We use cookies to distinguish you from other users and to provide you with a better experience on our websites. Close this message to accept cookies or find out how to manage your cookie settings.
Published online by Cambridge University Press: 08 August 2014
This paper reconsiders the dating of the Houma covenant texts in light of new findings from the Wenxian covenant texts. Dating of the Houma covenants has focused on matching certain names found in the Houma covenants to names and events in historical texts. These include the name of the sanctioning spirit invoked in the covenants, and that of the covenant lord overseeing the covenants. I argue that the sanctioning spirit is not, as is often proposed, a former lord of Jin, but a mountain spirit called Lord Yue, and, as such, has no bearing on the dating of the texts. I further argue that the use of the personal name of a Han lineage leader in the Wenxian covenants strongly supports the identification of the figure referred to as jia 嘉 in the Houma texts as the historical Zhao Jia (Zhao Huan Zi). I suggest that the mention of Zhao Jia in the recently published Chu-slips Xinian implies that Zhao Jia came to the leadership of the Zhao lineage around 442 B.C.E., well before 424 B.C.E., the date of his single-year reign reported in the Shi ji. I conclude that the Houma covenants include materials that may be linked to the Zhao Wu incident of the early fifth-century B.C.E., but that those materials in which Zhao Jia is named as the covenant lord probably date to sometime between 442 and 424 B.C.E.
本文根據溫縣盟書中的新發現重新考察侯馬盟書的年代問題。判斷侯 馬盟書年代的主要根據是侯馬盟書中能夠與歷史記載相關聯的一些人 名和地名。這些名字包括盟書中被召監督參盟人的神的名稱以及盟主 的名字。本文推測該神非晉國的某位先君，而是一位叫做 “岳公” 的山 神，因而對推測侯馬盟書的年代不構成限制。此外，根據溫縣盟書中 可確認為盟主人名的例子，可以推測侯馬盟書中的 “嘉” 字應該就是指 趙嘉 (趙桓子)。根據清華簡《繫年》中關於趙嘉的記載，本文進一步 推測趙嘉在公元前 442 年左右已經是趙氏的族長 ，比《史記》中記載 趙嘉 424 年即位早十幾年。根據這些分析，侯馬盟書中有的盟書應該 與公元前五世紀初的趙午事件有關，但是那些提到趙嘉的盟書的年代 大概在公元前 442 到 424 年之間。
My research on the Wenxian covenant texts has been made possible as a result of my collaboration on the publication project for these texts. I would like to acknowledge my gratitude to the excavators of the Wenxian covenants, Hao Benxing 郝本性 and Zhao Shigang 趙世綱 of the Henan Provincial Institute of Cultural Relics and Archaeology, and to Sun Xinmin 孫新民, the head of the Institute, for their support in the use of these materials in my research, and permission to use the images included in this paper. I thank Susan Roosevelt Weld for inviting me to join the project that she initiated with the excavators to photograph and digitalize the Wenxian texts. Research for this article was supported by: a grant from the University of Kansas New Faculty General Research Fund; a Fellowship for East Asian Archaeology and Early History from the American Council of Learned Societies, with funding from the Henry Luce Foundation; a Franklin Research Grant from the American Philosophical Society; a fellowship from the National Endowment of the Humanities (NEH). The NEH requires that the following statement be included: “Any views, findings, conclusions, or recommendations expressed in this publication do not necessarily reflect those of the National Endowment for the Humanities.” The University of Kansas generously provided supplemental salary funding during the period of my fellowships.
1. For the early sixth-century B.C.E. suggestion see Yumin, Li 李裕民, “Wo dui Houma mengshu de kanfa” 我對侯馬盟書的看法, Kaogu 考古 1973.3, 185–91Google Scholar. For the fourthcentury B.C.E. suggestion see Moruo, Guo 郭沫若, “Houma mengshu shitan” 侯馬盟 書試探, Wenwu 文物 1966.2, 4–6 Google Scholar; and Moruo, Guo, “Chutu wenwu er-san shi” 出土文 物二三事, Wenwu 1972.3, 2–10 Google Scholar.
2. Unless otherwise stated, extrapolated B.C.E. dates are taken from the Cihai 辭 海 chronology, see weiyuanhui, Cihai bianji, Cihai (Shanghai: Shanghai cishu, 1989), 5427–88Google Scholar.
