出三藏記集 8.6

8.6

天竺沙門曇摩蜱執本。佛護為譯。對而撿之。慧進筆受。與放光光讚同者。無所更出也。其二經譯人所漏者。隨其失處稱而正焉。其義異不知孰是者。輒併而兩存之。往往為訓其下。凡四卷。其一經五卷也。譯胡為秦。有五失本也。一者胡語盡倒而使從秦。一失本也。二者胡經尚質。秦人好文。傳可眾心非文不合。斯二失本也。三者胡經委悉至於嘆詠。丁寧反覆。或三或四。不嫌其煩。而今裁斥。三失本也。四者胡有義記正似亂辭。尋說向語文無以異。或千五百刈而不存。四失本也。五者事已全成。將更傍及。反騰前辭已乃後說而悉除此。五失本也。然般若經。三達之心覆面所演。聖必因時時俗有易。而刪雅古以適今時。一不易也。愚智天隔聖人叵階。乃欲以千歲之上微言。傳使合百王之下末俗。二不易也。阿難出經去佛未久。尊者大迦葉令五百六通迭察迭書。今離千年而以近意量截。彼阿羅漢乃兢兢若此。此生死人而平平若此。豈將不知法者勇乎。斯三不易也。涉茲五失經三不易。譯胡為秦。詎可不慎乎。正當以不開異言。傳令知會通耳。何復嫌大匠之得失乎。是乃未所敢知也。前人出經。支讖世高。審得胡本難繫者也。叉羅支越。斵鑿之巧者也。巧則巧矣。懼竅成而混沌終矣。若夫以詩為煩重。以尚書為質朴。而刪令合今。則馬鄭所深恨者也。近出此撮欲使不雜推經言旨。唯懼失實也。其有方言古辭。自為解其下也。於常首尾相違句不通者。則冥如合符。厭如復折。乃見前人之深謬。欣通外域之嘉會也。於九十章蕩然無措疑處。毫芒之間泯然無微疹。已矣乎。

The Indian śramaṇa Dharmapriya held the original text, and Buddhapāla translated it. Comparing and collating against it, Huijin served as scribal recorder. Where it was identical with the Fangguang and the Guangzan, nothing new was produced. Where the translators of those two sūtras had made omissions, the passages were emended and corrected accordingly at the points of error. Where the meanings differed and it was not known which was correct, both were retained together, often with explanatory notes provided below. In all there are four fascicles; one sūtra makes five fascicles. In translating from Hu into Qin, there are five losses of the original. First, the Hu syntax is completely inverted, yet it is made to conform to Qin: this is the first loss of the original. Second, Hu sūtras value unadorned substance, while the people of Qin favor literary refinement; to convey them so as to please the minds of the multitude, unless they are refined they will not suit: this is the second loss of the original. Third, Hu sūtras are minute and detailed; when coming to praises and eulogies, they are deliberate and repetitive, [saying things] three or four times without minding the tediousness, but these are now trimmed and removed: this is the third loss of the original. Fourth, Hu texts have expository notes that seem just like disorderly phrases; when one examines what has just been said, the text does not differ, yet sometimes a thousand or five hundred [words] are pruned away and not retained: this is the fourth loss of the original. Fifth, after an account is fully completed, when it is about to touch on something else collateral, it recurs back over the previous phrasing before stating the subsequent, but all of this is eliminated: this is the fifth loss of the original. Yet as for the Prajñā Sūtra, it was expounded by the three-fold attained mind and the face-covering tongue [of the Buddha]. The Sage invariably accorded with the times, but the customs of the times change; to prune the elegant antiquity to suit the present day is the first difficulty. Fools and the wise are separated as heaven from earth, and the Sage is impossible to scale; yet wanting to take the subtle words from over a thousand years ago and convey them so as to suit the debased customs under a hundred kings is the second difficulty. When Ānanda brought forth the sūtras, it was not long after the Buddha had passed; the Venerable Mahākāśyapa had five hundred [arhats] with the six supernatural knowledges examine them in turn and record them in turn. Now, separated by a thousand years, we measure and sever them with our immediate notions; those arhats were so cautious and apprehensive, yet these men of birth and death are so casual and complacent—is it not that those ignorant of the Dharma are daring? This is the third difficulty. Venturing into these five losses and undergoing the three difficulties, in translating Hu into Qin, how could one not be prudent? One should merely aim to prevent foreign words from remaining uncomprehended, conveying them so that their common sense is known, that is all. Why further criticize the merits and defects of the great masters? This is something I would never dare to claim to understand. Among earlier men who brought forth sūtras, Zhi Qian and Shigao accurately grasped the Hu originals, which are difficult to equal; Mokṣala and Zhi Qian were skilled at shaping and carving. Skilled they were indeed, yet I fear that once the openings were bored, Chaos met his end. As for considering the Odes redundant and the Book of Documents crude, and pruning them to suit the present day—this is what Ma [Rong] and Zheng [Xuan] deeply lamented. Recently, in bringing forth this excerpt, I wished to keep it unmixed and to fathom the doctrinal purport of the sūtra, fearing only to lose the truth. Where there were dialect words or archaic phrasing, I provided explanations below on my own. Where previously the beginning and end were mutually contradictory and the sentences did not make sense, now they matched silently like two halves of a tally, fitting together as if rejoined after being broken. Thereupon one saw the deep errors of the predecessors and rejoiced at the auspicious encounter of communicating with the foreign realms. In the ninety chapters, there is vastness with no place left for doubt; down to the finest hair's tip, all is obliterated without the slightest flaw. That is all.