Skip to content

Lakṣaṇa, “grammatical rule”


– Sergiu Al-George (1922-1981) –

The Sanskrit noun lakṣaṇa has many semantic values, all of them derived from its main meaning, i. e. “characteristic sign”, “distinguishing mark”, “indication”. Thus, for instance, in ritual literature lakṣaṇa means “catch-word” or “lines drawn on the sacrificial area”; in astrology, “auspicious mark”; in medicine, “symptom of a disease”; in iconography, “indicative attribute of a divinity”; in logic, “definition”; in grammatical terminology, among other things, it means “grammatical rule”. The first occurrence of this technical meaning goes back to the Rkpratiśakhya XIII, 31[1], and the word is frequently used with this meaning in post-Pāṇinian literature. In contradistinction to the above-mentioned technical uses of the word lakṣaṇa, the grammatical one, not being self-evident, requires some explanations.

In attempting to solve the equation: lakṣaṇa = grammatical rule, modern exegesis quotes, without further explanations, Patañjali’s famous gloss in his introduction to the Mahābhāṣya (after varttika 14), where he asserts that śabdo laksyah sūtram lakṣaṇam, “The word is the object of the rule (lit. «the signified object»), the sūtra [of Pāṇini] is the rule (lit. «the sign»)”. This statement, however, does not provide the reasons for the semantic change whereby the grammatical rule is assimilated to the significative structure, which holds a central position in Indian logic. Two different logical operations – definition (lakṣaṇa) and inference (anumāna) – are considered by Hindu logicians to belong to the relation of signification. Both of them are viewed as a mediated knowledge, a deductive cognition from antecedent to consequent; the sign is the antecedent and its admittance entails that of the consequent – the signified. Thus the Indian definition, where the definiens is called lakṣaṇa (a name which is extended to the whole definitional process), is not a mere “characterisation”, as is usually considered in Western translations, but a semiotic act. The definiens as lakṣaṇa is synonymous with liṅga, which qualifies the logical reason in the anumāna, the inferential reasoning: e. g., the existence of smoke is the “sign” (liṅga) of the existence of fire. “Pervasion” (vyāpti) is the common relation – according to Indian logic – lying at the basis of inference and definition as well[2].

We shall try to explain why in grammatical terminology “grammatical rule” is equated to “sign”, starting from the fact that Indian thought implicitly admits that the deductive relation between sign and signified may integrate distinct logical operations, being more comprehensive than them. According to the Indian point of view, the relation of signification is the most general consequence, that has been conceived along similar lines to those developed by European medieval logic; William of Ockham admitted as a valid logical consequence (consequentia bona), and therefore as an equivalent to conditional propositions, the consequence “from the definition to the definiendum” (a diffinitione ad diffinitum)[3]. What is revealing for our inquiry is the fact that Indian traditional grammar has resorted to this general relation – thus anticipating the logicians – in the analysis of the structural relations of language.

It is J.F. Staal’s credit to have been the first to discuss the affinity and even the filiation between grammatical rule and logical definition in India[4], but his inquiry – otherwise prefatory – does not make any reference to the general context of the relation of signification. The Dutch scholar deals with only one of the three categories of Pāṇinian sūtras, namely with the definitional rules (saṃjñāsūtra) for grammatical technical terms; although he also refers to the category of paribhāṣāsūtra, the metarules concerning the interpretative technique for the other rules (definitional and descriptive rules concerning the formation of speech units), and points out their logical value, he does not provide a direct explanation of the fact that not only the actual definitional rules, but also the other two types of rules are called lakṣaṇa. J.F. Staal’s paper remains, however, highly suggestive in this respect. Referring to J.F. Staal’s paper, B.K. Motilal asserts that

the scheme for a saṃjñā sūtra of Pāṇini roughly corresponds to the notion of «nominal or syntactical definition» of the modern formal logicians. Such definitions are explained as «conventions which provide that certain symbols or expressions shall stand (as substitutes or abbreviations) for particular formulas of the system»[5].

Even if we should retain only the denominative aspect of the saṃjñā rules, we have to admit that the nominal definition itself, just as a definition based on a distinguishing mark, contains an analytical relation; the only difference is that in the denominative definition the analytical relation is conventionally established, whereas in that based on a distinguishing mark it is pre-existent. The denominative definition implies a semiosis, since the technical term “stands for”, being an indirect mark of a reality; every proposition derived from such a definition is an analytical one or a “tautology”. It is worth recalling here that, according to the greatest logician of India, Dignāga, the name may be viewed as a logical reason (hetu, liṅga) through which its object is acknowledged; in summarizing Dignāga’s doctrine, Th. Stcherbatsky states that the respective relation

is here founded on Identity of objective reference, the deduction is analytical and the three aspects of the reason are realized; e. g.:

  1. This object is called a jar;
  2. Wherever such objects are found they are called jars;
  3. This name is never applied to a non-jar[6].

