Unknown Like Ḹ: Complications in Sanskrit "Alphabet Poem" Stotras

Abstract

Beginning with Rūpa Gosvāmin (late fifteenth century-1564), at least four poets composed poems in the Sanskrit virudāvalī genre that included akṣamayī kalikā s, rhythmic compositions wherein each successive segment begins with the letters of the Sanskrit varṇakrama, from a to kṣa, in sequence. The fact that a number of the letters of the varṇakrama rarely or never occur in word-initial position, including the nasals ṅa,ña, and ṇa, the retroflex consonants, the aspirated consonants jha and tha, the anusvāra and visarga, and the vowels ...,..., and especially ..., creates potential obstacles in the composition of such poems. This paper examines the four known works with these "alphabet poems," and identifies six strategies that the poets employed in coining epithets beginning with such challenging letters: 1) the use of ekākṣarakoṣa s, dictionaries of single-letter words, 2) reference to the shapes of the written letter, 3) onomatopoeia, 4) grammatical derivations, 5) fictitious sandhi-based back-derivations, and 6) metalinguistic puns. The paper thus argues that the difficulty these letters present was, for these poets, a desirable feature of this structure, allowing them to display their learning, whether in service of a royal patron or in a context of Vaiṣṇava devotion.

Headnote

Beginning with Rūpa Gosvāmin (late fifteenth century-1564), at least four poets composed poems in the Sanskrit virudāvalī genre that included akṣamayī kalikā s, rhythmic compositions wherein each successive segment begins with the letters of the Sanskrit varṇakrama, from a to kṣa, in sequence. The fact that a number of the letters of the varṇakrama rarely or never occur in word-initial position, including the nasals ṅa,ña, and ṇa, the retroflex consonants, the aspirated consonants jha and tha, the anusvāra and visarga, and the vowels ...,..., and especially ..., creates potential obstacles in the composition of such poems. This paper examines the four known works with these "alphabet poems," and identifies six strategies that the poets employed in coining epithets beginning with such challenging letters: 1) the use of ekākṣarakoṣa s, dictionaries of single-letter words, 2) reference to the shapes of the written letter, 3) onomatopoeia, 4) grammatical derivations, 5) fictitious sandhi-based back-derivations, and 6) metalinguistic puns. The paper thus argues that the difficulty these letters present was, for these poets, a desirable feature of this structure, allowing them to display their learning, whether in service of a royal patron or in a context of Vaiṣṇava devotion.

(ProQuest: ... denotes non-USASCII text omitted.)

As in many languages around the world, there exist in Sanskrit many "alphabet poems"- that is, acrostic poems wherein the words or lines begin with each letter of the language, in the standard order in which they are arranged.1 In Sanskrit, of course, this is not really an "alphabet" (derived from alpha,beta . . .), but rather the basic Sanskrit varṇakrama or sequence of letters (a,ā,i,ī. . .).2 A full history of such Sanskrit acrostic poems deserves to be written but is well beyond the scope of the present paper. Rather, this paper focuses on the particular manifestation of such varṇakrama acrostic poems as one of the possible components of Sanskrit virudāvalī praise poetry, a broader genre of composition, discussed briefly below, which is largely associated with the Gauḍīya Vaiṣṇava theologian and poet Rūpa Gosvāmin (late fifteenth century-1564). In Rūpa's terminology, these poetic segments stringing together alphabetically arranged epithets are called akṣamayī kalikā s ("rhythmic passages consisting of [the letters]a to kṣa").3 This paper will consider four such poems. Three are by Gauḍīya Vaiṣṇava poets: Rūpa Gosvāmin, Raghunandana Gosvāmin, and Kṛṣṇaśaraṇa. Rūpa's and Kṛṣṇaśaraṇa's poems are addressed to the god Kṛṣṇa, while Raghunandana's is in praise of the saint Caitanya, ontologically identified with Kṛṣṇa. The fourth poet, Raghudeva of Mithilā, composed epithets in praise of an unnamed king, possibly Shāh Jahān.

Aside from some minor constraints of meter and loose end rhyme, in most cases the selection or coinage of epithets beginning with each letter of the varṇakrama is fairly straightforward. In fact, a quick glance at the first two segments of the akṣamayī kalikā in Rūpa Gosvāmin's Govindavirudāvalī seems to confirm that this would be very easy poetry both to compose and to understand:acyuta jaya jayaārtakṛpāmaya("Triumph, triumph, infallible Acyuta, full of mercy for the distressed"). But just as the letters Q, X, and Z can cause difficulty in the writing of English alphabet books for children, a number of letters in the Sanskrit varṇakrama create complications.

It is valuable to bear in mind that the Sanskrit varṇakrama(or the codification of letters in any language for that matter) is a conceptual construct. As W. Freeman Twaddell writes, "It has long been known to phoneticians and linguists in general that the sounds of speech, even within the narrowest restrictions of time and place, even within the usage of a single individual, present an almost infinite variety" (1935: 5).4 While Twaddell focuses on defining the concept of a phoneme, a sound unit that is involved in semantic distinction, the categorization of sounds by early Sanskrit linguists was extremely sophisticated phonetically but was not strictly phonemic in its criteria. Linguistic traditions often explicitly codify the infinite variety into a finite set of sound types, based on one or more sets of criteria, which may correlate more or less strictly with a set of written characters. The specifications and abstractions of a given codification of sound types and corresponding letters in a language can create or avert challenges to poets playing with that system in the composition of alphabet poems and the like. Consider the English letter N. As any student of Sanskrit soon becomes aware, this letter is involved a variety of articulations, roughly parallel to the first four nasal consonants of Sanskrit. The velar-nasal pronunciation in a word like sing, parallel to Sanskrit's ṅa, never occurs in word-initial position, but as no separate letter represents this sound, an author of an English alphabet poem need not be troubled by this particular sound. The phonetically grounded choice to distinctly categorize this sound in the Sanskrit varṇakrama, and the distinct corresponding written symbol, thus create a difficulty for poets composing akṣamayī kalikā s.

This paper will consider the four published (complete)akṣamayī kalikā s,5 identified below, examining how the authors deal with the following letters: ... anusvāra(aṃ), visarga(aḥ), the velar and palatal nasals ṅa and ña, the retroflex consonants ṭa,ṭha,da,dha, ṇa, and the aspirated consonants jha and tha. One may immediately recognize that very few common Sanskrit words begin with letters like jha,d...a, or tha; that no common words begin with ...,...,ṅa,ña, or ṇa;6 and that the letter ... is not really an attested part of the language outside of certain linguistic or Tantric contexts. One may also notice that anusvāra and visarga are presented above as following a, since they cannot be uttered without following a vowel, but that even such complexes are not very common in word-initial position.

My analysis of these poems identifies six strategies that the poets employ in constructing epithets with these unusual word-initial letters: 1) the use of Sanskrit lexicons, especially the inventive ekākṣara-koṣa s, dictionaries of single-letter words, 2) reference to the shapes of the written letter, 3) onomatopoeia, 4) grammatical derivation, 5) fictitious sandhi-based back-derivation, and 6) metalinguistic punning. These strategies show a distinction between these akṣamayī kalikā poems and similar alphabet poems found in Tantra texts, such as the Kumārīsahasranāma in the Rudrayāmala Tantra, which takes frequent recourse to the Tantric associations of various letters.

While the focus of the present paper is on how these poets managed to address the difficulties in such compositions, the question of why they would choose to do so deserves brief consideration.7 This question may take three forms: How did the akṣamayī come to be included as a common kalikā type in the development of virudāvalī poetry... Why did Rūpa Gosvāmin feel inclined to adopt the virudāvalī genre and even compose a textbook on its structure, setting a model for his own tradition, and why did a parallel tradition of virudāvalī composition develop in Mithilā... Finally, why would a given virudāvalī poet choose to include this particular alphabetic component, given that, as discussed below, the range of possible components to include in a virudāvalī is virtually limitless and an akṣamayī kalikā is not required.8 The first question is difficult to answer, taking us into the murky pre-Rūpa history of virudāvalī poetry, but further research may eventually shed some light.9

Regarding Rūpa's interest in the virudāvalī genre, a few things can be said. I have previously speculated (Buchta 2014: 349-51) that Rūpa may have first encountered the virudāvalī genre while visiting Caitanya in Orissa, where the development of the virudāvalī genre seems to have migrated from its antecedents in Andhra, partly in light of evidence from Cintāmaṇi Miśra's Vāṅmayaviveka. And Delmonico has suggested (1990: 249-55) that Siṃhabhūpāla's court poet, Viśveśvara Kavicandra, who is among those I identify (2014: 332, 336) as involved in developing the antecedents of the virudāvalī genre, may have been a direct ancestor of Rūpa's. Neal Delmonico highlights the strong influence of Siṃhabhūpāla's Rasārṇavasudhākara on Rūpa, an influence shown particularly starkly throughout Måns Broo's 2011 study of Rūpa's textbook on dramaturgy, his Nāṭakacandrikā. Thus, the virudāvalī genre, though not developed by Rūpa, may have been part of his family's legacy. Furthermore, this genre appears to have been avant garde within Sanskrit poetic composition during Rūpa's time, and his engagement with it can be included in what may be characterized as a project on the part of Rūpa Gosvāmin and his Gauḍīya Vaiṣṇava cohort in Vṛndāvana of adapting various aspects of Sanskritic learned culture into Kṛṣṇa's service. For Rūpa and his cohort, the ultimate value of all knowledge was the creation of devotion to Kṛṣṇa; therefore, they worked to develop their own Kṛṣṇa-centered presentations of literary and intellectual traditions. Other examples of this project include Rūpa's textbook on dramaturgy, mentioned above, as well as his primary work adapting rasa theory into bhakti theology in his Bhaktirasāmṛtasindhu and Ujjvalamaṇi, and his nephew Jīva Gosvāmin's Sanskrit grammar, the Harināmāmṛtavyākaraṇa.10 How the virudāvalī genre came into Mithilā is not yet at all clear and awaits further research.

The choice, by Rūpa Gosvāmin and the two other Gauḍīya Vaiṣṇava poets, to include these difficult compositions within their virudāvalī s may be related to this same project of adapting intellectual traditions into Kṛṣṇa's service. That is, the need to apply one's deep knowledge of Sanskrit scholarly traditions would be an attractive feature of akṣamayī kalikā s, not a cause of avoidance. To compose the difficult epithets in these poems requires, and thus puts on display, deep study of Sanskrit grammatical and lexicographical traditions. To merely translate these epithets obfuscates the deep scholarship that they incorporate. Thus, in my analyses below, I often unpack the grammatical and lexicographical details behind them in some detail. Much less can be said for our court poet in Mithalā, Raghudeva, beyond a general comment that it may have been valuable to display his learning to and confer prestige upon his royal patron, though his poem has a less scholarly flavor than the other three.

