STELES AND STATUS: EVIDENCE FOR THE EMERGENCE OF A NEW ELITE IN YUAN NORTH CHINA

Abstract

Drawing on stele inscriptions in a Yuan-period ancestral graveyard, this article aims to shed light on the emergence and evolution of a Chinese office-holding family in North China under Mongol rule (thirteenth- to fourteenth century). Tracing the family's connections with Mongols, it argues that adaptation to the Mongolian patronage system was essential to obtaining and maintaining political status during the Yuan, and that the kin group was stratified with the patronized descent line monopolizing political privilege. In doing so, the article highlights the value of stone inscriptions in clarifying official status, patronage, and inheritance rights in North China during the Yuan period.

In 1299, Sun Gongliang 孫公亮 (1221–1300), who had once served the Mongols (who ruled in China from 1234 to 1368) as a Pacification Commissioner, installed a two-and-a-half-meter-tall massive granite stele at the center of the ancestral graveyard of his family in Hunyuan 渾源 county, northern Shanxi. Over the course of the next three decades, his descendants erected at least twelve more steles in the graveyard, most of which still exist in good shape today. In this paper, drawing on the epigraphy of this extraordinarily well-preserved, extensive Yuan-period family graveyard, I explore a Yuan northern Chinese family's strategy to perpetuate their political status in the court of the Great Khans, and to maintain their family's solidarity in their home region. My larger goal is to seek to answer a central question concerning Middle Period Chinese history: How different was the impact of the Mongol conquest on elites in the north and the south?

Pioneered by the monumental work of Robert Hymes, ¹ recent scholarship has elucidated that southern Chinese society under the Southern Song dynasty (1127–1279) and subsequent Mongol Yuan rule (1276–1368) saw the spread of local literati as a dominant social group in the majority of regions south of the Huai River. Fully aware of their predominance, during the Southern Song the local literati participated in local governance both on their own and by official request. In so doing, they enjoyed mostly harmonious relations with local administrators, who generally lacked adequate financial and human resources. Even after the Yuan conquest of the south in 1279, after which the civil service examination was suspended until 1313, local literati families continued to exercise considerable power in local government. ² After the reinstatement of the examination as the main avenue to officialdom by the Ming (1368–1644), local literati by and large consolidated their place in local society in China proper, although their patterns of dominance differed significantly in different regions. ³ In palpable contrast with the overall continuity of the southern literati elites, the Mongol conquest exerted a far more profound and lasting impact over society in the north. During the prolonged and devastating Jurchen-Mongol war (1211–34) that ravaged almost all of the north, the power of the former Jin literati families was so devastated that they almost entirely disappeared from contemporary records. Instead, a new group of office-holding families swiftly adapted to the new Mongol recruitment systems, and the resumption of the examination system seems not to have prompted their descendants to pursue success through the examinations as the Jin local elites had. ⁴ Meanwhile, in the south, as Richard Davis has skillfully demonstrated in the case of Shi clan in Mingzhou 明州 (modern Ningbo, Zhejiang), the political success of literati elites was facilitated by the proper management of kinship organization, and vice versa. Once a member obtained an official post through the examinations, literati elites and their families in Southern Song attempted to gain a firm foothold in officialdom for generations, not only by investing in the education of kinsmen, but by setting up an extensive network of contacts and affinal kin among prominent officials. ⁵ In local society, such families also expanded their sphere of influence by establishing local marriage ties with fellow elite families, while strengthening their own kinship solidarity. ⁶

With the system of recruitment being quite alien to the Song examination system (as will be discussed below), it can be assumed that the evolution of literati society under Mongol rule in the north would not follow the pattern of local literati evolution in the south under the Southern Song. Yet the scarcity of contemporary historical sources has frustrated the analysis of northern Chinese political elites and their kinship organization, until the recent uncovering of new epigraphic sources turned the tide. The steles in the Sun family graveyard, mostly unknown to academia until recently, provide the most extensive extant record on the topic, covering from 1190 to 1324 and fully illuminating the untold evolution of northern office-holding families during and after the turbulent Jin-Yuan transition. What makes the case of the Suns especially advantageous is the 1299 stele, falling into an epigraphic genre commonly called a “Stele of the Ancestral Graveyard” (xianying bei 先塋碑), which focused on recording genealogical information. More than three decades of fieldwork in North China since the 1980s has shed informative light on the mostly forgotten practice of northern families during the Jin and especially the Yuan periods of engraving their genealogy on steles, as opposed to writing them on paper as was done in the south. ⁷ This “rediscovery” of the Stele of the Ancestral Graveyard, which declined as an epigraphic genre after the fall of the Yuan, inevitably triggers the further question: How was kinship imagined and described in the inscription? The steles in the graveyard of the Sun family collectively provide insight into this question, as well as illuminating the evolution of northern elite families.

When it comes to achieving political success and transmitting it over generations, the Suns of Hunyuan were the northern equivalent of the Shis of Mingzhou. Under Mongol rule, the Suns produced four officials above rank 3 over at least four generations, including Sun Gongliang, the Pacification Commissioner himself (the Shis also produced officials for four generations, but did a little better, with three chief councilors). Considering together the Sun family's endeavors in office and local society, I will demonstrate the malleable strategies by which the Suns adapted to the non-Chinese ruling system, while also showing how the same system stimulated the stratification of Sun kinfolk, dividing them into two groups in accordance with their proximity to the Mongol ruling stratum. The case of the Suns will underline the path to success in the Yuan officialdom and how aspirants strove to achieve their goals, using methods that would have looked quite bizarre in the eyes of either the preceding Southern Song or the subsequent Ming-Qing political elites. To begin, I first discuss the significance of the 1299 stele and the political background of its installation.

