On the Condensation and Extension of Knowledge: The Sūtra Style in the Tolkāppiyam Poruḷatikāram1
p. 177-206
Texte intégral
cūttiram tāṉē
āṭi niḻaliṉ aṟiyat tōṉṟi
nāṭutal iṉṟip poruḷ naṉi viḷaṅka
yāppiṉuḷ tōṉṟa yāttu amaippatuvē. (TP 472)2
The earliest extant theoretical work in Classical Tamil language is, as is well-known, the venerable Tolkāppiyam. The oldest layers of its third, poetological part, the Poruḷatikāram (TP), may or may not be roughly contemporary with its first and second parts, concerned with phonetics and grammar.3 As for their style, however, the three parts are roughly similar, and in describing style we will address several related though distinct questions: the organisation, the transmission, the growth of knowledge. In other words, style opens a view on textual history, and textual history is not a dry philological exercise, but a means to gain insights into historical processes, a text (at least the text in question) being a testimony of a process rather than a snapshot of a momentary state or even an eternal teaching.
1The paper presented here, however, cannot hope to give more than a brief exposition of the problems. In many respects what is said will appear to be premature, but simply in order to explain why this should be so, it seemed necessary to put it into words. To give a rough outline of the conceptual horizon (restricted to the Akam part, that is, that part of poetics concerned with love lyrics), in Tamil there is a marvellously rich and extensive poetological tradition, beginning in, perhaps, the first centuries of the Christian era with brief commentaries (kiḷavi-s) on single poems and treatises/school traditions the phrasing of which still can be traced in the TP and the Iṟaiyaṉār Akapporuḷ (IA).4 The tradition moves on with texts that have been transmitted to this day, the IA, quite concise and probably more or less of one casting, and the TP, rather long and probably growing for centuries. Then follows a gap of several hundred years (which demonstrably represents a major break in the tradition), ended by a medieval renaissance of Classical poetry and poetics, attested in a wave of commentaries on poems and treatises beginning in about the 12thcentury with Nakkīraṉ’s commentary on the IA,5 or with Iḷampūraṇar (Iḷ.) and culminating in about the 15th century with Nacciṉārkiṉiyar (Nacc.). New poetological treatises are written too, such as the Akapporuḷ Viḷakkam (perhaps of the 12th century), the latest major one being the Ilakkaṇa Viḷakkam of the 18th century, both in their turn provided with commentaries.
2Now in order to understand what has happened early on in this enormous mass of transmitted materials it is necessary to keep apart two facts. On the one hand, it can be shown beyond doubt that already the early, that is, medieval, commentators were struggling hard with the text of the TP. The differences between Iḷ. and Nacc. testify not only to uncertainties in the textual transmission − differences in the wording of the root text and problems regarding the length as well as the sequence of sūtra-s − but also to major difficulties of interpretation: silent glossing-over of problems, contradictory or obviously biased explanations. This means that it will not be reasonable to rely on these commentaries. On the other hand, at least the outer surface of the materials transmitted bears all the signs of an age-long transmission process: under South Indian climatic conditions, manuscripts have to be reproduced roughly every 100 years, if not earlier, and it seems reasonable to assume that it is the "modern" experts on poetics who actually have copied and handed down also the time-honoured treatises and their commentaries and improved them to the best of their skill and knowledge. This means that in order to evaluate the stylistic disparities not only to be observed in the root text, but in the exegetical texts as well,6 one will have to trace peculiar phrasing and word usage even in the most recent specimens of the poetological idiom, if there is ever to be a chance of finding out who made changes and additions (and possibly also omissions?) and why.7
3To summarise, the goal will not be to deconstruct tradition, but rather to take it for what it is: a long process of interpretation, re-interpretation and misinterpretation, a struggle towards the integration of disparate concepts, a reconciliation of varying interests, a concession to the change of tastes. This will be quite a time-consuming task, and what I propose to do here is simply to start moderately at one corner: to understand as closely as possible the wording and structure of a part of the TP as it has come down to us, in order to find out what is typical of the text (and perhaps for the genre) and what is special. Even if it will not be possible for the time being to account in every way for this wording (occasionally it doesn´t even seem comprehensible), it will be useful to note its peculiarities. One major problem is of course already the wording of the text itself. The commentators, as has been noted, deviate from each other with respect to the sūtra-s, and at times also the text of the commentaries is uncertain. It will be unavoidable to collect manuscripts and find out at least what the testimonies actually are, which is impossible on the basis of the current editions.8 Keeping, however, to what is feasible at the moment − how does one describe the outstanding features of "our" TP text?
4First of all, it is not a prose text, it is a metrical one.9 Like the larger part of the poetry it is written in Akaval metre, the intricacies of which need not be discussed here. Important is that it consists of 4 metrical feet of rather variable length per line, and that, unlike in the poems, the penultimate line is not one foot short. This is already a first indication as to the possible length of sentences: the minimal sentence would be expected to comprise 1 line.10 This type of sentence is actually to be found, but the longest TP sentence makes up not less than 59 lines. The smallest unit of innertextual division is called a sūtra11, and this usually comprises one complete sentence (possibly with several dependent clauses), occasionally more than one. The word "sūtra" is not used as a term of self-reference by the TP itself, but it is the one used by the commentators from Nakkīraṉ onwards (cūttiram).12
5Now, the use of the word sūtra, even if it had been applied to the specimen in question only about a thousand years later,13 naturally raises a certain number of expectations, since in Sanskrit/Sanskrit philology it is connected with certain properties. This problem, to be sure, cannot be treated without considering the general question of the influence of Sanskrit thinking in the Tamil tradition. This is a most complicated issue, and it is necessary to restrict it here to a few essential points. As for the TP, some kind of Sanskrit influence can be observed on 4 levels (which at least partly have to be connected with different chronological layers, but to this we will come below).
Plainly evident is a reception of Sanskrit poetics, visible in the borrowing of technical terms such as uvamai/uvamam for upamā. On the basis of this Takahashi has established a first rough division of the 9 TP sections into unsystematic and un-Sanscritic, and therefore early (TP 1, 3-5, [2]), and systematic and Sanscritic, and therefore late (6-9).14 A detailed investigation of concepts, their provenance and their Tamil interpretation has yet to be carried out; moreover traces are also to be found in the "early" parts.
There is a share of Sanskrit vocabulary in the TP which is not easily explained by conceptual borrowing. Quite ordinary everyday words like kālai or nimittam are used in places where the Tamil equivalents supposedly would have done just as well. The same can be observed in the poetic texts, and there at least in some cases, the word-usage might be considered to betray an acquaintance with Sanskrit literature, as in the case of kāmam.
There is an obvious influence of Northern notions not so much on a theoretical (poetological) level, but on a more general cultural one, and it is paired with a conscious (and at times seemingly considerable)15 effort to avoid Sanskrit words, as in TP 89, starting with iṉpamum poruḷum aṟaṉum, presumably a transposition of the Sanskrit kāma, artha, dharma.
There is a distinct (and probably late) superimposition of Northern morals pervading the whole transmitted text of the TP: brahmins, varṇa-s and Vedas.
6The kind of influence, or for the time being rather a similarity, actually at issue in the present context is less straightforward to define. It is a question of form. At least some parts of the TP, which seem to have a theoretical concept, are written in concise, hierarchical and interdependent sentences at first sight not unlike the sūtra style of Sanskrit, and at a later stage the Tamil tradition itself chooses this term to refer to them. Whether this is more than just a parallelism is impossible to decide offhand, but a closer description might contribute to a basis for further investigation.16 In order to achieve this we will have to ask, first of all: what is characteristic of a sūtra in the Sanskrit tradition?
7Leaving aside the partly unsolved problems of origin and historical development17, the three basic features of a sūtra as denoting a rule in a theoretical treatise are conciseness18, hierarchy and sequence. A conception is packed into formulations as terse as possible19 and in fairly uniform constructions20. Statements are made on different levels of validity, a bit idealizingly we can discern meta-rules, rules and sub-rules. Subsequent sūtra-s can stand in a very close relationship, realised on the syntactical level by ellipsis, thus adding considerably to the appearance of brevity.
8On treatise level this means, in an ideal case: a lucid structure, where it is possible to follow the course of the argument in an organized whole, and the development of technical terminology. Ideal, because this is often not what we find when looking more closely at transmitted texts. In fact, we frequently encounter more complex formations where a supposed clear outline has been obscured by various accretions. There are basically two ways to account for such a state of affairs, which might be termed logical and chronological models. A logical model would proceed from the idea of a number of prescriptions and/or definitions enlarged by complements, thus in a way mirroring the complexities of the reality which is to be described by them (a concept to be found also in the designation of rules and sub-rules). A chronological model would transpose the obvious extensions on a temporal scheme, the idea of a nucleus gradually enlarged by additions. Both models need not be incongruent, but they tend to emphasize different aspects. While the logical one is well suited to explain a complex net of interrelations which may be adequate to the facts (the entity that is to be described or prescribed by the treatise: actual and/or acceptable poetry in the case of poetics), the chronological one is more adaptable to the explanation of breaks which can be explained as part of a historical development.
9A third model is envisaged in Srinivasan 1980, a study on the composition of the rasa section of the Nāṭyaśāstra. Here the obvious textual and argumentative disparities and the lack of positive evidence allowing for a convincing stratification have led the author to the conclusion that it is impossible to extract a nucleus (be it a conceptional or a temporal one) which has been enlarged by additions, that instead the text must have been composed, or rather compiled, as an addition of several fundamentally incongruent and partly simply irreconcilable notions. Now, the case of the Nāṭyaśāstra might still be open to discussion, but what Srinivasan´s study demonstrates all too lucidly is that the general level of argumentation in favour of textual division or stratification is still not subtle enough by a long way and needs a lot of further thinking, paying heed also to the special conditions of every particular text.
