Middle Bengali Metres^{^[i]}

Makoto Kitada

Note on the transcription method adopted in this paper

Modern Bengali pronunciation deviates considerably from the writing. Bengali terms are given in the form of one-to-one transliterations of its original spelling in Bengali script. Where necessary, the pronunciation is given in brackets [ ]. Thus e.g. namaskārˈ [nɔmoʃkar] denotes the term written written namaskārˈ in Bengali script, but pronounced [nɔmoʃkar]. The diacritic mark < ˈ > stands for an omitted word-final vowel A; e.g., namaskārˈ is actually written as namaskāra, but is pronounced in Modern Bengali as [nɔmoʃkar], with the final A dropped. In my view, this is the best, if not perfect, method of transcribing the Bengali language, for it is an exact transliteration of the original script, while at the same time giving adequate information on the actual pronunciation.

Part I: The theory of Middle Bengali Metres: Miśrabr̥tta and Dalˈbr̥tta

1. Miśrabr̥tta: The basic conception of Middle Bengali prosody

Classical Bengali, or Middle Bengali language, does not have the distinction of vowel length. Although the Bengali script, originally conceived for writing Sanskrit, contains letters representing the short and long vowels of Sanskrit (i.e. A, Ā, I, Ī, U, Ū, E, AI, O, AU), the distinction between short and long is ignored in actual pronunciation. I give here the Bengali vowel letters, followed by their respective pronunciations in modern Bengali: A [ɔ/o]^{^[ii]}, Ā [a], I [i], Ī [i], U [u], Ū [u], E [e]^{^[iii]}, AI [oi], O [o]^{^[iv]}, AU [ou]. Except for the diphthongs AI [oi] and AU [ou], all the vowels, are considered to have the same length.

As a result, Middle Bengali prosody differs significantly in its structure and patterning from that of Classical Hindi. In contrast to Classical Hindi meters, which are defined as combinations of short (laghu) and long (guru) syllables, Middle Bengali meters are simply defined by the number of the syllables or beats. More precisely, a Middle Bengali Metre consists of verse-units (B. padˈ <Skt. pada); each respective verse-unit contains a fixed number of syllables. In other words, a Bengali meter is defined by

• The number of verse units (padˈ), and

• The number of beats contained in each unit.

Therefore, in order to identify a certain meter, one simply needs first to count the units contained in the verse, and then count the number of the beats, which is equal to the number of the syllables.

For instance, the Paẏārˈ meter, one of the most popular Bengali meters, is defined as 8 + 6 beats, i.e. it consists of two verse-units, the first unit containing eight beats and the second six. An example of this meter is given below:

tomāre bujhāi bandhu ǀǀ tomāre bujhāi ǀ1aǀ

ḍākiẏā śudhāẏa more ǀǀ hena jana nāi ǀ1bǀ

anukhana gr̥he mora ǀǀ gañjaẏe sakale ǀ2aǀ

niścaẏa jānio muñi ǀǀ bhakhimu garale ǀ2bǀ

The original pronunciation may be reconstructed for Middle Bengali^{^[v]} as follows^{^[vi]}:

*[tomare bujhai bondhu ǀǀ tomare bujhai]

*[ḍakia ʃuddhayo more ǀǀ hæno jono nai]

*[onukhono grihe moro ǀǀ gɔñjɔyo ʃɔkole]

*[niʃcɔyo janio muñi ǀǀ bhokhimu gɔrole]

The verse is divided into two units: eight beats + six beats. At the same time, it consists of eight syllables and six syllables; at this stage of linguistic development, the number of beats equals that of syllables.

