1
AṢṬĀDHYĀYĪ
STRUCTURE: AN INTRODUCTION
2
INTRODUCTION
• Aṣṭādhyāyī is a grammatical device that seeks to provide a complete, maximally concise,
and theoretically consistent analysis of Sanskrit grammatical structure.
• Consists of a finite number of ordered rules that can derive an infinite number of correct
Sanskrit sentences.
• Based on the spoken language (bhāṣā) of Pāṇini’s time, and also gives rules on Vedic usage
and on regional variants.
• Sūtra-s are based on the principle of general and particular (sāmānya and viśeṣa)
• A key technique of generalization used by Pāṇini is to organize the sūtra-s into hierarchies
of generality.
COMMENTARIAL 3
LITERATURE
• Sūtra (Aṣṭādhyāyī): A succinct formulaic statement presenting a rule of grammar
• Vrtti (Kāśikāvr̥tti): Paraphrases sūtra-s complete with their implicit elements
• Vārttika (of Kātyāyana): Refinement (?) of sūtra-s
• Bhāşya (Patañjali): Detailed exposition of sūtra-s with examples, counter-examples
• Prakriyā (Rūpāvatāra, Prakriyākaumudī, Siddhāntakaumudī, etc): Application of the sūtra-s.
• Siddhānta (Vākyapadīya, Vaiyākaranasiddhāntakārikā, Vaiyākaranabhūṣaṇa, etc): Exposition of theoretical issues.
• Trimuni of Vyākaraṇā: Pāṇini, Kātyayana and Patañjali. According to the dictum of “yathottaraṁ munīnām prāmāņyam,”
Patañjali is the supreme authority in matters pertaining to Pāṇini.
• 10 pre-Pāṇinian grammarians mentioned in the Aṣṭādhyāyī are: Āpiśali, Kāśyapa, Gārgya, Gālava, Cākravarmaṇa,
Bhāradvāja, Śākaṭāyana, Śākalya, Senaka and Sphoṭāyana.
4
A N C I L L A RY T E X T S
The grammar consists of four components:
a. Aṣṭādhyāyī: a system of approx 4000 grammatical rules.
b. Śivasūtra-s: the inventory of phonological segments, partitioned by markers (anubandha-
s) to allow abbreviations for classes of segments to be formed (pratyāhāras).
c. Dhātupāṭha: a list of approx 2000 verbal roots, with subclassification and markers
encoding their morphological and syntactic properties.
d. Gaṇapāṭha: an inventory of classes of lexical items idiosyncratically subject to various
rules.
5
ŚIVASŪTRA-S
1 a i u Ṇ
2 r̥ l̥ K
3 e o Ṅ
4 ai au C
5 h y v r Ṭ
6 l Ṇ
7 ñ m ṅ ṇ n M
8 jh bh Ñ
9 gh ḍh dh Ṣ
10 j b g ḍ d Ś
11 kh ph ch ṭh th
c ṭ t V
12 k p Y
13 ś ṣ s R
14 h L
6
ŚIVASŪTRA-S
• so 'yam akṣara-samāmnāyo vāk-samāmnāyaḥ puṣpitaḥ phalitaś candra-tārakavat
pratimaṇḍito veditavyo brahmarāśiḥ / sarva-veda-puṇya-phalāvāptiśca asya jñāne bhavati
/ (MB, I: 36)
“This then is the listing of sounds, a listing of speech, bearing flowers and fruit, shining like
the moon and the stars, should be known as the collection of Brahman. The knowledge of this
[listing] leads to the attainment of the merit identical with that of the merit from all the
Vedas.”
• Kiparsky: If we did not have the text of the Aṣṭādhyāyī but merely a pre-theoretical
description of Sanskrit phonology, the main properties of Pāṇini’s grammar could be
inferred just from the way the phonemes of Sanskrit are organized in the Śiva-sūtra-s.
ŚIVASŪTRA-S
7
1 a i u Ṇ • If a list contains the sequence of elements
2 r̥ l̥ K
3 e o Ṅ [ . . . xp, xp+1, . . . xqQ] where Q is a marker, then the abbreviation
4 ai au C xpQ denotes the set xp, xp+1, . . . xq.
