Vuttodaya, The Composition of Metre


Principles of Indo-Āryan Metrical Composition

In the descriptions the following symbols are used:

= a light syllable
= a heavy syllable These two basic symbols are how light and heavy syllables are represented in European prosodies, but this is the opposite to those usually used in India, and as defined in Vuttodaya itself at v. 7.
= light or heavy syllable (anceps)
× = the syllable may be naturally light or heavy, but it is always taken as heavy, owing to the pause that follows it

The forms of verse structure in Indo-Āryan composition are based on three variables: the length of the line, and of the sections that make up a line; Normally separated by a diaeresis, or word break. the alternation of light and heavy syllables, which make rhythmic patterns within the line; and the collection of lines, similar or dissimilar, into a verse.

In the Indo-Āryan languages there is a differentiation between long and short vowels, which are very distinct and change the meaning of words; and similarly, consonant clusters are pronounced as such, and meanings change accordingly, depending on whether a consonant is single or in a cluster.

Because of these clear distinctions all syllables can be assigned a weight: short vowels in open syllables are light, all others are heavy, with light syllables being considered half the weight of the heavy ones. In later terminology, light syllables have one measure (mora, mātra), heavy syllables two.

This opposition between light and heavy syllables means that the languages have an inherent rhythm to them, even in the prose sections, but it is also easy to produce rhythmic cadences, openings and so on, which can be employed in verse.

And once the strict division of syllables into measures was established it also became possible to count the amount of measures in a line, and use that as the structural length. They were divided either into sections of four mātra, like the Gaṇacchandas (Āryā group) and Matrāsamaka metres; or into combinations of four, six and eight mātra, as in the Mātracchandas metres (Vetālīya group).

For our purposes we can divide the evolution of Indo-Āryan verse into three periods, always remembering though that the periods are not exclusive, and that in the first two periods especially structures were still very much evolving:

1) The Vedic Period
2) The MIA Period
3) The Classical Period

In synopsis, we can say that the length of the line in the early period was a simple syllable count, regardless of weight; but after the Vedic period a count of the measures in a verse could also be used as the defining quantity, with a light syllable being counted as one measure (mātra), and a heavy as two.

In the Vedic period the alternation of light and heavy syllables was mainly seen in the cadences to the line; by the MIA Period openings and cadences were very often defined; and the Classical Period saw the rise of verses with highly regulated rhythms, with the length of nearly every syllable defined, which in Classical Sanskrit, at least, became invariable.

Whereas in the Vedic Period one of the main structural principles was the alternation of lines having a different length within a verse, in the MIA Period this had already been discarded; and in the Classical Period the main metres were the samavtta metres which had four lines of the same length to the verse.

The sense was normally complete within the line in the Vedic Period, and there was a definite pause at the end of each line; but in the MIA Period the sense was completed with the pair of lines, and sometimes only with the whole verse; but by Classical times it was often streched over a number of verses.

Having given an outline, we can now give a concise description of the metres in each of these periods.


The Vedic Period

The verses in the Vedic period were quite simple structurally, having mainly 5, 8, 11 or 12 syllables to a line, and various combinations of lines to a verse. Modulation in the stress on the words gave them their rhythm, until at a later stage the alternation of light and heavy syllables in the cadence started to become important rhythmically.

By then the famous Gāyātrī verse was a line of eight syllables, repeated three times, having a cadence in each line which is usually, though not invariably, ¦⏑−⏑×.

The Anuṣṭubh line was similar to this, but repeated four times in the early period (⏓−⏓−¦⏑−⏑×). In the later Vedic Period it developed into a verse having a different cadence in the 1st and 3rd lines, so that eventually it could be described as being: ⏓⏓⏓⏓¦⏑−−צ¦⏓⏓⏓⏓¦⏑−⏑×.

The Triṣṭubh was a line of 11 syllables, mainly, but again not invariably, having the following profile in the cadence: ¦−⏑−×. The most common form was a verse having four lines, but we also find verses having only two lines, the dvipadā, and having three lines, the virāj.

The Jagatī was similar to the Triṣṭubh in most respects, but with an extra syllable in the cadence (¦−⏑−⏑×), which gave a 12 syllabic line. Triṣṭubh lines are quite often found in Jagatī verse, but it appears more as a matter of convenience, and not aiming at a definite structure which would repeat and form a new verse structure.

Another line of only 5 syllables had a pattern similar to the Triṣṭubh cadence: ¦⏓−⏑−×. When four of these lines made up a verse is was known as the Dvipadā Virāj.

Dissimilar line structures were also often combined in different ways to form a distinctive verse, so, for instance, the Uṣṇiḥ had three lines of eight, eight and twelve syllables (8, 8, 12), being therefore a combination of Anuṣṭubh and Jagatī.

