Paul Thieme has treated exactly this subject with the result, if my memory doesn't fail me, that Pāṇini's motivation - in conformity with late Vedic thought - would have been to formulate a grammatical "truth":
Thieme, Paul: Meaning and form of the 'Grammar' of Pāṇini. Studien zur Indologie und Iranistik 8-9 (1982): 1-34
[= Kleine Schriften 3, 1170-1201]


Prof. Dr. Walter Slaje
Hermann-Löns-Str. 1
D-99425 Weimar

Ego ex animi mei sententia spondeo ac polliceor

studia humanitatis impigro labore culturum et provecturum

non sordidi lucri causa nec ad vanam captandam gloriam,

sed quo magis veritas propagetur et lux eius, qua salus

humani generis continetur, clarius effulgeat.

Vindobonae, die XXI. mensis Novembris MCMLXXXIII.

2014-12-07 16:25 GMT+01:00 George Hart <>:
This discussion disregards the elephant in the room, in my opinion, and that is “why?”  I know of no other culture that went to such effort to describe/prescribe a language — and we should remember that there were other systems than Pāṇini’s.  The mindset that produced Pāṇini and other grammatical treatises also worked — and still works — for Tamil, which has the Tolkāppiyam, though it’s not as compact as Pāṇini.  If we can discover why so much effort was devoted to compacting Pāṇini’s formulations and why it was preserved so carefully we can gain insight into whether it was meant to be oral.  Surely it is far easier to memorize ślokas and other kinds of rhythmical language than Pāṇini’s sūtras — I do not believe their brevity has anything to do with memorization (I remember Ingalls once remarked on this).  I’m not convinced by the writing-materials argument either.  Pāṇini was impelled by a need for brevity, completeness and correctness, much as a computer programmer is.  Everyone knows the śloka about how a grammarian rejoices more if he can reduce the length of a sūtra by one syllable than he does if he has a son.  Isn’t it possible that Pāṇini was motivated by a desire for a kind of perfect elegance and nothing more?  But that still begs the question of “why.”

Perhaps it would be germane to look at the linguistic situation of N. India in Pāṇini’s time.  It’s hardly my field, but I would guess that Vedic Sanskrit was long since extinct as a spoken language and that everyday speech consisted of various sorts of Prakrits, Pali, and the like.  Here, the diglossia of modern Tamil may provide some insight.  Spoken Tamil is about as far from written Tamil as Prakrit/Pali is from Sanskrit.  Yet even though people do not speak it normally, formal written Tamil is cultivated assiduously and people all learn it.  It is used in many venues — political speeches, purāṇa movies, and most writing.  My guess is that in Pāṇini’s time, there was a need to define “elegant” language to be used for higher diction.  No doubt the Vedas also played a role.  Without definitive guides like Pāṇini’s grammar, Sanskrit would become corrupt — something we see in Buddhist Sanskrit texts.  To prevent this, a prescriptive tool like Pāṇini (or the Tolkāppiyam) is essential.

George Hart

On Dec 7, 2014, at 4:32 AM, Dipak Bhattacharya <> wrote:

To explain the unique phenomenon physiologically disregarding the Indian context may lead at best to downplaying it, and worse even to a theory of racial difference. Many earlier philologists did not admit the oral transmission in case of large prose texts like the Brāhmaṇas. Now the fact is accepted. So there was a unique development in India. As belonging to this new conscious generation Professor Houben is very welcome.

The contrary views are inevitable but the given aesthetic explanation does not explain their non-occurrence in early historical China or Greece. There is no reason to disregard the relatively late emergence of written literature in India which fits in with the phenomenon.

Pāṇini's Aṣṭādhyāyī is an easy text to remember because of its successfully structured composition. But see Whitney's comments in the 1880s that Thieme had to contradict. The Aṣṭādhyāyī was not easy for Whitney and Weber. The actual magnitude in normal language, along with its computer program like compactness and intertwining of the rules, will be apparent from the Kāśikā commentary. Normally, that is not got by heart. I did not mention Patañjali because of his metalinguistic discussions.

Two more points. The pedagogic need and oral transmission both contributed to the phenomenon. Secondly their continuance in India was most probably caused by both scarcity and perishability of the normal Indian paper that is palm leaf in late medieval India and earlier, perhaps, any leaf which was less durable. Birchbark was expensive for the village scholar.