At a later period, however, a phonetic change took place in Bengali comparable to the omission of the final short vowel A in Hindi^{^[vii]}: In Bengali, the vowel A, i.e. represented as the letter of A and normally pronounced as [o], came to be omitted word-finally in actual pronunciation due to the effect of the strong accent on the first syllable of the word: e.g. āja [*ajo] > [aj], saba [*ʃɔbo] > [ʃɔb], ghara [*ghɔro] > [ghɔr]. As a result of this change, which was well established by the 15^th century^{^[viii]}, the Bengali meters can no more be defined as simply as before, as the beats and syllables contained in a verse do not always agree in number. For instance, the Paẏārˈ verse quoted above is pronounced differently at this stage, and the following is the pronunciation heard in Bengal today (instances of omitted final vowel A [o] are marked in italics).

[tomare bujhai bondhu ǀǀ tomare bujhai]

[ḍakia ʃuddhae more ǀǀ hæno jon nai]

[onukhon grihe mor ǀǀ gɔñjɔe ʃɔkole]

[niʃcɔe janio muñi ǀǀ bhokhimu gɔrole]

This change of pronunciation caused the decrease in the number of syllables in the verse, which consequently is no longer equal to the number of beats. For example, the verse unit “hæno jon nai” now contains only five syllables^{^[ix]}, rather than six, and the unit “onukhon grihe mor”, with two instances of omitted word-final A, contains only six syllables vs. the expected eight beats. The number of the syllables does not equal that of the beats any more.

In order to solve this problem, a new rule was devised:

• A closed syllable at word-end is counted as two beats.

“Closed syllable at word-end” refers to those newly created by the loss of word-final A. Thus the closed syllables in the Paẏārˈ verse quoted above are marked in italics: jon, khon, and mor. By applying the new rule, this verse can still retain the original rhythmic structure of eight beats + six beats.

In conclusion, the definition of the Middle Bengali meters must be reformulated as follows:

The meter is defined by

• The number of verse-units, and

• The number of beats contained in each unit.

• In addition, each syllable is counted as one beat, except that closed syllables in word-final position are counted as two beats.

This modified definition of Middle Bengali meters is the one generally adopted in Bengal today. The meters thus defined are called “Miśrabr̥tta”, literally meaning “mixed mode”, because closed syllables are treated differently at word-end than elsewhere, i.e. two different modes are “mixed”.

2. Dalˈbr̥tta: Later development of prosody

In the period after the omission of the final vowel took place, there emerged a new type of prosody: Dalˈbr̥tta.

The definition of Dalˈbr̥tta is actually nothing else than the definition given at the first of all for Miśrabr̥tta:

• The number of verse-units (padˈ), and

• The respective number of beats contained in each unit.

The number of beats agrees with the number of syllables. However, Dalˈbr̥tta is based on the actual pronunciation, in which the final vowel is omitted.

The Paẏārˈ meter (8+6) in Dalˈbr̥tta thus appears as follows (an omitted final vowel is marked as a̸ )

Dalˈbr̥tta Paẏārˈ (8+6, or 4+4+4+2)^{^[x]}

āsāra̸ āśā ǀ āśā kebala̸ ǀǀ āśā mātrā ǀ holo ǀ1ǀ

citrera̸ kamala̸ yemona̸ ǀǀ bhr̥ṅga bhūle gelo ǀ2ǀ

khela̸bo bole ǀ phā̃ki diẏe ǀǀ nābāle bhūtalo ǀ3ǀ

ebāra̸

ye khelā khelāle māgo ǀǀ āśā nā purilo ǀ4ǀ

In order to grasp the meter, it is necessary to take account of the actual pronunciation:

[aʃar aʃa ǀ aʃa kebol ǀǀ aʃa matra ǀ hɔlo ǀ1ǀ]

[citrer kɔmol jæmon ǀǀ bhriṅgo bhule gælo ǀ2ǀ]

[khelbo bole ǀ phãki die ǀǀ nabale bhutolo ǀ3ǀ]

[ebar]^{^[xi]}

[je khæla kælale mago ǀǀ aʃa na phurilo ǀ4ǀ]

The number of syllables (= pronounced vowels) is 8 + 6, which corresponds to Paẏārˈ. Only the first unit of the second line [citrer kɔmol jæmon] is irregular in containing six syllables.