3. Henan, sheng wenwu yanjiusuo, “Henan Wenxian Dong-Zhou mengshi yizhi yi-hao kan fajue jianbao” 河南溫縣東周盟誓遺址一號坎發掘簡報, Wenwu 1983.3, 77, 78–89 Google Scholar. See also Shigang, Zhao and Li, Zhao 趙莉, “Wenxian mengshu de lishuo yanjiu” 溫 縣盟書的曆朔研究, Xinchu jianbo yanjiu 新出簡帛研究 (Beijing: Wenwu, 2004), 197–205 Google Scholar.
4. Xueqin, Li 李學勤, “Houma, Wenxian mengshu lishuo de zai kaocha” 侯馬、溫 縣盟書曆朔的再考察, Hua xue 華學, 3 (1998), 165–68Google Scholar.
5. Xueqin, Li, Eastern Zhou and Qin Civilizations, trans. Chang, K.C., (New Haven and London: Yale University Press, 1985), 40–52 Google Scholar. sheng, Shanxi wenwu gongzuo weiyuanhui, Houma mengshu 侯馬盟書 (Beijing: Wenwu, 1976), 2, 384 Google Scholar.
6. There is a date repeated on many tablets in one pit at Wenxian (with possibly one example from another pit). This date has a reign year and a matching date has been proposed, but the many variables involved in reconstruction of the early calendar mean there is potential for erroneous conclusions. Furthermore, as with the Houma tablets, we cannot assume covenants from different pits were carried out on the same day. For discussion of the date from Wenxian, see Li Xueqin, “Houma, Wenxian mengshu li shuo de zai kaocha” and also those works cited in n.3 above.
7. Xueqin, Li, ed., Qinghua Daxue cang Zhanguo zhujian (2) 清華大學藏戰國竹簡(貳) (Shanghai: Shanghai wenyi, 2011)Google Scholar.
8. Shanxi sheng, Houma mengshu.
9. Henan sheng wenwu yanjiusuo, “Henan Wenxian Dong-Zhou mengshi yizhi yi-hao kan fajue jianbao.”
10. This four-clause structure was described by Weld, Susan Roosevelt: “Covenant in Jin's Walled Cities: the Discoveries at Houma and Wenxian,” Ph.D. dissertation (Harvard University, 1990), 353–54Google Scholar.
11. Unless otherwise indicated, transcriptions will be given in an interpretative form, i.e., using the standard characters for the words I believe are denoted by the graphs in the palaeographic materials.
12. The labels for the Houma tablets include two numbers, the first is the number of the pit in which the tablet was found, the second is the number of the individual tablet. Thus 156:1 refers to tablet 1 from pit 156. HM is added here to indicate that the tablet is from Houma and not Wenxian. For a copy and image of this tablet, see sheng, Shanxi, Houma mengshu, 35, 123 Google Scholar.
13. I adopt an identification of the word here as ji 極 that was suggested by Chen Jian 陳劍: personal communication, February 22, 2009.
14. For a discussion of the scope of the term shi 氏 in the excavated covenant tablets, see Williams, Crispin, “Early References to Collective Punishment in an Excavated Chinese Text: Analysis and Discussion of an Imprecation from the Wenxian Covenants,” Bulletin of the School of Oriental and African Studies, 74.3 (2011), 437–67CrossRefGoogle Scholar.
15. For an annotated example of a complete Pledge Text, see sheng, Shanxi,Houma mengshu, 37–39 Google Scholar.
16. Kebin, Wei 魏克彬 (Crispin Williams), “Houma yu Wenxian mengshu zhong de ‘Yue Gong’” 侯馬與溫縣盟書中的‘岳公,’ Wenwu 2010.10, 76–83, 98 Google Scholar.
17. For a copy and image of this tablet, see sheng, Shanxi, Houma mengshu, 39–40, 151 Google Scholar.
18. For a copy and image of this tablet, see sheng, Shanxi, Houma mengshu, 33, 83 Google Scholar.
19. Although it is unclear from the photographs of this tablet, it appears that the editors of the Houma mengshu did not think the tablet had been damaged, in which case we should not assume that a reign year is missing from this formula. For example, when quoting the first two lines of the tablet on page 74 of the Houma mengshu they make no allowance for lacunae caused by a damaged tablet. If there was no damage to this part of the tablet, then the Houma dating formula does not include a reign year, unlike the formula used for the date in the Wenxian texts. See sheng, Shanxi, Houma mengshu, 74 Google Scholar.
