Further Reﬂections on Zhi Qian’s Foshuo Pusa Benye Jing : Some Terminological Questions

: Zhi Qian ( 支 謙 , ﬂ. ca. 220–257 CE), a proliﬁc Yuezhi-Chinese translator of Indian Buddhist scriptures into Chinese, is widely known for his broad range of styles and terminology. For several decades, his translation activities and his legacy in the history of Chinese Buddhist literature have been a rich ﬁeld of research, particularly within the context of the transmission of Buddhism from India to China. In the present article, as a follow-up study to “Buddhism for Chinese readers: Zhi Qian’s Literary Reﬁnements in the Foshuo pusa benye jing, ” recently published by the authors in this journal, we offer additional reﬂections on distinctive features of Zhi Qian’s language. We focus on four unusual and interesting renderings in the Foshuo pusa benye jing ( 佛 菩 薩 本 業 經 , T. 281), namely (1) santu 三 塗 ; (2) shezui 捨 罪 ; (3) kong 空 , wu xiang 無 想 and bu yuan 不 願 ; and (4) sishi buhu 四時 不 護 . Through an analysis of these words and phrases, we discuss Zhi Qian’s translation techniques and lexical idiosyncrasies, highlighting their signiﬁcance in our understanding of the dynamics of language contact and change in the early period of the Chinese Buddhist tradition. Thus, the paper investigates some key Buddhist terms as coined by the early translators on the basis of the classical Chinese and illustrates the semantic changes of the Chinese language taking place in the period as well as inﬂuence of Buddhist regimes of knowledge.


Introductory Remarks-The Historical Interaction between Buddhism and the Chinese Language
It is well known that the influx of Buddhism into China starting from the beginning of the first century CE changed the Chinese language and Chinese thought in many ways. To absorb the systems of knowledge embedded in Buddhism, terminologies had to be created by adopting existing Chinese terms and using them with new meanings expressing Buddhist concepts. This process had a lasting impact on Chinese culture in general, especially in the fields of religion, literature, philosophy, and even usage in every-day language.
In researching this process, we consider the work of the earliest translators, such as An Shigao (安世高, fl. ca. 148 CE), Lokaks . ema (Zhi Loujiachen 支婁迦讖,, Zhi Qian (支謙, fl. ca. 220-257 CE), and others. Many of their translations have been preserved, and they can serve as materials for assessing this process of Buddhist culturalisation. These outstanding pioneers created terminologies and styles with which to communicate Buddhism to the Chinese. Most of these translators were from Central Asia and knew Sanskrit and other Prakrits, such as Pāli and Gāndhārī, that were used to promulgate Buddhism. However, they were also naturalised Chinese inhabitants who were proficient in literary Chinese and were therefore able to communicate the rich Indian heritage of Buddhism as it was conveyed from India to an interested Chinese public. by modifying an earlier translation written in more readable Chinese; 4 (2) the underlying language of Zhi Qian's works is Gāndhārī or Prakrit containing Gāndhārī elements.
These results have greatly contributed to our understanding of Zhi Qian's translation style and vocabulary. However, it should also be noted that these results are not drawn from an analysis of the Benye jing, but of his other works, such as the Da mingdu jing 大明度 經 (T. 225), the Wuliang qingjing pingdengjue jing 無量淨平等覺經 (T. 361), and the Weimojie jing 維摩詰經 (T. 474), all of which are preserved in Indic languages. In the case of the Benye jing, there is no extant Indic original. Thus, this study refers to the Tibetan version of the larger Buddhāvatam . saka, the Saṅs rgyas phal po che (D 44), translated in about 800 CE by two Indian masters, Jinamitra and Surendrabodhi, and the Tibetan translator-editor Ye shes sde, from which we can draw conclusions concerning the Indic text that underlies the Tibetan translation. 5 When necessary, we also consult Zhi Qian's other works, the Da mingdu jing, the Weimojie jing, and the Faju jing 法句經 (T. 210), and their parallels in Sanskrit, Pāli, and Gāndhārī if needed.

