Sanskrit text and language
Sanskrit belongs to the Indo-European language family. Within this it forms, together with the closely related Iranian languages, the Indo-Iranian subfamily. Commonly three major forms of Sanskrit are distinguished, namely Vedic, classical Sanskrit, and epic Sanskrit. Vedic is the language of the Indic tribes who entered the South Asian subcontinent probably during the early to mid-2nd millennium BCE, migrating from the Iranian Plateau northwest of present-day Pakistan into the Punjab in eastern Pakistan, northwest of the contemporary Republic of India. The earliest preserved “document” of these tribes is the Ṛgveda. The language of its oldest parts is early Vedic, which is based on a western dialect. The less ancient parts of the Ṛgveda, the Atharvaveda, and the rest of vedic literature are composed in later Vedic, which displays more features deriving from central dialects. In the following centuries, the language continued to change, developing towards classical Sanskrit, which is, like later Vedic, based on a dialect of the central region of India. While it shares many features with later Vedic, the differences in phonology, morphology, and syntax are sufficient to distinguish Vedic and classical Sanskrit. Such a distinction did not receive much emphasis in the Brāhmaṇas, but the mere creation of the Saṃhitāpāṭhas (continuous recitation) from the Padapāṭhas (word-by-word recitation; see also below) already indicates that the language of the Vedas was perceived as distant. Yāska’s Nirukta as well as Pāṇini’s Aṣṭādhyāyī certainly display awareness of the difference between the language of the Vedas and their contemporary language (see esp. AṣṭA. 1.1.16). Pāṇini, for instance, explicitly distinguishes between the (metrical) Sanskrit language of the vedic texts (chandas) and that spoken in his time (bhāṣā). It was his grammar, then, which regulated Sanskrit so effectively that essentially all the classical literature of that language adheres to his rules. This “classical Sanskrit” was no longer subject to the normal laws of linguistic development. The “colloquial” forms of Sanskrit – the most important one being the so-called epic Sanskrit, which is the language of the Mahābhārata and Rāmāyaṇa , and, essentially, the Purāṇas – were, by contrast, still affected by language change.
While the vedic texts do not offer a proper name for their language of composition, in later eras the Indian word saṃskṛta was established as a characterization or denomination of the language we know by the (anglicized) term “Sanskrit.” The earliest usage of this term is found in the Rāmāyaṇa (3.16.14; 5.28.17–19). From about the beginning of the Common Era onwards, saṃskṛta was established as the usual way to refer to Sanskrit. Saṃskṛta, the verbal adjective of the verbal root kṛ- “to do,” along with the prefix sam “together,” generally means “put together,” “adorned,” “purified,” “prepared.” Essentially it refers to heightening the quality of something. Food, for instance, is called saṃskṛta when it is improved by cooking (Kāś. 4.2.16; 4.4.3). In a similar way, the language Sanskrit is believed to be “superior” due to its “built-up” or “regularly formed” character. It is said to be endued with the masculine noun saṃskāra, “composition,” “correct formation” (a masculine noun that, like saṃskṛta, is derivation of sam + kṛ-). With regard to the characterization of language, saṃskāra has two major aspects. Firstly, it may be used for a particularly distinct articulation, when it is intended to insist on sound being “formed correctly” (see MaBh. 1.5.7; MBh. 14.43.22; VākPad. 1.144). Secondly, saṃskāra is defined by Yāska and Kātyāyana (Nir. 2.1 and 4.1; VājPrāt. 1.1) as the regular word formation of Sanskrit, taught by Pāṇini. Saṃskṛta speech, that is, speech “which has saṃskāra” (saṃskāravatī), constitutes an adornment for the speaker (KumSaṃbh. 1.28; ŚatTr. 1.19). It is also saṃskṛta in the derived sense “made fit for,” “prepared for [a purpose],” this purpose characteristically being ritual.
While Sanskrit is claimed to be endowed with saṃskāra, other languages are, because of their lack of saṃskāra, regarded as corruptions of it (NāṭŚā. 17.2). These other languages, the Middle Indic vernaculars, are designated as prākṛta (Prakrit), “original,” “natural,” “ordinary” and even as apabhraṃśa (Apabhramsha), “corruptions” of the elevated Sanskritic speech (KSS. 1.6.147–148; KāvĀd. 1.32–33; KāvAl. 1.16; MaBh. 1.2.24; apaśabda [Apashabda; MaBh. 1.2.3–9; NāṭŚā. 14.5]). Sanskrit and Prakrit, however, seem to be equally appreciated by the Pāṇinīyaśikṣā
Sanskrit remains the major language of religious texts and textual discourse in classical Hindu traditions, so much so that it is sometimes spoken of as “Sanskritic Hinduism”: the Rāmāyaṇa and Mahābhārata, along with the Bhagavadgītā and Harivaṃśa, Purāṇas, Pāñcarātra Saṃhitās, Vaikhānasa Saṃhitās and Śaiva Āgamas, the Hindu Tantras, Stotras, and Māhātmyas, as well as the literature of the philosophical systems and treatises on dharma ( dharmaśāstra ), politics (arthaśāstra), architecture (śilpaśāstra), medicine ( āyurveda ), drama, and many other topics, all of which are composed in Sanskrit. The rich commentatorial literature (Bhāṣya) of India as well as manuals for ritual (Paddhati, Prayoga) continue to be composed in classical Sanskrit or a variety of regional Sanskrit. Apart from Rāmānuja, Madhva, and Nimbārka, most of the so-called sectarian Hindu traditions as well as the bhakti movements, however, use vernacular languages for their teachings.
The acceptance of the Veda is usually regarded as a sign of Hindu “orthodoxy,” and systems or sects that claim to be based upon additional, extravedic revelations are often regarded as illegitimate and unacceptable by “orthodox” thinkers. The Veda is invoked as the source and focus of the unity and identity of the Hindu religions. It is invoked against the internal, sectarian disintegration of the tradition, as well as against the “external” and “heterodox” challenges of, for example, Buddhism (Halbfass, 1991, 16).
In many postvedic Indian traditions, Sanskrit literature is divided into Śruti and Smṛti. The four Vedas, that is, Saṃhitās, Brāhmaṇas, Āraṇyakas, and Upaniṣads, are called Śruti ("Something Heard [from a Person of Authority]) and therefore highly reliable and often unquestionably true. Depending on the traditions, Śruti texts are regarded as eternal or composed by god. The Śruti texts and among them especially the Saṃhitās are carefully preserved, and much emphasis is placed on the purity and correctness of their recitation. By contrast, the concept of Smṛti ("Something Remembered") denotes less authoritative texts. It generally refers to post-upanisḥadic texts: it may denote the Dharmaśāstras alone (MaSm. 2.10) or the Vedāṅga, but it can also include Rāmāyaṇa, Mahābhārata, and Bhagavadgītā, as well as Purāṇas, Āgamas, and Tantras. The Smṛti texts tend to be more fluid in nature than the Śruti ones. Maybe originally the term Smṛti referred to specific personal memories regarding various customs and conventions, including legal ones, and the events of times past, that is, “tradition.” The term “Smṛti” in the relevant sense appears not to occur before Taittirīyāraṇyaka (1.2.1) and Lāṭyāyanaśrautasūtra (6.1.6.13), and “Śruti” not before Mānavaśrautasūtra (182.4). It is, however, by the Mīmāṃsakas that the two terms have been clearly conceptualized for the first time (see below). In later literature the opposition between Śruti and Smṛti is well known (see e.g. RaghV. 2.2).
The text of the Ṛgveda consists of hymns in praise of the vedic gods and goddesses. Its maṇḍalas (books) 2 to 7 are attributed to different families who collected and transmitted these texts orally. In contrast, our knowledge of the authorship of the individual hymns is hazy. However, some of them do mention the poet (kaví) or seer ( ṛ́ṣi ) who composed them (e.g. Viśvāmitra [ṚV. 3.53.12]). Probably around mid-4th century BCE, the Sarvānukramaṇī, a “general index” to the Ṛgveda, attributed each of the hymns to a particular named poet. The poets and seers, it is believed, have a special relation to the gods (ṚV. 1.139.9). And the poems are, like chariots, considered a means of establishing a relationship between men and gods (e.g. ṚV. 5.60.1). Indra, Agni, and the divine pair Mitra and Varuṇa as well as Soma are said to be the source of inspiration of those poems. According to Ṛgveda 8.59.6, it was Indra and Varuṇa who, in the beginning, endowed the ṛ́ṣis with thought and speech, which those seers then subsequently made stable (sthā́nāny asṛjanta) – that is, they made it into poetry (deváttam bráhma [ṚV. 1.37.4]). In Ṛgveda 4.11.2–3, the seer asks Agni for poetic inspiration, and in Ṛgveda 9.99.6 Soma is called the “Lord of Vision.” Although inspiration is given by the gods, the poets play an active role in “crafting” the hymns in ever-new forms. Some of the Indo-European formulas denoting this process of poetic composition are still present in the vedic texts as, for instance, the indication of poetry as carpentry (ápūrvyā purutámāny . . . vácāṃsy . . . takṣam [ṚV. 6.32.1]). The poets were conscious of the process of liturgical composition (ṚV. 10.71; 1.88.4; 6.9.6). In later texts the ṛ́ṣis are said to have “seen” the Veda or, more particularly, the Veda’s stanzas such as sāman, stoma, and mantra. Sometimes it is related that the dharma appeared before the ṛṣi’s eyes. The way of gaining this wisdom is explicitly connected with the notion of a visual perception (TaiĀ 2.9; AṣṭA. 4.2.7; Nir. 1.20; 2.11; 7.3; MBh. 18.5.33).
In the vedic literature, it was, as far as I can see, not questioned in which language songs and hymns for the gods and goddesses should be composed. Using Vedic without exception, the authors of the rgvedic hymns must have regarded this as the appropriate language. A topic of discussion is, however, which words (i.e. always Vedic words) are to be used in addressing the gods. Aitareyabrāhmaṇa (3.33), for instance, states that the gods love mysterious speech that is not directly accessible (parokṣapriya). The poets, therefore, will be careful to not choose plain words. In this context, it may be added that the Ṛgveda (6.24.6) alludes to riddling contests (brahmodya) of the poets – a real and potentially dangerous contest that consisted of encoded questions and equally encoded answers. The speech used for these occasions was also deliberately not easily accessible.
In Mīmāṃsā, the reliability and meaningfulness of the Veda is further backed up by postulating the eternalness of language. Words and the letters that constitute it are unchangeable and ubiquitous; it is only their particular manifestations, caused by articulation of the vocal organs, that are restricted to certain times and places. The meanings of words, being universal, are fixed as well. Finally, a primal, necessary, and unarbitrary (autpattika) relation between word (śabda) and meaning (artha) is postulated (MīmS. 1.1.5). Words could not be expressive of certain meanings as a result of artificial conventions. All this is, of course, valid for Sanskrit alone. Kumārila, a Mīmāṃsā teacher, specifically says that the knowledge of the inhabitants of Āryavārta is authoritative as regards words and their meanings (TVā. 220.3–4; for Patañjali’s notion of Āryavarta, see below).
Nyāya and Vaiśeṣika, the schools of logic and nature philosophy, accepted the authority of the Veda too. But these philosophical systems developed out of the tradition of debate and were not genuinely affiliated with the Veda. They defended the Veda in a retroactive manner. Nyāya postulated verbal testimony of a trustworthy person (āpta) as an independent means of obtaining knowledge (śabdapramāṇa [NyāS. and NyāSVṛ. 1.1.7–8]). The role of āpta is particularly important with regard to the special knowledge contained in the Veda. At an early stage of the Nyāya tradition, it was believed that the Veda was composed by “trustworthy persons” who were endowed with a direct perception of dharma (NyāS. and NyāSVṛ. 2.1.68). At a later stage of development, it was postulated that the Veda is the reliable word of Īśvara (ĀtTatVi. 436). The validity of the Veda was “verified” by comparison with the original āyurveda, whose truth and effectiveness are said to be supported by “empirical evidence.” As the Veda is composed by the same author(s), it must be infallible too (NyāSū. 2.1.68; Chemparathy, 1983, 40–57).
For Śaṅkara, the famous exponent of Advaita Vedānta, the Veda is authoritative; all the more so concerning the ultimate, liberating truth of ātman / brahman . Only by means of the Veda is human insight possible. It is a “sun which shines upon reality and appearance” (BrSBh. 2.1.1; BĀUBh. 2.1.20). Śaṅkara allows for cooperation of “revelation” (veda, śruti) and “reason” (yukti, tarka). However, “reason,” though being a valid form of inference, remains “human cognition” (puruṣabuddhi; Halbfass, 1991, 135), and as such it cannot claim any authority that would be independent from the Veda. One needs to follow “the path which has been shown by revelation and authoritative teachers.” The problem of the relation between revelation and reason is connected to Śaṅkara’s understanding of Śruti and Smṛti – the latter, that is, extravedic philosophical traditions, being subordinate to Śruti. They are acceptable only insofar as they serve the goal of understanding the Veda (BrSBh. 2.1.3).
The access to the Veda is restricted to certain parts of the vedic society. This restriction is common during the time of the Gṛhyasūtras. According to these texts, the initiation is performed only for Brāhmaṇas (Brahmans), Kṣatriyas/Rājanyas, and Vaiśyas (see e.g. GauGS. 2.10; HirGS. 1.1.17). In the Mīmāṃsā discussion, the right ( adhikāra ) to learn the Veda and thereby take part in the ethical realm of dharma is also explicitly linked to the varṇa ordering (MīmS. 6.1.33). According to later authors, most prominently Manu, male Brahmans were the only ones who could learn, teach, and use the Vedas. Śūdras were, in Manu’s view, not even allowed to hear these texts being recited (MaSm. 4.99; caste).
Like curses and oaths, ritual speech is performative. Being “speech acts,” ritual formulas are effective in the sense that the desired result is believed to happen by the act of uttering the according formula. During the oldest period of religious practice, gods and goddesses were begged for aid by the recitation of hymns. In later epochs, however, the power associated with correctly uttered ritual speech was thought to be absolute. It was believed that one can “force” the gods to grant the desired results of the ritual by the performance of correct actions and flawless liturgy (AitĀ. 1.3.2; 5.1.5). The importance of proper meaning and impeccable accentuation is illustrated by Taittirīyasaṃhitā (2.4.12.1; MaBh. 1.2.12): Tvaṣṭṛ, seeking to avenge his son’s death at Indra’s hands, performs a sacrifice whose purpose is the production of an enemy to destroy the god. Tvaṣṭṛ’s vendetta fails because – out of ignorance or inadvertence – he places the vedic pitch accent (udātta) on the first part of the compound rather than the last, thus forming an attributive compοund (bahuvrīhi) rather than the desired determinative compound (tatpuruṣa). In this way he creates a being “whose killer is Indra” (indraśatrú) rather than, as he had intended, a “killer of Indra (indraśátru).”
