A Brief Sketch of Sanskrit Development
After several thousand years, the relevancy of a language is often in question. But Sanskrit, the liturgical language of Hinduism and a central language of Jainism and Buddhism, continues to pervade the lives of millions around the world—in prayers, in poetry, and even in daily vocabulary. Sanskrit occupies a place in South Asia akin to Latin in Europe—spoken fluently by few, but acknowledged and respected by most. It is the foundation on which all modern northern Indian languages are created. In terms of vocabulary, Sanskrit-derived words or words taken directly from the language dominate nearly all Indian languages’ higher registers (including those of some Dravidian tongues). Tracing the development of Sanskrit is a complicated and immersive task because Sanskrit itself is so dense and diverse.
Sanskrit is an Indo-European language, a member of the Indo-Iranian group and the Indo-Aryan branch. In the Indian census of 1991, almost fifty thousand people listed Sanskrit as their mother tongue. This number has fluctuated intensely in the last few decades, rising as much as 714% from 1981 to 1991 and decreasing nearly 75% from 1991 to 2001 (India Census Bureau). This instability probably reflects the fact that Sanskrit, a dead language, is almost never used outside of formal or religious situations, and so the definition of a “fluent speaker” may continuously oscillate between bookish knowledge and conversing power. The three states with the highest population of “Sanskrit speakers,” according to the census, are those regions with notable traditions of Vedic study (Kushala 2). The Times of India reported that residents of the village Mattur in central Karnataka spoke Sanskrit as their vernacular (Kushala 1), but those numbers are not well reflected in the 2001 census, raising questions about the validity of that reputation.
Today, Sanskrit has a varied application. Government agencies in South Asia use Sanskrit phrases for their mottoes; Buddhist, Jain, and Hindu religious philosophers use Sanskrit for terminology; some Sanskrit words have entered English—avatar, pundit, yoga, and nirvana, for example; and students in India have the option of learning Sanskrit as a third language. Some computer programmers have even proposed the use of Sanskrit to structure computer languages because of its rigid grammar and unchanging syntax (Lakshmi 1). In Hindi and Kannada, two languages at opposites of the Indian linguistic spectrum, saṃskṛti—literally “Sanskritic”—is used to describe things that are traditional, and has a connotation of refinement and class.
To explore change in Sanskrit, one must start at the beginning. One of the most salient differences between Indo-Aryan languages and other branches of the Indo-European family is the presence of a series of aspirated and unaspirated retroflex consonants: ṭ, ṭh, ḍ, ḍh, ṇ, ṣ. In the Ṛgveda, the oldest known Sanskrit work, these consonants are present in a large number of words—an issue with which scholars have grappled for centuries. By the way, the Rgveda contains references to entertainments we play now, for example dice games which have transformed into online casino sites in our modern life.
Most hold that the influx of retroflex consonants is a result of interactions between Indo-Aryan speakers and speakers of a proto-Dravidian language, a member of a language family whose early alphabet may have included these sounds. According to this school of thought, such interaction must have occurred at an early stage in the development of Indo-Aryan languages—and must have been pervasive, owing to the frequency of retroflection in the Ṛgveda. By the time retroflection appeared in the Vedic canon, such sounds were established firmly in the language (Deshpande 129; Thirugnanasambandhan 3).
Scholars point to the fact that modern Dravidian languages use retroflex consonants in abundance to support their suspicion that Indo-Aryan retroflection is the effect of a Dravidian substratum acting on the existing language—possibly through bilingualism and the interpretation of certain Indo-European consonant clusters as retroflex instead of alveolar (Deshpande 130-3).
This view, while mostly accepted, must also involve examining the complex social environment in which Indo-Aryan and Dravidian peoples interacted. In the Ṛgveda, non-Aryans are derided as anāsa, akarman, and mṛdhavākya— snubnosed, non-religious, and speakers of a hostile language. The Aryans called their foes dāsas, slaves. Gradually, however, it seems that—ethnically, religiously, and linguistically—Aryans and non-Aryans mingled (Thirugnanasambandhan 2).