San Tu 三塗: The Three Times
Buddhism, with its metaphysics of the "three times", presented a highly abstract and philosophically complex concept of time, differing somewhat from the more concrete classical Chinese concept, reflected in terms connected with the concepts of generations, dynasties, years, seasons, months, and days. 6 The term san tu 三塗, literally meaning "three routes," is Zhi Qian's rendering of the Sanskrit 7 tryadhvan (the three times), i.e., past, present, and future (cf. BHSD, s.v. tryadhva(n)). Adhvan (G. adhvana; P. addhan) is a complex term that has different meanings in different contexts, for example, "way", "journey", "method", and "time" (cf. MW, s.v. adhvan), 8 of which the original meaning appears to be "road." In the Buddhist context, however, this term is generally used in connection with the concept of time (cf. BHSD, s.v. adhvan), and Chinese Buddhist translators, beginning at an early stage, have successfully conveyed this meaning. For example, the late third-century translator Dharmaraks . a (Zhu Fahu 竺法護, fl. ca. 280-308) renders the term tryadhvan as san shi 三世 (the three periods), which is the most widely accepted by later translators (Karashima 1998, p. 370, see also Kumārajīva's and Xuanzang's translations below). It is noteworthy in this regard that Zhi Qian translated tryadhvan as san tu 三塗 in the Benye jing as well as in his other works since the term san tu 三塗 is often interpreted as referring to the three unfortunate destinies (trayo durgatayah . )-namely, hell beings (naraka), animals (tiryag), and hungry spirits (preta)-because the character tu 塗 has a negative meaning, "mire". 9 Let us first look at the following examples, taken from the Weimojie jing 維摩詰 經 (T. 474), in which Zhi Qian alone understands tryadhvan in the sense of "way" (see the underlined parts): 10 In the Benye jing, the term san tu 三塗 appears six times (T. 281, 446c8, 449b3, 450b15, 450c4, 450c17), in each case corresponding to guoqu danglai jinxianzai 過去, 當來, 今現在 in Lokaks . ema's translation (T. 280, 445a19-20; t. 282, 453c11-12; T. 283, 455c22-23, 456a25-26, 456b22) 11 and to dus gsum (three times) or 'das pa daṅ|ma 'oṅs pa daṅ|da ltar byuṅ ba (past, future, and present) in the Tibetan version (D185, 256b4, 218b6, 250a1-2,251b7). 12  Tib. (D185, 251b7): dus gsum gyi ye shes brjod pa mdzad do As we will see below, Zhi Qian seems to have had access to Lokaks . ema's translation while working on the Benye jing. He appears to have deliberately chosen the term san tu 三塗 to translate tryadhvan or a similar expression instead of using the set of words guoqu danglai jinxianzai 過去, 當來, 今現在 (past, future, and future), which are semantic loanwords from atīta (gone by), anāgata (not come), and pratyutpanna (just arisen) (Feng 2004).
Thus, Zhi Qian's translation of the term adhvan, is based on the etymological meaning of the term tu 塗, road. However, other translators, such as Lokaks . ema, Kumārajīva, and Xuanzang have employed the metaphorical sense, namely time, as reflected by shi 世.
However, Zhi Qian did not always use the term san tu 三塗 for tryadhvan. In his other works, such as the Ban nihuan jing 般泥洹經 (T. 6, 184c15), the Fan moyu jing 梵摩渝經 (T. 76, 885b29, 885c5), and the Yizu jing 義足經 (T. 198, 176b6, 180b29), he used a different term, san shi 三世 (the three times) instead. The term san shi 三世 originally referred to "three generations" in the classical Chinese language, 13 but the word was used in the Buddhist translations to refer to the Indian concept of the three times of past, present, and future. 14 As such, it was widely accepted by later Buddhist translators, including Dharmaraks . a, Kumārajīva, and Xuanzang, and the meaning and usage in classical Chinese changed over time in the context of and under the influence of Buddhism.
As for the word san tu 三塗 as Zhi Qian's rendering of the three times, it is also notable that the same usage of the term san tu 三塗 occurs twice in the first chapter of the Da mingdu jing, once in the interlinear commentary (T. 225, 481a24, 481b2). 15 Modern scholars generally agree that the first chapter, Xing pin 行品 (T. 225, 478b20-482a26), 16 is not Zhi Qian's work (Lancaster 1969;Nattier 2008a). According to Zacchetti (2010, pp. 154-60), the interlinear commentary is related to Kang Senghui 康僧會 (fl. ca. 249-280) or his circle, active in Jianye 建業, the capital of the Wu dynasty. More textual evidence is needed; however, a more comprehensive and systematic investigation of the term san tu and other similar cases may shed further light on the question of who translated the Da mingdu jing.

Shezui 捨罪: To Leave the Secular Life
Karashima has shown that the underlying Indian texts used by Zhi Qian were transmitted into China in Gāndhārī or a Prakrit close to Gāndhārī and that, in the process of translation, Zhi Qian confused such vernacular or colloquial forms with those of the standard Sanskrit. 17 In the Da mingdu jing (T. 225), for instance, he often used the word shui 水 (water) where the Sanskrit equivalent isābhā (splendour, light). The Gāndhārī form of Sanskritābhā is ava, and the Gāndhārī form of Sanskritāpas (water) is also ava. Therefore, it is highly probable that here, Zhi Qian mistakenly translated Gāndhārī ava as "water", when it should be understood as "light". Specifically, the Zhi Qian's renderings for the heavens related to the second dhyāna, viz. shuiwei tian 水微天 (heaven of subtle water), wuliangshui tian 無量水天 (heaven of unlimited water), and shuiyin tian 水音天 (heaven of water sound), (T. 225, 485a12) should have been *guangwei tian 光微天 (heaven of subtle light), wuliangguang tian 無量光天 (heaven of unlimited splendour), and guangyin tian 光 音天 (heaven of radiant sound), as these terms correspond to parīttābha, apramān .ā bha, and abhāsvara, respectively (Coblin 1983, p. 244;Nattier 2008c;Karashima 2016).