Vāc appears as an independent force that is difficult to control. Even the vedic gods (devas) struggle to obtain and maintain her. In a famous myth related by Śatapathabrāhmaṇa (3.2.1.18–24), the devas and the asuras, both sons of Prajāpati, received the inheritance of their father: the gods obtained mind, sacrifice, and “that world”; the asuras obtained Vāc and “this world.” The gods then made Vāc their own by employing the sacrifice. The asuras – being deprived of Vāc and thus using improper speech in ritual – were defeated. In the following (ŚBr. 3.5.1.21–22; MaBh. 1.2.12), Vāc is slighted by the gods and asuras and goes away from them angrily. Only after having received a boon does she finally go over to the gods.
In the second half of the 1st millennium BCE, the vedic cult was systematized by the composition of the Śrautasūtras and Gṛhyasūtras, compendiums for the solemn and nonsolemn rituals, respectively. These compendiums prescribe many citations from the Veda, which are named as mantra, ṛc, yajus, sāman, or brahman, and so on, as liturgy in their ceremonies. As at least the Gṛhyasūtras still provide a direct or an indirect basis for contemporary ritual traditions, these Sanskrit citations are employed in Brahmanical and also popular Hindu traditions up to the present day, especially for the performance of the life-cycle rituals ( saṃskāras), although today’s priests and ritual participants might have little or no command of Sanskrit. The amount of information that mantras provide is, at least today, not their most significant aspect. They are, above all, “real” components of a past world, and as such they have to be employed and enacted, not “understood.”
Mantras are of utmost importance not only in orthodox Hindu traditions but also in Tantra. Here, they are considered soteriologically central acoustic formulas. Mantras are described as consisting of Sanskrit phonemes whose specific energy is that of Lord Śiva (ŚiSVim. 2.3). The highest form of the word lies beyond language. Tantric traditions extol bījamantras (lit. seed mantras) such as oṃ. These are syllables of extraordinary power that are clearly separated from ordinary language.
The attainment of speech – that is, the ability to learn and speak proper Vedic or Sanskrit – needs to be ritually secured shortly after birth. Usually the Gṛhyasūtras teach a ceremony for the “production of intelligence” (medhājanana) for the new-born child. The central element of this ritual is the reciting of a mantra into the child’s ear that contains the word medhā (see e.g. ĀśvGS. 1.15.2). In the ceremony taught by Śāṅkhāyanagṛhyasūtra (1.24.9–10), however, intelligence is equated with speech. Here, the word vāc is thrice muttered into the child’s ear, and the following mantra praises the great and sweet-sounding goddess Vāc and asks her to rejoice in the child. Similarly, in the medhājanana ritual of Kauśikasūtra (10.16–19), an Atharvaveda song (AVŚ. 4.30) is employed in which the goddess Vāc describes herself.
In ancient times not only was there a distinction drawn between proper and improper speech, that is, Sanskrit and non-Sanskrit, but social status (human/divine; male/female) was marked by differences within the Sanskrit language. Employing Indo-European notions, a distinction is made between words used by gods and words used by men. What the latter know as carman (skin) is called śarman (shelter) among the former (ŚBr. 1.1.4.4; 3.2.1.8). From Śatapathabrāhmaṇa (10.4.6.1) it appears that gods, gandharvas, asuras, and humans use different synonyms for the same thing meant. Thus, while later authors explicitly declare Sanskrit to be the “language of the gods” or the “heavenly language” (daivī vāc [KāvĀ. 1.32; VākPad. 1.182]), it appears that in vedic times a different way of expression within Sanskrit was ascribed to the gods (see ṚV. 1.164.45).
Women, in the Ṛgveda, also speak Sanskrit (in contrast with later times), but they do so differently from men. The vedic poets used certain linguistic features to characterize the speaker as female. Two of these features are the secondary -ka suffix and the perfect optative. The diminutive -ka forms in the Ṛgveda are employed in unusual, colloquial contexts, some of which involve women’s speech, often sexually charged or vulgar (see e.g. ṚV. 1.126.7; 10.86.7; VājSa. 23.22). Female speakers establish their identity by using perfect optatives (ṚV. 1.179; 10.10; 10.28.1), while in the whole of the Ṛgveda, subjunctives and imperatives are far more common (see Jamison, 2008).
Finally, the famous stanza of the rgvedic riddle–hymn informs us that "speech (vāc), [is] fourfold, [it] is measured out in quarters: those [only] wise Brahmans understand. Three portions, hidden [within a cave], do not move. The fourth portion of vāc, men speak" (ṚV. 1.164.45). Semantically complex, this stanza may allude to a social stratification within Sanskrit.
Maintaining its high or “sacral” status in later times, Sanskrit was and often remains restricted to the community of the Brahmans or, more generally, the twice born (men). Sāyaṇa in his commentary of the Ṛgveda offers no fewer than four interpretations of the above-cited stanza. However, in virtually all interpretations, the three hidden portions of vāc are assessed by an elite group, that is, the male Brahmanical world, or some aspect of it, and all equate the fourth portion of vāc to the language known as laukika (worldly) or vyavahārika (ordinary, mundane). Manu, however, treats the varṇa ordering separately from the languages spoken by different persons and groups. He says that
"all the castes (jāti) in the world that are outside those born from the mouth, arms, thighs, and feet are declared dasyus [by tradition, regardless of] whether their language is foreign (mleccha) or Ārya." (MaSm. 10.45)
Methodically, etymology (nirukta, explained, past participle of nir + vac-, “to express clearly”) can be considered as a continuation of the explanations of words in the Brāhmaṇas. Etymology is – most prominently – represented by Yāska’s Nirukta. Yāska opposes Kautsa for whom mantras are strictly meaningless. For him, there was no doubt that the vedic mantras have meaning. He holds that the words in the vedic verses are in principle the same as those employed in daily usage (Nir. 1.15–16).
Among the means by which the correct transmission of the vedic texts was achieved was the Padapāṭha (word-by-word recitation). While in the “continuous recitation” (Saṃhitāpāṭha) all the saṃdhis (euphonic junctions between words) are applied, in the Padapāṭhas each word of the text was repeated separately, hence unaffected by the rules of saṃdhi. Most compounds, derivates, and inflected forms are analyzed. Additionally, special patterns of repetition (vikṛti) of the separated words were constructed to minimize the risk of forgetting words. For instance, in the “step-by-step recitation” (Kramapātḥa) the saṃdhis are applied for pairs of words. In the “twisted recitation” (Jaṭāpāṭha) each pair of words is repeated thrice, the middle repetition being in inverted order.
Pāṇini describes a living language. He distinguishes in his grammar between chandas (sacred literature) and bhāṣā (spoken language), mentions usages particular to northerners (udīcās) and easterners (prācās), and notes a difference between usage north and south of the Vipāśā (Beas) River (AṣṭA. 4.2.74). Later grammarians such as Patañjali acknowledge that the language Pāṇini describes is the speech of persons referred to as śiṣṭa, an elite group of model speakers characterized as much by their moral qualities as by their language. In his Mahābhāṣya (3.173.19–174.15), Patañjali says that the śiṣṭas are restricted to the region of Āryāvarta, the region between the country Ādarśa, the mountain Kālakavana, the Himālayas, and the western Vindhya range (MaBh. 1.475.2–3). He believes that they alone speak proper Sanskrit, even without learning it with the help of grammar. Hence, the grammarians look up to them in formulating their rules. For Bhartṛhari the śiṣṭas are no longer a contemporary community of standard speakers of Sanskrit, but the ancient sages of a “golden age” of Sanskrit grammar. By his time, the founding fathers of Sanskrit grammar had achieved the status of ṛṣis (e.g. VākPad. 2.481). From about the 13th century, Patañjali is even believed to be an incarnation of the divine grammarian, the great snake Śeṣa.
In formulating his grammar, Pāṇini uses Sanskrit as a metalanguage to which he refers with the term upadeśa (instruction; e.g. AṣṭA. 1.3.2). Including regular Sanskrit words, its difference from proper Sanskrit is principally only visible in the artificial elements such as peculiarities in phonology, nominal flection, and the use of cases (Wezler, 1975, 29). Patañjali as well as Śabarasvāmin (5th cent. CE), the author of Mīmāṃsābhāṣya, recognized the difference between Pāṇini’s metalanguage and the object-language Sanskrit that he described. Unlike Sanskrit they did not regard the metalanguage as eternal (nitya) because in the process of word formation taught by Pāṇini, certain elements of the metalanguage have to be added and then removed as, for example, the end marker (anubandha) iṭ (see AṣṭA. 1.1.60). Kātyāyana, Patañjali, and Bhartṛhari use proper Sanskrit (as it is taught by Pāṇini) in analyzing this language. Their discussion of making statements about Sanskrit while at the same time using it for description centers round Pāṇini’s statement:
"A word [in a grammatical rule] which is not a technical term denotes its own form." (svaṃ rūpaṃ śabdasyāśabdasaṃjñā; AṣṭA. 1.1.68)
Thus, when the word agni (fire) is used in a grammatical statement, the signatum is not the fire but the same as the signans (Brough, 1951, 29).
In their analysis of Sanskrit, the grammarians made a distinction between “words” or other linguistic entities and the corresponding physical sounds. Against the common-sense view that a word consists of the actual sounds of the instance, they established the sphoṭa doctrine, which is of central importance for the philosophy of language. For Patañjali and Bhartṛhari, sphoṭa is the constant part of the speech, while dhvani or nāda denotes the sound that varies according to pronunciation and level of sound (see MaBh. 1.181.19–25; VākPad. 1.76–79). For Patañjali, sphoṭa can be a single phoneme or a fixed pattern of letters. For him, sphoṭa does not necessarily convey meaning. For Bhartṛhari, in contrast, sphoṭa is the “real word,” that is, the meaning-bearing unit that is without parts and sequence but revealed by sequential physical sounds. Similarly, Bhartṛhari teaches that a sentence and even its meaning are indivisible (VākPad. 2.7–16). He explains that in actual communication the meaning of a sentence is not understood as a sequence of the constituent word meanings but in a flash of understanding (pratibhā) as a single unit. Only by the deliberate subsequent division by grammarians through theoretical analysis does a sentence seem to have component words.
On the basis of the sphoṭa theory, a metaphysical superstructure was erected in which the transcendental word (śabdabrahman) was seen as the first principle of the universe. Bhartṛhari’s Vākyapadīya (1.1) begins with an assertion that the ultimate reality (brahman) is the imperishable principle of language and that the evolution of the entire world originates from this language reality in the form of its meaning. Language is the vehicle of all human activity, and the knowledge of language cannot come about without grammar. This grammar is a gateway to salvation, a treatment for the defilements of language, and the leading branch of knowledge. Through the science of grammar, which includes a language-based spirituality, one can attain the state of the highest language reality (VākPad. 1.11–22). Bhartṛhari’s effort to elevate grammar to the status of not only a philosophical system but also a spiritual path toward salvation is an elaboration on ideas scattered throughout Patañjali’s work. Patañjali discusses at length reasons to study grammar. In this context, he explains that one should learn grammar because the great god named in Ṛgveda (4.58.3) who is said to enter mortal beings is speech (śabda; MaBh. 1.3.20–22).
The Theravādins assume, as a matter of course, that the language of the Buddha was Magadhi and that it was the same as Pali, which is used for their canonical texts. They further regard Pali/Magadhi to be the original language of all beings (mūlabhāsā; RūSi. 2.1). In the Pali canon, it is prescribed that the Buddha’s words should be acquired in one’s own speech (sakāya niruttiyā; CuVa. 5.33.1 [= VinP. 2.139]). In order to preserve the Buddha’s words, two converted Brahman brothers had suggested putting it in chandas (lit. verse). The commentary sets chandas in relation to the Veda and, possibly, Sanskrit (vedaṃ viya sakkatabhāsāya vācanāmaggaṃ āropem; SāPā. 6.1214). However, the suggestion is objected to: “One who might put [the Buddha’s word in chandas] commits an offence an account of ‘wrong action’ (dukkaṭa).”
The canonical texts of the Jain traditions are composed in Ardhamagadhi, and later texts in Jain-Maharashtri, Jain-Shauraseni, and Apabhramsha. Like Sanskrit by the Brahman tradition, Ardhamagadhi is declared “the language of the gods” by the Jaina tradition (BhaVaī. 108). The 11th-century Śvetāmbara Jaina scholar Namisādhu expresses the same opinion in his commentary on Rudraṭa’s Kāvyālaṃkāra (devāṇaṃ bāṇī; KāvAl. 2.12). He further grants priority to prākṛta against saṃskṛta when he explains the former as a compound whose first constituent is prāk equivalent to pūrvam (before). Rudraṭa then claims that prākṛta refers to what was brought about “before” saṃskṛta. The latter is said to result from prākṛta through purification by grammar.
However, the early preference for vernaculars in Buddhism and Jainism was eventually abandoned, and both the traditions gradually switched to the use of Sanskrit. This shift may have been motivated by the rising (worldly) prestige of Sanskrit in the post-Maurya periods (see below). By the 1st century BCE, some Buddhists began to partially Sanskritize their texts. The resulting “mixed Sanskrit” has been called “Buddhist hybrid Sanskrit” or “quasi-Prakrit-cum-Sanskrit” and is the language of numerous important texts mostly by the Mahāsāṅghikas, such as the Mahāvastu, a long biography of the Buddha. By the 5th century CE, Buddhist traditions, that is not only the Mahāsaṅghikas but also the Sarvāstivādins, Mūlasarvāstivādins, Vaibhāṣikas, Sautrāntikas, and the various Mahāyāna traditions, adopted Sanskrit even for their canonical texts. Moreover, Buddhist poets and scholars have composed important works in Sanskrit. Aśvaghoṣa (2nd cent. CE) is the author of the poetical work Buddhacarita. And, to name just one further example, the famous Buddhist scholar Amarasiṃha has written the Amarakoṣa (or Nāmaliṅgānuśāsana [Instruction Concerning Nouns and Genders]), a synonymic dictionary of Sanskrit.