Before turning to the specific examples from the four poems, I will offer a brief introduction to the genre of virudāvalī poetry, wherein these akṣamayī kalikā s can be found, and to the four poets and their works, and will then discuss the variant terms for these akṣamayī kalikā s. I also present and translate the full akṣamayī kalikā from one of the four poets, Kṛṣṇaśaraṇa, to provide a sense of how the epithets coined with difficult initial letters fit into the larger panegyric context.

SANSKRIT VIRUDāvalī POETRY

Akṣamayī kalikā s appear as just one of the possible components in the composition of a Sanskrit virudāvalī. As the term suggests, a full virudāvalī is a compendium or 'row' (āvalī) of smaller poetic units primarily involving epithets (viruda) in praise of a king or a god. While the origins of virudāvalī poetry lie in royal panegyric, the oldest extant full example of the virudāvalī as an independent genre and the first extant full theoretical treatment of that genre are both devotional works dedicated to the god Kṛṣṇa written by Rūpa Gosvāmin, his Govindavirudāvalī and his (Sāmānya)-Virudāvalīlakṣaṇa, respectively, leading to the mistaken idea that he developed the genre.11

The virudāvalī genre of Sanskrit poetry involves a complex and elaborate structure, the full details of which are beyond the scope of this paper. I will offer here just a brief overview of the genre and how akṣamayī kalikā s fit within such poems. According to Rūpa's description and the model he presents,virudāvalī poems consist of three main components: 1) verses (śloka s) in classical meters such as śārdūlavikrīd...ita, which precede and follow 2)kalikā s, passages composed of units called kalā s with sustained patterns of rhythm and rhyme, followed finally by 3)viruda s, shorter strings formed on the same principles as kalikā s. Additionally, Rūpa notes that both kalikā s and viruda s should end with a word like dhīra('wise one') or vīra('hero'). Rūpa specifies that a complete virudāvalī composition should contain between five and thirty of such triadic units, with no fixed order for the various types of kalikā s.12

By contrast to the classical system of Sanskrit prosody wherein verses typically contain four quarters (pāda) with daṇd...a punctation after each pair of quarters, the kalikā and viruda structures described by Rūpa Gosvāmin are defined primarily by the rhythmic (tālaniyatā) segments called kalā s that comprise them.13 Other defining features of a given kalikā or viruda type include the alphabet acrostic of the akṣamayī kalikā s discussed here, as well as restriction on whether a heavy syllable has a long vowel and/or one of five types of consonant conjuncts (saṃyoga), or the sequential use of all seven case endings plus the vocative. In most cases,daṇd...a s are missing, allowing for a continuous flow through the entire kalikā or viruda, though sandhi is sometimes broken at the end of a kalā, as we will see in the example below. In practice,kalikā s and viruda s generally feature loose end rhyme after each kalā or set of kalā s.14 Rūpa enumerates forty-nine types of kalikā s in six categories, though he repeatedly reminds his reader that this is only a small sample of the possible variety.15 The akṣamayī kalikā s discussed here form one of the two kalikā subtypes listed under the kevalā kalikā or gadya category. While Rūpa includes these kalikā s in the gadya(nonmetrical) category and while his brief partial example in the Virudāvalīlakṣaṇa does not contain a consistent meter,16 his own full example in the Govindavirudāvalī, as well as the other examples I examine in this paper, are all in moric meters. The primary difference between kalikā s and viruda s is simply their length: a kalikā may consist of sixteen to sixty-four kalā s, while a viruda contains only two to ten kalā s.17 Otherwise, the categorization of kalikā types is used for viruda s as well. While the term virudāvalī highlights the viruda component, it is really the kalikā s that make up the defining bulk of virudāvalī poems.

FOUR AKṣAMAYĪ POETS AND THEIR VIRUDĀVALĪS

The four poems examined here include three devotional poems-two in praise of Kṛṣṇa and one in praise of Caitanya, regarded as ontologically identical to Kṛṣṇa-and one royal panegyric, possibly in praise of Shāh Jahān, composed from the sixteenth to the nineteenth century. As noted above, the earliest extant poem in the virudāvalī genre described here is Rūpa Gosvāmin's Govindavirudāvalī, included within the Stavamālā, a collection of Rūpa's shorter poetic works compiled by his nephew Jīva Gosvāmin. Rūpa's akṣamayī kalikā appears as the penultimate kalikā in the collection of twenty-eight śloka-kalikā-viruda triads,18 just as it is the penultimate kalikā type described in his Virudāvalīlakṣaṇa.

Of the four Gauḍīya Vaiṣṇava poets who wrote virudāvalī poems after Rūpa, two of them include akṣamayī kalikā s.19 One of these is Raghunandana Gosvāmin's Gaurāṅgavirudāvalī in praise of Caitanya.20 The text is not dated, but Raghunandana flourished in the early nineteenth century. He is described in William Adam's 1838 Third Report on the State of Education in Bengal as "the most voluminous native author I have met with" (p. 50). Raghunandana makes his indebtedness to Rūpa Gosvāmin's model explicit in his concluding verse, analogizing the connection between his poem and Rūpa's Govindavirudāvalī to the identity between Caitanya and Kṛṣṇa.21 The other Gauḍīya Vaiṣṇava akṣamayī kalikā is found in the Kṛṣṇavirudāvalī of the otherwise unknown Kṛṣṇaśaraṇa. Like Raghunandana, Kṛṣṇaśaraṇa explicitly acknowledges Rūpa's influence, describing his poem in the antepenultimate verse as sattamarūpānusāriṇī"following the most virtuous Rūpa." The editor Haridāsa Dāsa suggests that this Kṛṣṇaśaraṇa may have been a servant at the Govindadeva temple in Jaipur who lived until 1804, but he admits that there is no further evidence to support the identification (1944: 10 [preface]).

Parallel to, but seemingly not directly influenced by, the Gauḍīya Vaiṣṇava tradition of virudāvalī poetry, the genre experienced a flourishing in Mithila between the seventeenth and nineteenth centuries.22 Two of these poems have been published: Harideva's Jahāṅgīravirudāvalī in praise of the Mughal emperor and his younger brother Raghudeva's Virudāvalī in praise of a well-epitheted but unnamed royal patron. In the preface to Jagannātha Pāṭhaka's edition, G. C. Tripathi (1979: 1) notes a recent tradition identifying this patron as Shāh Jahān, which he claims "seems likely, but not certain." Harideva's work contains a letter-acrostic kalikā, but it only includes consonants and omits the tricky letters such as ṅa, so I do not consider it here. Raghudeva's work, on the other hand, does contain a complete akṣamayī composition.23

THE TERM AK&Sbdot;AMAYĪ AND ITS VARIANTS

So far I have referred to these poems as akṣamayī kalikā s. As noted above, the sense of this term is 'consisting of [the letters]a to kṣa'. In effect, this term is more closely parallel to "A-to-Z poems" than to "alphabet poems." However, two other terms are found as labels for these compositions in the four poems examined here:akṣaramayī('consisting of letters') and akṣamālā('a garland from a to kṣa'). In fact, even for Rūpa Gosvāmin's Govindavirudāvalī, only two of the five published editions and one of the three manuscripts I was able to examine had the label akṣamayī,24 the remainder containing the variant akṣaramayī. Likewise, the example in the single published edition of Rūpa's Virudāvalīlakṣaṇa, as well as the three published editions of Raghunandana's poem and the one published edition of Kṛṣṇaśaraṇa's poem all include the label akṣaramayī. Adding to the confusion, the two great modern Sanskrit-to-Sanskrit dictionaries, Tārānātha Tarkavācaspati's Vācaspatyam(1873-84: 1807) and Rādhākānta Deva's Śabdakalpadruma, in the edition revised by Vasu and Vasu (1886: 62), under their entries for the word kalikā, each block-quote nearly the entirety of the introduction from Baladeva Vidyābhūṣaṇa's commentary on the Govindavirudāvalī wherein he outlines the structure of the genre.25 The former has the reading akṣaramayī while the latter has akṣamayī.26 Meanwhile, Raghudeva's poem, in both published editions, employs the term akṣamālā.27 Thus, some rationale for why I have so far privileged the term akṣamayī is in order.

While the majority of sources, and the earliest sources available to us, read akṣaramayī, there are four factors suggesting that akṣamayī is the more likely original reading in Rūpa Gosvāmin's works. First, in the Virudāvalīlakṣaṇa, Rūpa initially provides a longer term for this kalikā type, fitting within a metrical structure:ādikṣāntāvali(i.e., [a-ādi]-[kṣaanta]-āvali: 'a string beginning with a and ending with kṣa'). This parallels the term akṣamālā discussed below. Still, the example that follows is labeled akṣaramayī in the sole published edition edited by Haridāsa Dāsa. Second, in Baladeva's commentary on the Govindavirudāvalī, he glosses the label as follows:akṣa(ra)mayīti|akārādyā kṣakārāntā prathamabhidā varṇyetyarthaḥ|28("The meaning of akṣa(ra)mayī is that the first type [of the kevalā kalikā category], which starts with the letter a and ends with the letter kṣa, is to be described"). This gloss, which parallels a gloss of Raghudeva's term akṣamālā discussed below, appears to break apart the compound a-kṣa, and does not account for the word akṣara at all. Thus, it makes more sense that Baladeva read the terms as akṣamayī rather than as akṣaramayī.

The third piece of evidence in favor of the term akṣamayī comes from Candraśekhara Bhaṭṭa's Vṛttamauktika, an encyclopedic work on prosody completed in 1619, which offers a detailed discussion of the virudāvalī genre. Throughout this discussion he cites Rūpa Gosvāmin's Govindavirudāvalī for the examples for every relevant kalikā type, in the end including the entirety of Rūpa's poem except for its opening and closing maṅgala verses. Candraśekhara consistently calls the kalikā type in question akṣamayī, including in metrical contexts where the variant akṣaramayī would destroy the meter, such as the following (Vinayasagar 1965: 262-64, 288):29

akārādikṣakārāntamātṛkārūpadhāriṇī|

viṣṇoḥ stutiparā seyaṃ kalikā'kṣamayī matā|| 8 ||

Having the form of the letters (mātṛkā) beginning with a and ending with kṣa,

Devoted to praise of Viṣṇu, this kalikā is known as akṣamayī.

Likewise, an earlier text on prosody, Cintāmaṇi Miśra's Vāṅmayaviveka, completed in 1573, less than a decade after Rūpa Gosvāmin's death, twice uses the term akṣāvalī in metrical passages that would not allow for a variant with akṣara(Kara 1973: 389, 412).30 This author knows Rūpa Gosvāmin, citing two verses from his Uddhavasandeśa(Kara 1973: 415), as well as attributing to Rūpa three kalikā examples not found in his extant corpus (Kara 1973: 393, 399). As I have argued previously (Buchta 2014: 347-51), it is more likely that Rūpa and Cintāmaṇi share a common source for their characterization of virudāvalī poetry than that Cintāmaṇi draws directly from Rūpa. Still, the theory of a common source supports a reading of akṣamayī in Rūpa's work.

