nature
and ultimately incompatible (Gellner 1983;Ho1998; Huntington 1997;
Address correspondence to Qing Lai, Department of Global and Sociocultural
Studies, Florida International University, Miami, FL 33199, USA.
E-mail: qlai@fiu.edu
Chinese Sociological Review,0:1–32, 2019
#2019 Taylor & Francis Group, LLC
ISSN: 2162-0555 print / 2162-0563 online
DOI: 10.1080/21620555.2019.1636218
Israeli 1977,1980,2002). However, researchers on China’s largest Muslim
group, Hui, have documented a variety of distinct Sino Muslim characters
across China’s geography and history (Atwill 2005; Gillette 2000; Gladney
1996,1998a,1998b,2003,2004;Hu2007; Jaschok and Shui 2013;Ke
2001; Liang 2006; Lipman 1997,2004;Ma2006,2011,2012,2013;Min
2011; Stewart 2017; Stroup 2017; Yang 2011).
Such literature has so far been limited to ethnographic and historical
studies. This article, for the first time, uses survey data to account for the
Sino Muslim identity in Xidaotang, one of the major Islamic sects in
China. Based on a random household sample and innovative designs, we
found a simultaneous saliency of Islamic religiosity and pro-Sino mental-
ity. We also found that this is to a great extent attributable to a body of
three hundred-year-old Islamic literature written in the Chinese language,
known as Han Kitab (kitab is the Arabic word for book), which continues
to construct a distinct Sino Muslim identity in both private and public
spheres of the grassroots- Muslim society.
The Rise of Sino Muslims and Han Kitab
Historically, it was migration and intermarriage that gave rise to the Sino
Muslim population. Unlike other Eastern Asian societies, China has been
home to Muslims for more than 1,300 years. Since the mid-seventh cen-
tury, Arab and Persian Muslims have come to trade and live in China via
the Maritime Silk Road. Muslim communities and mosques first appeared
in major seaport cities along China’s southern and eastern coast (e.g.,
Guangzhou, Quanzhou, Ningbo). Over time, business activities further led
Muslims to major urban centers upon China’s inland river network,
including Hangzhou on the Beijing-Hangzhou Grand Canal, and Nanjing,
Yangzhou, and Zhenjiang on the Yangtze River; they have also advanced
westward along the Yellow River to Kaifeng in Central China and Xi’an
in the Northwest. During the Tang (618–907) and Song dynasties
(960–1279), Muslims in China were largely foreign-looking and Arabic/
Persian-speaking. They were not only racial and religious minorities, but
also isolated as foreign traders from afar (Bai 2007; Leslie 1998;Yu1996:
1–73). In the Yuan Dynasty (1271–1368), while traders continued to come
from the sea, the Mongol conquerors brought great numbers of Central
Asian Muslims to inland China as bureaucrats, soldiers, traders, crafts-
men, and slave laborers. These Muslims from the west are referred to as
huihui in official documents (Yu 1996: 80), who constituted the main body
of the semu, the secondary class in Yuan’s racialized caste system. Though
there was evidence of increasing Sinicization (Chen and Chi’en 1966), the
huihui had largely remained a superior foreign class to the Han Chinese
2 CHINESE SOCIOLOGICAL REVIEW
under the Mongol reign—a truly fused identity of Sino Muslim was yet to
be constructed (Bai 2007; Leslie 1998).
What eventually made Sino Muslim possible were marriages. Until the
Ming Dynasty (1368–1644), intermarriages between the Muslims and Han
Chinese had been limited. However, in 1372, only four years after the Han
regained power, the state immediately outlawed the marriages within and
between the non-Han peoples, hoping to dissolve the remaining Mongol
and semu powers into the vast Han population (Bai 2007: 127; Yu 1996:
115). As a result of over two centuries of assimilative intermarriages, the
(more or less) Chinese-looking and Chinese-speaking huihui people finally
emerged across China as a sizeable Sinicized Muslim population toward
the end of the Ming regime (Yu 1996: 136–139). In modern Chinese schol-
arship, this population is regarded as the beginning of today’s Hui ethnic
group (or nationality).
Paradoxically, the rise of the Sinophone Muslim population was accom-
panied by a crisis of the Sino Muslim identity. By late Ming, the demo-
graphic Sinicization had made Chinese the lingua franca among the
huihui, whereas Arabic and Persian languages were only used—often in
fragmentary and mixed manners—for religious purposes in the mosques.
To make things worse, the Ming rulers and early Manchu (1644–1912) rul-
ers had repeatedly shut down China’s coastal borders to the Maritime Silk
Roads as well as the northwestern borders to Central Asia, causing a
shortage of religious staff who could read Koran and the Islamic cannons
written in Arabic and Persian (Yu 1996: 137–141). In response to such pre-
dicament, a grassroots-sponsored self-preservation movement of jingtang
(i.e., scripture hall, or mosque) training emerged in the late Ming. The
movement was pioneered by Hu Dengzhou (1522–1597), a Confucian
scholar and returned pilgrim, in the northwest and quickly spread to the
Muslim communities across the country. The jingtang training largely
involved mosque-based educational programs where a set of somewhat
standardized Islamic canons (originally in Arabic and Persian) were taught
largely in Chinese. Unlike typical educational activities in traditional
China, jingtang training aimed not to prepare scholar-bureaucrats for civil
services but religious staff for Mosque functionalities (Bai 2007; Leslie
1998;Ma2000:81–83).
The lack of religious literacy was not the only threat to China’s Islam.
Muslims in traditional China also had to face the coercive Han state (and
later the Manchu state run mainly by Han) and its ethnocentric scholar-
bureaucrats, especially in the east and south where Muslims were absolute
demographic and religious minorities (Ma 2000: 83; Yu 1996: 139). To
gain tolerance of the Han officialdom and society, Muslim scholars had to
master the keys to the Han elite mentality—high literacy and Confucian
knowledge. Consequently, a body of literature known as Han Kitab were
produced from late Ming to the end of imperial China. Aiming to
SUMMER 2019 3
interpret the Islamic cannons in Confucian terms, dozens of interpretive
translations and original writings were authored on Islamic philosophies,
rituals, laws, and sciences (Bai 2007;Ma2000:83–84). Prominent Muslim
Confucian scholars such as Wang Daiyu (c.1592–c.1658) and Liu Zhi
(c.1662–c.1724) of the Nanjing school, and Ma Zhu (1640–c.1710) and Ma
Dexin (a.k.a. Ma Fuchu, 1794–1874) of the Yunnan school had not only
earned important leeway for the Muslim ways among the Han elites, but
also became cultural icons in the contemporaneous jingtang training net-
work (Benite 2005). In the 1800 s most Han Kitab titles were reprinted at
least five times all over China (Leslie, Yang, and Yousef 2006).
Recent research on Han Kitab has mainly focused on its sources, pro-
duction, and content. Leslie (1981) compiled the bibliographic informa-
tion, book summaries, and authors’bio sketches of fifty-eight Han Kitab
titles, which were further expanded in his later more comprehensive biblio-
graphic guide to Islam in traditional China (Leslie et al. 2006). Notably,
he also identified an order of influence among the sources cited by Liu
Zhi—“a link with Central Asian Muslim sources, then Persian, and only
finally with the original Arabic sources”(Leslie and Wassel 1982: 104).
Benite (2005) offered a cultural history of Han Kitab production in rela-
tion to the organizational network of jingtang training. In addition, many
important Han Kitab texts have been translated to English and analyzed
from the angles of world philosophy and intellectual history (Frankel
2011; Lee 2015; Lipman 2016 Murata 2000,2017; Murata et al. 2009;
Petersen 2017; Tontini 2016). In the meantime, the Chinese academia also
celebrated the Han Kitab as an integral part of the Chinese history of phil-
osophy (e.g., Hua 2002;Ji2015; Jin 2008,2010,2016; Liang 2004; Sun
2011; Zhou 2002,2006).
A common finding of this body of research is that the Han Kitab is not
a simple juxtaposition of Confucianism and Islam, but an organic fusion
of the two intellectual and affectual systems. Despite its motivations,
Confucianism was more than just an instrument to promote Islam. Before
the Han Kitab authors could convince the sophisticated Han scholar-
elites, they had to first make peace with it themselves. The writing of the
Han Kitab, therefore, was not only a process of text production but also
identity negotiation. As a result, the Chinese and Islamic elements
appeared simultaneously and spontaneously throughout the Han Kitab lit-
erature, all pointing to a distinct Sino Muslim literati identity. Moreover,
Benite (2005) documented that the Han Kitab literature, through its
nationwide circulation, also facilitated the construction of a Sino Muslim
identity among the jingtang Muslim elites.
Unlike migration and intermarriage, the spread of Han Kitab was not
facilitated (let alone forced) by the state. Given its refined scholarly styles,
was the Han Kitab literature able to reach the grassroots society? More
importantly, did it contribute to a kind of Sino Muslim identity among
4 CHINESE SOCIOLOGICAL REVIEW
the vast ordinary Muslims besides the Muslim scholar elites? Is it still at
work in the contemporary China? Despite much attention to the Chinese
Muslims and Han Kitab, the social science literature has little to offer
regarding these important questions. In this article, we confront these
questions with a probability survey conducted in a major Islamic sect in
China, which we believe is the first of its kind, in the country’s northwest-
ern province, Gansu.