During the earlier centuries, the Jainas were less open to the Sanskritization movement. The logicians of the Śvetāmbara (and to a lesser extent Digambara) wrote commentaries and independent works in Sanskrit, but not in large numbers until the 8th and 9th centuries. An example of Jaina literature composed in Sanskrit is the Tattvārthadhigamasūtra by Umāsvati, a highly regarded philosophical treatise. From about the 13th century, the Śvetāmbara Jains of western India have produced extensive literature of narrative, chronicle, and hagiography written in Sanskrit.
The epics’ and Purāṇas’ originally oral character is revealed by their text structure. Both epics are set within a frame narrative that centers on the respective heroes. But in the course of time, more religious and didactic material was included in them, especially in the Mahābhārata. The Purāṇas are invariably framed in a dialogical structure, their contents being manifold from the beginning. On the one hand, they relate by definition “stories of the old days,” such as mythological narratives going back to the Vedas and the Mahābhārata. On the other hand, they contain a miscellany of customs, rituals, medical knowledge, and many other topics.
The original authors and oral transmitters of the various epic cycles would have been the sūtas or bards. The sūtas were attached to the courts of chieftains and recited in short songs and at major festivals the glorious deeds of their lords. These bards, most probably, accompanied their masters into battle or to the hunt as charioteers and told stories for their entertainment. It is generally agreed that the epics, having originated in a Kṣatriya context, were then taken over by Brahman redactors. For the Purāṇas, this view has also occasionally been claimed but more often rejected. Some of these texts, however, allude to their previous recitations in royal courts and villages.
The Indian tradition generally ascribes the authorship of the Mahābhārata (1.1.15) as well as many Purāṇas to the sage Kṛṣṇa Dvaipāyana, who is also called Vyāsa (lit. “Divider”) as he separated the Veda into its four parts (ṛc, yajus, sāman, and atharva). It is believed that he had direct access to the knowledge of everything past, present, and future. In some cases, the texts composed by Vyāsa are subsequently authorized by a god, often Brahmā. In other cases, the specific god of a Purāṇa is held as its original author (e.g. AgP. 1.7–8) who would then pass it on to Vyāsa or another sage. Several stages of transmission follow in the Mahābhārata and some Purāṇas until the respective narratives are told to humans. The place of transmission is often the Naimiṣa forest, and at some stage or the other the sūta Lomaharṣaṇa is involved in reciting the Mahābhārata or a puranic text. (For the Rāmāyaṇa’s view of its own creation, see below.)
The epics and Purāṇas are usually classified as Smṛti. Mahābhārata and Purāṇas are, however, also popularly called the “Fifth Veda” (MBh. 1.57.74; ChāU. 7.1.2) as these texts understand themselves as an expansion of the Veda (see e.g. VāyP. 1.200). They often claim to comprise all valuable knowledge and surpass all previous expressions of important knowledge. The classification of the Mahābhārata as Smṛti becomes especially controversial with regard to the Bhagavadgītā, as this text, being a later addition to the epic, is believed to be spoken by the god Kṛṣṇa. For many Hindus it is therefore deemed more authoritative than some of the Śruti texts themselves.
In early scholarship it had been speculated that because of the epics’ and Purāṇas’ generally “popular” nature, they must have been composed in one of the Prakrits and only later transposed into Sanskrit. This view has long been abandoned, but it is true that the Sanskrit of the epics (and Purāṇas) is different from Vedic as well as classical Sanskrit. This variety has sometimes been called “Kṣatriya Sanskrit,” but today it is generally referred to as epic Sanskrit. It is deemed a more “natural” Sanskrit than the classical variety following Pāṇini. In epic and puranic traditions themselves, the Sanskrit language is generally associated with the twice born. The social status implied by speaking Sanskrit is explicitly referred to in the Rāmāyaṇa. Whenever the sagacious monkey–hero Hanumān needs to establish his trustworthiness with one of the central epic figures, he must indicate that he can speak like a Brahman. Sanskrit is also used by demons when they disguise themselves as Brahmans (Rām. 3.16.14; 5.28.17–19). The demon-king Rāvaṇa, for instance, had approached Princess Sītā pretending he was a Brahman. Later in the plot, Hanumān therefore decides to address Sītā in “the language of men (mānuṣyaṃ vākyam, “a human utterance”), with meaning [but without saṃskāra],” that is most probably a Middle Indo-European dialect. He says, “if I speak saṃskṛtā vāc, as though I were a twice-born, Sītā will be frightened, thinking I am Rāvaṇa” (Rām. 5.28.18–19; 4.3.27–33).
The Bhāgavatapurāṇa, an important Sanskrit Purāṇa of about the 10th century, plays a major role in the propagation of the bhakti movement to North India. This movement originated in the south of the subcontinent where the Āḻvārs, a group of mystics (7th–9th cents.), composed their highly poetical, ecstatic songs in Tamil, the standard language of South India. The language choice of the Āḻvārs prevented the propagation of their songs to the Sanskritic traditions of North India, where Sanskrit was the lingua franca. These songs were adapted by the Bhāgavatapurāṇa into Sanskrit and thus made them available for traditions of North India. The Sanskrit epics, in contrast, have been adapted into various Dravidian as well as New-Indo-European languages. Kampaṉ (12th cent.) is the author of the Irāmāvatāram, a Tamil Rāmāyaṇa. In the 16th century, Tulsīdās composed another famous adaption of this epic, the Rāmcaritmānas, in Hindi. In the 1980s, both the epics were reworked for the first time for television broadcast in which a highly Sanskritized form of Hindi was used.
The rise in the public availability and use of Sanskrit must have begun slowly after the fall of the Mauryas at the hands of the Brahman Puṣyamitra Śuṅga (a contemporary of Patañjali; MaBh. 2.123.3–4), who assassinated the last Maurya king. An inscription at the end of the 1st century BCE from the region of Ayodhya, mentioning King Dhana(deva?) as “sixth [in generation] from the general Puṣyamitra,” is the sole Sanskrit document of the Shungas and the first preserved Sanskrit inscription. The revolutionary use of Sanskrit in the inscription might have aimed at religiopolitical prestige. The Shunga takeover strived to stabilize vedic religion. The dynasty after the Shungas is believed to be the Kanvas, to which the first inscription in real Sanskrit is attributed. It was found in Ghosundi, Rajasthan. These early Sanskrit epigraphs consist of one- or two-line records commemorating a vedic or quasi-vedic rite.
The earliest longer Sanskrit inscription is that of the Shaka ruler Rudradāman. It dates from approximately 150 CE and is located at Junagadh in the Girnar hills of Saurashtra, Gujarat. It commemorates the repair of a tank near Girnar, which was carried out by King Rudradāman. This inscription inaugurates the writing of benedictory introductions (praśasti) to royal inscriptions in a highly poetical and hyperbolic style. Rudradāman was a “foreign” ruler with a Sanskrit name (his ancestors still had Iranian names) belonging to the Iranian Kshatrapas. His choice of Sanskrit was a clear break with the tradition of writing in Prakrit. It appears to be a method to legitimize his reign and his acquisition of “Indian” identity. For similar reasons the Scythian and Kushana rulers (1st–3rd cents. CE) adopted a form of hybrid Sanskrit for their records. As foreigners, they were inclined to patronize the elite language in an effort to legitimize their rule and emphasize their own Indianization. This so-called Epigraphical Hybrid Sanskrit is neither fully Sanskrit nor fully Prakrit but combines the characteristics of both. By far the majority of those inscriptions have been found in the region of Mathura, whence they radiate toward the northeast and southwest and thus parallel the foundation and spread of the kingdoms of the Scythian and Kushana rulers in the heartland of India.
In the Gupta epoch (4th cent. CE), Sanskrit reached significantly beyond the domain of liturgy and its sacral auxiliaries. The adoption of Sanskrit for political records is mirrored in the rise of worldly poetry in Sanskrit, which is also intrinsically connected to the Gupta monarchy. Besides being established as the lingua franca, classical Sanskrit also became the epigraphic language par excellence during the reign of the early Gupta emperors. The turning point appears in the inscriptions of Samudragupta (mid- to late 4th cent.), especially the Allahabad Pillar Inscription, which is often held up as a model of the high classical literary style of the mixed prose and verse class (called campū). Sanskrit was also extensively used in coinage. Here, it was again under the Guptas that pure Sanskrit was adopted in coinage. The inscriptions on Samudragupta’s famous coins describe him as “the king of kings of irresistible prowess, the protector of the earth who wins heaven.” However, the western Kshatrapas, rulers of parts of the Shaka Empire, had used a form of Sanskrit for their coinage earlier.
Among other dynasties, mostly of the post-Gupta eras, mixed epigraphical records became common. While the pragmatic message of such documents is conveyed in the local language, that is, Prakrit and Tamil as well as Persian and New Indo-Aryan languages, the Sanskrit part is usually of religious and royal character. It is mainly used for an introductory invocation of the gods (maṅgalācaraṇa), a fixing of mythological and genealogical succession, a catalogue of kingly traits of the dynasty, and a eulogy of the ruling lord. These are all elements of what became the standard praśasti style.
Kāvya is, as a rule, associated with the glory of the royal courts. The central role that the figure of the king plays in Kālidāsa’s dramas and in his epic Raghuvaṃśa strongly suggests that he enjoyed royal patronage. Kālidāsa’s patron was probably Candragupta II, who ruled most of North India from about 375 to 415 CE. Likewise, Kaniṣka, a sovereign of foreign origin, is considered to have patronized Aśvaghoṣa, a Sanskrit poet of Buddhist faith, living in the 1st or 2nd century CE.
Kāvya requires considerable literary education to be composed and appreciated. It is a form of art characterized by a sometimes ostentatious display of erudition. The kavis choose their language carefully in order to achieve a particular effect. Extensive use of speech figures is made, especially of metaphor and simile. In the often hyperbolic compositions, varied and complicated meters are employed. The metrical poetry is usually divided into Laghukāvya (poetry of the minor form) and Mahākāvya (poetry of the major form). Poems of the major form are almost invariably divided into chapters and therefore also called Sargabandhas (“composed in cantos”). Poetry that is a mixture of prose and verse is called campū, while works written in the heavier prose style are subdivided into two categories, “(fictive) story” (Kathā) and “(true) story” (Ākhyāyikā; see Lienhard, 1984).
Kāvya also includes plays and drama, which are traditionally enacted in the theatre. The earliest extant textual source for drama is the Nāṭyaśāstra (2nd cent. CE), a Sanskrit dramaturgical manual attributed to Bharata. Today, the South Indian Sanskrit theatre kūṭiyāṭṭam is considered to be the only surviving traditional theatre form The text of the drama is recited in different pitches (svaras) similar to the traditional Veda recitation. By a system of hand gestures, the Sanskrit texts are simultaneously displayed according to Dravidian morphology.
Literacy is referred to in Kāvya almost from the genre’s beginning and often regarded as a precondition to Kāvya, as some traits of Kāvya can be understood properly only in written form, as, for instance, the figurative poetry (Citrakāvya). In some Citrakāvyas, double meaning is used throughout. These are called Yamakakāvya. The double epic Rāghavapāṇḍavīya tells both the Mahābhārata and Rāmāyaṇa stories simultaneously.
Major topics of Kāvya are love, nature, panegyric, moralizing, and storytelling. It mainly is situated in the context of kingship and court life. Often, literary devices are applied to traditional subjects and themes derived from early popular epics. Kālidāsa’s drama Raghuvaṃśa (The Raghu Lineage) offers the mythical prehistory of the Ikṣvāku and, concomitantly, of the Gupta family. His drama Kumārasaṃbhava (The Birth of the [God] Kumāra) is a camouflaged allusion to the birth of Kumāragupta (415 CE), the son of Candragupta II. Like this work, other Kāvyas are similarly built on mythological narrative or take up stories from the epics. Still others are devotional throughout. Jayadeva (Bengal, 12th cent.), for instance, describes in his Gītagovinda (Songs of the Cowherd) the love of Kṛṣṇa and Rādhā. Likewise the Karṇāndana (Delight of the Ears), a poem from the Rādhāvallabhī tradition, is centered on Kṛṣṇa’s lover Rādhā. However, the Kāvya genre is not predominantly religious, neither in intent nor in character.
The literary theories that developed with the invention of Kāvya define certain languages as permissible media for literary production and thereby exclude others. The number of primary literary languages is mostly three or four. Sanskrit and Prakrit are invariably mentioned; the third place is usually occupied by Apabhramsha or Deshabhasha ("Languages of Place"). The last language mentioned is Paishaci or Bhutabhasha ("Speech of Demons [ piśācas, bhūtas]). Bharata, in his Nāṭyaśāstra (17), discusses the speech forms that are to be represented on the stage, and who can use them. He considers three languages: Sanskrit, Prakrit, and Deshabhasha (see KSS. 1.6.147–148). Bhāmaha (7th cent.) teaches that Kāvya is “threefold in being composed in Sanskrit, Prakrit, or Apabhramsha” (KāvAl. 1.16; KāvĀd. 1.32). Rājaśekhara, in turn, claims to be an “expert in all languages” with a reference to the four literary languages, including Bhutabhasha (Balarām. 1.11; VāgAl. 2.1–3).
