Exploring the Intricate Usage and Interpretation Issues of “體” (tǐ) in Xuanzang’s Translation of Abhidharmakośabhāṣya

Abstract

This study delves into the intricate usage and interpretation issues of the Chinese term “體” (tǐ) in Xuanzang’s translation of the Abhidharmakośa (AKBh[X]) by providing a Sanskrit-Chinese comparative investigation. Xuanzang’s translations are pivotal in understanding certain Abhidharma scriptures, as some of them are the sole complete versions available. This study focuses on the term “體” in AKBh[X], evaluating its usage in relation to 16 corresponding Sanskrit equivalents and the instances where Xuanzang introduced “體” without a Sanskrit equivalent. The analysis uncovers translation errors, potential misinterpretations, and the lack of clarity in certain contexts, emphasizing the need for readers to be cautious and consult additional sources for a comprehensive understanding of his translations.

1. Introduction

Xuanzang (602–664) was a prominent translator in the history of Chinese Buddhism. His contributions were groundbreaking, evident in the extensive translations he and his team accomplished, encompassing various Buddhist scriptures from different traditions, such as Abhidharma, Prajñāpāramitā, and Yogacāra texts. Xuanzang’s translations1 also set new standards for translating Buddhist classics, rectifying numerous errors made in previous translations. His translations are categorized as “new translations”, while translations made prior to his era are referred to as “old translations”. Compared to the old translations, Xuanzang’s works stand out for their faithfulness and accuracy to the original texts, and a more consistent rendering of the translated language (Kuwayama and Hakamaya 1991, pp. 301–4).

The importance of Xuanzang’s translations to modern scholars is that some Buddhist scriptures, particularly certain Abhidharma texts, such as *Abhidharmanyāyānusāraśāstra, only exist in Chinese translations by Xuanzang. As a result, our knowledge of these texts depends entirely on his translations. However, accurately comprehending his translated terms is challenging due to the lack of parallel texts for comparison. Despite this challenge, grasping Xuanzang’s translation choices is crucial to correctly understanding these scriptures. Additionally, Xuanzang’s translations had a significant impact on the East Asian Abhidharma tradition, leading to the neglect of older translations in favor of his work, and his disciples extensively annotated his translations of the Abhidharmakośa (AKBh) (Willemen et al. 1998, pp. 136–37).

However, there are currently limited specialized studies on Xuanzang’s translations. Some representative studies, such as Sakurabe (1954), Kuwayama and Hakamaya (1991), Chen (2000), Wang (2014), Hirakawa et al. (2016), Delhey (2016), and Nehrdich (2023) have been conducted, utilizing methods including philology, grammar, linguistics, and textual studies. It is worth noting that although Hajime Sakurabe’s article (1954) explored the term “體” in AKBh[X], it did not conduct a comprehensive investigation, probably due to limitations in retrieval tools at that time. And its focus was mainly on the Sanskrit term svabhāva that was translated as “體”, overlooking other important instances and Xuanzang’s own additions of the term. Therefore, this study aims to complement the deficiencies in Sakurabe’s research by conducting a philological study specifically focusing on the term “體” in Xuanzang’s translation of AKBh (AKBh[X]).

The study aims to examine the correspondence between “體” and each corresponding Sanskrit term, not solely focusing on svabhāva. Furthermore, it will examine the cases where Xuanzang added the term in the absence of a Sanskrit equivalent. By thoroughly examining these occurrences, the study seeks to evaluate the usage of “體” in AKBh[X] and identify the issues or complexities that might arise in his rendering.

Moving forward, two questions are to be addressed: why the term “體” was selected as the primary focus of this study, and why AKBh was chosen as the main research text.

Firstly, “體” was chosen because it appears extensively in Xuanzang’s translations. In AKBh[X], it occurs 437 times, whereas in the translation by Paramārtha (AKBh[P]), it appears only 117 times. This notable discrepancy raises questions about Xuanzang’s preference for this translated term. Furthermore, “體” in AKBh[X] corresponds to not just one or two Sanskrit words, but in fact 16 Sanskrit words and suffixes. This prompts further inquiry as to why Xuanzang, renowned for his precision, would use one translated term in such a broad range of contexts. And considering “體” in Chinese philosophy signifies innermost essence and has ontological implications, its usage by Xuanzang may require careful examination to avoid potential misinterpretation within a Chinese philosophical context. Therefore, exploring Xuanzang’s usage of “體” in AKBh[X] can provide valuable insights into Xuanzang’s translations.

Secondly, AKBh was chosen because it is currently available in Sanskrit, Tibetan, and two Chinese translations2. This allows for a comparative study, comparing the Sanskrit manuscript with Xuanzang’s translations to identify possible Sanskrit originals for his translated terms. Additionally, AKBh provides a comprehensive introduction to the doctrinal system of the Sarvāstivāda school, covering almost all the topics relevant to this tradition (Sakurabe 1981, p. 36). Using it as the primary literature allows one to investigate the usage of specific translated terms across various aspects of Sarvāstivāda teachings, making the study more comprehensive.

2. The Overall Situation of “體” in AKBh[X]

The chosen Sanskrit text for this study was Pradhan’s second edition (1975). It is currently the most widely used critical edition, with only minor changes from Pradhan’s first edition. Upon comparison, it was observed that while AKBh[X] contains some explanatory content not present in Pradhan (1975), the majority of the sentences align with the Sanskrit originals in Pradhan (1975)3. Therefore, using Pradhan (1975) as a reference for comparison with AKBh[X] is a reasonable and appropriate approach.

