Introduction
Nandikeśvarakāśikā (also called Ādisūtrakāśikā, Ādisūtrakārikā,
Nandikeśvarakārikā) is a unique and distinctive text ascribed to Nandikeśvara (dates
unknown). Its uniqueness is at least partly attributable to the ways in which it
describes series of causal links and relationships between the Sanskrit alphabet and
stages of cosmic emanation. These stages, described in the text, as a graduated
unfoldment of categories of existence (tattvas), are synchronised with the letters of the
Sanskrit alphabet as these have been represented in the well-known Māheśvarāni
sūtrāni of Pāṇini’s Ashtādhyāyī (app. 500 BC)1, a foundational Sanskrit grammar text.
The Māheśvarāni sūtrāni, as we may know, group the alphabet into fourteen distinct
units of vowels and consonants based on their specific and internal consistencies. The
Nandikeśvarakāśikā (NK), remarkably, conjoins this unique sequence of phonological
letter-groups with a parallel emanation of sequential tattvas, or categories that are also
here significantly grouped into units. Perhaps, the intriguing feature of the cosmic
model or paradigm presented in the Nandikeśvarakāśikā is a series of complex causal
correlations between the Sanskrit phonological groups and tattvas which are
recognized here as emanating or originating from the former into existence.
Traditionally, the Māheśvarāni sūtrāni have been imbued with metaphysical
connotations specifically in the context of their sound-symbolism and association with
śabdabrahman or Word. However, such a sacred dimension to the letters in the
Pāṇinian sūtras has not been explicated or developed into a full-scale cosmogony in
any known texts prior to NK. Therefore, we find here a quite early and distinctive
usage of a ‘grammarians’ alphabet-system as a means of concrete manifestation of the
cosmos. Moreover, by establishing such links between the sūtras and cosmogony,
Nandikeśvara seems to have also created new links between the Sanskrit grammar of
Pāṇini and Śaiva Agama traditions.
In this paper, I present an edition together with English translation of
Nandikeśvarakāśikā in its distinct recession. This is a shorter version of this unique
and historically valuable text. I will try to give an outline of the available to me
research data regarding NK’s date and place. The text has a historical value because
traditionally it is considered to be an earliest text describing phonic cosmology.2 But
not only that, it is linked to the traditional grammatical elements. I also intend to
explore the specific ways in which the author, Nandikeśvara, has represented
relationships between the Pāṇinian groups of letters and groups of tattvas. I will try to
show how these groups of letters are creative cosmic principles, presided over by
specific deities, and the ways in which they serve to bring specific levels of existence
into being. It is my notion that the author may have viewed the Māheśvara sūtras,
which are groups of letters, as creative agencies, because these agencies in fact are
themselves groups, pairs, or triads that represent inherent aspects such as formless and
form, consciousness and power, will and knowledge, sovereignty or lordship and
multiplicity, etc. Instead of singularity, this text explores the universe as
fundamentally inter-relational.
3
Place and Date of Nanadikeśvarakāśikā
There seems to be much ambiguity and uncertainty with regards to the history of
Nanadikeśvarakāśikā as a text. We do know that Nandikeśvara is a name attributed to
various authors who wrote on a range of subjects. A grammarian called Nandikeśvara
is mentioned by Kṣirasvāmin in his commentary on Amarakośa and by Sayana in
Dhatuvivṛtti.3 In the canon of Jayadrathayāmala, a summary of which has been
presented by Dyzckowski,4 Nandinātha is an author of Krodhānalasaṁhitā. However,
one may not conclude from this that the Nandinātha of Jayadrathayāmala is the author
of NK.
As a name, Nandinātha, or Nandikeśvara, or simply Nandī, has several
associations: a deity associated with Śiva, a celestial being, a sage or a legendary
teacher who passed on various branches of Śaiva teachings. There are in fact several
texts where Nandī is mentioned as a legendary teacher. For instance, we know that
Nandikeśvara is purported to be the deity who once transmitted the knowledge of
Gandharvatantra on earth.5 According to Brahmayāmalatantra, Nandī was the source
who first revealed the knowledge of Raurava agama.6 Also, Nandinātha is mentioned
in the canon of Śaiva Siddhanta as a recipient of Raurava agama from the Tatpuruṣa
face of Sadāśiva.7 However, it is unlikely that these legendary authors could be
identical with our Nandikeśvara for several reasons not the least of which is the fact
that the style and the views reflected in NK are quite unique in themselves. Besides
this, it is apparent that most of the traditional canons reflected in the mentioned above
tantras and āgamas have little clear historical value.
At first glance, it may seem strange to note that perhaps our most reliable
sources of information about the actual history of the text may be some of the Śaiva
myths reflected in NK. For example, we know that there are specific myths
associating Śiva, or Maheśvara, in the form of Naṭarāja with a long-standing
grammatical tradition in Sanskrit literature. This form of Śiva is purported to have
delivered the Sanskrit alphabet to the world through the śiva sūtras. The earliest
instance of the myth of Pāṇini receiving akṣrasamāmnāya from Maheśvara occurs in
the versified Pāṇinīyaṣikśa, a text on phonetics, attributed to Pāṇini himself.8
Moreover, in the commentary of Upamanyu we find it noted that sages Nandikeśvara
and Patañjali also witnessed Śiva’s dance. Relying on this description, K.C. Pandey
concludes that Nandikeśvara was a contemporary of Patañjali. He adheres to a
traditional view that Nandikeśvara was a contemporary of both Pāṇini and Patañjali.9
However, we know that Pāṇini’s contemporaneity with Patañjali is considered to be
historically improbable by most scholars. Furthermore, the narrative of the dance
witnesses may be seen as a kind of rhetorical device for giving equal weight to the
above mentioned teachers (acaryas) rather bearing any kinship to historical facts.
However, it is interesting that Pandey suggests another ‘evidence’ that NK was known
to Patañjali. This is the occurrence of the word brahmarāśi in Mahābhāṣya (1.36) and
Kaiyata glossing the word as brahmatattva. Following Nāgeśa Baṭṭa, Pandey
concludes that Patañjali hinted at Nandikeśvara’s brahmatattva or brahmarūpa (in the
other recession) in the fourteenth verse. However, this evidence seems to be
insufficient to prove that Patañjali based himself on Nandikeśvara and not the other
way around.
It is perhaps significant that the traditional body of literature of Sanskrit
grammarians reveals little awareness of NK till much later in its development. It
4
seems that most grammarians preferred to remain either indifferent or were unaware
of any connection between Śiva and the sūtras till the time of Haradatta, a South
Indian commentator of Kāśikāvṛtti.10 It is also significant that Haradatta did not quote
the verse directly from Pāṇinīyaṣikśa, but from a compilation of traditional verses of
wisdom.11 The verse then appears as cited by Srilankan Dharmakirti in his Rūpāvatāra.
Deshpande concludes from several fragments of such evidence that the story of Pāṇini
receiving sūtras from Maheśvara was probably wide spread in southern parts of the
Indian subcontinent in the 10th century.12 However, none of the writers who
mentioned the famed Śiva-Pāṇini myth in their works actually referred themselves to
NK. It is in fact only Nāgeśa Bhaṭṭa, in his Laghuśabdenduśekhara, who actually notes
that Śiva himself offered the sūtras. Nāgesa may have done this on the basis of the
Pāṇinīyaṣikśa and the Kāśika of Nandikeśvara.13 Deshpande concludes that: “all
...factors, and especially the divine origins of Pāṇini’s grammar, are taken up by
Nāgeśa and woven into a powerful argument to explain the status and the authority of
the Sanskrit grammar.” According to him, the myth and the verse from NK about
Naṭarāja revealing sūtras has come to be widely used by traditional grammarians
from the time of Nāgeśa. However, the larger body of the text of NK seems to have
remained outside the context of grammatical literature for it seemingly has little
validity for grammar.