The Purauṣṇiḥ was the reverse of this, with a twelve syllable line, followed by two of eight syllables: (12, 8, 8). The Kakubh showed the other variant possible: (8, 12, 8).

Other combinations included the Bhatī: (8, 8, 12, 8); the Satobhatī: (12, 8, 12, 8); and there were also some very elaborate verse structures: Atiśakvarī: (8, 8, 8, 8, 8, 12, 8); and Atyaṣṭi: (12, 12, 8, 8, 8, 12, 8).


The MIA Period

The main collections of literature written in early MIA are the Pāḷi, For a description of the metres used in the Pāḷi Canonical period, see An Outline of the Metres in the Pāḷi Canon:, upon which much of what follows draws. Sanskritised Prākt and Ardha-Māgadhī scriptures, representing the Buddhist and Jaina teachings, which all contain a substantial amount of verse.

The main differences from the Vedic Period are three-fold:

1) The lines in the longer metres are more strictly defined, having set openings as well as cadences, resulting in less rhythmic freedom in the line.

2) There is a new organising principle, the measure (mātra, or matta), which is used to produce new verse structures.

3) The old dissimilar lines making up a verse become incidental, being only employed as a matter of expediency, not as a true structural principle, as in the earlier period.

The Anuṣṭubh, or Siloka as it is now known, is similar to the later structure in the Vedic Period: ⏓⏓⏓⏓¦⏑−−צ¦⏓⏓⏓¦⏑−⏑×; The prior line quite often will show one or other variation, but other forms of the posterior cadence are hardly ever found, only when the author couldn’t fit his ideas or vocabulary to the metre.

The earlier Triṣṭubh and Jagatī metres now have a more restricted profile, which can be defined like this: ⏓−⏑−¦⏓⏑⏓¦−⏑−[⏑]×. The opening and closing syllables are free, as are the ones in 5th and 7th position, but the rhythms of the opening and cadence are fixed now. The two metres are mixed in a verse ocassionally, but the tendency is to prefer one or the other.

During this time new metres arise, which have a different organising principle for the line, the measure. As mentioned earlier a heavy syllable is considered twice the length of a light one, and it is therefore possible to determine the amount of measures there are in a line.

The earlier metres were principally defined by syllabic count and cadence rhythm; and it is similar with the measure metres (Mattacchandas), except the count is now made of the overall measures in a line, the syllabic count being then variable.

These metres are further distinguished by being organised into pairs of lines, where the prior and posterior lines are of different length.

The Vetālīya has 14 measures in the prior lines, and 16 in the posterior, with a cadence: ¦−⏑−⏑×, giving an overall structure:

Prior Line:        ⏕⏕⏕¦−⏑−⏑×
Posterior Line: ⏕⏕⏕⏕¦−⏑−⏑×

The other common metre is the Opacchandasaka, which has 16 measures in the prior and 18 in the posterior, with an extended cadence: ¦−⏑−⏑−×.

Prior Line:        ⏕⏕⏕¦−⏑−⏑−×
Posterior Line: ⏕⏕⏕⏕¦−⏑−⏑−×

The bar metres (Gaṇacchandas), which also employ the measure to determine the length of the line, differ from these by not having fixed cadences, and by making a gaṇa of four measures the basic structure, which could then take one of the following forms: −− or ⏑−⏑ or ⏑⏑− or −⏑⏑ or ⏑⏑⏑⏑.

There were two basic structures to the line, the first one having this profile:

     1         2       3         4         5        6         7     8

and the second line looking like this:

     1         2       3         4         5     6      7     8

The internal rhythms are defined by the alternations of two rhythmic structures:

These metres also affected the syllabic metres as it came to be felt that as one heavy syllable is equal to two light syllables, then these could, on ocassion, substitute for each other, and the phenomena of resolution and replacement arose in the syllablic metres, so that we very often see the first syllable of a line being presumed to be heavy, and then resolved into two light syllables in the syllabic metres.

Other syllables could also follow this allowance, with the reservation that the syllables being resolved must be the first two syllables of a word. The compliment to this is seen much less often when two light syllables are replaced by a heavy one.

Although we do still see mixing of lines in verses at this stage in the development of Indo-Āryan metre, it is now a matter of expediency and it is not an organisational principle, as in the Vedic Period. Also a verse is now considered to be made up normally of four lines, and verses of other lengths, although found, are incidental, not structural.