Finally, as noted above, Raghudeva's kalikā is labeled akṣamālā, parallel to akṣamayī. And the commentator, Cakradhara Śarman Jhā, glosses this term with wording similar to Baladeva's gloss cited above:akārādikṣakārāntavarṇamāleyam akṣamālā"This garland of letters beginning with the letter a and ending with the letter kṣa is an akṣamālā." The term akṣamālā is also used for the garland of rudrākṣa beads worn by Śiva and used for japa by Śaiva devotees. In fact, this double meaning has a precedent in the Kulārṇavatantra 15.48-49: "The rosary (akṣamālā) is said to be of two kinds: created and uncreated. The created one is made of beads; the uncreated one [consists of] the syllables of the alphabet. It is called a-kṣa-mālā because it consists of the bead-like syllables from a to kṣa" (translation from Bühnemann 1992: 77).

It is plausible that Rūpa may have wanted to avoid the Śaiva associations with the term akṣamālā, and so used the term akṣamayī instead. This less familiar term in turn may have invited an intended "correction" to akṣaramayī. It is likely that by the time of the two later Gauḍīya Vaiṣṇava poets the term akṣaramayī had largely supplanted akṣamayī so that they only knew, and thus used, this later term.

tHe akṣa(ra)mayī kalikā from kṛṣṇaśaraṇa's kṛṣṇavirudāvalī

Before turning specific focus to the strategies employed by the four akṣamayī kalikā poets to coin epithets with difficult initial letters, I present and translate here a full example of the akṣamayī kalikā within its triadic structure (śloka-kalikā-viruda) from one of the virudāvalī s, Kṛṣṇaśaraṇa's Kṛṣṇavirudāvalī, to provide a sense of how the individual epithets discussed below fit into their larger context.31 This triadic unit occurs as the penultimate of twentyeight such triads, just as is the case for both of the other Gauḍīya Vaiṣṇava poets, Rūpa and Raghunandana, again showing the strong influence of Rūpa Gosvāmin's model within his tradition. The heading, however, shows the term akṣaramayī instead of akṣamayī, which, as noted above, appears to have taken over by this period.

A few characteristics of the poem presented here bear notice. First, there is no narrative or semantic coherence within the akṣamayī kalikā, nor within the larger triadic unit, nor across the entire Kṛṣṇavirudāvalī for that matter. Some epithets here describe Kṛṣṇa's beauty, such as his flowing locks of hair; others reference his various activities, such as his lifting of the mountain Govardhana; while yet others make theological claims about the salvific power of his name and so on. Beyond Kṛṣṇa as the single object of praise, it is the form in terms of the meter, end rhyme, and alphabetic organization, not the content, that provides unity to this kalikā, just as the organization of tripartite units centering around each kalikā, rather than any narrative, ties together the virudāvalī as a whole. Second, the preceding and following śloka s and the final viruda do not participate in the alphabetic structure of the kalikā. Rather, they serve merely as framing elements around the kalikā, despite the apparent centrality of the viruda component in the term virudāvalī. Finally, most of the epithets here, even if clever or charming, do not involve any particular difficulty in coinage. The epithet paṅkajalocana (lotus-eyed), for example, is commonplace in Sanskrit. The opportunities that the difficult letters present for these poets to display their knowledge appear by the by. The successful coinage of these difficult epithets allows the kalikā to achieve completeness and lends an element of scholarly playfulness to an otherwise straightforward composition.

akṣaramayī||

Consisting of Letters

taralitataralodyattārahārāvalībhir

valayitam atulaśrīnīlaśailāvabhāsam|

kucaghusṛṇaviliptaṃ gopasīmantinīnāṃ

kṣaṇam api mama lakṣyaṃ kṛṣṇavakṣaḥ purastāt|| 113 ||

Encircled by strings of bright pearls rising up from a swinging central gem, lustrous like the unparalleled Blue Mountain, smeared with saffron from the breasts of cowherd women-may Kṛṣṇa's chest be visible before me, even just for a moment.

avyayapada jayaārtiharāhvaya

indumukhodayaīśakṛpāśraya

utpalavibhramaūrddhvagavikrama

ṛddhidasaṃkatha...bhukulākṣata

...takayajñasakha ...vad anarthaka

eṇadṛśekṣitaaindramadakṣata

ojaḥpuñjaka aupamanindaka

aṃśubhir ujjvala aḥivakuṇd...ala

kañjavarekṣaṇa khaṇd...alabhakṣaṇa

garjitakuñjara ghusṛṇasupiñjara

ṅokṛtakalaruta calitālakavṛta

chalitāmbujabhavajala jottamarava

jhaṅkṛtakaṅka ṇañokṛtasurasana

ṭaṅkṛtabahusukha ṭhākṛtividhumukha

ḍiṇd...imaghoṣaṇa ḍhakkāravaraṇa

ṇārthagatasmaya taraladṛśidvaya

thutkṛtakuvalaya dīnajanāśraya

dharadhṛtinirbhaya navanītapriya

paṅkajalocana phaṇimadamocana

bakamukhadāraṇa bhavabhayatāraṇa

mañjulamaṇd...anayudhi ripukhaṇd...ana

rādhārañjana lalitadṛgañjana

vikasitamukhavaraśa śadharasundara

ṣaṭpadarucidhara sarasamanohara

haladharasahacarakṣapitamahībhara||dhīra|| 114 ||

[a] With your imperishable abode, triumph! [ā] With a name that removes misery,

[i] And lofty face like the rising moon, [ī] a foundation of mercy for Śiva,

[u] With the illusive beauty of a blue lotus, [ū] and upward stride,

[ṛ] With stories that bestow prosperity, [...] uninjured by hosts of demons,

[...] ...taka's friend in sacrifice, [...] without motive, just as the vowel...has no purpose,

[e] Beheld by doe-eyed women, [ai] a wound to Indra's pride,

[o] Heaped with vigor, [au] scorning any object of comparison,

[aṃ] Shining with beams of light, [aḥ] with earrings like [the two dots of] avisarga,

[k] With eyes like choice lotuses, [kh] eating a morsel,

[g] Making an elephant (Kuvalayāpīḍa...) roar, [gh] painted yellow with saffron,

[ṅ] Humming the soundṅo, [c] encompassed with flowing locks,

[ch] Tricking the lotus-born Brahmā, [j] roaring like the best of rainclouds,

[jh] With bracelets tinkling with the soundjham, [ñ] and a great tongue resoundingño,

[ṭ] Causing great joy to resound, [ṭh] with a face like theṭha-shaped full moon,

[ḍ] Resounding like ad...iṇd...imadrum, [ḍh] delighting in the sound of thed...hakkādrum,

[ṇ] Free of pride because of knowledge, [t] with a pair of flitting eyes,

[th] Spitting on blue lotuses, [d] a shelter for afflicted people,

[dh] Fearless while holding the mountain (Govardhana), [n] fond of fresh butter,

[p] Lotus-eyed, [ph] dispelling the pride of the snake (Kāliya),

[b] Rending the beak of Baka, [bh] delivering (devotees) from the fears of worldly life,

[m] With beautiful ornaments, [y] annihilating your enemies in battle,

[r] Delighting Rādhā, [l] with charming eye makeup,

[v] With a choice, blossoming face, [ś] beautiful as the hare-bearing moon,

[ṣ] Bearing the beauty of a bee, [s] captivating as a lake full ofrasa

[h] Companion of the plow-bearing Baladeva, [kṣ] quelling the earth's burden,

Wise one!

karṇe kampitakarṇikārakalikaḥ kandarpakelikriyā-

kalyākalyavikalpanātikutukī kaiśorakālakramaḥ |

kiñcit kuñcitakomalālakakulaḥ kādambinīkandalaḥ

kṛṣṇaḥ kelikalāpakīlitakacaḥ kaṃ vaḥ kriyāt kāmadaḥ|| 115 ||

With the bud of a karṇikāra flower swinging from his ear,

So curious about distinguishing between helpful and not

In the activity of love-play,

In the course of his early youth,

With locks of soft hair, slightly curly,

Rumbling like a bank of clouds,

May Kṛṣṇa, with hair untied during the art of love-play,

Bestower of desires, bring joy to you.

rādhāgurutara-bādhāśatahara

riṅgāśritanaṭa raṅgāhṛtapaṭa

nāthāsukhaśatam āsūdaya hita||dhīra|| 116 ||

Remover of hundreds of Rādhā's heaviest pains

With dance infusing your every movement, stealing [the gopīs'] clothes for play,

O benevolent lord, destroy [my] hundreds of sorrows!

Wise one!

STRATEGIES IN THE COMPOSITION OF AK&Sbdot;amayī KALIKĀ s

As noted above, a number of letters create complications for the composition of akṣamayī kalikā s: the vowels ...,..., and ..., the anusvāra and visarga, the aspirated consonants jha and tha, the nasals ṅa and ña, and the retroflex consonants. How, then, is a poet to compose such a poem including epithets starting with each of these letters... Of course, having command over a vast vocabulary, perhaps with lexicons on hand to assist, is helpful. For the letter jha, for example, Rūpa Gosvāmin uses the solidly attested if not particularly common word jhaṣa to coin the bahuvrīhi compound jhaṣavarakuṇd...ala'with makara[-shaped] earrings'.32 But such attested words will only take a poet so far. In the remainder of this paper, I will examine how the four poets identified above tackled these tricky contexts in the compositions of their poems, identifying the six primary strategies noted above: 1) the use of ekākṣarakoṣa s, 2) reference to the shapes of the written letter, 3) onomatopoeia, 4) grammatical derivation, 5) fictitious sandhi back-derivation, and 6) metalinguistic punning. The examples from the four poets, discussed comprehensively below, are laid out in Table 1.

THE USE OF EKĀkṣARAKO&Sbdot;a s

Sanskrit poets and their audiences often delight in wordplay. Sanskrit scholars, going back at least to Puruṣottamadeva in the twelfth century, have facilitated such word play by compiling numerous (often inventive) lexicons, including ekākṣarakoṣa s (lexicons of monosyllabic words). One might expect these poets to rely heavily upon such resources, but as will be shown, it is only Raghudeva who uses them extensively.