20. Xueqin, Li, “Houma, Wenxian mengshu lishuo de zai kaocha,” 166 Google Scholar.
21. For copies and images of these tablets, see sheng, Shanxi, Houma mengshu, 41–42, 154 Google Scholar.
22. For a copy and image of this tablet, see sheng, Shanxi, Houma mengshu, 49, 159 Google Scholar.
23. sheng, Shanxi, Houma mengshu, 77 Google Scholar.
24. For a convenient collection of passages from texts which recount this incident, see sheng, Shanxi, Houma mengshu, 421–29Google Scholar.
25. There are further minor variations in the wording: HM 1:40 leaves out the fu 腹 of fuxin 腹心; part of tablet HM 1:42 is missing and the missing section includes the section where we would expect to find the characters qi fuxin yi 其腹心以. Based on their calligraphic style and several other unusual characteristics of these three texts, they were almost certainly written by a single scribe. For copies of the tablets see sheng, Shanxi, Houma mengshu, 171–72Google Scholar.
26. For a copy and image of this tablet, see sheng, Shanxi, Houma mengshu, 86, 167 Google Scholar. Tablets HM 1:23 and HM 1:24 appear to have the same variant, but the title is not fully legible.
27. I would conjecture that the first-person pronoun yu 余 that appears earlier in the HM 16:3 text refers to jia himself. Thus, in this text jia first announces sacrifice to Lord Yue, thus invoking the spirit, he then says that he himself will take some action with respect to the decrees given at the Ding Temple and Ping Altar, and then he goes on to order his ministers to take some action, and says that if they do not obey, this will trigger the imprecation. The imprecation would have applied to the ministers, and perhaps jia himself, and the spirit called on to sanction this is almost certainly Lord Yue, invoked at the start of the text.
28. Shi ji 史記 (Beijing: Zhonghua, 1959)Google Scholar, 43.1796–97.
29. Li Xueqin's article gives 424 B.C.E., Yang Kuan 楊寬 suggests 425 B.C.E. and Hirase Takao 平勢隆郎 gives 426 B.C.E. See Xueqin, Li, “Houma, Wenxian mengshu li shuo de zai kaocha,” 166 Google Scholar; Kuan, Yang, Zhanguo shiliao biannian jizheng 戰國史料編年 輯證 (Shanghai: Shanghai renmin, 2001), 56 Google Scholar; Takao, Hirase, Shinpen Shiki Tōshū Nenpyō 新編史記東周年表 (Tokyo: Tōkyō Daigaku Tōyō Bunka Kenkyūjo, 1995), 142–43, 469, 640 Google Scholar. In the discussion below this date will be given as 424 B.C.E.
30. Shi ji 43.1789, 1796–97.
31. See the Suo yin 索隱 commentary for Shi ji 43.1796–97.
32. For the identification as Duke Qing of Jin, see Zhenwu, Wu 吴振武, “Guanyu Wenxian mengshu zhong de ‘Qing Gong’” 關於溫縣盟書中的 ‘ 公’, Xinchu jianbo yanjiu 新出簡帛研究 (Beijing: Wenwu, 2004), 206–7Google Scholar. For the identification as Duke Chu of Jin, see Ming, Gao 高明, “Houma zaishu mengzhu kao” 侯馬載書盟主考, Guwenzi yanjiu 古文字研究 1 (1979), 103–15Google Scholar.
33. sheng, Shanxi, Houma mengshu, 65–68 Google Scholar. Zhang Han 張頷 was the main author of the sections in the excavation report that analyze the texts. See also Han, Zhang, Zhang Han xueshu wenji 張頷學術文集 (Beijing: Zhonghua, 1995)Google Scholar.
34. sheng, Shanxi, Houma mengshu, 65 Google Scholar.
35. sheng, Shanxi, Houma mengshu, 74–77 Google Scholar.
36. Lan, Tang 唐蘭, “Houma chutu Jinguo Zhao Jia zhi mengzaishu xinshi” 侯馬 出土晉國趙嘉之盟載書新釋, Wenwu 1972.8, 31–35, 58 Google Scholar. Ming, Gao, “Houma zaishu mengzhu kao,” 108–11Google Scholar.