The same pattern of confusion is also found in the Benye jing. Before focusing on the case of the term shezui 捨罪, we first give four examples, presumably arising from the confusion between Gāndhārī/Prakrit and Sanskrit in order to verify the possibility that the Benye jing was also translated from Gāndhārī or a Prakrit that is similar to Gāndhārī: (a) the same heaven names, shuiwei tian, 水微天, wuliangshui tian 無量水天, and shuiyin tian 水音天, occur in the first chapter; 18 (b) the word jingjin 精進 (effort) attached to the names of various buddhas, such as Ru jingjin 入精進, Xing jingjin 行精進, and Jin jingjin 盡 精進, probably resulted from the confusion between a certain Prakrit form related to vid-(to know) and the Sanskrit vīrya (G. viya, viriya, cf. GD, s.v. viriya). 19 This is supported by the Tibetan and later Chinese equivalents rig byed and zhi 智 in the corresponding passages in the larger Buddhāvatam . saka (D44, 177a1-178a7; T. 278, 418b19-c28); (c) the word shou 首, used in bodhisattvas' names, like Jingshou 敬首, Zhishou 智首, and Baoshou 寶首, reflects Zhi Qian's confusion between the Prakrit form of Skt.śrī (glory) and that of Skt. sira (head) (Karashima 1992, p. 277); (d) the word lin 林 (forest), included in the names of buddhaks . etras, reflects Zhi Qian's interpretation of the term vana as "forest." However, the term mdog (colour, appearance) is found in the corresponding passages in the Tibetan version (D44, 177a1-178a7). The Sanskrit equivalent of Tib. mdog is varn . a, and, in the sound changes from Sanskrit to Gāndhārī, the alveolar tap r [R] can undergo complete assimilation to a following consonant (Baums [2010] 2019, pp. 6-7). Therefore, we can assume that G. vana was differently interpreted by Zhi Qian and by those who produced the Sanskrit manuscripts on which the Tibetan is based. 20 We now turn to the expression shezui 捨罪, which literally means "to abandon sin". In the Benye jing, it corresponds to pravraj-, which literally means "to go forth" but refers to "the act of leaving the secular life," or rather "the lower ordination." The term occurs only once in the Benye jing, but, with its Sanskrit, Gāndhārī, Pāli, and Tibetan equivalents, it is more common in the Faju jing (T. 210). 21 Let us first take a look at the following passage from the Benye jing and its parallels in Lokaks . ema's translation and the Tibetan Buddhāvatam . saka: 22 Zhi Qian (T. 281, 447c13-14): 請求捨罪 當願衆生 得成就志 學不中悔.
[When a bodhisattva] seeks to become aśraman . a (lit. to get rid of sins), he should wish for all living beings to attain accomplishment and resolution without quitting in the middle of learning.
When a bodhisattva seeks to become aśraman . a, he thinks: may all livings beings reach and attain accomplishment without quitting or halting.
Given the context and structure, namely the bodhisattva's activity and the related wish for all beings' benefit, one can easily observe that the underlined phrases agree with one another. Yet it is still unclear why the term shezui 捨罪was chosen to translate the Indic word underlying zuo shamen 作沙門 and rab tu 'byuṅ ba, which is probably a certain Gāndhārī or Prakrit form of Skt. pravraj-(to become a monk). A possible explanation is that Zhi Qian figuratively rendered Skt. pravrajas "to abandon sins" in the light of the fact that living a secular life inevitably entails vices. Fortunately, however, the term shezui 捨 罪also appears several times in another of Zhi Qian's works, the Faju jing, and its use there can shed further light on this issue, as the Faju jing is currently available in various Indic versions, including Gāndhārī. Let us examine the following two verses and their parallels: Here, it is notable that in the first of the verses above, (a) shezui 捨罪 is used in the sense of "leaving the secular world", corresponding to Skt. pravraj-(e.g., Skt. pravrajya, pravrajita; P. pabbajjā, pabbajita; Tib. rab tu byuṅ ba), while in the second verse, (b) shezui directly refers to its literal meaning, as seen in the Pāli equivalent in the second verse (d) pāpañ ca bāhetvā (keeping away from sins). Another noteworthy point is that the Gāndhārī form of Skt. pāpa (evil, sin) is pava, as found in the phrase (c) baheti pavan . a in the second verse (see also GD, s.v. pava). In the sound changes from Sanskrit to Gāndhārī, the clusters of plosives followed by r [R] are generally unchanged (Baums [2010] 2019, pp. 7-8); accordingly, the regular form of Skt. pravrajin Gāndhārī is pravaj(or -y)-(e.g., pravaja, pravayadi). 23 However, irregular changes are also attested, for instance, pavaja, pavaje, parvaja for pravaja and pavayadi, parvayadi, padaïdi for pravayadi (cf. GD, s.v. pravaja, pravayadi). Therefore, it is possible that the supposed Indic original form that Zhi Qian translated was pavaj(or -y)-(e.g., pavaja, pavayadi). Furthermore, he may have rendered it as shezui 捨罪 by adopting the meaning of G. pava (evil, sin). However, it is still unlikely that Zhi Qian failed to understand the meaning of the term since he also used the expression zuo shamen 作沙門 (to become aśraman . a) in the Benye jing (T. 281, 447c19-20) shortly after the citation above (see the underlined passages): Zhi Qian (T. 281, 447c19-20): 已作沙門 當願衆生 受行佛意 開導天下 Having become aśraman . a, he should wish for all living beings to receive and practice the disposition of the Buddha, which guides beings in the world.
When a bodhisattva becomes aśraman . a, he thinks: may all living beings, when they becomeśraman . as, save others as the Buddha did.
We can see here that Zhi Qian adopted Lokaks . ema's rendering zuo shamen 作沙門 and added the particle yi 已, indicating the completion of an action. It is clear that yi zuo shamen has the same meaning here as shezui in the first verse from the Faju jing that is quoted above. Some other examples can be found in Zhi Qian's translations of the Weimojie jing and the Faju jing: UvTib. 232b3-4: bzod pa dka' thub mchog ste bzod pa ni || myaṅan 'das pa mchog ces saṅs rgyas gsuṅ || rab tu byuṅ ba gźan la 'tshe ba daṅ || gźan la gnod pas dge sbyoṅ ma yin no || * No parallels in DhpG.