The Classical Period

The Classical Period is characterised by the use of an array of highly defined and invariable metres of various lengths and shapes, which became the norm; the allowance for a compound to cross the seam in a pādayuga, even in long lines; the completed sense, which was previously restricted to the verse, to be extended to two or more verses; and for one of the longer metres to conclude a chapter with a flourish.

The Śloka metre, which was previously quite free, especially in the prior line, became much more restricted at this time, with the pathyā cadence (¦⏑−−×) in the prior line being much preferred to any other variation.

Of the Samvutta metres the Triṣṭubh and Jagatī have been standardised, with even the break now having a fixed shape, and they have been analysed into their different structures, which are named:

−−⏑−¦−⏑⏑¦−⏑−×       Indravajrā (Indavajirā)
⏑−⏑−¦−⏑⏑¦−⏑−×       Upendravajrā (Upindavajirā)

⏑−⏑−¦−⏑⏑¦−⏑−⏑×       Vaṁśasthā (Vaṁsaṭṭhā)
−−⏑−¦−⏑⏑¦−⏑−⏑×       Indravaṁśā (Indavaṁsā)

These are the most widely used of the Classical metres, either indivually, or in combination, in which case they are known as Upajāti, and the 16 different varieties of Upajāti having Indravajrā and Upendravajrā have all been given individual names. See v. 63 below for a list

Vasantatilakā, a 14-syllablic metre, which is derived from the Indravajrā with an extended break, was also very popular:


Other long metres which were popular include:

⏑⏑⏑⏑⏑⏑−−,−⏑−−⏑−×       Mālinī
⏑−−−−−,⏑⏑⏑⏑⏑−−⏑⏑⏑×       Sikhariṇī
−−−−,⏑⏑⏑⏑⏑−,−⏑−−⏑−×       Mandakkantā
−−−⏑⏑−⏑−⏑⏑⏑−,−−⏑−−⏑×       Saddūlavikkīḷita


The Prosodic Literature

Chandaḥ Śāstram

The most important of the works on Sanskrit prosody that survives is Śrī Piṅgala’s Chandaḥ Śāstram, as it is from Piṅgala’s system that all the other classical works on Indian prosody, including Vttaratnākara, and its translation Vuttodaya, have been derived. Śrī Piṅgala’s work itself, however, is by no means a first attempt at writing a prosody, but has to be seen as the culmination of a long development in the science, which had attained a great deal of precision by the time it was written.

The work itself sometimes quotes the views of other teachers who define various matters differently – e.g. in the description of the Vaktra metre (5:18), we find the line: sarvataḥ saitavasya but according to Saitava, all (the lines should show a certain variation). Or, again, following the description of the Vasantatilakā metre (7:8): vasantatilakā tbhau jau gau; we find 2 sūtras (7:9): siṁhonnatā kāśyapasya according to Kaśyapa (the name is) Siṁhonnatā; (7:10): uddharṣiṇī saitavasya according to Saitava (the name is) Uddharṣiṇī.

Chandaḥ Śāstram (The Science of Metre) is divided into 8 sections (adhyāya), having a total of 307 sūtras in all. The first two sections deal with general matters concerning the gaṇa system, heavy and light syllables, measures (mātrā), the gods associated with the metres, and so on; the 3rd section describes the Vedic metres; the 4th continues the description of the Vedic metres; before turning to the classical (laukika) metres, beginning with the measure metres (mātrācchandas), in the order: Āryā, Vaitālīya, Mātrāsamaka, & Gītyāryā. The order in which the metres are dealt with here is not strictly logical, as the Mātrāsamaka & Gītyāryā are gaṇacchandas metres, and varieties of the Āryā group of metres; while the Vaitālīya group are mātrācchandas metres, based on somewhat different principles. Nor is the order historical as, according to Warder PM, the Vaitālīya group of metres probably emerged first, followed by the Gīti, Ariyā and the other gaṇacchandas metres. The 5th section describes the Vaktra, Viṣamavtta, and Ardhasamavtta metres; the 6th deals with the Samavtta metres from Gāyātrī, six syllables, to Jagatī, 12 syllables; section 7 continues from Atijagatī, 13 syllables, up to the Daṇḍaka metres, more than 26 syllables. The last section, no 8, is a kind of supplement to the preceding, which includes extra metres and directions for the making of tables.

As is normal in the genre many of the sūtras are extremely terse, and are not easily understood by the uninitiated. The first 5 sūtras in Chandaḥ Śāstram, for instance read: dhī-srī-strī m || 1 ||; varā sā y || 2 ||; kā guhā r || 3 ||; vasudhā s || 4 || sā te kva t || 5 || – those who already know something about the science may be able to understand that this is the beginning of a description of the gaṇa system, others though will simply be mystified, without the aid of a commentary.