Among our four poets, Rūpa Gosvāmin appears quite hesitant to avail himself of the simple solution of relying on ekākṣarakoṣa s. Aside from his use of ṭha to mean 'full moon', discussed below as a reference to the shape of the letter, there is only one other place where he might be understood to rely on an ekākṣara word, his epithet ...bhudayāpara, if ...bhu is read as a compound: either ...-bhu or ṛ-ṛbhu, as I suggest below. In his commentary Baladeva Vidyābhūṣaṇa interprets the epithet as meaning "intent on [showing] mercy to the gods" (...bhuṣu deveṣu dayāpara he). His justification for this interpretation begins with a citation of Amarakoṣa 1.1.16:ādityā ...bhavo'svapnāḥ, listing these three words as synonyms, each meaning 'gods'. Note, however, that this is not the actual reading of the passage, which instead has short ṛ, presenting the fairly common noun ṛbhu.33 Perhaps aware of the suspicion his citation might raise, Baladeva goes on to offer an etymological derivation of the word, a variant of a common traditional etymology of ṛbhu. He begins by citing Pāṇini's Aṣṭādhyāyī 3.2.179 to claim that the affix ...u(a short u affixed after a verb to create a nominal stem) can be added after the verb√bhū in the formation of a name.34 Baladeva then indicates that the sense of ... here is the mother of the gods, i.e., Aditi, and cites Amaracandrakavi's Ekākṣaranāmamālikā(2cd:. . . ṛ ... jñeye devadānavamātarau"ṛ and ... should be understood as the mothers of the gods and of the demons [respectively...]"). That is, the word ...bhu is taken etymologically to mean 'born from the mother of the gods', wherefrom the conventional meaning 'god' derives. The citation Baladeva provides, however, appears to indicate instead that ṛ means the mother of the gods while ... means the mother of the demons, though the Medinīkoṣa does explicitly list 'mother of the gods' (devāmbāyām) as a meaning of ....35 The complications of Baladeva's analysis aside, his citation of an ekākṣarakoṣa can point to other analyses, assuming that Rūpa was aware of these meanings attributed to ṛ and .... It may be that Rūpa used ...-bhu to mean 'those born from the mother of the demons (i.e., the demons themselves)' intending to describe Kṛṣṇa as 'full of mercy for demons', a reference to stories such as Kṛṣṇa's liberation of Pūtanā.36 Or ...bhu here may be read as a dvandva compound (ṛ-ṛbhu) so that the epithet ...bhudayāpara means 'intent on [showing] mercy to the gods and their mother'.

A similar question of analysis arises for Raghunandana's epithet for Caitanya,...ṣigaṇastuta. Baikuṇṭhanātha Ghoṣāla's Bangla translation appears to interpret ...ṣi here as just an alternate spelling of ṛṣi so that the epithet means "praised [or 'praiseworthy' in the translation] by the hosts of sages" (he ṛṣigaṇera stabanīya), though such an alternate spelling is not otherwise attested, even in dvirūpakoṣa s (lexicons of words with two spellings), to my knowledge. If this was Raghunandana's intention, perhaps he took Rūpa's ...bhu as an alternate spelling of ṛbhu, thus providing precedent. But I suspect that he took Rūpa's ...bhu as a compound as I have done above and formed his own parallel dvandva compound ṛ-ṛṣi, so that the epithet means 'praised by the mother of the gods and by hosts of sages'. Indeed, Haridāsa Dāsa accounts for such a meaning in his Sanskrit commentary, though he takes ...ṣi as ...-ṛsi(...kāro devamātā ṛṣigaṇāśca taiḥ stuta"praised by them: the letter ..., i.e., the mother of the gods, and the hosts of sages").

Neither Kṛṣṇaśaraṇa's nor Raghudeva's epithets beginning with ... appear to involve dvandva compounds, though they seem to take ... in opposite senses. Kṛṣṇaśaraṇa calls Kṛṣṇa ...bhukulākṣata, 'uninjured by hosts of demons', deriving ...bhu much as Baladeva did, but taking ... as 'mother of the demons'.37 Raghudeva, on the other hand, calls his patron ...bhavamaṇd...ana, which seems to mean 'an ornament to the gods', or 'having a god on his ornament'. The commentator Cakradhara takes this to mean "having Gaṇeśa as his ornament" (...bhava gaṇeśo maṇd...anaṃ bhūṣaṇaṃ yasya tādṛśa), with ... referring to Śiva, citing an ekākṣarakoṣa to that effect (...kāras tripurāntakaḥ"The letter ...[means] destroyer of the three cities [i.e., Śiva]").38 It is noteworthy that in these cases, once ṛbhu is understood as a compound ṛ-bhu, with ṛ meaning the mother of the gods and bhu understood as a derivate of the verb√bhū, parallel compounds, substituting ... for ṛ or including alternate derivates from √bhū such as bhava, become possible.

All three poets after Rūpa appeal to ekākṣarakoṣa meanings for the letter ṇa, which is typically identified with positive meanings such as 'joy' and 'knowledge'. This is not surprising for the Gauḍīya Vaiṣṇava poets, in light of the common practice of attributing meaning to the two syllables in the name Kṛṣ-ṇa.39 Raghunandana describes Caitanya as ṇātmakavigraha'whose body consists of joy', perhaps.40 Kṛṣṇaśaraṇa describes Kṛṣṇa with the epithet ṇārthagatasmaya. Haridāsa Dāsa, in his Sanskrit commentary on Kṛṣṇaśaraṇa's poem, cites an ekākṣarakoṣa to the effect that ṇa means 'knowledge' (ṇakāraḥ kathyate bodhaḥ), and in his Bangla translation renders this compound to mean "embodying knowledge and free of pride" (tumi bodhasvarūpa o garvarahita). I would suggest that a rendering that ties the elements of this compound together more closely might be preferable, such as 'free of pride because of knowledge'. Raghudeva likewise uses an ekākṣara meaning of ṇa, describing his patron as one 'whose [every] fragment of misfortunate is destroyed by knowledge' (ṇahatavipallava).

Raghudeva, in fact, goes much further in his appeal to ekākṣarakoṣa meanings, drawing on them for all the most difficult letters. He describes his patron as ...caraṇasevana('serving the feet of ... [the mother of the gods...]'),...citasadguṇa('having accumulated good qualities from ... [his mother...]'),ṅacayasurasamaya('filled with a taste for many sensual objects'), ñasamarpitadhana('bestowing wealth upon singers'), and thataṭāhṛtagaja('bringing [his] elephants to the slope of the mountain').41

REFERENCE TO THE SHAPES OF LETTERS

As noted above, Rūpa Gosvāmin uses ṭha in the sense of 'full moon', a meaning commonly found in ekākṣarakoṣa s. The connection of this meaning to the shape of the written letter in the Nāgarī script warrants considering this strategy separately, particularly as this strategy is employed by the two later Gauḍīya Vaiṣṇava poets. Rūpa addresses Kṛṣṇa straightforwardly as ṭhanibhānanavara('with a fine face like the full moon'). Raghunandana's epithet is particularly striking: Caitanya is ṭhajayinakhāñcala('whose fingernail tips defeat ṭha'). A reference to the full moon would not make sense in describing the distal edge of Caitanya's fingernails. However, if one considers the shape of the letter ṭha in the Bangla script, one can understand this as a reference to the crescent moon, not the full moon. (See Table 2 for the shapes of ṭha in Nāgarī and Bangla.) Finally, Kṛṣṇaśaraṇa makes the reference to the shape of the letter more explicit in his epithet:ṭhākṛtividhumukha('with a face like the ṭhashaped moon, i.e., a full moon'). He likewise references the shape of visarga (written almost like a colon, with two dots or small circles) without, to my knowledge, any corresponding meaning in ekākṣarakoṣa s. Kṛṣṇa, decorated with an earring dotting each ear, is aḥivakuṇd...ala('with earrings like [the two dots of] the visarga').42

ONOMATOPOEIA

Just as the shape of letters was referenced in the composition of these poems, their sounds also played an important role, with many instances of either direct or indirect onomatopoeia, especially for the letters ṅa,jha,d...ha, and tha. Most of the words used by these poets were already conventionalized, through attestation and/or inclusion in lexicons, but their onomatopoeic quality was still clear and widely acknowledged. By indirect onomatopoeia, I refer to the names of drums:d...amaru,d...hakkā, and d...iṇd...ima, which are understood to derive from the sound they make.43 Each of the poets includes the names of at least one of these drums, not surprising for the Gauḍīya Vaiṣṇavas, given their emphasis on communal singing of Kṛṣṇa's names.44 Rūpa's epithet d...amaraghaṭāhara('removing massive tumult') likewise involves the similarly onomatopoeic word d...amara'tumult'.

For the letter ṅa Rūpa and Raghunandana after him both rely on one of the most foundational sources, the Dhātupāṭha, where √ṅu(in the sense '[making] sound'śabde) occurs as the only verb beginning with ṅa. I have never seen the verb attested in a nongrammatical context outside of these poems, but the root is included in all but one of the Dhātupāṭha s in Palsule's concordance (1953: 41).45 Rūpa derives the participle ṅuta for his epithet ṅutamuralīrata ('immersed in your resounding flute'), while Raghunandana derives the verbal noun ṅuti to describe Caitanya as ṅutijitakokila('defeating the cuckoo with your song'). Their uses of this root again will be discussed below as examples of grammatical derivation. I suspect that Kṛṣṇaśaraṇa may have followed the onomatopoeic spirit of Rūpa's example, without awareness of this obscure verb root, in his epithets for Kṛṣṇa ṅokṛtakalaruta('with a humming sound made like ṅo') and ñokṛtasurasana('with a great tongue making a sound like ño').46

Each of the Gauḍīya Vaiṣṇava poets employs the trope of the object of praise "spitting on" someone or something to express superiority, with the onomatopoeic thūtkṛta(or thutkṛta in the two later poems). In Rūpa's epithet, Kṛṣṇa is 'spitting on wagtails' (thūtkṛtakhañjana) with his flirtatious flitting eyes. For Raghunandana, 'a mere bit of devotion to whose [= Caitanya's] feet spits on liberation' (thutkṛtamuktikadarapadabhaktika). And for Kṛṣṇaśaraṇa, Kṛṣṇa is 'spitting on blue lotuses' (thutkṛtakuvalaya).

Additionally, Raghunandana describes Caitanya as jhanaditinūpurañoṅūyākara('making abundant music with his anklebells [tinkling with the sound]jhanat'). This epithet will be discussed further below. Likewise, Kṛṣṇaśaraṇa describes Kṛṣṇa 'with bracelets tinkling with the sound jham' (jhaṅkṛtakaṅkaṇa). Kṛṣṇaśaraṇa also uses an onomatopoeic word in an extended, non-sound-related sense. He describes Kṛṣṇa as ṭaṅkṛtabahusukha. Normally,ṭaṅ means the 'twang' of a bow. Perhaps the sense of this epithet could be rendered as 'causing great joy to resound.'47

Finally, Raghudeva incorporates two rather novel examples of onomatopoeia. He describes his patron as aḥkāritanṛpa('causing [rival] kings to make [the gasping sound] aḥ'). He also uses the sound d...haṇad...haṇa(otherwise unknown to me) to derive the epithet d...haṇad...haṇitāhita. According to Cakradhara,d...haṇad...haṇa is a sound of grief, so Raghudeva's patron is described as "one for whom those not beneficent to him, i.e., his enemies, make the grieving sound d...haṇad...haṇa" (d...haṇad...haṇitāḥ śokaśabdaṃ kurvāṇā ahitāḥ śatravo yasya tādṛśa).