śamana̸ damana̸ ǀǀ rābana̸ rājā ǀǀ rāmaṇa̸ damana̸ rāma̸ ǀǀ

śamana̸ bhabana̸ nā haẏa̸ gamana̸ ye laẏa̸ rāmera̸ nāma̸ ǀǀ (Kr̥ttibāsˈ, Kiskindākāṇḍa, SENˈ 1995, p. 20)

[ʃɔmon dɔmon ǀǀ rabon raja ǀǀ ramon dɔmon ram ǀǀ]

[ʃɔmon bhɔbon ǀǀ na hɔy gɔmon ǀǀ je lɔy ramer nam]

In this excerpt, the two rhyming words [ram] and [nam] seem to be exceptionally counted as two beats, although vowels are usually counted as one beat in Dalˈbr̥tta.

Middle Bengali examples of Dalˈbr̥tta contain such anomalies. Although modern theoretical treatises on Bengali prosody consider Dalˈbr̥tta to be an established mode next to Miśrabr̥tta, and although modern poets consciously adopt Dalˈbr̥tta in their works, the real situation in the period in which verses in Dalˈbr̥tta first appeared seems to have been different. In fact, Middle Bengali poets seem to have been not fully aware that they were composing in Dalˈbr̥tta. In their composition, they switch from Miśrabr̥tta to Dalˈbr̥tta––probably unconsciously––when their expression comes nearer to colloquial styles of speech. For instance, Dalˈbr̥tta is observed in the dialogue passages in Kr̥ttibāsˈ’ Rāmāyana.

Part II: List of Middle Bengali Metres

In the following, a list of classical Bengali meters is given. This list, though not exhaustive, includes various meters belonging to the three genres, Miśrabr̥tta, Dalˈbr̥tta and Kalābr̥tta. The samples are taken from SENˈ 1995, pp. 9—31. The recordings of these samples are contained in the audio CD. They were recited by two persons. One is Mr. Durgāpadˈ Datta, who recorded plain recitations without any melody; and the other is Ms. Rādhārāṇī Datta, the singer of the Śrī Śrī Mahāprabhukīrtanˈ Sampradāẏˈ, a Pālā Kīrtanˈ group from Brāhmaṇˈḍihi in Bā̃kuṙā district.

The recitations by Ms. Datta, are the product of an experiment: these were recorded in the evening when we invited her group to perform in Bibaṙˈdā village in Bā̃kuṙā district, West Bengal. Just before her performance on stage, I interviewed her. I gave her the sample verses which she saw for the first time, and let her recite these spontaneously. Ms. Datta was thus compelled to analyze the metrical structure of each verse and recite it on the spot, without any preconceived notions. My intention was to observe the state of affairs in actual practice, namely how the meters are perceived and interpreted by the performer herself at first sight, i.e., in a condition as free as possible from and as uncontaminated as possible by scholarly theories and analyses, with which her performance actually has very little to do. As a result, some recordings are “imperfect”: she sometime stuttered or hesitated to recite, when she had difficulties grasping the metrical structure. In most cases, not she, but the texts given her to recite were responsible for her stuttering, for these Middle Bengali texts very often contained anomalies in meter. Probably the state of affairs at the time when these texts were composed was not entirely unlike this. The “imperfection” in the recordings is not a defect, but I intentionally allowed the performer to make mistakes, not to do it perfectly, so as to observe the actual practice.

The four most popular meters in Middle Bengali lyrics are: Paẏārˈ, Laghu Tripadī and Dīrgha Tripadī.