Xidaotang
The Xidaotang is commonly distinguished as one of the major Hui Islamic
sects in China, standing next to Gedimu,Yihewani, and four Sufi schools
(i.e., Khufiyya,Jahriyya,Qadariyaa,Kubrawiyya)(Ma2000). Shortly after
it was founded in Southern Gansu at the turn of the twentieth century, it
began to draw extensive and lasting attention from the Chinese academia,
yielding a sizeable body of research literature (e.g., Institute for Minzu
Studies, Qinghai Minzu College, and Northwest Minzu College Institute
for Northwestern Minzu Studies 1987;S.Min2010,J.Min2011). Possibly
due to its religious prominence, in 2017 Xidaotang’s Min Shengguang
became the first Chinese Islamic leader to publish a substantial collection
of research work (Min 2017). In the English literature, Lipman (1997:
186–199) and Dillon (1999: 131–152) documented Xidaotang’s history up
to 1949, based mainly on Chinese historiographic writings (e.g., Ma 1983;
XSJ 1987). More recently, ethnographic research has been conducted in
Xidaotang with a focus on its economic history (Hille 2010,2015,2018).
The Xidaotang is distinctive in several ways. Its communal socioeco-
nomic organization inspired by ummah (Gao 1999;Min1995; Zhou 1997)
and its phenomenal success as traders between the Tibetans and Han (e.g.,
Hille 2010,2015,2018;Min2011; Zhang 2010) are frequent topics in
research. In terms of religion, what makes Xidaotang stand out is its con-
nection to the Han Kitab. In fact, the Han Kitab literature, especially the
writings by Liu Zhi, occupied a central place in the teaching of
Xidaotang’s founder, Ma Qixi (1857–1914). Unlike the concurrent Islamic
movements under Sufi or Wahhabi influences, Ma Qixi did not introduce
any new canons or rituals beyond China’s northwestern borders. Rather,
he drew upon the Han Kitab and explicitly emphasized on the Sinic iden-
tity in the interpretation of Islamic meanings. In the research on Han
Kitab, too, there is an awareness of its connection to the Xidaotang (e.g.,
Frankel 2016; Cherif-Chebbi 2016; Lee 2015:30–32, 202–203).
Ma Qixi’s emphasis on the Han Kitab, and his appreciation for Liu Zhi
of the Nanjing school, may have to do with his ancestral memory and aca-
demic training. Born in Lintan, Ma Qixi shared the local collective mem-
ory that their huihui ancestors came in 1379 in an imperial army from
SUMMER 2019 5
Nanjing to put down the local rebellions (Gu et al. 2002: 217–218). Since
then, Gedimu had been the only Muslim way until Sufism entered Lintan
in the late 1700s. Ma Qixi grew up in a time when Lintan Muslims were
divided into at least three competing Khufiyya Sufi orders (known as men-
huan in Chinese)—namely, Huasi, Beizhuang, and Dingmen. Such division
may have resulted in an anxiety of religious identity which motivated Ma
Qixi’s quest for the true meaning of Islam. Son of a Beizhuang ahong,Ma
Qixi began to study Chinese and Confucian classics with local Han schol-
ars since age eleven. He excelled in the Confucian learning and earned the
prestigious civil service examination credential, xuecai, in his early twen-
ties. Instead of advancing to higher ranks, Ma Qixi returned to Islamic
learning for eleven years, studying mainly the Han Kitab. He began to
teach Han Kitab titles and Confucian classics in 1890 and finally founded
Xidaotang in 1907 (Lipman 1997;Ma2000: 113–122; Min 2011:81–89). It
is important to note, though, that Xidaotang was not the only sect that
celebrated Han Kitab—in fact, authors like Liu Zhi had extensive influen-
ces among the northwestern Sufi orders, including the aforementioned
Dingmen menhuan (Ma 2000:224) and various menhuans following
Jahriyya (Jin 1996).
Besides the conceptual relevance, the Xidaotang also offers a methodo-
logical advantage for survey research. The Xidaotang members mostly live
in Lintan, Gansu. Such geo-demographic concentration makes it possible
to construct a sampling frame. A sampling frame is rarely achievable for
religious populations in China (but see Lai and Mu 2016; Lai and
Thornton 2015; Mu and Lai 2016; Tang et al. 2016; Zang 2007,2012 for
some examples of survey research that involve Islamic groups). The
Chinese government does not collect or publish demographic data by reli-
gion, except for some aggregate estimates occasionally (State Council
2018). By ethnicity, we know that in 2010 there are 10.6 million Hui peo-
ple in China, half (50 percent) of whom live in four northwestern provin-
ces/autonomous regions (i.e., Ningxia, Gansu, Xinjiang, and Qinghai) in
more or less concentrated communities (China Statistical Bureau 2012).
Gansu was China’s traditional northwestern frontier and hence the heart-
land of many Sino Muslim traditions. Whereas other Hui Islamic sects are
widely distributed large populations with enormous internal heterogeneity,
most Xidaotang followers have concentrated in one single county in
Southern Gansu—Lintan.
Data Collection
Our data were collected in Lintan County in 2018. Given our research objec-
tives, both sampling and fielding require creative solutions. The first chal-
lenge is to reach the universe of the study. So far as we know, no Chinese
6 CHINESE SOCIOLOGICAL REVIEW
administrative data systematically identify religious affiliation at the individ-
ual level, not to mention adherence to Islamic sects. In order to build a sam-
ple, we need a list of all Xidaotang members or households. The best chance
for an enumeration of the Xidaotang population is the annual commemora-
tive Amal of Ma Qixi’s martyrdom on May 19 on the Chinese lunar calen-
dar. In 2014 we took the chance of the one hundredth commemorative Amal
to construct a list of all participating Xidaotang households, including those
who lived out of Lintan. Using that as the sampling frame, we then selected
two hundred households at random. During the survey, in each household
our interviewers would make a list of all its adult members, from which one
respondent would be randomly determined for the interview. If refused, the
interviewer would turn to their eligible neighbors (i.e., Xidaotang members)
in the order of ascending street number until an interview was successfully
conducted. Throughout the study, we encountered only six refusals due to
time conflicts, which were all successfully replaced, yielding a sample of two
hundred individual respondents. As shown in Table 1, 90 percent of the
respondents were interviewed in Lintan, and the rest who lived elsewhere
were interviewed over the phone.
The second challenge stems from the high mobility of Xidaotang mem-
bers. As reviewed before, the Xidaotang has strong traditions of trading
and education. It is known that a significant proportion of Xidaotang
households in Lintan have members doing businesses or going to schools
elsewhere. According to our survey, the proportions are 73.4 percent and
51.6 percent, respectively. Therefore, there is a nontrivial probability that
the selected individual of a household in Lintan is not available for inter-
view. Most people do come back to Lintan for Ma Qixi’s Commemorative
Amal, but it is a busy and time-limited event, and the students studying
elsewhere are not always excused for religious reasons. To increase the
chance of capturing the selected household members, we fielded the survey
during the Spring Festival in early 2018. Though Xidaotang Muslims do
not celebrate the Chinese Spring Festival, most travelers return to Lintan
for family reunions as all businesses and schools across the country are
practically closed during the national holiday. Eventually, we accom-
plished a study sample that was very similar to the known sociodemo-
graphic features of the Xidaotang population (Cf. Table 1).
The third challenge has to do with the highly diverse Islamic landscape
in Lintan. Like other parts of northwestern China, the relationships
between different Islamic sects in Lintan are complicated. As a result, sur-
veys that touch on religious topics can expect a lot of reluctance to com-
ply. Besides refusal, respondents are likely to provide convenient answers
(e.g., middle categories, Don’t Know) or “correct”answers, leading to sur-
vey satisficing and social desirability bias. To guarantee data quality, we
recruited college and graduate students from the Xidaotang—including
local elite families—to conduct the interviews. After training, they were
SUMMER 2019 7
Table 1
Sample Description by Han Kitab Exposure
Overall
Number of Han Kitab
books at home
Exposure to Han
Kitab at services
Mean (SD)/% Low High Low High Analytic N
Male (%) 55 57 53 52 58 200
Age 39.2 (16.5) 34.2 45.9 39.9 38.3 199
Urban residence (%) 58 58 56 54 61 200
Years of education 9.6 (5.4) 10.4 8.5 8.7 10.5 200
Residential Muslim concentration (1–6) 3.6 (1.5) 3.6 3.6 3.7 3.6 199
County of current residence (%) 195
Lintan (Gannan, Gansu) 90 89 92 88 94
Zhuoni (Gannan, Gansu) 5 5 5 6 4
Luqu (Gannan, Gansu) 1 1 1 2 0
Hezheng (Linxia, Gansu) 1 2 0 2 0
Guanghe (Linxia, Gansu) 2 2 1 1 2
Xicheng (Beijing) 1 0 1 0 0
Xindu (Chengdu, Sichuan) 1 1 0 1 0
Source: Xidaotang Survey 2018 (N¼200).
8 CHINESE SOCIOLOGICAL REVIEW
able to thoroughly understand the survey instrument and effectively help
the respondents go through the questionnaire in local vernacular.