GRAMMATICAL DERIVATION

As shown above,√ṅu, the sole verbal root beginning with the letter ṅa, was used by Rūpa Gosvāmin and Raghunandana Gosvāmin for this tricky letter. But no root whatsoever begins with the letter ña. Nevertheless, through reduplication, forms of√ṅu can be derived beginning with the letter ña.48 Perhaps inspired by forms listed in Dhātapāṭha commentaries, both Rūpa and Raghunandana use intensive derivatives of√ṅu, which show such reduplication.49 Thus, Rūpa addresses Kṛṣṇa as ñoṅūyitadala('making a leaf whistle to resound loudly'). Raghunandana is best read not as creating a separate epithet for the letter ña, but as joining two kalā s together into the longer compound epithet noted above: jhanaditinūpura-ñoṅūyākara('making abundant music with his anklebells [tinkling with the sound]jhanat').50 Though only these two words involve such derivation, the strategy is distinct enough to warrant separate enumeration.

SANDHI-BASED BACK-DERIVATION

The anusvāra and especially the visarga present challenges for composing akṣamayī poems. Neither can truly occur at the beginning of a word, as they inherently follow vowels, but even in their conventionalized form occurring after a, very few common words begin with aṃ and no common words begin with aḥ.51 These letters do occur in ekākṣarakoṣa s, but perhaps surprisingly none of these poets availed themselves of such an option. We saw Kṛṣṇaśaraṇa and Raghudeva deal with aḥ cleverly by utilizing the written shape and onomatopoeia, respectively. Rūpa Gosvāmin, on the other hand, and Raghunandana following him, instead employed what I would call a sandhi-based back-derivation. Rūpa's epithet for Kṛṣṇa is aṣṭāpadapaṭa('with golden clothing'). There is, in fact, no visarga here, but Rūpa appeals to an imagined underlying visarga that could have been replaced by ṣ following the usual sandhi rules, as though aṣṭāpada were originally aḥ-ṭāpada. Likewise, Raghunandana addresses Caitanya as one 'about whom [every] critique has vanished' (astagadūṣaṇa), similarly imagining asta as aḥ-ta.

The anusvāra is not as problematic, as words such as aṃsa('shoulder'),aṃśuka('clothing'), and aṃśu('ray'), used by Rūpa, Raghunandana, and Kṛṣṇaśaraṇa respectively, are common in the language. Still, Raghudeva addresses his patron as antaruditakṛpa('with compassion welling up internally'). Whether Raghudeva similarly imagined a back derivation where aṃ-tar becomes antar or whether he would justify this epithet based on the use of the written anusvāra symbol as a shortcut for representing all nasals is not clear.

METALINGUISTIC PUNNING

In the examples discussed thus far, these poets have drawn on their knowledge of the linguistic scholarship in Sanskrit intellectual traditions. A final strategy employed by the three Gauḍīya Vaiṣṇava poets engages directly with the Sanskrit grammatical tradition with epithets for perhaps the two most difficult letters,... and ..., that involve metalinguistic puns.52

The short vowel ... has had a dubious status in the Sanskrit language going back all the way to Kātyāyana's (ca. third centurybce)Vārttika s on Pāṇini's grammar. A vārttika on the second sūtra in the Akṣarasamāmnāya, where ... is listed among the letters of the language, provides three reasons for this inclusion: "The mention of the letter ... is for the sake of [names given by] one's own free choice, for imitating [the pronunciation of] someone without the ability [to enunciate properly], and for [rules pertaining to] prolated vowels and the like" (...kāropadeśaḥ yadṛcchāśaktijānukaraṇaplutyādyarthaḥ; Kielhorn 1880: 19). Based on this vārttika, Patañjali's Mahābhāṣya offers an elaborate discussion of the propriety of the inclusion of ..., and mentions ...taka both as a freely chosen name (i.e., a name without etymology) and as a mispronunciation of a name ...taka.53 Aside from such names and mispronunciations, the letter ... is normally found in Sanskrit only in words like k...pta and cak...pe for which, together with forms like kalpate, one may posit a verbal root √k...p. However, in Pāṇini's Dhātupāṭha, the root is listed as kṛpŪ, and it is cited in the Mahābhāṣya as kṛpA.54 Aṣṭādhyāyī 8.2.18 (kṛpo ro laḥ: "l is the substitute for the r of [the root]kṛp") then provides for the proper substitution to obtain forms such as k...pta or kalpate.

Rūpa thus turns the difficulty of finding a word starting with ... into an advantage, providing a name of Kṛṣṇa that comments on the very rarity of the letter in question through a metalinguistic pun. After all, the first extant attestation of word-initial ... is in the compound ...-kāra, when Patañjali asks of Pāṇini's Akṣarasamāmnāya, "Why is there mention of the letter ......" (...kāropadeśaḥ kimarthaḥ). Thus, Rūpa addresses Kṛṣṇa as ...diva kṛpekṣita. Kṛṣṇa is ...T-iva,55 like the short ..., insofar as he is kṛpekṣita. On one reading, the compound kṛpekṣita is kṛpA-īkṣita, 'seen in the verb root√kṛp'. On the other reading, the compound is kṛpā-īkṣita 'seen through [his] mercy', as Kṛṣṇa can be seen only when he bestows his mercy upon a devotee. Raghunātha creates a similar pun to address Caitanya:...diva ṛtāmita. Just as short ... goes to the state of being ṛ(ṛtām ita) in the Dhātupāṭha, Caitanya is honest and immeasurable (ṛta-amita).

Kṛṣṇaśaraṇa stays within the metalinguistic realm, but does not incorporate a pun into his epithet beginning with .... He addresses Kṛṣṇa as ...takayajñasakha('...taka's friend in sacrifice'). It appears that Kṛṣṇaśaraṇa here draws upon Kātyāyana's example, presenting ...taka as a generic name, much as Devadatta is used in philosophical literature, here serving as a kind of "John Q. Devotee."

If the standing of short ... is dubious, that of long ... is even more so. Although the vowel is acknowledged as far back as the Kātantra(no later than ca. 400ce),56 it is basically unattested, at least until the proliferation of ekākṣarakoṣa s beginning in the twelfth century. Thus, Rūpa addresses Kṛṣṇa as ...vad alakṣita("unperceived like ..."). Baladeva reconciles this with Rūpa's previous epithet by adding the comment "without your mercy" (kṛpāṃ vinā). Both Raghunandana and Kṛṣṇaśaraṇa similarly begin their epithets with ...vad. For Raghunandana, Caitanya is asādhita('not effected', i.e., eternal), just as ... does not come about as a result of sandhi.57 For Kṛṣṇaśaraṇa, Kṛṣṇa is anarthaka('without motive') just as ... has no meaning or purpose in the language.

TWO MISCELLANIOUS EPITHETS

Two additional epithets formed in ways that do not fit any of these larger strategies deserve mention. For the letter ṇa, where all other poets had drawn upon ekākṣarakoṣa s, Rūpa's epithet involves what might be thought of as a Prakritization, spelling nakhara with an initial retroflex ṇa:ṇakharadhṛtācala('holding the mountain [Govardhana] with your fingertip'). Raghunandana, in what may again best be taken as a long compound epithet spanning two kalā s, calls Caitanya d...amarudhṛgīśvarad...huṇd...hyakṛpābhara('full of mercy sought by the lord who holds a d...amaru[Śiva]').58 The potential participle d...huṇd...hya here derives from√d...huṇd...h, a non-Pāṇinian verbal root found only in the Dhātupāṭha s in the traditions of Kāśakṛtsna and Vopadeva (Palsule 1953: 166). Such Prakritization and appeal to non-Pāṇinian grammatical traditions, however, are not found elsewhere in these poems.

CONCLUSION

The poems explored in this paper are clearly not simple pedagogical acrostics. Rather, they show a level of complexity, involving even grammatical inside jokes, that would make their pedagogical use untenable. In fact, these complexities have caused minor problems in the transmission of these texts. Rather than serving a purpose to educate young students, these poems display playful pāṇd...itya.

Each of these poets wrote their akṣamayī kalikā s within the inherited structures of Sanskrit virudāvalī poetry. Exactly how, and more importantly why,akṣamayī kalikā s came to be included as a common component of virudāvalī poetry is not fully clear. There may well have been pedagogical motivations behind the antecedents of this poetic genre, as the sāptavibhaktikā kalikā type using all of the Sanskrit case endings seems possibly to suggest. But whatever the origins behind the akṣamayī kalikā structure, each of these poets also made a conscious choice to include this type in their virudāvalī. And each approached the composition of these poems with scholarly complexity. Bear in mind that the three other currently available Sanskrit virudāvalī poems do not feature complete akṣamayī kalikā s. We may never know if Jīva Gosvāmin's incomplete (in extant form)Gopālavirudāvalī ever had, or would have had, an akṣamayī kalikā, while Viśvanātha Cakravartin's Nikuñjakelivirudāvalī omitted this kalikā type. Even more to the point, Harideva's Jāhaṅgīravirudāvalī did include a version of this kalikā type, but specifically omitted vowels (thus avoiding the difficulties associated with ...,..., and ...) and difficult consonants such as ṅa and ña.

Each of the four poets studied here approached the composition of these poems with their own preferences. Rūpa largely avoided ekākṣarakoṣa meanings, while Raghudeva relied on them almost exclusively. Kṛṣṇaśaraṇa, more than any of the other poets, used onomatopoeia extensively.