1. Paẏār 8 + 6 　　　　 WAV11K WAV11k

Miśrabr̥tta Paẏārˈ (Vaishnava Padavali, Hare Kr̥ṣṇa, Caṇḍīdāsˈ)

tomāre bujhāi bandhu ǀǀ tomāre bujhāi ǀ1aǀ^{^[xii]}

[tomare bujhai bondhu ǀǀ tomare bujhai]

ḍākiẏā ʃudhāẏa more ǀǀ hena jana nāi ǀ1bǀ

[ḍakia ʃuddhae more ǀǀ hæno jon nai]

anukhana gr̥he mora ǀǀ gañjaẏe sakale ǀ2aǀ

[onukhon grihe mor ǀǀ gɔñjɔe ʃɔkole]

niścaẏa jānio muñi ǀǀ bhakhimu garale ǀ2bǀ

[niʃcɔe janio muñi ǀǀ bhokhimu gɔrole]

In this case, the word bujhāi [bujhai] in the first line (1a) is counted as three beats, bu + jhā + i [bu + jha + i].

2. Laghu Tripadī (6 + 6 + 8)

Tripadī means “that which consists of three padas or verse units.” There are two kinds of Tripadī: Laghu (“light”) Tripadī and Dīrgha (“long”) Tripadī. Both Laghu Tripadī and Dīrgha Tripadī are often simply referred to as Tripadī, although their structural difference is clear-cut.

I begin with Laghu Tripadī, literally “the light Tripadī”, i.e. the short version of the verse containing three units. Its structure is 6 + 6 + 8.

Miśrabr̥tta Laghu Tripadī WAV11K WAV11K

gharera bāhire ǀǀ daṇḍe śatabāra ǀǀ tile tile āise yāẏa ǀ1ǀ

[ghɔrer bahire ǀǀ dɔṇḍe ʃɔtobar ǀǀ tile tile aiʃe jae]

mana ucāṭana ǀǀ niśbāsa saghana ǀǀ kadambakānane cāẏa ǀ2ǀ

[mon ucaṭon ǀǀ niʃʃaʃ ʃɔghono ǀǀ kɔdombokanone cae]

(Baiṣṇaba Padābalī, Hare Kr̥ṣṇa, Caṇḍīdāsˈ)

In the third unit of the first line, āise [aiʃe] is counted as two syllables, or two beats, i.e. āi + se [ai + ʃe].^{^[xiii]}

Another example of Laghu Tripadī (6 + 6 + 8)

Vidyāpati, Mitra-Majumˈdārˈ (SENˈ 1995, p.17) WAV11K

bāli bilāsini ǀ jatane ānali ǀ ramana karabi rākhi ǀ1ǀ

[bali bilaʃini jɔtone anoli rɔmono kɔrobi rakhi]

jaise madhukara ǀ kusuma na tola ǀ madhu piba mukhamākhi ǀ2ǀ

3. Dīrgha Tripadī (8 + 8 + 10)

Dīrgha Tripadī, meaning literally “the long Tripadī”, is the long version of the verse containing three units. Its structure is 8 + 8 +10.

Miśrabr̥tta Dīrgha Tripadī (Jñānˈdāsˈ)^{^[xiv]} WAV11K WAV11K

rajanī śāṅana ghana ǀǀ ghana deẏā garajana ǀǀ rimijhimi śarade bariśe ǀ1ǀ

[rɔjoni ʃaṅon ghɔno ǀǀ ghɔno dea gɔrojon ǀǀ rimijhimi ʃɔrode boriʃe]

pālaṅke śaẏana raṅge ǀǀ bigalita cīra aṅge ǀǀ ninda yāi manera hariṣe ǀ2ǀ

[paloṅke ʃɔyon rɔṅge ǀǀ bigolito cir ɔṅge ǀǀ nindo jai moner horiʃe]

The closed word-final syllables (marked in italics) are each counted as two beats.

There exist other meters than the four representative meters, Paẏārˈ, Laghu Tripadī and Dīrgha Tripadī, although they occur far less frequently.