10Now, apart from the question of whether there is a historical relationship between the Sanskrit and the Tamil material (an at least formal influence of a/the Sanskrit sūtra style on Tamil poetological texts), we can ask whether the hermeneutical background developed in Sanskrit philology to deal with this kind of text is suitable for describing a similar phenomenon in the Tamil tradition. In order to test such an assumption, the goal of the following considerations will be to analyse the structure of one treatise on the micro-level (the formation of sentences which constitute sūtra-s) and on the macro-level (the more or less comprehensive treatment of a theme). As an exemplary case, I have chosen the Akattiṇai-iyal of the TP (TP 1), which can be called an intermediate type between a well-structured and organized whole, such as is to be found in the IA, and a mere conglomerate of relevant material, a compilation in the sense of Srinivasan, like the Kaḷavu-iyal of the TP (TP 3). The text (as read by Iḷ. plus the deviations of Nacc.) with a (no doubt sadly preliminary) annotated translation is given as an appendix.21
11When turning to the microstructure, the sūtra, we have to face first of all the problem of sūtra-division. It can be quite different with the different commentaries22, and the rationale behind this is easy to discern: Iḷ. splits according to content (which occasionally leads to very elliptical sentences, as in the case of TP 6+7), while Nacc. heeds the metrical boundaries. This means that there is at least one sentence per sūtra (however elliptical), and possibly more than one (rarely if ever more than two).23
12Length, as has been mentioned already, has the astonishing range of 1 line (a metrical line of 4 cīr with Nacc., with Iḷ. even 3 cīr) to 59 lines. The latter is the case with TP 144, the kiḻavōṉ sūtra of the Kaṟpu-iyal, and it is the longest but by no means an exception. Especially among the speaker sūtra-s there are several well above 30 lines. TP 1, however, retains an intermediate position also in this respect. While 1 line is fairly frequent (18 of 55/57 sūtra-s in TP 1), the average lies with 2 lines (23 sūtra-s in TP 1), and the rest are longer (4 times 3 lines, 4 times 4 lines, 6 longer than 4 lines). The longest here is sūtra 43 with 24 lines (also the one for the man as a speaker).
13There are two sets of fairly developed technical terminology. One of them pertains to poetics and mostly is explicitly defined in the course of the argument; parts of it are shared also by the IA (and, of course, the subsequent tradition). The other is a basic inventory of phrases employed in shaping rules, let us call it the sūtra idiom, which is not totally homogenous (and probably changing in the course of time), but common to the whole Tolkāppiyam as well as the IA. It makes use of a limited number of constructions, the most important of which shall be presented here.
14Let us first of all take a look at the structure of main sentences. The three basic types are definition, application and extension sūtra-s, less frequent are prohibitions. A definition usually consists of a nominal sentence "X [is] Y", or "what is called X (eṉa, eṉappaṭuvatu) [is] Y". Often the predicate noun precedes the subject, but not generally; in TP 6+7, for example, there is a chiasm. An application very often has the form "Y (nom.) is suitable (uriyatu, sg., uriya, pl.) for X (dat.)", or "Y (nom.) [is] with X (loc.)", i.e. certain occasions of speech are connected with a particular speaker or certain times with a particular setting. Extensions are either of the same form as applications, making use of a conceding "also" (-um), or they state that something is not to be excluded (kaṭivarai il-). Prohibitions are mostly negated applications. All of these main sentences can be concluded by a kind of authorization: "so they say, so say the scholars" (eṉpa, eṉmaṉār pulavar).24
15Interesting is the way sūtra-s are put in sequence. As can be gathered from table 1 (see below), there are thematic hierarchies of definitions, that is, for example, settings connected with basic coordinates consisting in places and times. Repetitions of the hierarchic nexus are avoided, they have to be established by sequence. If references are made, it is with the help of pronouns, which contributes considerably to the density of the exposition and constitutes a major textcritical and exegetical problem: some sūtra-s are not placed unanimously within the tradition, and their meaning can be considerably different depending on the context they are put in (see, for example, TP 16 with notes). Another case is that there are pronouns which have no proper reference point in the preceding sūtra, which means that in the given context they are hard to make sense of (see, for example, TP 54 and note).25
16The sub-constructions are even more uniform than the main sentences. They are usually enumerations of predicate nouns or enumerations of conditions for a predicate noun. Depending on whether they refer to things, facts, actions or occasions they consist of nouns, verbal nouns, locative phrases or clauses ending in an absolutive or a conditional. Enumerations can, but need not be coordinated by "and" (-um) and may be concluded by a summarising tokaii. To indicate the incompleteness of an enumeration a piṟavum, "and others" can be added at the end. Thus it is to be explained that the long sūtra-s are not basically different from the shorter ones; they contain lengthy subordinate enumerations.
17A most remarkable trait, now, is revealed by syntactical analysis: while the short sūtra-s as a rule represent well-formed sentences, the long ones more often than not depict a quasi-anacolouthic agglutination of clauses: mixed nouns, verbal nouns, locatives, absolutives, conditionals (or 5 lines of conditionals followed by two lines of locatives and the like), in one line coordinated by -um, but not in the next, tokaii or piṟavum which are not concluding the enumeration, in bad cases even clauses which have no point of reference in the rest of the sentence. These things are markedly worst in the speaker sūtra-s, not only in TP 1, but all the more in 3 and 4. The most conspicuous example in TP 1 is s. 43 (see translation + notes). Here, the overall impression is that people must have added sub-themes or occasions of speech for quite some time.
18Thus far for the structure of sentences. As for the thematic structure, it is, despite various accretions and several commentatorial misrepresentations, still remarkably clear.26 We can discern three rather loosely connected parts, one dealing with the settings (tiṇai-s) and their constitutive elements (mutal-poruḷ, uri-p., karu-p.), one concerned with the basic dramatic situation for poetry, namely separation (pirivu), and the different speakers, and finally some addenda to part I and II plus several unrelated ones.

table 1. the structure of TP 1
In other words, there are two counter-active movements of condensation and extension discernible on a macro- as well as on a micro-level: a systematic and hierarchic text, the structure of which is obscured by addition and insertion of sūtra-s − simple and brief sentences enlarged by adding clauses up to the very limit of intelligiblity. An explanation that might account for such a structure and which can at least claim to be plausible, even if it might be yet another thing to actually prove it, is that there was development over a longer period. Combining considerations of thematic plausibility with the distinction of stylistic features, it is possible to derive a
Tentative chronological model in 4 layers for the development of the Tolkāppiyam Poruḷatikāram Akattiñai-iyal27

Applying this model to the text, we might divide it up as follows:
TP Akattiṇai-iyal
I. the settings
1:eḻutiṇai 2nd layer |
2: aintiṇai 2nd layer |
3: mutal-, karu-, uri-poruḷ 1st layer |
4: mutal: nilam (5) (?) + poḻutu of aintiṇai (6,7,9,10,11) 1st layer |
5 : gods and landscapes 2nd/4th layer? |
8,12: paṉi and piṉpaṉi 2nd layer? |
16, 17, 18?: in the situation of pirivu 1st layer? |
13, 14, 18?: tiṇai-mayakku 2nd layer |
15 : uri-s 1st layer |
19: karu: elements 1st layer |
20: mayakku 2nd layer |
21-25: karu: people |
II. the dramatic situation
26, 27, 34, 35: pirivu 1st layer |
28-33: complements 3rd/4th layer? |
36f.: restrictions 4th layer |
38, 41-44: the dramatis personae 2nd + 3rd layer? |
39-40: restrictions 4th layer |
III. addenda
19The last part of the section doesn´t seem to have a thematic nexus at all and simply consists of diverse addenda.
45: speaking of the past 2nd/3rd layer? |
46, 47:complement on mayakku 2nd/3rd layer? |
walking through the desert and remembering his last encounter with HER). Still, this might also be a further specification as typical of layer 3, especially since the last sūtra here is either badly composed or irrelevant − unless it mirrors a later discussion of marapu, the word usage (thus the title of the late TP 9). |
48-51: insertion on uvamam-s (subject of TP 7) 3rd layer |
52-54: complement on the 2 additional tiṇai-s 2nd/3rd layer? |
55: addendum on metres (subject of TP 8) 3rd layer |
56, 57: addendum on proper names 3rd/4th layer? |
Appendix: Annotated Translation of TP Akattiṇai-iyal
1. (seven tiṇai-s)32
kaikkiḷai mutalāp peruntiṇai yiṟuvāy
muṟpaṭak kiḷanta veḻutiṇai yeṉpa.
20They say the seven settings expressed clearly before
[are what] begins with33 kaikkiḷai [and] ends with peruntiṇai.34
2. (5-1 tiṇai-s)
avaṟṟuḷ
naṭuva ṇaintiṇai naṭuvaṇa toḻiyap
paṭutirai vaiyam pāttiya paṇpē.
21Among these,
the five in the centre, apart from the one in the centre,
[have] the quality of dividing the earth [surrounded] by permanent (?) waves.
3. (mutal- karu- uri- poruḷ)
mutalkaru vuriporu ḷeṉṟa mūṉṟē
nuvaluṅ kālai muṟaiciṟan taṉavē
pāṭaluṭ payiṉṟavai nāṭuṅ kālai.
22These three, named sense-units (poruḷ)35 [namely] basis (mutal)36, ingredient (karu)37 [and] mood (uri)38, are important in the order at the time one utters [them]39− when one examines what is employed in song.
4. (mutal = nilam + poḻutu)
mutaleṉap paṭuvatu nilampoḻu tiraṇṭiṉ
iyalpeṉa moḻipa viyalpuṇarn tōrē.
23What is called mutal [has] the nature of two, [namely] region [and] time,
say those who have understood [its] nature.
5. (mullai, kuṟiñci, marutam, neytal)
māyōṉ mēya kāṭuṟai yulakamum
cēyōṉ mēya maivarai yulakamum
vēntaṉ mēya tīmpuṉal lulakamum
varuṇaṉ mēya perumaṇa lulakamum
mullai kuṟiñci maruta neytaleṉav
colliya muṟaiyāṟ collavum paṭumē.