Consequently, we achieved near perfect response rates on almost all items
on the questionnaire (see result tables for details). Also, survey satisficing
and social desirability bias are minimal thanks to the trusts and close
informal relationships among the Xidaotang members.
Analytical Design and Results
The main goals of the empirical analysis are to quantitatively assess the
Han Kitab tradition and its implications in Xidaotang’s Islamic religiosity
and Sinic identity. Tables 2 and 3summarize a series of indicators of the
Han Kitab exposure at home and at Jumma (the congregational prayer on
Fridays), respectively. Table 1 provides an overview of the sample.
Tables 4 and 5describe various outcomes measuring the Islamic religiosity
and Sinic affinity—both the sample averages and conditional means by
Han Kitab exposure. Tables 6 and 7present formal tests on Han Kitab’s
influences on those outcomes based on regression models.
Exposure to Han Kitab
We estimate the exposure to Han Kitab in both private and public spheres
of life. First, given Liu Zhi’s central place in Xidaotang’s teaching, we list
six of his books and ask if the respondents have them at home. As shown
in Table 2, a great majority (72 percent) have Wu Geng Yue, a colloquial
Table 2
Han Kitab Books at Home
Mean/% SD
Do you have the following Han Kitab books at home? (%)
Wu Geng Yue 《五更月》72.0
Wu Gong Shi Yi 《五功释义》50.0
Tianfang Sanzijing 《天方三字经》38.5
Tiangfang Zhisheng Shilu 《天方至圣实录》37.5
Tianfang Dian Li 《天方典礼》37.0
Tianfang Xing Li 《天方性理》36.0
Number of Han Kitab books at home (0–6) 2.7 2.3
Number of Han Kitab books at home (rescaled to 0–1) 0.45 0.38
Source: Xidaotang Survey 2018 (N ¼200).
Note: All books are authored by Liu Zhi (c.1662-c.1724).
SUMMER 2019 9
Table 3
Han Kitab Authors and Books Cited at Jumma (Friday Congregational Prayer)
Liu Zhi Wang Daiyu Ma Zhu Ma Fuchu Ma Lianyuan
刘智王岱舆马注马复初马联元
c.1662–c.1724c.1592–c.16581640–c.1710s 1794–1874 1841–1903
Mean 2.8 2.0 1.9 1.7 1.4
SD 0.8 0.9 0.8 0.8 0.7
% distribution
Never 9.5 35.0 36.0 48.7 69.3
Occasionally 14.0 38.5 38.5 37.7 24.6
Often 62.0 22.5 22.5 11.6 4.0
Every time 14.5 4.0 3.0 2.0 2.0
Analytic N 200 200 200 199 199
10 CHINESE SOCIOLOGICAL REVIEW
Liu Zhi Wang Daiyu Ma Zhu Ma Fuchu
Wu Geng
Yue
Wu Gong
Shi Yi
Tianfang
Dian Li
Tianfang
Xing Li
Tianfang
Zhisheng Shilu
Qingzhen
Da Xue
Qingzhen
Zhinan
Si Dian
Yao Hui
《五更月》《五功释义》《天方典礼》《天方性理》《天方至圣实录》《清真大学》《清真指南》《四典要会》
c.1710 c.1710 c.1709 c.1710 c.1724 c.1640s c.1680–1710 1859
Mean 2.9 2.5 2.4 2.3 2.3 1.9 2.0 1.4
SD 0.7 0.9 0.8 0.8 0.8 0.8 0.8 0.7
% distribution
Never 4.0 16.0 15.1 20.6 21.6 34.0 28.5 70.0
Occasionally 16.5 29.5 38.7 35.7 34.7 43.5 46.0 22.0
Often 67.0 46.5 41.7 39.2 40.2 18.5 21.5 6.0
Every time 12.5 8.0 4.5 4.5 3.5 4.0 4.0 2.0
Analytic N 200 200 199 199 199 200 200 200
Exposure to Han Mean SD Analytic N
Kitab at Jumma (0–1) 0.38 0.19 198
Source: Xidaotang Survey 2018 (N¼200).
Note: All dates in this table are based on the biographical sketches in Bai 2007.
SUMMER 2019 11
Table 4
Religiosity by Han Kitab Exposure
Overall
Number of Han
Kitab books
at home
Exposure to
Han Kitab at
Jumma
Mean (SD)/% Low High Low High
Self-assessed strength of faith (0–10) 9.4 (1.3) 9.2 9.7 9.2 9.6
Frequency of pray (1–6) 4.1 (1.6) 3.6 4.6 3.8 4.3
Never (%) 7 9485
Less than once a week (%) 13 17 6 14 11
Once or more a week (%) 22 25162814
Once a day (%) 13 12141214
2–4 times a day (%) 22 24181527
5 times a day (%) 25 12422228
Frequency to attend services in Mosque (1–6) 4.6 (1.4) 4.4 4.8 4.5 4.7
Never (%) 0 0000
Less than once a year (%) 2 3030
A few times a year (%) 35 38313832
Less than once a week (%) 8 9687
Once a week (%)131312816
More than once a week (%) 44 37524244
12 CHINESE SOCIOLOGICAL REVIEW
Fasting (1–3) 2.9 (0.3) 2.9 2.9 2.9 2.9
Never (%) 1 1010
Not every year (%) 8 10688
Every year (%) 92 90949192
Hajj (1–5) 2.7 (0.8) 2.6 2.7 2.6 2.7
No plan (%)10713 14 5
Not decided (%) 22 24192321
Not yet, but will apply (%) 64 68585969
Not yet, but applying (%) 3 1514
Yes (%) 3 0631
Self-identified Sunni (%) 51.5 48.7 55.3 38.4 63.6
Source: Xidaotang Survey 2018 (N¼200).
SUMMER 2019 13
Table 5
Social Distances by Han Kitab Exposure
Overall
Number of Han
Kitab books at home
Exposure to Han
Kitab at Jumma Cronbach’s
alpha
Factor
eigenvalue
(loadings)Mean/% SD Low High Low High
Hui social distance index (0–100) 95.5 (18.0) 92.8 99.2 96.8 94.1 0.82 2.50
To live next door (%)9998999998(0.25)
To do business with (%) 96 93 100 98 94 (0.78)
To invite to dinner at home (%) 98 96 100 99 96 (0.61)
To have as daughter-in-law (%)95 92999694 (0.85)
To have as son-in-law (%)9592999694 (0.85)
Han social distance index (0–100) 73.2 (18.5) 68.8 79.1 73.8 72.4 0.58 1.21
To live next door (%)9491989593(0.42)
To do business with (%)9694999795 (0.51)
To invite to dinner at home (%)88 81968887 (0.62)
To have as daughter-in-law (%)34 234732 34 (0.49)
To have as son-in-law (%) 10 7 13 11 8 (0.37)
Tibetan social distance index (0–100) 73.3 (20.7) 69.4 78.6 74.0 72.4 0.63 1.43
To live next door (%)9288969291(0.54)
To do business with (%)9593969792 (0.58)
To invite to dinner at home (%)86 80938685 (0.68)
To have as daughter-in-law (%)31 22422733 (0.46)
To have as son-in-law (%) 11 9 14 12 10 (0.37)
14 CHINESE SOCIOLOGICAL REVIEW
Uighur social distance index (0–100) 75.0 (27.6) 73.2 77.4 73.2 76.6 0.73 2.07
To live next door (%)8987928593(0.67)
To do business with (%)8786878588 (0.79)
To invite to dinner at home (%)84 82878484 (0.78)
To have as daughter-in-law (%)40 36453444 (0.50)
To have as son-in-law (%)2422272622 (0.38)
Saudi Arabian social distance index (0–100) 65.9 (30.9) 61.8 71.3 62.7 68.9 0.80 2.41
To live next door (%)8279857984(0.73)
To do business with (%)8381858085 (0.78)
To invite to dinner at home (%)82 78868083 (0.75)
To have as daughter-in-law (%)36 29453041 (0.61)
To have as son-in-law (%)2516362029 (0.57)
Egyptian social distance index (0–100) 64.2 (31.7) 61.4 67.9 61.2 67.0 0.81 2.50
To live next door (%)8078817782(0.80)
To do business with (%)8180827884 (0.78)
To invite to dinner at home (%)80 77847881 (0.84)
To have as daughter-in-law (%)27 23332232 (0.55)
To have as son-in-law (%)191326 14 23 (0.49)
Iranian social distance index (0–100) 62.8 (31.8) 60.8 65.7 60.9 64.6 0.80 2.43
To live next door (%)7877797680(0.82)
To do business with (%)7977807879 (0.80)
To invite to dinner at home (%)79 77827880 (0.82)
(continued)
SUMMER 2019 15
Table 5 Continued.