Caupadī (verse with four units, 6 + 6 + 6 + 5) WAV11K

Ex. Rasamañjarī, pragalbhā adhīrā (SENˈ 1995, p. 23) Miśrabr̥tta

kona phūle bãdhu ǀǀ pāna kare madhu ǀǀ haẏe ele yadu ǀǀ poṙāte more ǀ1ǀ

[kon phule bõdhu ǀǀ pan kɔre modhu ǀǀ hɔe ele jodu ǀǀ poɽate more]

ālatā kajjala ǀǀ sindūra ujjala ǀǀ jāgiẏā bikala ǀǀ naẏana ghore ǀ2ǀ

[alota kɔjjol ǀǀ ʃindur ujjol ǀǀ jagia bikol ǀǀ nɔyon ghore]

Ekābalī (6+5) (Miśrabr̥tta) WAV11K

Ex.Rasamañjarī, Youbanˈ Kathanˈ (SENˈ 1995, p.24)

nārīra youbana ǀǀ baṙa duranta ǀ1aǀ

[narir joubon ǀǀ bɔṙo duronto]

śarīrera mājhe ǀǀ poṣe basanta ǀ1bǀ

ʃorirer majhe ǀǀ poʃe bɔʃonto]

binoda bināne ǀǀ bināẏe beṇī ǀ2aǀ

[binod binane ǀǀ binae beni]

puruṣe daṃśite ǀǀ poṣe sāpinī ǀ2bǀ

[puruʃe doṅʃite ǀǀ poʃe ʃapini]

bhārata-candrera ǀǀ bhāratībhoga ǀ3aǀ

[bharot cɔndrer ǀǀ bharotibhog]

youbanete kara ǀǀ youbana bhoga ǀ3bǀ

[joubonete kɔro ǀǀ joubon bhog]

Another example for Ekābalī (6 + 5) (Miśrabr̥tta)

Vidyāpati, Mitra-Majumˈdārˈ (SENˈ 1995, p.24)

cumbana naẏana ǀ kājara gelā ǀǀ

[cumbono nɔyono kajor gela]

damane adhara ǀ khaṇḍita bhelā ǀǀ

[dɔmone ɔdhoro khoṇḍito bhela]

pīna paẏodhara ǀ nakhara mandā ǀǀ

jani mahesara ǀ sikhara candā ǀǀ

na mukha bacana ǀ na cita thīre ǀǀ

kā̃pa ghana ghana ǀ sabe sabīre ǀǀ

This verse is composed in the Maithili language and its meter is a Maithili one, but in Bengal it was reinterpreted as a Bengali meter by Bengali readers.

Ekˈpadī + Tripadī Miśrabandha (8, 6 + 6 + 8) (Miśrabr̥tta) WAV11K

This verse consists of a verse/unit in 8 beats and Laghu Tripadī.

(Caṇḍīdāsˈ)

sai,

[ʃoi]

ki āra baliba tore ǀǀ

[ki ar bolibo tore]

kona puṇyaphale ǀǀ se hena bãdhuẏā ǀǀ āsiẏā milala more ǀǀ

[kon punnophɔle ǀǀ ʃe hæno bõdhua ǀǀ aʃia milol more]

The first word of this verse sai [ʃoi] is an interjection, a kind of Auftakt and is not counted in the meter.

Dbipadī^{^[xv]} + Tripadī Stabakˈ (8 + 8, 6 + 6 + 8)

Combination of Dbipadī (two units) and Laghu Tripadī.

Rhyme: A A, B B A

Ex. Annadāmaṅgalˈ, Mālinīmarma (SENˈ 1995, p.23) WAV11K

mālinī bujhila marma ǀǀ koṭāle jānāẏa dharma ǀǀ

[malini bujhilo mɔrmo ǀǀ koṭale janae dhɔrmo]

homa-kuṇḍa bali ǀǀ bujhi more chali ǀǀ sundarera ei karma ǀǀ

[hom kuṇḍo boli ǀǀ bujhi more choli ǀǀ ʃundorer ei kɔrmo]

Note that the closed syllables in word-internal position, such as the first syllables of marma [mɔrmo], dharma [dhɔrmo] and sundarera [ʃundorer], are NOT long; they are counted as one beat.