Deshpande agrees that Indo-Aryan retroflection is an absorption of Dravidian consonants, but he argues that retroflection did not enter Indo-Aryan languages until later, and that retroflex consonants in the Ṛgveda are a consequence of Vedic students (perhaps predominantly in southern India, he argues, where Dravidian languages are spoken) adapting trends in their vernacular to alveolar consonantal clusters in the Ṛgveda, which may have been compiled before retroflection was added to Indo-Aryan alphabets. Exacerbating this trend was the tradition of completely oral transmission: the Vedas were taught without written records until recently.
Proto-Dravidian exerted other influences on Sanskrit and its descendants. In Sanskrit, there are three classifications of number: singular, plural, and dual. This third classification, the dual, is found in a very small number of languages, a group which doesn’t include the modern Indo-Aryan languages. Dravidian languages do not have a dual, either. Thirugnanasambandhan believes that the absence of the dual in all modern Indian languages is evidence of the structural effect a Dravidian substratum had on a changing Sanskrit (11).
There are other elements in the modern Indo-Aryan languages that point to non-Indo-European influence. Postpositions, the only adposition in Indian languages, are far more common in these languages than in other Indo-European languages. Several northern Indian languages have the option of using the concept of “we” with and without including the party being addressed, a feature also found in Dravidian languages. Modern Indo-Aryan languages’ verbal tenses are often formed by using participles and subsequent attachments, another southern trend (Caldwell 57).
Non-Aryan elements, then, exerted a huge influence on Indo-Aryan languages, which distinctively set this branch off from other Indo-European languages.
Pāṇini, a Sanskrit grammarian from the fourth century B.C.E. who set out the rules that defined classical Sanskrit, made an early distinction between the two faces of the language: bhāṣa, “colloquial language,” and chandas, “the language of the Vedas”—but by bhāṣa, of course, he really meant the colloquial language of the educated upper class. So other forms of Indo-Aryan and non-Aryan languages were thus deemed substandard—and speakers of those forms were placed in the lower parts of the social hierarchy (Deshpande 2). From the point of view of upper-class citizens, there was a definite prejudice against the development of vernaculars. But another influential grammarian, Patañjali, wrote in his commentary on Pāṇini that word forms, verb tenses, and noun declensions that were not used by common speakers should not be used—in short, grammar should be a complete democracy, and the people decide what works and what doesn’t (Bhandarkar 24).
Looking into a Sanskrit dictionary, one finds a distinction between the definition of Sanskrit, saṃskṛta, “refined,” and Prakrit, prākṛta, “natural.” There was a view of Sanskrit as something that had been polished, smoothed, and developed for ages, making it infallible and unquestionable and beautiful. The Prakrits, on the other hand, were utilitarian languages that developed organically from Sanskrit through different phonological and morphological trends and a migration of Indo-Aryan peoples across ancient India (Pischel 21). This delineation persisted at the first signs of the Prakrits’ development, with Sanskrit authors lamenting the vulgarization of their cultivated language (Thirugnanasambandhan 15). Some Prakrits, however, were recognized with time as valuable cultural and literary—particularly playwriting—languages (Pischel 24).
Grammarians classified Prakrit words into three categories: tatsama, words identical with Sanskrit; tadbhava, words derived from Sanskrit; and deśya, words not derivable from Sanskrit. Tatsama words were generally borrowed to elevate the literary registers of Prakrit languages, whereas tadbhava words went through the natural phonetic changes that all Prakrit language experienced between Sanskrit and themselves (Thirugnanasambandhan 15; Kahrs 227). This is particularly significant because tadbhava words may exist alongside the tatsama words from which they were derived: for example, Prakrit paum, “flower,” coexists with padma, “lotus,” from which it was borne. Most of the linguistic research about the history of Indo-Aryan languages centers along the development of tadbhava words.
Bhandarkar attributes much of the transformation from Sanskrit into modern Indic tongues to phonetic decay, a tendency of speakers to economize the effort involved in producing a sound or pronouncing a word. He points to one of John Locke’s aphorisms, “Labor for labor’s sake is against nature,” to describe this trend. According to his view, consonantal clusters or diphthongs are lost in translation from one generation to the next until an atypical or non-standard pronunciation becomes the norm. Education, he argues, is an institution which arrests this decay—but in ancient India, the lack of widespread education for most of the population meant that sounds lost their clarity and specificity (Bhandarkar 30).