The connection of NK with Kashmiri Shaiva tradition shows more textual
evidences. But these evidences are interpreted differently by researchers. Pandey
points out to lines similar in NK with Spandakārika14 and with Pratyabhijñā
Hṛdayam.15 He suggests that the NK “...is the earliest voluntaristic philosophy, which
was developed... by the thinkers of Kashmir ... in the light of monism.” He concludes
about Kashmiri origins of the text as well as about Kashmir Shaivas like
Abhinavagupta and Kshemaraja drawing from NK. One can also find common with
KSH terminology and notions such as citkalā (consciousness), being the supreme Lord
and aham (absolute I).16 Yet, it could be Nandikeśvara echoing Kaśmiri texts and not
vice-verse. Moreover, the text seems to be not known to Kashmir Shaivas. Rafaele
Torella in his interview to Deshpande sheared his impressions about later character of
the text: "I don't remember seeing any reference to Nandikeśvara's Kāśikā in any
Kashmir Shaiva texts (even where one might most have expected to find it, like the
Paratrimśikavivārana or Tantraloka, ch. III)... it is not an old text,.. but a late one
which rather presupposes the doctrines of the so called ‘Kashmir Shaivism’ at a postexegetical
stage.” Therefore, taking into consideration a striking resemblance of NK
with some verses from Kashmiri texts, it is more likely to conclude that NK was
written after Abhinavagupta’s time, that is not before the eleventh century CE.
In his article, Deshpande agues that NK is a later text which could be composed
in Tamil Nadu after 800 CE.17 In regard to Southern origins of the text and not
Kashmirian, Deshpande points out that the first verse of NK mentions Naṭarāja, whose
worship was not known in Kashmir, but was spread in Chidambaram region.
Additionally, Upamanyu mentioned among sages not only Nandikeśvara but also
Patañjali and Vyāghrapād, whose shrines are found in the yard of Naṭarāja’s temple in
Chidambaram.18 The story of Patañjali and Vyāghrapād is as early as the earliest
temple of Śiva in Chidambaram founded by Hiraṇyavarman of the early Pallava
dynasty.19 So, Chidambaram seems to be the most possible place where the Naṭarāja-
Pāṇini myth was developed.
5
Similar conclusions can be made from the paper by S. Palaniappan available on
the Indology webpage.20 On the basis of the Tamili literary tradition, Palaniappan adds
that there are certain elements in Tamil mythology that could contribute to the myth of
Pāṇini and Śiva. For example, a cult of Dakṣiṇamūrti of Madurai created a similar
myth of Tamil grammar taught by Shiva to Agastya. Then a dance treatise Tamil
kuutta nUl says that from one side of Śiva's drum, Sanskrit letters come and Tamil
letters come from the other. Interestingly, the story of the Lord Naṭeśa teaching Pāṇini
fourteen sūtras was already known in the early Chola period of worship of the Lord
Dakṣiṇamūrti, whose cult was supported by the Pandyan dynasty. There is a place of
worship of the Lord Vyākaraṇadānaperumāl called Vyākaraṇasāna-mandapa, a place
where Śiva is believed to have appeared before Pāṇini.21 Palaniappan says that the
myth of Śiva-Pāṇini connection could serve as a political tool to support Shaiva rulers
of Chidambaram in their political competition with Pandyan rulers of Madurai from 6th
to 10th centuries. This is a very possible period when NK could be composed to
increase the authority of Chidambaram with not less powerful religious figures, cults,
and doctrines.
One more evidence of Nandikeśvara’s connection to Chidambaram comes from
the Shaiva Siddhanta tradition. Nandikeśvara is considered to be a legendary founder
of some Shaiva Siddhanta schools. Tirumūlar, the author of Tirumantiram in the Tamil
language where much of the Shaiva Siddhanta provisions have been formulated, is
considered to be a direct disciple of Nandikeśvara. In his work, Tirumūlar names those
who received Śiva’s grace at Chidambaram. Among them are the four kumāras
beginning with Sanaka, then Panañjali and Vyāghrapāda.22 Nandikeśvara also speaks
of sages beginning with Sanaka who benefited from the Śivasūtras revealed by
Naṭarāja. Upamanyu extends this list to Panañjali and Vyāghrapāda. So, there is not
only a claim of succession in lineage, but also the common mythology connected to
the shrine of Naṭarāja in Chidambaram which has been employed by the authors from
this lineage. Goudriaan places Tirumular in the seventh century of CE, on the basis of
evidences presented by Zvelebil.23 This can lead to a conclusion that Nandikeśvara was
either Tirumular’s contemporary or lived before him. A further study of references to
NK and its myths in Tamil sources as well as relations between Tamil Siddhas and
Kashmir Pandits could reveal more details regarding origins and the role of this text.
Māheśvarāni Sūtrāni and their role in Nandikeśvarakāśikā: a tantric
connection?
An introduction to the māheśvarāni sūtrāni in the Pāṇinian system leads one
quickly to a recognition and understanding that the unique clustering and arrangement
of the Sanskrit letters has been designed here to serve a rather precise function. The
Sanskrit syllabary, as we know, has been reorganised in these well-known sūtras and
placed into discrete units, or groups, according to their phonetic characteristics. For
example, we find that semivowels are placed together in one group, the nasalised
letters are in another, aspirated are in yet another group, and so on. It is also quite
well-known that these groupings carry the possibility for an efficient implementation
of grammatical rules of Sanskrit as a language.
6
For example, in the Pāninian system, the grammatical rule 8.4.53 jhalāṁ
jaśjhaśi requires that voiced un-aspirated consonants (jaś) replace spirants (jhal)
occurring before voiced consonants (jhaś). Here, jaś jhal and jhaś are pratyahāras, or
compressed groups of letters. They are formed with the help of a final extra consonant
which is a ‘terminal sign’, or a marker, called anubandha. These markers serve as a
means of formation of a coding system expressed through phonetic keys called
pratyahāras, or compressed groups of letters. For example, the four sūtras - a i u ṇ ṛ ḷ
k e o ṅ ai au c - are abbreviated into one pratyāhāra ‘ac’ where ‘a’ is the initial vowel
and ‘c’ is the final marker, and the whole pratyāhāra means ‘all vowels’. Similarly,
‘jaś’ would mean consonants j b g ḍ d where ‘a’ is added for pronunciation purposes
and in order not to mix the letters with anubandhas.
In this way, instead of enumerating all the phonemes significant for
grammatical operations, Pāṇini uses one short key-word. Then, the key-words are used
to form grammatical rules and technical terms such as ‘aj-anta’ or ‘hal-anta’ (nominal
stems ending in vowels or ending in consonants). So, in the grammatical context, the
rearrangement of phonemes and the use of markers are highly valid for they serve to
form meta-linguistic elements which contribute to efficiency and brevity of the
grammatical text. Such a system of coding is rational and each symbol has a concrete
semantic meaning.
An interesting feature of Nandikeśvarakāśikā as a text is how such a Pāṇinian
grammatical-system of grouped letters has been linked here to a seemingly distinct
system of the categories of cosmic emanation known - mostly in tantric cosmology - as
tattvas. It is as if the author of this text believed that there was a kind of similarity and
sympathetic resonance between these two quite different systems - one relating to
core-rules of Sanskrit grammar as a language and the other to hidden grammatical
‘rules’ of cosmic manifestation. Both of these systems seem to involve the ability to
distinguish specific groups or units - letters in one case and tattvas, or categories of
creation, in the other - based on an inbuilt universal logic, or grammar, if you will. By
means of such a ‘grammar’, both language and physical creation gain an internal
consistency of laws, rules, and regulations from which they emerge and which are
maintained through countless instances of their application in reality.