24The world of the forest residence, connected with (?)40 Māyōṉ,
and the world of the black41 mountains, connected with (?) Cēyōṉ
and the world of the sweet flood, connected with (?) Vēntaṉ,
and the world of the great sands, connected with (?) Varuṇaṉ42,
[these] are called43 mullai, kuṟiñci, marutam [and] neytal,
in the order they have been named.
6. + 7.44 (mullai; kuṟiñci)
kāru mālaiyu mullai kuṟiñci
kūtir yāma meṉmaṉār pulavar.
25Rainy season and evening [is] mullai,
kuṟiñci [is] cool season [and] midnight,
say the scholars.
8. (paṉi)
paṉiyetir paruvamu muritteṉa moḻipa.
26The season facing the dew is also suitable, they say.45
| |
| vaikuṟu (Nacc.) |
|
27That daybreak [and] dawn46 [are] marutam,
that afternoon is neytal, appears to be true (?)47.
11. (the middle tiṇai)
naṭuvunilait tiṇaiyē naṇpakal vēṉiloṭu
muṭivunilai maruṅkiṉ muṉṉiya neṟittē.
28The setting situated in the middle has the tradition48
of being(?)49 the result50 of midday together with summer.
12. (piṉpaṉi)
piṉpaṉi tāṉu muritteṉa moḻipa.
29The late dewy season51 itself52 is also suitable, they say.
13. (mix of tiṇai-s; not with nilam)
tiṇaimayak kuṟutaluṅ kaṭinilai yilavē[.]
nilaṉoruṅku mayaṅkuta lilleṉa moḻipa
pulaṉaṉ kuṇarnta pulamai yōrē.
30Also to have the mix of settings is not a state to be excluded.
[There] is no mixing so that [different] regions come together,
say the erudite ones who have well understood the teachings (?).53
14. (no uri mix)
uripporu ḷallaṉa mayaṅkavum peṟumē.
31That those apart from the sense-units [related to] mood are mixed also occurs.54
15. (uri) | |
puṇartal pirita lirutta liraṅkal | |
ūṭa lavaṟṟi ṉimitta meṉṟivai | livaṟṟi (Nacc.) |
tēruṅ kālait tiṇaikkurip poruḷē. |
32Uniting, separating, waiting, grieving, quarrelling
− the occasions55 of these, when one examines them56,
[are] the sense-units [related to] mood for the settings.57
16.58 (pirivu) | |
iruvakai pirivu nilaipeṟat tōṉṟalum | tōṉṟiṉum (Nacc.) |
uriya tāku meṉmaṉār pulavar. |
33It59 is suitable60 to appear when the state of separation is obtained, which [is of] two kinds, say the scholars.61
17. (pirivu 1)
koṇṭutalaik kaḻitalum pirintava ṇiraṅkalum
kaḻiyiṉum…iraṅkiṉum (Nacc.)
uṇṭeṉa moḻipa vōriṭat tāṉa.
34Departing with [HER] and [HER] grieving there in being separated,
[this] is of one order, they say.62
18. (pirivu 1)
kalanta poḻutuṅ kāṭciyu maṉṉa.
351. Even at the time [the two lovers] are together, the sight63 [is] thus64 too.
2. Also times that are mixed65 [are of] the same sight.
3. Times and sights (?) that are mixed [are] thus [too].
19. (karu)
mutaleṉap paṭuva tāyiru vakaittē
teyva muṇāvē māmaram puṭpaṟai
ceyti yāḻiṉ pakutiyoṭu tokaii
avvakaip piṟavuṅ karuveṉa moḻipa.
36What belongs to the two kinds that have been called "basis" (mutal),
[that is, ] god66, food, animal, tree, bird, drum,
action/profession together with the type of melody (?)67,
these kinds and other [are called] "ingredients" (karu) they say.
20. (karu mix)
ennila maruṅkiṟ pūvum puḷḷum
annilam poḻutoṭu vārā vāyiṉum
vanta nilattiṉ payatta vākum.
37In whatever68 region flower and bird [appear],
even if they don´t come with the time of that region,
they become the fruit of the region where they have come.69
21. (karu people)
peyarum viṉaiyumeṉ ṟāyiru vakaiya
tiṇaitoṟu marīiya tiṇainilaip peyarē.
38The names [denoting] the staying in one setting, joined with each particular setting, are of two kinds, [namely] name and occupation.70
22. (karu people)
āyar vēṭṭuva rāṭūut tiṇaipeyar
āvayiṉ varūuṅ kiḻavaru muḷarē.
39Setting [-related] names of males [are] (for example?) āyar (herdsmen), vēṭṭuvar (hunters);
the designations for HIM are such as come in this place.71
23. (karu people) | |
ēṉōr maruṅkiṉu meṉṉuṅ kālai | pāṅkiṉum (Nacc.) |
āṉā vakaiya tiṇainilaip peyarē. |
40As for the others, when one considers it,
there are innumerable kinds of names [denoting] the staying in one setting.72

25. (karu people)
ēvaṉ marapi ṉēṉōru muriyar
ākiya nilaimai yavaru maṉṉar.
According to someone´s (?)76 usage, others are suitable too;
they are thus under the conditions that have arisen.77
26.78 (pirivu 2)
ōtal pakaiyē tūtivai pirivē.
41Reciting, enmity [and] message [are reasons for] separation.79
27. (pirivu 2)
avaṟṟuḷ
ōtalun tūtu muyarntōr mēṉa.
42Among these, reciting and message are for highborn people.
28. (pirivu 2) | |
tāṉē cēṟalun taṉṉoṭu civaṇi | civaṇiya (Nacc.) |
ēṉōr cēṟalum vēntaṉ mēṟṟē. |
43Going himself and others going in his company [is] with the king.80
29. (pirivu 2)
mēviya ciṟappi ṉēṉōr paṭimaiya
mullai mutalāc colliya muṟaiyāṟ
piḻaittatu piḻaiyā tākal vēṇṭiyum
iḻaitta voṇporuṇ muṭiyavum pirivē.
44Separation [is] in order to achieve the appointed bright wealth,
even if wanting that which is failing to be/become unfailing,
by the sorts that have been taught beginning with mullai as an example (?),81
for (?) others of wished-for superiority.82
30. (pirivu 2)
mēlōr muṟaimai nālvarkku murittē.
45For the four83 the right of the ones above84 is suitable.
31. (pirivu 2)
maṉṉar pāṅkiṟ piṉṉōr ākupa.
46Those who are [mentioned] subsequently belong to85 the chiefs.
32. (pirivu 2)
uyarntōrk kuriya vōtti ṉāṉa.
47They are suitable to the highborn ones according to the recitation. (?)86
in [the case of separation for] recitation.87
33. (pirivu 2) | |
vēntuviṉai yiyaṟkai vēnta ṉorīiya | vēntaṉi ṉorīiya |
(Nacc.) | |
ēṉōr maruṅkiṉu meytiṭa ṉuṭaittē. |
48The natural work of the king possesses room of occurrence also on the part of others except the king.
34. (pirivu 2)
poruḷvayiṟ piritalu mavarvayiṉu murittē.
49Also separating on account of wealth is suitable for those.
35. (pirivu 2)
uyarntōr poruḷvayi ṉoḻukat tāṉa.
50[Separating] on account of wealth [is] for highborn ones in accordance with conduct.88
36.89 (pirivu 2)
munnīr vaḻakka makaṭūuvō ṭillai.
511. [Going] by sea [is] not custom together with a woman. (Iḷ.)
2. The three kinds [of separation are] not custom together with a woman. (Nacc.)90
37.91 (maṭal)
ettiṇai maruṅkiṉu makaṭūu maṭaṉmēl
poṟpuṭai neṟimai yiṉmai yāṉa.
52There is no natural (?) literary usage (?)92 of [being] on the palmyra horse for a woman in any setting.
38. (naṟṟāy) | |
taṉṉu mavaṉu mavaḷuñ cuṭṭi | |
maṉṉu nimitta moḻipporuṭ ṭeyvam | |
naṉmai tīmai yaccañ cārtaleṉṟu | |
aṉṉa pirivu mavaṟṟoṭu tokaii | |
muṉṉiya kāla mūṉṟoṭu viḷakkit | mūṉṟuṭaṉ (Nacc.) |
tōḻi tēettuṅ kaṇṭōr pāṅkiṉum | |
pōkiya tiṟattu naṟṟāy pulampalum | |
ākiya kiḷaviyu mavvaḻi yuriya. |
53Those kinds of words that express being lonely/grieving93
are suitable to the real mother, at the occasion (?)94 when [they] have gone,
to the confidante and the onlookers,
pointing out herself and him and her,
saying to approach the fear of good [and] evil
in the god by the meaning of the words of a lasting occasion (?),95
together with those and others of that kind,
[and] making clear with three (?)96 the former time.97
39. (tāy)
ēmap pērūrc cēriyuñ curattum
tāmē cellun tāyaru muḷarē.
54And mothers98 are such as go themselves
through the streets of joyous/confused big villages and deserts.
40. (THEY?)
ayalō rāyiṉu makaṟci mēṟṟē.
Even if they [are]99 in the neighbourhood, [it is to be considered as] separation.100
41. (tōḻi) | |
talaivaru viḻuma nilaiyeṭut turaippiṉum | |
pōkkaṟ kaṇṇum viṭuttaṟ kaṇṇum | |
nīkkaliṉ vanta tammuṟu viḻumamum | |
vāymaiyum poymmaiyuṅ kaṇṭōr cuṭṭit | |
tāynilai nōkkit talaipeyarttu koḷiṉum | |
nōymikap perukit taṉṉeñcu kaluḻntōḷai | |
aḻintatu kaḷaiyeṉa moḻintatu kūṟi | kaḷaiiya voḻintatu (Nacc.)101 |
vaṉpuṟai neruṅki vantataṉ ṟirattōṭu | |
eṉṟivai yellā miyalpuṟa nāṭiṉ | |
oṉṟit tōṉṟun tōḻi mēṉa. |
551. When speaking of the state of excellence that will come to pass,
and 2. when letting [them] go, and 3. when sending [them] off,102
and 4. when recovering herself103, having looked at the state of the mother,
after onlookers had pointed out truth and lie
and the excellence of having them [back?], who had met with opposition (?),
together with 5. the elements of coming [and] approaching with encouragement
saying what is worded splendidly104 about breaking down,
to her, who has dimned in her heart since pain has become very great,
all these [and what]105 appears to agree with situations of the same nature
[are the occasions of speech that are] with the confidante.