5 h y v r Ṭ
6 l Ṇ • For example,
7 ñ m ṅ ṇ n M a. iK = i, u, r̥ , l̥
8 jh bh Ñ
9 gh ḍh dh Ṣ b. aC = a, i, u, r̥ , l̥, e, o, ai, au = vowels
10 j b g ḍ d Ś
c. yaṆ = y, v, r, l
11 kh ph ch ṭh th
c ṭ t V d. jhaL = obstruent
12 k p Y
e. haL = consonants
13 ś ṣ s R
14 h L f. aL = vowels+consonants
ŚIVASŪTRA-S
8
1 a i u Ṇ • Consider the sūtra, इको यणचि (iko yaṇaci)
2 r̥ l̥ K
3 e o Ṅ = iKaḥ (genitive case) yaṆ (nominative case) aCi (locative case)
4 ai au C
Here,
5 h y v r Ṭ
6 l Ṇ • iK = i, u, r̥, l̥ ; yaṆ = y, v, r, l ; aC = a, i, u, r̥, l̥, e, o, ai, au
7 ñ m ṅ ṇ n M
8 jh bh Ñ इको यणचि = replace iK with yaṆ before aC
9 gh ḍh dh Ṣ
Example:
10 j b g ḍ d Ś
11 kh ph ch ṭh th
दचि + अत्र → दि् य् + अत्र → दध्यत्र ।
c ṭ t V
12 k p Y
मिु + इचि → मि् व् + इचि → मध्विचि ।
13 ś ṣ s R
14 h L
DHĀTU-PĀṬHA 9
• List of approx 2000 verbal roots
• Pāṇini identifies primarily two lexical terms: prakrti (base), and pratyaya (affix). Bases are of two types: dhātu
(verbal roots), and prātipadika (nominal stems); affixes are of five types: kr̥t, tiṅ, sup, taddhita, and strī.
• A combination of the different types of bases with the different types of affixes gives us a variety of words:
a. [Root + Affix]Root = desideratives, intensives, causatives.
b. [Word + Affix]Root = denominal verbs.
c. [Root + Affix]Stem: primary (kr̥t) suffixes.
d. [Word + Affix]Stem: secondary (taddhita) suffixes.
e. [Word + Word]Stem: compounding.
f. [Root + Affix]Word: verb inflection.
g. [Stem + Affix]Word: noun inflection.
10
DHĀTU-PĀṬHA
• भू सत्तायाम् (to exist, to become, to be, to happen) => भवचि
• एिँ वृद्धौ (to grow, to increase, to prosper, to extend, to swell, to rise) => एििे
• स्पिध स
ँ ङ्घर्षे (to challenge, to compete, to conflict, to rival, to struggle) => स्पिध िे
• गािृँ प्रचिष्ठाचिप्सयोर्ग्ध न्थे ि (to praise oneself, to desire, to tie together, to string
together, to seek, to stay together, to stand firmly, to compile) => गाििे
• बािृँ िोडने, रोटने (to oppress, to torment, to harass, to destroy, to infect, to torture) =>
बाििे
• नािृँ याच्ञोपिापैश्वयाध शीष्षु (to seek help, to beg, to irritate, to bless, to master =>
नाििे
11
GAṆAPĀṬHA, ETC
• Inventory of classes of lexical items subject to specific rules
• These sets are of two types: some have a definite number of nominal stems, while some are
open-ended.
• The authorship of the Gaṇapāṭha is doubtful
• Uṇādisūtra-s: extend the Pāṇinian technique to analyze irregularly formed derivatives
from roots.
• Phitsutra: deals with the accentuation of linguistic items that are not developed through a
derivational process from underlying bases and affixes.
• Liṅgānuśāsana: deals with assigning gender to nominals based on their structure and
meaning.
12
TYPES OF SŪTRA-S
• Definitions (saṁjña-sūtra-s): Introduce the technical terms of the grammar
• Meta-rules (paribhāṣā-s): Play a pivotal role in the correct interpretation and application
of the operational rules. They may be classified into three categories:
(a) vācanikī: explicitly stated by Panini in the Aṣṭādhyāyī,
(b) jñāpakasiddha: those which Panini must have implicitly assumed;
(c) nyāyasiddha: axiomatic; norms or standards of the outside world.
• Headings (adhikāra-sūṭra-s): Headings supply a common element for a group of rules
• Operational rules (vidhi-sūtra-s): Carry out four basic types of operations on strings:
replacement, affixation, augmentation, and compounding.
13
OPERATIONS: FORMAT
The basic format of an operational rule is:
• [1a] A → B / C_D
[1b] AGenitive BNominative CAblative Dlocative
• [2a] A B → C
[2b] A → A A
• [3] ANominative BInstrumental → ‘A is compounded with B’
• Rule A has priority over rule B if A is
a. ordered after B (para)
b. applicable whether or not B applies (nitya)
c. conditioned internally to B (antaraṅga)
d. applicable in a proper subset of the cases to which B is applicable (apavāda)
A N U V R̥ T T I 14
• Pāṇini uses a metalinguistic device called anuvrtti/recurrence for brevity.