Two points are noteworthy here: (1) Zuo shamen in Zhi Qian's translation of the Weimojie jing was later replaced by Kumārajīva with chujia 出家, literally meaning "to leave home", and this is also the translation adopted by Xuanzang. 24 Regarding the expression zuo shamen 作沙門, here, it is uncertain whether Zhi Qian learned it from Lokaks . ema's translation, as seen in the Benye jing (T. 280, 451c25-27), or whether he borrowed it from the older translation of the Vimalakīrtinirdeśa by Yan Fotiao 嚴佛調 (fl. ca. 180 CE), now lost, on which Zhi Qian's translation is perhaps based (Karashima 2016). Yet zuo shamen does not appear in the Fajing jing 法鏡經 (T. 322), the only surviving text attributed to Yan Fotiao (Nattier 2008b, p. 177).
(2) In Zhi Qian's translation of the Faju jing, the two terms, shezui and zuo shamen, appear together. Since there is nothing in this verse in any of the texts that suggests the literal meaning of the word shezui 捨罪 (to abandon sin), we can assume that shezui is used in the sense of pravraj-, and thus that the phrase shezui zuo shamen 捨罪作沙門 in the third pāda corresponds to the Sanskrit expression pravrajito . . .śraman . o bhavati (P. pabbajito . . . saman . o hoti; BHS: pravrajito . . .śaman . o hoti; Tib. rab tu byuṅ ba . . . dge sbyoṅ . . . yin no).
In sum, we can learn from the examples examined so far that Zhi Qian intentionally rendered the Gāndhārī (or Prakrit) form of pravrajas shezui 捨罪 even though he knew the expression zuo shamen 作沙門, which appears in the Benye jing and his other works. Further, the underlying Indic form of shezui may be pava-, from which Zhi Qian could derive the meaning of "evil, sin" and not merely his figurative interpretation of "leaving a secular life (or becoming a monk)" as "abandoning sins." In Indian hermeneutics, as practised by Hindu and Buddhist writers and commentators alike, one would explain words with etymologies that were sometimes not at all in accordance with scientific etymologies. With pāpañ ca bāhetvā (doing away with sin) as an etymology for pravrajyā is definitely such a case, and one may indeed call the procedure a folk-etymology or pseudo-etymology. However, a better term might be "intentional etymology", as in its Buddhist context, it better expresses the intention of the word pravarajyā, though its (true) etymological reference is "leaving (home)".
Thus, we would like to contend that Zhi Qian's choice of the Chinese equivalent shezui is based on this principle of Indian hermeneutics to convey the Buddhist intention of the expression rather than a misunderstood reading of the Gandhārī word pavajor Pāli pabbajas etymologically connected to pāpa. 25 In addition, there is a notable case found in the Fo kaijie fanzhiaba jing 佛開解梵志阿颰 經 (T. 20), which is traditionally attributed to Zhi Qian but not included in the list of his genuine works prepared by Nattier (2008b, pp. 177-78;2019, p. 821). In this sūtra, the term shezui 捨罪 is used in the sense of pravrajas follows: After the expression (a) yu zuo shamen 欲作沙門, "to desire to become a monk", it seems reasonable to interpret the following phrase (b) nai ke shezui 乃可捨罪 as "then, [one] may become a monk" and not as "then [one] may abandon sins." The Taishō edition as well as the second Koryȏ edition on which the Taishō edition is based contain the term shezui 捨 罪, whereas the Sixi zang 思溪藏 (1126-1138), the Puning zang 普寧藏 (1277-1290), and the Jingshan zang 徑山藏 (1589-1712) read shejia 捨家 (abandon home) as indicated in the Taishō note (see above). Thus, the correct, or rather original, reading here is presumably shezui 捨 罪, as presented in the second Koryȏ edition, even though the term shejia 捨家 sounds more natural in the given context. It is also likely that this modification was made by the editor(s), who did not know that the term shezui referred to pravrajin Zhi Qian's terminology or who thought that the term does not adequately convey the original meaning of the text. Further, although this occurrence of a single term, shezui, is limited evidence, it may support the supposition that the Fo kaijie fanzhiaba jing is Zhi Qian's genuine work or a text related to him. 26 2.3. Kong 空, Wu Xiang 無想, and Bu Yuan 不願: Emptiness, Signlessness, and Wishlessness The word kong 空 is a good example of the Buddhist influence on Chinese terminologies, particularly in terms of their semantic changes. It used to simply mean "hollow" or "nothing", similar to wu 無, but its meaning was later extended to reflect the concept of sūnyatā with much greater philosophical depth. 27 Even as late as the Dang dynasty, we find, parallel with the Buddhist connotation, what sems to be the original meaning of the word kong, where the term sikong 司空 is used for a title of Amoghavajra (Bukong 不空, 704-774), meaning "Imperial craftsman" or "Imperial architect" (cf. Hucker 1985, p. 450, nr. 5687). The fact that the word is originally associated with craftmanship is also evident in the etymology of the character kong 空. 28 The character kong may be a phonetic loan to represent the concept of "empty", but the old Sino-Tibetan words kong, etc., seem also to connote "boring a hole", etc., as documented, e.g., in Naxi, Tamang, and Western Tani, underpinning the meaning expressed by the character. 29 In pre-Buddhist Chinese, the two characters xiang 想, usually equivalent to sam . jñā, and xiang相, most often used to represent nimitta, have often been used in the same sense from the very early period of Chinese Buddhist translation (cf. Karashima 1998, pp. 474-75;Karashima 2010, pp. 517-18;Vetter 2012, pp. 116, 210-11). Rather, the two terms represent a case of semantic continuity, as xiang想 would in early texts mostly mean "to perceive, conceptualise", while xiang相 would carry the connotation of "mutually, each other, meeting with" but also "to see, to perceive". The classical terms were thus the natural choices for the early translators, though xiang 相 would absorb the special Buddhist meaning of "quality, distinguishing mark (in perception)," that is, nimitta, as a technical term. The terms sam . jñā and nimitta are also interrelated in Buddhist epistemological theory-sam . jñā being defined as nimittagraha (cf., e.g., sam . jñā nimittodgrahan .ā tmikā in the Abhidharmakośa 1.14 (Pradhan 1967, p. 10)). The two xiangs are, though, often intermingled in Chinese translations, at least as we have them handed down.