37. Ming, Gao, “Houma zaishu mengzhu kao,” 111–13Google Scholar. The pit 105 tablets had not been published when Tang Lan wrote his article.
38. Ming, Gao, “Houma zaishu mengzhu kao,” 108–11Google Scholar.
39. See n.32 above.
40. Xueqin, Li, “Houma, Wenxian mengshu li shuo de zai kaocha,” 167 Google Scholar.
41. The following section is largely based on Wei Kebin (Crispin Williams), “Houma yu Wenxian mengshu zhong de ‘Yue Gong.’”
42. Haruki, Emura 江村治樹, “Kōma Meisho Kō” 侯馬盟書考, Uchida Ginpū Hakushi Shūju Kinen Tōyōshi Ronshū 内田吟風博士頌壽紀念東洋史論集 (Tokyo: Dohosha, 1978), 97 n.43Google Scholar.
43. Examples from the Houma covenant texts are prefixed with “HM” and are taken from Shanxi sheng, Houma mengshu. All other examples are from the Wenxian covenants.
44. For , see Mengjia, Chen 陳夢家, “Dong Zhou mengshi yu chutu zaishu” 東周 盟誓與出土載書, Kaogu 1966.5, 277 Google Scholar. For , see Lan, Tang, “Houma chutu Jinguo Zhao Jia zhi mengzaishu xinshi,” 31 Google Scholar, and also see Han, Zhang, “‘Houma mengshu’ congkao xu” ‘侯馬盟書’ 叢考續, in Han, Zhang, Zhang Han xueshu wenji, 91–109 Google Scholar (first published in Guwenzi yanjiu 古文字研究 1 [1979], 78–102 Google Scholar). For 出, see Ming, Gao, “Houma zaishu mengzhu kao,” 108–11Google Scholar. For , see Wu Zhenwu, “Guanyu Wenxian mengshu zhong de ‘Qing Gong.’” For 舌, see Jiahao, Li 李家浩, “Yan zhong mingwen kaoshi”鐘銘 文考釋, Zhuming zhongnian yuyanxuejia zixuanji, Li Jiahao juan 著名中年語言學家自選 集•李家浩卷 (Hefei: Anhui jiaoyu, 2002), 68 n.1Google Scholar.
45. Based on the similarity of the hand we can be confident that one scribe wrote all four tablets.
46. Shen, Xu 許慎, Shuowen jiezi (Beijing: Zhonghua, 1963), 5 Google Scholar (mu 目.5a), 3 (mu 目.1b).
47. sheng, Shanxi, Houma mengshu, 353 Google Scholar.
48. Shuowen jiezi, 206 (10a yinbu 犾部.15b).
49. Ming, Gao, Zhongguo guwenzixue tonglun 中國古文字學通論 (Beijing: Wenwu, 1987), 153–54Google Scholar.
50. I use the Old Chinese reconstruction system of William H. Baxter and Laurent Sagart. At the time of writing, their most recent set of reconstructions is available on the website of the Centre de Recherches Linguistiques sur l'Asie Orientale (http://crlao.ehess.fr/document.php?id=1217). For the posthumous names of the Jin and Zhou lords, see, for example, weiyuanhui, Cihai bianji, Cihai, 5429–39Google Scholar.
51. Karlgren, Bernhard, Grammata Serica Recensa (Taibei: SMC Publishing Inc.), 312–13Google Scholar.
52. Shuowen jiezi, 190 (9b shanbu 山部.1a).
53. Zhongshu, Guo 郭忠恕, Han jian 汗簡, in Han jian Guwen sisheng yun 汗簡古文四 聲韻, ed. Ling, Li 李零 and Xin'guang, Liu 劉新光 (Beijing: Zhonghua, 1983), 26 Google Scholar (51a).
54. Song, Xia 夏竦, Guwen sisheng yun 古文四聲韻, in Han jian Guwen sisheng yun, 73 (5.6b, 7a).Google Scholar
55. As mentioned, Emura Haruki made the point that the Shi ji uses the term dazhong to refer to mountains, see n.42 above.