In ancient Chinese, “體” can function as an independent word and is defined with various meanings, such as “body”, “form”, “expression”, “essence”, “intrinsic nature”, “subject”, and “principles”, according to the Comprehensive Chinese Dictionary (《漢語大字典》) (2010, pp. 4708–9). Additionally, as a morpheme, “體” can combine with other morphemes to create compound words, such as “自體” (one’s own body, itself) and “實體” (objective existence). Interestingly, even when used as a single morpheme, “體” retains the meanings found in compound words such as “自體” “实體”, and so on. In AKBh[X], the term “體” is encountered both as an individual word and as part of compound words alongside other morphemes. When investigating Xuanzang’s translation of “體”, therefore, it is essential to consider its usage in both standalone and compound-word contexts.

Through a thorough search, it was found that there are a total of 437 instances of the term “體” in AKBh[X]. These occurrences can be traced back to 16 Sanskrit words and suffixes, which include svabhāva, ātman, dravya, bhāva, ātmabhāva, jāti, svarūpa, sat, aṅgapratyaṅga, -tā, -tva, artha, aikya, dravyabhāva, sattva, and śarīra. Further details are as follows (

Table 1. ‘體’ in AKBh[X].

	Chapter1	Chapter2	Chapter3	Chapter4	Chapter5	Chapter6	Chapter7	Chapter8	Chapter9	Total
Xuanzang’s own addition	20	51	37	37	46	41	17	10	39	298
svabhāva	7	6	3	15	6	10	2	2	3	54
ātman	2	5	2	2	2	2	1	2		18
dravya		11	2	4	2			3	3	25
bhāva		2	3	2	1		1			9
ātmabhāva		2	2	1	3	1	1			10
jāti		4		1						5
svarūpa		1								1
sat		1								1
aṅgapratyaṅga			2	1						3
~tā	1									1
~tva	1	1		1						3
artha		3		1						4
aikya				1						1
dravyabhāva					1					1
sattva									1	1
śarīra		2								2
Total	31	89	51	66	61	54	22	17	46	437

):

The table above reveals that among all occurrences of “體” in AKBh[X], the ones added by Xuanzang himself without the Sanskrit equivalents are the most numerous. When there are corresponding Sanskrit words, “體” most frequently corresponds to svabhāva, with a total of 54 instances. Following this, there are 25 instances corresponding dravya and 18 instances corresponding to ātman. Additionally, “體” appears frequently with bhāva, totaling nine instances. Moreover, there are ten instances of the compound term ātmabhāva (composed of ātman and bhāva) and one instance of dravyabhāva (composed of dravya and bhāva). Notably, Xuanzang also translated the terms artha, svarūpa, sat, -tā, -tva, aikya, sattva, sat, and śarīra into “體”.

In the subsequent phase of this study, a meticulous investigation will be carried out to analyze the occurrences of “體” in AKBh[X], first focusing on the correspondence between “體” and various Sanskrit terms, followed by an analysis of cases where there is no corresponding Sanskrit term. In instances where there exists a corresponding Sanskrit term, our approach will involve initially discerning the meaning of the equivalent term for “體” in the Sanskrit original. Subsequently, we will evaluate the accuracy of Xuanzang’s utilization of “體” and its derivative compounds in translating the respective Sanskrit term. This analysis will also encompass an exploration of potential misinterpretations that may arise from Xuanzang’s choices. In cases where no Sanskrit equivalents exist, our analysis will involve categorizing the instances where Xuanzang added the term “體”.

3. “體” and artha

In this section, we will delve into the essential but often overlooked correspondence between the term “體” and the Sanskrit word, artha. Despite being translated as “體”, artha only appears in four instances, but these occurrences provide valuable insights into the potential inadequacies and challenges faced in Xuanzang’s translations. Notably, three of these instances are concentrated in the second chapter of AKBh, specifically in the discourse on the thought (citta) and thought-concomitant (caitasa). The following are the specific occurrences of artha in AKBh[X]:

cittaṃ mano ‘tha vijñānamekārthaṃ

(2.34ab)

cinotīti cittam | manuta iti manaḥ | vijānītīti vijñānam | citaṃ śubhāśubhairdhātubhiriti cittam | tadevāśrayabhūtaṃ manaḥ | āśritabhūtaṃ vijñānamityapare | yathā cittaṃ mano vijñānamityeko ‘rthaḥ evaṃ

cittacaitasāḥ | sāśrayālambanākārāḥ samprayuktāśca

(2.34bcd)

eko ‘rthaḥ | [061|22-062|05]

Modern Translation4:

Also, Thought (citta), Mental Faculty (manas), and Consciousness (vijñāna) share the same meaning (artha).

(2.34ab)

[It] accumulates, thus [it] is Thought. [It] contemplates, thus [it] is Mental Faculty. [It] discriminates, thus [it] is Consciousness. Others assert (apare) that what accumulates from the gathering of good or bad elements (dhātu) is Thought, the same it serving as the basis (āśraya) is Mental Faculty, the same it serving as the dependence is Consciousness. How about Thought, Mental Faculty and Consciousness being the same entity, in the same way that

Thoughts and Thought-concomitants having a basis, cognitive object (ālambana), modes of activity (ākāra)

(2.34bcd)

is the same entity (arthaḥ).