The choice from among the primary literary languages was, for all theorists, largely determined by the genre in which one wrote. Thus, for instance, the Mahākāvya and the Ākhyāyikā were to be written in Sanskrit, the Skandhaka (that is, Mahākāvya with different metrical organization) and Gāthā (song) only in Prakrit (KāvAl. 1.28, 37–38; ŚṛṅPr. 11; DhvaĀL.3.7). Rājaśekhara, in his Kāvyamīmāṃsā, says that
"a given topic will be best treated in Sanskrit, another in Prakrit, or Apabhraṃśa, or the language of the spirits, bhūtabhāṣā; others still in two, or three, or all four languages. The writer whose mind is sharp enough to distinguish [among these possibilities] will win such fame as spreads across the universe."(KāvMī. 9; trans. Pollock, 2006, 94)
The four languages are far from being equally valued. Bharata regards the idealized speakers of the Sanskrit language as pure (saṃskṛta) and contrasts them to barbaric speakers ( mlecchas; NāṭŚā. 17.28). In Sanskrit drama, as for instance in Kālidāsa’s plays, Sanskrit is the language spoken by twice-born men: the king, his advisers, and others of high status in religious or political spheres. It is occasionally spoken by a woman of learning (see MālA.). However, women were, as a rule, linguistically characterized by Prakrit, that is, Shauraseni. Similarly, the Brahman buffoon (Vidūṣaka) as well as various minor characters also spoke different Prakrits. The hierarchy between (the) different languages is most explicitly illustrated by Rājaśekhara, in his tale of the origins of literary culture. He imagines a poetry giant (Kāvyapuruṣa) – whose mouth consists of Sanskrit, arms of Prakrit, groin of Apabhramsha, feet of Paishaci, and chest of mixed language (KāvMī. 3) – as the primal being of literature. Daṇḍin praises Sanskrit in his 17th-century work on literary theory, the Kāvyādarśa (Mirror on Literature; 1.33), by characterizing it as “the language of the gods (daivī vāc), taught [to men] by the great sages.”
The “purity” and “built-up” character of Sanskrit serves Kāvya better than any other languages. Because, as Bhāmaha in his Kāvyālaṃkāra (1.16), states, Kāvya denotes “a unity of word and meaning,” a text in which form and content require and receive equal attention. Especially Sanskrit offers writers a rich opportunity of varying linguistic elements and of putting them in any order they choose so as to obtain the maximum effect. The great majority of authors concentrate on the sound qualities of their sentences and verses and do not hesitate to make full use of the possibilities their language offers them. The language of Kāvya went even further in that poets were constantly enlarging the existing number and scope of synonyms. In certain fields old Indian poets have built up a real synthetic vocabulary that is peculiar to the field.
Sanskrit has retained its religiously superior status in the modern period. It remains the ritual language par excellence of classical Hinduism as well as many popular Hindu traditions. Its literature continues to be regarded with respect, even though the understanding of religious sentiment nowadays almost always takes place through one’s vernacular. Vivekananda, the founder of Ramakrishna Mission, suggests that
"the ideas should be taught in the language of the people; [but] at the same time, Sanskrit education must go along with it."
He views Sanskrit not as a medium or a spiritual message, but rather as a means to raise the condition of the lower castes. Swami Dayananda Saraswati, founder of the Arya Samaj, directed his message in his first years of preaching mainly to the top layer of the Indian society, that is, the Brahman community to whom he spoke in Sanskrit. Later he switched to Hindi, with Sanskrit playing only a nominal role in his teachings. The Theosophical Society in its aim to return to traditional Indian values also ascribes Sanskrit a high status. H.S. Olcott, one of the founders of the Theosophical Society, therefore established the Adyar Library, which collects manuscripts up to the present day.
Up to the 18th century, Sanskrit unquestionably remained a very important language of arts and sciences alike. Jagannātha Paṇḍita (17th cent.) is one of the major Sanskrit poets, and Nāgeśabhaṭṭa Kāle (d. 1755) a respected grammarian. He composed the Paribhāṣenduśekhara. Moreover, many Sanskrit commentaries on a variety of subjects are handed down to us. Though a controversially discussed question (see Pollock, 2001; Hanneder, 2002), it is a matter of fact that up to the present day, Sanskrit remains an important language of Kāvya poetry and, to a lesser extent, traditional scientific literature. Many poems, short novels, novellas, and Mahākāvyas are even nowadays written in Sanskrit (for examples, see Hanneder, 2002).
Sanskrit is still actively encouraged in Indian education. The traditional way of learning Sanskrit is within the guruśiṣyasambandha (master–disciple relationship). Thus, the major location in which speakers learn the language was not the home but the household of a guru or other institutions of learning (pāṭhāśālā and saṃskṛtavidyāpīṭha). The Kendriya Sanskrit Vidyapeeth in Tirupati has, since its founding, conducted all of its instruction in Sanskrit, whether on traditional or modern subjects. The traditional system was considerably weakened by the introduction of a dual system of Sanskrit education. The methods of teaching at government schools, colleges, and universities follow the English system. However, up to the 1960s, students at Varanasi Sanskrit University also had to speak and understand Sanskrit well enough to attend classes taught in this language (Hock, 1992, 18).
It is an often-cited fact that the census of India lists a few thousand people who claim Sanskrit as their mother tongue. In the year 1971 it listed 2,212 persons, 6,106 in 1981, 49,736 in 1991, and 14,135 in 2001. There is nothing in the data to indicate the extent to which people claiming Sanskrit as their mother tongue or as an auxiliary language can actually speak it. The facts suggest that the large numbers of persons declaring Sanskrit as their mother tongue in the censuses either are exaggerating or have a different interpretation of the term “mother tongue,” not as “native language” but as language toward which one feels just the same as toward his or her own mother. The oral use of the language is generally limited to the recitation of short verses (ślokas). However, even though Sanskrit is nowadays not spoken by a significant number of persons, Sanskrit words are included in śuddha (pure) Hindi with slight changes such as the omission of final -a. Furthermore, Sanskrit is used as a source of new words for new concepts in Indian languages, so that a speaker of any Indian language knows some Sanskrit words, albeit in adapted forms. Sanskrit was, however, among the other 14 (now 22) spoken languages listed that were introduced into the Eighth Schedule of the Constitution of India (1949) as a recognized language of the new State of India. In the mid-20th century, there have even been attempts to substitute English as the official language by a simplified Sanskrit. The Sanskrit Commission, which met in 1957 parallel to the language commission, did not see any scope for Sanskrit as an official language but only as a language of classical Hinduism. The commission advised the government to integrate Sanskrit into the school curricula, advocate traditional Sanskrit schools, preserve manuscripts, and so on. The inclusion of Sanskrit among the other recognized languages still ensures funding for Sanskrit colleges and universities, and for a national organization to stimulate the study of the language. The year 1999–2000 was declared the “Year of Sanskrit.”
Sanskrit also gets support from Hindu nationalist and fundamentalist movements. The Rashtriya Swayamsevak Sangh (RSS) passed in March 1958 a resolution concerning the national language policy. Though Hindi is wished for as the official language, Sanskrit is viewed as the root language, the “Rashtra Bhasha par excellence.” It is suggested that “instruction of Sanskrit should be compulsorily imparted” (RSS Resolves; cited by Burg, 1996, 375–376). Madhavrao Sadshivrao Golwalkar, leader of the RSS from 1940 to 1973, views “the queen of languages, the language of the gods” as the unifying source not only for North Indian languages but also for the Dravidian languages (Golwarkar; cited by Burg, 1996, 377). According to N.S. Rajaram, a fiery hindutva propagandist, Sanskrit is exploited in yet another way. He declares the language of Harappa to be “late Vedic” Sanskrit – some two thousand years before the language itself existed. By “decoding” the Harappa seals, he tries to turn up “missing links” between Harappan and vedic cultures in order to support his hindutva revisions of history (Witzel & Farmer, 2000, 6).
In ethnological and sociological discourse, “Sanskritization” denotes a process by which a lower caste attempts to raise its status and to rise to a higher position in the caste hierarchy. Here, the term – often being called into question – is used to refer to Sanskrit culture rather than to the language itself. Sanskritization may take place through the adoption of vegetarianism and the worship of “Sanskritic deities,” or by engaging the service of Brahmans for ritual purposes. Similarly, rulers sought to legitimize their authority by linking it to Brahmanical institutions such as ritual and textual traditions and by patronizing Brahman groups. The spread of the use of Sanskrit as a language of high-cultural expression is an important though not indispensable feature of this process. Sanskrit texts and expressions or, more generally, “Sanskritic culture” are often perceived to be most ancient, most pure, and hierarchically the most elevated, though the “other” traditions have, through history and still today, left their imprint upon Sanskritic Hinduism. Within Hindu traditions, rendering something in Sanskrit, whether orally or in written form, is a way of linking “new” expressions to a sacred past.
While the vedic texts do not offer a proper name for their language of composition, in later eras the Indian word saṃskṛta was established as a characterization or denomination of the language we know by the (anglicized) term “Sanskrit.” The earliest usage of this term is found in the Rāmāyaṇa (3.16.14; 5.28.17–19). From about the beginning of the Common Era onwards, saṃskṛta was established as the usual way to refer to Sanskrit. Saṃskṛta, the verbal adjective of the verbal root kṛ- “to do,” along with the prefix sam “together,” generally means “put together,” “adorned,” “purified,” “prepared.” Essentially it refers to heightening the quality of something. Food, for instance, is called saṃskṛta when it is improved by cooking (Kāś. 4.2.16; 4.4.3). In a similar way, the language Sanskrit is believed to be “superior” due to its “built-up” or “regularly formed” character. It is said to be endued with the masculine noun saṃskāra, “composition,” “correct formation” (a masculine noun that, like saṃskṛta, is derivation of sam + kṛ-). With regard to the characterization of language, saṃskāra has two major aspects. Firstly, it may be used for a particularly distinct articulation, when it is intended to insist on sound being “formed correctly” (see MaBh. 1.5.7; MBh. 14.43.22; VākPad. 1.144). Secondly, saṃskāra is defined by Yāska and Kātyāyana (Nir. 2.1 and 4.1; VājPrāt. 1.1) as the regular word formation of Sanskrit, taught by Pāṇini. Saṃskṛta speech, that is, speech “which has saṃskāra” (saṃskāravatī), constitutes an adornment for the speaker (KumSaṃbh. 1.28; ŚatTr. 1.19). It is also saṃskṛta in the derived sense “made fit for,” “prepared for [a purpose],” this purpose characteristically being ritual.
While Sanskrit is claimed to be endowed with saṃskāra, other languages are, because of their lack of saṃskāra, regarded as corruptions of it (NāṭŚā. 17.2). These other languages, the Middle Indic vernaculars, are designated as prākṛta (Prakrit), “original,” “natural,” “ordinary” and even as apabhraṃśa (Apabhramsha), “corruptions” of the elevated Sanskritic speech (KSS. 1.6.147–148; KāvĀd. 1.32–33; KāvAl. 1.16; MaBh. 1.2.24; apaśabda [Apashabda; MaBh. 1.2.3–9; NāṭŚā. 14.5]). Sanskrit and Prakrit, however, seem to be equally appreciated by the Pāṇinīyaśikṣā
The Texts of Vedic and Hindu Religions
All of the early brahmanical texts, which, in a broader sense, center on the ritual cult (see below), are composed in Vedic or Sanskrit. This literature basically consists of the Veda and Vedāṅga ("Limbs of the Veda"). The Veda is threefold, comprising Ṛgveda, Sāmaveda, and Yajurveda, or fourfold, including Atharvaveda. It consists of Saṃhitās, Brāhmaṇas, Āraṇyakas, and Upaniṣads. The Vedāṅga includes works on phonetics, grammar, etymology, and metrics as well as on astronomy and ritual. The vedic literature is organized in different schools (śākhās, lit. branches), which belong to either one of the four Vedas. To cite just some of many examples: the śākhās of Śākala and Aitareya belong to the Ṛgveda, Jaiminīya and Kauthuma to the Sāmaveda, Maitrāyanī and Taittirīya to the Black Yajurveda, and Vājasaneyī and Pāraskara to the White Yajurveda. Paippalāda and Śaunaka are śākhās of the Atharvaveda.Sanskrit remains the major language of religious texts and textual discourse in classical Hindu traditions, so much so that it is sometimes spoken of as “Sanskritic Hinduism”: the Rāmāyaṇa and Mahābhārata, along with the Bhagavadgītā and Harivaṃśa, Purāṇas, Pāñcarātra Saṃhitās, Vaikhānasa Saṃhitās and Śaiva Āgamas, the Hindu Tantras, Stotras, and Māhātmyas, as well as the literature of the philosophical systems and treatises on dharma ( dharmaśāstra ), politics (arthaśāstra), architecture (śilpaśāstra), medicine ( āyurveda ), drama, and many other topics, all of which are composed in Sanskrit. The rich commentatorial literature (Bhāṣya) of India as well as manuals for ritual (Paddhati, Prayoga) continue to be composed in classical Sanskrit or a variety of regional Sanskrit. Apart from Rāmānuja, Madhva, and Nimbārka, most of the so-called sectarian Hindu traditions as well as the bhakti movements, however, use vernacular languages for their teachings.
The acceptance of the Veda is usually regarded as a sign of Hindu “orthodoxy,” and systems or sects that claim to be based upon additional, extravedic revelations are often regarded as illegitimate and unacceptable by “orthodox” thinkers. The Veda is invoked as the source and focus of the unity and identity of the Hindu religions. It is invoked against the internal, sectarian disintegration of the tradition, as well as against the “external” and “heterodox” challenges of, for example, Buddhism (Halbfass, 1991, 16).
In many postvedic Indian traditions, Sanskrit literature is divided into Śruti and Smṛti. The four Vedas, that is, Saṃhitās, Brāhmaṇas, Āraṇyakas, and Upaniṣads, are called Śruti ("Something Heard [from a Person of Authority]) and therefore highly reliable and often unquestionably true. Depending on the traditions, Śruti texts are regarded as eternal or composed by god. The Śruti texts and among them especially the Saṃhitās are carefully preserved, and much emphasis is placed on the purity and correctness of their recitation. By contrast, the concept of Smṛti ("Something Remembered") denotes less authoritative texts. It generally refers to post-upanisḥadic texts: it may denote the Dharmaśāstras alone (MaSm. 2.10) or the Vedāṅga, but it can also include Rāmāyaṇa, Mahābhārata, and Bhagavadgītā, as well as Purāṇas, Āgamas, and Tantras. The Smṛti texts tend to be more fluid in nature than the Śruti ones. Maybe originally the term Smṛti referred to specific personal memories regarding various customs and conventions, including legal ones, and the events of times past, that is, “tradition.” The term “Smṛti” in the relevant sense appears not to occur before Taittirīyāraṇyaka (1.2.1) and Lāṭyāyanaśrautasūtra (6.1.6.13), and “Śruti” not before Mānavaśrautasūtra (182.4). It is, however, by the Mīmāṃsakas that the two terms have been clearly conceptualized for the first time (see below). In later literature the opposition between Śruti and Smṛti is well known (see e.g. RaghV. 2.2).
