Fifty Key Thinkers on Language and Linguistics

Margaret Thomas

• Publisher: Routledge
• Year: 2012
• Print ISBN: 9780415373036
• eBook ISBN: 9781136707490
• Topic: Languages & Linguistics
• Subtopic: Linguistics
• Index: Languages & Linguistics

What was the first language, and where did it come from? Do all languages have properties in common? What is the relationship of language to thought? Fifty Key Thinkers on Language and Linguistics explores how fifty of the most influential figures in the field have asked and have responded to classic questions about language. Each entry includes a discussion of the person's life, work and ideas as well as the historical context and an analysis of his or her lasting contributions. Thinkers include:

• Aristotle
• Samuel Johnson
• Friedrich Max Müller
• Ferdinand de Saussure
• Joseph H. Greenberg
• Noam Chomsky

Fully cross-referenced and with useful guides to further reading, this is an ideal introduction to the thinkers who have had a significant impact on the subject of Language and Linguistics.

Fifty Key Thinkers on Language and Linguistics

Pāṇini (Fourth or Fifth Century BCE)

We know very little about the Indian grammarian Pāṇini. He probably lived around 500 BCE, probably in the northwest of the subcontinent in what is now Pakistan. His name has traditionally been associated with a massive corpus known as the Aṣṭādhyāyī (lit. ‘eight chapters’), the earliest extant grammar of Sanskrit. The stunning complexity and finesse of this work set a lasting high-water mark for Indian language scholarship. It has also had significant impact on the study of language in the west.

The Aṣṭādhyāyī describes, for native speakers, an elite version of Sanskrit spoken in Pāṇini’s lifetime. We don’t know for sure, but can only assume that Pāṇini was motivated to compose it as a means of defining and preserving the structure of the language because of its ritual significance in Indian culture, even though Vedic Sanskrit was not exactly the Sanskrit that Pāṇini described, and even though his observations about dialectal variation suggest that his goal was not narrowly prescriptive. We can also only assume that Pāṇini participated in the general cultural assumptions of his day, which made it a matter of spiritual merit to bring implicit knowledge to consciousness: in this case, implicit knowledge of linguistic structure (Emeneau 1955: 146).

But in another sense – because the Aṣṭādhyāyī is so ambitious, so intricate, and has been studied by so many, over such a long interval – we know a great deal about Pāṇini The centrepiece of the Aṣṭādhyāyī corpus of almost 4000 compact rules, or sūtras, collectively referred to as the sūtra-pāṭha. The Aṣṭādhyāyī also subsumes three other texts, for which Pāṇini may or may not have been responsible: an inventory of sounds organized into classes (the śiva-sūtras); a catalogue of about 2000 verbal roots, subclassified by their morpho-syntactic properties (the dhātu-pāṭha); and a list of nominal stems and other lexical items subject to idiosyncratic rules (the gaṇa-pāṭha).

We also know a lot about Pāṇini in the sense that a giant literature of exegesis and commentary has grown up around the Aṣṭādhyāyī, created over centuries by traditional Indian scholars, and more recently by Indologists and by westerners trained by Indian scholars. Moreover, we know that Pāṇini’s work has profoundly influenced Euro-American language science in several waves, sometimes by infusing genuine novelty, and sometimes by seeming to validate developments that occurred independently in western study of language.

Many who have studied the Aṣṭādhyāyī depict it as a work of genius, or of extraordinary, possibly unique, linguistic insight. But no-one is Pa depicts it as easy to read. One can hardly ‘read’ it at all, in the normal sense. The Aṣṭādhyāyī is a collection of pithy statements about Sanskrit phonetics, phonology, morphology, syntax and (in passing) semantics, presented with extreme conciseness in a kind of algebraic metalanguage to facilitate memorization. (In fact, Pāṇini’s rules seem to have remained remarkably stable over centuries of oral transmission.) The material is organized to reduce redundancy to the starkest minimum. For example, Pāṇini exploited the classes of sounds set out in the śiva-sūtras, by referring to the first ordered member of the sequence plus a single, strategically chosen, additional sound that indicated the last ordered member to be included. Likewise, at other levels of grammatical organization, he defined an elaborate metalanguage. For example, sūtra 3.4.69 establishes ‘l’ as a cover term for tense and mood markers. One subclass, which combines ‘l’ with ‘T.’, labels present indicative (‘lAṬ’), perfect (‘lIṬ’), imperative (‘lOṬ’), etc.; another subclass, with ‘Ṅ’, labels imperfect‛IAṄ ’), aorist (‘lUṄ’), etc. (Katre 1987: 340). This metalanguage allows Pāṇini to manipulate classes and subclasses of forms in an efficient manner. His rules admit processes of affixation, doubling and substitution (of one element by another, or by a zero form). Multiple rules may apply in sequence in the derivation of any one form: Max Müller (cited by Staal 1972: 138–39) famously recounted Pāṇini’s derivation of the aorist of the root jāgṛ ‘to wake’, in the course of which nine rules apply. Along the way, intermediate forms may be produced that modern linguists would describe as highly ‘abstract’, that is, which display features that do not appear in the forms’ familiar, end-state, representations.