In several systems of tantric metaphysics, phonemic cosmology is an essential
feature.24 The phonemes are perceived as being alive, filled with power, and as sacred
beings of some kind – deities, in fact. Traditionally, such a sacralization of Sanskrit
phonemes is rooted in a concept of ‘sonic power’ (vāc śakti) wherein each letter of the
alphabet is synchronized with a vibratory pattern of sound that articulates a particular
aspect of reality into existence.25 In this context of creative power, the phonemes
acquire a very specific meaning which is not the same as linguistic, or semantic,
meaning. The attribution of meaning to the Sanskrit alphabet there depends on
religiously constructed representational power of sound. A reference point of each
letter then, becomes a stage, or a station in the process of unfoldment of a hierarchical
cosmos. Therefore, in order to explain and to understand the sacredness of phonemes
and their meaning, each phoneme ought to be seen in a cosmological context. Each
part of the Sanskrit alphabet here becomes more like a symbolic sign endowed with
creative or destructive power and serves a cosmological function within the tantric
paradigm.
Andre Padoux, who has written extensively on tantric cosmology and the
concept of speech or ‘vāc’, describes the latter as a ‘creatrix’ and “a symbol of
7
…Godhead… the force that creates, maintains, and upholds the universe.”26 In his
seminal work, ‘Vāc:.. in selected Hindu Tantras’, Padoux shows how this concept of
‘vāc’ or ‘Word’ is expressed in tantric systems at two overlapping levels - as a cosmic
manifestation and as a human act of speech:
All… developments of the Word… occur homologously within man or the
cosmos…Thus…the creative act (is) an utterance which is a human act…reversing the
order (we) see in this act nothing but the reproduction at the human level of an
archetypal, divine act or process… the universe emerges within divine consciousness,
through… stages or levels of speech…the categories…of the cosmic manifestation arise
concurrently with the Sanskrit phonemes (varṇa) arranged in their grammatical order,
while grammar – as well as traditional phonetics - will serve to account for the
cosmology.27
The Sanskrit phonemes, or varṇas, represent sound-elements which constitute
the world of expressed objective forms (vācya) including all the basic elements and
categories of creation from the mahābhutas to the tattvas of human senses (indriyas)
and inner mental organs (antaḥkarana). The grammatical units of speech, in this view,
are transformed into symbols that express a functional cosmology. Such a tantric
approach to the letters of the Sanskrit alphabet is somehow similar to mantras,
especially bija mantras, which do not have semantic meaning and are called as nonlinguistic,
but which are meaningful because they serve the purpose of mapping and
representing an emergent universe.
If we keep to this perspective, then it is logical that a text like the NK could
find deeper correlations between the grammatology of the Māheśvara sūtras and their
cosmological role as creative matrices of a manifesting universe. To articulate this
deeper dimension and function of Māheśvara sūtras and to distinguish it from the
purely grammatical level of language, Nandikeśvara says that anubandhas are for
grammatical use only while the letters have deeper esoteric meanings beyond this.28
Nandikeśvara’s unique methodology of this sacred dimension seems to be in a
functional complementarity to the existing arrangement and grouping of the Sanskrit
letters as we find them in the Māheśvara sūtras. A remarkable feature here is that the
cosmological grammar of emanating levels of being – tattvas - appears altogether in
concert with the linguistic component and its ‘grammar’. He emphasizes the
importance of grouping of individual letters and syllables into units, or sūtras, where
each element in a unit is inseparably related to the others. Here, a sūtra, as a composite
of letters, represents a complete stage in the creation. This may be because the creative
agencies – sounds or letters as deities - and the constituent elements of the universe
emanating from them are envisioned to appear entirely in, and as, groups or units.
Both sūtras and the cosmological categories are found grouped in pairs, triads,
in units of five and even in units of eight. Just as in the sūtras, where the letters in
each group have common features, so also apparently do the tattvas. The relationships
between the elements in one unit can be of different kinds but are always close and
inseparable so that one cannot exist without the others. Within the groups of letters -
especially at the level of vowel-units - there tends to be an inherent or inbuilt polarity
of pairs between each creative agency, or deity. So, for example, the first group of
letters - ‘aiuṇ’ - is presided over by Brahma who signifies the a-kāra principle which
is formless and constitutive of pure awareness or pure knowledge (jñaṇamaṭram,
citkalā). This ‘alpha’ deity is coupled with the next letter of the group - ‘i’ - that is
female and signifies the pure energy of will or desire (icchā śakti). Their togetherness,
8
or androgyny, is expressed through the third letter of the group - u-kāra - which is
expressed in the form of Viṣṇu, the ‘Pervader’.
This tendency to represent an inherent gender-based polarity within each group
of letters and within their representative deity persists like a strain throughout the text
and is repeated in different ways in each successive sūtra. Within the groups of letters,
consisting of even more than two, we notice that this polarity of gender between letters
may somehow be critical to the morphology of the cosmic creative process. Each
created unit seems to have two aspects: a pure awareness factor, or cognitive factor,
which tends to be masculine in nature and a corresponding dynamic movement
towards proliferation, multiplicity, and objectivity that is somehow feminine.
Such repeated polarities indicate that both aspects are constitutive to the very
order of creation. On the one hand, creation may be viewed in purely process terms,
implying dynamic change and movement from a subtle state to increasingly gross and
fragmented levels of being. On the other hand, the creative agency somehow exceeds
and transcends the frontiers of the created world, and may be described as a
"substance of substances", implying statism and inherent existence from its own side
(svabhāva). Therefore, there is a polarity between two forms that a deity assumes:
Brahma and Māyā, Puruśa and Prakṛtī, Meśvara and Maheśvarī. These two poles have
a strong tantric resonance and remind a play of creation through the exchange between
Śiva and Śakti. These two aspects are always conjoined and explicitly recognised in
the texts as inseparable from each other.
An even more interesting feature of NK is the element of causal relationship
introduced in the text as a link between the two corresponding ‘grammars’ - of
language and of cosmic manifestation through tattvas. We note that a kind of primacy
seems to be given here to the former as effective cause, source, and origin, of the
latter. The letters are somehow viewed as inherently potent and imbued with power -
in fact, they possess creative agency (karaṇatva) to sound the entire world into being.
As creators, or cosmic actors, the letters are in themselves full-fledged sacred deities
with names like Brahma, Maheśvara and Viṣṇu. The letters represent or express the
deities and their causal functions. In order to underline causative powers of the letters,
Upamanyu, the commentator of Nandikeśvara, gives an example of the letter ‘e’
which purportedly derives from a combination of letters ‘a’ and ‘i’. According to his
reading of the Nandikeśvara text, the letters in the very first sūtra, ‘have causative
power to produce other letters’. Similarly, these first letters, known as Brahma and his
power of will (icchā śakti), operate through distinct stages of manifestation of the
cosmos originating the succeeding stages of enfoldment of the universe.
One must note at this point, that as we follow the text through its intricate
patterns of sound, their deities, and the issuing cosmic tattvas, there persists a pattern
of continuity between the two operative systems – of language with its core-rules of
grammar and categories of cosmic emanation. Both share seeming similarities in
activities, functions, or traits and are continuously interwoven with each other. One of
the distinct ways to identify letters as deities or tattvas is to involve verbal or phonetic
similarities between two systems. For example, the words agra (first) and akṣara
(imperishable) traditionally characterise the Absolute, or Brahma. The word agra is
also a characteristic of the letter ‘a’ and the word akṣara means ‘a letter’ in Sanskrit.