AKBh[X]:

今當辯此名義差別。頌曰：心意識體一。

論曰：集起故名心。思量故名意。了別故名識。復有釋言：淨不淨界，種種差別，故名為心。即此為他作所依止，故名為意。作能依止，故名為識。故心意識，三名所詮。義雖有異而體是一。

[頌曰：]心心所有依。所緣行相。相應亦爾。

如心意識，三名所詮，義異體一。諸心心所，名有所依，所緣行相，相應亦爾。名義雖殊，而體是一。

[T29, p. 21c17-26]

As we know, the AKBh authored by Vasubandhu comprises both Kārikā and Bhāṣya. The former corresponds to verses, while the latter is its prose auto-commentary. However, the viewpoints represented in the Kārikā and Bhāṣya diverge. The Kārikā represent the position of the Sarvāstivāda school, while the Bhāṣya present the opposing view, associated with the Sautrāntika school. The above Sanskrit excerpt brilliantly illustrates this distinction.

Here, Vasubandhu employed two clever tactics. Firstly, he skillfully manipulated the term artha in the Kārikā and Bhāṣya. In Kārikā 2.34ab which represents the standpoint of the Sarvāstivāda, it is stated that Thought (citta), Mental Faculty (manas), and Consciousness (vjñāna) share the same meaning (artham). As “artham” in 2.34ab is a neuter gender noun, the term “artham” is understood to mean “meaning”5. However, in the Bhāṣya, Vasubandhu subtly altered “artham” to the masculine noun “arthaḥ”, changing the meaning of “artha” to “entities” or “objects”.

Secondly, Vasubandhu employed another clever strategy by utilizing the structure “yathā…evam…” (just as… in this way…) to connect Kārikās 2.34ab and 2.34bcd. The issue is that Kārikā 2.34ab discusses only Thought, Mental Faculty, and Consciousness, while Kārikā 2.34bcd exclusively addresses Thought and Thought-concomitant. From the Kārikās alone, there seems to be no logical connection between 2.34ab and 2.34bcd. However, through the “yathā…evam…” construction in the Bhāṣya, Vasubandhu connects 2.34ab with 2.34bcd, suggesting that just as Thought, Mental faculty, and Consciousness are one entity, so too are Thought and Thought-concomitant, a correlation that does not exist in the Kārikās alone. This allows Vasubandhu to present a view aligned with the Sautrāntika school, while the original Kārikās reflect the Sarvāstivāda perspective. As we are aware, the Sarvāstivāda maintains that Thought and Thought-concomitant are two separate entities, whereas the Sautrāntika views them as one.

Vasubandhu’s intricate maneuvering is truly impressive. Without referring to the original Sanskrit text and relying solely on Xuanzang’s translation, however, we would be unaware of Vasubandhu’s subtle intentions. Xuanzang translated both “artham” and “arthaḥ”, which have distinct meanings, as “體”. He rendered Kārikā 2.34ab as “心意識體一” (Thought, Mental faculty, and Consciousness are one 體). And in the Bhāṣya, he translated the lines containing “arthaḥ” as “義雖有異而體是一” (although the meanings differ, the 體 is one) and “而體是一” (yet the 體 is one).

Indeed, while “體” carries multiple meanings in Chinese, the close proximity and repeated use of “體” in the same context could lead readers to overlook any subtle differences in its meaning. Without referring to the original Sanskrit text, readers may remain unaware that the first occurrence of “體” corresponds to “artham”, while the subsequent two occurrences correspond to “arthaḥ”. Furthermore, “體” in Chinese can carry meanings such as “entities” and “objects”, supporting Xuanzang’s translation of “arthaḥ” as “體”. However, according to the Comprehensive Chinese Dictionary (《漢語大字典》) (2010, pp. 4708–9), “體” does not encompass the meaning of “meaning”, making Xuanzang’s rendering of “artham” inaccurate or even incorrect from this perspective.

As a result, Xuanzang’s translation in this case falls short of capturing the subtleties of the original text.

In Chapter 4 of AKBh, there is also an occurrence of the term artha translated by Xuanzang as “體”. The Sanskrit original along with its corresponding translations is presented below.

akarmasvabhāvam apy asti trividhaṃ manoduścaritaṃ cetanārthāntarabhūtam abhidhyā vyāpādaḥ, mithyādṛṣṭiśca |

[237|17-237|18]

Modern Translation：The three kinds of mental faculty having non-karma as svabhāva also exist, being an entity different from Intention (cetana), [they] are desiring, ruining, and wrong view.

AKBh[X]：然意惡行復有三種，謂非意業貪瞋邪見，貪等離思別有體故。

[T29, p. 84b2-3]

Upon comparing the Sanskrit original, it becomes evident that although Xuanzang’s translation is not a word-for-word rendition, we can still determine that his use of “體” corresponds to “artha”. Specifically, “cetanārthāntarabhūtam” is rendered by Xuanzang as “貪等離思別有體 (desiring and others exist apart from Intention with another 體) “. In the Sanskrit original above, “artha” conveys the meaning of “entities” or “objects”. Given that “體” can also be interpreted as “ entities “ or “objects”, Xuanzang’s choice to translate “artha” as “體” appears justifiable from this perspective. However, as we will see next, the translation like this may also give rise to some potential misconceptions.