Overall
Number of Han
Kitab books at home
Exposure to Han
Kitab at Jumma Cronbach’s
alpha
Factor
eigenvalue
(loadings)Mean/% SD Low High Low High
To have as daughter-in-law (%)26 23292130 (0.50)
To have as son-in-law (%)1510211217 (0.44)
Malay social distance index (0–100) 60.7 (31.4) 58.7 63.4 58.6 62.7 0.80 2.47
To live next door (%)7979797682(0.82)
To do business with (%)7977817880 (0.80)
To invite to dinner at home (%)78 76807778 (0.82)
To have as daughter-in-law (%)26 23292229 (0.50)
To have as son-in-law (%)1712241420 (0.44)
Pakistan social distance index (0–100) 52.0 (27.1) 49.3 55.6 50.3 53.7 0.78 1.68
To live next door (%)7876817680(0.65)
To do business with (%)8282817984 (0.64)
To invite to dinner at home (%)78 75827878 (0.31)
To have as daughter-in-law (%)22 18272024 (0.64)
To have as son-in-law (%)1410191315 (0.59)
Source: Xidaotang Survey 2018 (N¼200).
16 CHINESE SOCIOLOGICAL REVIEW
poem of 636 characters that interprets the Islamic faith with a moon meta-
phor. Half of the sample have Wugong Shiyi, a text explaining the five pil-
lars of Islam. Near 40 percent have Tianfang Sanzijing, a 1,344-character
text of traditional Chinese verbal education for small children that also
teaches basic Islamic knowledge. As for the three more scholarly works,
the percentages are lower, but none are lower than one third of the sam-
ple. Specifically, 37.5 percent have Tianfang Zhisheng Shilu (biography of
Muhammad), 37 percent have Tianfang Dian Li (Law and Rituals of
Islam), and 36 percent have Tianfang Xing Li (Metaphysics of Islam). On
average, each respondent has 2.7 books that are listed in our question-
naire. A great majority (80.5 percent) have at least one of these six books
at home. Note that Liu Zhi is the most influential but not the only Han
Kitab authors, so we underestimate the real exposure to Han Kitab at
home. Even so, these results provide convincing evidence that the grass-
roots-level Xidaotang society are under substantial exposure to the Han
Kitab at home.
In addition, to assess the respondent’s Han Kitab exposure in the pub-
lic sphere of life, we ask how often they hear ahongs cite the Han Kitab
authors and books at the Friday congregations. As shown in Table 3, Liu
Zhi is by far the most popular author, followed by Wang Daiyu, Liu Zhi’s
predecessor in the Nanking School of Han Kitab, and three other authors
of the Yunan School (i.e., Ma Zhu, Ma Fuchu, and Ma Lianyuan). In
particular, 14.5 percent of the sample reported that Liu Zhi’s name was
heard every time they attended the Friday services. Only 7.5 percent have
never heard any of those names during the services (results not reported in
table). Regarding the books, the five listed titles by Liu Zhi, again, are
more frequently cited at the congregations. Like the home exposure, Wu
Geng Yue and Wugong Shiyi are more popular than the three more schol-
arly books. A total of 79.5 percent reported that Wu Geng Yue was cited
“often”or “every time”during the congregations. For his other four
books, including the three scholarly titles, this percentage is well above 40
percent. In contrast, books by other authors are significantly less cited at
Jumma, yet only a minority (around 30 percent) have never heard of the
listed books by Wang Daiyu and Ma Zhu. The only exception is Ma
Fuchu’sSidian Yaohui (Essentials of Four Canons), which 70 percent have
never heard it cited at the congregations. This is possibly because the book
was finished in 1859 and had not attained canonic status when Ma Qixi
began his teaching in 1890. Taken together, only 1 percent of our entire
sample, or 2 respondents, reported that they had never heard ahong cite
any of the listed books (results not shown in table).
To sum up, in contemporary times the Xidaotang Muslims continue to
have extensive exposure to the Han Kitab, which exists in both private
and public spaces. As a Xidaotang tradition, they hold respect for the
Nanking School, especially Liu Zhi. Unlike in academia, it is Liu Zhi’s
SUMMER 2019 17
Table 6
Han Kitab’s Influences on Religiosity
Self-assessed
strength of faith
Frequency
of pray Fasting Hajj Sunni
Number of Han Kitab books at home (0–1) 0.17 (0.27) 0.33 (0.24) 0.06 (0.06) 0.24 (0.16) 0.77 (0.39)
Exposure to Han Kitab at Jumma (0–1) 1.55 (0.52) 1.64 (0.45) 0.05 (0.12) 0.13 (0.30) 32.20 (32.95)
Frequency to attend services in Mosque (1–6) 0.16 (0.08)0.36 (0.07) 0.03 (0.02)†0.04 (0.04) 1.13 (0.16)
Male (ref.¼female) 0.22 (0.22) 0.42 (0.19)0.08 (0.05) 0.30 (0.13)4.08 (1.59)
Age (in years) 0.01 (0.01) 0.03 (0.01) 0.00 (0.00) 0.01 (0.00)†1.01 (0.01)
Urban residence (ref.¼rural) 0.36 (0.19)†0.28 (0.17) 0.04 (0.05) 0.01 (0.11) 1.24 (0.43)
Years of education (in years) 0.01 (0.02) 0.09 (0.02) 0.00 (0.01) 0.03 (0.01)1.01 (0.04)
Residential Muslim concentration (1–6) 0.07 (0.07) 0.01 (0.06) 0.01(0.02) 0.05 (0.04) 0.82 (0.10)†
Constant 7.31 (0.62) 1.15 (0.54)2.64 (0.14) 2.01 (0.36) 0.10 (0.11)
County of current residence (results suppressed)
R
2
0.2202 0.5569 0.1044 0.1908 0.1811
Analytic N 191 191 191 191 190
Source: Xidaotang Survey 2018 (N¼200).
Notes: Logit model is estimated for Sunni identification and the coefficients are odds ratios. All other models are OLS models.
†p<0.1, p<0.05, p<0.01, p<0.001.
18 CHINESE SOCIOLOGICAL REVIEW
Table 7
Han Kitab’s Influences on Social Distances
Chinese Foreign Muslims
Hui Han Tibetan Uighur Saudi Arabian Egyptian Iranian Malay Pakistani
Number of Han Kitab books at home (0–1) 7.939.957.98†3.25 10.84 1.53 0.13 1.39 4.19
(3.7) (3.8) (4.3) (6.0) (6.8) (6.9) (7.0) (6.9) (5.9)
Exposure to Han Kitab at Jumma (0–1) 4.90 7.33 10.25 4.71 12.10 14.52 6.45 9.56 5.23
(7.5) (7.8) (8.7) (12.2) (13.8) (14.2) (14.2) (14.1) (12.0)
Frequency to attend services in
Mosque (1–6)
0.92 0.02 0.08 1.63 0.18 1.34 0.33 0.46 0.50
(1.1) (1.2) (1.3) (1.8) (2.1) (2.1) (2.1) (2.1) (1.8)
Self-assessed strength of faith (0–10) 0.03 0.19 0.15 2.47 0.54 1.90 1.86 2.16 1.63
(1.0) (1.1) (1.2) (1.6) (1.9) (1.9) (1.9) (1.9) (1.6)
Frequency of pray (1–6) 1.88 0.23 0.16 1.06 0.25 1.87 1.20 1.16 0.15
(1.2) (1.2) (1.4) (1.9) (2.2) (2.2) (2.2) (2.2) (1.9)
Fasting (1–3) 12.87 2.32 0.92 5.01 6.14 0.63 0.12 0.21 2.92
(4.4) (4.6) (5.1) (7.2) (8.2) (8.4) (8.4) (8.3) (7.1)
Hajj (1–5) 0.77 1.25 1.40 0.24 0.58 3.99 2.36 2.90 1.35
(1.7) (1.8) (2.0) (2.8) (3.2) (3.3) (3.3) (3.2) (2.8)
Self-identified Sunni (ref.¼others) 1.68 2.79 3.40 2.32 0.34 4.12 2.07 3.69 0.16
(2.8) (2.9) (3.2) (4.5) (5.1) (5.3) (5.3) (5.2) (4.5)
(continued)
SUMMER 2019 19
Table 7 Continued.
Chinese Foreign Muslims
Hui Han Tibetan Uighur Saudi Arabian Egyptian Iranian Malay Pakistani
Male (ref.¼female) 0.69 3.91 6.10†10.422.74 5.20 4.64 5.09 0.22
(3.1) (3.2) (3.6) (5.1) (5.7) (5.8) (5.9) (5.8) (5.0)
Age (in years) 0.31 0.44 0.51 0.410.30 0.480.60 0.520.59
(0.1) (0.1) (0.1) (0.2) (0.2) (0.2) (0.2) (0.2) (0.2)
Urban residence (ref.¼rural) 1.01 0.05 0.89 1.81 3.26 1.73 1.68 2.06 0.98
(2.6) (2.8) (3.1) (4.3) (4.9) (5.0) (5.0) (5.0) (4.2)
Years of education (in years) 0.39 0.760.911.140.82 1.04 1.601.281.37
(0.3) (0.4) (0.4) (0.6) (0.6) (0.6) (0.6) (0.6) (0.5)
Residential Muslim concentration (1–6) 0.23 0.24 0.06 1.10 4.112.51 2.00 2.55 2.20
(0.9) (0.9) (1.0) (1.5) (1.7) (1.7) (1.7) (1.7) (1.4)
Constant 53.71 38.9045.2340.30 8.93 27.90 26.44 27.45 9.65
(16.2) (16.9) (18.8) (26.5) (29.9) (30.6) (30.8) (30.4) (26.0)
County of current residence (results suppressed)
R
2
0.2170 0.2233 0.2305 0.1307 0.1384 0.1420 0.1314 0.1352 0.1380
Source: Xidaotang Survey 2018 (N¼200; Analytic N¼191).