But there were also commonalities between these poets. All, for example, used the names of drums for the letters d...a and/or d...ha. Rūpa Gosvāmin's influence on the later Gauḍīya Vaiṣṇava poets, Raghunandana and Kṛṣṇaśaraṇa, explicitly acknowledged by them as noted above, can be clearly and consistently seen, from referencing the shape of the letter ṭha, to the sound of spitting for the letter tha, to the metalinguistic puns involving ... and .... This influence is especially prominent in the case of Raghunandana, who follows Rūpa's use of√ṅu in both its basic and intensive forms. Raghunandana explicitly claims that his poem shares a kind of ontological identity with Rūpa's, like Caitanya's with Kṛṣṇa. And yet he shows novelty as well, playing off the shape of the Bangla ṭha, distinct from its Nāgarī counterpart.59

These poets still found value in engaging the "prestige economy of Sanskrit" (Pollock 2009: 59-74), alive enough in their respective times and places. For Raghudeva, we might understand his poem infused with scholarly learning as meant in part to reflect well on his patron, an argument that could be made more robustly if only he told us who his patron was. For the Gauḍīya Vaiṣṇava poets, on the other hand, their poems can be better understood in the larger project of appropriating traditional Sanskrit knowledge systems and subordinating them to the ultimate goal of bhakti(devotion) to Kṛṣṇa. This project involved major works like Rūpa Gosvāmin's Bhaktirasāmṛtasindhu, which reworked rasa theory into a tool for an emotion-based theology of bhakti, and Jīva Gosvāmin's Harināmāmṛtavyākaraṇa, a grammar that replaced established technical terms with names of (or associated with) Kṛṣṇa. As perhaps a smaller intervention, Rūpa brought the virudāvalī genre to life in the service of praising Kṛṣṇa and brought some delightful wordplay along with it.

Sidebar

Author's note: I wish to thank Alex Roy and Scott AnderBois for talking through some of the linguistics issues involved in the paper with me, and Catherine Nelli, Srinaath Perangur, Stephanie Jamison, and the anonymous reviewers for helpful comments on earlier drafts. I also wish to thank Sourish Das at the Bhaktivedanta Resarch Center in Kolkata, who, with Catherine Nelli's assistance, arranged to provide me copies of the manuscript that the center had digitized. Finally, I thank Satoshi Ogura, who headed the Digitization of the Matsya Project Microfilms, with funding from the Japan Society for the Promotion Science and the Information Resources Center at the Research Institute for Language and Cultures of Asia and Africa, Tokyo University of Foreign Studies, as well as Kiyokazu Okita who helped me to access the digitized Matsya Project manuscript.

Footnote

1. See Buchta 2023: 121 for a brief selection of examples of alphabet poems in various languages.

2. I use the word "letter" here and throughout the paper broadly, following Allen (1953: 15-16), not in the sense of the written symbol, but rather in the sense of the abstracted "sound-unit" as part of a codified phonetic inventory of the language. With the exception of certain consonant conjuncts, including kṣa discussed below, the Brahmibased Indic scripts involved a one-to-one (or in the case of vowels, two-to-one) correspondence with the sound-units included in the varṇakrama, allowing for a close identification of sound and script.

3. The virudāvalī genre discussed here should be distinguished from its Andhra-based antecedent, the birudāvalī (with initial b instead of v), treated by Pandiri Sarasvati Mohan (1972: 114) and Jamal Jones (2018: 92-101), and the possibly unrelated Hindi virudāvalī poetry treated by Allison Busch (2018). Rūpa Gosvāmin's dates are considered by Delmonico (1990: 279-80). See below for variant terms for these poems. While kṣa is typically not counted as a separate individual letter in the major Sanskrit grammatical traditions, being rather a conjunct of ka and ṣa, it is often included as the final letter in lexicons going back at least to the twelfth century (Vogel 1979: 326, 330, 351, 354, 356, 367, 376), in Tantric texts (Bühnemann 1992), and in the Buddhist Lalitavistara. The use of a distinct written symbol to represent kṣa presumably played a central role in its eventual inclusion as a distinct letter in some varṇakrama s; Allen (1953: 78-79) notes that its realization is in some cases phonologically distinct.

4. See also Allen 1953: 9-20.

5. As noted below, one published virudāvalī poem, by Harideva, contains an alphabetically arranged kalikā, but it omits vowels and difficult consonants.

6. As the ...gveda-Prātiśākhya notes (ed. Shastri 1931: 18) (in the presumably interpolated introductory passage),padādyantayor na ...-kāraḥ svareṣu("Among the vowels, the letter ... does not occur at the beginning or end of a word"). Likewise, Monier-Williams notes for ṅa that, "No word in use begins with this letter" (1899: 380), and the entries given under the headings of a number of these letters is less than a page.

7. See also Buchta 2023: 122-23 for further discussion of the potential motivations behind the composition of akṣamayī kalikā s, particularly focused on Raghunandana.

8. As noted below, not all published virudāvalī s contain an akṣamayī kalikā. Furthermore, Harideva provides an example of a partial alphabet poem that omits the difficult letters, and Cintāmaṇi Miśra offers an option of substitutions for difficult letters (see n. 30).

9. See Buchta (2014: 321-57) for a preliminary attempt to sketch the history and antecedents of the virudāvalī genre.

10. On Jīva's Harināmāmṛtavyākaraṇa see Blinderman 2024.

11. On the antecedent birudāvalī treated in Andhra-based alaṅkāra-śāstra s see Mohan 1972: 114; Jones 2018: 92-101. On the tracing of the virudāvalī genre back to these antecedents, and the refutation of claims by S. K. De and Jan Brzezinski that Rūpa is to be credited with developing the genre see Buchta 2014: 322-37.

12.Virudāvalīlakṣaṇa 2-3:kalikā-śloka-virudair yutā vividha-lakṣaṇaiḥ | kīrtti-pratāpa-śauṭīryasaundaryonmeṣa-śālinī || kalikādy-anta-saṃsargi-padyā doṣa-vivarjitā | śabdād...ambara-sambaddhā kartavyā virudāvalī ||"Filled with kalikā s,śloka s, and viruda s of various types, possessing fame, splendor, haughtiness, beauty, and radiance, with verses accompanying kalikā s at their beginning and end, free of faults, with intensity of speech, one should compose a virudāvalī."

Virudāvalīlakṣaṇa 101:virudaḥ kalikā cānte dhīra-vīrādi-śabda-bhāk | na pañca-trikato nyūnā nādhikā triṃśatas trikā | virudālī prakarttavyā paurvāparya-vikalpataḥ ||"A viruda and a kalikā have a word like dhīra or vīra at the end. One should compose a virudāvalī of no less than five and no more than thirty triads, the order being optional."

13.Virudāvalīlakṣaṇa 5ab:kalā nāma bhavet tālaniyatā padasantatiḥ|"What is called kalā is a group of words fixed by rhythm."

14. This loose end rhyme can be seen in the akṣamayī kalikā of Kṛṣṇaśaraṇa presented in full below. Typically, the rhyme begins from at least the penultimate vowel in each kalā. Thus, the first two kalā s (in fact, the first four) end -aya. However, we also see corresponding aspirated and unaspirated consonants involved in rhyme as in the kalā s beginning with ṛ and ...(ending in -atha and -ata), occasionally different penultimate vowels as in the kalā s beginning with e and ai(ending in -ita and -ata), or dental and retroflex nasals as in the kalā s beginning with jh and ñ(ending in -aṇa and -ana).

15. For example,Virudāvalīlakṣaṇa 64:evaṃprakārair aparair madhuraśliṣṭayogataḥ | viśliṣṭaśithilādīnāṃ yogataś ca paraḥ śatāḥ | caṇd...avṛttaprabandhāḥ syuḥ kas tān kārtsnyena vakṣyati ||"Together with such preceding types [formed] according to the use of madhura and śliṣṭha[conjuncts], there may be hundreds more caṇd...avṛtta compositions according to the use of viśliṣṭa,śithila, and other [i.e.,hrādin, conjuncts]. Who can speak about them exhaustively..."

16. The example occurs after Virudāvalīlakṣaṇa 97ab. The first line appears to be corrupt. The following two pairs of kalā s each share an internal metrical structure.

17.Virudāvalīlakṣaṇa 12:adhikāś cec catuḥṣaṣṭhir nyūnā dvādaśa tāḥ kalāḥ | etābhyo nādhikā kāryā nyūnāś cāpi na paṇd...itaiḥ ||"If the kalā s are in higher quantity, then they [may be up to] sixty-four, and if they are fewer, they [may be down to] sixteen. The learned should not make them more or less than these two numbers." This restriction is mentioned in the discussion of the first, and by far most pervasive major kalikā category, called caṇd...avṛtta, though examples of kalikā s in the other categories mostly adhere to this restriction as well. For akṣamayī kalikā s, the number of kalā s is determined by the number of letters in the varṇakrama, in this case fifty. Virudāvalīlakṣaṇa 99:dvi-catuḥ-ṣad...-daśaś cātra kalās tu virude matāḥ | daśabhyo nādhikāḥ kāryāḥ kalās tu virude budhaiḥ ||"There are understood to be two, four, six, or ten kalā s in a viruda. The wise should compose no more than ten kalā s in a viruda."

18. There are, by another count, twenty-ninekalikās in total, as Rūpa includes both of the two variants ofmiśra-

kalikāwithin one triad.

19. The two that do not are Jīva Gosvāmin's incompleteGopālavirudāvalīand Viśvanātha Cakravartin's

Nikuñjakelivirudāvalī.

20. For further details about Raghunanda and his work and a complete translation of hisakṣamayī kalikāsee

Buchta 2023.

21. See Buchta 2023: 117-18.

22. Details in Buchta 2014: 337-45.

23. Strictly speaking, Raghudeva presents his akṣamayī as the viruda of the final triadic unit, not heeding the limitation Rūpa had indicated, noted above, of a maximum of ten kalā s in viruda.

24. The Śāstrī and Parab (1903) and the Devī (1960) editions of the Stavamālā and the Bhaktivedānta Research Centre's manuscript of the Govindavirudāvalī contain the reading akṣamayī. The Rāmanārāyaṇa (1885) and Purīdāsa (1946) editions of the Stavamālā, the Paṭṭanāyaka (1982) edition of the Govindavirudāvalī, and the Tübingen and Matsya Project manuscripts of the Stavamālā contain the reading akṣaramayī. In those instances where Baladeva Vidyābhūṣaṇa's commentary accompanies the text (Śāstrī and Parab, Rāmanārāyaṇa, Paṭṭanāyaka [excluding the introductory section], Tübingen, and Matsya Project), the commentary is consistent with the root text both in the introduction to the Govindavirudāvalī and in the commentary on this label. Purīdāsa is the only editor to list the manuscripts and editions consulted and to occasionally note variant readings, though not offering a full critical edition. In addition to the Śāstrī and Parab and Rāmanārāyaṇa editions, Purīdāsa consults one from Gaurāṅgagranthamandira in Varāhanagara and two from the Baṅgīyasāhityapariṣad (1946: 2). Still, he does not note the reading akṣamayī found in Śāstrī and Parab, though his appendix listing the kalikā types in the Govindavirudāvalī does note akṣamayī as a variant name of the kalikā type.