Pāṇini also dispensed with complete sentences in stating his rules, using instead bare-bone fragments from which all predictable content has been elided. He ordered the sūtras to maximize what is predictable and therefore can be deleted, not according to topic. This means that the meaning of any one sūtra might depend on the content of multiple preceding and following sūtras.

These characteristics of the Aṣṭādhyāyī render it essentially unreadable, and barely translatable. But recall that Pāṇini created it to be inscribed in memory, not on pages. For example, sūtra 2.1.11 appears in a discussion of compound words. It consists in its entirety of the word vibhāṣā, ‘optionally’ (but see below with respect to this translation). Its purport in this context is to signal that compounds defined in subsequent sūtras (up to sūtra 2.2.35, as traditionally inferred) all freely co-occur with uncompounded forms. By another inference, then, compounds defined by the sūtras immediately preceding 2.1.11 are not optional. Falling within the scope of optionality established by 2.1.11 is sūtra 2.2.6, consisting in its entirety of the word náÑ, ‘not’.With this one word, in this context, Pāṇini communicated that the Sanskrit negative particle may optionally (as defined by 2.1.11) combine with a particular class of syntactically connected nominals (as defined by another earlier sūtra, 2.1.4), to form a particular resulting class of compounds (whose properties are defined in sūtra 2.1.22) (Katre 1987: 105–28; Cardona 1988: 243, 254).

At this very fine level of resolution, expressed in arch-economical manner, Pāṇini inched over the whole tissue of Sanskrit’s complex morpho-phonetics and morpho-syntax. The interpretive challenges posed by the Aṣṭādhyāyī have inspired many to analyse, elucidate, exemplify or counter-exemplify its claims. Indian tradition highly values textual exegesis, so that such works have often been elevated to share the status of the text they address. The earliest well developed commentary on the Aṣṭādhyāyī to which we have access today is by Kātyāyana (third century BCE). The most famous is Patañjali’s Mahābhāṣya (second century BCE), which also comments on Kātyāyana’s work; the Mahābhāṣya has itself become the object of a large exegetical industry. Both Kātyāyana’s and Patañjali’s works discuss the validity, formal properties and inter-relationships of Pāṇini’s sūtras. Both remain essential tools for understanding the Aṣṭādhyāyī

Europeans’ first experience of Pāṇini’s work and its long-standing, self-referential, commentarial tradition was as language students. Although the Aṣṭādhyāyī was not designed for foreign language instruction, British colonial administrators and scholars employed it, and the literature to which it gave rise, in that capacity when they began to learn Sanskrit starting in the late 1700s (Staal 1972: 33–34). As westerners both in India and back in Europe became familiar with Sanskrit, a new understanding opened to them of the historical relationships among the languages that they then began to arrange on branches of a single Indo-European family tree. Their recognition of the unmistakable relatedness of Sanskrit to Latin, Greek and other European languages set in motion comparative-historical linguistics, a central intellectual and cultural achievement of the nineteenth century (see Wilhelm and Friedrich von Schlegel; Franz Bopp). This constitutes one of the most far-reaching effects of Pāṇini’s grammar on linguistics and language study worldwide.

Another effect of Pāṇini’s work occurred as western scholars came to appreciate its structure as a descriptive grammar. Several waves of such appreciation have propagated through western language science. Among admirers of Pāṇini have been the English linguist J. R. Firth, who studied Indian linguistic tradition first-hand in India, and the American structuralist Leonard Bloomfield. Bloomfield characterized Pāṇini’s descriptive grammar of Sanskrit as ‘one of the greatest monuments of human intelligence and … an indispensable model for the description of languages’ (Bloomfield 1929: 268). In his analyses of Algonquian languages, Bloomfield adopted Pāṇinian concepts and apparatus, including the notions of form classes, suppletion, exocentric versus endocentric compounds, and zero forms. Above all, Bloomfield’s avoidance of redundancy and his commitment to rigour in the statement of linguistic principles reveal Pāṇini’s influence (Rogers 1987). That influence flowed forward to the post-Boomfieldian scholars who followed his lead in trying to create a formal science of linguistics.

Other waves of Pāṇini’s influence constitute not infusions of novel terms or ideas, as much as recognition that certain concepts favoured in modern linguistics have long existed outside Euro-American language science, in radically different contexts. Pāṇini’s conception of a grammar as a system of intricate, interacting rules (stated with maximum simplicity) that operate on a body of lexical items is also foundational to early generative grammar. Generative grammar resembles Pāṇini’s descriptive technique in its receptivity to abstractness, and in distinguishing underlying versus surface forms. In addition, modern computer science and research on artificial intelligence has found Pāṇini’s work congenial, including its decomposition of a grammar into a web of rules that are ordered, recursive (can apply iteratively), and context-sensitive (Bhate and Kak 1993).