Notes: All models in this table are OLS models. †p<0.1, p<0.05, p<0.01, p<0.001.
20 CHINESE SOCIOLOGICAL REVIEW
more colloquial writings, such as Wu Geng Yue and Tianfang Sanzijing,
that have gained more popularity at the grassroots level.
For parsimony, we constructed two composite scales—“number of Han
Kitab books at home”(see Table 2) and “exposure to Han Kitab at
Jumma”(see Table 3). The first scale is the simple sum of the listed books
at home that ranges from 0 to 6. The second scale is a principle compo-
nent factor extracted from all thirteen items in Table 3. As the thirteen
items are highly correlated (Cronbach’s alpha ¼0.89), the factor bears a
very high eigenvalue of 5.24 and positive loadings on all thirteen items. To
facilitate the interpretation, we constrained both scales between 0 and 1.
Note that the unconditional means of these two scales do not have natural
interpretations—they are created to capture the variances of Han
Kitab exposure.
What do people under different levels of Han Kitab exposure look like?
Table 1 provides sociodemographic profiles by the level of exposure. First,
we dichotomized the sample based on their scores on the home exposure
scale into the upper 42.5 percent and the lower 57.5 percent (the variation
does not allow for equal-size division). Those have more books at home
are much older (45.9 vs. 34.2) and less educated (8.5 vs. 10.4), which sug-
gests Han Kitab books are more popular among the older cohorts.
However, the pattern is reversed if we group by exposure at Jumma, where
more exposure is associated with younger age (38.3 vs. 39.9) and more
education (10.5 vs. 8.7). Therefore, we cannot jump to the conclusion that
the Han Kitab exposure has declined over time. However, one pattern is
clear and consistent—those under more Han Kitab exposure are more
likely to be residents in Lintan (92 percent vs. 89 percent, and 94 percent
vs. 88 percent).
Han Kitab’s Influences on Islamic Religiosity
We now move on to Han Kitab’s influence on Xidaotang members’
Islamic religiosity. Our data allow us to examine their strength of faith,
key Islamic practices, and their identification with the international sectar-
ian divisions. Specifically, we use a self-reported rating to indicate the
strength of faith. Regarding behavioral practices, we were able to ask
about three out of the five pillars of Islam—the daily prayers, the annual
fasting during the Ramadan, and the pilgrimage to Mecca (i.e., Hajj). The
other two were missed because practically all Xidaotang Muslims can recite
Shahada (declamation of faith) in Arabic, and data on donation would
reveal respondents’financial privacy. As mosques play a central role in
the religious life of the northwestern Muslims, the survey also investigated
how often they attended services in Mosques. Finally, respondents were
asked if they self-identify with the major sects in the Muslim world (i.e.,
Sunni or Shia). With these outcomes, we examine Xidaotang Muslims’
SUMMER 2019 21
Islamic belief and behaviors and how Han Kitab contribute to them.
Tables 4 and 6report descriptive patterns and formal tests, respectively.
Three major findings emerge from Table 4. First, Xidaotang followers
are pious Muslims. On a scale from 0 to 10, our respondents rated their
strength of faith at a very high average of 9.4. This is consistent with their
frequent prayers (60 percent pray once or more per day) and service
attendances (57 percent goes to Mosques more than once a week).
Moreover, an overwhelming majority (92 percent) observe fasting every
year. Though only 3 percent have been to Mecca, 67 percent have plans to
go on Hajj. Second, though Xidaotang can be traced to Hanafi, one of the
four Sunni schools of jurisprudence, its members are not well informed of
this. In fact, only 51.5 percent of the sample confidently identified with
Sunni. Our fieldwork leads us to believe that they understand the Sunni
identity as a piece of religious knowledge, which has little political implica-
tion. The third finding has to do with the impact of Han Kitab. For all
five outcomes, the exposure to Han Kitab—both at home and at
Jumma—produces positive effects, driving our respondents to stronger
faith, greater engagement with Islamic practices, and higher likelihood to
self-identify with Sunni.
The Han Kitab influences observed in Table 4, however, are bivariate
patterns based on sample data. Are those influences independent of the
demographic and socioeconomic compositions? Can they be generalized to
the Xidaotang population? To answer these questions, we move on to
examine the model results in Table 6. As shown in the table, regression
models are estimated for all outcomes in Table 4 except mosque attend-
ance. This is chiefly because one of our key predictors, “exposure to Han
Kitab at Jumma,”is conditional on going to mosque, thus making it an
illogical pretreatment outcome. A second methodological reason is that
since we aim to estimate the net effect of the “exposure to Han Kitab at
Jumma”, we must consider mosque attendance as a control variable.
Fortunately, the three pillars—prayer, fasting, and hajj—are still valid
dependent variables measuring behavioral aspect of the Islamic religiosity.
For simplicity, we linearized those outcomes using integer scoring (see
Table 4 for descriptives) and estimated OLS models to predict them. For
the binary Sunni identity, we performed logistic transformation to estimate
the log odds ratios. The coefficients reported in Table 6 are converted to
the odds ratios for easy interpretation.
As shown in Table 6, the coefficients on both Han Kitab scales are all
positive except for the Sunni model, where the first scale bears an odds
ratio of 0.77. The overall message is that both types of Han Kitab expos-
ure increase the Islamic religiosity of our sample respondents even in the
presence of demographic and socioeconomic confounders. In terms of
inferential results, however, the Han Kitab exposure in the private sphere
(i.e., number of books at home) has no statistically significant impact on
22 CHINESE SOCIOLOGICAL REVIEW
any of the outcomes. On the other hand, more frequent references to the
Han Kitab at Jumma significantly increase one’s strength of faith, fre-
quency to pray, and Sunni awareness. The conclusion is clear—Han Kitab
matters, and its influences on Islamic religiosity mainly stem from religious
activities in the public sphere. Also, such influences do not exist for fasting
or hajj at the population level, possibly due to the lack of variation in the
fasting variable and the imperfect linearity of our hajj measure.
Han Kitab’s Influences on Sinic Identity
Now that we know Han Kitab contributes to Xidaotang’s Islamic religios-
ity, does it also help with their Sinic identity? We understand Sinic identity
as the extent to which one self-identifies with the Chinese. The direct
measurement, therefore, should be a Likert scale to assess statements like
“I consider myself as Chinese.”However, ethnoreligious studies in China
are known to suffer from social desirability bias—researchers can expect
that near 100 percent of the respondents would choose the politically cor-
rect option “strongly agree.”Therefore, we choose to assess the identity
with social distance scales instead.
Identity is best revealed in comparison. Accordingly, our empirical
strategy is to use social distance indicators to evaluate Xidaotang Muslims’
psychological distances with various social groups from China and other
countries. If a Sinic identity is present, we anticipate observing closer psy-
chological distances to the Chinse groups than the foreigners. For the
China part, we included Hui, Han, Tibetans, and Uighurs. For foreign
Muslims, we selected Saudi Arabians, Egyptians, Iranians, Malays, and
the Pakistani. As Xidaotang is part of the Hui designation, Hui is inter-
preted in our survey as other Hui groups such as Gedimu, Yihewani,
Salaffiya, or various Sufi orders in the northwest. Unlike the Han and
Tibetans, Uighurs are Muslims with Turkic ancestry living in Xinjiang,
which borders on Gansu in the northwest. Our choices of Muslim coun-
tries cover major Sunni (Saudi Arabia) and Shia (Iran) powers in the
Middle East. Egypt has long been regarded by the Chinese Muslims as an
intellectual center of the Islamic world. As the northwestern Muslims
maintain strong bonds with the Muslims in Central and Southeastern
Asia, we also included Pakistan and Malaysia (a Muslim majority country
with a large Chinese diaspora). Therefore, all the Chinese groups and for-
eign Muslim countries are well recognizable among our respondents,
which strengthens the validity of our attitudinal measurements.
As shown in Table 5, regarding each social group, we asked our
respondents if they were willing to live next-door to them, do businesses
with them, invite them over for dinner at home, have them as daughter-in-
law, and have them as son-in-law. The group-specific endorsement rates in
the first column reveal distinct orders of preference. For the first three
SUMMER 2019 23
social dimensions (i.e., neighbors, businesses, dinners at home), the
Xidaotang members clearly favor the Chinese groups more than foreign
Muslims; within the Chinese groups, the preference order goes from Hui
to Han to Tibetans and finally to the Uighurs without a single exception.
However, when it comes to marital choices, the patterns become sys-
tematically different due to stronger preferences for Muslim partners—
both domestic and foreign. For potential daughter-in-law, the Uighurs
become more popular, changing the within-Chinese order to Hui (95 per-
cent), Uighur (40 percent), Han (34 percent), and Tibetan (31 percent).