RISE OF NEW EPIGRAPHIC PRACTICE AND AN ELITE STRATUM IN THE YUAN NORTH

As is well known, during the same period that Sun Gongliang installed the stele, descent groups in the south began investing in weaving centrifugal collateral descent lines into cohesive lineages by compiling genealogical texts and establishing ancestral halls. ⁸ In the north, by contrast, only ten genealogical texts are known to have been compiled. ⁹ In discussing the emergence of genealogical texts in the social and economic context of the south, the dominant narrative appears to attribute the lack of genealogical texts in the north to the massive southward population outflow from the region during the Song-Jin transition and the subsequent economic stagnation of the north. ¹⁰ There is no denying that economic development was one of the factors that enabled southern descent groups to strengthen their organization. Yet no empirical research corroborates the impact of economic stagnation on genealogy compilation in the north. For that matter, we do not even know if the northern economy did in fact fall into stagnation. Rather, the deficiency in textual sources itself may be a key to better understanding the formation of the northern descent groups during the Jin-Yuan, as genealogies were not necessarily always written down on paper.

During the Jin-Yuan (1115–1368) period, although individual steles could differ greatly in size, title, and sometimes content, the emergent genre of “Steles of Ancestral Graveyards” as a whole pursued the same goal: recording family genealogy. Such steles commented on or commemorated matters of ancestry, ancestral virtue, and family history, and were in general given titles such as “Stele/Epitaph for the Ancestral Graveyard” (xianingbei 先塋碑, qianbiao 仟表), “Stele Recording Ancestral Virtue” (xiandebei 先德碑, shidebei 世德碑), “Stele Clarifying Ancestry” (zhaoxianbei 昭先碑), “Stele Clarifying Ancestral Virtue” (zhaodebei 昭德碑), or “Tombstone for Family X” (X shi mubiao X 氏墓表). While the total number of extant examples is 24 before the end of the Jin (prior to 1234), it spikes up to 252 during the Yuan, especially after the 1260s. ¹¹ This seemingly rather abrupt emergence of a new epigraphic genre is the most remarkable feature of the extant Yuan stele inscriptions. Without exception, the steles were installed in ancestral graveyards and, quite notably, many had a genealogical chart engraved on the verso, as did the 1299 stele of the Suns. Steles of almost identical style were also erected, albeit with different titles, when an ancestral graveyard was established or relocated. In these cases, the steles were entitled “Stele Commemorating the Establishment/ Relocation of the Ancestral Graveyard” (xinying bei 新塋碑 or qianying bei 遷塋碑). To circumvent these unwieldy phrases, hereafter, I refer to the genre simply as “genealogical steles.” In the south, the rise of ancestral halls and genealogical texts gradually eclipsed the significance of graveyards in ancestral worship and the formation of descent groups. ¹² Highlighting this disparity more clearly, this particular genre of stele was widespread only within the former territory of the Jin dynasty, and only a handful of cases can be found scattered across South China, mostly erected by families that migrated from the north after the fall of the Southern Song. ¹³ The genealogical steles provide us with a different approach to the elusive questions concerning the formation of descent groups in the north.

The rise of a new way to narrate and record ancestry and genealogy seems to have corresponded with the enormous social, political, and cultural transitions brought about by the Mongol conquest of North China and its accompanying devastation. As previous conquerors like the Khitans and Jurchens had done, the Mongols amalgamated their indigenous ruling institutions to the Chinese traditional administrative system. The new ruling elites who arose amid the ruin of the Jurchen Jin dynasty differed substantially from the literati elites who had almost monopolized official posts in the preceding two centuries. In stark contrast to the way Song literati elites acquired posts through the highly institutionalized civil service examination system, under Mongol rule those who aspired to higher status took rather diverse routes, such as becoming clerks, joining the army, and, most importantly, establishing personal connections with those in power. Mongol princes and the members of the imperial house, including the khan himself, maintained groups of retainers (called kesig in Middle Mongolian; the term also meant a member of a kesig, while its plural form kesigten pointed to collective members of a kesig; the term was transliterated into Chinese as qiexue 怯薛) and appointed them to posts under their control, bypassing any institutionalized recruitment and promotion systems. The examination system remained only a minor route in terms of the number of candidates it sent into officialdom throughout the Yuan. Furthermore, once a person occupied a position in a kesig, his descendants were usually expected to inherit the prestigious status over the generations as long as the bloodline of the prince persisted, or unless that of the retainer fell into disgrace. The spread of genealogical steles precisely coincided with the establishment of the new elite stratum, most of whom showed off in stele inscriptions their success in acquiring connections with Mongols during the Jin-Yuan transition, and their status as office-holding households. ¹⁴

The coincidence of new elites and the new genre becomes particularly evident when the Yuan steles are compared with the limited number of similar steles erected under Northern Song and Jurchen rule. In fact, engraving genealogies on steles was originally a custom practiced mainly among commoners and low-ranking officials, but rapidly spread as an elite custom after the rise of new local elites in connection with the Mongols. ¹⁵ Proving descent from an ancestor who had served the Mongol Empire (especially the early khans) and contributed to its success became important under the Yuan, which institutionalized the traditional Mongolian notion called genjiao in Chinese. ¹⁶ This tradition can be roughly interpreted in English as the subordinate affiliation with a master's household (usually a Mongolian princely or imperial bloodline) over generations, and became one of the decisive factors in hunting for an official post, entering a kesig, or maintaining political status, as discussed below.

OFFICIAL CAREERS OF THE SUNS AND THEIR GENJIAO BACKGROUND

With regard to political careers and social background, the Suns of Hunyuan were one of the most successful among the emergent Yuan northern office-holding families. Before Sun Wei 孫威 (1183–1240) distinguished himself as an armorer in the service of Činggis Qan (r. 1206–27), the steles installed by the family, both those extant in the graveyard and already lost but recorded in the Yuan-Ming-Qing literary collections and local gazetteers, tell almost nothing but the name of several patriarchs, all of whom were peasants. During the Zhenyou 貞祐 period (1213–16), the Jin was facing Mongolian aggression in the northern Shanxi region. Sun Wei, wanting to help defend the region, started by joining the Jin army at Xijing 西京 (modern Datong). Sometime in 1215, the general of the garrison troop ordered him to travel from Datong all the way to Hebei, where the fiercest battles of the initial stage of the Mongol invasion were being fought. At the risk of his life, Sun Wei was to ask the Jin emperor Xuanzong 宣宗 (r. 1213–24) whether the garrisons should defend their city, as the emperor had recently abandoned the capital Yanjing and fled to Kaifeng. After successfully completing this mission, he was made centurion in the garrison troop. Yet, after the Mongols captured the city, sometime before 1218, Sun Wei somehow switched sides. In the biography of his son, Sun Gongliang, he abruptly appears with the rank of battalion commander in the Righteous Army (yijun qianhu 義軍千戶) in the region newly conquered by Mongols. ¹⁷ A “righteous army” usually meant an army organized locally without dynastic intervention, and this implies Sun Wei joined the Mongol army with the soldiers formerly under his command in the Jin army or newly recruited in the region.