4. “體” and svabhāva

The significance of svabhāva in Abhidharma doctrines needs no overemphasis. It is not only directly related to the definition of dharma but also widely applied in various contexts throughout the literature. According to Seiji Kimura, AKBh contains a total of 196 occurrences of svabhāva (2002). Among these 196 instances, 54 were translated by Xuanzang as “體” or the compound terms containing “體”. Based on the investigation conducted in this study, these 54 usages of svabhāva can be broadly categorized into three types: categorical usage, denoting “itself”, and representation of real entities (dravya).

4.1. The Svabhāva Used to Denote “Itself”

Xuanzang translated svabhāva, which carries the meaning of “itself”, as “體” or its derived compounds. For example:

yathā payaścātapaśca sarpiṣaḥ śyānatvavilīnatvayor na tu punastatsvabhāvau |

[061|02-061|03]

Modern Translation: And, for example, water and sunlight are the [causes] of the solidified state and the melted state of the pure butter, not [its solidified state and the melted state] itself.

AKBh[X]: 如水日光，是凝釋因，體非凝釋。

[T29, p. 21b23-24]

saṃskṛtasya hi dharmasya svabhāvavarjyāḥ sarvadharmāḥ kāraṇahetuḥ |

[082|26]

Modern Translation：A conditioned dharma takes all dharmas except itself as its Efficient cause (kāraṇahetu).

AKBh[X]: 論曰：一切有為，唯除自體，以一切法為能作因。

[T29, p. 30a17-18]

In the above two examples, Xuanzang translated svabhāva, which conveys the meaning of “itself”, as “體” and “自體”. In Chinese, “體” can mean “entities themselves”, while “自體” specifically refers to “itself”. Considering the context and the surrounding language, it becomes evident that “體” and “自體” in these instances are intended to convey the meaning of “itself”. Therefore, Xuanzang’s translation here is highly appropriate and accurate in this regard. But in the following usage, Xuanzang’s translation of svabhāva as “體” could potentially lead to some misunderstandings.

4.2. The Svabhāva Used for Categorization and to Denote Real Entity

The so-called categorical usage of svabhāva refers to its function in distinguishing one or more things from others. Takumi Fukuda (1988, p. 62) mentioned that svabhāva can be used to form various categories, and Kimura (2002, p. 316) explicitly stated that svabhāva serves a categorization function in AKBh. It is worth noting that this “categorization” occurs not only within the same level but also across different levels. In other words, svabhāva can be employed to differentiate entities at various hierarchical levels. The Sarvāstivada may state “X has Y as svabhāva” and “Y has Z as svabhāva”, where X and Y belong to distinct levels.

The following examples illustrate this usage of svabhāva in AKBh.

[1] ato ya īraṇāsvabhāvo dharmaḥ sa vāyuriti karṃaṇā ‘sya svabhāvo ‘bhivyaktaḥ |

[008|24-008|25]

Modern Translation：Hence (atas), a dharma having mobility (īraṇā) as svabhāva is Wind (vāyu). Therefore (iti), its svabhāva is manifested by efficacies (karman).

AKBh[X]：故應風界動為自性。舉業顯體故亦言輕。

[T29, p. 3b12-13]

Xuanzang translated the first occurrence of “svabhāva” in this sentence as “自性” and the second occurrence as “體”. However, it is evident that in this sentence, both “svabhāva” have the same meaning, referring to the defining characteristic of Wind as a fundamental material element (mahābhūta). In other words, through the svabhāva of mobility, Wind is distinguished from other elements. Xuanzang’s translation of “舉業顯體” can be understood as “manifesting the svabhāva through karman”, which aligns with the meaning of the original Sanskrit text. Notably, in this case, Xuanzang translated two “svabhāva” of identical meaning into two different Chinese terms, possibly due to considerations of the four-character style of translation. The following examples also demonstrate the same usage of svabhāva.

sarveṣu ca dharmasvabhāveṣv ekaṃ dharmāyatanam |

[016|14-016|15]

Modern Translation: Also, in the case where all the sense-spheres (āyatana) have dharma as svabhāva, there is one sense-sphere of dharma.

AKBh[X]: 又十二處體皆是法，唯於一種立法處名。

[T29, p. 6a9-10]

yathaivāmohaḥ kuśalamahābhūmiko nāvadhāryate prajñāsvabhāvatvāt |

[056|16]

Modern Translation: It is due to having Understanding (prajñā) as svabhāva, for instance, non-delusion (amoha) is not considered as a wholesome permeating factor (kuśala-mahā-bhūmika).

AKBh[X]: 如無癡善根，慧為體故，非大善地法。彼亦應爾。

[T29, p. 19c19-20]

yadi tayaiva rūpād abhinnasvabhāvaḥ pudgalaḥ prāpnoti rūpa eva vā tatprajñaptiḥ |

[463|26-463|27]

Modern Translation: If it were through that [knowledge of the Form], the pudgala would have svabhāva that is not distinct from the Form, or just have a name for it with respect to the Form.

AKBh[X]: 若色能了即能了此，則應許此體即是色，或唯於色假立於此。

[T29, p. 153c3-5]

Next, when svabhāva was used to indicate real entity or substantial existence, Xuanzang also translated it as “體” in Chinese. For instance:

tasmād abhinna eṣāṃ caturṇāṃ saṃvarāṇāṃ tribhyaḥ svabhāvaḥ |

[206|06-206|07]

Modern Translation：Therefore (tasmāt), the four types of Restraints (saṃvara)’s svabhāva is not distinct from [other] three.