25. The relevant entry is not found in the original 1821 Bangla-script edition of the Śabdakalpadrumaḥ prior to the revision by Vasu and Vasu. Thus, this entry is later than, and likely directly influenced by, the Vācaspatyam. Nevertheless, the Śabdakalpadrumaḥ cites a further passage from Baladeva's commentary not found in the Vācaspatyam under the entry for viruda, so Vasu and Vasu must have had direct access to the commentary.

26. In the Śabdakalpadrumaḥ, the word akṣamayī is broken across the line boundry as akṣa-mayī, adding a slight degree of uncertainty as to whether this spelling was intended.

27. More specifically, the Cakradhara edition has akṣamālā viruṣam[sic], while the Pāṭhaka edition has akṣamālāvirudam, but notes a variant reading akṣamālāvīravirudam.

28. This is the reading of the Tübingen manuscript. I have included the syllable ra in parentheses as being in question. Śāstrī and Parab read the first two words of the gloss as a compound:akārādyakṣakārāntā. The Matsya Project manuscript introduces an extra ma:prathamamabhidā. Rāmanārāyaṇa's edition follows the insertion of the extra ma and then changes the dā to dhā, apparently expecting the word abhidhā('name') instead of bhidā ('type'), so that it reads (spacing strictly represented):akārādyākṣakārāntā prathama mabhidhā varṇyetyarthaḥ. Paṭṭanāyaka shows further variants, including the interesting kṣarāntā:akārādyā kṣarāntā prathamamabhidhākā barṇṇyā ityarthaḥ.

29. As a humorous typo, this section is headed,akṣa akṣamayī kalikā, where the first word should be atha.

30. Cintāmaṇi's treatment of the akṣāvalī kalikā shows a few distinctive features. Rather than simply start each line with the letters of the alphabet, each line also ends with the letter that the next line will start with. Cintāmaṇi's example, attributed to his uncle, includes a retroflex la between ha and kṣa, further suggesting the likely Andhra antecedents to the genre. Relevant to the topic of this paper, Cintāmaṇi includes a comment that since ...,..., and ... are impossible at the beginning or end of a word, it is acceptable to use rū,lu, and lū, respectively (...-...-...nāṃ padasyādāv ante ca sthityasambhave | rū lu lū ca prayoktavyā yathākramam itīṣyate| [... emended from ṛ and rū emended from ru, confirmed by the example] Kara 1973: 389). Note that the use of u/ū in these substitutes corresponds to the pronunciation of these Sanskrit letters in the broader southeastern region including Odissa and Andhra Pradesh. Given this acknowledged leniency and other liberties taken in the examples with difficult letters, as well as apparent corruptions in the textual transmission, this example will not be thoroughly treated in this paper, though concurring examples of the strategies discussed here will be mentioned in footnotes.

31. For a full translation of Rūpa Gosvāmin's akṣamayī kalikā see Buchta 2014: 306-7; for Raghunandana's see Buchta 2023: 118-20.

32. That jhaṣavara should be taken to mean the makara, Kāmadeva's vehicle, likely alludes to Bhagavadgītā 10.31, where Kṛṣṇa identifies himself with the makara among fish (jhaṣāṇāṃ makaraś cāsmi).

33. Only Rāmanārāyaṇa's edition has short ṛ in Baladeva's citation.

34.bhuvaḥ sañjñāntaroḥ iti bhavateḥ sañjñāyāṃ d...upratyayaḥ. There are some complications with this appeal to Pāṇini's grammar. First, the sūtra cited indicates the affixing of KviP(from 3.2.177), not ...u, which is introduced in the following sūtra. That sūtra(3.2.180:viprasambhyo d...vasañjñāyām) introduces ...u with specific conditions: that √bhū follows vi,pra, or sam, and that the word derived is not a name. It may be that Baladeva meant to allude to the vārttika to this sūtra:d...uprakaraṇe mitadrvādibhyaḥ upasaṅkhyānam"Regarding ...u, [there should be] inclusion [of its occurrence] after elements such as mitadru." The Kāśikāvṛtti gives śambhu as an example of the application of this vārttika. Bhānujī Dīkṣita cites this vārttika in his etymology of ṛbhu in his commentary on the Amarakoṣa. Thus, there is precedent for the derivation that Baladeva proposes here, even if his citation is imprecise.

35.Medinīkoṣa,avyayānekārthavarga 6c-7b:... . . . devāmbāyāṃ danau cāpi bhairave danuje gatau"...[occurs] in the sense of the mother of the gods and Danu, as well as Bhairava, [a demon] born of Danu, and motion." Puruṣottamadeva's Ekākṣarakośa, verse 3cd (Ramaṇīkavijaya 1964: 76) has a similar pair of entries, but makes it clear that the identification of ṛ and ... with the mother of the gods and the demons is respective (ṛkāro devamātā syād ...kāro danujaprasūḥ"the letter ṛ is the mother of the gods; the letter ..., the mother of the demons").

36. See, for example,Bhāgavata Purāṇa 3.2.23, where Uddhava describes Kṛṣṇa as dayālu for his treatment of Pūtanā (bakī).

37. Haridāsa Dāsa's Bangla translation offers a rather convoluted reading with ...bhu meaning 'god', appearing to take this as a kind of contorted bahuvrīhi: "the hosts of gods are able to be uninjured only in your shelter" (devagaṇa tomāra āśrayeï akṣata haïte pāre).

38. Cakradhara does not explicitly cite the source of this entry, but it is identical to 10d in the Ekākṣarīmātṛkākośa edited by Ramaṇīkavijaya. Vogel (1979: 370-71) claims that this is a reprint of the Mātṛkānighaṇṭa of Mahīdāsa published in the Dvādaśakośānāṃ Saṅgrahaḥ published in 1865, but I have not been able to access that text.

39.Mahābhārata 5.68.5, for example, which identifies ṇa as expressing joy (ṇaśca nirvṛtivācakaḥ), is cited in Caitanyacaritāmṛta 2.9.30.

40. Ghoṣāla's edition has ṇarūpa, which was accidentally followed in Buchta 2023: 119.

41. These glosses follow the interpretation of Cakradhara, who cites Puruṣottamadeva's Ekākṣarakoṣa. Tarkavācaspati (1873-84: 1598) cites a Varṇābhidhāna with many more meanings attributed to these letters, especially ... and ..., allowing for a number of other possible interpretations. The example in Cintāmaṇi Miśra's Vāṅmayaviveka includes ṅarahitanigamaśatair api durvaca, describing Viṣṇu as "difficult for even hundreds of scriptures free of sensuality to describe."

42. Haridāsa Dāsa's edition removes the visarga according to sandhi rules, but I have restored it here. The shape of the visarga is referred to in Durgasiṃha's commentary on the Kātantra:kumārīstanayugākṛtiḥ('having the shape of a young woman's breasts'), similarly referencing the fact that there are two dots or circles, and here, more so, the roundness of the dots or circles.

43. Mallinātha, for example, offers etymologies such as d...hagiti vyaktaṃ kāyatīti d...hakkā"It makes the sound d...hak clearly, thus [it is called]d...hakkā."

44. Rūpa:d...hakkitakaratala('making claps like a d...hakkā'); Raghunandana:d...amarudhṛgīśvarad...huṇd...hyakṛpābhara('full of mercy sought by the lord who holds a d...amaru[Śiva]', discussed below); Kṛṣṇaśaraṇa: d...iṇd...imaghoṣaṇa('resounding like a d...iṇd...ima'),d...hakkāravaraṇa('with delight in the sound of the d...hakkā'); Raghudeva:d...amaruravādṛta('fixated on the sound of the d...amaru'); Cintāmaṇi:d...amarukarād api yogapathe dṛd...ha ('firmer on the path of yoga even than Śiva with the d...amaru in his hand'),d...hakkād...haṅkṛtiśaṅkitaripugaṇa('making hosts of enemies fearful with the sound d...haṅ from his d...hakkā').

45. Monier-Williams cites only the Dhātupāṭha and commentary on Pāṇini in his entry for this root.

46. In his Sanskrit commentary Haridāsa Dāsa explicitly acknowledges that these words are used as imitative sounds (ṅokṛtañokṛtaśabdayoḥ śabdānukaraṇe tātparyam"The purpose of the words ṅokṛta and ñokṛta is for the imitation of sound").

47. Cintāmaṇi Miśra's example uses ṭaṅkṛta in its normal sense:ṭaṅkṛtaśārṅgadhanūddhatikarmaṭha('expert in striking with his Śārṅga bow, made to twang'; emended from dhanuddhati). It likewise addresses Viṣṇu as ṭhaṇitivipāṭitakaiṭabhagalanad...a('cutting the reed-like neck of Kaiṭabha with the sound ṭhaṇ'; emended from ṭhaṇṇiti[for meter] and keṭhabha).

48. The shift from velar to palatal consonants in the first iteration of a reduplicated syllable is accounted for in Aṣṭādhyāyī 7.4.62:kuhoścuḥ('In place of a velar stop or h, [in the first iteration of a reduplicated syllable (abhyāsasya from 7.4.58)] there occurs a palatal stop').

49.Aṣṭādhyāyī 6.1.9 accounts for reduplication of a monosyllabic verb root before the desiderative and intensive suffixes. Maitreyarakṣita's Dhātupāṭha commentary lists the intensive third-person singular present tense form ñoṅūyate for √ṅu. Monier-Williams cites the desiderative form ñuṅūṣate as attested in commentaries on Aṣṭādhyāyī 7.4.62, but I have not been able to trace such a citation.

50. All editions read ñoṅūyāpadakara, presumably carrying over pada from the line above. I have emended the reading based on meter and end rhyme.

51. Entries beginning with aṃ occupy only half a page in Monier-Williams, and no entries are listed beginning with aḥ.

52. Cintāmaṇi Miśra's example has a similar metalinguistic pun for the letter ña:ña ivobhayapadayogitayā sphuṭa. Viṣṇu is "manifest as connected with both pada s, like ña," perhaps a reference to Viṣṇu's heavenly existence and his manifestation on earth through his avatāra s. Thus, Viṣṇu is like the Pāṇinian code-letter Ñ, which marks verbs that can take both parasmaipada and ātmanepada endings, as per Aṣṭādhyāyī 1.3.72.

53. For a condensed but highly informative discussion of these issues see Cardona 1969: 6-7, 38-39.

54. The Mahābhāṣya discussion occurs in the commentary on Akṣarasamāmnāya 3-4 (Kielhorn 1880: 25-26). See Cardona 1969: 38 n. 88.

55. The term ...T, i.e.,... marked with a code letter T, refers exclusively to the short vowel and not to a corresponding (savarṇa) long vowel, according to the metarule stated in Aṣṭādhyāyī 1.1.70:taparastatkālasya('[a vowel] followed by t is a technical term for [only] a vowel of that length').