In any case, as Sun Wei had learned armor smithy from his brother-in-law, he gradually emerged as the leader of a group of armorers in North China and Mongolia in the service of the advancing Mongol army. One day, he happened to seize an opportunity to present an extremely well-crafted set of armor to Činggis Qan himself. As a reward for his outstanding service, Činggis Qan granted him the Mongolian title yeke uran (transliterated in Chinese as yeke wulan 也可兀蘭), meaning fine or great artisan, ¹⁸ with permission to let his descendants serve in the imperial court as head armorers. ¹⁹ After the death of Činggis Qan, Sun Wei somehow grabbed the chance to submit a set of armor to the next khan, Ögödei Qaγan (r. 1229–41), who shot an arrow at it and found the arrow did not penetrate it at all. Impressed by Sun Wei's outstanding craftsmanship, Ögödei Qaγan granted him a golden tablet (jinfu 金符) that assured Sun Wei autonomy over appointments of armor smiths, and perhaps also certain judicial authority over them as well. ²⁰ During the reign of Ögödei Qaγan, Sun Wei and his son, Gongliang, continued to assemble armorers, mainly from among prisoners of war in North China. Eventually they came to preside over arsenals in Shuntian 順天, Anping 安平 (both in modern Hebei), Huaizhou 懷州, and Henan (which at the time literally meant “south of the Yellow River”). Sun Wei and Sun Gongliang privately covered the living expenses of the armorers until the official stipend system was established in the 1260s. At this point, it is highly likely that Sun Wei belonged to the kesig of a Mongolian prince, perhaps that of Tolui (1192–1232), the fourth and youngest legitimate son of Činggis Qan.

Sun Gongliang, as Sun Wei's eldest son, was born and grew up in the “northern steppes” (mobei 漠北) among the ruling elites of the Mongol Empire. ²¹ Having spent his childhood in the itinerant imperial court, as did other hereditary subjects, he inherited his father's position upon Sun Wei's death in 1240. Gongliang spent the two following decades as the leader of armorers in the direct service of successive khans. He presented armor every year, and as a reward for his service, Möngke Qaγan (r. 1251–59) ordered that he inherit the title of his father, yeke uran. As genjiao persisted over generations, Gongliang and his kinsmen were extremely fortunate to serve the direct descendants of Činggis Qan, the most powerful and honored figures in the Mongol Empire. Among the imperial descent lines, as their career demonstrates, the Suns seem to have served exclusively the descendants of Tolui. Before his ascension to the imperial throne, Qubilai Qaγan (r. 1260–94), the fourth son of Tolui, had once directed Sun Gongliang to supply one hundred pieces of armor to his princely office. This connection with the future khan would further accelerate the prosperity of the Suns. In 1261, in the second year of Qubilai's reign, Sun Gongliang was promoted to investigating censor (jianchayushi 監察御史) after presenting sixty pieces of armor to Qubilai at the battle of Šimültü Naγur, the decisive battle fought in modern southern Inner Mongolia during the civil war between Qubilai and his younger brother Arïq Buka (Chin. Alibuge 阿里不哥, d. 1266). In the Mongol Yuan bureaucracy, it was quite common for a person to be appointed to a post completely irrelevant whatever role he had in his kesig, if the master found his talent suitable for the mission. Having proved himself as an astute investigating censor, Gongliang advanced rapidly in regular administration, allegedly with the backing of the ruling khan himself.

In the meantime, the Suns never abandoned the duty assigned by Činggis Qan. Sun Wei had been made governor-in-chief of armorers in Shuntian, Anping, Huaizhou, Henan, and Pingyang circuits (Shuntian Anping Huaizhou Henan Pingyang denglu jiajiang du zongguan 順天安平壞州河南平陽等路甲匠都總管). The post became hereditary (shijue 世爵), ²² and would be inherited by major male descendants of Sun Wei. Sun Gongliang's eldest son, Sun Gong 孫拱 (1241–1306), upon reaching his majority, was assigned to be the lead armorer in Shuntian and Hejian 河間 circuits, the regions included on jurisdiction of the Suns. Then as his father had done, he presented armor to Qubilai, during the siege of Xianghan 襄漢 during the early 1270s, to have his claim to be an heir approved. At the time, Sun Gong was superintendent of armorers in Baoding 保定circuit (the former Shuntian circuit). Then, almost following his father's path, Sun Gong also advanced in regular administration by imperial appointment. After Sun Gongliang resigned the post of pacification commissioner in Zhexi circuit and minister of the Branch Ministry of Works (Zhexidao xuanweishi jian xinggongbu shangshu 浙西道宣慰使兼行工部尚書), Sun Gong held ranks such as vice prefect of Baoding circuit (Baodinglu zhizhong 保定路治中), governor of armorers in Dadu circuit (Dadulu junqi renjiang zongguan 大都路人匠總管), and vice minister of works (gongbu shilang 工部侍郎), all of which were related to supervising artisans. After the death of Qubilai, Sun Gong hurriedly obtained an imperial audience with the new khan, Temür Qaγan (r. 1294–1307), and reconfirmed the imperial patronage after being introduced with the declaration “This is the grandson of yeke uran.” ²³ Sun Gong's younger brother Qing 擏 (1249–96) entered the kesig, presumably that of Qubilai or his short-lived heir, Činkim (Chin. Zhenjin 真金, 1243–86), as his father allegedly had, at age ten or so. Sun Qing was later appointed supervisor-in-chief of paper money in the various circuits (zhulu jiaochao doutiju 諸路交鈔都提舉). Similarly, in 1287, Sun Qian 孫謙 (1255–98), the eldest son of Sun Gong, was accepted to enter the kesig of Činkim, and later presented armor to Qubilai, just as his father and grandfather had, during the Rebellion of Nayan (1287–88). At the time, Sun Qian was supervisor of armor smiths in Baoding circuit (Baoding dengchu jiajiang tiju 保定等處甲匠提舉), a post his father had held. After the death of Sun Gong, the eldest male descendant of the direct bloodline of Sun Wei in each succeeding generation obtained official posts mostly relating to the armor smithy. Typically, they acquired a post in regular administration after conceding the post of supervisor to their eldest son. In the last generation recorded on the steles, Sun Xie 孫諧 (dates unknown), the son of Sun Qing and the family patriarch after the premature death of Sun Qian, was vice commissioner of pacification in the Hedong and Shanxi circuits (Hedong Shanxi xuanwei fushi 河東山西宣慰副使), while his younger brother Yi 誼 (dates unknown) was supervising armorers affiliated with the central court.