Consider the example: (3.1.62) acaḥ karmakarttari: It has two elements, “acaḥ” & “karmakarttari.” In order for it to be sensible, it
must borrow elements from the previous sūtra-s:
3.1.1 pratyayah,
3.1.2 parasca,
3.1.22 dhātorekāco halādeḥ kriyāsamabhihāre yaṅ
3.1.43 cli luṅi,
3.1.44 cleḥ sic
3.1.60 ciṇ te padaḥ
3.1.61 dīpajanabudhapūritāyipyāyibhyo'nyatarasyām
It now reads,
ajantād dhātoh parasya cleh pratyayasya karmakarttari taśabde parataś ciņādeśo bhavaty anyatarasyām:
Affix “cli” is optionally replaced by the affix “ciṇ” when the former occurs after a dhatu ending in aC and when “ta,” a
replacement of luṅ, which denotes karmakartr (object treated as if agent) follows.
I T - SA M J Ñ A
15
• 1.3.2 upadeśe'janunāsika it [An anunāsika (nasal) aC (vowel) present in an upadeśa is called “it”]
• 1.3.3 hal antyam [The ending “haL” (consonant)]
• 1.3.4 na vibhaktau tusmāḥ [Not the letters - ि्, थ्, द् , ि्, न्, स्, म् – of the affixes called vibhakti or
(inflective affixes)]
• 1.3.5 ādirñiṭuḍavaḥ [The initial चञ , टु , डु ]
• 1.3.6 ṣaḥ pratyayasya [The र्ष् of affix].
• 1.3.7 cuṭū [The letters - ि् , छ् , ज्, झ्, ञ्, ट् , ठ् , ड् , ढ् , ण्]
• 1.3.8 laśakvataddhite [The letters ि् , श्, क्, ख्, ग्, घ्, ङ् of non-taddhita affixes]
• 1.3.9 tasya lopaḥ [It is elided/deleted]
IT-SAMJÑA-EXAMPLE
16
• 1.3.2 upadeśe'janunāsika it [An anunāsika (nasal) aC (vowel) present in an • डु पिँर्ष् (prakr̥ti - dhātu)
upadeśa is called “it”]
• टु ओचँ श्व (prakr̥ti - dhātu)
• 1.3.3 hal antyam [The ending “haL” (consonant)]
• 1.3.4 na vibhaktau tusmāḥ [Not the letters - ि् , थ्, द् , ि्, न्, स्, म् – of the • िुमुन् (pratyaya)
affixes called vibhakti or (inflective affixes)] • शस् (pratyaya -
• 1.3.5 ādirñiṭuḍavaḥ [The initial चञ , टु , डु ] vibhakti)
• 1.3.6 ṣaḥ pratyayasya [The र्ष् of affix].
• चञचमदाँ (prakr̥ti –
dhatu)
• 1.3.7 cuṭū [The letters - ि् , छ् , ज्, झ्, ञ्, ट् , ठ् , ड् , ढ् , ण्]
• डु कृञ् (prakr̥ti – dhatu)
• 1.3.8 laśakvataddhite [The letters ि् , श्, क्, ख्, ग्, घ्, ङ् of non-taddhita
• ष्ट्रन् (kr̥t pratyaya)
affixes]
• 1.3.9 tasya lopaḥ [It is elided/deleted] • ष्ट्रन्ष्कन् (taddhita
17
STRUCTURE
• There is interdependency among the sūtra-s;
• Interdependency is of two types: intradomain and interdomain.
• Further, the Astadhyāyī is divided into two basic organizational units: sapādasaptādhyāyī
'the first seven books and one quarter' and tripādī 'the last three quarters’.
• This twofold division is shown by the sūtra, 8.2.1 pūrvatrāsiddham which states that rules
of the last three quarters (tripādī) are treated as suspended (asiddha) in view of rules of the
first seven books and one quarter.
=> lack of interaction among rules of the two units
18
SY N TA X
• Pāṇini accounts for sentence structure by a set of grammatical categories which allow syntactic
relationship to be represented as identity at the appropriate level of abstraction: karaka-s.
• A sentence is seen as a little drama played out by an Agent and a set of other actors, which may
include Goal, Recipient, Instrument, Location, and Source.
• Kārakas are the categories in terms of which the assignment of case and other morphological
elements is formulated.
• The key principle is that every kāraka must be expressed (abhihita) by a morphological element, and
none can be expressed by more than one.
• Another principle ensures that any given argument gets only one role. For example, in dhanuṣā
vidhyati ‘he pierces by means of a bow’ (i.e. with arrows shot from a bow), dhanuṣ ‘bow’ gets only
the instrumental role.
19
T H A N K YO U !