AKBh[X]: 故四律儀非異三體。

[T29, p. 72c8-9]

Here, the “svabhāva” should be understood to refer to real entities or substantial existences, because earlier in the text, it was mentioned that there are eight types of Restraints (saṃvarāṇa), but at the level of real entities (dravyatas), there are only four6.

In the provided examples, Xuanzang consistently translated the “svabhāva” used for categorization and to denote real entity as “體” in Chinese. While “體” carries a rich set of meanings in Chinese, when applied to abstract entities and material phenomena that are not easily perceivable in our daily experience, such as in the sentence “夷雅之體, 無待韋弦” (The intrinsic nature of elegance does not rely on external influences) where it is applied to elegance, “體” is often understood as “intrinsic nature”. In the mentioned examples, “svabhāva” is applied to Wind, sense-sphere (āyatana), Understanding (prajñā), Form (rūpa), and Restraints (saṃvara). According to the Abhidharma doctrines, Wind is one of four fundamental material elements. Although considered as concrete, Wind as a material element is not easily and accurately perceivable in in our daily experience. Consequently, readers may naturally interpret “體” applied to Wind as “intrinsic nature”. Similarly, because Understanding is an abstract mental factor, and sense-sphere and Form represent categories that are also abstract, while Restraints are abstract entities, it becomes easy to interpret the use of “體” applied to these entities as signifying their “intrinsic nature”.

However, this understanding may lead to a misconception when encountering phrases such as “dharma’s svabhāva”. It might create the misunderstanding that within the dharma of the Sarvāstivāda school, there is another intrinsic nature of dharma that exists separately from the dharma themselves. Yet, in the Sarvāstivāda view, a fundamental characteristic of dharma lies precisely in its mereological independence (Westerhoff 2018, p. 71). As illustrated in example [1], when Wind is described as having “mobility” as its svabhāva, it does not imply the existence of a separate “mobility” nature distinct from the phenomenon of Wind. Instead, according to the Abhidharmaprakaraṇapādaśāstra, “What is the Wind as elementary substance? It is called lightness and other mobility (風界云何？謂輕等動性 [T26, p. 692c12])”, indicating that the Wind itself is “mobility”7.

In summary, regardless of whether svabhāva is used for categorization, represents “itself”, or denotes real entities, Xuanzang consistently translates it as “體” or the compound words with “體”. When svabhāva refers to “itself”, Xuanzang’s translation of “體” and “自體” is clear and unambiguous. However, when svabhāva is used for categorization or denotes real entities, translating it as “體” may cause the misunderstanding that there is an intrinsic nature distinct from phenomena within the entities.

5. “體” and ātman, dravya, bhāva, ātmabhāva, dravyabhāva as Well as Other Sanskrit Equivalents

In addition to svabhāva, Xuanzang frequently translated the Sanskrit terms ātman, dravya, and bhāva, as well as the compound words ātmabhāva and dravyabhāva, as “體” or compounds with it. In AKBh, some usages of ātman, dravya, and dravyabhāva are similar to svabhāva. The following will examine the correspondence between “體” and these Sanskrit terms and identify potential issues that may arise in their translations.

5.1. ātman

In AKBh, the Sanskrit term ātman, when translated by Xuanzang as “體”, appears in three different contexts: representing “itself”, indicating “composing of / being included in”, and used for categorization.

When ātman represented “itself”, Xuanzang frequently translated it as “自體”. Just as in the case of svabhāva, Xuanzang’s translation of “自體” for the meaning of “itself” is clear and unambiguous.

However, the appropriateness of Xuanzang’s translation becomes questionable when ātman is used in the other two contexts. Firstly, for ātman used in categorization, Xuanzang translated it as a single morpheme “體”; for example, “saṃjñā nimittodgrahaṇātmikā” [010|16] (AKBh[X]: 想取像為體 [T29, p. 3c28]). Since the usage of ātman for categorization aligns with the usage of svabhāva for categorization, the drawback of translating svabhāva as “體” also applies here. In this case, it may lead to the misunderstanding that there is an intrinsic nature distinct from phenomena within things.

Next, when ātman appears as the last element in a compound word, it can signify “composing of / being included in”. Xuanzang also translated this type of ātman as “體”. While “體” indeed encompasses the meaning of “inclusion” according to the Comprehensive Chinese Dictionary (《漢語大字典》) (2010, pp. 4708–9), the problem arises, for instance, when readers encounter Xuanzang’s translation “體唯三” (tryāyatanātmakaḥ [152|18]). They might misunderstand the meaning of “體” here, interpreting it as a noun denoting “essence” or something else.

In conclusion, while Xuanzang consistently translates ātman as “體” or compounds with it, the appropriateness of the translation depends on the context in which these Sanskrit terms appear. While the translation as “自體” for representing “itself” is clear and unambiguous, translating ātman as “體” for categorization or representing inclusion might lead to potential misunderstandings and misinterpretations.

5.2. Dravya

In AKBh, the term dravya referring to real entity was translated by Xuanzang as “體” “实體” “體类” “體实”, and “有體实”. Where Xuanzang translated it as “实體” and other compounds above, there is relatively little ambiguity, since these carry the meaning of “real entity” in Chinese. If he simply translated it as “體” where it applied to something abstract, however, the potential for readers to misinterpret “體” as “intrinsic nature” remains, as highlighted earlier.