As most of the Suns started their service in the khan's (or imperial crown prince's) kesig, they naturally acquired near-native fluency in Mongolian. Sun Gongliang “had thorough knowledge of ceremony and etiquette in the imperial court” and was also famous for “cracking timely jokes in Mongolian while translating [Chinese into Mongolian and vice versa].” ²⁴ Sun Gong also spoke Mongolian without any problem. ²⁵ Sun Qing was also renowned for his Mongolian language skill, both speaking and writing. ²⁶ At least two kinsmen are known to have had Mongolian names. Sun Gong was called Dei Temür (Chin. Taitiemu'er 太帖穆而) by Qubilai ²⁷ and Sun Qian called himself Bei Buqa (Chin. Bobuhua 伯不華) in Mongolian. ²⁸ As seen below, some of their descendants appear with Mongolian names in the genealogical chart of the Suns. Hsiao Ch'i-ch'ing has persuasively argued that non-Chinese persons could attain Chinese literati identity through learning the classics, while retaining their original “ethnic” identity, during the Yuan. ²⁹ Similarly, Michael Brose explicitly illustrates, in the case of Uyghur official families in Yuan China, a non-Han Chinese person could develop identity as a literatus while maintaining and emphasizing his non-Han origin. Indeed, pursuing Chinese high culture—including taking the civil service examinations, compiling genealogies, and conducting ancestral worship—was a part of their strategy to acquire legitimate status among the new ruling elites in an empire that comprised multiple cultures. ³⁰ The case of the Suns demonstrates that the opposite was also true. In fact, contemporary cases suggest that to acquire the patronage necessary to enter and advance in the Yuan government, Chinese subjects to a great extent needed to assimilate themselves to Mongol political tradition and culture. ³¹ Having themselves adapted to the political structure and tradition of the Mongol Empire, the Suns exemplify one way to maintain and further promote one's political status in the near-highest ruling stratum of the empire at the time.

The official careers of the kinsmen of the Suns also clearly demonstrate that genjiao, the trans-generational connection with a master, was the key to their office-holding. Although it needed to be renewed at the passing of each patriarch in the subject's and master's households through the submission of armor to the ruling khans, genjiao in general ensured the subject's descent line stable paths to officialdom for generations. In the case of the Suns, having a genjiao with the imperial bloodline granted them steady official status under the Mongol Empire. To retain this privileged treatment, they did everything required: passing down the skill of smithing armor, presenting armor to khans at key political turning points, and sending their eldest sons to the imperial kesig. The steles installed by the Suns also reveal that they sought to record their faithful service from the beginning of Mongol rule and display it to their descendants and other visitors to their graveyard.

How did the Sun family manage their kinship ties in their home county while remaining in the upper stratum of the Yuan ruling elite? Did they attempt to exert influence over the locale through marriage ties the way their southern counterparts did? To answer these questions requires shifting our attention from the court to the graveyard in northern Shanxi, where the Suns had lived for generations.

THE GRAVEYARD IN HUNYUAN

The Suns’ home county, Hunyuan, is located approximately 30 kilometers southeast of Datong, the regional capital of northern Shanxi. The tombs are collectively named Sun Gongliang jiazu fenmu 孫公亮家族墳墓 ³² by the State Bureau of Cultural Relics, after the highest ranking member of the descent group during the Mongol period. Unfortunately, we cannot even speculate on the original size of the graveyard as the whole area has become farmland. The extant steles are scattered about most extensively in a rectangular area in the southeasternmost part of the plain, which stretches for about 170 meters east to west, and 70 meters south to north.

According to the report of a field survey conducted by a group of northern Shanxi archaeologists, there were eleven steles in the graveyard in 1987. ³³ In my first visit to the site in March 2008, I found twelve steles, including a fragment of a stele that seemed to have been recently dug from the ground. Besides the steles, there are twelve stone animals and eight stone statues of civil and military officials, all lying on the ground. Although the site is called “family tombs” (jiazu fenmu), no burial mound can be identified on the flat field. Today, about half of the populace in the village are surnamed Sun and claim ancestry tracing back to the Suns in the Yuan period. In my interviews, elderly inhabitants of the village testified that initially there were more steles than now, until an irrigation project in the 1970s used the steles for building stones. Their testimony helps explain the fact that the empty pedestals (beizuo 碑座) in the graveyard outnumber the existing twelve steles. Three out of twelve steles have completely fallen on their front sides, leaving no clue as to their titles and contents as none of the back sides of the three steles have any script. Table 1, “Steles in the Sun Family Graveyard,” provides information on the nine legible steles in chronological order. All the steles seem to have been hewed out from fine plutonic rocks (most likely granites).