The most important commentary on the Chandaḥ Śāstram is that of Śrī Halāyudhabhaṭṭa, which is called Mtasañjīvanī (Bringing the Dead back to Life). This commentary not only explicates the sūtras, but also gives standard examples of the metres described therein, and provides much supplementary information about the science.



Vttaratnākara by Śrī Kedārabhaṭṭa differs from the Chandaḥ Śāstram in a number of important respects. For one thing it does not describe the Vedic metres at all, but concerns itself exculsively with the classical metres. The format is also changed from memorial sūtras to memorial kārikās, so that virtually the whole book, and therefore all the prosodic rules, are themselves written in verse.

Not only that, but the rules describing the metres are written in the form of the metre itself, so that the rule is also an example of the metre. For instance, the description of Indravajrā (v. 91) reads:

syādindravajrā yadi tau jagau gaḥ

and is therefore written in the metre it is describing. To produce a description that is also an example, of course, requires great ingenuity and skill of the part of the writer.

The idea that the rule should also be the example is also found occasionally in Chandaḥ Śāstram. For example the opening sūtras quoted above on the gaṇa system, scan as examples of the gaṇas themselves dhī-śrī-strī m −−− || 1:1 ||; varā sā y ⏑−− || 1:2 ||; kā guhā r −⏑− || 1:3 ||; vasudhā s ⏑⏑− || 1:4 || sā te kva t −−⏑ || 1:5 || etc. SED quotes vasudhā in the meaning anapæst–but it seems the word is only an illustration of an anapæst (ragaṇa), and does not actually mean that. Occasionally the rules for the metre also scan as examples: kumāralalitā jsaug || 6: 3 || ⏑−⏑⏑⏑−− , but this appears to be just a coincidence. It is possible, however, that it may have had some influence upon the origination of the idea that the rules themselves should exemplify the metre.

Vttaratnākara is divided into 6 sections: the first deals with the terminology (saṁjñābhidhāna); the second with the measure metres (mātrāvtta), in the order: Āryā, Vaitālīya, Vaktra, Note that the inclusion of Vaktra is an innovation of Śrī Kedārabhaṭṭa’s, albeit an unhappy one, as Vaktra is not a mātrācchandas metre at all. Chandaḥ Śāstram deals with Vaktra seperately at the beginning of the Vtta metres. Mātrāsamaka, Dvipāda. The third section concerns the metres that have the same line repeated 4 times to make up a verse (samavtta), with rules and examples from one syllable to 26, In my edition of this text I have noted that the metres from one syllable up to five syllables are probably additions to the text made at a later date. and then the Daṇḍaka class of metres that have more than 26 syllables. The fourth section deals with the metres that have two dissimilar lines that are repeated (ardhasamavtta) to make up a verse. The fifth with the metres that are dissimilar in all their lines (viṣamavtta). The last section deals with the tables that can be made of the various metres, and their use in identifying the metres, etc.



As is to be expected from what is more or less a translation, Vuttodaya follows the same outline as Vttaratnākara, having the same 6 sections, dealing approximately with the same material, and it is noticeable that both Vuttodaya & Vttaratnākara differ from Chandaḥ Śāstram, in much the same way. Both leave out the Vedic metres and when, for instance, the sequence of metres in any particular section differs between the prosodies, then Vuttodaya nearly always agrees with Vttaratnākara; See the inclusion of Dakkhiṇantikā (VR 37, Vutt 30); and Acaladhiti (VR 53, Vutt 36); and cf. the sequence of metres in the Mātrāsamaka & Ardhasamavtta sections. For a detailed comparison of the material found in the three prosodies, see: which is, of course, very significant for defining the relationship between the 3 prosodies.

However, Vuttodaya is not slavish in copying Vttaratnākara, as Ven. Saṅgharakkhita deals with his most immediate source quite creatively: rearranging, editing, and rewriting the material whenever he has felt it to be necessary.

We may note here some of these changes, beginning first with the editorial omissions. The list of generic titles of the metres (chandas) that is found near the end of the first section has been omitted in Vuttodaya. In Vttaratnākara the metres up to 26 syllables, have been illustrated. Vuttodaya, however, only illustrates the metres up to 22 syllables. Moreover the whole class of Daṇḍaka metres, which have more than 26 syllables have also been excluded. Similarly the Dvipāda metres which are described as a part of the Mātrācchandas class of metres in the Sanskrit work, have been excluded in the Pāḷi work.

Further all the metres that have been described in the Viṣamavtta section in Vttaratnākara have been excluded from Vuttodaya. Instead their place has been taken by a description of the Vatta metre, which has been moved to the Visamavutta section in Vuttodaya, Again this is not a happy innovation, as Vatta is no more a Visamavutta metre, than it is a Mattacchandas metre. although in the Sanskrit source it is found in the Mātrācchandas section.