Notably, Xuanzang occasionally translated “dravyatas” with the ablative case suffix “tas” as the compounds including “实體”, “有體实”, or “实有體”, which have little difference in meaning. It seems that he did not translate the ablative case ending. In modern literal translations, we generally render dravyatas as “in the aspect of real entities” or “as real entities”, and other such phrases in prepositional structures. However, this literal translation style might have been considered verbose and less elegant by ancient Chinese translators. Therefore, in cases where the semantics remain unchanged, Xuanzang often adopted semantic equivalence. Thus, Xuanzang’s translation of “dravyatas” as “实體”, “实有體”, or “有體实” is fully understandable.

Furthermore, from a word-by-word comparison perspective, Xuanzang also translated one instance of dravya as “自體类”, as shown below:

na tvalpakād vedanādidravyāt prabhūtaṃ vedanādidravyamamutpadyata ityetāvadevātroktam |

[098|21-098|22]

Modern Translation: But (tu) multiple Sensations (vedanā) and other real entities do not arise from fewer Sensations and other real entities. It is just like that has been said here.

AKBh[X]: 且於受等自體類中，無少生多。以說等義。

[T29, p. 36c8-9]

Xuanzang’s translation of “dravya” as “自體类” in the context of “vedanādidravyāt” may initially appear perplexing, as “自體类” seems to evoke the term svajāti rather than dravya. However, considering the preceding context, we find that Xuanzang’s choice was intended to maintain consistency in his Chinese translation. Earlier in the text, “svajāti” has indeed been mentioned, and the previous themes have consistently revolved around “jāti”. In this particular sentence, perhaps Vasubandhu switched to using “dravya” to provide a more specific explanation.

From this perspective, Xuanzang’s later use of “自體类” actually signifies “svajāti-dravya”. In other words, to ensure a smoother flow in the Chinese translation, Xuanzang added the term “svajāti” (AKBh[X]:自类). Consequently, in this context, “dravya” is effectively translated simply as “體”.

This demonstrates how Xuanzang strived to maintain consistency and clarity in his translation, adapting certain terms to better align with the evolving themes and context of the text. By incorporating “svajāti” into the translation, Xuanzang aimed to make the reading experience more fluid and comprehensible to his Chinese audience.

5.3. bhāva, ātmabhāva, and dravyabhāva

It is noteworthy that Xuanzang also translated certain instances of bhāva in AKBh as “體”. These occurrences of bhāva carry various meanings, such as “state”, “existing”, “existence”, and in the compound word bhikṣubhāva, it signifies “monkhood”. Xuanzang consistently used “體” to translate all these diverse meanings of bhāva.

When bhāva represents “existing” or “existence”, its translation as “體” aligns well with the context and conveys a clear meaning, given that “體” also carries such implications. However, when bhāva signifies “state”, as in the example “kukṛtasya bhāvaḥ kaukṛtyam” [057|18] (AKBh[X]: 惡所作體，名為惡作 [T29, p. 20b7]), as “體” in Chinese does not include the precise meaning of “state”, Xuanzang’s translation of “體” seems rather ambiguous, potentially leading to misunderstandings.

Additionally, Xuanzang translated bhikṣubhāva, which refers to “monkhood”, as “比丘體” (bhikṣu’s 體), where the meaning of “體” is also quite ambiguous and difficult to understand precisely, even with the context.

Furthermore, Xuanzang’s translation of the compound term ātmabhāva as “自體”, which can be understood to denote body or self-body, fits well with the context.

For dravyabhāva applied to abstract entities, which shares a synonymous meaning with dravya, Xuanzang translated it as “體”, introducing the same potential drawback as in the case of dravya translations, wherein “體” may be interpreted as “intrinsic nature”.

5.4. Other Sanskrit Terms and Suffixes

Apart from the cases mentioned earlier, Xuanzang also translated another nine Sanskrit terms and suffixes into “體” or compounds with “體”. They were -tā, -tva, svarūpa, jāti, sat, śarīra, aṅgapratyaṅga, aikya, and sattva.

Among these, Xuanzang generally translated jāti meaning “species” or “category” as “體类”, aṅgapratyaṅga signifying “limbs” or “body parts” as “肢體” or “支體”, and aikya conveying “identity” or “unity” as “一體”. These translations are clear and in line with the original meanings.

Furthermore, Xuanzang rendered svarūpa representing “itself” as “體相”, śarīra meaning “body” as “體”, and sattva denoting “existence” or “being” as “體”. And he translated sat signifying “existence” as “有體”, meaning “existing entity”. Although these translations are not as explicit as the previous examples, they still convey relatively clear meanings with context.

It is essential to note that Xuanzang also translated the suffixes -tā and -tva as “體”. In Sanskrit, -tā and -tva represents abstract qualities or states. Since “體” itself can also imply “quality”, Xuanzang’s choice to translate them as “體” is reasonable from this perspective. In such cases, however, readers might mistakenly associate “體” with svabhāva, leading to potential misunderstandings of the sentence’s intended meaning.

Additionally, in modern language, for sentences with the structure of Genitive case + -tva, such as “tasya tadekatva”, the structure is generally translated as “they are the same”, without turning “-tva” into a separate word. In AKBh, however, Xuanzang often translated -tva / -tā into terms such as “體”, which could be considered a characteristic of his translation style.