Table 1. Steles in the Sun Family Graveyard

The report of the survey in 1987 lists two other steles: “Yuan gu chaoliedafu Hedong Shanxilu xuanweifushi Sungong mu” 元故朝列大夫河東山西路宣慰副使孫公墓 (dated 1335, 145 cm tall, 110 cm wide) and “Hanlinxueshichengzhi rongludafu zhizhigao jian xiuguo dayuan shenchuanqun gu shanshi………..guanzhuan zhenggong zhimu” 翰林學士承旨榮祿大夫知制誥兼修國大元神川郡故善士□□□□ ^† 貫篆証公之墓 (dated 1329, 155 cm tall, 93 cm wide, with a pedestal). ³⁴ The size of the existing three fallen steles indicates that two of them are likely the missing two steles. While the steles of “Sun Gongliang jiazu fenmu” were completely neglected in the successive Ming-Qing local gazetteers of the Hunyuan county and Shanxi province, several recent scholars have published transcriptions of the steles. ³⁵

In terms of number of extant steles, this is undoubtedly one of the best-preserved family graveyards from the Yuan in the north. ³⁶ As was typical in the north, it is a concentration of tombs of kinsfolk, called “kinship burial” (zuzang 族葬), in which the pattern of burial mounds was to be arranged in accordance with genealogical order (zhaomu 昭穆). Unfortunately, we have no way to figure out the general plan for the graveyard, since it has been disturbed at least once. However, as discussed below, the current placement of the steles strongly suggests it was established based on a certain relational or generational order.

The uniqueness of the graveyard is sharply contrasted with the lack of any genealogical information in the adjacent villages, including Xiliucun. When I asked those surnamed Sun in the village to substantiate their ancestry, everyone referred to the genealogical stele (Stele 1 in the Table) in the graveyard. In contrast with the south (and quite commonly in the north), as long as elderly persons can remember, there has not been a genealogical text or an ancestral hall in the village. The oldest building in the village is a small village theater (xitai 戲臺) with a commemorative stele dated 1886, located near the western entrance of the village, where the villagers used to stage dramas during the Qingming festival. The graveyard seems to have been one of the most, if not the most, prominent cores in the religious and social domains of the village, well into the Republican period. ³⁷ After all, the graveyard used to, and still does, carry considerable significance in the life of villagers in Xiliucun. Yet, first and foremost, for those who pay close attention to the stele inscriptions, it still functions, precisely as planned in the Yuan period, to display the family genealogy of the Sun family, whose members acquired extremely prestigious positions in the Mongol Empire some seven centuries ago. And the genealogical information helps us explore how a Chinese office-holding household in close to the uppermost tier of the Yuan bureaucracy defined kinship and attempted to manage it.

DEFINING KINSHIP IN GENEALOGICAL STELES

Just like other Yuan genealogical steles, ³⁸ Stele 1 has a genealogical chart engraved on its verso, entitled “Genealogy of the Sun Clan” (Sunshi zongzu shipu 孫氏宗族世譜), which clearly tells of the purpose of the erection (see Figure 1). The chart elaborately records the names and official titles of kin, and even notes when the whereabouts of someone who disappeared during the Jin-Yuan transition were unknown, for instance: “his whereabouts, after he moved to Luzhou, is unknown” (遷潞州不知所在). Also, the chart meticulously includes the name of all the spouses of the kinsmen as of 1299, going much further than southern genealogical texts of the period, which usually excluded female kin. Of 183 names in seven generations, given the advanced age of Sun Gongliang, most of the kin from the first to his (fourth) generation had most certainly passed away. That would mean that about 120 Suns would have celebrated the 1299 erection of the stele, which falls in the average range of the size of descent groups recorded in Yuan genealogical steles. The number of family members, in cases where the stele refers to it, usually never exceeds one hundred (yiqian zhi 一千指, or yibai kou 一百口) by much. In several exceptional cases, the inscription counts affinal relatives as family members to swell the size of the group, such as, “Now we have grandsons and great-grandsons of affinal relatives, numbering a hundred and several dozens. We must do our share to record them, for [otherwise] later when some of them become prominent, no one will be able to trace them” (今女族有孫，其孫曾乃至百數十人。務所紀，後有興者，固無以究). ³⁹ In this case, assuming the parents of the grandchildren and great grandchildren were mostly alive, the number of family members likely exceeded two hundred. Compared to the post-Song-Yuan southern lineages, the relatively small scale of the Suns, together with that of other Yuan northern descent groups, can be attributed to the “shallowness” of their generational depth, averaging four to five generations as recorded in the Jin-Yuan genealogical steles. ⁴⁰ It must be noted that as in the case of the Suns, except for oral history, individual biographical records (zhuang 狀), and funerary steles (if any), most of the Yuan installers of genealogical steles allegedly did not have any form of written genealogical account to work from, largely because of the social upheavals during the Song-Jin and Jin-Yuan transition. ⁴¹ Contemporary inscriptions frequently state this explicitly, as in “the ancestral genealogy has been lost” (shipu wangsang世譜亡喪). As the sole genealogical record of a descent group, a genealogical stele's narration typically begins with an apical ancestor who established the graveyard and traces his descendants and their spouses who had been or would be buried therein.

Figure 1 “Genealogy of the Sun Clan” in Hunyuan county, Shanxi

At the same time, perhaps an equally important possible cause making the northern descent groups relatively small was the consistent influence of the notion of “nine generations” (jiuzu 九族). This traditional notion delineates a descent group consisting of four generations preceding ego and another four generations following ego. In some cases, genealogical stele inscriptions refer to this ancient notion to define kinship. One of them, “Jining lishi zuying bei” 濟寧李氏祖塋碑 (dated 1195), emphasizes, “In the reigns of the great emperors under whom people prospered, it was fitting to conduct proper birth and funerary rituals, respect ancestry, and teach the harmonization of the nine generations of kin” (而况遭際聖時, 人人漸沐, 生事死葬, 尊祖報本, 與夫敦睦九族之教). ⁴² One and a half centuries later, “Miaoshi zongpai zhi tu” 苗氏宗派之圖 (dated 1343), also specifies the criteria according to which ancestors were included on the stele saying: “[Offering sacrifices based on] the system of five degrees of mourning circle across nine generations has been an obligation since ancient times” (以五服九族立法之制, 乃古今之通義也). ⁴³ The clear tendency to include ancestors only for three to four generations in genealogical stele inscriptions and charts implies that the installers had no intention of expanding the sphere of kinship to relatives who had a common ancestor more than four generations back, which is what their southern counterparts would attempt to achieve in their genealogical practices. ⁴⁴ Even in the case of the Wangs in Pucheng 蒲城, Shaanxi, who were commended by the central court around 1328 for living together for six generations, the genealogical stele inscription, “Pucheng yimen wangshi xianying beiming 蒲城王氏先塋碑銘” (dated 1355), almost completely omits their ancestral history before the Yuanguang 元光 period (1222–23), only briefly remarking, “Prior to that, since the end of the Five Dynasties, our ancestors had lived together for five generations” (其先自五季末，已五世同居). ⁴⁵ Referring to Patricia Ebrey's pathbreaking work on descent groups, we can find a strong “jia 家 orientation” that focused on “tangible material and human entities” (family members and ancestral graveyards) and is conceived “overwhelmingly in unitary, lineal terms,” rather than in terms of the “zong 宗 orientation” that paid much attention to the hierarchical harmony of kinsfolk within mourning circles. ⁴⁶ Meanwhile, as Ebrey has pointed out, the former is a quite archaic orientation of kinship formation, which can be traced back at least to the Han period. ⁴⁷ In the case of the Suns, we find considerable longevity of the orientation in the Jin-Yuan north, in spite of repeated cataclysms after the demise of the Tang.