He gives an example of the formation of the Sindhi and Gujarati locative case—a situation of “yesterday’s syntax” becoming “tomorrow’s morphology.” The Sanskrit word madhya, meaning “middle,” is attached to the word grāma, “village,” to form a compound grāma–madhya, meaning “in the middle of the village” or “in the village.” Over time, madhya became majjha, an “easier” word to pronounce, which eventually became maha and finally mā. Grāma became gām through assimilation and apocope, and so grāma–madhya completed the change into gām-mā, the Sindhi and Gujarati locative case of “village” (Bhandarkar 32). One may describe this trend in terms of assimilation: consonantal clusters made up of two different consonants tend to transform into a cluster defined by one of those two consonants. For example, Sanskrit sarva, “all,” changes into Prakrit savva. Lenition, too, fits into Bhandarkar’s idea of Sanskrit trends of economy—ś and ṣ have been reduced to s in all cases in Prakrit languages. To speakers who may not have encountered Sanskrit words (i.e. non-Aryans), certain consonant clusters may have been difficult to produce or seemed different to inexperienced ears, thus producing new, nonstandard forms.
These Prakrit forms were defined early on as apabhramśa, “corrupt,” a term which soon encompassed a new language trend in itself. Apabhramśa appeared after Prakrit languages and represents the transitional state between Prakrits and modern languages. Apabhramśa was the idiom of many Jain writers, themselves propagators of a religion, Jainism, that some considered nonstandard (Bhayani 3). Bhayani describes Apabhramśa as “the standard Prakrit colloquialized,” meaning Apabhramśa was one step further along in the development of Sanskrit (5). Ratcliffe points out that several literary languages coexisted during the Middle Indo-Aryan period (a time spanning the sixth to the twelfth centuries): classical Sanskrit, the Prakrits, and Apabhramśa—but, essentially, the three groups are different registers of the same language; a case of “triglossia” (16). The social stratification of medieval India meant that Brahmins used Vedic Sanskrit, scholars and writers used classical Sanskrit, stage writers used Prakrit languages, and Jain religious writers used Apabhramśa. At the bottom of the social stratum, the immediate predecessors to modern northern Indian languages began to emerge (Ratcliffe 17). Apabhramśa continued to be used in literature until the seventeenth century, coexisting with increasingly colloquial forms of itself, until it was completely supplanted by these regional languages, the modern northern Indian languages (Bhayani 9).
Indo-European languages traveled an incredible distance from the Proto-Indo-European Urheimat eastward, and within India, Sanskrit speakers moved across the subcontinent and encountered new peoples. Over generations, this “refined” tongue changed its phonological and morphological rules to cope with social and sociolinguistic factors. The results of this shift are particularly important because of the enormous number of Indo-Aryan speakers today, and the enduring influence of Sanskrit alongside its daughter languages.
Bhandarkar, R. G. Wilson Philological Lectures (On Sanskrit and the Derived Languages). 1st ed. Vol. IV. Pune, India: Bhandarkar Oriental Research Institute, 1877.
Bhayani, H. C. Apabhramsa Language and Literature. Vol. 7. Delhi, India: B.L. Institute of Indology, 1989.
Caldwell, Robert. A Comparative grammar of the Dravidian or South Indian family of languages. Madras, India: University of Madras, 1976.
Deshpande, Madhav M. Sanskrit and Prakrit : Sociolinguistic Issues. New York: Motilal Banarsidass (Pvt. Ltd), 1993.
Kahrs, Eivind G. “What is a tadbhava word?” Indo-Iranian Journal 35 (2006): 225-49.
Kushala, T. N. “This village speaks God’s language.” The Times of India 13 Aug. 2005.
Lakshmi, Rama. “Summer Camps Revive India’s Ancient Sanskrit.” The Washington Post 15 June 2008.
Languages and Mother Tongues. Republic of India. Census Bureau. Office of the Registrar General & Census Commissioner. New Delhi, 2001.
Pischel, R., and Subhadra Jha. Comparative Grammar of the Prakrit Languages. 2nd ed. Delhi, India: Motilal Banarsidass, 1965.
Ratcliffe, Robert R., and Vit Bubenik. A Historical Syntax of Late Middle Indo-Aryan (Apabhramsa). Boston: John Benjamins Company, 1998.
Thirugnanasambandhan, P. Sanskrit-Tamil Contact. Thiruvananthapuram, India: International School of Dravidian Linguistics, 1992.