6. Cases Where Xuanzang Added “體” Himself8

Xuanzang added the term “體” 290 times, which accounts for 68% of all occurrences of “體” in AKBh[X], indicating that in over half of the cases “體” was added by Xuanzang himself and does not have a corresponding Sanskrit original. There are two main categories of cases where Xuanzang added “體”:

(1) The entire sentence containing “體” does not have a corresponding Sanskrit equivalent.

(2) The sentence containing “體” has a corresponding Sanskrit equivalent. Within this category, we can further divide it into: (2.1) Instances where “體” lacks a corresponding Sanskrit word in the sentence but may have a corresponding word found in the context of the surrounding text. (2.2) Instances where “體” does not have a corresponding Sanskrit word anywhere in the context.

Let us begin by examining examples falling under category (1):

dvividhaṃ hi prema｜[060|09]

Modern Translation: Because (hi) Affection (preman) is of two kinds.

AKBh[X]: 愛敬別者，愛謂愛樂，體即是信。然愛有二。

[T29, p. 21a25-26]

In this example, the sentence “愛敬別者，愛謂愛樂，體即是信” (The difference between Affection and Respect is that Affection is thirst, 體 is Faith) are all explanatory content added by Xuanzang in the translation. By the Sanskrit-Chinese collation, it was found that Xuanzang sometimes incorporated explanatory content into his translations. In such cases, since the entire sentence is explanatory, the added “體” by Xuanzang can generally be understood based on the context in preceding text. For the example above, in Kārikā 2.32c, it says “prema śraddhā” [060|08] (Affection is Faith), and therefore, the “體” added by Xuanzang in the sentence “愛謂愛樂，體即是信” (Affection is thirst, 體 is Faith) may refer to the affection itself, just like the usage of svabhāva to signify “itself”.

Next, let us consider situation (2.1), where the sentence containing the added “體” has a corresponding Sanskrit original. The added “體” lacks a corresponding Sanskrit word in the specific sentence, but can be found in the surrounding context.

buddhadharmasaṃghāvetyaprasādāḥ śraddhāsvabhāvāḥ āryakāntāni ca śīlāni śīlam iti dve(du) dravye(du) bhavataḥ |

[387|06-387|07]

Modern Translation：The serenity based on trusting faith of the Buddha, Dharma, and Sangha has Faith as svabhāva. And the beloved ones of the respectable have Restraints [as svabhāva]. On the level of existence, there are Restraint [and Faith] such two real entities.

AKBh[X]：謂於佛等三種證淨以信為體，聖戒證淨以戒為體故唯有二。

[T29, p. 133b25-26]

Based on the “buddhadharmasaṅghāvetyaprasādāḥ śraddhāsvabhāvāḥ”, it can be inferred that “āryakāntāni ca śīlāni” omits the term “svabhāva”. Xuanzang noticed this and thus supplemented it during the translation.

Next, let us consider situation (2.2), where the “體” added by Xuanzang does not have a corresponding Sanskrit word in the preceding or following context. Here are a few examples of this situation:

[2] tatra svabhāvavikalpo vitarkaḥ |

[022|22]

Modern Translation: Here, Initial Inquiry’s (vitarka) svabhāva is conceptualizing activity (vavikalpa).

AKBh[X]: 自性分別，體唯是尋。

[T29, p. 8b5]

[3] ātmanyasati kathamādhyātmikam bāhyaṃ vā |

[027|06]

Modern Translation: If the selves do not exist, how can there be an inner or outer [self]?

AKBh[X]: 我依名內，外謂此餘。我體既無，內外何有？

[T29, p. 9c18-19]

In Example [2], Xuanzang translated “svabhāva” as “自性” and then added the term “體” to his translation. According to the context, “體” here refers to entities or things. In Example [3], Xuanzang translated “ātman” as “我” (self / I) and added “體” after “我”. In this context, “體” possibly represents entities or things. In both examples, the added “體” seems to function as an expletive and does not significantly alter the meaning of the sentences. In the following instance, however, he chose to replace the original Sanskrit term with “體”.

[4] yathā suvarṇabhājanasya bhittvā’nyathā kriyamāṇasya saṃsthānānyathātvaṃ bhavati na varṇānyathātvam |

[296|11-296|12]

Modren Translation: For example, after a gold object is broken, [the other objects] made from it in different ways have different shapes, and their colors are not different.

AKBh[X]：如破金器作餘物時。形雖有殊而體無異。

[T29, p. 104c4-5]

In Example [4], according to the Sanskrit–Chinese comparison, it is easy to assume that Xuanzang translated “varṇa” (color) as “體”. However, the Chinese term “體” does not carry the meaning of “color”, and the Sanskrit word “varṇa” does not have the same usage as “體”. Therefore, in terms of meaning, “varṇa” and “體” cannot correspond to each other, suggesting that “體” in this context was added by Xuanzang himself.

It is worth noting that for the same sentence, Paramārtha translated it as “譬如打破金器作別莊嚴具，有別形相故有異。不由物類異故異，色等同故 [T29, p. 258a2-3]” (For example, when breaking a golden object to create various ornaments, due to different forms, there are differences. Because of the sameness of color, the differences do not result from a distinction in substance). It is evident that Paramārtha accurately translated “varṇa” as “色” (color) based on the Sanskrit meaning. The question remains: why did Xuanzang translate the Sanskrit original as “體無異” (體 is not different)?