42. (kaṇṭōr) | |
poḻutu māṟu muṭkuvarat tōṉṟi | |
vaḻuvi ṉākiya kuṟṟaṅ kāṭṭalum | |
ūratu cārppuñ cellun tēyamum | cārvuñ (Nacc.) |
ārva neñcamoṭu ceppiya vaḻiyiṉum | kiḷaviyum (Nacc.) |
puṇarntōr pāṅkiṟ puṇarnta neñcamoṭu | |
aḻintatetir kūṟi viṭuppiṉu māṅkat | |
tāynilai kaṇṭu taṭuppiṉum viṭuppiṉum | |
cēynilaik kakaṉṟōr celaviṉum varaviṉum | |
kaṇṭōr moḻital kaṇṭa teṉpa. |
56[When] showing106 the fault that has become a mistake (?)
since dread of the time and the way appears to come up (?),
and 2. in the situation of speaking (?) with affectionate heart
of the shelter of the village and the land [they] are going to,
and 3. when sending [them] on after having spoken against breaking down
with a heart which has become close to those who are united,
and 4. when stopping the mother and sending her on at the sight of [her] state,
and 5. at the going and coming of those who have departed to a distant place,
[in these situations] it is seen that the onlookers speak, they say.
43.107 (HE) | |
oṉṟāt tamariṉum paruvattuñ curattum | |
oṉṟiya moḻiyoṭu valippiṉum viṭuppiṉum | tōḻiyoṭu (Nacc.) |
iṭaicura maruṅki ṉavaṭama reytik | |
kaṭaikoṇtu peyartaliṟ kalaṅkaña reytik | |
kaṟpoṭu puṇarnta kauvai vuḷappaṭa | |
appāṟ paṭṭa vorutiṟat tāṉum | |
nālatu ciṉmaiyu miḷaimaiya tarumaiyun | |
tāḷāṇ pakkamun takutiya tamaitiyum | |
iṇmaiya tiḷivu muṭaimaiya tuyarcciyum | |
aṉpiṉa takalamu makaṟciya tarumaiyum | |
oṉṟap poruḷvayi ṉūkkiya pāliṉum | |
vāyiṉuṅ kaiyiṉum vakutta pakkamōṭu | |
ūtiyaṅ karutiya vorutiṟat tāṉum | |
pukaḻu māṉamu meṭuttuvaṟ puṟuttalun | |
tūtiṭai yiṭṭa vakaiyi ṉāṉum | |
ākit tōṉṟum pāṅkor pāṅkiṉum | |
mūṉṟaṉ pakutiyu maṇṭilat tarumaiyum | |
tōṉṟal cāṉṟa māṟṟōr meṉmaiyum | |
pācaṟaip pulampalu muṭinta kālattup | |
pākaṉōtu virumpiya viṉaittiṟa vakaiyiṉum | |
kāvaṟ pāṅki ṉāṅkōr pakkamum | |
parattaiyi ṉakaṟciyiṟ pirintōṭ kuṟuki | |
irattalun teḷittalu meṉaviru vakaiyōṭu | |
uraittiṟa nāṭṭaṅ kiḻavōṉ mēṉa. |
571. When considering with agreeing words his people
who are not agreeing and the season and the desert, and sending [them] off,108
2. when he leaves from accompanying [her]109
after her people have approached [them] in the midst of the desert,
and 3. with one constituent110 that comes to that situation (?)111,
including affliction when they have united with wedded love,
after disturbing grief approaches (?!),
and 4. in the situation(?) when he considers because of wealth, to which he is not
agreeing, the fewness of days and the rareness of youth
and the part of the... (?) male (?) and the appropriate calmness (?)
and the wretchedess of having nothing and the highness of possession
and the expanse of loving and the difficulty of separation,
and 5. with one constituent that [he] is intent on gain,
together with the part that is distributed by mouth and hand (?!)
and 6. encouraging, raising praise and honour,
and 7. in the situation that a messenger is intervening/that he is interposed as a messenger112,
and 8. at the side of the companion (s) who appear to have become (?)113
and 9. lamenting in the warcamp (?)
the part of three (?) and the difficulty of the circle
and the superiority of the enemies worthy for appearing
and 10. in the situation of opening114 a work, which has been longed for by the
charioteer in a time, when [it] is completed (?),
and 11. at the side of such ones close to protection (?!),
12. together with the two situations of begging and affirming
after having approached the one he had separated from in a separation on account of
another woman/a courtesan,
[in such situations] the chiefdom over the constituents of speech [is] with HIM.
44. (others)
eñci yōrkku meñcuta lilavē.
58Also for the remaining [people] they (these situations of speech) are not to be left behind.115
45. (speaking of the past)
nikaḻntatu niṉaittaṟ kētuvu mākum.
59It becomes/there is116 also a reason117 for thinking of what has happened.
46. (speaking of the past)
nikaḻntatu kūṟi nilaiyaluñ tiṇaiyē.
60When talking118 of what has happened the setting also has to stay. (?)119
47. (poruḷ? tiṇai?)
marapunilai tiriyā māṭciya vāki
viravum poruḷum viravu meṉpa.
61Mingling [of settings]120 and sense-units (porul) are mingled gloriously
without changing the state of word-usage121, they say.
48.122 (tiṇai: uḷḷuṟai +ēṉai)
uḷḷuṟai yuvama mēṉai yuvamameṉat
taḷḷā tākun tiṇaiyuṇar vakaiyē.
62Implicite simile and other similes,
these are without fail (?) situations to realize the setting.
49. (tiṇai + uḷḷuṟai)
uḷḷuṟai teyva moḻintatai nilameṉak
koḷḷu meṉpa kuṟiyaṟin tōrē.
63Implicite simile takes what speaks of/what excepts123 the deity as region,
say those who have understood the signs.
50. (uḷḷuṟai uvamam) | |
uḷḷuṟut titaṉō ṭottuporuṇ muṭikeṉa | |
uḷḷuṟut turaippatē yuḷḷurai yuvamam. | tiṟuvatai (Nacc.) |
64Implicite simile [is] what is said with implication,
65[with the thought] ’may the appropriate sense become complete124 along with this which has been implied’.
51. (eṉai uvamam)
ēṉai yuvaman tāṉuṇar vakaittē.
The other similes [are] of the kind to be realized by themselves.
52. (kaikkiḷai)125
kāmañ cālā viḷamai yōḷvayiṉ
ēmañ cālā viṭumpai yeyti
naṉmaiyun tīmaiyu meṉṟiru tiṟattāṉ
taṉṉoṭu mavaḷoṭum tarukkiya puṇarttuc
colletir peṟāaṉ colli yiṉpuṟal
pullit tōṉṟuṅ kaikkiḷaik kuṟippē.
66Approaching with pain which is not appropriate for pleasure
a youthful girl who is not appropriate for desire,
uniting what torments with himself and with her
in the two constituents of good and evil,126
having pleasure speaking as one who doesn´t obtain an answer,
in connection [with these] kaikkiḷai [is] intended to appear.
53. (peruntiṇai)
ēṟiya maṭaṟṟiṟa miḷamai tīrtiṟam
tēṟuta loḻinta kāmattu mikutiṟam
mikka kāmattu miṭaloṭu tokaii
ceppiya nāṉkum peruntiṇaik kuṟippē.
67The constituent of the palmyra horse mounted, the constituent of youth ending,
the constituent of desire, which has left behind believing, exceeding,
together with the strength of excessive desire,
as the said four peruntiṇai [is] intended.
54. (kaikkiḷai?)
muṉṉaiya nāṉku muṉṉataṟ keṉpa.
The four previous ones [are] for the preceding, they say.127
55. (kali + paripāṭu) | |
nāṭaka vaḻakkiṉu mulakiyal vaḻakkiṉum | |
pāṭal cāṉṟa pulaṉeṟi vaḻakkam | |
kaliyē paripāṭ tāyiru pāviṉum | pāṅkiṉum (Nacc.) |
uriyatāku meṉmaṉār pulavar. |
68In dramatic (?)128 usage and in wordly usage
the usage of the scholarly path worthy for singing (= poetry?)
is suitable to be in the warp of two, namely Kali and Paripāṭu,
say the scholars.
56. (proper names in Akam)
makka ṇutaliya vakaṉain tiṇaiyum
cuṭṭi yoruvarp peyarkoḷp peṟāar.
When human beings are denoted, with respect to the five settings of Akam129 individuals don´t obtain getting names.
57. (aḷavutal? in Akam/Puṟam)
puṟattiṇai maruṅkiṟ porunti ṉallatu
akattiṇai maruṅki ṉaḷavuta lilavē.
69Besides being fit in the Puṟam settings,
70blending (?)130 is not [allowed] in the Akam settings.
Bibliographie
Literature
Primary Sources
Akapporuḷ Viḷakkam (+ Nampís urai): Ed. by Gōvintacāmi Piḷḷai, 1962.
Iṟaiyaṉār Akapporuḷ (+ Nakkīraṉ́s urai): Ed. by SISSWPS, Tinnevelly, Rep. 1964.
Tolkāppiyam Poruḷatikāram. Ed. + Transl. + Comm. by P.S. Subrahmanya Sastri, Part I Madras 1949, Part II Madras 1952, Part III Madras 1956.
Tolkāppiyam Poruḷatikāram + Iḷampūraṇaŕs urai. Ed. by SISSWPS, Tinnevelly, Rep. 1967.