On the other hand, the compound homa-kuṇḍa is treated as two words, and homa- [hom] is counted as long (i.e. as two beats).

Dīrgha Dbipadī + Dīrgha Tripadī Miśrabandha (10 + 10, 8 + 8 + 10)

Combination of a long (dīrgha) Dbipadī (two units) and a Dīrgha Tripadī.

Rhyme: A A, B B A

Miśrabr̥tta. Ex. Annadāmaṅgalˈ, Bidyārˈ Ākṣepˈ (SENˈ 1995, p.23) WAV11K

kā̃de bidyā ākula kuntale ǀǀ dharā tite naẏanera jale ǀǀ

[kā̃de bidda akul kuntale ǀǀ dhɔra tite nɔyoner jɔle]

kapāle kaṅkaṇa hāne ǀǀ adhīra rudhira bāne ǀǀ ki haila ki haila ghana bale ǀǀ

[kɔpale kɔṅkono hane ǀǀ ɔdhir rudhir bane ǀǀ ki hoilo ki hoilo ghɔno bɔle]

The diphthong [oi] of haila [hoilo] is counted as one beat.

Pañcakalˈ-parbikˈ Caupadī (10 + 10 + 10 + 5) WAV11K

The verse consist of four units which are based on five beats.

Miśrabr̥tta. Ex. Rasamañjarī, Pīṭhamarda (SENˈ 1995, p. 24)

ramanī ratna ǀ sahenā ā̃ca ǀǀ ṭūṭaẏe agni ǀ paraśe kāca ǀǀ

[rɔmoni rɔtno ǀ ʃɔhena ãc ǀǀ ṭuṭɔe ogni ǀ pɔroʃe kac]

karite māna ǀ dibe nā sthāna ǀǀ dibe nā sthāna ǀ1ǀ

[korite man ǀ dibe na sthan ǀǀ dibe na sthan]

ki kare kṣobha ǀ sahe rāmāra ǀǀ abalā jāti ǀ mr̥du ākāra ǀǀ

[ki kɔre khobh ǀ ʃɔhe ramar ǀǀ ɔbola jati ǀ mridu akar]

jbalaẏe agni ǀ nahe se māna ǀǀ nahe se māna ǀ2ǀ

[jɔlɔe ogni ǀ nɔhe ʃe man ǀǀ nɔhe ʃe man ǀǀ]

I now give examples of Dalˈbr̥tta.

Dalˈbr̥tta Paẏārˈ (8 + 6, or 4 + 4 + 4 + 2)^{^[xvi]} WAV11K

āsāra̸ āśā ǀ āśā kebala̸ ǀǀ āśā mātrā ǀ holo ǀ1ǀ

citrera̸ kamala̸ yemona̸ ǀǀ bhr̥ṅga bhūle gelo ǀ2ǀ

khela̸bo bole ǀ phā̃ki diẏe ǀǀ nābāle bhūtalo ǀ3ǀ

ebāra̸

ye khelā khelāle māgo ǀǀ āśā nā purilo ǀ4ǀ

(Omitted final vowels are marked as a̸ )

[aʃar aʃa ǀ aʃa kebol ǀǀ aʃa matra ǀ hɔlo ǀ1ǀ]

[citrer kɔmol jæmon ǀǀ bhriṅgo bhule gælo ǀ2ǀ]

[khelbo bole ǀ phãki die ǀǀ nabale bhutolo ǀ3ǀ]

[ebar]^{^[xvii]}

[je khæla kælale mago ǀǀ aʃa na phurilo ǀ4ǀ]

The first unit of the second line [citrer kɔmol jæmon] irregularly contains six syllables.

