Devotional, Covenantal and Yogic:
Three Episodes in the Religious Use of Alphabet and Letter
from a Millennium of Great Vehicle Buddhism.*
Dan Martin
Version: April 24, 2004
Devotional, Covenantal and Yogic: Three Episodes in the Religious Use of Alphabet and Letter from a Millennium of Great Vehicle Buddhism. Contained in: Sergio La Porta and David Shulman, eds., The Poetics of Grammar and the Metaphysics of Sound and Sign, Jerusalem Studies in Religion and Culture series no. 6, Brill (Leiden 2007), pp. 201-229. The published version is the one that ought to be cited, and not this pre-published version.
That would be like teaching the alphabet to a Buddha!
— Tibetan proverb.[1]
While it is true that teaching the alphabet to people who already know so much more would be frivolous if not laughable, in the pages ahead the alphabet as such will be taken very seriously, and not simply assumed. Far from being 'simple,' it was clear even before beginning the first word of this paper, that the topic is far too broad, and lined with intriguing side paths branching off in many directions. As a time span, a millennium naturally resists encapsulation and invites sketchiness and generalizations which we should do our best to resist, especially during the process of succumbing. The discussion will be divided into three parts, roughly dividing the millennium into thirds: 1. the first three centuries of the common era, 2. the next three centuries, and 3. the following centuries ending in the vicinity of the eleventh century. Perhaps these time spans do somehow, or at least well enough for present purposes, correspond to the three themes of this paper: devotional, covenantal and yogic.
For the many who in some degree or another appreciate Buddhism as a philosophy, and dislike what they know as 'religion,' no offense is intended. Of course, Buddhism has much philosophy in any sense of the term. But for now we will be looking at aspects of Buddhism that are very likely to be overlooked by the philosophers. Alphabet usages such as those considered here certainly interconnect in various interesting ways with Buddhist philosophy, psychology, ethics, language science and so forth, but for economy of time, space and ability, it will not be possible to say very much along these lines. We will look rather more at things that might be termed, in the broader and older (and most definitely not the recent socio-political or journalistic) sense of the word, 'cultic.' To emphasize the cultic just means to attempt to explore the areas surrounding religious practices, and especially practices intended to honor whatever is most highly regarded in a particular religion. In the beginning it should suffice to suggest that, as a general principle, the devotional and other religious usages of letters are in every case somehow and in some degree tied up with or inspired by the sacredness of scriptural texts (whether orally recited or written), as well as the sacredness of the figure of the Buddha Himself. At times, like full-blown religious symbols, or even like physical relics, the letters may place believers directly in contact with sacramental powers or blessings. But that being admitted, my own ideas about the general picture are constantly shifting, perhaps even shifting during the act of writing. Nothing is permanent, and least of all, structures.
Part One: Devotional.
In December of 2002, I visited the ruins of Kapilavastu, which have never been properly excavated, even if some ruined buildings and gateways have been exposed. Well, it is at least the Kapilavastu on the Nepalese side of the border, since India also lays claims to the Buddha's childhood home.[2] The following story, told in the tenth chapter of His biography as found in the Lalitavistara Sūtra, takes place in Kapilavastu. One day the young prince, and future Buddha, Siddhārtha set out, at the head of a procession of ten thousand children, to visit an elementary school headed by Viśvāmitra ('Everybody's Friend').[3] The future Buddha really had no need to go to school, of course, and He immediately demonstrated to the schoolmaster Viśvāmitra that He already knew about sixty-four different scripts (the names of the scripts are listed). The children together recited the Sanskrit alphabet, and after each syllable, through the blessings of the future Buddha, a phrase rang out as if from nowhere, one which began with that same syllable. In effect, this appears to be very much like the well-known English Abecedarium "A is for apple. B is for boy. C is for cat." Only in this case the usual Sanskrit alphabet is used, and each letter comes at the beginning of a word or phrase expressing a basic Buddhist concept:
When they pronounced the short A, the sound of this phrase emerged: "All compounded things are impermanent" (Anityaḥ sarvasaṃskāraḥ).