Tolkāppiyam Poruḷatikāram 1-5 + Nacciṉārkkiṉiyaŕs urai. Ed. by SISSWPS, Tinnevelly, Rep. 1968.
Tolkāppiyam Poruḷatikāram 6-9 + Pērāciriyaŕs urai. Ed. by SISSWPS, Tinnevelly, Rep. 1966.
Secondary Sources
Aravamuthan, T.G. 1930: "The Oldest Account of the Tamil Academies". JOR IV (1930) 183- 201, 289-317.
Buck D./Paramasivam K. 1997: The Study of Stolen Love: A Translation of Kaḷaviyal eṉṟa Iṟaiyaṉār Akapporuḷ with Commentary by Nakkīraṉ. Scholars Press of Atlanta, Georgia 1997.
Ilakkuvanar, S. 1963: Tholkāppiyam (in English) with Critical Studies. Kuṟaḷ ṉeṟi Publishing House, Madurai 1963.
Klaus, C. 2000: "Zu den Śrautasūtras." In: Indoarisch, Iranisch und die Indogermanistik. Arbeitstagung der Indogermanischen Gesellschaft vom 2. bis 5. Oktober 1997 in Erlangen. Hrsg. Von B. Forssman und R. Plath. Wiesbaden 2000, 177-190.
— — 2001: "Zur Entstehung des Pātimokkhasutta der Theravādin." WZKS 45 (2001) 23-39.
Renou, L. 1963: "Sur le genre du sūtra dans la littérature sanskrite." JA 251 (1963) 165-216.
Selvamony, N. 1989: Tolkaapiyam: akattiNai iyal. Sobitham 1989.
Srinivasan, S.A. 1980: On the Composition of the Nāṭyaśāstra. StII Monographie I, Reinbek.
Staal, F. 1992: "Sūtra." In: Kalātattvakośa. A Lexicon of the Fundamental Concepts of the Indian Arts. Vol. II: Concepts of Space and Time. Delhi 1992, p. 303-314.
Takahashi, T. 1991, "The List Collating the Sūtra Numbers among the Various Texts of the Tolkāppiyam" In: The Memories of the Institute of Oriental Culture, 65-119 [in Japanese]
———— 1995: Tamil Love Poetry and Poetics. E.J. Brill, Leiden 1995.
———— 2002: "Before Grammar: Issues on Reading Some Classical Tamil Texts." (Lecture given at the 17th European Conference on Modern South Asian Studies), Kolam <www.unikoeln.de/phil-fak/indologie/kolam>, Heidelberg, 13th Sept. 2002.
von Hinüber, O. 2001: "Nochmals über das Pātimokkhasutta. Anmerkungen zu K. Klaus: ´Zur Entstehung des Pātimokkhasutta der Theravādin´." WZKS 45 (2001) 41-58.
Wezler, A. 2001: "Zu der Frage des ´Strebens nach äuẞerster Kürze´ in den Śrautasūtras." ZDMG 151 (2001) 351-366.
Wilden, E. 2000: "The Position of the tuṟai-s in the Poetological Tradition of Caṅkam Literature". In: Pandanus 2000, Workshop on Natural Symbolism in Indian Literatures, ed. J. Vacek, Prague, 263-283.
———— 2002a: "Towards an Internal Chronology of Old Tamil Caṅkam Literature Or How to Trace the Laws of a Poetic Universe (A Review of H. Tieken´s Kāvya in South India)." WZKS 46 (2002) 105-133.
———— 2002b: "Announcement of a Critical Edition of the Naṟṟiṇai." (Lecture given at the 17th European Conference on Modern South Asian Studies), Kolam <www.unikoeln.de/phil-fak/indologie/kolam>, Heidelberg, 13th Sept. 2002.
———— 2003 (forthcoming): Literary Techniques in Old Tamil Caṅkam Poetry: the Kuṟuntokai; ch. III. Studies in the Poetological and the Poetical Tradition, p. 125-238.
Zvelebil, K. 1973: "The Earliest Account of the Tamil Academies". IIJ 15.2 (1973) 109-135.
Notes de bas de page
1 For reading and discussing the issues of this paper I want to thank Dominic Goodall, Jean-Luc Chevillard and Sascha Ebeling.
2 TP 472: "A sūtra itself is created to appear in verse, in order to explain well the meaning, without examination, appearing to be known like the image in a mirror."
3 For the general question of dating Caṅkam texts see Wilden 2002a, for a summary of the discussion concerning the Tolkāppiyam see Takahashi 1995: 16f.
4 On the stock of phrases common to the kiḷavi-s, the TP and the IA see Wilden 2000 and Wilden 2003 (forthcoming).
5 The relation between IA and Nakkīraṉ on the one hand and that between the TP commentators and Nakkīraṉ on the other is still in need of detailed investigation. The dating of Nakkīraṉ´s commentary is rather vague and unconvincing. Zvelebil 1972: 121f. proceeds from external data: Nakkīraṉ´s not being referred to by Iḷ., but by Pērāciriyar (commentator on TP 6-9), which would point to a date between these two. This would mean some time within the 12th century. Things are complicated by a discussion about a lost commentary on the IA written by Iḷ., but if Iḷ. (provided it was the same man and not just someone with the same name) in his TP commentary doesn´t mention Nakkīraṉ, it might either mean that he didn´t have access to the earlier commentary, that he deliberately was silent about it, or that it simply wasn´t earlier, but later. For unclear reasons, however, Zvelebil doesn´t argue along these lines, but instead takes Nakkīraṉ´s quotations from the Pāṇṭikkōvai as a decisive hint for dating. Now certainly, if the Kōvai refers to a Pāṇṭiya king of the 8th century, this means that the IA commentary cannot have been written before that period (as is pointed out already by Aravamuthan 1930: 296), but there is no prima facie reason that it should have been contemporary, as is silently assumed by Zvelebil. As for style, diction and morphology, Nakkīraṉ rather seems to belong into the vicinity of the medieval commentators, even if he isn´t quite as prolix as Nacc.
6 A cogent case in point might be the kiḷavi-s. A first tentative search for manuscripts in preparation of a critical edition of one of the lyrical anthologies, the Naṟṟiṇai, has brought to light that in different manuscripts there may be different kiḷavi-s. If one considers the matter with all due caution, the obvious conclusion is that the relation between a poem and its kiḷavi cannot have been looked upon as indissoluble: people must have set out to write new kiḷavi-s, and it is, for the time being, impossible to ascertain when this practice stopped. Moreover, the phrasing of certain kiḷavi-s shows affinities to the medieval commentary idioms, which certainly gives no upper limit, though perhaps a lower limit for their origin. (See Wilden 2002b)
7 7One calamity not to be underestimated in this context is the damage happening to manuscripts.
8 A collection of the remaining witnesses (presumably desintegrating quickly) and the preparation of critical editions is as yet a desideratum not only in Caṅkam philology, but in Classical Tamil philology in general.
9 Or perhaps one should say that, rather like with the Śloka in Sanskrit technical writing, this metre creates the impression of a kind of rhythmical prose.
10 Whether this rule of thumb can have any claim to being compulsory is open to discussion. Among the commentators at least Iḷ. occasionally segments 3 metrical feet as a sūtra. In the cases to be found in the TP text to be analysed below, however, Nacc. decides differently, and he has reasons of syntax to do so. In any case the relation of metre and syntax is an open question in Caṅkam language in general.
11 Note that the term chosen by tradition is sūtra and not kārikā, which can be seen as a hint at least that the text of the TP was not conceived of as poetry but as versified instruction.
12 Note that the Tamil word nūl, the usual word for "thread", and thus expected to be an equivalent for sūtra, is explained in the preface of Nakkīraṉ´s commentary on the IA − which is one of the earliest available discussions on the matter − to denote a treatise, the smallest unit of which is termed cūttiram (see also note 17 for details). There are also two portions in the TP, (Ceyyuḷ-iyal = TP 468ff. and Marapu-iyal = TP 642ff.) slightly different in terminology, but agreeing as far as nūl and cūttiram are concerned, which define the types of treatises and their elements. Both are obviously late and not of the same hand, and both are in need of special investigation.
13 From the text of the TP itself it is impossible to discern whether its sūtra-s were already named that way at the time of their composition. And since manuscripts are so young, even the word usage of colophons probably wouldn´t be of great help. Once again the only relevant text between the TP and its commentators might be Nakkīraṉ´s commentary on the IA plus the even later preface to that commentary, both of which also employ cūttiram. Another minor chance might lie in the study of the word usage of the few old and anonymous commentaries preserved for the anthologies, that is, those on the Aiṅkuṟunūṟu and Patiṟṟuppattu plus the two partial ones on the Puṟanāṉūṟu and the Akanāṉūṟu.
14 See Takahashi 1995: 23f.; see also Wilden 2000 and 2003 (forthcoming).
15 Remarkable is TP 89 tuṟaiyamai nalyāḻt tuṇaimaiyōr instead of kantaruvam in the parallel in IA 1.
16 One possible channel of stylistic inspiration, for example, might have been Sanskrit grammar. Though the grammatical parts of the Tolkāppiyam clearly show an independent approach to describing a language most definitely not comparable to Sanskrit in many respects, some of the categories of analysis betray a familiarity with the Pāṇinian (?) system, as the section on compounds, a phenomenon hard to detect in Tamil (a task dutifully taken up by the commentators). Whether the reception of, say, the Aṣṭādhyāyi has shaped in any way the formulation of the Tolkāppiyam grammatical aphorisms, and whether these, in their turn, have been the model for the slightly later (?) TP aphorisms, for the moment has to remain a matter of mere speculation. An influence of Sanskrit poetics, on the other hand, is tangible only at a rather late stage of textual formation and thus probably cannot have shaped the formal outline.