PROPERTY, MARRIAGE, AND KINSHIP SOLIDARITY

In reference to political privilege, there seems to have been a hierarchical ramification among the Suns, such that descent lines differed in the levels of political privilege they enjoyed. In addition to recording ancestral history, the inscription of Stele 1 shows that another prominent purpose of its installation was to prove the family's close connection and loyalty to the imperial household. Spelling out the meritorious service of the successive patriarchs since the reign of Činggis Qan, the inscriptions of the successive patriarchs and the two imperial edicts engraved on Stele 4 (see Table 1, “Steles in the Sun Family Graveyard”) collectively corroborate the clan members’ service under Mongol rule for four generations. As mentioned above, maintaining and proving genjiao was essential for the continuous prosperity of office-holding families. Christopher Atwood persuasively argues that, in the Mongol ruling stratum at the time, the knowledge of one's ancestry and lineage was crucially important. As the notion of genjiao demonstrates, the number of generations in the service of khans, or a princely bloodline, represented the proximity to the khan. Consequently, genealogical information was frequently referred to in confirming one's political and social status in officialdom. ⁴⁸ Ruth Dunnell has also pointed out, in her discussion of an office-holding family who were descendants of a Tangut general, “[t]hus the offspring of non-Chinese immigrants and Chinese alike who had joined the early Mongol conquest established and assiduously cultivated the documentary record of their family's service to secure and expand the hereditary privileges they came to enjoy in the new socio-political order.” ⁴⁹ The Yuan statutes concerning inheritance of official status by protection (yin 蔭) also prescribe that the applicant needed to report the official careers (i.e. genjiao) of ancestors (fuzu qianhou lishi genjiao 父祖前後歷仕根腳), together with official certificates and a “colored genealogy” (caihua zongzhi 綵畫宗枝) to prove his inheritance right. Furthermore, the submitted materials were to be confirmed by the relatives of the same descent line (fangqin 房親). ⁵⁰ Apart from preservation of genealogical information, Sun Gongliang's installation of the genealogical stele can also be understood in this context. The khans after the later half of the thirteenth century legitimized the emergent epigraphic practice (i.e., engraving genealogies on steles) by ordering eminent scholar-officials to compose inscriptions for meritorious subjects as a symbol of imperial grace, which was then to be inscribed on a stele that the khan granted to the same subject. ⁵¹ In the Yuan system, one needed to prove he/she had hu'jaur and how generationally deep it was, when the person was to obtain or inherit an official post. ⁵² A stele inscription, from which one could make a portable rubbing, appears to have been an ideal material on which to record one's genjiao. ⁵³ As a rank 3 official who had a direct connection with the imperial household, Sun Gongliang was most likely aware that preserving a genealogy in the form of a stele inscription would further facilitate his family's political success based on their superior hu'jaur. Written by Liu Yin 劉因 (1249–93), a Jixian academician and an eminent northern scholar, in narrating the family history, the inscription concludes: “Yet lord [Gongliang] is not yet old and can exert himself for the state. Gong [Gongliang's eldest son] and other clan members are also all engaged in service. Someday I [Liu Yin] will again examine their successes and failures” (然而公未老，尚能盡力國家，而拱等才且亦皆任事。予他日又可以攷其淺深厚薄於此). ⁵⁴ Apparently, the installation of the genealogical stele was the clan's momentous endeavor in order to commemorate their success under Mongol rule, and record their genealogy for future use.