When they pronounced the long Ā, the sound of this phrase emerged: "Nonself" (Ātmaparahita, 'beyond self' or 'welfare of self and others').
When they pronounced the short I, the sound of the phrase emerged: "Sense faculties are vast" (Indriyavaipulya).
When they pronounced the long Ī, the sound of this phrase emerged: "Beings have many contagious diseases" (Ītibahula, 'hosts of calamities')...[4]
The text continues similarly through the letters of the Sanskrit alphabet. The list of sixty-four (there are actually sixty-eight in the Tibetan version) scripts is very interesting in itself, although we will not go deeply into it.[5] It includes not only ordinary human scripts used in our world, but scripts used in other parts of the universe, and scripts of various non-human entities. It begins with Brāhmī and Kharoṣṭhī, two very early Indian scripts, but also mentions scripts of south India, what may be Greek script (Yavana), and so forth. The Sanskrit script as we know it today, called Devanāgarī, is not to be found among them, and it is essential to be aware that the Sanskrit letters did not exist in their current shapes until relatively recently, and Devanāgarī became the dominant script for writing Sanskrit only in around the 18th century. It is perhaps worth noting, too, that some very good scholars believe that the Kharoṣṭhī script descended from an eastern Aramaic script. Like Semitic scripts, Kharoṣṭhī was written from right to left. Brahmī script, which as far as I know originated in India, is written from left to right.[6]
John Brough's 1977 article clearly demonstrated that the earlier of the two Chinese translations of the Lalitavistara Sūtra, done by Dharmarakṣa in the year 308 ce, employs an entirely different alphabet in this passage. It most certainly is not an 'alphabet,' in the sense that it represents all the letters used to write any particular language. Of the several learned articles written on what is now known as the Arapacana syllabary, only the earliest ones called it, inaccurately, the Arapacana alphabet.[7]
Incidentally, today every Tibetan knows the Arapacana primarily as part of a mantra invoking the Bodhisattva of wisdom and learning, Mañjuśrī.[8] Several years ago, I spent a summer in Himachal Pradesh, at the town of Gangcan Kyishong just above Dharamsala and just below the residence of the His Holiness the Dalai Lama. I remember several times being awakened in the morning by the sound of a child shouting very loudly from the rooftop "Oṃ A-ra-pa-tsa-na Dhiḥ! Dhiḥ! Dhiḥ! Dhiḥ! Dhiḥ! Dhiḥ! Dhiḥ! Dhiḥ!" The syllable Dhiḥ, which was repeated in shrill and rapid machine-gun-like bursts until running out of breath, is considered by the experts (not necessarily so by the child) as a 'seed-syllable' for generating the visualized form of Mañjuśrī, the Bodhisattva of wisdom and learning, although it surely derives from the Sanskrit root dhī, which means 'to think.' A-ra-pa-tsa-na is just a Tibetan pronunciation for Arapacana. This mantra clearly means something — well, at least one should not really think of it as made up of nonsense syllables, or that it is verbalized without intentionality and purpose — and part of that meaning is surely to be found in the history of the Arapacana syllabary itself. That contemporary Tibetan schoolchildren, in the morning before going to school, might be heard reciting the first part of an 'alphabet' used in a two-millennia-old story of Buddha's school visit is certainly in itself an impressive feat of cultural memory. We should add that this practice is not done only by children, but by monks as well.[9]
Of the other Buddhist scriptures in which the Arapacana syllabary appears,[10] probably the most important of them, the 25,000 Perfection of Insight Sūtra, was translated by Dharmarakṣa into Chinese in the year 286, so we may be quite sure that we are dealing with scriptures available somewhere in the Indian subcontinent in the earliest centuries of the Common Era. Here is the beginning of the 25,000 passage:
The syllable [letter] A is access point of all dharmas, since they are from the beginning unproduced (Ādyanutpannatva).
The syllable RA is access point of all dharmas, since they are free of impurity ['dust'] (Rajo 'pagatatva).
The syllable PA is access point of all dharmas, because they point to the ultimate truth (Paramārthanirdeśa).