Similarities in words characterising two entities here lead to identification of the letter
‘a’ with the Absolute. Even more interesting feature is establishing links based on
similarity in one letter. As in case of the sūtra ‘kapay’, the letter ‘pa’ represents
9
puruṣa, or the letter ‘sa’ represents sattva in the sūtra ‘śaṣasar’. Here, it seems that the
meaning of the word emerges from the initial syllable the word. Such attribution of
meaning to sounds echoes traditional tantric views that constituent syllables or sounds
of a word somehow possess lexical meaning which contributes to the meaning of the
whole word.29 By means of identification of syllables with words, it seems these
syllables become identified with categories and things those words denote. Thus,
connections between letters and words bring the alphabet into the realm of
metaphysics.
To sum up, the Māheśvara sūtras in NK depart from its purely grammatical
functions and appear to be a symbolic system with references which can be interpreted
only through other texts. Passages similar to NK or with some dissimilarites or
passeges extending topics touched in this text are found in other tantric texts which
reveal symbolism of each word employed in the NK. The association of letters with
deities and their creative powers is a typical feature of tantric traditions based on the
fundamental notion of the creative power of sound. However, through the text of
Nandikeśvara itself, one can notice specific ways of reasoning to connect grammatical
sūtras and tantric metaphysics. The author and his later commentator Upamanyu
establish links by means of traditional methods, which basically employ similarities
specifically found here between sūtras and cosmological models. The linking methods
include verbal similarities and other similarities in features and functions such as
grouping into units, polarisation, causality, or possibility of compression into
pratyahāras. In this way, Māheśvara sūtras in particular and grammar in general are
brought into the realm of sacred cosmology and represent cosmic creative powers.
Such sacralization does not effect implementation of the sūtras in grammatical rules,
but is supposed to reveal a metaphisical layer of the human sounds from advatistic
tantric perspective. A phoneme here is the underlying structure and animating power
of both, the language and the universe. It represents that level where a sound generates
not only words, but also what those words denote, or tattvas in our case. That is why
grasp and understanding of creative powers of tantra as phonemes grouped into sūtras,
reflecting upon it leads to elimination of seeming differentiations between part and
whole, subject and object, word and meaning, signifier and signified, cause and effect
and ultimately individual modality and its divine source.
Sources
The text of NK together with Upamanyu’s ṭīka (also called Tattvavimarśinī)
presented here is based on two manuscripts of Nepalese provenance. The manuscripts are
probably either copies of each other or they have one common source because they
represent one version of the text with practically no discrepancies in readings as well as
there are many shared errors. Collation with the published editions of NK by N. C.
Vedantatirtha30 and by K.S. Balasubramani31 has revealed that the Mss represent a distinct
line of transmission of the text. Variant readings in the NK verses tend to prove that the
Mss’s readings could exist in certain versions of the text. One of the deviations of the
Nepalese manuscripts is that they include only fifteen verses out of twenty seven in the
main text, but most of the other twelve verses appear as a part of the commentary.
Practically all the verses have variant readings some of which occur uniquely only in the
Mss. The manuscripts also depart from the editions in a way that the commentary is shorter
and the sentences seem to be rearranged, omitted, or truncated. The titles of texts cited in
10
the Mss appear differently, or partially, or are missing altogether. These could indicate that
the Mss belong to later stages of transmission based on the original, more complete text.
Therefore, I used Vedantatirtha’s version (V) as an aid to establish possibly original
readings and adopt them in cases where both Mss have errors. I have not intended to
produce a critical edition, but some editorial work was necessary to make the text coherent
for further translation. The text of the Mss is clear and close enough to V to adapt some of
its readings which possibly reflect earlier source Mss. I attempted to represent faithfully
the Mss’s version. Therefore, I preferred the Mss’s readings unless they are unacceptable.
Substantive corrections and emendations made on the basis of Vedantatirtha’s version as
well as variant readings in the NK verses have been documented in the footnotes.
Notes
1 Kiparsky, 1994.
2 Padoux, André, 1992, pg. 224.
3 Bhattacharya, J.N., pg. 947.
4 Dyczkowski, Mark, 1988, pg.108.
5 Gandharvatantram 1.9.
6 Banerji, 1992, pg. 85.
7 http://saivaphilosophy.blogspot.de/2010/01/saiva-agamas.html
8 yenākṣrasamāmnāya adigamya maheśvarāt|
kṛtsnaṁ vyākaraṇam proktam tasmai pāṇinaye namaḥ||
Deshpande, pg. 447; Cardona, 1976, pg. 179ff.
9 Pandey, Kahti Chandra, 1986, pp. 49, 50.
10 Deshpande, pg. 448. Haradatta is a South-Indian Śaiva scholar from Chola area. He wrote a
commentary Padamañjari on Kāśikāvṛtti (The date is uncertain. According to Renou, 1940 pg. 29
Haradatta lived in the 13th century CE).
11 Ibid. pg. 449a.
12 Ibid.
13 Ibid, pg. 445.
14 yasyonmeṣanimeṣābhyām jagataḥ pralayodayau (SpandaK 3)
yasyonmeṣanimeṣābhyām vyaktāvyaktam idam jagat (NK 3)
15 svecchayā svabhittau viśvam unmilayati (PratyabhijñaHr 2)
svecchayā svasya cicchaktau viśvam unmilayaty asau (NK 12)
16 Padoux, André, 1992, pg. 225.
17 Deshpande, pg. 453.
18 Ibid. pg. 453.
19 Śivaramamurti, C. 1974, pp.74, 75.
20 http://indology.info/email/members/palaniappan/patanjali/
21 Śivaramamurti, 1974, pp.74, 75.
22 http://indology.info/email/members/palaniappan/patanjali/
23 Goudriaan, Teun, 1983, pg. 23. Zvelebil, Poets, pp. 18, 73.
24 I can give an example of a few texts belonging to distinct tantric systems with emphasis on phonemic
cosmologies like Śāradātilaka tantram, Parātriśikavivaraṇa of Abhinavagupta, or Tantrasadbhāva 3.67
where the phonic energy is expressed as Kuṇḍalinī.
25 Padoux, André, 1992, pg. 49.
26Ibid., pg. X.
27Ibid., pg. XI.
28 NK: atra sarvatra sūtreṣu antyavarṇacaturdaśam|
dhātvarthaṁ samupādiṣṭaṁ pāṇinyādīṣṭasiddhaye||2||
Translation: Here in all the sūtras, the fourteen final letters are taught to achieve desired efficiency of
Pāṇini’s etc. [grammar] in [conveying] the meaning of dhātus (verbal roots).
29 For example, in Kulārṇava tantra 7.7: guśabdas tv andhakāraḥ syāt ruśabdas tan nirodhakaḥ|
11
andhakāranirodhatvāt gurur ity abhidhīyate||
Translation: The syllable gu signifies darkness, ru - what restrains it. He who restrains darkness of
ignorance is known as guru.
30 Nandikeśvara-ka śikā: with the commentary of Upamanyu. Edited by Narendra Chandra Vedantatirtha.
Calcutta: Metropolitan Printing & Publishing House, 1937.