Moreover, 36 percent of our respondents are willing to have a daughter-
in-law from Saudi Arabia, which places a foreign Muslim group above the
Chinese Han and Tibetans. This favor does not exist for other foreign
Muslims, possibly due to the popular association between hajj and Saudi
Arabia. Finally, in terms of son-in-law, the norm of religious homogamy
increases dramatically. Not only Uighurs (24 percent) lead Tibetans (11
percent) and Han (10 percent) by big margins, all foreign Muslim groups
are preferred over the Tibetans and the Han Chinese, making those two
non-Muslim groups the least desirable potential spouses for their daugh-
ters. This is not surprising given the patriarchal Chinese cultural context—
it is widely believed that Muslim women cannot keep their Islamic ways if
they marry out of the Muslim society.
Overall, our Xidaotang respondents are psychologically closer to the
Chinese groups than foreign Muslims, especially when marriage is not
involved. To see if Han Kitab plays a role here, we need a more parsimo-
nious approach. First, we constructed a composite social distance index
for each social group and then compare its means by levels of Han Kitab
exposure. For simplicity and consistency, each social distance index is
exacted as a principal component factor and then rescaled to 0–100. For
each index, Table 5 reports the Cronbach’s alpha of its five component
items, as well as the factor eigenvalue and loadings. The five-item reliabil-
ity is rather high for all the social groups—all the alpha values are well
above 0.7 except Han (0.58) and Tibetans (0.63), possibly due to the con-
flicting views between marriage and other social dimensions for the Han
and Tibetans. Moreover, the eigenvalues are all greater than 1 and the fac-
tor loadings are all positive and large values, meaning that they are all
good measures in the direction of closer social distance. It is also worth
noting that across all the nine social groups, the eigenvalues for the
second-best factors are all smaller than 1.
As the social distance indices are all factor-based scales, the uncondi-
tional values are not interpretable. However, by comparing their condi-
tional means by Han Kitab exposure in Table 5, we can tell if the Han
Kitab drives our respondents psychologically closer to each social group.
By the number of Han Kitab books at home, the patterns are very clear
and consistent—higher exposure leads to closer social distance regardless
24 CHINESE SOCIOLOGICAL REVIEW
of the social group. However, the observed effects of the congregation
exposure is more complicated—while it draws the respondents closer to
Uighur Muslims and all the foreign Muslims, it works the opposite way
for Hui (94.1 vs 96.8), Han (72.4 vs. 73.8), and Tibetans (72.4 vs. 74).
Nonetheless, these may just be random disturbances as such differences
are very small in magnitude.
Due to the distinct social and religious characters of the rated groups,
the index outcomes must depend on not just the Han Kitab but also many
other religious and social factors. We now proceed to multiple regression
models to formally test the net effect of the Han Kitab tradition. In add-
ition to the demographic and socioeconomic variables, in these models we
also perform a full control for the Xidaotang religiosity—including self-
rated strength of belief, frequency of prayers and mosque attendances,
fasting, hajj, and Sunni identity. Thus estimated, the coefficients on the
Han Kitab measures can be interpreted as the net effects with greater
methodological rigor.
As shown in Table 7, net of all other influences, the home exposure to
Han Kitab drives Xidaotang Muslims significantly closer to other Hui
groups (7.93), the Han Chinese (9.95), and their Tibetan neighbors
(7.98)—all are Chinese nationalities/ethnicities. While the Uighur coeffi-
cient is also positive (3.25), it is not statistically significant—so are all
other foreign Muslims. On the other hand, the exposure at congregations
does not have any significant impact across all the models. Thus, we come
to three conclusions. First, the exposure to Han Kitab does shorten
Xidaotang members’psychological distances to the Chinese people, except
for the Uighurs. Second, this Han Kitab effect works for the Chinese
groups regardless of religion but does not apply to foreign Muslims.
Third, unlike our previous findings with the Islamic religiosity, Han
Kitab’s influences on such Sinic affinity work not through the public
sphere of religious life but the exposure at home.
Conclusions and Discussions
In this article we have examined the individual-level coexistence of two
expansive and seemingly incompatible civilizations—Islam and China.
Literature indicates that distinct Sino Muslim identities have been under
construction through integration and resistance by numerous Muslims ori-
ginated from Arab, Persia, and Central Asia who came to live in China.
As a result of two centuries of coercive marital assimilation during Ming
Dynasty (1368–1644), the huihui people emerged across China as a popula-
tion of (more or less) Chinese-looking, Chinese-speaking Muslims. This
was followed by a Han Kitab movement which lasted for about one cen-
tury and a half, during which a collection of Islamic literature written in
SUMMER 2019 25
Chinese were systematically produced and gradually acquired near canon-
ical status. Itself a product of the Sino Muslim knowledge elites, the Han
Kitab tradition has penetrated through China’s space-time and social
strata and continued to construct the Sino Muslim identities in the grass-
roots-level societies.
This article uses Xidaotang as a specimen to examine this unique reli-
gious-cultural heritage and its contemporary relevance. Based on an
innovative household survey in 2018, we come to three major conclusions.
First, the over three hundred-year-old Han Kitab written in traditional
Chinese, especially Liu Zhi’s works, continue to be widely accessible in
both private and public spheres of life in Xidaotang. Unlike among the
academic elites, it is Liu Zhi’s more colloquial writings that have gained
greater popularity among ordinary Muslims. Second, we observe a simul-
taneous saliency in Islamic religiosity and Chinese affinity, which we argue
is characteristic of the Sino Muslim identity. Finally, we identify a linkage
between the Han Kitab tradition and this Sino Muslim identity.
Specifically, those under more Han Kitab influences at religious services
tend to be more pious, more diligent in Islamic practices, and more likely
to be aware of their Sunni identity; whereas the private-sphere exposure to
Han Kitab draws Xidaotang Muslims psychologically closer to various
Chinese social groups regardless of religion, while the same effect does not
exist for foreigners, even if they are Muslims.
As this study is the first survey research on the Xidaotang—in fact, to
our knowledge the first probability survey on any Islamic sect in China—
our analysis also presents many empirical results of distinct value besides
the focal findings. We found that the global sectarian identity of Sunni
does not occupy a central place in the Xidaotang mentality. As pious as
they are, only a half our respondents know for sure that they are Sunni.
Furthermore, our analysis of the social distance indices shows that the
Sunni awareness neither pushes them closer to any of the Sunni countries
nor pulls them away from Iran (see Table 7). This means that, to
Xidaotang Muslims, the Sunni identity is chiefly understood as a piece of
religious knowledge with little sociopolitical implications. To us, this lack
of statistical significance is a finding of great substantive importance.
Based on our field experience, we propose that this could be a general fact
of the grassroots-level Chinese Muslim societies. More empirical efforts
should be made to assess such basic facts.
Of course, our cross-sectional data have its obvious limitations in terms
of enabling causal arguments. Nonetheless, it is important to note that our
analysis does not stem from a causal argument. We believe that the expos-
ure to Han Kitab and the Sino Muslim identity have mutually reinforced
each other in the historical dynamics. By formulating the Sino Muslim
identity as the dependent variable, we do not intend to posit that the
causal relation cannot go the other way around. Rather, this analysis
26 CHINESE SOCIOLOGICAL REVIEW
simply performs a standard regression decomposition to understand the
variation in our focal substantive outcome—Sino Muslim identity.
Finally, we acknowledge the plurality of the Sino Muslim identities.
Like all major ethnoreligious groups in China, identity is never a homoge-
neous construct (e.g., Jankowiak 2013). By using the term Sino Muslim
our purpose is to emphasize on the simultaneity and the spontaneity of
the Islamic and Sinic characters. The Xidaotang case offers a distinct vista
onto our argument, but it is by no means the only child of the Islamic
spirituality and the Chinese society. For example, Ma Qixi is often quoted
for claiming “Jielian (Liu Zhi) planted the seeds, Guanchuan blossomed,
and I will bear fruits.”Indeed, Guanchuan is Ma Mingxin (1719–1781)
who founded China’s largest Sufi order, Jahriyya, which opened the door
to other possibilities of Sino Muslim identities. Also, we do not posit that
Sino Muslim identities can only be constructed retrospectively through
traditional cultural forces. The contemporary China’s shifting ethnic poli-
cies and its globalizing ambitions, especially those Eurasian development
programs, are continuing to shape the identities of the Muslims in China
on an ongoing basis.
China’s twenty-three million Muslims house immense sociocultural het-
erogeneity. Given China’s fast-growing arms of globalization (e.g., Belt
and Road initiatives), the world can expect increasingly extensive contact
between China and the Islamic world, where various Sino Muslims might
play a crucial role (e.g., Ho 2013). It is our belief that the social sciences
have an urgent need to understand this theoretically interesting and sub-
stantively important population. With this paper, we contribute a first
quantitative attempt to understand an evolving Sino Muslim character
that is found very much alive in China’s southern Gansu.
Acknowledgments
This research was supported by the Morris and Anita Broad Research Fellowship
by the Steven J. Green School of International and Public Affairs, Florida
International University, and the Provost Humanities Research Grant by the
Florida International University. I am grateful for Yumei Ding and Junqing Min
for the data collection. I also thank Zheng Mu, Carlos Grenier, Lana Shehadeh,
Yumei Ding, and Junqing Min for their helpful comments on the paper.
References
Atwill, David G. 2005. The Chinese Sultanate: Islam, Ethnicity, and the Panthay
Rebellion in Southwest China, 1856–1873. Stanford: Stanford University Press.