In the translation itself the material at hand is dealt with in several different ways, there are transliterations, with or without insignificant changes in the words employed; rearrangements; partial translations, and total rewrites. Below is a list showing the different ways in which the material has been dealt with, with examples. It should be noted that the inclusion of a line in one or other or the specified catorgories is often a rather subjective choice, and this table is only meant to be indicative of the types of translation that has been evolved.

37; 44; 45; 47; 50; 52; 53; 56; 57; 60; 75; 76; 77; 81; 83; 87; 88; 89; 90; 91; 94; 95; 103; 109; 121.

e.g. v. 44 – VR: Tyau cet-Tanumadhyā; Vutt.: Tyā ce Tanumajjhā.

Transliterations with inconsequential changes in words:
29; 36; 39; 40; 42; 51; 54; 55; 58; 65; 67; 69; 70; 71; 74; 82; 86; 97; 100; 104; 111; 122; 123; 131.

e.g. v. 36 – VR: Dvikaguṇitavasulaghu-r-Acaladhti-r-iha; Vutt.: Dvikavihatavasulahu-r-Acaladhiti-r-iha. Both guṇita & vihata both mean multiplied.

As can be seen, most of the first two classes of changes occur in the Samavutta section (vv. 44–104).

10; 46; 48; 49; 66; 72; 79; 96; 98; 99; 105.

e.g. v. 48 – VR: Māṇavakaṁ bhāttalagāḥ; Vutt.: Bhā talagā Mānavakaṁ.

Transliterations that contain word translations:
38; 63; 78; 80; 93; 98; 99; 101; 117; 118; 124.

e.g. v. 78 translates: Bhujaṅgaprayātaṁ bhavedyaiścaturbhiḥ; with Bhujaṅgappayātaṁ bhave Vedayehi; the only change being the use of the bhūtasaṅkhyā Bhūtasaṅkhyā words are words that invariably have numbers attached to them by convention, as will be explained below. See the commentary to v. 10. word Vedayehi for caturbhiḥ.

4; 5; 6; 7; 8; 9; 11; 12; 13; 14; 17; 18; 19; 20; 21; 22; 23; 24; 25; 26; 27; 28; 30; 31; 33; 34; 35; 41; 43; 68; 73; 96; 102; 106; 107; 108; 110; 112; 113; 114; 115; 116; 119; 120; 125; 126; 127; 128; 129; 130; 132.

e.g. VR: Trīrajau galau bhavedihedśena lakṣaṇena Vttanāma;
Vutt.: Vuttam-īdisan-tu nāmato rajā rajā rajā garū lahū ca.

Notice that the major rewrites occur in Chapters 1 (Saññāparibhāsāniddesa; vv. 1–16); 2 (Mattāvuttaniddesa; vv. 17–43); 4 (Addhasamavuttaniddesa; vv. 105–115); & 6 (Chappaccayavibhāga; vv. 127–132).

Ven. Saṅgharakkhita sometimes condenses the material, as in the summary of the types of syllabic metre given in vv. 11 & 12; or in the description of the Ariyā Vipulā & Capalā, v. 20, which in Vttaratnākara have a verse each (vv. 26 & 27), Vuttodaya reduces that to a line each. He is also occasionally able to give more complete descriptions of the metres (cf. vv. 17 & 18); or simply better and clearer descriptions (cf. vv. 30–32).

Some verses in Vuttodaya find no direct parallels in the Sanskrit original. These include the verses at the beginning and end of the work (vv. 1, 2, Though there are dedicatory verses in Vttaratnākara, they are not real parallels. 133–137); and the verses at the end of the first chapter where the list of generic metres has been excluded from the Pāḷi work, and a summary is included instead (vv. 15–16).

Very rarely there are descriptions of metres that are not found in the Sanskrit work as we now receive it, and it is a moot point which cannot readily be answered: whether these have been lost from the original, or were additions found in the copy Ven. Saṅgharakkhita was translating, or were included by the translator himself from other sources when composing the work (see vv. 59 & 85).


Ven. Saṅgharakkhitattherapāda

The Vuttodaya itself is a masterful work by one of the most important figures of the Pāḷi literary renaissance that flourished in Sri Lanka around the 12th and 13th centuries. As can be seen from Ven Saṅgharakkhita’s teachers and disciples as outlined below he stands at the heart of this renaissance, and his works were vital in a number of ways in the Pāḷi tradition.