17 The topic has been treated at some length by Renou 1963. See also Staal 1992; for a recent renewed discussion see Klaus 2000, 2001, Wezler 2001, von Hinüber 2001. An interesting side-issue is that the distinction between sūtra as a rule in a theoretical treatise and Sūtra as a treatise made up of sūtra-s (graphical denotation introduced by Klaus 2000) doesn´t seem to be prominent in Tamil. At least as far as poetics is concerned, there is no work bearing a title ending in -cūttiram. This is well in keeping with the word usage as expounded in the preface to Nakkīraṉ´s commentary on the IA (cf. IA p. 12). There a treatise is termed nūl, its sections atikāram, the subsections ōttu, and the smallest units cūttiram. Such a view on the matter is corroborated in the case of grammatical literature in the title Naṉṉūl (i.e. nal nūl), and here a Tamil transposition of the Sanskrit titles ending in -sūtra might be considered. It should, however, be noted that this preface (cf. IA p. 14) gives a gloss of Tamil nūl with Sanskrit tantra (tantiram).
18 I won´t take into consideration the frequently repeated idea that a sūtra generally is prone to be so concise as not to be comprehensible without proper exposition and explanation by a commentary (for a recent example see Buck/Paramasivam 1997: Xf.). Should such a commentary (be it in oral or written form) have been in existence in the case of the TP, there is no positive indication that it was available to Iḷ. and Nacc., both of whom have all too many obvious problems in rendering the sūtra wording intelligible.
19 The degree of conciseness is, to be sure, a variable notion, but the point might be the wish to avoid unnecessary redundancy, an endeavour that can range from the mere avoidance of repetitions in subsequent sūtra-s to actual compression.
20 There are perhaps sūtra types as many as there are Sūtras − cf. Renou ib. p. 181ff. −, but a lack of variety is notable in the constructions to be found in any given Sūtra.
21 My numbering follows the edition cum translation of Sastri 1949f. Further available (minimally annotated) translations are Ilakkuvanar 1963 and Selvamony 1989.
22 It might be interesting to see how those manuscripts that are not accompanied by a commentary split the texts − if there are any (at least for the IA I have seen one).
23 For a tabular concordance of Tolkāppiyam sūtra divisions according to the different editions and commentators see Takahashi 1991.
24 The distribution of these authorizing statements as well as their precise impact are a great puzzle as yet. All that can be said with confidence is that they refer to some kind of past or present expertise.
25 On this problem of disrupted sequence betrayed by the use of pronouns see Takahashi 2002.
26 One of my general questions was whether the notion of a convincing macrostructure, that is, of a coherent concept in the text, might be used as an argument when trying to solve textual difficulties. The more I read, the stronger my feeling that the idea of coherence is highly relevant when weighing the probabilities of possible translations for sūtra-s which lack a convincing traditional interpretation and/or don´t fit in well with their context. And I cannot see this as the hybris of "Western" interpretation − rather it is a kind of respect due to the authors: what they thought and wrote is supposed to make sense.
27 The following commentary on structure and status of the single sūtra-s and the stages of their development will be somewhat condensed in order to show the broad outline. For fuller argument and philological details see the footnotes to the translation in the appendix.
28 One can speculate here whether the notion of its being the middle one among the 5 tiṇai-s might have given rise to the analogous imbedding of the 5 tiṇai-s between the two younger settings kaikkiḷai and peruntiṇai.
29 Noticeable is that the first extensive insertion which doesn´t make a contribution to systematic concerns is to be found at the end of the first sub-unit, that is, at a clear break in the overall structure.
30 For thematic traces of such a trend in the poems and kiḷavi-s of the Kuṟuntokai see Wilden 2003: 198ff. (forthcoming).
31 On the growth of the number of speakers in the TP see Wilden 2003: 135f. (forthcoming).
32 Apart from its conspicuous position (preceding the actual exposition of the tiṇai-s) and the fact that the old part of the poetic corpus makes use of only five tiṇai-s if at all, this sūtra exhibits no clear stylistic or linguistic trace of being a later interpolation. What arouses suspicion, however, is part of the content: muṟpaṭa kiḷanta "clearly expressed before" − what is this supposed to mean in a first sūtra? This recalls strangely the beginning of the insertion in TP Ceyyuḷ-iyal (TP 488) and the first sūtra of the IA. Now, even if we are willing to consider a change of position, there is no place in the TP where this reference could be taken to refer to something already taught. This means, it probably has to be seen in connection with other frequent kinds of reference to former authority such as eṉmaṉār pulavar. So, one sad but not too unlikely possibility of explaining it is that the whole tiṇai section of the TP is quite a late composition/redaction of partly older/heterogenous materials (not before the time of the Kalittokai): the tiṇai exposition referred to in TP 488 and IA 1 may no longer exist, though it must have been one of the sources, since there are phrasal correspondences. If what follows here were the original tiṇai section, it wouldn´t make sense to talk of the five middle tiṇai-s, in other words the first sūtra has not simply been prefixed to an existing sequence.
33 mutal-ā: this mere verbal root ā is analysed by Iḷ. and Nacc. unanimously as an infinitive (glossed by āka). One would like to have parallels. Another possibility might be to take it as an absolutive to the verb mutalu-tal.
34 Syntax is not very satisfactory in this sentence: if the first line is supposed to be the predicate noun for the second, it is at least totally unmarked as such. A little awkward is moreover the iṟuvāy, which probably has to be analysed simply as a noun meaning "end" (-vāy being in this case a suffix which doesn´t change the semantics of the root iṟu). Then we have to take peruntiṇaiyiṟuvāy as a veritable bahuvrīhi: "that whose end is peruntiṇai" − unless we want to understand something deviating considerably from the traditional view, namely that the seven tiṇai-s are the end (in the sense of the complete row) of the great tiṇai-s beginning with kaikkiḷai. This solution would strangely coincide with TP 488, where we have kaikkiḷai mutalā vēḻ perun tiṇaiyum, "the seven big settings beginning with k.".
35 poruḷ as a technical term is very difficult. In this threefold distinction it might mean something like "level of sense", "kind of sense-conveying unit".
36 The technical sense of mutal might be here something like "fundamental element" of place and time (see next sūtra).
37 karu as a technical term refers to the regional elements or implements.
38 There doesn´t seem to be a satisfactory etymological analysis for uri. Has it to be connected with uri-tal "to peel off" (i.e. "bark" in the sense of emotive outer covering)?
39 nuvalum kālai muṟai: is this equivalent to colliya muṟaiyāl/ṉ (s. 5)? The glosses are different, but if this refers to poruḷ utterances uttered in a poem, what could it mean? The element named first is the decisive one? Or does the clause simply refer to the order of utterance here in the sūtra (thus the traditional interpretation)? Since kālai is regularly used to form actual temporal clauses (cf. nāṭuṅ kālai in the next line), this solution doesn´t seem very satisfactory. This is a case where one might suspect also textual corruption.
40 mēya might be considered here to be a mere connective particle as in the medieval language (suggestion by Sascha Ebeling), which would be a linguistic indication of lateness. In this case it would have to be explained as a contraction of mēviya (of the verb mēvu-tal "to be dear to") "which is dear to" − "to belong to, to be connected with" (such the gloss given by Nacc.). Another possibilty to consider is, however, an infinitive of the root mey-tal, in poetry usually "to graze", but also meaning "to roam ": "the world of the forest residence, where Māyōṉ roams, …"
41 maivarai: mai "collyrium" is used in the lyrics for example of the KT several times as a metaphor for "cloud", so most probably this means "clouded mountain", that is, mountains so high as to touch the sky (see Nacc.´s gloss: vāṉ taṅkiya varai).
42 At least Varuṇaṉ is suspicious in this enumeration of tiṇai deities. But again there is no conclusive stylistic/linguistic reason to suppose an interpolation, except for the length, and the sūtra as such is necessary in this place. Here a complement within a given sūtra seems to be a plausible explanation.
43 What is the significance of the -um with collavum?
44 Here, as well as in 9, the sūtra division according to Iḷ. doesn´t respect metrical boundaries. Nacc. takes 6+7 as well as 9+10 together.
45 If we take, against the traditional interpretation, sūtra-s 6 + 7 together and read 8 as an additional/exceptional rule to both, we get a description much more in tune with the actual poetry: the paṉi poems are at least as often situated in the forest region as in the mountains.
46 Why vaikaṟai plus viṭiyal, since both refer to daybreak? Both Il. and Nacc. seem to see a difference, since both of them coordinate the two by -um. T.V. Gopal Iyer (oral comm.) explains the former as applying to 2-6 in the morning, the latter as 6-10.
47 ātal meypeṟa tōṉṟum: does this elaborate predicate signify anything specifically (which would distinguish neytal and marutam from the other tiṇai-s), or is it just a rarer specimen of the predicative idioms connected with former authority?
48 neṟi is probably not just another word for "way" such as āṟu and vaḻi, but the "traditional path" (thus the interpretation of both commentators). Then muṉṉiya neṟi would be "the traditional path, which is [actually] followed".
49 maruṅkiṉ: this locative is difficult to construe. It might be possible to read it not as a mere locative suffix, but take maruṅku as an equivalent to nilam: "the setting situated in the middle has the tradition [of being situated] in a region that is the result of midday together with summer."
50 muṉivu-nilai, literally "end-state", is best taken as a compound meaning "result". The sūtra no doubt refers to the specific pālai region, the point being that the setting of pālai is to be placed in any region in the state of extremity caused by summer drought. (Nacc. seems to understand it like this, but he restricts the occurrence to kuṟiñci and mullai regions.)
51 What might piṉpaṉi be from the point of view of the poetry? And suitable for what? For the middle setting? Does this make sense?
52 tāṉ might be here not much more than an emphatic particle, used to make up a complete metrical foot. Or it might emphasize something special, piṉpaṉi, whatever might be the relevance.
53 This injunction seems to make sense: it is easier to imagine, say, a seashore region in times other than afternoon, and animals (or even plants) straying into other areas, than a mountain on the seashore, though in the poety the division is less distinct than in theory − there are extensive descriptions of forests in mountain poems, for example.