56. There is evidence for this phenomenon as early as the Shang period. Sarah Allan, for example, points out that in the oracle bones the Shang spirit Di 帝 has “ministers” chen 臣, suggesting that “Di commanded a celestial court, like that of the early ruler.” David Pankenier argues: “In both Shang and Zhou … we have the projection into the supernatural realm of a socio-political model of sovereignty and hierarchical order.” See Allan, Sarah, “On the Identity of Shang Di 上帝 and the Origin of the Concept of a Celestial Mandate (Tian Ming 天命),” Early China 31 (2007), 7–8 CrossRefGoogle Scholar. Pankenier, David, “The Cosmo-Political Background of Heaven's Mandate,” Early China 20 (1995), 166 CrossRefGoogle Scholar.
57. Li ji jijie 禮記集解, ed. Xidan, Sun 孫希旦 (Beijing: Zhonghua, 1989), 347 Google Scholar (“Wang zhi” 王制 13.5.2).
58. Moruo, Guo, Shiguwen yanjiu Zu Chu wen kaoshi 石鼓文研究詛楚文考釋 (Beijing: Wenwu, 1982)Google Scholar.
59. Zaixing, Zhang 張再興, “‘Wen’ ‘huang’ kaobian” ‘文’ ‘皇’考辯, Yuyanwenzixue (Renda fuyin baokan ziliao 人大複印報刊資料) 2008.5, 101–9Google Scholar.
60. For an overview of mountain worship in early China, see Kleeman, Terry F., “Mountain Deities in China: The Domestication of the Mountain God and the Subjugation of the Margins,” Journal of the American Oriental Society 114.2 (1994), 226–38CrossRefGoogle Scholar. For this practice in the Eastern Han, see Brashier, Kenneth E., “The Spirit Lord of Baishi Mountain: Feeding the Deities or Heeding the yinyang ,” Early China 26–27 (2001–2002), 159–231 CrossRefGoogle Scholar.
61. For the Baoshan slips, see Wei, Chen 陳偉, Baoshan Chu jian chutan 包山楚簡初探 (Wuhan: Wuhan Daxue, 1996)Google Scholar. Li Xueqin discusses the Qin Yin jade tablets and provides a transcription in Xueqin, Li, “Qin yuban suoyin” 秦玉牘索隱, Gugong bowuyuan yuankan 故宮博物院院刊 2000.2, 41–45 Google Scholar. I have adopted Li's dating and identification of the Qin king as Huiwen here. For a discussion of these points, and an English translation of this text, see Pines, Yuri, “The Question of Interpretation: Qin History in Light of New Epigraphic Sources,” Early China 29 (2004), 4–14 CrossRefGoogle Scholar.
62. Chunqiu Zuo zhuan zhu 春秋左傳注, ed. Bojun, Yang 楊伯峻 (Beijing: Zhonghua, 1990), 989–90Google Scholar (Xiang 襄 11.3).
63. Wanli, Qu 屈萬里, “Yue yi jigu” 岳義稽古, Qinghua xuebao 清華學報 2.1 (1960), 53–68 Google Scholar. Cited in Kleeman, , “Mountain Deities in China,” 227 Google Scholar.
64. Shi ji 43.1781.
65. Shi ji 5.174–5.
66. Shi ji 43.1794–5.
67. Hao Benxing, personal communication, September 2008.
68. Shi ji 44.1838. See the Suo Yin commentary on the name Han Wu Zi 韓武子.
69. While completing this article, I have determined that this name is Qu 取, the name of Han Lie Hou 韓烈侯 (r. 399–387 B.C.E.), see Kebin, Wei (Williams, Crispin), “Wenxian mengshu T4K5, T4K6, T4K11 mengci shidu” 溫縣盟書 T4K5, T4K6, T4K11 盟辭釋讀, Chutu wenxian yu guwenzi yanjiu 出土文獻與古文字研究 5 Google Scholar (forthcoming).
70. Hirase gives 423–408 B.C.E., Hirase, , Shinpen Shiki Tōshū Nenpyō, 144–48, 641 Google Scholar.
71. Li Xueqin discusses this and other naming conventions in Xueqin, Li, “Xian Qin renming de ji ge wenti” 先秦人名的幾個問題, Lishi yanjiu 歷史研究 1991.5, 106–11Google Scholar.