Composition and Language of the Veda
Relevant to the distinction of Śruti and Smṛti, the Sanskrit texts of vedic and Hindu religions differ greatly with regard to their actual as well as their ascribed authorship. The latter extends from the postulated authorlessness, by which the text is considered eternal, to its creation by a god, another somehow supranormal being, or a human.The text of the Ṛgveda consists of hymns in praise of the vedic gods and goddesses. Its maṇḍalas (books) 2 to 7 are attributed to different families who collected and transmitted these texts orally. In contrast, our knowledge of the authorship of the individual hymns is hazy. However, some of them do mention the poet (kaví) or seer ( ṛ́ṣi ) who composed them (e.g. Viśvāmitra [ṚV. 3.53.12]). Probably around mid-4th century BCE, the Sarvānukramaṇī, a “general index” to the Ṛgveda, attributed each of the hymns to a particular named poet. The poets and seers, it is believed, have a special relation to the gods (ṚV. 1.139.9). And the poems are, like chariots, considered a means of establishing a relationship between men and gods (e.g. ṚV. 5.60.1). Indra, Agni, and the divine pair Mitra and Varuṇa as well as Soma are said to be the source of inspiration of those poems. According to Ṛgveda 8.59.6, it was Indra and Varuṇa who, in the beginning, endowed the ṛ́ṣis with thought and speech, which those seers then subsequently made stable (sthā́nāny asṛjanta) – that is, they made it into poetry (deváttam bráhma [ṚV. 1.37.4]). In Ṛgveda 4.11.2–3, the seer asks Agni for poetic inspiration, and in Ṛgveda 9.99.6 Soma is called the “Lord of Vision.” Although inspiration is given by the gods, the poets play an active role in “crafting” the hymns in ever-new forms. Some of the Indo-European formulas denoting this process of poetic composition are still present in the vedic texts as, for instance, the indication of poetry as carpentry (ápūrvyā purutámāny . . . vácāṃsy . . . takṣam [ṚV. 6.32.1]). The poets were conscious of the process of liturgical composition (ṚV. 10.71; 1.88.4; 6.9.6). In later texts the ṛ́ṣis are said to have “seen” the Veda or, more particularly, the Veda’s stanzas such as sāman, stoma, and mantra. Sometimes it is related that the dharma appeared before the ṛṣi’s eyes. The way of gaining this wisdom is explicitly connected with the notion of a visual perception (TaiĀ 2.9; AṣṭA. 4.2.7; Nir. 1.20; 2.11; 7.3; MBh. 18.5.33).
In the vedic literature, it was, as far as I can see, not questioned in which language songs and hymns for the gods and goddesses should be composed. Using Vedic without exception, the authors of the rgvedic hymns must have regarded this as the appropriate language. A topic of discussion is, however, which words (i.e. always Vedic words) are to be used in addressing the gods. Aitareyabrāhmaṇa (3.33), for instance, states that the gods love mysterious speech that is not directly accessible (parokṣapriya). The poets, therefore, will be careful to not choose plain words. In this context, it may be added that the Ṛgveda (6.24.6) alludes to riddling contests (brahmodya) of the poets – a real and potentially dangerous contest that consisted of encoded questions and equally encoded answers. The speech used for these occasions was also deliberately not easily accessible.
Eternality and Authorlessness of the Veda
With later Hindu philosophical traditions, it is believed that the Veda has no author, or at least no human author. Some traditions even declare it to be uncreated and therefore eternal. The genuinely exegetic tradition of Mīmāṃsā, one of the six systems of classical Hindu philosophy, has developed dependent on the Veda. The central notions of Mīmāṃsā are the “authorlessness” (apauruṣeyatva) and “eternity” (nityatva) of the Veda as well as its “self-sufficient validity” (svataḥprāmāṇya). According to Mīmāṃsā, the Veda must be authorless because there is no reminiscene of an author, human or divine, or any other proof of its having been composed. Nevertheless it exists and is handed down from generation to generation. It is further argued that, since the Veda has no author who, by definition, would be fallible, the cognition to which it gives rise must be true; all the more so since Mīmāṃsā holds that all knowledge is valid unless its cause is defective. The Mīmāṃsā tradition makes a clear distinction between Śruti and Smṛti. Śruti generally means “(Veda) actually now perceived aurally (in recitation)”; Smṛti, it is said, means “(Veda) that is remembered,” that is, material that had once been heard in recitation but today is only recoverable from present reformulations (ŚāBh. 77; TVā. 76.4–5; 94). The philosophers of Mīmāṃsā unambigously regard the Smṛti as based on the Veda (or Śruti; NyāMañ. 1.372.9–373.6).In Mīmāṃsā, the reliability and meaningfulness of the Veda is further backed up by postulating the eternalness of language. Words and the letters that constitute it are unchangeable and ubiquitous; it is only their particular manifestations, caused by articulation of the vocal organs, that are restricted to certain times and places. The meanings of words, being universal, are fixed as well. Finally, a primal, necessary, and unarbitrary (autpattika) relation between word (śabda) and meaning (artha) is postulated (MīmS. 1.1.5). Words could not be expressive of certain meanings as a result of artificial conventions. All this is, of course, valid for Sanskrit alone. Kumārila, a Mīmāṃsā teacher, specifically says that the knowledge of the inhabitants of Āryavārta is authoritative as regards words and their meanings (TVā. 220.3–4; for Patañjali’s notion of Āryavarta, see below).
Nyāya and Vaiśeṣika, the schools of logic and nature philosophy, accepted the authority of the Veda too. But these philosophical systems developed out of the tradition of debate and were not genuinely affiliated with the Veda. They defended the Veda in a retroactive manner. Nyāya postulated verbal testimony of a trustworthy person (āpta) as an independent means of obtaining knowledge (śabdapramāṇa [NyāS. and NyāSVṛ. 1.1.7–8]). The role of āpta is particularly important with regard to the special knowledge contained in the Veda. At an early stage of the Nyāya tradition, it was believed that the Veda was composed by “trustworthy persons” who were endowed with a direct perception of dharma (NyāS. and NyāSVṛ. 2.1.68). At a later stage of development, it was postulated that the Veda is the reliable word of Īśvara (ĀtTatVi. 436). The validity of the Veda was “verified” by comparison with the original āyurveda, whose truth and effectiveness are said to be supported by “empirical evidence.” As the Veda is composed by the same author(s), it must be infallible too (NyāSū. 2.1.68; Chemparathy, 1983, 40–57).
For Śaṅkara, the famous exponent of Advaita Vedānta, the Veda is authoritative; all the more so concerning the ultimate, liberating truth of ātman / brahman . Only by means of the Veda is human insight possible. It is a “sun which shines upon reality and appearance” (BrSBh. 2.1.1; BĀUBh. 2.1.20). Śaṅkara allows for cooperation of “revelation” (veda, śruti) and “reason” (yukti, tarka). However, “reason,” though being a valid form of inference, remains “human cognition” (puruṣabuddhi; Halbfass, 1991, 135), and as such it cannot claim any authority that would be independent from the Veda. One needs to follow “the path which has been shown by revelation and authoritative teachers.” The problem of the relation between revelation and reason is connected to Śaṅkara’s understanding of Śruti and Smṛti – the latter, that is, extravedic philosophical traditions, being subordinate to Śruti. They are acceptable only insofar as they serve the goal of understanding the Veda (BrSBh. 2.1.3).
Speech and Language in the Vedic and Hindu Traditions
Learning the Veda
By the performance of initiation (upanayana), a young man commences the learning of vedic texts and gradually acquires access to them. Usually during the initiation ritual itself or after a certain interval, the student is taught the stanza called gāyatrī or sāvitrī (ṚV. 3.62.10) by his teacher. The gāyatrī is sometimes said to contain the whole Veda in its condensed form. This is the only part of the Ṛgveda most practitioners nowadays know. In the following years of study, the students ideally learn the whole Veda. In order to become eligible for the study of certain parts of the Veda, they need to keep observances ( vratas). According to Śāṅkhāyaṇagṛhyasūtra (2.11.9–12), the śukriyavrata, śākvaravrata, vrātikavrata, and aupaniṣadavrata need to be followed. These vratas have to be kept, according to the commentary, before the study of the certain texts. The śukriyavrata prepares the student for the learning of the Ṛgveda. Śakvaravrata and vrātikavrata are the observances necessary for the study of the Mahānāmnī and Mahāvrata texts which are both part of the Aitareyāraṇyaka. The aupaniṣadavrata appears to belong to the Upaniṣads. The named vratas need to be followed for 3 days, 12 days, a year, or as long as the teacher may think is appropriate. The “secret” parts of the Veda, that is, the Āraṇyaka texts, which are taught to selected students only, are studied under special conditions. Even the teacher has to follow certain observances for three nights. Then, student and teacher leave the village and go to the forest (araṇyaka), where the student, having covered his head with a turban, listens to his teacher reciting the “secret” parts of the Veda (ŚāṅkhGS. 2.12). The completion of the Veda studies is marked by the samāvartana ritual. Here, the alumnus takes his absolving bath and is then smeared with sandal paste, dressed with new garments, and decorated with a wreath of flowers. He puts on sandals and takes up a staff and a parasol. This splendid appearance sets him apart from the Veda student, who is not allowed to adorn himself.The access to the Veda is restricted to certain parts of the vedic society. This restriction is common during the time of the Gṛhyasūtras. According to these texts, the initiation is performed only for Brāhmaṇas (Brahmans), Kṣatriyas/Rājanyas, and Vaiśyas (see e.g. GauGS. 2.10; HirGS. 1.1.17). In the Mīmāṃsā discussion, the right ( adhikāra ) to learn the Veda and thereby take part in the ethical realm of dharma is also explicitly linked to the varṇa ordering (MīmS. 6.1.33). According to later authors, most prominently Manu, male Brahmans were the only ones who could learn, teach, and use the Vedas. Śūdras were, in Manu’s view, not even allowed to hear these texts being recited (MaSm. 4.99; caste).
Ritual Speech and the Goddess Vāc
The ritual cult and with it the liturgy being the common point of reference of the Vedas, the Sanskrit language is central to the construction of the Brahmanical religious tradition. “Speech” (vāc) is marked as divinely revealed. Already in the late parts of the Ṛgveda (10.90.9), we hear that the verses (ṛc), the songs (sāman), and the ritual formulas (yajus) arose from the primordial sacrifice offered by the gods, that is, they arose from the sacrificed body of the cosmic giant, the ultimate ground of existence. Speech, and more specifically ritual or metrical speech, is even deified. Since rgvedic times the feminine word vāc, which denotes sound and speech alike, serves as the name of the goddess “Speech.” The younger books of the Ṛgveda contain hymns devoted exclusively to Vāc (see ṚV. 1.164; 10.71; 10.125; AVŚ. 4.30). However, comparably little mythology is ascribed to this goddess. Her personification is basically one of grammatical gender and remains so until the post-rgvedic period, when she has blended with Sarasvatī (ŚBr. 3.9.1.7, 9; AitBr. 3.1), who is then called Vāgdevī or Girdevī. Nevertheless, within a limited realm of religious-philosophical speculation, Vāc came to occupy a commanding position as “The One” (ékaṃ sát [ṚV. 1.164.46]; tád ékam [ṚV. 10.129.2–3]; brahman [AitBr. 4.21.1]). It is clear from the vedic corpus that she was believed to have creative power. Important for the present context, she is said to be the creator of the Vedas (ŚBr. 5.5.5.12) and identified with the vedic liturgy (yā vāk sark [ChāU. 1.3.4; 7.2.2]). As such she carries the sacrifice to the gods (ŚBr. 1.4.4.2).Like curses and oaths, ritual speech is performative. Being “speech acts,” ritual formulas are effective in the sense that the desired result is believed to happen by the act of uttering the according formula. During the oldest period of religious practice, gods and goddesses were begged for aid by the recitation of hymns. In later epochs, however, the power associated with correctly uttered ritual speech was thought to be absolute. It was believed that one can “force” the gods to grant the desired results of the ritual by the performance of correct actions and flawless liturgy (AitĀ. 1.3.2; 5.1.5). The importance of proper meaning and impeccable accentuation is illustrated by Taittirīyasaṃhitā (2.4.12.1; MaBh. 1.2.12): Tvaṣṭṛ, seeking to avenge his son’s death at Indra’s hands, performs a sacrifice whose purpose is the production of an enemy to destroy the god. Tvaṣṭṛ’s vendetta fails because – out of ignorance or inadvertence – he places the vedic pitch accent (udātta) on the first part of the compound rather than the last, thus forming an attributive compοund (bahuvrīhi) rather than the desired determinative compound (tatpuruṣa). In this way he creates a being “whose killer is Indra” (indraśatrú) rather than, as he had intended, a “killer of Indra (indraśátru).”
Vāc appears as an independent force that is difficult to control. Even the vedic gods (devas) struggle to obtain and maintain her. In a famous myth related by Śatapathabrāhmaṇa (3.2.1.18–24), the devas and the asuras, both sons of Prajāpati, received the inheritance of their father: the gods obtained mind, sacrifice, and “that world”; the asuras obtained Vāc and “this world.” The gods then made Vāc their own by employing the sacrifice. The asuras – being deprived of Vāc and thus using improper speech in ritual – were defeated. In the following (ŚBr. 3.5.1.21–22; MaBh. 1.2.12), Vāc is slighted by the gods and asuras and goes away from them angrily. Only after having received a boon does she finally go over to the gods.
In the second half of the 1st millennium BCE, the vedic cult was systematized by the composition of the Śrautasūtras and Gṛhyasūtras, compendiums for the solemn and nonsolemn rituals, respectively. These compendiums prescribe many citations from the Veda, which are named as mantra, ṛc, yajus, sāman, or brahman, and so on, as liturgy in their ceremonies. As at least the Gṛhyasūtras still provide a direct or an indirect basis for contemporary ritual traditions, these Sanskrit citations are employed in Brahmanical and also popular Hindu traditions up to the present day, especially for the performance of the life-cycle rituals ( saṃskāras), although today’s priests and ritual participants might have little or no command of Sanskrit. The amount of information that mantras provide is, at least today, not their most significant aspect. They are, above all, “real” components of a past world, and as such they have to be employed and enacted, not “understood.”