As the case of the Suns suggests, the beneficiary status of genjiao was limited to a legitimate heir in the direct descent line of the first ancestor who entered into the genjiao relation with a Mongol master. Unlike preceding dynasties, the Yuan strictly limited the scope of protection to a single kinsman. A similar policy seems to have been applied to genjiao holders, supposedly to prevent overuse of the privilege. Naturally, elucidating the inheritance right to be a candidate must have been crucially important when it came to sharing in the privilege. This must be the context needed to understand why descendants of Sun Gui 孫貴 (dates unknown), one of the younger cousins of Sun Wei and relatives in the fourth collateral degree of Sun Gong, appear to have been far less successful when it comes to office-holding. These Suns also installed at least two steles in the same graveyard after 1324, some 15 meters east of Stele 1. One of them, Stele 7, has the epitaph of Sun Fu 孫撫 (1233–85), a grandson of Sun Gui, who was conscripted to serve a Mongol garrison troop stationed somewhere along the Southern Song border in 1255. Although assigned to the post of provisional battalion commander (quan qianhu 權千戶) on the basis of his war exploits, he had to return home because of a disease and died a commoner. His younger brother Sun Hui 孫揮 (1240–1314), for whom Stele 8 was installed, never left Hunyuan and also died without an official title. Interestingly, the weathered scripts on the verso of Stele 8 strongly indicate that the stele originally had a genealogical record of Sun Hui's descendants, presumably with the dates of births or deaths. ⁵⁵ The names still recognizable on the upper half of the verso (“Nan Tuohuan 男脱歡,” “Nan Tu…dai 男兎□歹,” “Sunnü Mian 孫女綿”) also do not appear in the genealogical chart on Stele 1. Among these names, Toγon (Tuohuan) was recorded as a grandson of Sun Hui in the inscription on the front side. This strongly suggests the genealogical record on Stele 8 was engraved to record the kinfolk in this branch, mostly born after 1299, to preserve their own genealogy separately from the main branch. The extant steles in the graveyard indicate that after 1299 the Suns did not attempt to compile another comprehensive genealogy in any form. The separate genealogical record on Stele 8 may suggest that the gap between the branches had significantly widened by 1324. Although sharing the graveyard, the two sons of Sun Fu, Sun Liang 孫諒 and Sun Tan 孫譚 (both dates unknown), appear on Steles 7 and 8 without official titles, while in Stele 5, their cousin in the main branch, Sun Qian 孫諧 (dates unknown), is recorded to have reached the position of gentleman of Chaolie (Chaolie dafu 朝列大夫), ranked 4b, working as vice pacification commissioner in Hedong Shanxi circuit (Hedong Shanxi xuanwei fushi 河東山西宣慰副使). It must be noted that the shared generational character (yan 言) corroborates that the internal bond between the branches was firm at the time. Also, as the location of the steles itself indicates, the main branch occupied the western part while the collaterals used the eastern part of the graveyard, centered on Stele 1, and generational order appears to have been observed strictly. As a whole, the kinship solidarity of the Suns was not shaken. Yet at the same time, in principle the fruits of genjiao privilege were granted exclusively to the direct descendants of Sun Gongliang.

Meanwhile, the collaterals appear to have been self-reliant with respect to property. In the Jin-Yuan north, financial ties between descent lines were weak at best, or almost nonexistent. Except for the ancestral graveyard, whose ownership was in general shared among descent lines, genealogical steles and other contemporary material sources never mention corporate property in any form, such as a land trust. A 1281 legal judgment on distribution of parental property, stipulates that “when family property must be distributed, if one has acquired property through his official or military service, or from his wife's household through marriage, it is not subject to the restrictions of equal distribution” (應分家財，若因官及隨軍、或妻家所得財物，不在均分之限). This indicates that family property, except for that accrued personally, was to be distributed equally among siblings upon their parents’ deaths. ⁵⁶ Of course, in principle, filial sons were not to distribute family property, but rather live together and share the ancestral property. We have several records from the Jin-Yuan period commemorating kin groups for living together for five to six generations. ⁵⁷ The commemorative records themselves, however, testify that the formation of such kinship solidity was a rare occurrence. In the case of the Suns, distinct descent lines prove to have owned family property separately. Upon his return from military service, Sun Fu faced his brothers’ persistent demand to distribute their parental property. To avoid the breakup of the branch, Sun Fu, as the eldest among the siblings, had to distribute the property, if only reluctantly. ⁵⁸ According to conventional custom, Sun Fu had the authority to handle his parents’ property without consulting with kinsfolk in other branches. We do not know if the branch had any shared property. Yet the fact that no contemporary genealogical stele refers to any form of shared property in a branch tells a lot.

Marriage patterns appearing in the steles in the graveyard also seem to suggest that, in addition to being financially independent, the distinct branches did not share marriage strategies. Each branch chose spouses among neighbors in the county. The genealogical chart on Stele 1 testifies that male descendants of Sun Gui, one of the younger cousins of Sun Wei and the grandfather of ex-Provisional Battalion Commander Sun Fu, married women surnamed Li 李 (six cases), Zhang 張 (four cases), Zhao 趙 (three cases), Hou 侯 (three cases), Wang 王 (two cases), and (one case each) Gu 谷, Shi 史, Yao 姚, Liu 劉, Bai 白, Shao 邵, Lian 連, and He 賀. Sun Gui himself married two women, surnamed He 何 and Dong 董, respectively. As the wives were recorded without their birthplace, we do not know if the kinsmen married locally or trans-locally. In contrast, the female descendants of the same branch appear with their husband's places of residence (or birth) as follows: Li Xiu 李秀 of Dengjiazhuang 鄧家莊, Wu Yong 武用 of Yaocun 姚村, Liang Xiang 梁享 of Hunyuan, Zhang Zhongyuan 張仲元 of Xiliucun 西劉村, Wang Boxiang 王伯祥 of Sunjiawan 孫家灣, Zhang X 張□ of Bicun 畢村, Wang Jin 王進 of Anping 安平, Liu Xin 劉信 of Yaocun, Hu Deshan 胡德山 of Bicun, Jia Shichang 賈世昌 of Xujiacun 戌家村, Yin Yuanzhen 尹元貞 of Hunyuan, Wang Shanqing 王善慶 of Hunyuan, Liu Fujin 劉福進 of Qimaocun 戚毛村, Zhang Rong 張榮 of Zhangjiazhuang 張家莊, Zhang Zhongxin 張仲信 of Zangjiazhuan 藏家莊, Bai Zhongrong 白仲榮 of Xiliucun, X Xde □□德 of Xiliucun, and He Geng 賀庚 of Bicun. Except Anping, which is supposed to be a county in modern Hebei, the other settlements were supposedly located in Hunyuan county. Yaocun, Xiliucun, and Bicun each appear multiple times, among which Xiliucun could have been another spelling, or an earlier name of Xiliucun 西留村, the home village of the Suns. The marriage pattern of Sun Gui's branch contrasts with that of the main branch. The male descendants of Sun Wei married women surnamed Hao 郝 (two cases), and (one case each) Liang 梁, Zhang 張, Dula 篤剌, Jiao 焦, Li 李, Shang 尙, Qu 瞿, and Lü 呂. Similarly, no clear marriage policy can be observed, except for the case of Sun Qing, who married an obviously non-Chinese woman surnamed Dula, which may suggest his connection with a Mongol princely household.