We should add that, according to V.N. Misra’s detailed analysis, beyond their purely nominal shape, the saṃjñā rules point at definite categories of hierarchical relations existing between the structural units of the language[7].

Reasons for qualifying as lakṣaṇa not only the definitional rules, but other types of Pāṇini’s sūtras too, may be provided by the various acceptations of the word lakṣaṇa in Pāṇini’s own text. The great Indian grammarian uses the word lakṣaṇa in dealing with some infra-lexical (the unilateral dependence of a suffix on its function) and syntactical relations[8]. Among the syntactic devices for expressing the sign/signified relations, Pāṇini indicates some conditional subordinations (III, 3, 8-9), the absolute constructions of the locative (II, 3, 37) and the genitive (II, 3, 38), the present active and medium participle (III, 2, 126); by establishing syntactic equivalents, Pāṇini assigns the same value to some units of the simple sentence, as for instance to certain prepositions of the karmapravacanīya group (I, 4, 84; I, 4, 90-91), to some invariable compounds  made up by some of these prepositions (II, 1, 14-16) and to the suffixes of the instrumental case (II, 3, 21). In all these subordinations the content of the protasis or its equivalent is called lakṣaṇa, “sign”, because it expresses an antecedent of the content expressed by the main clause.

One of the above-mentioned constructions, frequently present even in the framework of ‎Aṣṭādhyāyī‘s sūtras, is that of the locative absolute. Because of the abridged style of the treatise, the absolute construction does not appear in its standard form – a participle accompanying a noun –, but under an incomplete form; either the noun is missing, and only the participle is present (e. g., II, 3, 1), or the noun alone is present, and the participle to be supplied is sati (the present participle of the verb AS-, “to be “; e. g., I, 4, 23); at other times, the two terms of the locative absolute construction are fused into one, the latter only being in the locative (e. g., III, 3, 144). The absolute value of the locative in Pāṇini is the same as in the nominal style of technical texts[9], that of a conditional clause, and it expresses the conditions indicating a certain grammatical or metagrammatical operation. Patañjali calls this locative viṣayasaptami, “field locative”, “locative of the range [of validity]”, in contradistinction to parasaptami, “locative of the following one”, which marks a word indicating that the preceding one is to be substituted, as stipulated in metarule I, 1, 66.

The viṣayasaptami of the grammatical operations is the most used, mainly in the operations leading to the generation of expression units, when a transition is made from the abstract plane to the concrete one: the locative marks the units of content or grammatical functions, “terms of the higher strata”[10], for the expression of which concrete units in the expression chain are indicated. These rules have the general form: “When A is to be expressed, then A’ is valid (or indicated)”. The viṣayasaptami of the metagrammatical operations, that of the paribhāṣās, marks the indication of an operation in interpreting and applying a grammatical rule. We should note the frequent occurrence of this syntactical construction in the oldest axiomatic syntax of the paribhāṣās: over a third of the paribhāṣā list – as given by Nāgojībhaṭṭa – resorts to this syntactic form. Both types of rules formulated with the viṣayasaptami, grammatical or metagrammatical, are the best illustration of rule II, 3, 37 (yasya ca bhāvena bhāvalakṣaṇam), where Pāṇini states that

[the seventh suffix (= the locative) is valid] also for expressing that by the action of which another action is signified.

The fact that a grammatical rule may be illustrated by the very grammatical structure of a rule only supports the discovery made by transformational grammar that “linguistic «structure» is always relative not just to the data or corpus but also to the grammatical theory describing the data”[11]; both for Pāṇini and for transformational grammar, grammar was a grammar of rules, language and the language of rules being one.

The logical value of the locative absolute – qualified as sign – becomes more evident when we confront it with its role in the syntax of the logical texts. This grammatical form is very frequently resorted to when definitions are formulated and it expresses the definiens which, according to Indian logic, is a distinguishing mark (lakṣaṇa); this stylistic value of the locative was noticed by the Indian logicians themselves[12]. The locative absolute may also express the logical antecedent in a rule stating the unilateral dependence (vyāpti) between sign and signified, a dependence on which inferential reasoning (anumāna) is based; thus, the general rule: “Whenever there is no fire, there is no smoke” may be formulated either as: yatra vahnir nāsti tatra dhūmo ‘pi nāsti – the protasis being a relative clause – or as: agnyabhāve dhūmo ‘pi nāsti – the protasis being a locative absolute. Accordingly, the locative absolute appears as a syntactical equivalent to the relative clause when it expresses the protasis of a logical rule.