The etymological connotations of pran . idhi and pran . idhāna would be "place at, in, before" and in the Buddhist use, evidently to "place your mind" at "some object". Thus, in the Mahāyāna context, the word acquires the meaning of "wish" but with both positive and negative connotations. First, it denotes the bodhisattva's wish or promise to attain bodhi, more or less a synonym of bodhicitta. In the other use, as is relevant here in the case of trīn . i vimoks . amukhāni (three doors to liberation), the pranidhāna still means wish but refers to wishful thinking to be done away with, similar to that of constructing qualities (nimitta) of objects and then wanting them, though they are of course empty, constructed, and illusory. The meaning of yuan 願 before it was "invaded" by the special and well-defined connotations of Buddhism, is gleaned from a number of classical texts as "to long for (often persons)" but also "deep desire" in general. In this way, the case of yuan 願 is that of semantic continuity but is recreated as (two!) Buddhist technical terms; yuan 願 acquires a much more defined philosophical and religious connotation when adopted by Lokaks . ema, Zhi Qian, and the other translators.
The words kong 空, wu xiang 無想, and bu yuan不願are Zhi Qian's translations, often in a set of three (空無想不願) forśūnyatā (emptiness), animitta (signlessness), and apran . ihita (wishlessness), which represent the three doors to liberation (trīn . i vimoks . amukhāni). According to Zacchetti (2008, p. 472), this string of three words, particularly the term bu yuan, is rare, mostly only occurring in the texts dating back to the Three Kingdoms period. Indeed, the Parthian An Shigao, who is the first historical figure of Chinese Buddhism active in Luoyang (洛陽) during the Later Han dynasty, uses kong 空, bu xiang 不想, and bu yuan不 願 forśūnyatā, animitta, and apran . ihita (Vetter 2012, p. 224). Soon afterwards, however, the terms bu xiang 不想 and bu yuan不願 seem to have been replaced by wu xiang 無想 and wu yuan無願, as can be seen in the works of Lokaks . ema and Dharmaraks . a (Karashima 1998, pp. 475-76;Karashima 2010, pp. 521-22), and wu xiang and wu yuan are now generally accepted for animitta and apran . ihita (cf. DDB, s.v. 無相, 無願, 空無相無願). In this regard, it is notable that Zhi Qian, who was later than Lokaks . ema, still used bu yuan for apran . ihita, but adopted wu xiang for animitta at the same time. The set of words kong, wu xiang, and bu yuan is found only once in the Benye jing but more often in Zhi Qian's other works. 30 For instance, it occurs five times in the Da mingdu jing (T. 225, 478c3, 478c11, 479a7-8, 496c8, and 505b27), 31 twice in the Foshuo weimojie jing (T. 474, 526b18-19 and 526b21), and once each in the Fanmoyu jing 梵摩渝經 (T. 76, 884b17) and the Faju jing (T. 210, 561b13). Let us look at the following sentences that include this set of words: (a) The words that he often recited can all be trusted by people, so there is no one who does not respect or follow him. The action produced by his body is nothing but pure; regarding all sūtras that he looked into, there was nothing left to understand. (b) He one-pointedly controlled his mind, (d) contemplated the enlightenment, and (c) was attentive in entering into meditation.
Here, even though Zhi Qian's translation does not completely agree with Lokaks . ema's, the two Chinese versions seem related to one another to some extent. That is to say, (a) yanjian xinyong 言見信用, (b) jiangxin zhengyi 降心正意, and (c) shenian ruchan 攝念入禪 in T. 280 correspond to the (a) duosuo churu yuzhe, ren jie xinyong zhi 多所出入語者, 人皆信用 之, (b) yixin jiangyi 一心降意, and (c) shenian ruchan 攝念入禪 in T. 281, respectively. There is, however, no counterpart for the terms (d) kong 空, wu xiang 無想, and bu yuan 不願 in Lokaks . ema's translation, apart from the word xiao 曉 in the phrase siwei mingxiao 思惟明 曉 (to be awakened to what he thinks). Thus, it seems safe to say that a counterpart of the expression, kong, wu xiang, and bu yuan, is only found in the Tibetan translation and not in Lokaks . ema's translation. Since the Tibetan translation has the corresponding set of words, stoṅ pa ñid, mtshan ma med pa, and smon pa med pa, and since it is impossible to derive the concepts kong, wu xiang, and bu yuan without referring to sources other than Lokaks . ema's translation, it is likely that Zhi Qian had access to an Indic original, although we do not know what version it was. We can also assume that, when Zhi Qian worked on the Benye jing, he not only consulted an already existing translation as we will see below, but he also consulted, at least to some degree, the Indic original.