śamana̸ damana̸ ǀǀ rābana̸ rājā ǀǀ rāmaṇa̸ damana̸ rāma̸ ǀǀ

śamana̸ bhabana̸ nā haẏa̸ gamana̸ ye laẏa̸ rāmera̸ nāma̸ ǀǀ

(Kr̥ttibāsˈ, Kiskindākāṇḍa, SENˈ 1995, p.20)

Actual pronunciation:

[ʃɔmon dɔmon ǀǀ rabon raja ǀǀ ramon dɔmon ram ǀǀ]

[ʃɔmon bhɔbon ǀǀ na hɔy gɔmon ǀǀ je lɔy ramer nam ǀǀ]

Another example of Dalˈbr̥tta Paẏārˈ (4 + 4 + 4 + 2)

Vaishnava Padavali, Locanˈdās. (SENˈ 1995, p.19) WAV11K

āra sunyācha ǀ ālo saï ǀǀ gorā bābhera ǀ kathā ǀǀ

[ar ʃunnacho alo ʃoi gora babher kɔtha]

koṇera bhetara ǀ kulabadhū ǀǀ kāndyā ākula ǀ tathā ǀǀ

[koner bhetor kulbodhu kanda akul tɔtha]

kisera sādhana ǀ kisera rāṙhana ǀǀ kisera haladi ǀ bāṭā ǀǀ

[kiʃer ʃadhon kiʃer raṙhon kiʃer hɔldi baṭa]

ā̃khira jale ǀ buka bhijhila ǀǀ bhyāsā gela ǀ pāṭā ǀǀ

[ãkhir jɔle buk bhijhilo bhæʃa gælo paṭa]

Dalˈbr̥tta Dīrgha Tripadī (8 + 8 + 10) WAV11K WAV11K

Locanˈdāsˈ (SENˈ 1995, p.19)

dīghala dīghala ǀ cā̃cara cūla ǀǀ

tāẏa diẏāche ǀ cā̃pāra phūla ǀǀ

kunda mālatīra ǀ mālā beṙhā ǀ jhuṭā ǀ1ǀ

[dighol dighol ǀ cãcor cul ǀǀ]

[tay diache ǀ cãpar phul ǀǀ]

[kundo maltir ǀ mala beṙha ǀ jhuṭa ǀ1ǀ]

candana mākhā ǀ gorā gāẏa ǀǀ

bāhu dolāiẏā ǀ calyā jāẏa ǀǀ

lalāṭa upara ǀ bhubana mohana ǀ phõṭā ǀ2ǀ

[cɔndon makha ǀ gora gay ǀǀ]

[bahu dolaiya ǀ colla jay ǀǀ]

[lɔlaṭ upor ǀ bhubon mohon ǀ phõṭa ǀ2ǀ]

In the second verse, the unit bāhu dolaiẏā [bahu dolaiya] contains an irregular number of syllables.^{^[xviii]} The pair of rhyming words gāẏa [gay] and jāẏa [jay] must be exceptionally counted as two beats, as in Miśrabr̥tta.

Dalˈbr̥tta Caupadī (8 + 8 + 8 + 8) WAV11K

Rāmˈprasādˈ Senˈ (SENˈ 1995, p.26)

mr̥tyuñjaẏa ǀ upayukta ǀ sebāẏa habe ǀ āśumukta ǀǀ

[mrittuñjoe ǀ upojukto ǀ ʃebay hɔbe ǀ aʃumukto ǀǀ]

ore, sakali sambhabe tāte ǀ paramātmāẏa miśāibe ǀǀ

[ore] [ʃɔkoli ʃɔmbhobe tate ǀ pɔromattay ǀ miʃaibe ǀǀ]

prasāda bale ǀ manabhāẏā ǀ chāṙa kalpa ǀ taru chāẏā ǀǀ

proʃad bɔle ǀ monobhaya ǀ chaṙo kɔlpo ǀ toru chaya ǀǀ]

ore, kāṭā br̥kṣera ǀ tale giẏe ǀ mr̥tyubhaẏaṭā ǀ ki eṙābe ǀǀ

[ore] [kaṭa brikkher ǀ tɔle gie ǀ mrittubhɔyṭa ǀ ki eṙabe ǀǀ]