Amongst his surviving works are books on Pāḷi prosody, ornamentation, grammar and also a sub-commentary on the vinaya rules. The first two particularly have been enormously influential ever since they were written and formed the basis of the teaching of Pāḷi literature all through the Medieval period down to the present day.

Nearly all works in the Pāḷi tradition on both prosody and ornamentation that follow him are but commentaries, elaborations and illustrations of his works, and never did a school arise to rival his central position in the tradition.

This was the period that saw two major developments in Pāḷi literature: the works on the sub-commentaries to the Canonical texts, which update the collected wisdom of the teachers in the post-commentarial period; and consolidated works on various aspects of Pāḷi grammar that had been lacking for the most part up till that time.

Ven. Saṅgharakkhita’s status was assured not only as a literary figure, but also as a central member of the Saṅgha by his appointment to the position of Mahāsāmi following the passing of his teacher Ven. Mahākassapa of Udumbaragiri, and he became the second person to hold that place in the Saṅgha, which eventually became the position of Saṅgharāja in many traditions.

Below I give a number of charts which illustrate his position in regard to his teachers, his contemporaries and his disciples, and the major works that they produced in what has to be regarded as one of the most fruitful periods in Buddhist literary history, and well illustrate why this period in known as a renaissance in learning in the tradition.





Texts & Abbreviations

This edition of Vuttodaya was prepared from a comparison of the following twelve manuscript and printed editions, which I list and assess here.

Palm Leaf Manuscripts

MK = Palm Leaf Manuscript from Kedārakārāmaya, Godagama, Kosgoda, Sri Lanka. A photocopy of this manuscript was supplied to me by the National Archives, Colombo; the copy is quite good, but a few lines in Chapter 5 cannot be read, and the variants are therefore not recorded. The manuscript is clearly, even if a little carelessly, written. There are numerous mistakes and crossings out, which are sometimes a little hard to see – I have occassionally given the scribe the benefit of the doubt where the copy is not clear. The scribe distinguishes very clearly between -i- and -ī-, despite the many variants in this regard, and there can be no doubt about what he intended. There are some interesting variants like always reading Patthyā with the conjunct -tth- .

MP = Palm Leaf Manuscript No 277291, University of Peradeniya Arts Library. Small, neat handwriting with numerous crossings out and inconsistencies. A number of times the readings adopted go against the metre, especially in the Ariya section, and I get the impression that the scribe didn’t fully understand what he was writing. There are also some idiosycnrasies like normally writing -n- where retroflex -ṇ- is normally found (e.g. gana; lakkhana, vanna)–though here again the scribe is inconsistent; also quite often writes ṇa when intending na, as in ṇa(gaṇa).

MR14 = Palm Leaf Manuscript in Book No 14, Rāja Mahā Vihāra, Ridigama. Scruffy handwriting, sometimes hard to read. Some omissions and many crossings out. Numerous unmetrical readings accepted by the scribe. Different tradition from MR93 below. Writes guru for garu. Quite often writes ṇa when intending na, as in ṇa(gaṇa).

MR93 = Palm Leaf Manuscript in Book No 93 leaves Ka–Kī, Rāja Mahā Vihāra, Ridigama. Different tradition from the manuscript listed above. This is a finely written manuscript, with occasional omissions. It has many unmetrical readings, especially in the Ariya section, which spoil the metre. Normally takes the spellings guru and patthyā and Vetāḷiya.

Printed Editions

Fry = Vuttodaya (Exposition of Metre), edited by Major G.E. Fryer (Calcutta, 1877). Includes extracts from the following Burmese commentaries: Chandosārattha; Vacanatthajotikā Ṭīkā; & Kavisāra, examples of the Samavutta metres (presumably taken from one or other of the commentaries), and additional explanatory notes by the editor himself. There are a number of mistakes in the edition, which can only reasonably be put down to the printer (Burmese Tradition).

Wim = Vuttodaya, with a Sinhalese Translation, edited by Rev. M. Wimalajoti (Colombo, 1888). Includes an edition of the Sinhalese Sannaya. The variants have been taken from the text with the Suddhipatraya (page unnumbered) taken into account. I have a very poor photocopy of this work, and occasionally I have taken the reading from the Sannaya when the intended reading in the text is in doubt (Sri Lanka Tradition).

Dhm = Vuttodaya, by Udammiṭa Dhammarakkhitatissa Sthavira: (Balapitiya, 1916). Includes editions of the Sannaya, Vivaraṇa, and Vuttodayaṭīkā. The variants have been taken from the text at the beginning of the book, with the Suddhipatraya, pg 72 ff. taken into account. Occasionally I have taken the reading from the text as printed in the Sannaya when the intended reading in the text at the beginning of the book is in doubt (Sri Lanka Tradition).