31 Nandikeśvara-ka śikā: with the commentary of Upamanyu. Edited by K.S. Balasubramanian. Chennai:
Nandikeśvarakāśikā (also called Ādisūtrakāśikā, Ādisūtrakārikā,
Nandikeśvarakārikā) is a unique and distinctive text ascribed to Nandikeśvara (dates
unknown). Its uniqueness is at least partly attributable to the ways in which it
describes series of causal links and relationships between the Sanskrit alphabet and
stages of cosmic emanation. These stages, described in the text, as a graduated
unfoldment of categories of existence (tattvas), are synchronised with the letters of the
Sanskrit alphabet as these have been represented in the well-known Māheśvarāni
sūtrāni of Pāṇini’s Ashtādhyāyī (app. 500 BC)1, a foundational Sanskrit grammar text.
The Māheśvarāni sūtrāni, as we may know, group the alphabet into fourteen distinct
units of vowels and consonants based on their specific and internal consistencies. The
Nandikeśvarakāśikā (NK), remarkably, conjoins this unique sequence of phonological
letter-groups with a parallel emanation of sequential tattvas, or categories that are also
here significantly grouped into units. Perhaps, the intriguing feature of the cosmic
model or paradigm presented in the Nandikeśvarakāśikā is a series of complex causal
correlations between the Sanskrit phonological groups and tattvas which are
recognized here as emanating or originating from the former into existence.
Traditionally, the Māheśvarāni sūtrāni have been imbued with metaphysical
connotations specifically in the context of their sound-symbolism and association with
śabdabrahman or Word. However, such a sacred dimension to the letters in the
Pāṇinian sūtras has not been explicated or developed into a full-scale cosmogony in
any known texts prior to NK. Therefore, we find here a quite early and distinctive
usage of a ‘grammarians’ alphabet-system as a means of concrete manifestation of the
cosmos. Moreover, by establishing such links between the sūtras and cosmogony,
Nandikeśvara seems to have also created new links between the Sanskrit grammar of
Pāṇini and Śaiva Agama traditions.
In this paper, I present an edition together with English translation of
Nandikeśvarakāśikā in its distinct recession. This is a shorter version of this unique
and historically valuable text. I will try to give an outline of the available to me
research data regarding NK’s date and place. The text has a historical value because
traditionally it is considered to be an earliest text describing phonic cosmology.2 But
not only that, it is linked to the traditional grammatical elements. I also intend to
explore the specific ways in which the author, Nandikeśvara, has represented
relationships between the Pāṇinian groups of letters and groups of tattvas. I will try to
show how these groups of letters are creative cosmic principles, presided over by
specific deities, and the ways in which they serve to bring specific levels of existence
into being. It is my notion that the author may have viewed the Māheśvara sūtras,
which are groups of letters, as creative agencies, because these agencies in fact are
themselves groups, pairs, or triads that represent inherent aspects such as formless and
form, consciousness and power, will and knowledge, sovereignty or lordship and
multiplicity, etc. Instead of singularity, this text explores the universe as
fundamentally inter-relational.
3
Place and Date of Nanadikeśvarakāśikā
There seems to be much ambiguity and uncertainty with regards to the history of
Nanadikeśvarakāśikā as a text. We do know that Nandikeśvara is a name attributed to
various authors who wrote on a range of subjects. A grammarian called Nandikeśvara
is mentioned by Kṣirasvāmin in his commentary on Amarakośa and by Sayana in
Dhatuvivṛtti.3 In the canon of Jayadrathayāmala, a summary of which has been
presented by Dyzckowski,4 Nandinātha is an author of Krodhānalasaṁhitā. However,
one may not conclude from this that the Nandinātha of Jayadrathayāmala is the author
of NK.
As a name, Nandinātha, or Nandikeśvara, or simply Nandī, has several
associations: a deity associated with Śiva, a celestial being, a sage or a legendary
teacher who passed on various branches of Śaiva teachings. There are in fact several
texts where Nandī is mentioned as a legendary teacher. For instance, we know that
Nandikeśvara is purported to be the deity who once transmitted the knowledge of
Gandharvatantra on earth.5 According to Brahmayāmalatantra, Nandī was the source
who first revealed the knowledge of Raurava agama.6 Also, Nandinātha is mentioned
in the canon of Śaiva Siddhanta as a recipient of Raurava agama from the Tatpuruṣa
face of Sadāśiva.7 However, it is unlikely that these legendary authors could be
identical with our Nandikeśvara for several reasons not the least of which is the fact
that the style and the views reflected in NK are quite unique in themselves. Besides
this, it is apparent that most of the traditional canons reflected in the mentioned above
tantras and āgamas have little clear historical value.
At first glance, it may seem strange to note that perhaps our most reliable
sources of information about the actual history of the text may be some of the Śaiva
myths reflected in NK. For example, we know that there are specific myths
associating Śiva, or Maheśvara, in the form of Naṭarāja with a long-standing
grammatical tradition in Sanskrit literature. This form of Śiva is purported to have
delivered the Sanskrit alphabet to the world through the śiva sūtras. The earliest
instance of the myth of Pāṇini receiving akṣrasamāmnāya from Maheśvara occurs in
the versified Pāṇinīyaṣikśa, a text on phonetics, attributed to Pāṇini himself.8
Moreover, in the commentary of Upamanyu we find it noted that sages Nandikeśvara
and Patañjali also witnessed Śiva’s dance. Relying on this description, K.C. Pandey
concludes that Nandikeśvara was a contemporary of Patañjali. He adheres to a
traditional view that Nandikeśvara was a contemporary of both Pāṇini and Patañjali.9
However, we know that Pāṇini’s contemporaneity with Patañjali is considered to be
historically improbable by most scholars. Furthermore, the narrative of the dance
witnesses may be seen as a kind of rhetorical device for giving equal weight to the
above mentioned teachers (acaryas) rather bearing any kinship to historical facts.
However, it is interesting that Pandey suggests another ‘evidence’ that NK was known
to Patañjali. This is the occurrence of the word brahmarāśi in Mahābhāṣya (1.36) and
Kaiyata glossing the word as brahmatattva. Following Nāgeśa Baṭṭa, Pandey
concludes that Patañjali hinted at Nandikeśvara’s brahmatattva or brahmarūpa (in the
other recession) in the fourteenth verse. However, this evidence seems to be
insufficient to prove that Patañjali based himself on Nandikeśvara and not the other
way around.
It is perhaps significant that the traditional body of literature of Sanskrit
grammarians reveals little awareness of NK till much later in its development. It
4
seems that most grammarians preferred to remain either indifferent or were unaware
of any connection between Śiva and the sūtras till the time of Haradatta, a South
Indian commentator of Kāśikāvṛtti.10 It is also significant that Haradatta did not quote
the verse directly from Pāṇinīyaṣikśa, but from a compilation of traditional verses of
wisdom.11 The verse then appears as cited by Srilankan Dharmakirti in his Rūpāvatāra.
Deshpande concludes from several fragments of such evidence that the story of Pāṇini
receiving sūtras from Maheśvara was probably wide spread in southern parts of the
Indian subcontinent in the 10th century.12 However, none of the writers who
mentioned the famed Śiva-Pāṇini myth in their works actually referred themselves to
NK. It is in fact only Nāgeśa Bhaṭṭa, in his Laghuśabdenduśekhara, who actually notes
that Śiva himself offered the sūtras. Nāgesa may have done this on the basis of the
Pāṇinīyaṣikśa and the Kāśika of Nandikeśvara.13 Deshpande concludes that: “all
...factors, and especially the divine origins of Pāṇini’s grammar, are taken up by
Nāgeśa and woven into a powerful argument to explain the status and the authority of
the Sanskrit grammar.” According to him, the myth and the verse from NK about
Naṭarāja revealing sūtras has come to be widely used by traditional grammarians
from the time of Nāgeśa. However, the larger body of the text of NK seems to have
remained outside the context of grammatical literature for it seemingly has little
validity for grammar.