Dalbr̥tta Ekˈpadī + Dbipadī (8 ǀ 8 ǀǀ 8) WAV11K

ebāra kālī ǀ tomāẏa khāba ǀ

[ebar kali tomae khabo]

khāba) khāba go dīna ǀ daẏāmaẏī ǀǀ

[khabo khabo go din dɔyamoi]^{^[xix]}

tārā) guṇḍayoge janma āmāra ǀ

[tara guṇḍojoge jɔnmo amar]^{^[xx]}

guṇḍayoge ǀ janamile ǀǀ se haẏe ye mā ǀ kheko chele ǀǀ

[guṇḍojoge jɔnomile ʃe hɔe je ma kheko chele]

Part III Kalābr̥tta, the prosody of Vaishnava lyrics

Kalābr̥tta, the third type of prosody, is completely different from Miśrabr̥tta and Dalˈbr̥tta in nature. It refers to the meters of Brajbhāṣā and Maithili lyrics as these were imported into Bengali for Vaishnava lyrics. These are defined as combinations of laghu (short) and guru (long) syllables. Although the Bengali language does not distinguish between long and short syllables, syllables may be artificially lengthened in recitation according to the metric schema. For instance, we have,

A unit consisting of five syllables, but sung in seven beats: (SENˈ 1995, p.25) WAV11K

⏑⏑ −⏑−⏑⏑ −⏑−⏑⏑ −⏑−⏑⏑ −⏑−

jaẏa) kr̥ṣṇa keśaba ǀ rāma rāghaba ǀ kaṃsa dānaba ǀ yātana ǀ1ǀ

[jɔyo kriʃno keʃobo ramo raghobo kɔṅʃo danobo jatono]

jaẏa) padmalocana ǀ nandanandanaǀ kuñjakānana ǀ rañjana ǀ2ǀ

[jɔyo pɔddolocono nɔndonondono kuñjokanono rɔñjono]

(The syllables which are pronounced as long are written in italics.)

Toṭakˈ ( ⏑⏑−, ⏑⏑–, ⏑⏑−, ⏑⏑− )

ramanī mani nāgara-rāja kabi ǀ

rati-nātha binindita cāruchabi ǀǀ

Reconstructed pronunciation

*[rɔmoni moni nagoro-rajo kobi]

*[roti-natho binindito caruchobi]

The syllables marked in italics are artificially pronounced as long in recitation. In Kalābr̥tta, the meters are treated purely on the level of the script. If a syllable is represented as long in the script, it is counted as long (guru). In other words, the method of counting Brajbhāṣā or Maithili metrics is mechanically applied to Bengali, even though this language has lost syllable length distinctions.

The following verse by Vidyāpati, which was originally composed in Maithilī meter, is recited as Laghu Tripadī (6 + 6 + 8) in Bengali. (SENˈ 1995, p.18) WAV11K

pahili pirīti ǀ parāṇa ā̃tara ǀ takhane aisana rīti ǀǀ

[pohili piriti pɔrano ãtoro tɔkhone oiʃono riti]

se ābe kabahu ǀ heri na heraṣi ǀ bheli nima tīti ǀǀ

sājani) jibathu sae pacāsa ǀǀ

sahase ramani ǀ raẏani khepathu ǀ morāhu tanhika āsa ǀǀ

Abbreviations

B. = Bengali

H. = Hindi

Skt. = Sanskrit

Bibliography

SENˈ, NĪLˈRATANˈ, 1995, Ādhunikˈ bāṃlā chanda. Prathamˈ Parba, Kalkātā: De’jˈ Pābˈliśiṃ, [History of Modern Bengali Prosody. First Part. Calcutta, Dey’s Publishing].