Dīp = Vrittodaya, by Sri Deepankara Sthavira Swami (Alutgama, 1925). Includes the Vttadīpikā on the Vttodayavaṇṇanā; and the Vttodaya Pariśiṣṭaya written by Ven. Dīpaṅkara himself. The variants have been taken from the text with the Suddhipatraya (page unnumbered) taken into account (Sri Lanka Tradition).

Laṅk = Laṅkānanda Vyākhyā sahita Vuttodaya, by Ācārya Paṇḍita Labugama Laṅkānanda Svāmin Vahanse (Colombo, 1946). Prints the text at the beginning of the book, and repeats it in the lemma to the Exposition (Vyākhyā), and again in the Gloss (Sannaya). The variants have normally been taken from the text at the beginning of the book, occasionally checked against the Vyākhyā, when the text is unclear, or the intended reading in the text is in doubt (Sri Lanka Tradition).

Sid = Saṅgharakkhita’s Vuttodaya, A Study of Pāli Metre, by R. Siddhartha. First Edition: Calcutta 1929; Second Edition: Delhi 1981. This book includes the text, translation, and a very useful (though occasionally incorrect) commentary by Siddhartha himself. The variants have normally been taken from the text, but occasionally I have taken the reading from the translation or commentary, when the intended reading in the text is in doubt (Burmese Tradition).

Kat = Vuttodaya, by Īchirō Katayama, in Buddhist Studies, Bukkyō Kenkyū Vol 3 (Japan, 1973). This is said to be a hybrid based on 3 editions, one each from Burma (1968), Sri Lanka (= L), and Calcutta (Fryer, 1887), and some commentaries, but it seems to actually follow the 1968 Burmese edition published at the Buddhasāsana Press, Rangoon. I have been unable to obtain a copy of this edition, despite my best efforts, but the readings may be inferred from Kat and Th below. The edition has some peculiarities, like always writing aḍḍha- (here addha-), but without noting the variants (Burmese Tradition).

Th = Thai Edition, edited by Ajahn Gandhasārābhivaṁsa. This is virtually identical in the readings to Kat above, but without the peculiarities, which suggests to me that this edition is simply a transcript of the same Burmese version of Vuttodaya, which would account for the very few differences between this and its predecessor (Burmese Tradition).

Other Prosodies Prepared for Publication

For a proper understanding of Vuttodaya it was necessary also to prepare editions of a number of other prosodies so as to include parallels, variations and so on.

ChŚā = Chhandas Śāstra by Śrī Piṅgalanāga, with the commentary Mitasañjīvanī by Śrī Halāyudha Bhaṭṭa, edited by Paṇḍit Kedāranāth of Jaypur (ed), (3rd edition Bombay, 1938). The earliest surviving work on Sanskrit prosody, written as sūtras, rather than as kārikās, as is the case with VR, ChM, SB, Vutt. etc.

VR = Vttaratnākara by Śrī Kedārabhaṭṭa, edited by Ānandajoti Bhikkhu, 2003.

ŚB = Śrutabodha of Kālidāsa, edited by Vāsudev Laxmaṇ Shāstrī Paṇśīkar, (Bombay, 1906).

Other Texts Prepared for Publication

A Comparative Table of the Metres found in Chandaḥśāstra, Vttaratnākara, & Vuttodaya compiled by Ānandajoti Bhikkhu (January, 2004). A comparative table of the metres found in three seminal texts in the Sanskrit and Pāli traditions, showing how these texts relate to each other.

Examples of Classical Metres from Mahāvaṁsa & Cūlavaṁsa from the edition by
Wilhelm Geiger (1908, republished London, 1958). Over seventy examples of classical Pāli metres as found mainly in the colophons to the chapters in the Mahāvaṁsa, and its continuation, the Cūlavaṁsa.

Studies in Ven. Buddhadatta’s Prosody (Vinayavinicchaya; Uttaravinicchaya;
Abhidhāvatāra; Buddhavaṁsa-aṭṭhakathā) (2005). An analysis and study of the metres in Ven Buddhadatta’s four main commentaries.

Metre Tables (Chandaḥprasthāra) compiled by Ānandajoti Bhikkhu (Sept. 2005/2549; revised version 2.0 Aug. 2013). A listing of the metres known to occur in Sanskrit and Pāli literature, along with explanations of how to make metre tables, an index and a searchable database.

Jinacaritaṁ by Venerable Medhaṅkara Thera, edited by Ānandajoti Bhikkhu together with a metrical analysis (2007). A new edition of the Medieval Pāli text retelling the Life of the Buddha. Includes notes on variant readings, with an analysis of the metre of the verses.