As in the case above, the place names and official titles associated with husbands of female descendants provide a clue to their marriage circles: Zhao Xiang 趙襄, magistrate of Anping county; Li Xiang 李享, clerk of the interpreters’ institute; Liu Juji 劉巨濟, superintendent of artisans in Hangzhou circuit; Zhao Yu 趙瑀, tax commissioner of Qizhou; and Wang Chuqian 王處謙 of Baoding. It is worth noting that the patriarch of the Suns in each generation was charged with supervising the armory smithy exactly in the regions where the husbands of the kinswomen held their offices. Supposedly closely related to this tendency, the wife of Sun Gong, surnamed Hao, was also a daughter of Hao Deyan 郝德言 (dates unknown), governor of arrow smiths in Zhending circuit (Zhending gongjiang zongguan 真定弓匠總管). It is hard to know if the Suns established trans-generational marriage ties with the families in their jurisdiction, northern Hebei. Yet it appears to be true that the main branch, the direct descendants of Sun Wei, expanded and maintained their official network through marriages, while the collateral branch kept on finding spouses in the home county. Recent studies on northern office-holding families in the Yuan also suggest the same was true among other contemporary office-holding families in the north. While only major male members of the direct descent line established trans-prefectural marriage ties with office-holding families, the great majority of kinfolk remained local in their marriage networks. ⁵⁹ As the Sun clan grew steadily larger over time, the difference between the marriage patterns of distinct descent lines became clearer, as the prestige of being direct descendants of Sun Wei became a prerequisite to obtaining an official rank and thus finding a spouse from fellow or subordinate office-holding families.

The seemingly irregular marriage pattern of the Suns had a lot in common with that of the Xues in Ming North China, as Khee Heong Koh discusses: “The statistics [extracted from Xue Genealogy] suggest that even within the same line, there was a lack of a strong diachronic pattern of marriage alliances with a particular surname group.” ⁶⁰ Koh’s remarks comfortably fit into the abovementioned Ebrey’s definition of “jia” orientation in imagining descent. The same “jia” orientation appears to have persisted beyond the devastating Yuan-Ming transition. The sole, obvious difference is that although connection with Xue Xuan 薛瑄 (1389–1464), the progenitor of both the Hedong school and the clan, conferred certain cultural and social prestige on his direct descendants, having him as direct ancestor “did not turn into tangible economic and social benefits for his descendants.” ⁶¹ In the meantime, genjiao not only defined lineal descent, but brought official and political benefits in very practical ways and demonstrated the political superiority of the main branch and their marriage patterns. For successive patriarchs, their goal was to maintain and improve their genjiao, which decisively defined one's political status in the Mongol Empire. In a sense, the Yuan kinship organization among the northern office-holding families was formed under a substantially non-Chinese political system, in contrast with that of southern lineages, which was facilitated mostly by economic and social interests at the local level.

CONCLUSION

It is a mere coincidence that Sun Wei, the first patriarch of the Suns of Hunyuan under Mongol rule, lived in approximately the same period as Shi Miyuan 史彌遠 (1164–1233), who realized the golden age of the Shi clan of Mingzhou. However, comparison of the two office-holding households, which prospered in the north and south respectively, illustrates the significant disparity in the evolution of political elite families in Middle Period China. While establishing solid footholds in the two distinct political systems, the Suns and Shis prospered by attaining high official ranks over generations in different ways: the former adapting themselves to the Mongol traditional master-servant bond system, and the latter relying on the examination and protection systems. Of course, the case of the Suns does not necessarily suggest that the former strategy overwhelmingly prevailed in the north. Less successful office-holding families without a powerful genjiao and fledgling aspirants without any connections could opt to take the examinations after its resumption in 1313. Yet contemporary historical sources strongly suggest such aspirants nonetheless preferred to establish a connection if possible. A 1317 source impressively records that every day crowds of hopeful clients, including many from southeast China, pressed at the main gate of the palace in the capital Dadu to obtain a connection with a passing Mongol prince or high-ranking official, even after the resumption of the examinations. ⁶² Until the end of Mongol rule, it remained true that to enter into the upper stratum of the ruling elite, one inevitably needed a genjiao. ⁶³

The equally evident difference between the northern and southern political elite families is the way they recorded their genealogies. Unlike the genealogical texts that came to help unite collateral kinsmen into a lineage, the case of the Suns demonstrates that the installation of a genealogical stele functioned to strictly clarify the membership of an office-holding descent line, presumably in order to elucidate who were to be given the priority to benefit from the political privilege granted by the Mongol regime. This evolution of northern descent groups under centuries of non-Chinese rules does not, again, necessarily contradict our conventional understanding of Chinese kinship organization. In the south, members of a lineage similarly used to be stratified in accordance with their economic conditions and political power derived from office-holding. The same similarity can also be observed in marriage patterns. In the case of the Suns, the stratification of kinsfolk is also reflected in the marriage patterns of different descent lines, much as in southern cases such as that of the Shis. Yet the two parallel trajectories did not merge into one grand stream during the Yuan, as the northern elite families seem to have focused on strengthening their ties with the ruling stratum of the empire, while paying little attention to expanding kinship locally. In the meantime, with fewer opportunities to establish prestigious political connections, southern families could only expand their affinal networks locally.

As such, the Mongol conquest did mark a point of divergence, such that the pattern of success in the leading elite strata in North and South China during the Yuan went different ways. And this impact of Mongol rule may have lasted beyond the demise of Mongol rule in China. When it comes to political success, the mediocre performance of Ming-Qing northern local literati in the examination system may be traced back to Mongol rule. Gu Yanwu 顧炎武 (1613–82), one of the most important figures in Chinese intellectual history during the Ming-Qing transition, testified, “Today, when it comes to teaching Elementary Learning, southerners let their students make couplets. This may be a traditional way of learning, which has been transmitted from one generation to the next since the Tang-Song; yet the northerners do not practice it at all”?. ⁶⁴ Although Gu attributed the unimpressive performance of the northerners to their lack of talent, we know that his observation is unfair. Unlike the south, the north experienced a century-long rupture in the examination system during the Yuan, as a consequence of which the “traditional way of learning” in preparation for the examinations had not had been “transmitted from the Tang-Song.” In the longer perspective, the different type of elite pattern represented by the Suns resulted in a divergent regional evolution in Chinese history.