Mantras are of utmost importance not only in orthodox Hindu traditions but also in Tantra. Here, they are considered soteriologically central acoustic formulas. Mantras are described as consisting of Sanskrit phonemes whose specific energy is that of Lord Śiva (ŚiSVim. 2.3). The highest form of the word lies beyond language. Tantric traditions extol bījamantras (lit. seed mantras) such as oṃ. These are syllables of extraordinary power that are clearly separated from ordinary language.
Sociolinguistics of Sanskrit
Since early times language has been an important component of the vedic or Indian (religious) identity. The “others” (dásyus) are referred to as “without (proper) language” (anā́so . . . mṛdhrávācaḥ, lit. “without mouth . . . of abusive speech” [ṚV. 5.29.10]). As noted above, the asuras, being deprived of Vāc, lost the ability to speak proper Sanskrit. Instead of the correct he’rayo ‘rayaḥ (hail friends!), they used an Eastern dialect he ‘lavaḥ! he ‘lavaḥ! (Thieme, 1955, 437–438).The attainment of speech – that is, the ability to learn and speak proper Vedic or Sanskrit – needs to be ritually secured shortly after birth. Usually the Gṛhyasūtras teach a ceremony for the “production of intelligence” (medhājanana) for the new-born child. The central element of this ritual is the reciting of a mantra into the child’s ear that contains the word medhā (see e.g. ĀśvGS. 1.15.2). In the ceremony taught by Śāṅkhāyanagṛhyasūtra (1.24.9–10), however, intelligence is equated with speech. Here, the word vāc is thrice muttered into the child’s ear, and the following mantra praises the great and sweet-sounding goddess Vāc and asks her to rejoice in the child. Similarly, in the medhājanana ritual of Kauśikasūtra (10.16–19), an Atharvaveda song (AVŚ. 4.30) is employed in which the goddess Vāc describes herself.
In ancient times not only was there a distinction drawn between proper and improper speech, that is, Sanskrit and non-Sanskrit, but social status (human/divine; male/female) was marked by differences within the Sanskrit language. Employing Indo-European notions, a distinction is made between words used by gods and words used by men. What the latter know as carman (skin) is called śarman (shelter) among the former (ŚBr. 1.1.4.4; 3.2.1.8). From Śatapathabrāhmaṇa (10.4.6.1) it appears that gods, gandharvas, asuras, and humans use different synonyms for the same thing meant. Thus, while later authors explicitly declare Sanskrit to be the “language of the gods” or the “heavenly language” (daivī vāc [KāvĀ. 1.32; VākPad. 1.182]), it appears that in vedic times a different way of expression within Sanskrit was ascribed to the gods (see ṚV. 1.164.45).
Women, in the Ṛgveda, also speak Sanskrit (in contrast with later times), but they do so differently from men. The vedic poets used certain linguistic features to characterize the speaker as female. Two of these features are the secondary -ka suffix and the perfect optative. The diminutive -ka forms in the Ṛgveda are employed in unusual, colloquial contexts, some of which involve women’s speech, often sexually charged or vulgar (see e.g. ṚV. 1.126.7; 10.86.7; VājSa. 23.22). Female speakers establish their identity by using perfect optatives (ṚV. 1.179; 10.10; 10.28.1), while in the whole of the Ṛgveda, subjunctives and imperatives are far more common (see Jamison, 2008).
Finally, the famous stanza of the rgvedic riddle–hymn informs us that "speech (vāc), [is] fourfold, [it] is measured out in quarters: those [only] wise Brahmans understand. Three portions, hidden [within a cave], do not move. The fourth portion of vāc, men speak" (ṚV. 1.164.45). Semantically complex, this stanza may allude to a social stratification within Sanskrit.
Maintaining its high or “sacral” status in later times, Sanskrit was and often remains restricted to the community of the Brahmans or, more generally, the twice born (men). Sāyaṇa in his commentary of the Ṛgveda offers no fewer than four interpretations of the above-cited stanza. However, in virtually all interpretations, the three hidden portions of vāc are assessed by an elite group, that is, the male Brahmanical world, or some aspect of it, and all equate the fourth portion of vāc to the language known as laukika (worldly) or vyavahārika (ordinary, mundane). Manu, however, treats the varṇa ordering separately from the languages spoken by different persons and groups. He says that
"all the castes (jāti) in the world that are outside those born from the mouth, arms, thighs, and feet are declared dasyus [by tradition, regardless of] whether their language is foreign (mleccha) or Ārya." (MaSm. 10.45)
Sanskrit and the Veda in Science and Philosophy
Understanding and Preserving the Language of the Veda
The Indian traditions, from early on, had an astonishing level of reflection on their language. Grammatical interest arose in the first place in connection with the necessity of preserving, intact, the sacred texts of the Veda. While the Brāhmaṇas, Padapāṭhas, and Prātiśākhyas (see below) as well as Yāska in his Nirukta confined themselves more or less to the corpus of the Veda as such, the later grammarians directed their analysis at the spoken language. In order to preserve the Vedas, an extensive literature concerning correct wording, pronunciation, and understanding of the Vedas was produced. The Brāhmaṇas, manuals of liturgical justification, are concerned with the meaning of utterances used in ritual. They offer pseudo-etymologies of words to explain their significance. The etymologies of ancient India have nothing to do with the origin or the history of the words concerned. They establish links to the “essence” of the thing named. The fire god (Agni), Śatapathabrāhmaṇa (2.2.4.2) says, is thus so-called because he was created first (agre).Methodically, etymology (nirukta, explained, past participle of nir + vac-, “to express clearly”) can be considered as a continuation of the explanations of words in the Brāhmaṇas. Etymology is – most prominently – represented by Yāska’s Nirukta. Yāska opposes Kautsa for whom mantras are strictly meaningless. For him, there was no doubt that the vedic mantras have meaning. He holds that the words in the vedic verses are in principle the same as those employed in daily usage (Nir. 1.15–16).
Among the means by which the correct transmission of the vedic texts was achieved was the Padapāṭha (word-by-word recitation). While in the “continuous recitation” (Saṃhitāpāṭha) all the saṃdhis (euphonic junctions between words) are applied, in the Padapāṭhas each word of the text was repeated separately, hence unaffected by the rules of saṃdhi. Most compounds, derivates, and inflected forms are analyzed. Additionally, special patterns of repetition (vikṛti) of the separated words were constructed to minimize the risk of forgetting words. For instance, in the “step-by-step recitation” (Kramapātḥa) the saṃdhis are applied for pairs of words. In the “twisted recitation” (Jaṭāpāṭha) each pair of words is repeated thrice, the middle repetition being in inverted order.
- -
- Saṃhitāpātḥa: 1 2 3 4
- -
- Padapātḥa: 1 / 2 / 3 / 4
- -
- Kramapātḥa: 1 2 / 2 3 / 3 4
- -
- Jaṭāpāṭha: 1 2 2 1 1 2/ 2 3 3 2 2 3/ 3 4 4 3 3 4/ etc.
Grammar and Language Philosophy
In India, formal grammar (vyākaraṇa) was cultivated to a degree of sophistication hardly found elsewhere in the world, let alone so early in history. Quite early on grammarians also concerned themselves with the philosophical-religious aspects of language associated with release from the cycle of births and union with an ultimate being. In these fields, Sanskrit was apparently the only object language that the Indian authors found worthy of study. The four most prominent names in the fields of grammar and language philosophy are Pāṇini, Kātyāyana, Patañjali, and Bhartṛhari. Pāṇini (c. 500 BCE) is the author of the Aṣṭādhyāyī, in which he codified the grammar of Sanskrit in about four thousand brief rules. The first preserved addition to Pāṇini’s Aṣṭādhyāyī is Kātyāyana’s Vārttika (c. 250 BCE). The third of the “grammar ṛṣis” is Patañjali, who composed his Mahābhāṣya in the mid-2nd century BCE. He treats Pāṇini’s Aṣṭādhyāyī as well as Kātyāyana’s Vārttika. Along with grammar, he concerns himself with matters of the philosophy of language. Bhartṛhari (probably 5th cent. CE) is acknowledged as the greatest systematic philosopher of language medieval India produced. He is the author of the Vākyapadīya (or Trikāṇḍī) and most probably of the Mahābhāṣyaṭīkā (or Mahābhāṣyadīpikā). The Indian tradition also ascribes to Bhartṛhari the Vākyapadīyavṛtti, a commentary on the Vākyapadīya.Pāṇini describes a living language. He distinguishes in his grammar between chandas (sacred literature) and bhāṣā (spoken language), mentions usages particular to northerners (udīcās) and easterners (prācās), and notes a difference between usage north and south of the Vipāśā (Beas) River (AṣṭA. 4.2.74). Later grammarians such as Patañjali acknowledge that the language Pāṇini describes is the speech of persons referred to as śiṣṭa, an elite group of model speakers characterized as much by their moral qualities as by their language. In his Mahābhāṣya (3.173.19–174.15), Patañjali says that the śiṣṭas are restricted to the region of Āryāvarta, the region between the country Ādarśa, the mountain Kālakavana, the Himālayas, and the western Vindhya range (MaBh. 1.475.2–3). He believes that they alone speak proper Sanskrit, even without learning it with the help of grammar. Hence, the grammarians look up to them in formulating their rules. For Bhartṛhari the śiṣṭas are no longer a contemporary community of standard speakers of Sanskrit, but the ancient sages of a “golden age” of Sanskrit grammar. By his time, the founding fathers of Sanskrit grammar had achieved the status of ṛṣis (e.g. VākPad. 2.481). From about the 13th century, Patañjali is even believed to be an incarnation of the divine grammarian, the great snake Śeṣa.
In formulating his grammar, Pāṇini uses Sanskrit as a metalanguage to which he refers with the term upadeśa (instruction; e.g. AṣṭA. 1.3.2). Including regular Sanskrit words, its difference from proper Sanskrit is principally only visible in the artificial elements such as peculiarities in phonology, nominal flection, and the use of cases (Wezler, 1975, 29). Patañjali as well as Śabarasvāmin (5th cent. CE), the author of Mīmāṃsābhāṣya, recognized the difference between Pāṇini’s metalanguage and the object-language Sanskrit that he described. Unlike Sanskrit they did not regard the metalanguage as eternal (nitya) because in the process of word formation taught by Pāṇini, certain elements of the metalanguage have to be added and then removed as, for example, the end marker (anubandha) iṭ (see AṣṭA. 1.1.60). Kātyāyana, Patañjali, and Bhartṛhari use proper Sanskrit (as it is taught by Pāṇini) in analyzing this language. Their discussion of making statements about Sanskrit while at the same time using it for description centers round Pāṇini’s statement:
"A word [in a grammatical rule] which is not a technical term denotes its own form." (svaṃ rūpaṃ śabdasyāśabdasaṃjñā; AṣṭA. 1.1.68)
Thus, when the word agni (fire) is used in a grammatical statement, the signatum is not the fire but the same as the signans (Brough, 1951, 29).
In their analysis of Sanskrit, the grammarians made a distinction between “words” or other linguistic entities and the corresponding physical sounds. Against the common-sense view that a word consists of the actual sounds of the instance, they established the sphoṭa doctrine, which is of central importance for the philosophy of language. For Patañjali and Bhartṛhari, sphoṭa is the constant part of the speech, while dhvani or nāda denotes the sound that varies according to pronunciation and level of sound (see MaBh. 1.181.19–25; VākPad. 1.76–79). For Patañjali, sphoṭa can be a single phoneme or a fixed pattern of letters. For him, sphoṭa does not necessarily convey meaning. For Bhartṛhari, in contrast, sphoṭa is the “real word,” that is, the meaning-bearing unit that is without parts and sequence but revealed by sequential physical sounds. Similarly, Bhartṛhari teaches that a sentence and even its meaning are indivisible (VākPad. 2.7–16). He explains that in actual communication the meaning of a sentence is not understood as a sequence of the constituent word meanings but in a flash of understanding (pratibhā) as a single unit. Only by the deliberate subsequent division by grammarians through theoretical analysis does a sentence seem to have component words.
On the basis of the sphoṭa theory, a metaphysical superstructure was erected in which the transcendental word (śabdabrahman) was seen as the first principle of the universe. Bhartṛhari’s Vākyapadīya (1.1) begins with an assertion that the ultimate reality (brahman) is the imperishable principle of language and that the evolution of the entire world originates from this language reality in the form of its meaning. Language is the vehicle of all human activity, and the knowledge of language cannot come about without grammar. This grammar is a gateway to salvation, a treatment for the defilements of language, and the leading branch of knowledge. Through the science of grammar, which includes a language-based spirituality, one can attain the state of the highest language reality (VākPad. 1.11–22). Bhartṛhari’s effort to elevate grammar to the status of not only a philosophical system but also a spiritual path toward salvation is an elaboration on ideas scattered throughout Patañjali’s work. Patañjali discusses at length reasons to study grammar. In this context, he explains that one should learn grammar because the great god named in Ṛgveda (4.58.3) who is said to enter mortal beings is speech (śabda; MaBh. 1.3.20–22).
Sanskrit in Jainism and Buddhism
The prestige of Sanskrit was refused by those who questioned the authority of the Vedas. Mahāvīra and Buddha (5th to 4th cents. BCE), the founders of Jainism and Buddhism, respectively, were born in the region to the east of the north-central region of Āryāvarta, the religious center of the Brahmanical tradition. In contrast to the Brahmans, Mahāvīra and Buddha developed religious teachings that offered the possibility of salvation to everyone. In their zeal to reach the masses, they taught not in Sanskrit but in Middle-Indo-Aryan dialects. Jainism and Buddhism came to regard the Prakrits, which they used, as languages of high status and great religious value.The Theravādins assume, as a matter of course, that the language of the Buddha was Magadhi and that it was the same as Pali, which is used for their canonical texts. They further regard Pali/Magadhi to be the original language of all beings (mūlabhāsā; RūSi. 2.1). In the Pali canon, it is prescribed that the Buddha’s words should be acquired in one’s own speech (sakāya niruttiyā; CuVa. 5.33.1 [= VinP. 2.139]). In order to preserve the Buddha’s words, two converted Brahman brothers had suggested putting it in chandas (lit. verse). The commentary sets chandas in relation to the Veda and, possibly, Sanskrit (vedaṃ viya sakkatabhāsāya vācanāmaggaṃ āropem; SāPā. 6.1214). However, the suggestion is objected to: “One who might put [the Buddha’s word in chandas] commits an offence an account of ‘wrong action’ (dukkaṭa).”