Dāṭhāvaṁsa by Ven. Dhammakitti edited by R.D Rhys Davids & R. Morris (JPTS 1884). An edition of the Medieval Pāli text telling the story of the tooth relic of the Buddha, with an analysis of the metre of the verses.

Namakkārapāḷi saha Saṅkhepayojanā, The Reverence Text with the Short Word-Commentary edited and translated by Ānandajoti Bhikkhu (April 2013). A Pāli and English line by line (interlinear) version of this medieval text from Myanmar praising the Buddha, written is elaborate metres, together with its word-commentary.

Tigumbacetiyathomanā, Praise of the Tigumba Shrine (re-edited and translated by Ānandajoti Bhikkhu (August 2013). Pāli and English line by line (interlinear) version of a chanting text from Myanmar praising the Shwedagon Pagoda, which illustrates the variations that occur in the Siloka metre.

Narasīhagāthā from the Jātakanidāna.

Other texts re-edited and quoted from, but not yet prepared for publication:

Ganthārambhakathā to Dīghanikāyaṭṭhakathā
Ganthārambhakathā to Saṁyuttanikāyaṭṭhakathā
Jinālaṅkāra by Ven. Buddharakkhita
Pajjamadhu by Ven Buddhappiya
Jinavaṁsadīpa by Ven. Medhānanda.

Other Works Consulted

Chhandomañjarī of Gaṅgādāsa, edited by Vāsudev Laxmaṇ Shāstrī Paṇśīkar, (Bombay, 1906).

Jayadāman, by H. D. Velankar (Bombay, 1949). A very useful work by the great Vedic and Classical scholar, containing his editions of Jayadeva’s Jayadevacchandas, Jayakīrti’s Chandośāsana, Kedārabhaṭṭa’s Vttaratnākara and Hemacandra’s Chandonuśāsana; as well as the author’s own invaluable collection of more than 850 classical metres Vttakusumoccaya, it also contains a useful introduction.

Sanskrit Prosody: its Evolution, by Amulyadhan Mukherji (Calcutta, 1976). An interesting study, which is quite suggestive in places, but ultimately spoiled by the author’s ignorance of the MIA stage in Indo-Āryan prosody, which leads him to many wrong conclusions.

The practical Sanskrit-English dictionary. Apte, Vaman Shivaram. Revised and enlarged edition (Poona, 1957-1959).

This Edition

As can be seen from the titles listed above the work on Vuttodaya has occupied my time on and off for a good deal of the past fifteen years. The main text, which compares fifteen manuscripts and editions, was ready around the end of 2006; but I needed reliable editions of other prosodies and Pāḷi literary texts before the work could be brought to completion.

Even now the work is in something of an imperfect state as I have not made editions of the Vuttodayaṭīkā or the Vuttodayasannaya, and some of the main works I have quoted from to illustrate the metres, like the Jinavaṁsadīpa and the Mahāpaṇāmapaṭha remain unpublished.

However all works must be considered in process, rather than in completion, and it seems to me best to publish the work now, and hopefully get some feedback on it from other scholars and students working in the field, so that a corrected edition can be prepared at a later date.

The text of Vuttodaya is written in 137 succint kārikās that are little more than memorial verses that often need rather elaborate commentaries to explain all that has been said or implied in them, and the text as it is presented below is an attempt to do just that, placing the text in historical perspective and explaining its meaning.

The structure is presented in this way: first there are the antecedents to the text, quoted from two texts that had the most influence on Ven. Saṅgharakkhita’s work: Chandaḥśāstra and Vttaratnākara.

This is followed by the verse, or kārikā itself, together with a metrical analysis, which, in the descriptions of the metres is also an example of the metre it is describing. Then comes a sannaya, or word-analysis which parses the words and, where it is deemed neceesary, identifies its grammar.

This is followed by the direct translation of the verse, and a commentary, often rather long, on its meaning, and what other information can be added to elucidate the verse and matters of prosody in general.

Where possible this is then followed by a profile of the metre involved and an example from the literature. These have been quoted according to priority, if there was an example I was aware of in Pāḷi literature I have preferred to quote that, and only secondly to quote works that were evidently written in illustration of the verses themselves.

I have written introductions to the various sections giving an overview of the matter presented in then, and many charts, lists and other explanatory materials have been added to the text so that it has eventually become something of a compendium of prosodic knowledge, and it is hoped that it will serve as such for the interested student.

A work of such complexity is bound to have mistakes, inconsistencies, etc. in it; there could also be lack of clarity in places and things that could have been explained at more length, so I would be very grateful to anyone who could help improve the text by pointing these out, in hope of making a better edition later on.

Ānandajoti Bhikkhu
April 2016