The connection of NK with Kashmiri Shaiva tradition shows more textual
evidences. But these evidences are interpreted differently by researchers. Pandey
points out to lines similar in NK with Spandakārika14 and with Pratyabhijñā
Hṛdayam.15 He suggests that the NK “...is the earliest voluntaristic philosophy, which
was developed... by the thinkers of Kashmir ... in the light of monism.” He concludes
about Kashmiri origins of the text as well as about Kashmir Shaivas like
Abhinavagupta and Kshemaraja drawing from NK. One can also find common with
KSH terminology and notions such as citkalā (consciousness), being the supreme Lord
and aham (absolute I).16 Yet, it could be Nandikeśvara echoing Kaśmiri texts and not
vice-verse. Moreover, the text seems to be not known to Kashmir Shaivas. Rafaele
Torella in his interview to Deshpande sheared his impressions about later character of
the text: "I don't remember seeing any reference to Nandikeśvara's Kāśikā in any
Kashmir Shaiva texts (even where one might most have expected to find it, like the
Paratrimśikavivārana or Tantraloka, ch. III)... it is not an old text,.. but a late one
which rather presupposes the doctrines of the so called ‘Kashmir Shaivism’ at a postexegetical
stage.” Therefore, taking into consideration a striking resemblance of NK
with some verses from Kashmiri texts, it is more likely to conclude that NK was
written after Abhinavagupta’s time, that is not before the eleventh century CE.
In his article, Deshpande agues that NK is a later text which could be composed
in Tamil Nadu after 800 CE.17 In regard to Southern origins of the text and not
Kashmirian, Deshpande points out that the first verse of NK mentions Naṭarāja, whose
worship was not known in Kashmir, but was spread in Chidambaram region.
Additionally, Upamanyu mentioned among sages not only Nandikeśvara but also
Patañjali and Vyāghrapād, whose shrines are found in the yard of Naṭarāja’s temple in
Chidambaram.18 The story of Patañjali and Vyāghrapād is as early as the earliest
temple of Śiva in Chidambaram founded by Hiraṇyavarman of the early Pallava
dynasty.19 So, Chidambaram seems to be the most possible place where the Naṭarāja-
Pāṇini myth was developed.
5
Similar conclusions can be made from the paper by S. Palaniappan available on
the Indology webpage.20 On the basis of the Tamili literary tradition, Palaniappan adds
that there are certain elements in Tamil mythology that could contribute to the myth of
Pāṇini and Śiva. For example, a cult of Dakṣiṇamūrti of Madurai created a similar
myth of Tamil grammar taught by Shiva to Agastya. Then a dance treatise Tamil
kuutta nUl says that from one side of Śiva's drum, Sanskrit letters come and Tamil
letters come from the other. Interestingly, the story of the Lord Naṭeśa teaching Pāṇini
fourteen sūtras was already known in the early Chola period of worship of the Lord
Dakṣiṇamūrti, whose cult was supported by the Pandyan dynasty. There is a place of
worship of the Lord Vyākaraṇadānaperumāl called Vyākaraṇasāna-mandapa, a place
where Śiva is believed to have appeared before Pāṇini.21 Palaniappan says that the
myth of Śiva-Pāṇini connection could serve as a political tool to support Shaiva rulers
of Chidambaram in their political competition with Pandyan rulers of Madurai from 6th
to 10th centuries. This is a very possible period when NK could be composed to
increase the authority of Chidambaram with not less powerful religious figures, cults,
and doctrines.
One more evidence of Nandikeśvara’s connection to Chidambaram comes from
the Shaiva Siddhanta tradition. Nandikeśvara is considered to be a legendary founder
of some Shaiva Siddhanta schools. Tirumūlar, the author of Tirumantiram in the Tamil
language where much of the Shaiva Siddhanta provisions have been formulated, is
considered to be a direct disciple of Nandikeśvara. In his work, Tirumūlar names those
who received Śiva’s grace at Chidambaram. Among them are the four kumāras
beginning with Sanaka, then Panañjali and Vyāghrapāda.22 Nandikeśvara also speaks
of sages beginning with Sanaka who benefited from the Śivasūtras revealed by
Naṭarāja. Upamanyu extends this list to Panañjali and Vyāghrapāda. So, there is not
only a claim of succession in lineage, but also the common mythology connected to
the shrine of Naṭarāja in Chidambaram which has been employed by the authors from
this lineage. Goudriaan places Tirumular in the seventh century of CE, on the basis of
evidences presented by Zvelebil.23 This can lead to a conclusion that Nandikeśvara was
either Tirumular’s contemporary or lived before him. A further study of references to
NK and its myths in Tamil sources as well as relations between Tamil Siddhas and
Kashmir Pandits could reveal more details regarding origins and the role of this text.
Māheśvarāni Sūtrāni and their role in Nandikeśvarakāśikā: a tantric
connection?
An introduction to the māheśvarāni sūtrāni in the Pāṇinian system leads one
quickly to a recognition and understanding that the unique clustering and arrangement
of the Sanskrit letters has been designed here to serve a rather precise function. The
Sanskrit syllabary, as we know, has been reorganised in these well-known sūtras and
placed into discrete units, or groups, according to their phonetic characteristics. For
example, we find that semivowels are placed together in one group, the nasalised
letters are in another, aspirated are in yet another group, and so on. It is also quite
well-known that these groupings carry the possibility for an efficient implementation
of grammatical rules of Sanskrit as a language.
6
For example, in the Pāninian system, the grammatical rule 8.4.53 jhalāṁ
jaśjhaśi requires that voiced un-aspirated consonants (jaś) replace spirants (jhal)
occurring before voiced consonants (jhaś). Here, jaś jhal and jhaś are pratyahāras, or
compressed groups of letters. They are formed with the help of a final extra consonant
which is a ‘terminal sign’, or a marker, called anubandha. These markers serve as a
means of formation of a coding system expressed through phonetic keys called
pratyahāras, or compressed groups of letters. For example, the four sūtras - a i u ṇ ṛ ḷ
k e o ṅ ai au c - are abbreviated into one pratyāhāra ‘ac’ where ‘a’ is the initial vowel
and ‘c’ is the final marker, and the whole pratyāhāra means ‘all vowels’. Similarly,
‘jaś’ would mean consonants j b g ḍ d where ‘a’ is added for pronunciation purposes
and in order not to mix the letters with anubandhas.
In this way, instead of enumerating all the phonemes significant for
grammatical operations, Pāṇini uses one short key-word. Then, the key-words are used
to form grammatical rules and technical terms such as ‘aj-anta’ or ‘hal-anta’ (nominal
stems ending in vowels or ending in consonants). So, in the grammatical context, the
rearrangement of phonemes and the use of markers are highly valid for they serve to
form meta-linguistic elements which contribute to efficiency and brevity of the
grammatical text. Such a system of coding is rational and each symbol has a concrete
semantic meaning.
An interesting feature of Nandikeśvarakāśikā as a text is how such a Pāṇinian
grammatical-system of grouped letters has been linked here to a seemingly distinct
system of the categories of cosmic emanation known - mostly in tantric cosmology - as
tattvas. It is as if the author of this text believed that there was a kind of similarity and
sympathetic resonance between these two quite different systems - one relating to
core-rules of Sanskrit grammar as a language and the other to hidden grammatical
‘rules’ of cosmic manifestation. Both of these systems seem to involve the ability to
distinguish specific groups or units - letters in one case and tattvas, or categories of
creation, in the other - based on an inbuilt universal logic, or grammar, if you will. By
means of such a ‘grammar’, both language and physical creation gain an internal
consistency of laws, rules, and regulations from which they emerge and which are
maintained through countless instances of their application in reality.