Sishi Buhu 四時不護: Four Methods of Winning [People] Over
Finally, we will present an additional terminological issue found in the Benye jing in the Taishō edition. The phrase sishi buhu 四時不護, which literally means "four periods of non-guards", is found nowhere else in the entire Taishō edition. No variant reading is given in the Taishō edition. However, this expression appears as sishi buhu 四事不護 (four kinds of non-guards) in the corresponding passage in the second Koryȏ 高麗 edition (13th century CE) as well as in all other ancient printed editions available to us, namely the Qisha zang 磧砂藏 (completed in 1322), the Hongwu nan zang 洪武南藏 (completed in 1402), and the Qing zang 清藏 (completed in 1738). 32 Furthermore, another important modern edition, the Zhonghua dazangjing 中華大藏經, has the same phrase, sishi buhu, in the corresponding place without indicating any variant readings. Therefore, the same reading is probably found in all of the editions collated in the Zhonghua dazangjing, namely the Sixi 思溪 edition (completed in 1138), the Puning zang 普寧藏 (completed in 1290), the Yongle nan zang 永樂 南藏 (completed in 1420), and the Jingshan zang 徑山藏 (completed in 1712). 33 More notably, the phrase sishi bu hu 四事不護 is found in a parallel passage in the Foshuo dousha jing 佛兜 沙經 (T. 280): Zhi Qian (T. 281, 446c14-16): 悉爲我現佛行、佛智、佛神、佛力、佛定、無量變 化隨時。四時不護、四無所畏、十八不絶、一切敏慧、無上道徳, 衆事敷露。 He shows us the Buddha's practice, the Buddha's knowledge, the Buddha's supernormal power, the Buddha's ability, the Buddha's samādhi, and (the Buddha's) immeasurable transformation as the situation demands. He also presents all qualities [of the Buddha, namely], the four kinds of non-guards, the four forms of fearlessness, the eighteen distinctive abilities, all kinds of wisdom, and the utmost merit.
Lokaks . ema (T. 283, 445b8-15) Two inferences can be made from these passages: (1) It is very probable that the phrase sishi buhu 四時不護 in the Benye jing in the Taishō edition is not a misinterpretation but an editorial error for the phrase sishi buhu 四事不護. It is likely that the error was the result of confusing the characters shi 事 and shi 時 (perhaps influenced by the compound suishi 隨時, underlined in the previous phrase), even though the words are pronounced in different tones. This case is a good example that highlights the importance of directly consulting multiple editions in the study of Chinese Buddhist texts and not just relying on the given apparatus. 34 (2) According to Karashima, the phrase sishi buhu 四事不護 is Lokaks . ema's misinterpretation of catursam . grahavastu (four means of attracting people) (Karashima 2011, p. 464), which is more often translated as si shefa 四攝法 (four methods of winning (people) over). Since there is no counterpart in the corresponding passage in the Tibetan Buddhāvatam . saka (D44, 176b2-7), it can be assumed that Zhi Qian directly consulted Lokaks . ema's translation here, which is now known as the Foshuo dousha jing, and that he borrowed the expression sishi bu hu 四事不護 from there even though it was initially misinterpreted. Indeed, it is difficult to explain why Lokaks . ema translated sam . graha, which means "seizing, grasping" as buhu 不護 (not guarding, not being on guard), 35 and why Zhi Qian subsequently adopted this translation without modification. 36 This is an example of how investigating a mistranslation may lead to further topics of discussion: in this case, the textual relationship between the Benye jing and the Foshuo dousha jing and Zhi Qian's translation procedure.

Conclusions
This paper is a follow-up to the authors' previous study (Han et al. 2021) dealing with Zhi Qian's stylistic embellishments in the Foshuo pusa benye jing (佛菩薩本業經, T. 281). Zhi Qian played a significant role in the introduction and spread of Buddhism in China during the Three Kingdoms period (ca. 220-280 CE), and the Benye jing, one of his most influential pieces among his works, clearly documents his unique translation style and techniques. 37 In the present paper, we examined Zhi Qian's four particular renderings in the Benye jing, viz., (1) santu 三塗, (2) shezui 捨罪, (3) kong 空, wu xiang 無想 and bu yuan 不 願, and (4) sishi buhu 四時不護, and discussed their semantic interpretations and their later reception in the history of Chinese Buddhist translation. Specifically, it was shown that the terms santu 三塗and shezui 捨罪 are Zhi Qian's unique renderings of tryadhvan (the three times) and pravraj-(to become a monk), respectively, and that the two cases are examples of Zhi Qian's lexical idiosyncrasies that may shed further light on the questions of how to identify his genuine works. The set of words kong 空, wu xiang 無想, and bu yuan 不願 allowed us to see the dynamic process of language change, or rather evolution, during the formative period of Chinese Buddhism, and Zhi Qian's place therein. The term sishi buhu 四時不護, probably an editorial error only occurring in the Taishō edition, demonstrated the necessity of consulting multiple editions in the study of Chinese Buddhist texts. The last two cases, kong, wu xiang, bu yuan, and sishi buhu, also showed how the Benye jing was produced, that is, through the use of an Indic original source but also with reference to Lokaks . ema's pre-existing translation.