The canonical texts of the Jain traditions are composed in Ardhamagadhi, and later texts in Jain-Maharashtri, Jain-Shauraseni, and Apabhramsha. Like Sanskrit by the Brahman tradition, Ardhamagadhi is declared “the language of the gods” by the Jaina tradition (BhaVaī. 108). The 11th-century Śvetāmbara Jaina scholar Namisādhu expresses the same opinion in his commentary on Rudraṭa’s Kāvyālaṃkāra (devāṇaṃ bāṇī; KāvAl. 2.12). He further grants priority to prākṛta against saṃskṛta when he explains the former as a compound whose first constituent is prāk equivalent to pūrvam (before). Rudraṭa then claims that prākṛta refers to what was brought about “before” saṃskṛta. The latter is said to result from prākṛta through purification by grammar.
However, the early preference for vernaculars in Buddhism and Jainism was eventually abandoned, and both the traditions gradually switched to the use of Sanskrit. This shift may have been motivated by the rising (worldly) prestige of Sanskrit in the post-Maurya periods (see below). By the 1st century BCE, some Buddhists began to partially Sanskritize their texts. The resulting “mixed Sanskrit” has been called “Buddhist hybrid Sanskrit” or “quasi-Prakrit-cum-Sanskrit” and is the language of numerous important texts mostly by the Mahāsāṅghikas, such as the Mahāvastu, a long biography of the Buddha. By the 5th century CE, Buddhist traditions, that is not only the Mahāsaṅghikas but also the Sarvāstivādins, Mūlasarvāstivādins, Vaibhāṣikas, Sautrāntikas, and the various Mahāyāna traditions, adopted Sanskrit even for their canonical texts. Moreover, Buddhist poets and scholars have composed important works in Sanskrit. Aśvaghoṣa (2nd cent. CE) is the author of the poetical work Buddhacarita. And, to name just one further example, the famous Buddhist scholar Amarasiṃha has written the Amarakoṣa (or Nāmaliṅgānuśāsana [Instruction Concerning Nouns and Genders]), a synonymic dictionary of Sanskrit.
During the earlier centuries, the Jainas were less open to the Sanskritization movement. The logicians of the Śvetāmbara (and to a lesser extent Digambara) wrote commentaries and independent works in Sanskrit, but not in large numbers until the 8th and 9th centuries. An example of Jaina literature composed in Sanskrit is the Tattvārthadhigamasūtra by Umāsvati, a highly regarded philosophical treatise. From about the 13th century, the Śvetāmbara Jains of western India have produced extensive literature of narrative, chronicle, and hagiography written in Sanskrit.
Epics and Purāṇas
Compared to vedic literature, the epics, that is, Mahābhārata and Rāmāyaṇa, and the Purāṇas are different in character as well as intent. Their common center is the art of storytelling, and they are accessible to a broader audience than vedic literature. These texts are rather fluid in nature as, in their oral stages, their transmission was altogether different from the exact and carefully protected recitation of the Vedas. The meaning of the stories, not sound, was crucial, and additions and modifications were made freely. The many versions of the texts, none of them being “the original,” only became fixed and stable relative to the vast majority of other realizations of the texts when they were finally written down. The north central version of the Sanskrit Mahābhārata includes in the text itself the story of being written down. Here, the god Gaṇeśa is said to have written the Mahābhārata on palm leaves (MBh., app. I/1).The epics’ and Purāṇas’ originally oral character is revealed by their text structure. Both epics are set within a frame narrative that centers on the respective heroes. But in the course of time, more religious and didactic material was included in them, especially in the Mahābhārata. The Purāṇas are invariably framed in a dialogical structure, their contents being manifold from the beginning. On the one hand, they relate by definition “stories of the old days,” such as mythological narratives going back to the Vedas and the Mahābhārata. On the other hand, they contain a miscellany of customs, rituals, medical knowledge, and many other topics.
The original authors and oral transmitters of the various epic cycles would have been the sūtas or bards. The sūtas were attached to the courts of chieftains and recited in short songs and at major festivals the glorious deeds of their lords. These bards, most probably, accompanied their masters into battle or to the hunt as charioteers and told stories for their entertainment. It is generally agreed that the epics, having originated in a Kṣatriya context, were then taken over by Brahman redactors. For the Purāṇas, this view has also occasionally been claimed but more often rejected. Some of these texts, however, allude to their previous recitations in royal courts and villages.
The Indian tradition generally ascribes the authorship of the Mahābhārata (1.1.15) as well as many Purāṇas to the sage Kṛṣṇa Dvaipāyana, who is also called Vyāsa (lit. “Divider”) as he separated the Veda into its four parts (ṛc, yajus, sāman, and atharva). It is believed that he had direct access to the knowledge of everything past, present, and future. In some cases, the texts composed by Vyāsa are subsequently authorized by a god, often Brahmā. In other cases, the specific god of a Purāṇa is held as its original author (e.g. AgP. 1.7–8) who would then pass it on to Vyāsa or another sage. Several stages of transmission follow in the Mahābhārata and some Purāṇas until the respective narratives are told to humans. The place of transmission is often the Naimiṣa forest, and at some stage or the other the sūta Lomaharṣaṇa is involved in reciting the Mahābhārata or a puranic text. (For the Rāmāyaṇa’s view of its own creation, see below.)
The epics and Purāṇas are usually classified as Smṛti. Mahābhārata and Purāṇas are, however, also popularly called the “Fifth Veda” (MBh. 1.57.74; ChāU. 7.1.2) as these texts understand themselves as an expansion of the Veda (see e.g. VāyP. 1.200). They often claim to comprise all valuable knowledge and surpass all previous expressions of important knowledge. The classification of the Mahābhārata as Smṛti becomes especially controversial with regard to the Bhagavadgītā, as this text, being a later addition to the epic, is believed to be spoken by the god Kṛṣṇa. For many Hindus it is therefore deemed more authoritative than some of the Śruti texts themselves.
In early scholarship it had been speculated that because of the epics’ and Purāṇas’ generally “popular” nature, they must have been composed in one of the Prakrits and only later transposed into Sanskrit. This view has long been abandoned, but it is true that the Sanskrit of the epics (and Purāṇas) is different from Vedic as well as classical Sanskrit. This variety has sometimes been called “Kṣatriya Sanskrit,” but today it is generally referred to as epic Sanskrit. It is deemed a more “natural” Sanskrit than the classical variety following Pāṇini. In epic and puranic traditions themselves, the Sanskrit language is generally associated with the twice born. The social status implied by speaking Sanskrit is explicitly referred to in the Rāmāyaṇa. Whenever the sagacious monkey–hero Hanumān needs to establish his trustworthiness with one of the central epic figures, he must indicate that he can speak like a Brahman. Sanskrit is also used by demons when they disguise themselves as Brahmans (Rām. 3.16.14; 5.28.17–19). The demon-king Rāvaṇa, for instance, had approached Princess Sītā pretending he was a Brahman. Later in the plot, Hanumān therefore decides to address Sītā in “the language of men (mānuṣyaṃ vākyam, “a human utterance”), with meaning [but without saṃskāra],” that is most probably a Middle Indo-European dialect. He says, “if I speak saṃskṛtā vāc, as though I were a twice-born, Sītā will be frightened, thinking I am Rāvaṇa” (Rām. 5.28.18–19; 4.3.27–33).
The Bhāgavatapurāṇa, an important Sanskrit Purāṇa of about the 10th century, plays a major role in the propagation of the bhakti movement to North India. This movement originated in the south of the subcontinent where the Āḻvārs, a group of mystics (7th–9th cents.), composed their highly poetical, ecstatic songs in Tamil, the standard language of South India. The language choice of the Āḻvārs prevented the propagation of their songs to the Sanskritic traditions of North India, where Sanskrit was the lingua franca. These songs were adapted by the Bhāgavatapurāṇa into Sanskrit and thus made them available for traditions of North India. The Sanskrit epics, in contrast, have been adapted into various Dravidian as well as New-Indo-European languages. Kampaṉ (12th cent.) is the author of the Irāmāvatāram, a Tamil Rāmāyaṇa. In the 16th century, Tulsīdās composed another famous adaption of this epic, the Rāmcaritmānas, in Hindi. In the 1980s, both the epics were reworked for the first time for television broadcast in which a highly Sanskritized form of Hindi was used.
Royal Inscriptions and Classical Poetry
Around the beginning of the 1st millennium CE, Sanskrit was transformed into comparatively widely available language. It was employed for royal inscriptions on monuments, land-grant plates, seals, and coins. It also became the most esteemed medium for courtly literature and classical poetry. The conservative values that Sanskrit came to represent made the language increasingly regularized and artificial. The widespread use of Sanskrit during the Gupta period is at least partially responsible for the term “Classical Age” that is given to this era.Royal Inscriptions
The earliest inscriptions and thereby the earliest records of literacy found in India are composed in various Prakrits, that is, Middle Indo-Aryan languages, not in Sanskrit. Only after three or four centuries of literacy, Sanskrit, the linguistic parent of those languages, is to be found in inscriptions. Both Buddhist and Brahman rulers avoided Sanskrit for worldly records in earlier centuries. The oldest attested inscriptions, those of the Maurya emperor Aśoka, who sponsored Buddhism and reigned approximately 272–231 BCE, are composed in the vernacular language of Magadha, a form of Prakrit. The Satavahanas (230 BCE–220 CE), who clearly supported Brahmans and stressed traditional vedic or Hindu religion in their inscriptions, also used Prakrit in their records. In the Kathāsaritsāgara (6.108–164) and Bṛhatkathāmañjarī (6.35–52), fun is made of the Satavahana king who did not know Sanskrit. Embarrassed by his lack of command over this language, he ordered the Brahman Śarvavarmin to compose a “quick Sanskrit grammar” for him, the Kātantra. Despite this late caricature, the Satavahanas’ choice of Prakrit against Sanskrit was probably not due to a lack of knowledge of Sanskrit. Rather, it seems to be the pure and carefully regulated orality of Sanskrit that made this language inappropriate for publicly written worldly records.The rise in the public availability and use of Sanskrit must have begun slowly after the fall of the Mauryas at the hands of the Brahman Puṣyamitra Śuṅga (a contemporary of Patañjali; MaBh. 2.123.3–4), who assassinated the last Maurya king. An inscription at the end of the 1st century BCE from the region of Ayodhya, mentioning King Dhana(deva?) as “sixth [in generation] from the general Puṣyamitra,” is the sole Sanskrit document of the Shungas and the first preserved Sanskrit inscription. The revolutionary use of Sanskrit in the inscription might have aimed at religiopolitical prestige. The Shunga takeover strived to stabilize vedic religion. The dynasty after the Shungas is believed to be the Kanvas, to which the first inscription in real Sanskrit is attributed. It was found in Ghosundi, Rajasthan. These early Sanskrit epigraphs consist of one- or two-line records commemorating a vedic or quasi-vedic rite.
The earliest longer Sanskrit inscription is that of the Shaka ruler Rudradāman. It dates from approximately 150 CE and is located at Junagadh in the Girnar hills of Saurashtra, Gujarat. It commemorates the repair of a tank near Girnar, which was carried out by King Rudradāman. This inscription inaugurates the writing of benedictory introductions (praśasti) to royal inscriptions in a highly poetical and hyperbolic style. Rudradāman was a “foreign” ruler with a Sanskrit name (his ancestors still had Iranian names) belonging to the Iranian Kshatrapas. His choice of Sanskrit was a clear break with the tradition of writing in Prakrit. It appears to be a method to legitimize his reign and his acquisition of “Indian” identity. For similar reasons the Scythian and Kushana rulers (1st–3rd cents. CE) adopted a form of hybrid Sanskrit for their records. As foreigners, they were inclined to patronize the elite language in an effort to legitimize their rule and emphasize their own Indianization. This so-called Epigraphical Hybrid Sanskrit is neither fully Sanskrit nor fully Prakrit but combines the characteristics of both. By far the majority of those inscriptions have been found in the region of Mathura, whence they radiate toward the northeast and southwest and thus parallel the foundation and spread of the kingdoms of the Scythian and Kushana rulers in the heartland of India.
In the Gupta epoch (4th cent. CE), Sanskrit reached significantly beyond the domain of liturgy and its sacral auxiliaries. The adoption of Sanskrit for political records is mirrored in the rise of worldly poetry in Sanskrit, which is also intrinsically connected to the Gupta monarchy. Besides being established as the lingua franca, classical Sanskrit also became the epigraphic language par excellence during the reign of the early Gupta emperors. The turning point appears in the inscriptions of Samudragupta (mid- to late 4th cent.), especially the Allahabad Pillar Inscription, which is often held up as a model of the high classical literary style of the mixed prose and verse class (called campū). Sanskrit was also extensively used in coinage. Here, it was again under the Guptas that pure Sanskrit was adopted in coinage. The inscriptions on Samudragupta’s famous coins describe him as “the king of kings of irresistible prowess, the protector of the earth who wins heaven.” However, the western Kshatrapas, rulers of parts of the Shaka Empire, had used a form of Sanskrit for their coinage earlier.
Among other dynasties, mostly of the post-Gupta eras, mixed epigraphical records became common. While the pragmatic message of such documents is conveyed in the local language, that is, Prakrit and Tamil as well as Persian and New Indo-Aryan languages, the Sanskrit part is usually of religious and royal character. It is mainly used for an introductory invocation of the gods (maṅgalācaraṇa), a fixing of mythological and genealogical succession, a catalogue of kingly traits of the dynasty, and a eulogy of the ruling lord. These are all elements of what became the standard praśasti style.