Bai, Shouyi. 2007. Zhongguo Huihui Minzu Shi [an Ethnic History of Chinese
Huihui]. Beijing: Zhonghua Press.
SUMMER 2019 27
Benite, Zvi Ben-Dor. 2005. The Dao of Muhammad: A Cultural History of Muslims
in Late Imperial China. MA: Harvard University Press.
Chen, Yuan, and Hsing-Hai Chʻien. 1966. Western and Central Asians in China
under the Mongols: Their Transformation into Chinese. CA: Monumenta Serica
at the University of California.
Cherif-Chebbi, Leila. 2016. “Between ‘Abb al-Wahhab and Liu Zhi: Chinese
Muslim Intellecturals at the Turn of the Twenty-first Century.”In Islamic
Thought in China. Sino-Muslim Intellectual Evolution from the 17th to the 21st
Century, edited by Jonathan Lipman, 197–232. Edinburgh: Edinburgh
University Press.
China Statistical Bureau. 2012. Zhongguo 2010 Nian Renkou Pucha Ziliao [2010
Population Census of the People’s Republic of China]. Beijing: China Statistics
Press.
Dillon, Michael. 1999. China’s Muslim Hui Community: Migration, Settlement and
Sects. Richmond, Surrey: Curzon.
Frankel, James D. 2011. Rectifying God’s Name: Liu Zhi’s Confucian Translation
of Monotheism and Islamic Law. HI: University of Hawai’i Press.
Frankel, James D. 2016. “Liu Zhi: The Great Integrator of Chinese Islamic
Thought.”In Islamic Thought in China. Sino-Muslim Intellectual Evolution from
the 17th to the 21st Century, edited by Jonathan Lipman, 34–54. Edinburgh:
Edinburgh University Press.
Gao, Zhanfu. 1999. “Guanyu Xidaotang dajiating zhuzhi de diaocha yu yanjiu
[An Investigation on the Xidaotang’s Big Family Organization].”Gansu Minzu
Yanjiu 2:63–78.
Gellner, Ernest. 1983. Muslim Society. Cambridge, UK: Cambridge University
Press.
Gillette, Maris Boyd. 2000. Between Mecca and Beijing: Modernization and
Consumption among Urban Chinese Muslims. Stanford, CA: Stanford University
Press.
Gladney, Dru C. 1996. Muslim Chinese: Ethnic Nationalism in the People’s
Republic. Cambridge, MA: Harvard University Press.
Gladney, Dru C. 1998a. “Clashed civilizations? Muslim and Chinese identities in
the PRC.”In Making Majorities: Constituting the Nation in Japan, Korea,
China, Malaysia, Fiji, Turkey and the United States, edited by Dru C. Gladney,
106–31. Stanford: Stanford University Press.
Gladney, Dru C. 1998b. Ethnic Identity in China: The Making of a Muslim
Minority Nationality. New York: Harcourt Brace College Publishers.
Gladney, Dru C. 2003. “Islam in China: Accommodation or Separatism?”The
China Quarterly 174:451–67. doi:10.1017/S0009443903000275.
Gladney, Dru C. 2004. Dislocating China: Reflections on Muslims, Minorities, and
Other Subaltern Subjects. Chicago: Univ. of Chicago Press.
Gu, Jiegang, Deyi Li, Xianjia Cheng, and Tuo Ming. 2002. Xibei Kaocha Riji [a
Journal of Northwest Feild Studies]. Lanzhou: Gansu People’s Press.
Hille, Marie-Paule. 2010. “Taoshang jingli de pianduan: Jingying zhishi de chuan-
cheng jiqi dangdai zhuanxing [Episodes of Taozhou Hui business history: Trade
experiences, transmission of merchant knowledge, and contemporary trans-
formations].”In Zhongguo Jinxiandai Hangye Wenhua Yanjiu [Studies on
Occupational Cultures in Modern China], edited by Christian Lamouroux,
218–61. Beijing: National Library Press.
Hille, Marie-Paule. 2015. “Rethinking Muslim-Tibetan Trade Relations in Amdo.”
In Muslims in Amdo Tibetan Society: Multidisciplinary Approaches, edited by
Marie-Paule Hille, Bianca Horlemann, and Paul K. Nietupski, 179–206.
Lanham, MD: Lexington Books.
28 CHINESE SOCIOLOGICAL REVIEW
Hille, Marie-Paule. 2018. “Interacting with The ‘Himalayan Ummah’. The Case of
Xidaotang, a Chinese Muslim Community from Lintan.”HIMALAYA, the
Journal of the Association for Nepal and Himalayan Studies 38(2):13.
Ho, Ping-Ti. 1998. “In Defense of Sinicization: A Rebuttal of Evelyn Rawski’s
“Reenvisioning the Qing.”The Journal of Asian Studies 57(01):123–55. doi:10.
2307/2659026.
Ho, Wai-Yip. 2013. “Mobilizing the Muslim Minority for China’s Development:
Hui Muslims, Ethnic Relations and Sino-Arab Connections.”Journal of
Comparative Asian Development 12(1):84–112. doi:10.1080/15339114.2012.
749119.
Hu, Yunsheng. 2007. Chuancheng yu Rentong: Henan Huizu Lishi Bianqian Yanjiu
[Heritage and Identity: The Historical Changes of the Hui in Henan].
Yingchuan: Ningxia People’s Press.
Hua, Tao. 2002. “Wenming Duihua: Zhongguo Musilin yu Feimusilin de Duihua
[Dialogue between Civilizations: Dialogue between Chinese Muslims and Non-
Muslims].”Hui Studies 4:38–43.
Huntington, Samuel P. 1997. The Clash of Civilizations and the Remaking of World
Order. India: Penguin Books India.
Institute for Minzu Studies, Qinghai Minzu College, and Northwest Minzu
College Institute for Northwestern Minzu Studies. 1987. Historical Materials on
Xidaotang [Xidaotang shiliao ji].
Israeli, Raphael. 1977. “Muslims in China: The Incompatibility between Islam and
the Chinese Order.”T’oung Pao 63(4):296–323. doi:10.1163/156853277X00105.
Israeli, Raphael. 1980. Muslims in China: A Study in Cultural Confrontation.
London: Curzon.
Israeli, Raphael. 2002. Islam in China: Religion, Ethnicity, Culture, and Politics.
Lanham, MD: Lexington Books.
Jankowiak, William. 2013. “Urban Mongols the Flourishing of Multiple Identities
in a Vibrant Ethnic Community.”Chinese Sociological Review 45(3):53–73. doi:
10.2753/CSA2162-0555450303.
Jaschok, Mari, and Jingjun Shui. 2013. The History of Women’s Mosques in
Chinese Islam. Richmond, Surrey: Curzon.
Ji, Fangtong. 2015. Yi Ru Huitong Yanjiu [Studies on Islam-Confucianism
Synthesis]. Yinchuan: Ningxia People’s Press.
Jin, Yijiu. 1996. “Liuzhi sixiang zai zhongguo musilin zhong de yingxiang [The
influences of Liu Zhi’s thoughts among Chnese Muslims].”Gansu Minzu Studies
3–4:48–70.
Jin, Yijiu. 2008. Wang Daiyu Sixiang Yanjiu [a Study on the Thoughts of Wang
Daiyu]. Beijing: Minzu Press.
Jin, Yijiu. 2010. Zhongguo Yisilan Tanmi: Liuzhi Yanjiu [an Inquiry into Chinese
Islam: Liu Zhi Studies]. Beijing: Renmin University of China Press.
Jin, Yijiu. 2016. Zhongguo Yisilan Xianxian: Ma Zhu Sixiang Yanjiu [Chinese
Islamic Sage: A Study on the Thoughts of Ma Zhu]. Beijing: Social Science
Literature Press.
Ke, Fan. 2001. “Maritime Muslims and Hui Identity: A South Fujian Case.”
Journal of Muslim Minority Affairs 21(2):309–32. doi:10.1080/
1360200120092888.
Lai, Qing, and Zheng Mu. 2016. “Universal, yet Local: The Religious Factor in
Chinese Muslims’Perception of World Developmental Hierarchy.”Chinese
Journal of Sociology 2(4):524–46. doi:10.1177/2057150X16674126.
Lai, Qing, and Arland Thornton. 2015. “The Making of Family Values:
Developmental Idealism in Gansu, China.”Social Science Research 51(0):
174–88. doi:10.1016/j.ssresearch.2014.09.012.
SUMMER 2019 29
Lee, David. 2015. Contextualization of Sufi Spirituality in Seventeenth-and
Eighteenth-century China: The Role of Liu Zhi (c.1662–c.1730). Eugene, OR:
Wipf and Stock Publishers.
Leslie, Donald. 1981. Islamic literature in Chinese, late Ming and early Chʻing:
Books, authors, and associates. Canberra: Canberra College of Advanced
Education.
Leslie, Donald. 1998. The Integration of Religious Minorities in China: The Case of
Chinese Muslims. Canberra: Australian National University Canberra.
Leslie, Donald, Yang Daye, and Ahmed Yousef. 2006. Islam in Traditional China:
A Bibliographical Guide. London: Routledge.