In this paper, we also illustrated some semantic changes of the Chinese language taking place in the period as well as the influence of Buddhist regimes of knowledge. The semantic continuities, or disruptions, between classical Chinese and Buddhist Chinese seem to be quite understudied though the topic potentially has great importance for understanding the Chinese history of language as well as the history of Chinese thought. In view of this, as the authors of this paper, we would contend that the semantic changes of the Chinese language deserve further study. For the spread of Buddhist thought in China, see (Braarvig 2012). On Chinese lexicography and on language diffusion and dominant/dependent languages in general, see (Braarvig 2018b(Braarvig , 2018a, respectively. 2 The term "sinicisation" as used here is concerned with the early Chinese Buddhist translators' various attempts to produce more readable translations for the readers of Chinese during the Eastern Han dynasty and early Weijin period. For the meaning and context of the term "sinicisation" used in our study, see (Han et al. 2021, pp. 11-12n6). 3 Notably, a critical reading of the Benye jing reveals its similarity to the poetic prose fu 賦 of classical Chinese literature. Fu, which is generally translated as "rhapsody", "epic poem", or "poetic exposition", was the dominant literary genre of the Han dynasty, mainly composed and recited at the imperial courts or among the cultured people of the time. According to Knechtges, it can be characterised as "a mixture of prose and verse, relatively long lines, parallelism, elaborate description, dialogue, extensive cataloging, and difficult language" (Knechtges 1976, p. 13). Indeed, the Benye jing represents itself as a highly polished piece of Buddhist literature, in which one can easily find such characteristics, most notably, the insertion of verse into a prose structure, which is one of the distinguishing formal features of the fu. We plan to investigate this point further in a future publication. On the historical importance and literary characteristics of the Chinese literary form, fu, in the context of classical Chinese literature, see (Knechtges 1976;Kern 2003). 4 However, the other terminologies of the earlier translations were also learned creations and as such, were very far from vernacular and colloquial language, under which category they have sometimes been subsumed by sinological scholarship, notwithstanding the bisyllabic nature of Buddhist terms. The reason for the bisyllabic terms was probably a means to catch the Buddhist semantics that were foreign to the Chinese conceptual world at the time, where the monosyllabic classical language could not cope with novel concepts without modifications. 5 According to the Zhiyuan fabao kantong zonglu 至元法寶堪同總, a comparative Sino-Tibetan catalogue compiled by Qing Jixiang 慶吉祥 in the late 13th century, the Tibetan Buddhāvatam . saka was translated from the Chinese translation. Yet, recent studies have shown that it was not based on the Chinese version but on the Sanskrit original. For more details on the textual history of the Tibetan Buddhāvatam . saka, see (Hamar 2007, pp. 165-67). 6 However, the word shi 時 for Skt. samaya, etc., (e.g., ekasmin samaye ...) is retained in Buddhism but is never used for the Buddhist abstract concept of time and rather denotes a day or a point of time.
Zürcher mentions the term san tu 三塗in his study on the Buddhist influence on early Daoist notions and terminology, but he treats it in the context of the three evil destinations not as the three times. See (Zürcher [1980(Zürcher [ ] 2013. 10 All citations in this article from the Taishō edition have been re-punctuated by the authors, based on their reading of the texts. 11 It is questionable whether the word san shi 三世 was used by Lokaks . ema. According to Nattier (2008b, p. 176), two texts, the Daoxing banruo jing 道行般若經 (T. 224) and the unrevised prose potion of the Banzhou sanmei jing 般舟三昧經 (T. 418), are identified as Lokaks . ema's core texts, while the other texts attributed to him are probably related to his circle and not to Lokaks . ema himself. The term tryadhvan occurs three times in the As . t . asāhasrikā Prajñāpāramitā (Vaidya 1960, pp. 75, 95), but there is no counterpart in the corresponding parts of the Daoxing banruo jing, the Lokaks . ema's translation of the As . t . asāhasrikā Prajñāpāramitā (Karashima 2011, pp. 145n14, 183). Furthermore, the term san shi 三世 is found only once in the verse potion of the Banzhou sanmei jing (T. 418, 911c28) and in the Pusa shizhu xingdao pin (T. 283, 455b11), both texts that may have been revised or refined by his circle. 12 There is no counterpart for T. 281, 450c4 and T. 282, 456a25-26 in the Tibetan version. 13 Some examples follow: (1) 陪臣執國命, 三世希不失矣。"When the subsidiary ministers of the great officers hold in their grasp the orders of the state, as a rule the cases will be few in which they do not lose their power in three generations" (Lunyu 論語/ Jishi 季氏/ 2, ICS Lunyu: 16.2/45/26, tr. James Legge). (2) 去國三世, 爵祿有列於朝, 出入有詔於國, 若兄弟宗族猶存, 則反告於宗 後。"(But) if he (or his descendants) has been away from the state for three generations, and if his dignity and emoluments be (still) reckoned to him (or his representative) at the court, and his outgoings and incomings are announced to the state, and if his brothers or cousins and other members of his house be still there, he should (continue to) send back word about himself to the representative of his ancestor. (Even) after the three generations, if his dignity and emoluments be not reckoned to him in the court, and his outgoings and incomings are (no longer) announced in the state, it is only on the day of his elevation (to official rank) that he should follow the ways of his new state" (Liji 禮記/曲禮下 Quli II/ 85, tr. James Legge). (3) 越人三世其 君, 王子搜患之, 逃乎丹穴。"