In several systems of tantric metaphysics, phonemic cosmology is an essential
feature.24 The phonemes are perceived as being alive, filled with power, and as sacred
beings of some kind – deities, in fact. Traditionally, such a sacralization of Sanskrit
phonemes is rooted in a concept of ‘sonic power’ (vāc śakti) wherein each letter of the
alphabet is synchronized with a vibratory pattern of sound that articulates a particular
aspect of reality into existence.25 In this context of creative power, the phonemes
acquire a very specific meaning which is not the same as linguistic, or semantic,
meaning. The attribution of meaning to the Sanskrit alphabet there depends on
religiously constructed representational power of sound. A reference point of each
letter then, becomes a stage, or a station in the process of unfoldment of a hierarchical
cosmos. Therefore, in order to explain and to understand the sacredness of phonemes
and their meaning, each phoneme ought to be seen in a cosmological context. Each
part of the Sanskrit alphabet here becomes more like a symbolic sign endowed with
creative or destructive power and serves a cosmological function within the tantric
paradigm.
Andre Padoux, who has written extensively on tantric cosmology and the
concept of speech or ‘vāc’, describes the latter as a ‘creatrix’ and “a symbol of
7
…Godhead… the force that creates, maintains, and upholds the universe.”26 In his
seminal work, ‘Vāc:.. in selected Hindu Tantras’, Padoux shows how this concept of
‘vāc’ or ‘Word’ is expressed in tantric systems at two overlapping levels - as a cosmic
manifestation and as a human act of speech:
All… developments of the Word… occur homologously within man or the
cosmos…Thus…the creative act (is) an utterance which is a human act…reversing the
order (we) see in this act nothing but the reproduction at the human level of an
archetypal, divine act or process… the universe emerges within divine consciousness,
through… stages or levels of speech…the categories…of the cosmic manifestation arise
concurrently with the Sanskrit phonemes (varṇa) arranged in their grammatical order,
while grammar – as well as traditional phonetics - will serve to account for the
cosmology.27
The Sanskrit phonemes, or varṇas, represent sound-elements which constitute
the world of expressed objective forms (vācya) including all the basic elements and
categories of creation from the mahābhutas to the tattvas of human senses (indriyas)
and inner mental organs (antaḥkarana). The grammatical units of speech, in this view,
are transformed into symbols that express a functional cosmology. Such a tantric
approach to the letters of the Sanskrit alphabet is somehow similar to mantras,
especially bija mantras, which do not have semantic meaning and are called as nonlinguistic,
but which are meaningful because they serve the purpose of mapping and
representing an emergent universe.
If we keep to this perspective, then it is logical that a text like the NK could
find deeper correlations between the grammatology of the Māheśvara sūtras and their
cosmological role as creative matrices of a manifesting universe. To articulate this
deeper dimension and function of Māheśvara sūtras and to distinguish it from the
purely grammatical level of language, Nandikeśvara says that anubandhas are for
grammatical use only while the letters have deeper esoteric meanings beyond this.28
Nandikeśvara’s unique methodology of this sacred dimension seems to be in a
functional complementarity to the existing arrangement and grouping of the Sanskrit
letters as we find them in the Māheśvara sūtras. A remarkable feature here is that the
cosmological grammar of emanating levels of being – tattvas - appears altogether in
concert with the linguistic component and its ‘grammar’. He emphasizes the
importance of grouping of individual letters and syllables into units, or sūtras, where
each element in a unit is inseparably related to the others. Here, a sūtra, as a composite
of letters, represents a complete stage in the creation. This may be because the creative
agencies – sounds or letters as deities - and the constituent elements of the universe
emanating from them are envisioned to appear entirely in, and as, groups or units.
Both sūtras and the cosmological categories are found grouped in pairs, triads,
in units of five and even in units of eight. Just as in the sūtras, where the letters in
each group have common features, so also apparently do the tattvas. The relationships
between the elements in one unit can be of different kinds but are always close and
inseparable so that one cannot exist without the others. Within the groups of letters -
especially at the level of vowel-units - there tends to be an inherent or inbuilt polarity
of pairs between each creative agency, or deity. So, for example, the first group of
letters - ‘aiuṇ’ - is presided over by Brahma who signifies the a-kāra principle which
is formless and constitutive of pure awareness or pure knowledge (jñaṇamaṭram,
citkalā). This ‘alpha’ deity is coupled with the next letter of the group - ‘i’ - that is
female and signifies the pure energy of will or desire (icchā śakti). Their togetherness,
8
or androgyny, is expressed through the third letter of the group - u-kāra - which is
expressed in the form of Viṣṇu, the ‘Pervader’.
This tendency to represent an inherent gender-based polarity within each group
of letters and within their representative deity persists like a strain throughout the text
and is repeated in different ways in each successive sūtra. Within the groups of letters,
consisting of even more than two, we notice that this polarity of gender between letters
may somehow be critical to the morphology of the cosmic creative process. Each
created unit seems to have two aspects: a pure awareness factor, or cognitive factor,
which tends to be masculine in nature and a corresponding dynamic movement
towards proliferation, multiplicity, and objectivity that is somehow feminine.
Such repeated polarities indicate that both aspects are constitutive to the very
order of creation. On the one hand, creation may be viewed in purely process terms,
implying dynamic change and movement from a subtle state to increasingly gross and
fragmented levels of being. On the other hand, the creative agency somehow exceeds
and transcends the frontiers of the created world, and may be described as a
"substance of substances", implying statism and inherent existence from its own side
(svabhāva). Therefore, there is a polarity between two forms that a deity assumes:
Brahma and Māyā, Puruśa and Prakṛtī, Meśvara and Maheśvarī. These two poles have
a strong tantric resonance and remind a play of creation through the exchange between
Śiva and Śakti. These two aspects are always conjoined and explicitly recognised in
the texts as inseparable from each other.
An even more interesting feature of NK is the element of causal relationship
introduced in the text as a link between the two corresponding ‘grammars’ - of
language and of cosmic manifestation through tattvas. We note that a kind of primacy
seems to be given here to the former as effective cause, source, and origin, of the
latter. The letters are somehow viewed as inherently potent and imbued with power -
in fact, they possess creative agency (karaṇatva) to sound the entire world into being.
As creators, or cosmic actors, the letters are in themselves full-fledged sacred deities
with names like Brahma, Maheśvara and Viṣṇu. The letters represent or express the
deities and their causal functions. In order to underline causative powers of the letters,
Upamanyu, the commentator of Nandikeśvara, gives an example of the letter ‘e’
which purportedly derives from a combination of letters ‘a’ and ‘i’. According to his
reading of the Nandikeśvara text, the letters in the very first sūtra, ‘have causative
power to produce other letters’. Similarly, these first letters, known as Brahma and his
power of will (icchā śakti), operate through distinct stages of manifestation of the
cosmos originating the succeeding stages of enfoldment of the universe.
One must note at this point, that as we follow the text through its intricate
patterns of sound, their deities, and the issuing cosmic tattvas, there persists a pattern
of continuity between the two operative systems – of language with its core-rules of
grammar and categories of cosmic emanation. Both share seeming similarities in
activities, functions, or traits and are continuously interwoven with each other. One of
the distinct ways to identify letters as deities or tattvas is to involve verbal or phonetic
similarities between two systems. For example, the words agra (first) and akṣara
(imperishable) traditionally characterise the Absolute, or Brahma. The word agra is
also a characteristic of the letter ‘a’ and the word akṣara means ‘a letter’ in Sanskrit.