54 What is the relation to the previous sūtra? There it is taught that tiṇai-s may be mixed, except for the basic regions, i.e. presumably with respect to time and specific elements (karu). Is this one to be read as a special injunction to forbid the mix of uri-s (the mix of uri-s with uri-s, i.e. ambivalent moods? the combination of uri-s with different regions?)? Apart from making no sense at all from the point of view of the poetry, it is inconsistent in itself. Neither before nor in what follows the correlation of one uri with one tiṇai is taught − quite contrary to the modern vulgata version of Tamil poetics.
55 nimittam is, apart from varuṇaṉ and an occasional kālai, the first clear Sanskritism in this section of which there are quite few compared with the other "old" parts of the TP.
56 Is ivai to be connected with tēruṅ kālai (slightly unusual with the idiom) or is it the direct predicate noun to urip poruḷ, which is further qualified by avaṟṟiṉ nimittam eṉṟu: "the uri-poruḷ-s [are] these, or rather the occasions of these, [namely] uniting etc. "?
57 This sūtra makes it absolutely clear that there is not one uri related to one tiṇai. There is no correlation here, and if we go back to the sūtra-s relating settings with place and time (6ff.), we find two enumerations identical as to sequence (mullai − kuṟiñci − marutam − neytal − naṭuvaṇ) which doesn´t conform with the correlation established by the commentators (puṇartal − kuṟiñci, pirital − pālai, iruttal − mullai, iraṅkal − neytal, ūṭal − marutam).
58 Sastri places the pirivu sūtra here after the uṟi sūtra, but according to Iḷ. as well as Nacc. it precedes the sūtra-s on tiṇai mix and follows directly the mutal sūtra-s (i.e. Iḷ. 13, Nacc. 11). Strangely enough he doesn´t at all comment on this fact. Since the subject is elliptical this position would result in a completely different statement, namely to the point that mutal application (or even making tiṇai poetry) is appropriate in the case of separation. There are two ways of explaining: either the pirivu sūtra has been misplaced by the commentators (or in a part of the subsequent transmission, which might even mean the modern editors of the commentary editions), because there is a clear sequence between 16 and 17 as above. Or the sūtra-s in between have been interpolated, but this would imply several problems. While it might well be possible that the tiṇai mix is an addition here (though there is no stylistic indication that would make this plausible), the same is not true for the uṟi sūtra, which is indispensable. This is one of the cases where one would desperately want to have a look at the manuscripts.
59 What is suitable for what? Iḷ. reads two sentences: there are two kinds of separation, and both are suitable for pālai. Nacc. reads a sub-sūtra to the preceding one: both kinds of separation are suitable during the after-dews. But given the basic character of this differentiation of pirivu, doesn´t it make more sense to see this as a further characterisation of mutal? The differentiation of the world (spacial and temporal as given in the preceding sūtra-s) lyrically takes place in the situation of separation, which is of two kinds.
60 uriyatu, n.sg., can, when this positioning of the sūtra is tenable after all, only refer to uri-poruḷ as a plurale tantum.
61 The impact of this sūtra may be twofold. On the one hand it might be read as a subsequent exposition of one of the uri-s, namely pirital (followed rather eclectically by iraṅkal, but not by the others). On the other hand it is possible to separate pirital and pirivu and see pirivu as a higher category: in the general state of separation (pirivu) the uri-s will manifest themselves (as a basic thematic or modal subdivision).
62 This sūtra might be taken as the basic definition of pirivu, which has two kinds (iruvakai), namely HIS going away either with or without HER. The ōriṭattu in this case wouldn´t refer to these two being the first vakai, but it might emphasize their falling under the same category despite basic differences. Structurally this is the only difference of interest between the kinds of separation. True, from the thematic point of view actually going away together is just one fairly rare special case, but it is quite important with respect to the protagonists − the poems where the mother ponders about their whereabouts, which are very prominent in the AN, and in the speaker sūtra-s of this section the mother gets the first place.
63 kāṭci is difficult here. It might either be used as a technical term − something like "on poḻutu mix one takes the same view", or it might have thematic implication: "THEIR actually seeing each other (which implies a mix of poetic times?) is still of the same pirivu order, i.e. not to be seen as a deviation from the rule.
64 The most likely interpretation of aṉṉa here is that it refers back to the predicate noun of the previous sūtra: oṟiṭattu, i.e. it still belongs under pirivu.
65 By the commentators kalanta poḻutum is taken to refer to THEIR being together, but in the light of the sūtra-s on tiṇai mix it might be preferrable to read it as a synonym to maya-ttal and understand that the mix of poḻutu-s (poetic times, besides karu the only elements that may be mixed) doesn´t imply change of the basic feature of pirivu.
66 teyvam, here another Sanskritism.
67 What does yāḻiṉ pakuti denote? Do we have to think of a kind of musical version of the poems? How old is this?
68 Here the interrogative base e- apparently is used in an indefinite sense. This is strange, but how else to take it?
69 So this sūtra seems to regulate tiṇai mix with respect to karu-s. But why only flowers and birds? Is this a "realistic" rule? − of both of these there are so many, and they are indeed not clearly correlated to particular regions.
70 This is an important sūtra since it throws light on our general problem of generic names. Is paratavar the name of a tribe of fishermen, or is it the word for fishermen (who might have been organised in a tribal way, or even as a caste)? The TP doesn´t offer a solution, but sees already the same phenomenon.
71 āyar is not a frequent term; it appears just once in the old anthologies (PN 390.1) and lots of times in Kal. vēṭṭuvar is there occasionally, but also not frequent, and both not for HIM. Why these choice of examples and not some really common ones? Perhaps it would be better to separate into two sūtra-s, one referring to the tribe/occupation names and one drawing the parallel to the nāṭaṉ type, the latter without examples because it is self-evident.
72 What is the impact of this sūtra? Is this not a restriction, as usual, but on the contrary, an extension, i.e. there are not only family- and occupation names? Then it should go together with the second part of the previous sūtra, because the designations for HIM are structurally of the same type as the others, but they are mostly based on residence (nāṭaṉ "the man from a land", cērpaṉ "the man from the coast") or even more complicated (makiḻnaṉ "the one who gives joy "?). A few of the other regional names are also based on residence (kāṉavar "forest inhabitants").
73 pāṅkiṉ is not to be counted among the numerous regular locative suffixes, but it seems hardly possible to see anything else here.
74 The exact meaning of viṉaivalar is doubtful. Etymologically it could mean either "those who are able to work" or "those working under instruction "; in any case in combination with aṭiyōr "slaves" it seems to be a social classification.
75 The reading of puṟattu as a second predicate noun is grammatically problematic, since we expect a neuter plural (puṟatta), as would parallel ila. It might, however, be possible to postulate the special sandhi with following eṉṉutal (puṟatta-eṉmaṉār). (The commentary explanation that this refers to there being outside aintiṇai, but allowed for peruntiṇai and kaikkilai, is not supported by the text, but systematically possible.) Alternatively it would be possible to read an extra injunction for Puṟam poetry ("they are not to be excluded in Puṟam"), but this is problematic in the context. There has been no mention of the division into Akam and Puṟam, and the description so far clearly pertains to Akam alone.
76 ēvaṉ, actually interrogative pronoun, is problematic here, but it seems once again best to take it as an indefinite: there are also people (poets? poeticians?) who deal with yet more kinds of person. This seems to be a concession to actual occurrence. On the other hand both commentators readēval, a verbal noun here. This would be a usage which is imperative. This doesn´t make much sense unless read as a late insertion.
77 ākiya nilaimai might refer to the state of affairs as it is to be found in the actual poetry of the day. So this once again might be read as an extensional sūtra making concessions to actual usage (as against the normative usage). The persons of karu thus take a whole 5 sūtra-s, which might mean that on this point there has been a lively controversy.
78 The following renewed discussion of pirivu is peculiar in several respects. Firstly, it is disproportionately long, comprising sūtra-s 26-36. Secondly, from the structural point of view it is of no avail at all, and thus at discrepancy with what has been taught so far. It is a "sociclogical" analysis of the phenomenon of pirivu, enumerating different sets of reasons for seperation and different kinds of people for whom these reasons may be valid. Nothing of this kind has any repercussions with either poems or kiḷavi-s. We know, however, that this must have been an old concern, because a similar discussion (with the only phrasal correspondences to be found in this section) is to be found in the IA, though considerably more concise and comprehensible (= IA 35- 40).
79 ōtal, pakai, tūtu: these three are traditionally interpreted as study, warfare and diplomatic errand.
80 This sūtra might make sense if taken elliptical. While the uyarntōr separate for studying or delivering a diplomatic message, the king goes to make war (pakai "enmity"), the third element of the basic enumeration of s. 26.
81 patimaiyam = Skt. pratimā-?
82 This sūtra is an enigma and its different, widely disparate interpretations would make a whole paper. Neither wording, nor context nor correspondences of any kind give a clue to what might have been said in the first three lines, while the fourth adduces wealth as a reason for separation, indeed the only one so far to have a clear correspondence to kiḷavi-s and poetry.
83 nālvar is interpreted unanimously as a reference to the four Sanskrit varṇa-s. If this is correct (and I don´t see any way of explaining it otherwise on the basis of the immediate context) it is prone to make the sūtra an alien element in the context (see also the next note).
84 mēlōr: if this is to be seen as a synonym to the usual uyarntōr, it might be a first stylistic hint for a difference of origin, which in itself wouldn´t carry much weight, but in combination with the nālvar it might make another point. It might however, also be interpreted as textual reference (complementary to the piṉṉōr "those mentioned subsequently" of the next sūtra): "the ones mentioned above", and then the sūtra would have to be seen as a rather democratic extensional rule: not only highborn ones and kings are allowed to separate for study, enmity, message and wealth, but all the four varṇa-s, but this wouldn´t make much sense in the light of what follows, namely further complicated rules of social distinction.