The people of Yue *three times in succession killed their ruler, and the prince Sou, distressed by it, made his escape to the caves of Dan" (Zhuangzi 莊子/ Rangwang 讓王/ 3, ICS Zhuangzi: 28/82/1, Harvard-Yenching Zhuangzi Yinde: 77/28/15, tr. James Legge). *The underlined san shi 三世 should be translated as "three generations" not as "three times". Legge's translation is incorrect. (4) 今王發明惠, 諸侯之士來歸義者, 今使復之三世, 無知軍事。"Now, if Your Majesty will issue a favourable proclamation to the effect that those soldiers of the various feudal lords who will come and submit, will be granted exemption for three generations, without hearing anything of military affairs" (Shangjun Shu 商君書/Laimin 來民/ 4, ICS Shangjunshu: 15/20/1, tr. J. J. L. Duyvendak). For more references, cf. Chinese Text Project (https://ctext.org) s.v. 三世 (accessed on 1 June 2021). 14 For a detailed discussion of the concept of time in Indian context, see (Balslev 1983). 15 The term occurs once more in the latter part (Chapter 7) of the Da mingdu jing (T. 225, 488a9), but it is not related to this argument since it refers to "the three evil destinies," which is a more frequent usage of the term san tu 三塗. 16 Skt. sarvākārajñatācaryā prathamah . parivartah . (Chapter 1: Practice of the knowledge of all aspects). 17 See (Karashima 2006, p. 363;Karashima 2016;Nattier 2007a, pp. 115-17 ( 19 Nattier suggests veda for the reconstruction of the original word derived from vid- (Nattier 2007a, pp. 116-17). The sound change from G. -d-to Skt. -y-is common since both are palatals. OIA -e-is also changeable to -i-in M, AMg, JM. Cf. (Pischel 1981, p. 90 For the attribution of the Faju jing 法句經 (T. 210) and the process of its translation, see Zhi Qian's Faju Jing Xu 法句經序, "Preface to the Translation of the Dharmapāda" see (Cheung and Lin 2006, pp. 58-63). 22 All translations in this article are by the authors. 23 The palatal plosive j (é) is weakened to y (j) when surrounded by vowels (Baums [2010] 2019, p. 4).

24
The term chujia 出家 is a new expression imported into the Chinese language through the influence of Buddhism (Guang 2012, p. 160). 25 That such hermeneutics are employed as very conscious processes when translating Buddhist litarature into Chinese, and, indeed, into Tibetan, is well documented in Chinese and Tibetan biingual lexicography (Braarvig 2018b). See also (Braarvig et al. 2017) for an example of Indian monolingual synonym lexicography and (Kahrs 1998) for the Indian tradition of etymological analysis. 26 Park, too, suggests in his posthumous work that the Fo kaijie fanzhiaba jing may be a genuine translation of Zhi Qian (Park 2012, pp. 28-36), but his argument is not widely accepted by other scholars (see, for an example, Nattier 2019).

27
For more discussion of such semantic development in the Chinese lexicon, including the term kong 空, see (Sun 2006;Guang 2012). 28 Perhaps it referred to the holes drilled by craftsmen or to the spaces created inside houses. Compare this character with xue 穴 (cave, hole) for meaning and kong 工 (tool, craft) for pronunciation. 29 Cf. Sino-Tibetan Etymological Dictionary and Thesaurus (STEDT: https://stedt.berkeley.edu/~stedt-cgi/rootcanal.pl/etymon/ 820) (accessed on 29 June 2021). 30 On Zhi Qian's authentic translations, see (Nattier 2008b, pp. 177-78;2019, pp. 820-21). 31 However, the following two points should be noted. First, the order of the three words is different, viz. 空-不願-無想. Second, the first three occurrences are found in the first chapter of the Da mingdu jing, which is not considered to be Zhi Qian's genuine work by modern Buddhist scholars, such as Lancaster, Karashima, and Nattier. This is because regarding wording and sentence style, the Da mingdu jing can be divided into two parts: (1) from the beginning to the end of Chapter 1, including the interlinear commentary and the use of unusual vocabulary with an irregular prosodic style that appears not to be ascribed to Zhi Qian; (2) the remainder of the sūtra, Chapters 2-27, written in elegant four-character prosody without interlinear commentary, which is considered to be Zhi Qian's authentic translation. For more discussion on the attribution of the Da mingdu jing, see (Lancaster 1969;Katsuzaki 1985;Nattier 2008a). 32 The references are as follows: (1)  The second Koryȏ edition may be a feasible option for additional textual witnesses since it is fully digitalised and made accessible online (digital photos and e-text available from the Academy of Buddhist Studies at Dongguk University: http: //abchome.dongguk.edu) (accessed on 31 May 2021). Additionally, the Zhonghua dazangjing, compiled in Beijing between 1984 and 1988, seems to be a good option, as it contains critical apparatuses from various pre-modern editions at the end of each juan 卷. However, it should be noted that the reliability of the apparatus is not completely guaranteed. For more discussion of the Dazangjing Studies from a textual critical perspective, including the history of various editions and relationships among them, see (Zacchetti 2005, pp. 74-140). 35 The term si shi 四事 probably corresponds to Skt. catur . . . vastu.