Finally, Kiparsky (1979) made a striking sociolinguistic discovery about the Aṣṭādhyāyī that has achieved acceptance by at least some Sanskritists. Kiparsky scrutinized the distribution of three conjunctions in Pāṇini’s text that are all conventionally translated as ‘or’ (vā; vibhāṣā; anyatarasyām) to show that, far from being synonyms, each one actually posits a different logical relationship between the conjoined terms: vā means ‘preferably’; vibhāṣā means ‘preferably not’; anyatarasyām means ‘optionally’. Re-reading the sūtra-pāṭha by the light of these distinctions shifts our understanding of many passages. It also led Kiparsky to argue that Pāṇini deserves more credit as a keen observer of linguistic variation.

It is worth noting that the admiration, even awe, with which Pāṇini has generally been received in the West has not been unanimous. Aside from those who have found the Aṣṭādhyāyī simply impenetrable, there have been some, such as the late-nineteenth-century American philologist William Dwight Whitney, who dismissed it on other grounds. Whitney denied that the object of Pāṇini’s description could ever have been a natural human language, claiming it was instead only an artificial, ‘special and peculiar’, version of Sanskrit (Whitney 1884: 282). His evidence was that he could not locate in Sanskrit literature many of the roots recorded in the dhātu-pāṭha and that, in any case, the Aṣṭādhyāyī was an improbably convoluted, pedantic system that could not reflect any real language.

Whitney was mistaken, of course: the language Pāṇini described certainly existed, as a genuine product of human cognition and social life. But Whitney’s scepticism testifies to how incredible (in the literal sense of the word) Pāṇini’s work might seem outside the context of its own century and culture.

Bibliography

A respected modern translation is S. M. Katre’s (1987) Aṣṭādhyāyī of Pāṇini. Katre writes for linguists, not Indologists, rendering the Sanskrit in roman transliteration. An accessible English translation of the first 75 sūtras, by H. T. Colebrooke, appears in Kielhorn (1891). Cardona (1988) provides background information and elucidates much of the content of the sūtra-patha, but does not offer a formally organized translation. Rocher (1975) surveys additional translations and editions of the Aṣṭādhyāyī, including the dhātu-pātha and gaṇa-pāṭha also the commentarial literature.

Further reading

Bhate, S. and Kak, S. (1993) ‘Pāṇini’s grammar and computer science’, Annals of the Bhandarkar Oriental Research Institute, 72, 79–94.

Bloomfield, L. (1929) ‘Review of Konkordanz Pāṇini-Candra by B. Liebich’, Language, 5, 267–276.

Cardona, G. (1976) Pāṇini: A Survey of Research, The Hague: Mouton.

Cardona, G. (1988) Pāṇini: His Work and its Tradition: Vol. 1. Background and Introduction, New Delhi: Motilal Banarsidass.

Emeneau, M. B. (1955) ‘India and linguistics’, Journal of the American Oriental Society, 75, 145–153.

Katre, S. M. (1987) Aṣṭādhyāyī of Pāṇini, Austin, TX: University of Texas Press.

Kielhorn, F. (1891) ‘Die Colebrooke’schen Pāṇini Handschriften der Königlichen Bibliothek zu Göttingen’, Nachrichten von der Königlichen Gesellschaft der Wissenschaften und der Georg-Augusts-Universität zu Göttingen, 3, 101–112.

Kiparsky, P. (1979) Pāṇini as a Variationist, Poona/Cambridge, MA: Poona University Press/MIT Press.

Rocher, R. (1975) ‘India’, in T. A. Sebeok (ed.), Current Trends in Linguistics:Vol. 13. Historiography of Linguistics, The Hague: Mouton.

Rogers, D. E. (1987) ‘The influence of Pāṇini on Leonard Bloomfield’, Historiographia Linguistica, 14, 89–138.

Staal, J. F. (1972) A Reader on the Sanskrit Grammarians, Cambridge, MA: MIT Press.

Whitney, W. D. (1884) ‘The study of Hindu grammar and the study of Sanskrit’, The American Journal of Philology, 5, 279–297.

Plato (c. 428/7–c. 349/7 BCE)

The ancient Greek philosopher Plato was a key figure in intellectual history, whose ideas have contributed to the construction of major institutions in western culture. In this sense his influence pervades Euro-American social and political organization, intellectual life and education. Moreover, Plato recognized language as foundational to philosophy. This made it important to him to probe the nature of language, and has made his work enduringly relevant to both philosophy and linguistics. Plato wrote one of the first extended reflections on language, the dialogue Cratylus, which modern scholars still find provocative. Some of his other writings and ideas also continue to be referenced in western language science up to the present day.

Plato was born into a distinguished and well connected Athenian family. He met the philosopher Socrates (469–399 BCE) through family connections, and later became attached to him as a student. On some accounts (Taylor 1926/2001: 3–4), it was the persecution and eventual execution of Socrates that turned Plato away from a career in politics. But although he chose philosophy over public life, he ma...

Table of contents

• Cover
• Half Title
• Title Page
• Copyright
• Contents
• Alphabetical list of contents
• Chronological list of contents
• Acknowledgements
• Introduction
• Fifty Key Thinkers on Language and Linguistics
• Glossary
• Index