56. For a discussion of the Kātantra and its date see Scharfe 1977: 162-63.

57. I take Raghunandana here to be alluding to discussion in the Mahābhāṣya and the Kāśikāvṛtti on Aṣṭādhyāyī 1.1.9 and 6.1.101, where long ... is ruled out as a single substitute for ṛ+.... For analysis of other interpretations of this epithet see Buchta 2023: 125.

58. For an alternate interpretation that reads these two kalā s separately see Buchta 2023: 119.

59. While the indebtedness to Rūpa is seen most starkly in the composition of these difficult epithets, one can see potential traces of his influence in the case of some, though not all, of the easier letters. For example, compare Rūpa's paṅkajasamapada('with feet like lotuses') and phaṇinutamodita('pleased by the praise of the snake [Kāliya]') with Kṛṣṇaśaraṇa's paṅkajalocana('lotus-eyed') and phaṇimadamocana('dispelling the pride of the snake').

References

REFERENCES

Manuscripts Consulted

Rūpa Gosvāmin.Govindavirudāvalī. From Gopivallabhapura, Midnapore. Digitized by the Bhaktivedanta Research Center, Kolkata. gen, Shelfmark Ma I 211. (https://opendigi.ub.uni-tuebingen.de/opendigi/MaI211)

-.Stavamālā, with the Commentary of Baladeva Vidyābhūṣaṇa. Owned by Pradīpa Kumāra Banarjī, VBM. Digitized by the Matsya Project and the Vrindavan Research Institute.

Primary Works Cited

Amaracandrakavi.Ekākṣaranāmamālikā. In Ekākṣaranāma-Koṣasaṅgraha, ed. Ramaṇīkavijaya. Rājasthāna Purātana Granthamālā, vol. 64. Jodhpur: Rājasthāna Prācyavidyā Pratiṣṭhāna, 1964.

Amarasiṃha.Amarakoṣa (Nāmaliṅgānuśāsana), with the Commentary of Bhānuji Dīkṣita, ed. Haragovinda Śāstrī. Kashi Sanskrit Series, vol. 198. Varanasi: Chowkhamba Sanskrit Series Office, 1970.

Candraśekhara Bhaṭṭa.Vṛttamauktika, ed. M. Vinayasagar. Rājasthāna Purātana Granthamālā, vol. 79. Jodhpur: Rajasthan Oriental Research Institute, 1965.

Cintāmaṇi Miśra.Vāṅmayaviveka, ed. Karuṇākara Kara. Bhubaneswar: Orissa Sahitya Akademi, 1973.

Ekākṣarīmātṛkākośa. In Ekākṣaranāma-Koṣasaṅgraha, ed. Ramaṇīkavijaya. Rājasthāna Purātana Granthamālā, vol. 64. Jodhpur: Rājasthāna Prācyavidyā Pratiṣṭhāna, 1964.

Harideva Miśra.Jahāṃgīra-virudāvalī, ed. Jagannātha Pāṭhaka. Prayāga: Gaṅgānātha Jhā Kendrīya Saṃskṛta Vidyāpīṭham, 1978.

Kṛṣṇaśaraṇa.Kṛṣṇavirudāvalī, ed. Haridāsa Dāsa. Navadvīpa: Haribola Kuṭīra, 1944.

Maitreyarakṣita.Dhātupradīpa, ed. Srish Chandra Chakravarti. Rajshahi: Varendra Research Society, 1919.

Medinīkara.Medinīkoṣa (Nānārthaśabdakoṣa), ed. Jagannāth Śāstrī Hośhing. Kashi Sanskrit Series, vol. 41. Varanasi: Chowkhamba Sanskrit Series Office, 1968.

Pāṇini.Aṣṭādhyāyī, with the Mahābhāṣya of Patañjali, ed. F. Kielhorn. Bombay: Government Central Book Depot, 1880.

-.Aṣṭādhyāyī, with the Kāśikāvṛtti of Vāmana and Jayāditya, ed. Dwarika Das Shastri and Kalika Prasad Shukla. Prachya Bharati Series, vol. 2. Varanasi: Prachya Bharati Prakashan, 1965-67.

Puruṣottamadeva.Ekākṣarakoṣa. In Ekākṣaranāma-Koṣasaṅgraha, ed. Ramaṇīkavijaya. Rājasthāna Purātana Granthamālā, vol. 64. Jodhpur: Rājasthāna Prācyavidyā Pratiṣṭhāna, 1964.

Raghudeva Miśra.Virudāvalī, with the Commentary of Cakradhara Śarman Jhā, ed. Muralīdhara Śarman Jhā. Kāśī: Prābhākarī-yantrālaya, 1903.

-.Virudāvalī, ed. Jagannātha Pāṭhaka. Prayāga: Gaṅgānātha Jhā Kendrīya Saṃskṛta Vidyāpīṭham, 1979. [Repr. from Journal of Ganganatha Jha Kendriya Sanskrit Vidyapeetha 34.1-2 (1978).]

Raghunandana Gosvāmin.Gaurāṅgavirudāvalī, ed. Haridāsa Dāsa. Navadvīpa: Haribola Kuṭīra, 1942. -.Gaurāṅgavirudāvalī, ed. Haridāsa Śāstrī. Mathurā: Śrī-Gadādhara-Gaurahari Presa [Press], 1984.

-Gaurāṅgavirudāvalī, ed. Baikuṇṭhanātha Ghoṣāla. Calcutta: Hari Press. No date. [Orig. publ. in Sajjanatoṣaṇī.]

Rg-Veda-Prātiśākhya, with the Commentary of Uvaṭa, vol. 2, ed. Mangal Deva Shastri. Allahabad: The Indian Press, 1931.

Rūpa Gosvāmin.Govindavirudāvalī, with the Commentary of Baladeva Vidyābhūṣaṇa, ed. Dolagobinda Paṭṭanāyaka. Kaṭaka: Bidyāpurī, 1982.

-.Stavamālā, with the Commentary of Jīva Gosvāmin [sic; Baladeva Vidyābhūṣaṇa], ed. Rāmanārāyaṇa Vidyāratna. Baharamapura: Rādhāramaṇa Yantra, 1885.

-.Stavamālā, with the Commentary of Śrījīvadeva [sic; Baladeva Vidyābhūṣaṇa], ed. Bhavadatta Śāstrī and Kāśīnāth Pāṇdurang Parab. Kāvyamālā Series, vol. 84. Bombay: Nirṇaya-Sāgara Press, 1903.

-.Stavamālā, ed. Purīdāsa. Mayamanasiṃha [Mymensingh]: Śacīnātharāya-Caturdhurīṇa, 1946.

-.Stavamālā, ed. Aparṇā Devī. Kalikātā: Gauḍīya Miśana [Mission], 1960.

-.Virudāvalīlakṣaṇa, ed. Haridāsa Dāsa. Navadvīpa: Haribola Kuṭīra, 1941.

Secondary Works Cited

Adam, William. 1938.Third Report on the State of Education in Bengal; Including Some Account of the State of Education in Behar, and a Consideration of the Means Adapted to the Improvement and Extension of Public Instruction in Both Provinces. Calcutta: G. H. Huttmann, Bengal Military Orphan Press.

Allen, W. S. 1953.Phonetics in Ancient India. London: Oxford Univ. Press.

Blinderman, Radha. 2024. Why Kṛṣṇa and Śakti Have Their Own Grammars: The Case of Jīva Gosvāmin and Balarāma Pañcānana. PhD diss., Harvard University.

Broo, Måns. 2011. Drama in the Service of Kṛṣṇa: Rūpa Gosvāmin's Nāṭaka-Candrikā. In Pūrvāparaprajñābhinandanam : East and West, Past and Present. Indological and Other Essays in Honour of Klaus Karttunen, ed. Bertil Tikkanen and Albion M. Butters. Pp. 55-65. Studia Orientalia, vol. 110. Helsinki: Finnish Oriental Society.

Buchta, David. 2014. Pedagogical Poetry: Didactics and Devotion in Rūpa Gosvāmin's Stavamālā. PhD diss., University of Pennsylvania.

-. 2023. An 'Alphabet Poem' for Caitanya: The Akṣaramayī Kalikā in Raghunandana's Gaurāṅgavirudāvalī.Journal of Vaishnava Studies 32.1: 115-29.

Bühnemann, Gudrun. 1992. On Puraścaraṇa: Kulārṇavatantra, Chapter 15. In Ritual and Speculation in Early Tantra: Studies in Honor of André Padoux, ed. Teun Goudriaan. Pp. 61-106. Albany: State Univ. of New York Press.

Busch, Allison. 2018. The Poetics of History in Padmakar's Himmatbahādurvirudāvalī. In Text and Tradition in Early Modern North India, ed. Tyler Williams, Anshu Malhotra, and John Stratton Hawley. Pp. 260-81. New Delhi: Oxford Univ. Press.

Cardona, George. 1969. Studies in Indian Grammarians: I. The Method of Description Reflected in the Śivasūtras.Transactions of the American Philosophical Society, new ser. 59.1: 3-48.

Delmonico, Neal. 1990. Sacred Rapture: A Study of the Religious Aesthetic of Rupa Gosvamin. PhD diss., University of Chicago.

Deva, Rādhākānta, Varadāprasāda Vasu, and Haricaraṇa Vasu, eds. 1886.Śabdakalpadrumaḥ. Chowkhamba Sanskrit Series, vol. 93. Rev. ed. Varanasi: Chowkhamba Sanskrit Series Office.

Jones, Jamal. 2018. A Poetics of Power in Andhra: 1323-1450 CE. PhD diss., University of Chicago.

Mohan, P. Sarasvati, ed. 1972.The Camatkāracandrikā of Śrī Viśveśvara Kavicandra: Critical Edition and Study. Delhi: Sanskrit Book Depot.

Monier-Williams, Monier. 1899.A Sanskrit English Dictionary: Etymologically and Philologically Arranged with Special Reference to Cognate Indo-European Languages. Rev. ed. Oxford: Clarendon Press.

Palsule, G. B. 1953. Concordance of the Sanskrit Dhātupāṭhas.Bulletin of the Deccan College Post-Graduate and Research Institute 15.1: 1-201.

Pollock, Sheldon. 2009.The Language of the Gods in the World of Men: Sanskrit, Culture, and Power in Premodern India. Delhi: Permanent Black.

Scharfe, Hartmut. 1977.Grammatical Literature. Wiesbaden: Otto Harrassowitz.

Tarkavācaspati, Tārānātha, ed. 1873-84.Vācaspatyam. Chowkhamba Sanskrit Series, vol. 94. Varanasi: Chowkhamba Sanskrit Series Office.

Twaddell, W. Freeman. 1935.On Defining the Phoneme. Language Monographs, vol. 16. Baltimore: Waverly Press.

Vogel, Claus. 1979.Indian Lexicography. Wiesbaden: Otto Harrassowitz.