I tried to treat the matter more extensively elsewhere. I heartily encourage the dialogue but at the same time can assure that the related problems and issues are of considerable magnitude not accommodable in a forum discussion.


On Sun, Dec 7, 2014 at 4:31 PM, Dominik Wujastyk <> wrote:
Respecfully, I disagree.

Many generalizations are made about oral literature, but empirical or quantitative evidence to support such assertions is less common.  By quantitative evicence, I mean, for example, counting actual volumes of material, analyzing its uses, the relation of comprehension to volume, and so forth.  Do we even know, really, the answer to even such a simple question as "what makes a text easy to memorize?"  For example, one migh think, prima facie, that the proliferation of very similar sentences and paragraphs, such as we see in the Tripitaka, would lead directly to error.  But apparently this extensive repetition (which we also see in sastric texts like the Mahabhasya) was a positive contribution to accurate transmission, not a negative one.  Another simple question would be, does the amount of text to be memorized have any impact on accuracy or ease of memorization?  Or is the volume of text only related to the amount of time necessary to memorize it?  It seems likely that "memory muscles" would get stronger the more a person practiced.

In short, there are many questions, including ones bordering on neurology, that bear investigation as part of an exploration of the factors affecting memorization.

The idea that Panini's grammar contains abbreviations etc. simply as a side-effect of oral transmission seems to me to be such an unfounded assertion.  We know for a fact that gargantuan amounts of literature were memorized by Vaidika brahmanas and Buddhist monks.  By comparison, the Astadhyayi is a tiny text.  I would hypothesize that in a cultural milieu that supported memorization and where students had sufficient resources of time, memorizing the Astadhyayi was really quite easy. 

My own studies of Panini's system lead me to believe that the various techniques of abbreviation it uses are motivated by a sense of beauty, efficiency, internal coherence and perhaps even intellectual playfulness.  Exactly what mathematicians call "elegance."  I think we can ask whether the Astadhyayi's internal concision is not better understood as a product of intensive use by an expert community over a period of time, like a tool becoming smooth in the hand of an artisan.  Were mnemotechnical devices not developed so intensively in Paninian grammar because they provided vaiyakaranas with aesthetic pleasure and astonishment, in ancient times as much as today?

Dominik Wujastyk

On 7 December 2014 at 10:18, Dipak Bhattacharya <> wrote:

Wonder but not unimaginable.

One can be fairly certain that the many characteristic features of the Aṣṭādhyāyī , terse prose-- sentences without finite verb, not an essential feature of contemporary ritual sūtras, obsession with brevity (lāghava), enumerative ie non-explicative definition, almost total absence of argumentation, the abridgement of material worth some thousands of pages of discussion, into a few pages of 4004 sūtras,  can be explained only by the postulation of an environment of entirely oral transmission of knowledge. So at a time not later than the 4th century BCE. The Ṛgvedic teaching and training could not be different.

The real wonder is that the practice continued even after writing material became available sufficiently.


On Fri, Dec 5, 2014 at 10:23 PM, Jan E.M. Houben <> wrote:
Would it really be possible that vaidika pandits learn by heart such huge amounts of text, and are able to reproduce them correctly, syllable by syllable and tone by tone?
The following clips demonstrate the predominant orality and memory culture in current Vedic schools.
Among these clips, the second is a proof that at least the group of students studying the Saamaveda on that day (pre-dawn hour) in February 2001 did not use any hidden piece of paper to read the text of their lengthy chants: see what happened to their chant when there is a failure of electricity and the light goes off.
The chanting tradition followed here is that of the rare Ranayaniya school of the Saamaveda.
The fourth clip shows the performance of the Pravadbhaargava saaman earlier studied by the students.
Intended public of these clips:
students in Indology, Indian Studies, Ritual Studies, History of Education, History of Music, Ethnomusicology.


Prof. Dr. Jan E.M. Houben,
Directeur d Etudes « Sources et Histoire de la Tradition Sanskrite »
Ecole Pratique des Hautes Etudes, Sciences historiques et philologiques,
Sorbonne – 54, rue Saint-Jacques
CS 20525 – 75005 Paris – France. 

INDOLOGY mailing list

INDOLOGY mailing list

INDOLOGY mailing list

INDOLOGY mailing list