Classical Poetry (Kāvya)
Kāvya denotes a highly educated literary style employed in the court epics of India from the early centuries of the Common Era. The appearance of these worldly texts, which are predominantly composed in Sanskrit, coincides more or less in time with the adoption of Sanskrit in royal inscriptions. Unlike the authorship of Vedas, Śāstras, and Purāṇas, the Indian tradition regards, the authors of Kāvya as ordinary humans. In opposition to gods, seers, and sages of the vedic and epic eras, the poet (kavi) is a normal human being. He is, however, of refined taste and great versatility and has a broad education acquired by wide reading as well as, generally, a command of several languages, particularly Sanskrit. The great masters of the Kāvya form were Aśvaghoṣa, Kālidāsa, Bāna, Daṇḍin, Magha, Bhāvabhūti, and Bharavi. Vālmīki is often considered the first poet and his Rāmāyaṇa the first poem (BuCar. 1.43). The god Brahmā confirms Vālmīki’s Rāmāyaṇa to be a new invention (Rām. 1.2.29–36). What makes the work “literature” is the poet’s recording of a personal response to real and not mythic human experience. Vālmīki utters the primal metrical line (śloka) when he witnesses an act of violence in the forest: a pair of krauñca birds (type of curlew) in amorous play being killed by a hunter. Vālmīki invented the śloka because of his grief about the act: “I was overcome with pity,” he says, “and this [metrical line] issued forth from me – it must be poetry and nothing else” (Rām. 1.2.17). The śloka was only subsequently authorized by the god Brahmā (Rām. 1.2.30).Kāvya is, as a rule, associated with the glory of the royal courts. The central role that the figure of the king plays in Kālidāsa’s dramas and in his epic Raghuvaṃśa strongly suggests that he enjoyed royal patronage. Kālidāsa’s patron was probably Candragupta II, who ruled most of North India from about 375 to 415 CE. Likewise, Kaniṣka, a sovereign of foreign origin, is considered to have patronized Aśvaghoṣa, a Sanskrit poet of Buddhist faith, living in the 1st or 2nd century CE.
Kāvya requires considerable literary education to be composed and appreciated. It is a form of art characterized by a sometimes ostentatious display of erudition. The kavis choose their language carefully in order to achieve a particular effect. Extensive use of speech figures is made, especially of metaphor and simile. In the often hyperbolic compositions, varied and complicated meters are employed. The metrical poetry is usually divided into Laghukāvya (poetry of the minor form) and Mahākāvya (poetry of the major form). Poems of the major form are almost invariably divided into chapters and therefore also called Sargabandhas (“composed in cantos”). Poetry that is a mixture of prose and verse is called campū, while works written in the heavier prose style are subdivided into two categories, “(fictive) story” (Kathā) and “(true) story” (Ākhyāyikā; see Lienhard, 1984).
Kāvya also includes plays and drama, which are traditionally enacted in the theatre. The earliest extant textual source for drama is the Nāṭyaśāstra (2nd cent. CE), a Sanskrit dramaturgical manual attributed to Bharata. Today, the South Indian Sanskrit theatre kūṭiyāṭṭam is considered to be the only surviving traditional theatre form The text of the drama is recited in different pitches (svaras) similar to the traditional Veda recitation. By a system of hand gestures, the Sanskrit texts are simultaneously displayed according to Dravidian morphology.
Literacy is referred to in Kāvya almost from the genre’s beginning and often regarded as a precondition to Kāvya, as some traits of Kāvya can be understood properly only in written form, as, for instance, the figurative poetry (Citrakāvya). In some Citrakāvyas, double meaning is used throughout. These are called Yamakakāvya. The double epic Rāghavapāṇḍavīya tells both the Mahābhārata and Rāmāyaṇa stories simultaneously.
Major topics of Kāvya are love, nature, panegyric, moralizing, and storytelling. It mainly is situated in the context of kingship and court life. Often, literary devices are applied to traditional subjects and themes derived from early popular epics. Kālidāsa’s drama Raghuvaṃśa (The Raghu Lineage) offers the mythical prehistory of the Ikṣvāku and, concomitantly, of the Gupta family. His drama Kumārasaṃbhava (The Birth of the [God] Kumāra) is a camouflaged allusion to the birth of Kumāragupta (415 CE), the son of Candragupta II. Like this work, other Kāvyas are similarly built on mythological narrative or take up stories from the epics. Still others are devotional throughout. Jayadeva (Bengal, 12th cent.), for instance, describes in his Gītagovinda (Songs of the Cowherd) the love of Kṛṣṇa and Rādhā. Likewise the Karṇāndana (Delight of the Ears), a poem from the Rādhāvallabhī tradition, is centered on Kṛṣṇa’s lover Rādhā. However, the Kāvya genre is not predominantly religious, neither in intent nor in character.
The literary theories that developed with the invention of Kāvya define certain languages as permissible media for literary production and thereby exclude others. The number of primary literary languages is mostly three or four. Sanskrit and Prakrit are invariably mentioned; the third place is usually occupied by Apabhramsha or Deshabhasha ("Languages of Place"). The last language mentioned is Paishaci or Bhutabhasha ("Speech of Demons [ piśācas, bhūtas]). Bharata, in his Nāṭyaśāstra (17), discusses the speech forms that are to be represented on the stage, and who can use them. He considers three languages: Sanskrit, Prakrit, and Deshabhasha (see KSS. 1.6.147–148). Bhāmaha (7th cent.) teaches that Kāvya is “threefold in being composed in Sanskrit, Prakrit, or Apabhramsha” (KāvAl. 1.16; KāvĀd. 1.32). Rājaśekhara, in turn, claims to be an “expert in all languages” with a reference to the four literary languages, including Bhutabhasha (Balarām. 1.11; VāgAl. 2.1–3).
The choice from among the primary literary languages was, for all theorists, largely determined by the genre in which one wrote. Thus, for instance, the Mahākāvya and the Ākhyāyikā were to be written in Sanskrit, the Skandhaka (that is, Mahākāvya with different metrical organization) and Gāthā (song) only in Prakrit (KāvAl. 1.28, 37–38; ŚṛṅPr. 11; DhvaĀL.3.7). Rājaśekhara, in his Kāvyamīmāṃsā, says that
"a given topic will be best treated in Sanskrit, another in Prakrit, or Apabhraṃśa, or the language of the spirits, bhūtabhāṣā; others still in two, or three, or all four languages. The writer whose mind is sharp enough to distinguish [among these possibilities] will win such fame as spreads across the universe."(KāvMī. 9; trans. Pollock, 2006, 94)
The four languages are far from being equally valued. Bharata regards the idealized speakers of the Sanskrit language as pure (saṃskṛta) and contrasts them to barbaric speakers ( mlecchas; NāṭŚā. 17.28). In Sanskrit drama, as for instance in Kālidāsa’s plays, Sanskrit is the language spoken by twice-born men: the king, his advisers, and others of high status in religious or political spheres. It is occasionally spoken by a woman of learning (see MālA.). However, women were, as a rule, linguistically characterized by Prakrit, that is, Shauraseni. Similarly, the Brahman buffoon (Vidūṣaka) as well as various minor characters also spoke different Prakrits. The hierarchy between (the) different languages is most explicitly illustrated by Rājaśekhara, in his tale of the origins of literary culture. He imagines a poetry giant (Kāvyapuruṣa) – whose mouth consists of Sanskrit, arms of Prakrit, groin of Apabhramsha, feet of Paishaci, and chest of mixed language (KāvMī. 3) – as the primal being of literature. Daṇḍin praises Sanskrit in his 17th-century work on literary theory, the Kāvyādarśa (Mirror on Literature; 1.33), by characterizing it as “the language of the gods (daivī vāc), taught [to men] by the great sages.”
The “purity” and “built-up” character of Sanskrit serves Kāvya better than any other languages. Because, as Bhāmaha in his Kāvyālaṃkāra (1.16), states, Kāvya denotes “a unity of word and meaning,” a text in which form and content require and receive equal attention. Especially Sanskrit offers writers a rich opportunity of varying linguistic elements and of putting them in any order they choose so as to obtain the maximum effect. The great majority of authors concentrate on the sound qualities of their sentences and verses and do not hesitate to make full use of the possibilities their language offers them. The language of Kāvya went even further in that poets were constantly enlarging the existing number and scope of synonyms. In certain fields old Indian poets have built up a real synthetic vocabulary that is peculiar to the field.
Status of Sanskrit in Colonial and Modern India
Albeit in a limited sense, Sanskrit is a living language up to the present day. It still has important religious, cultural, educational, and political influence in South Asia and among the Hindu diaspora all over the world.Sanskrit has retained its religiously superior status in the modern period. It remains the ritual language par excellence of classical Hinduism as well as many popular Hindu traditions. Its literature continues to be regarded with respect, even though the understanding of religious sentiment nowadays almost always takes place through one’s vernacular. Vivekananda, the founder of Ramakrishna Mission, suggests that
"the ideas should be taught in the language of the people; [but] at the same time, Sanskrit education must go along with it."
He views Sanskrit not as a medium or a spiritual message, but rather as a means to raise the condition of the lower castes. Swami Dayananda Saraswati, founder of the Arya Samaj, directed his message in his first years of preaching mainly to the top layer of the Indian society, that is, the Brahman community to whom he spoke in Sanskrit. Later he switched to Hindi, with Sanskrit playing only a nominal role in his teachings. The Theosophical Society in its aim to return to traditional Indian values also ascribes Sanskrit a high status. H.S. Olcott, one of the founders of the Theosophical Society, therefore established the Adyar Library, which collects manuscripts up to the present day.
Up to the 18th century, Sanskrit unquestionably remained a very important language of arts and sciences alike. Jagannātha Paṇḍita (17th cent.) is one of the major Sanskrit poets, and Nāgeśabhaṭṭa Kāle (d. 1755) a respected grammarian. He composed the Paribhāṣenduśekhara. Moreover, many Sanskrit commentaries on a variety of subjects are handed down to us. Though a controversially discussed question (see Pollock, 2001; Hanneder, 2002), it is a matter of fact that up to the present day, Sanskrit remains an important language of Kāvya poetry and, to a lesser extent, traditional scientific literature. Many poems, short novels, novellas, and Mahākāvyas are even nowadays written in Sanskrit (for examples, see Hanneder, 2002).
Sanskrit is still actively encouraged in Indian education. The traditional way of learning Sanskrit is within the guruśiṣyasambandha (master–disciple relationship). Thus, the major location in which speakers learn the language was not the home but the household of a guru or other institutions of learning (pāṭhāśālā and saṃskṛtavidyāpīṭha). The Kendriya Sanskrit Vidyapeeth in Tirupati has, since its founding, conducted all of its instruction in Sanskrit, whether on traditional or modern subjects. The traditional system was considerably weakened by the introduction of a dual system of Sanskrit education. The methods of teaching at government schools, colleges, and universities follow the English system. However, up to the 1960s, students at Varanasi Sanskrit University also had to speak and understand Sanskrit well enough to attend classes taught in this language (Hock, 1992, 18).
It is an often-cited fact that the census of India lists a few thousand people who claim Sanskrit as their mother tongue. In the year 1971 it listed 2,212 persons, 6,106 in 1981, 49,736 in 1991, and 14,135 in 2001. There is nothing in the data to indicate the extent to which people claiming Sanskrit as their mother tongue or as an auxiliary language can actually speak it. The facts suggest that the large numbers of persons declaring Sanskrit as their mother tongue in the censuses either are exaggerating or have a different interpretation of the term “mother tongue,” not as “native language” but as language toward which one feels just the same as toward his or her own mother. The oral use of the language is generally limited to the recitation of short verses (ślokas). However, even though Sanskrit is nowadays not spoken by a significant number of persons, Sanskrit words are included in śuddha (pure) Hindi with slight changes such as the omission of final -a. Furthermore, Sanskrit is used as a source of new words for new concepts in Indian languages, so that a speaker of any Indian language knows some Sanskrit words, albeit in adapted forms. Sanskrit was, however, among the other 14 (now 22) spoken languages listed that were introduced into the Eighth Schedule of the Constitution of India (1949) as a recognized language of the new State of India. In the mid-20th century, there have even been attempts to substitute English as the official language by a simplified Sanskrit. The Sanskrit Commission, which met in 1957 parallel to the language commission, did not see any scope for Sanskrit as an official language but only as a language of classical Hinduism. The commission advised the government to integrate Sanskrit into the school curricula, advocate traditional Sanskrit schools, preserve manuscripts, and so on. The inclusion of Sanskrit among the other recognized languages still ensures funding for Sanskrit colleges and universities, and for a national organization to stimulate the study of the language. The year 1999–2000 was declared the “Year of Sanskrit.”
Sanskrit also gets support from Hindu nationalist and fundamentalist movements. The Rashtriya Swayamsevak Sangh (RSS) passed in March 1958 a resolution concerning the national language policy. Though Hindi is wished for as the official language, Sanskrit is viewed as the root language, the “Rashtra Bhasha par excellence.” It is suggested that “instruction of Sanskrit should be compulsorily imparted” (RSS Resolves; cited by Burg, 1996, 375–376). Madhavrao Sadshivrao Golwalkar, leader of the RSS from 1940 to 1973, views “the queen of languages, the language of the gods” as the unifying source not only for North Indian languages but also for the Dravidian languages (Golwarkar; cited by Burg, 1996, 377). According to N.S. Rajaram, a fiery hindutva propagandist, Sanskrit is exploited in yet another way. He declares the language of Harappa to be “late Vedic” Sanskrit – some two thousand years before the language itself existed. By “decoding” the Harappa seals, he tries to turn up “missing links” between Harappan and vedic cultures in order to support his hindutva revisions of history (Witzel & Farmer, 2000, 6).
Sanskritization
In the study of Indian languages, certain parts of linguistic changes are sometimes named “Sanskritization.” On the one hand, non-Sanskrit words are “Sanskritized” in the way that they are adopted into Sanskrit in accordance with known phonetic correlations. On the other hand, modern Indo-European languages, for example, Hindi and Marathi, make extensive use of Sanskrit vocabulary and are thereby “Sanskritized.” This heavy form of Hindi is called śuddha (pure) and is predominantly employed in formal and religious contexts.In ethnological and sociological discourse, “Sanskritization” denotes a process by which a lower caste attempts to raise its status and to rise to a higher position in the caste hierarchy. Here, the term – often being called into question – is used to refer to Sanskrit culture rather than to the language itself. Sanskritization may take place through the adoption of vegetarianism and the worship of “Sanskritic deities,” or by engaging the service of Brahmans for ritual purposes. Similarly, rulers sought to legitimize their authority by linking it to Brahmanical institutions such as ritual and textual traditions and by patronizing Brahman groups. The spread of the use of Sanskrit as a language of high-cultural expression is an important though not indispensable feature of this process. Sanskrit texts and expressions or, more generally, “Sanskritic culture” are often perceived to be most ancient, most pure, and hierarchically the most elevated, though the “other” traditions have, through history and still today, left their imprint upon Sanskritic Hinduism. Within Hindu traditions, rendering something in Sanskrit, whether orally or in written form, is a way of linking “new” expressions to a sacred past.
Comments
Post a Comment