Upon examining the preceding text, we find the expression “na dravyānyathātvam” [296|11] (not different from the real entities) where Xuanzang translated “dravya” as “體”. Therefore, it can be inferred that Xuanzang’s usage of “體” in Example [4] is also derived from the earlier “dravya”. Based on Xuanzang’s translation, the sentence now reads: “The broken golden vessel made into other objects may have different shapes, but the ‘體’ remains the same.” Here, “體” appears to refer to the gold. While this interpretation does not significantly alter the overall meaning of the sentence, it does involve adding Xuanzang’s own understanding in the translation process.

Xuanzang’s preference for using “體” is further evident in his frequent translation of dharma as “法體” (法 refers to dharma). For instance:

[5] syād eṣa doṣo yadi dharmāt kāritram anyat syāt |

[298|04]

Modern Translation: If the Activities (kāritra) differs from the dharma, there would be this fault.

AKBh[X]: 若許作用異法體者，可有此失。

[T29, p. 105a16]

Upon careful examination, we can indeed find that in Example [5], the “體” originates from the preceding term “ātman”9 representing “itself”. Therefore Xuanzang’s translation of “法體” would mean the dharma itself. However, if one were to read Xuanzang’s translation without referring to other sources, it would be easy to misinterpret “法體” as “dharma’s 體”, referring to the intrinsic nature of dharma, leading to a misunderstanding.

7. Conclusions

In AKBh[X], Xuanzang extensively employed the term “體” and its derived compounds. When corresponding to Sanskrit words, “體” and its composites are linked to 16 Sanskrit terms and suffixes: svabhāva, ātman, dravya, bhāva, ātmabhāva, jāti, svarūpa, sat, aṅgapratyaṅga, -tā, -tva, artha, aikya, dravyabhāva, sattva, and śarīra. In instances where no Sanskrit equivalents existed, Xuanzang added 292 instances of “體”.

Xuanzang’s usage of the term “體” was highly intricate. Not only did he translate Sanskrit words such as svabhāva, ātman, and bhāva into “體”, each with distinct meanings and usages, but he also introduced various examples of “體” on his own. However, not all instances of “體” in AKBh[X] are accurate and faithful representations of the original Sanskrit texts. There are several issues with his usage:

1. Translation errors: Xuanzang mistranslated the term “artham”, which denotes “meaning” in Kārikā 2.34ab, as “體”. However, “體” does not carry the connotation of “meaning”. Additionally, Xuanzang conflates “artham” with “arthaḥ”, translating both as “體”, causing confusion between the two.

2. Potential misinterpretations: Xuanzang’s translation of svabhāva and ātman as “體” when used in a categorical sense or to denote real entities might lead to misconceptions, suggesting the existence of an intrinsic nature distinct from phenomena. Similarly, rendering dravya and dravyabhāva applied to abstract entities as “體” and dharma as “法體” could also contribute to such misunderstandings. Furthermore, when ātman appears at the end of a compound and means “consisting of”, Xuanzang translated it as “體”, further adding to possible confusion.

3. Lack of clarity: Due to the extensive usage of “體” throughout AKBh[X], without Sanskrit counterparts for reference, it can be challenging to determine the precise meaning of “體” in certain contexts. For instance, expressions such as “惡所作體” (kukṛtasya bhāvaḥ) and “比丘體” (bhikṣubhāva) remain ambiguous unless compared with their corresponding Sanskrit originals, which respectively signify “the state of doing something wrong” and “monkhood”.

Based on the above, a question arises: Why did Xuanzang, renowned for his accurate translations, have such a strong preference for using “體” and its compounds, even in cases where other Chinese terms were available or in the absence of corresponding Sanskrit words, to the extent that it caused some ambiguity and even errors in translation? The author speculates two primary reasons. Firstly, “體” in Chinese has a rich range of meanings, allowing it to cover numerous Sanskrit terms, thereby reducing the burden of translation. Secondly, “體” holds a central position in Chinese classical philosophy, forming a complementary pair with “用” (yòng), where the former denotes “essence” and the latter “function”. “體” is often seen as fundamental and intrinsic, while “用” represents its external manifestation. As Xuanzang and his translation team were deeply influenced by Chinese classical philosophy, perhaps it was inevitable that this core concept of “體” would find its way into their translations. Notably, Sakurabe pointed out that the representative Sanskrit original for “體” in AKBh[X] is svabhāva, referring to the essential nature of phenomena, contrasting with karman (efficacy) (1954, p. 265). This judgement appears to have been influenced by “the theory of 體and 用” (體用論) in traditional Chinese philosophy.

In such cases, it becomes crucial for readers to be aware of the potential interpretational pitfalls that may arise from relying solely on Xuanzang’s translation and to consult other materials for a more comprehensive understanding of the context and nuances of his translated work.

Moreover, besides the aforementioned issues, it is worth mentioning that not only does “體” in AKBh[X] correspond to various Sanskrit words, but even for the same Sanskrit term in identical usage, Xuanzang may use different translations. As pointed out in this study, svabhāva is sometimes translated as “體”, while at other times it is translated as “自性”. Hence, Xuanzang’s translated terms are not entirely consistent. Therefore, without the original Sanskrit or other translations, it is unwise to interpret different renderings of Xuanzang’s terms as indications of varying corresponding Sanskrit words.