85 Here a second rather clear instance of pāṅkiṉ as a locative suffix.
86 At the very latest this sūtra makes clear that either textual or transmissional order are awry. What is suitable, and in the plural (uriya), is elliptical and not to be discerned from the preceding sūtra-s. And the meaning of ōttu here is not at all clear. To read it as a reference to the Veda as is done by Nacc. lacks any kind of systemic or contextual plausibility, if it is not to be read as a sign of moralistic interpolation (cf. the brahmins intruding in Kaḷavu, Kaṟpu and Ceyyuḷ insertion).
87 Thus Iḷ., but that doesn´t make sense since this is already taught in sūtra 27.
88 What is the meaning of oḻukattu here? The commentators take it to refer to caste dharma (for which there is no evidence in the text), unless we assume it belongs to the same set of interpolations as the nālvar sūtra. It might, however, also be a technical term referring to suitability, but for this one would want parallels.
89 This sūtra, whatever the factual meaning, is a special rule only loosely fitting the context and lacking the ordinary sūtra to which it might be the restriction − unless it is totally misplaced here.
90 The crux here is the expression munnīr, taken by Iḷ. as a word occasionally to be found in the poetry (especially in PN) and denoting the sea (as having three kinds of water). This doesn´t make any sense, because travelling by sea has never been mentioned nor is it a lyrical topos. Nacc. takes it to refer to three kinds of pirivu, which might make sense if we put the sūtra into the vicinity of s. 26 (i.e. ōtal, pakai, tūtu). He sorts out, however, ōtal, tūtu and poruḷ, which would be better from a systematic and thematic point of view (war being the prerogative of the king and thus a special case). This solution, however, poses a major morphological problem: how to account for the formation of munnīr?
91 This sūtra is with absolute certainty either misplaced or badly interpolated. Its only contextual defence is that it also deals with women (making use of the same unusual word makaṭūu as the preceding s.).
92 poṟpuṭai neṟimai iṉmai: while certainly being a technical predication expressing a prohibition, it is not one of the familiar ones. Neither precise impact nor origin are traceable.
93 pulampalum ākiya kiḷavi: this presumably refers to the situation of the mother being left behind when THEY elope together, but how to explain the construction? Is this simply a pointed way of attribution?
94 What can be the meaning of tiṟattu here? Both commentators decide to take the pōkiya as an elliptical reference to THEIR going away together, and indeed this information seems necessary for the sūtra, since the mother speaks only then. This means, however, to take tiṟam, "constituent, element", as an equivalent of kālam, "time" − thus the gloss given by Nacc..
95 Whatever is this supposed to mean?
96 Does this refer to the persons mentioned in line 1, i.e. the mother herself, HIM and HER?
97 Here is one case of tokaii not concluding an enumeration, an agglutination of clauses which looks quite anacolouthically.
98 Note the change of idiom from naṟṟāy to tāyar, either a (generalising) plural or the usual honorific. Probably this sūtra is a later extension.
99 The ayalōr are not identified, but probably this refers rather to the eloped couple than to the mother (s), as the commentators seem to imply by glossing it with pirintār.
100 Note the unusual word for separation employed here, akaṟci, not pirivu.
101 "talking about breaking down [and] dying in order to remove [the pain] "?
102 Pōkkal appears to be terminological here, i.e. the confidante advising the lovers to take resort to going away together, while viṭuttal might refer to the confidante´s actually seeing them off.
103 Koḷiṉ: here koḷ once more in its modern auxiliary function.
104 kaḷaiyeṉa adverbially modifying moḻintatu?
105 Syntax is as so often awkward here, and actually the tōṉṟum hangs syntactically on thin air.
106 Either this sūtra is badly anacolouthic and the kāṭṭalum has to be read as coordinate with the iṉum-forms or I don´t understand the structure at all.
107 At the very latest this sūtra is quite as bad as the speaker sūtra-s in the Kaḷavu and Kaṟpu sections: as heavily reworked, as unsystematic, as anacolouthic. Definitely not an earlier and simpler state of affairs. But then, how to explain their interrelation at all? Do we have to assume a state of affairs when the whole TP text was compilated, but additions could be made wherever they were felt necessary?
108 viṭuppiṉum has either to be read as a very elliptical construction (as do the commentators) or we do have to assume that the subject here is not HE himself, but perhaps the confidante who first ponders with him cicumstances and occasion and then sees the eloping couple off (cf. previous s.).
109 If peyartaliṉ is the next item on the list, it is not coordinated by -um, and it is unclear how it is to be construed with kaṭaikkoṇṭu.
110 How to explain the function of orutiṟattāṉum? Has it to be read as elliptical: "in [situations that also have this] one constituent "?
111 Is pāl to be understood here and in the following as "fate" in the sense of actual momentary situation of the lovers?
112 The first reading might be the one chance of finding in the TP something similar to the messenger theme occasionally to be found in the kiḷavi-s and frequent in the lyrics, but the second seems to be in accordance with sūtra 26.
113 āki tōṉṟum seems to be part of an incomplete clause, or at least I have no idea how to understand it all by itself.
114 Can tiṟa here have the meaning of "beginning" or does it say that HE discloses his decision to set off in order to accomplish work?
115 eñciyōr: this is hopefully just a synonym for the usualēṉōr/piṟavōr (cf. also oḻintōr in 498). In this case we have a shiboleth sūtra: besides the usual speakers also others are possible. Note that Nacc. takes the opportunity to introduce HER (a most surprising omission indeed) and the foster-moster (probably an invention of later poetics; see Kaḷavu).
The scopus of the ila still remains unclear (and this again looks like an interrupted sequence); it might be HIS speech situations or any of the ones enumerated so far. Or we have simply to complement with kiḷavi.
116 This sūtra must be dislocated, because it doesn´t make sense in this context. The immediate surroundings provide no noun for complementing the ellipsis: what is theētu? On the other hand it might also be possible to read a general statement as to poems also dealing with the past. This seems to be the strategy of the commentators.
117 ētu = Skt. hetu-?
118 kūṟi: the absolutive is quite awkward here. If this clause is meant to give the condition for the main sentence we would expect a conditional or a nominal construction with locative.
119 This sentence is just as bad though it at least might belong together with the preceding one. The point might be that a reference to what has happened before (for example in another place and time: HE on his way remembering their being together) doesn´t change the tiṇai-classification for the whole poem (which is oriented by the actual situation of the speaker).
120 How to understand viravu here? (The first has to be taken as a noun, the second as a verb, I suppose, and the -um´s are here coordinating viravum and poruḷum.) Is it another word for tiṇaimayakku, as the commentators seem to think? Then it probably belongs to another textual layer. In this case the sūtra would refer to the possibility of extending the mixture of tiṇai-s to the level of poruḷ-s − which wouldn´t be great news after the rules 13f., 18. And the wording is not convincing at all: why say that it is the mingling which is mingled with poruḷ?
121 = without deviating from the fixed word-usage.
122 The point here is obviously not to define the two kinds of simile, but to establish the relation between tiṇai and simile (which is intimate indeed: if there is tiṇai poetry, it is to be found in connection with the inset poems). Is this miscellaneous material simply added at the end?
123 According to sandhi dissolution we can read moḻintatai or oḻintatai, which seems to yield contrary meanings, but since deities are mentioned so rarely in the poems at all, it might be more useful to read the latter with the commentaries. Why on earth here the accusative marking? These instances can be counted on a hand for the whole of TP.
124 muṭikeṉa seems to be one of the "irregular" cases of optative in -ka plus eṉ. This would be a highly unusual construction for a TP sūtra: embedded direct and modal speech.
125 What follows here, without any contextual imbedding, is the description of the two additional settings which is missing in the beginning of the section. Note that it is not a structural description at all, there probably being no poruḷ-s for kaikkiḷai and peruntiṇai, because the "world" as contained in the five settings is already complete. It is a thematic description presupposing the dramatic conception of the speaker sūtra-s, actually it contains additional speech situations for HIM.
126 The meaning of the 2 medium lines is not at all self-evident. Nacc. connects the puṇarttu with the process of HIS speaking, but that doesn´t clarify the internal syntax of the lines.
127 With goodwill this can be read as the four tiṟam-s taught in the previous sūtra also being applicable to kaikkilai − which either makes peruntiṇai a subtype of kaikkilai, or it doesn´t make sense at all.
128 nāṭakam: again a clear Sanskritsm, but this time a very strange one. What has drama to do with Akam poetry? Would it be conceivable that nāṭakam in Tamil can be used as an equivalent to ceyyuḷ − as Nacc. seems to understand it: "in ornate speech "? Which Sanskrit texts have to be presupposed for this? And is this an even later complement (later than TP 8), which feels also the development of metres has to be accounted for somehow?
129 Do we have to assume here akaṉ instead of akam?
130 What can be the meaning of aḷavutal here? The commentary interpretation of taking this as referring still to proper names in Akam and Puṟam is at least not obvious, and the word deviation akaṉ in 56 but akam in 57 doesn´t exactly encourage reading these two as actually belonging together.
Auteur
-
Eva wilden
Eva Wilden (b. 1965) is a researcher at the École Française d’Extrême-Orient in Pondicherry where she is engaged in a project of the preservation and edition of Classical Tamil manuscripts. After studying Indology and Philosophy at the University of Hamburg she took a doctorate on Vedic ritual and afterwards specialised in Classical Tamil under the guidance of S.A. Srinivasan. Her habilitation comprised a translation of the Kuṟuntokai, and currently she is working on a critical edition plus translation of the Naṟṟiṇai, and on a partial edition and translation of the Akanāṉūṟu together with J.-L. Chevillard. She has been Associate-Editor for this volume.
Le texte seul est utilisable sous licence Licence OpenEdition Books. Les autres éléments (illustrations, fichiers annexes importés) sont « Tous droits réservés », sauf mention contraire.
La création d'une iconographie sivaïte narrative
Incarnations du dieu dans les temples pallava construits
Valérie Gillet
2010
Bibliotheca Malabarica
Bartholomäus Ziegenbalg's Tamil Library
Bartholomaus Will Sweetman et R. Ilakkuvan (éd.) Will Sweetman et R. Ilakkuvan (trad.)
2012
