Indian Buddhist textual production—championed under a missionary ideology that resisted the Brahmanical rhetoric of the exclusivity of Sanskrit—has always conceived of itself as operating within and targeting a diversified multilingual context (of both literary and vernacular languages). This awareness, joined with a nominalistic and conventional view of language, found expression in the development of an array of distinct textual and hermeneutical practices, which persisted even during the growing Sanskritization of the North Indian branch of the tradition from around the second century CE.

One of these practices was the appropriation of Sanskrit etymological analysis (nirvacana, nirukti) into a primarily commentarial technique. Applied in this way, etymology was not about revealing an underlying intrinsic structure of language (as in the case of some Brahmanical schools of thought, see Chapter 2.2), nor was it a means for gaining insight into the temporal dimension of language (as in the case of historical etymology). Rather than a way of excavating semantic meaning, it was utilized as a way to negotiate and create meaning in commentarial praxis. As demonstrated in the text excerpts below, one of the rather unique expressions of this approach was the carrying over of Sanskrit etymologies across languages (in the case before us, into Chinese).

The Buddhist conception of etymology as a largely hermeneutical device differed in an important way from the way in which it was conceived in Sanskrit Brahmanical sources, which understood it—schematically put—as a means for exploring the interconnections of language and uncovering the ontological deep structure that was seen to be embedded in it. This approach was grounded in a language metaphysics that took Sanskrit as consisting of a fixed semantic system corresponding to real existents, and in which the essence of a thing was to be uncovered by identifying the meaning of the term that denotes it (primarily nouns, all taken to be derivations of verbal roots). The method for doing so was therefore etymological analysis, which sought—either through grammatical analysis or by pointing out commonalities—to trace nouns to their verbal roots, and by doing so to identify their decisive meaning.

In opposition to the Brahmanical view of language, Indian Buddhist schools of thought championed a view of language as utterly conventional, a system whose signs do not refer to real existents but are forever mere interpretants of other signs. Under this framework, linguistic structures and interrelations—revealed either by formal analysis or by descriptive practices such as etymological analysis—were seen merely as an intrinsic feature of language as a self-referential realm, and as having no purchase on reality as it truly is. So while Buddhists keenly adopted the Brahmanical Sanskrit etymological techniques (with their reduction of nouns into verbal roots, and as a semantically rather than historically oriented analysis), they saw these techniques as serving not to uncover the deep structure of language but primarily as an interpretative and argumentative tool in the elucidation of Buddhist philosophical and literary texts.

The text excerpts below all focus on Buddhist etymological glosses of the Sanskrit word śāstra, a generic term for a treatise (by a human author, in contrast to sūtra, i.e. scripture, which is ascribed to a Buddha or a bodhisattva). All the sources before us are from around the first half of the first millennium, and while they vary in their sectarian affiliation (some are affiliated with Northern Abhidharma schools, some with the Mahāyāna Yogācāra school), they nonetheless form a tightly connected intertextual realm insofar as they refer to, explicitly comment upon, and quote each other.

Passages I, II, and III are from treatises traditionally ascribed to the influential Buddhist philosopher Vasubandhu (typically dated to around the late fourth to early fifth century CE). According to the traditional biographies, Vasubandhu’s life, during the Gupta reign in India, was marked by two acts of conversion: first from the Sarvāstivāda to the Sautrāntika school of Indian Abhidharma, and then to Mahāyāna, into the Yogācāra school, of which he is considered one of the founding figures.

Passage I provides an etymology of the term “treatise” (śāstra) that is taken from the opening lines of Vasubandhu’s Commentary on the Treasury of Abhidharma (Abhidharmakośabhāṣya), an encyclopedic source for Abhidharma doctrinal thought. This etymological gloss is rather straightforward, and in terms of Sanskrit classical grammar, a correct analysis.

Passage II provides an alternative etymology of the term given by a different work by Vasubanhdu, The Proper Mode of Scriptural Exegesis (Vyākhyāyukti, extent only in Tibetan translation), a protocol for the writing of commentaries and exegesis written from a distinctively Mahāyāna perspective. Here Vasubandhu uses a more fanciful etymological gloss to argue that scripture (the Buddha’s speech) should be seen as the ultimate treatise.

Passages III and IV demonstrate how both these alternative glosses are then taken up and woven together to form a new commentarial synthesis. Both passages are taken from a thread of commentaries on Distinguishing the Middle from the Extremes (Madhyāntavibhāga). Traditionally, the Madhyāntavibhāga is considered to be a revealed work ascribed to Maitreya, who is said to be a Bodhisattva removed from complete Buddhahood by only one birth. According to the tradition, Maitreya pronounced the work in verse form to Asaṅga (ca. fourth to fifth century CE), one of the founding figures of the Yogācāra school, who in turn made it available to Vasubandhu, who composed the commentary (Madhyāntavibhāga-bhāṣya). To this chain of commentaries is then added Sthiramati’s (ca. sixth century CE) super-commentary (Madhyāntavibhāga-bhāṣya-Ṭīkā) on Vasubandhu’s commentary.

As is the custom in these works, Vasubandhu opens his commentary with a dedication (Passage III), in which, however, he refers to the revealed work as a “treatise” (śāstra). This presents a serious interpretative difficulty for the ensuing super-commentary of Sthiramati, since, as I mentioned above, in Buddhist Mahāyāna lore a treatise is typically used as a generic term for a scholastic work composed by an ordinary human author, whereas revelatory works ascribed directly to the Buddha or bodhisattvas like Maitreya are called sūtra. The issue at stake, it should be clarified, is more than just getting the terminology right: it bears on the fundamental question of the text’s authority. It is this question, therefore, that Sthiramati addresses in his ensuing super-commentary (Passage IV). For this purpose, he provides a definition of a treatise that uses etymology to reinstate its status and authority as equal to that of any other revelatory text. It is important to note that in this passage, Sthiramati provides two alternative etymologies of śāstra, taken from Vasubnadhu’s sources mentioned above (in Passages I and II), and which constitute two different interpretations, neither of which is exclusive.

This feature, which is ubiquitous in Buddhist Sanskrit lore, emphasizes yet again the foremost interpretative function of etymology in the Buddhist context. The Buddhist conception does not view meaning as something that lies in the temporal evolution of language or reflects its deep structure. It is not something to be discovered in language, but something to be created with language, that is, in the context of its use.

This conception of etymology and meaning, in turn, sheds light on the otherwise rather puzzling instances in which Sanskrit etymological glosses are carried across languages within the Buddhist realm. Passage V is a distinct example of such a case: it is a translation of Passage I from Sanskrit into Chinese by Paramārtha (499–569 CE), an Indian monk who worked in, among other places, the Chinese imperial capital. While Paramārtha faithfully translates Vasubandhu’s original etymological gloss, he also supplements it and synthesizes it with the additional alternative etymology we encountered above (in Passage II, and also in Passage IV).

The crossover of an etymology-derived meaning from one language into another—without any acknowledgment of this transfer or adjustments to the target language—is already something of a peculiarity, as it seems to obliterate the explanatory force of such an analysis. Cases like these—and they are ubiquitous in the translations of Buddhist texts across East Asia—were often explained away by scholars as reflecting either the translator’s ignorance of the original etymology, or, more commonly, the translator’s reverential treatment of the Sanskrit and the original text. Neither explanation, however, can adequately make sense of the case before us (and many other such cases), in which the translator both knew his Sanskrit well and intentionally altered the meaning in the process of the translation. This move makes perfect sense, however, once we consider it in light of the broader Buddhist approach to etymology outlined above—as primarily an interpretative, tradition-making, commentarial tool.