72. The most obvious example being that of the main enemy named in the Houma covenants as Zhao Hu 趙弧. If the covenants naming Zhao Jia as leader reflect a conflict with Zhao Huan, then one would expect Zhao Hu to refer to Zhao Huan, but the suggestion that hu 弧 could loan for the name Huan 浣 is highly problematic. The reconstructions are: 弧 hu < hu < *[g]ʷˤa and 浣 huan < hwanX < *[ɢ]ˤo[n]ʔ. The initial consonants are of different types (labialized and nonlabialized) and the vowels and codas are different. On this basis, hu 弧 is not a suitable graph to write the name huan 浣. (I am grateful to William H. Baxter for comments on this issue: personal communication, August 25, 2012.)
73. Susan R. Weld has suggested this may be the case, see Weld, , “Covenant in Jin's Walled Cities,” 432 Google Scholar.
74. This is the approach proposed by Wang Guowei 王國維 in the 1920s, for which he used the term “er zhong zhengju fa 二種證據法 “the two-types-of-evidence method.” Li argues for this approach in his book Zouchu yigu shidai 走出疑古時代, rev.ed. (Shenyang: Liaoning Daxue, 1997)Google Scholar. See also Guowei, Wang, Gushi xin zheng: Wang Guowei zuihou de jiangyi 古史新證–––王國維最後的講義, ed. Tao, Song 宋韜 (Beijing: Qinghua University, 1996)Google Scholar. See also Fischer, Paul, “Authentication Studies (辨偽學) Methodology and the Polymorphous Text Paradigm,” Early China 32 (2008–2009), 1–43 Google Scholar, particularly pages 31–36.
75. Gao Ming also makes this point: Gao Ming, “Houma zaishu mengzhu kao,” 104. There is, it should be acknowledged, evidence from the transmitted histories that appears to suggest the Zhao lineage used meng to refer to their leader regardless of his position in the birth order. If this was the case for the scribe who wrote the variant using meng in the Houma covenants, then this evidence would not be relevant to the question of Zhao Jia's relationship to Zhao Xiang Zi. Tsang Chi-hung 曾志雄 makes this point, citing the study of Fang Xuanchen 方炫琛, see Chi-hung, Tsang, “A Study of Alliance Pacts Unearthed at Houma” Houma mengshu yanjiu 侯馬盟書研究, Ph.D. dissertation (University of Hong Kong, 1993), 67–68 Google Scholar; Xuanchen, Fang, Zuo zhuan renwu minghao yanjiu 左傳人物名號研究, Ph.D. dissertation (Taibei: Guoli Zhengzhi Daxue 國立政治大學, 1983), 512, 569–70Google Scholar. Another approach to this problem is to consider the circumstantial evidence the texts provide for the approximate ages of Zhao Xiang Zi and Zhao Jia at the time when they are recorded as having been active leaders. In his 1998 article, Li Xueqin makes the point that if Zhao Jia was Zhao Xiang Zi's younger brother, he would have been a very old man by 424 B.C.E., the time of the succession struggle. Gao Ming notes that Zhao Jian Zi, Zhao Jia's father in this scenario, is reported as already being an active leader in 517 B.C.E. If Zhao Jia had been born around this time, he would have been very elderly by 424 B.C.E., and not fit to engage in a major conflict with Zhao Huan. This is no longer a problem if Zhao Jia was Zhao Xiang Zi's son. He would then have been middle-aged around 440 B.C.E., the time at which evidence from the Xinian, discussed below, implies that he was taking an active leadership role. See Xueqin, Li, “Houma, Wenxian mengshu li shuo de zai kaocha,” 167 Google Scholar; Ming, Gao, “Houma zaishu mengzhu kao,” 104 Google Scholar.
76. Li Xueqin, Qinghua Daxue cang Zhanguo zhujian (2).
77. Xueqin, Li, Qinghua Daxue cang Zhanguo zhujian (2), 186 Google Scholar.
78. The term used is sefu zuoyou 嗇夫左右.
79. This covenant is discussed in Williams, Crispin, “Ten Thousand Names: Rank and Lineage Affiliation in the Wenxian Covenant Texts,” Asiatische Studien LXIII.4, 959–89Google Scholar.
80. See n. 69 above.
81. Xueqin, Li, “Houma, Wenxian mengshu li shuo de zai kaocha,” 167 Google Scholar.
82. Xueqin, Li, Qinghua Daxue cang Zhanguo zhujian (2), 189 Google Scholar.