Similarities in words characterising two entities here lead to identification of the letter
‘a’ with the Absolute. Even more interesting feature is establishing links based on
similarity in one letter. As in case of the sūtra ‘kapay’, the letter ‘pa’ represents
9
puruṣa, or the letter ‘sa’ represents sattva in the sūtra ‘śaṣasar’. Here, it seems that the
meaning of the word emerges from the initial syllable the word. Such attribution of
meaning to sounds echoes traditional tantric views that constituent syllables or sounds
of a word somehow possess lexical meaning which contributes to the meaning of the
whole word.29 By means of identification of syllables with words, it seems these
syllables become identified with categories and things those words denote. Thus,
connections between letters and words bring the alphabet into the realm of
metaphysics.
To sum up, the Māheśvara sūtras in NK depart from its purely grammatical
functions and appear to be a symbolic system with references which can be interpreted
only through other texts. Passages similar to NK or with some dissimilarites or
passeges extending topics touched in this text are found in other tantric texts which
reveal symbolism of each word employed in the NK. The association of letters with
deities and their creative powers is a typical feature of tantric traditions based on the
fundamental notion of the creative power of sound. However, through the text of
Nandikeśvara itself, one can notice specific ways of reasoning to connect grammatical
sūtras and tantric metaphysics. The author and his later commentator Upamanyu
establish links by means of traditional methods, which basically employ similarities
specifically found here between sūtras and cosmological models. The linking methods
include verbal similarities and other similarities in features and functions such as
grouping into units, polarisation, causality, or possibility of compression into
pratyahāras. In this way, Māheśvara sūtras in particular and grammar in general are
brought into the realm of sacred cosmology and represent cosmic creative powers.
Such sacralization does not effect implementation of the sūtras in grammatical rules,
but is supposed to reveal a metaphisical layer of the human sounds from advatistic
tantric perspective. A phoneme here is the underlying structure and animating power
of both, the language and the universe. It represents that level where a sound generates
not only words, but also what those words denote, or tattvas in our case. That is why
grasp and understanding of creative powers of tantra as phonemes grouped into sūtras,
reflecting upon it leads to elimination of seeming differentiations between part and
whole, subject and object, word and meaning, signifier and signified, cause and effect
and ultimately individual modality and its divine source.
Sources
The text of NK together with Upamanyu’s ṭīka (also called Tattvavimarśinī)
presented here is based on two manuscripts of Nepalese provenance. The manuscripts are
probably either copies of each other or they have one common source because they
represent one version of the text with practically no discrepancies in readings as well as
there are many shared errors. Collation with the published editions of NK by N. C.
Vedantatirtha30 and by K.S. Balasubramani31 has revealed that the Mss represent a distinct
line of transmission of the text. Variant readings in the NK verses tend to prove that the
Mss’s readings could exist in certain versions of the text. One of the deviations of the
Nepalese manuscripts is that they include only fifteen verses out of twenty seven in the
main text, but most of the other twelve verses appear as a part of the commentary.
Practically all the verses have variant readings some of which occur uniquely only in the
Mss. The manuscripts also depart from the editions in a way that the commentary is shorter
and the sentences seem to be rearranged, omitted, or truncated. The titles of texts cited in
10
the Mss appear differently, or partially, or are missing altogether. These could indicate that
the Mss belong to later stages of transmission based on the original, more complete text.
Therefore, I used Vedantatirtha’s version (V) as an aid to establish possibly original
readings and adopt them in cases where both Mss have errors. I have not intended to
produce a critical edition, but some editorial work was necessary to make the text coherent
for further translation. The text of the Mss is clear and close enough to V to adapt some of
its readings which possibly reflect earlier source Mss. I attempted to represent faithfully
the Mss’s version. Therefore, I preferred the Mss’s readings unless they are unacceptable.
Substantive corrections and emendations made on the basis of Vedantatirtha’s version as
well as variant readings in the NK verses have been documented in the footnotes.
Notes
1 Kiparsky, 1994.
2 Padoux, André, 1992, pg. 224.
3 Bhattacharya, J.N., pg. 947.
4 Dyczkowski, Mark, 1988, pg.108.
5 Gandharvatantram 1.9.
6 Banerji, 1992, pg. 85.
7 http://saivaphilosophy.blogspot.de/2010/01/saiva-agamas.html
8 yenākṣrasamāmnāya adigamya maheśvarāt|
kṛtsnaṁ vyākaraṇam proktam tasmai pāṇinaye namaḥ||
Deshpande, pg. 447; Cardona, 1976, pg. 179ff.
9 Pandey, Kahti Chandra, 1986, pp. 49, 50.
10 Deshpande, pg. 448. Haradatta is a South-Indian Śaiva scholar from Chola area. He wrote a
commentary Padamañjari on Kāśikāvṛtti (The date is uncertain. According to Renou, 1940 pg. 29
Haradatta lived in the 13th century CE).
11 Ibid. pg. 449a.
12 Ibid.
13 Ibid, pg. 445.
14 yasyonmeṣanimeṣābhyām jagataḥ pralayodayau (SpandaK 3)
yasyonmeṣanimeṣābhyām vyaktāvyaktam idam jagat (NK 3)
15 svecchayā svabhittau viśvam unmilayati (PratyabhijñaHr 2)
svecchayā svasya cicchaktau viśvam unmilayaty asau (NK 12)
16 Padoux, André, 1992, pg. 225.
17 Deshpande, pg. 453.
18 Ibid. pg. 453.
19 Śivaramamurti, C. 1974, pp.74, 75.
20 http://indology.info/email/members/palaniappan/patanjali/
21 Śivaramamurti, 1974, pp.74, 75.
22 http://indology.info/email/members/palaniappan/patanjali/
23 Goudriaan, Teun, 1983, pg. 23. Zvelebil, Poets, pp. 18, 73.
24 I can give an example of a few texts belonging to distinct tantric systems with emphasis on phonemic
cosmologies like Śāradātilaka tantram, Parātriśikavivaraṇa of Abhinavagupta, or Tantrasadbhāva 3.67
where the phonic energy is expressed as Kuṇḍalinī.
25 Padoux, André, 1992, pg. 49.
26Ibid., pg. X.
27Ibid., pg. XI.
28 NK: atra sarvatra sūtreṣu antyavarṇacaturdaśam|
dhātvarthaṁ samupādiṣṭaṁ pāṇinyādīṣṭasiddhaye||2||
Translation: Here in all the sūtras, the fourteen final letters are taught to achieve desired efficiency of
Pāṇini’s etc. [grammar] in [conveying] the meaning of dhātus (verbal roots).
29 For example, in Kulārṇava tantra 7.7: guśabdas tv andhakāraḥ syāt ruśabdas tan nirodhakaḥ|
11
andhakāranirodhatvāt gurur ity abhidhīyate||
Translation: The syllable gu signifies darkness, ru - what restrains it. He who restrains darkness of
ignorance is known as guru.
30 Nandikeśvara-ka śikā: with the commentary of Upamanyu. Edited by Narendra Chandra Vedantatirtha.
Calcutta: Metropolitan Printing & Publishing House, 1937.
31 Nandikeśvara-ka śikā: with the commentary of Upamanyu. Edited by K.S. Balasubramanian. Chennai:
Kuppuswami Sastri Research Center, 2009.
No comments:
Post a Comment