The comparisons thus established between the syntactical structure of the grammatical rule and that of the two logical operations – viewed by Indian logicians as deductive operations based on a relation of signification – as well as the logical value of the word lakṣaṇa in Pāṇini may provide a clue to the equation: lakṣaṇa = grammatical rule. Accordingly, the word lakṣaṇa, when applied to the grammatical rule, must be taken in its logical sense; the most suitable translation is indicium, “indication”, “criterion”, instead of “characterisation”, which is rather ambiguous, because it suggests a mere descriptive operation.

Just as the Stoics, who considered both the validity of the premiss and the revelation of the conclusion to be a derivation from sign to signified[13], Indian traditional grammar qualifies both the internal relations between the elements of a rule and the relation between the rule itself and the linguistic form which it generates as a relation between sign and signified. The grammatical rule expresses relations of an analytical nature, enabling operations of building correct language forms, in the same way as a logical rule can serve as the main premiss which makes possible the operation of deducing a valid conclusion.

The discussion above brings into the foreground an important fact: unlike the Greeks, the Indians qualified as semiosis the application of the condition of validity not only in logical operations, but in linguistic ones as well; thus they opened a large perspective to the understanding of the oneness of the linguistic and logical structures. As for the antecedence of grammatical rules as against logical rules or laws – a thesis upheld by some modern logicians which say, namely , that logical laws are merely instances of grammatical rules – the Indian cultural context is revealing, because this antecedence is here historically attested.

 [1]. L. Renou, “Les connexions entre le rituel et la grammaire en sanskrit”, JA 233 (1941-1942), p. 128.

 [2]. Th. Stcherbatsky, Buddhist logic I, Leningrad 1932, p. 146 and n. 5, points out that in Indian logic the definition is an “abridged” or a “defective” inference; D. H.H. Ingalls, Materials for the study of Navya-nyaya logic, Cambridge (Mass.) 1951, p. 81, further remarks that the three fallacies of the definition in Nyāya (ativyāpti, “overpervasion”, avyāpti, “nonpervasion” and asambhāva, “impossibility”) “are essentially the same as the traditional hetvabhasa‘s or fallacies of the hetu”. Cf. also B.K. Motilal, “The intensional character of lakṣaṇa and samkara in Navya-Nyāya”, IIJ 8 (1964-1965), 2, p. 86:

In each case of a true definition, it will be possible to formulate a sort of miniature syllogistic inference, of the form: «A (is) B, because C», where the definiendum will occupy the subject position (paksa), «distinct from others» will be the sadhya, and the definiens will be the hetu.

 [3]. Cf. A. Dumitriu, Istoria Logicii, Bucharest 1969, p. 405, where other types  of valid consequences are listed: “A proprietate ad proprium”, “Ab uno synonymo ad aliud”, etc., which have not been accepted by the mathematical logic.

 [4]. J.F. Staal, “The theory of definition in Indian logic”, JAOS 81 (1961), 2, pp. 122-124.

 [5]. B.K. Motilal, “The intensional character”, p. 85, n. 1.

 [6]. Th. Stcherbatsky, Buddhist logic, p. 459.

 [7]. V. N. Misra, The descriptive technique of Pāṇini. An introduction, The Hague-Paris 1966, pp. 84-101.

 [8]. This problem has been discussed in more detail in our paper “Sign (lakṣaṇa) and propositional logic in Pāṇini”, (supra, pp. 105-119).

 [9]. H. Jacobi, “Über den nominalen Stil des wissenschaftlichen Sanskrits”, Indogermanische Forschungen 14 (1903), p. 244:

die Verwendung des Lokativus zur Umschreibung von Konditional- und Konzessivsätzen von dem Gebrauche des Lok[ativus] absol[utus] ausgegangen ist.

  [10]. V. N. Misra, The descriptive technique, p. 103.

  [11]. E. Bach, An introduction to transformational grammars, New York-Chicago-San Francisco 1964, p. 29.

 [12]. The Nyāyabodhini commentary on Tarkasamgraha 13, (definition of air) is formulated thus: rūparahitatve sati sparśavattvam vayor lakṣaṇam, “Lack of colour existing, possessing touch is the definition of air”; it further explains: satisaptamya viśistarthakataya rūparahitatvaviśistasparśavattvam vayor lakṣaṇam, “Since the locative absolute has a qualifying meaning, possessing touch, qualified by lack of colour, is the definition of air”.

  [13]. Stoic fragment II, 221, apud Sextus Empiricus, Adversus mathematicos VIII, 245.