Leslie, Donald Daniel, and Mohamed Wassel. 1982. “Arabic and Persian Sources
Used by Liu Chih.”Central Asiatic Journal 26(1/2):78–104.
Liang, Jingyu. 2006. Niujie: Yige Chengshi Huizu Shequ de Bianqian [Ox Street:
Changes of an Urban Hui Community]. Beijing: Minzu University of China
Press.
Liang, Xiangmin. 2004. Liuzhi Jiqi Yisilan Sixiang Yanjiu [Studies on Liu Zhi and
His Islamic Thoughts]. Lanzhou: Lanzohu University Press.
Lipman, Jonathan. 1997. Familiar Strangers: A History of Muslims in Northwest
China. Seattle: University of Washington Press.
Lipman, Jonathan. 2004. “White Hats, Oil Cakes, and Common Blood: The Hui
in the Contemporary Chinese State.”In Governing China’s Multiethnic
Frontiers, edited by Morris Rossabi, 19–52. Seattle: University of Washington
Press.
Lipman, Jonathan. 2016. Islamic Thought in China. Sino-Muslim Intellectual
Evolution from the 17th to the 21st Century. Edinburgh: Edinburgh University
Press.
Ma, Qiang. 2006. Liudong de Jingshen Shequ: Renleixue Shiye Xia de Guangzhou
Musilin Zhemati Yanjiu [Mobile Spiritual Communities: An Anthropological
Study on the Guangzhou Jama’at]. Beijing: Chinese Social Science Press.
Ma, Qiang. 2011. Huifang Neiwai: Chengshi Xiandaihua Jincheng Zhong de Xi’an
Yisilanjiao Yanjiu [in and out of Hui Neighborhoods: Urban Xi’an Islam in the
Modernization]. Beijing: Chinese Social Science Press.
Ma, Tong. 1983. Zhongguo Yisilan Jiaopai yu Menhuan Zhidu Shilue [a History of
Muslim Factions and the Menhuan System in China]. Yinchuan: Ningxia
People’s Press.
Ma, Tong. 2000. Zhongguo Yisilan Jiaopai yu Menhuan Zhidu Shilue (Xiuding
Ban) [a History of Muslim Factions and the Menhuan System in China (Revised
Edition)]. Yinchuan: Ningxia People’s Press.
Ma, Xuefeng. 2013. Cong Jiaomen Dao Minzu: Xinan Biandi Yige Shaoshu Mnzu
Shequn de Minzu Shi [from Religion to Ethnicity: History of a Minority
Community in China’s Southwestern Borderland]. Beijing: Social Science
Literature Press.
Ma, Yan. 2012. Yige Xinyang Qunti de Yimin Shijian: Yiwu Musilin Shehui
Shenghuo de Minzuzhi [Immigrant Experiences of a Faith Community–an
Ethnography of Muslims in Yiwu]. Beijing: Minzu University of China Press.
Min, Junqing. 2011. Zhongjianren: Liudong yu Jiaohuan: Lintan Huishang Yanjiu
[Middleman–Mobility and Exchange: A Study on Hui Businessmen from Lintan].
Beijing: Minzu University of China Press.
Min, Shengguang. 1995. “Lishi Shang Yisilan Wuma Dui Xidaotang de Yingxiang
[the Influences of Islamic History of Ummah on Xidaotang].”Shijie Zongjiao
Yanjiu [World Religion Studies] 1:1–7.
Min, Shengguang. 2010. Zhongguo Yisilanjiao Xidaotang Wenji [Research on the
Chinese Xidaotang: An Anthology]. Lanzhou: Gansu Minzu Press.
30 CHINESE SOCIOLOGICAL REVIEW
Min, Shengguang. 2017. Xifeng Lungao [Xifeng Research Anthology]. Beijing:
Religion Culture Press.
Mu, Zheng, and Qing Lai. 2016. “Micro-Macro Interactions in Ethno-Religious
Homogamy among Hui Muslims in Contemporary China: The Roles of
Residential Concentration and Aging.”Journal of Muslim Minority Affairs
36(1):88–105. doi:10.1080/13602004.2016.1153823.
Murata, Sachiko, William C. Chittick, and Wei-Ming Tu. 2009. The Sage Learning
of Liu Zhi: Islamic Thought in Confucian Terms. MA: Harvard Univ Council on
East Asian.
Murata, Sachiko. 2000. Chinese Gleams of Sufi Light: Wang Tai-Yu’s Great
Learning of the Pure and Real and Liu Chih’s Displaying the Concealment of the
Real Realm. With a New Translation of Jami’s Lawa’ih from the Persian by
William C. Chittick. NY: SUNY Press.
Murata, Sachiko. 2017. The First Islamic Classic in Chinese: Wang Daiyu’s Real
Commentary on the True Teaching. NY: SUNY Press.
Petersen, Kristian. 2017. Interpreting Islam in China: Pilgrimage, Scripture, and
Language in the Han Kitab. UK: Oxford University Press.
State Council, Chinese. 2018. Zhongguo Baozhang Zongjiao Xinyang Ziyou de
Zhengce he Shijian [Policies and Practices to Guarantee Religious Freedom in
China]. Beijing: People’s Press.
Stewart, Alexander Blair. 2017. Chinese Muslims and the Global Ummah. London
& NY: Routledge, Taylor & Francis Group.
Stroup, David R. 2017. “Boundaries of Belief: Religious Practices and the
Construction of Ethnic Identity in Hui Muslim Communities.”Ethnic and
Racial Studies 40(6):988–1006. doi:10.1080/01419870.2016.1193211.
Sun, Zhenyu. 2011. Wang Daiyu Liu Zhi Pingzhuan [a Critical Biography of Wang
Daiyu and Liu Zhi]. Nanjing: Nanjing University Press.
Tang, Wenfang, Yue Hu, and Shuai Jin. 2016. “Affirmative Inaction: Education,
Language Proficiency, and Socioeconomic Attainment among China’s Uyghur
Minority.”Chinese Sociological Review 48(4):346–66. doi:10.1080/21620555.
2016.1202753.
Tontini, Roberta. 2016. Muslim Sanzijing: Shifts and Continuities in the Definition
of Islam in China. Leiden: Brill.
Yang, Xiaochun. 2011. Sanzaju Huizu Jingji yu Huihan Minzu Guanxi Yanjiu: Yi
Shandongsheng Zaozhuangshi Taierzhuangqu Weili [the Economy of Scattered
Hui Minorities and Han-Hui Relations: The Case of Taierzhuang, Zaozhuang,
Shandong]. Beijing: Minzu University of China Press.
Yu, Zhengui. 1996. Zhongguo Lidai Zhengquan yu Yisilanjiao [Chinese Historical
Regimes and Islam]. Beijing: Xinhua Press.
Zang, Xiaowei. 2007. Ethnicity and Urban Life in China: A Comparative Study of
Hui Muslims and Han Chinese. London & NY: Routledge.
Zang, Xiaowei. 2012. “Uyghur Islamic Piety in Urumchi, Xinjiang.”Chinese
Sociological Review 44(4):82–100. doi:10.2753/CSA2162-0555440404.
Zhang, Shihai. 2010. “Minguo shiqi Xidaotang shangye fazhan yanjiu [Xidaotang’s
business development during the Republic of China].”In Zhongguo Jinxiandai
Hangye Wenhua Yanjiu [Studies on Occupational Culture in Modern China],
edited by Christian Lamouroux, 209–17. Beijing: National Library Press.
Zhou, Chuanbin. 1997. “Xidaotang Moshi yu Huizu Chuantong Zhengzhi Wenhua
[Xidaotang Model and the Traditional Hui Political Culture].”Ningxia Shehui
Kexue [Ningxia Social Science] 2:71–6.
Zhou, Xiefan. 2002. “Sufei Zhuyi yu Ming Qing Zhiji de Zhongguo Yisilanjiao
[Sufism and China’s Islam in Ming and Qing].”Xibei Di Er Minzu Xueyuan
Yuanbao [Journal of No. 2 Northwest Minzu College] 1:23–34.
SUMMER 2019 31
Zhou, Xiefan. 2006. “Qingzhen Dadian Qianyan [Foreword to Qingzhen
Collection].”Shijie Zongjiao Yanjiu [World Religion Studies ] 2006(2):147–8.
About the Authors
Qing Lai (qlai@fiu.edu) currently teaches at Florida International University
as an assistant professor of sociology. He graduated with a PhD in sociology
from the University of Michigan. He received graduate training at the
Population Studies Center and Survey Research Center, Institute for Social
Research. His research interests include Chinese Muslims, development, dem-
ography, and social stratification. He has published peer-reviewed articles in
Social Science Research, Journal of Muslim Minority Affairs, Chinese
Sociological Review, Chinese Journal of Sociology, Research in the Sociology
of Work, Natural Hazards, and Proceedings of the National Academy of
Sciences of the United States of America (PNAS). His current research
focuses on the Muslim population in China. He analyzes census and survey
data to examine Chinese Muslims’population heterogeneity and their rela-
tions with the Han majority, the Chinese State, and the Islamic world. He is
collaborating with multiple Chinese institutions to collect survey data on
Chinese Muslims.
32 CHINESE SOCIOLOGICAL REVIEW