The syllable CA is access point of all dharmas, because of the nonapplicability of death and rebirth (Cyavanopapatty-anupalabdhitva, or 'there is apprehension neither of decease nor of rebirth').[11]
But there are several indications that these sūtras, along with the Arapacana syllabaries they contain, originated (or at least were redacted) in the northwestern part of the Indian subcontinent. Not least of the pieces of evidence, we find among the famous stone-carved reliefs of Gandhāra depictions of the visit to Viśvāmitra's school. In these friezes, which have been studied by Richard Salomon, we see the young Siddhārtha seated with a stylus in His hand ready to make letters on a slate that has a handle extending out one side (shaped rather like a cricket bat). By the way, just a few years ago I saw writing slates that work in the same way, with the same basic shape, being used by young student monks in Lhasa, Tibet. At least they work in the same way. One does not write on them with sticks of chalk as we do on modern blackboards. Instead one must evenly coat the surface with powdered chalk, which is then scratched away with a stylus so that the dark background is exposed. It is an excellent way to practice calligraphy without wasting precious writing materials. To return to Gandhāra, when letters are depicted on these slates they are in Kharoṣṭhī script, showing letters from the beginning of the Arapacana syllabary in their proper sequence. Therefore, the schoolhouse narrative of the earliest Chinese translation of the Lalitavistara Sūtra, in its use of the Arapacana syllabary, finds outside verification in stone friezes from Gandhāra in roughly the same centuries, and these friezes contain some part of the Arapacana syllabary in the Kharoṣṭhī script, the script that was in wide use in Gandhāra in those times. Salomon makes a good argument that the local Gāndhārī dialect, written in Kharoṣṭhī script, is behind the Arapacana syllabary (but, beware; it is most definitely not the case that the Arapacana is an alphabet of Kharoṣṭhī). The replacement of the Arapacana by the regular Sanskrit alphabet in the story of Siddhārtha's day in school would be just another example of the process of Sanskritization (making into more perfect Sanskrit) that many other Great Vehicle (Mahāyāna) scriptures underwent during their textual transmissions.
Salomon also studies another inscription carved on the back side of a frieze with a different subject (perhaps the presentation of a bride to Siddhārtha). It has not only a part of the beginning of the Arapacana syllabary, but also an array of numbers off to one side. In this particular case, Salomon raises the possibility that the letters and numbers might have somehow been meant to serve the builders as a key to the arrangement of scenes on the wall. Salomon does not necessarily believe this is the correct explanation (after all it is not at all clear how this would have worked... Were other friezes actually marked with numbers and syllables? We are not told), and other possibilities remain to be explored. Khettry (2001) suggests that the inscription was placed on the stone for the sake of gaining merit, and for all we know, this explanation might supply sufficient motive. Today we will put aside the many philological complications and leave some of the larger questions in abeyance. Even if the specific mysteries cannot all be solved, I believe it is possible to bring more light to the general principles involved.
Although the possibility had been raised in the 1950's by Thomas and Lamotte, John Brough (in 1977) was perhaps the first to clearly articulate the idea, recently endorsed as the most likely explanation by Salomon, which is that the Arapacana syllabary originated in a list of significant words, or head-words, taken from some so-far unidentified Buddhist scripture. The head-syllables of these significant words (and phrases) were then abstracted to form a mnemonic key for remembering the scripture (or passage of scripture) as a whole. In other words, by memorizing the syllabary, the entire text can be brought to mind.
There are certain key terms used in these early Great Vehicle scriptures that I believe need to be understood a little better. In the first place, however much we may insist on it, there is no necessary difference between an alphabet and a syllabary in Sanskrit. Both are called by the name mātrikā, a word which we might translate by 'grandmother' (the Tibetan uses two different translations in different contexts, both of which may also be translated as 'grandmother'), but also when used figuratively, as 'source' or 'origin' (very much like the English cognate 'matrix').[12] Secondly, in Sanskrit, unlike English, it is not usually necessary to distinguish between (written) letter and (spoken) phoneme. After all, unlike western Eurasian alphabets, the letters of the Sanskrit alphabet are scientifically arranged according to their phonological values and unlike English they remain consistent in their pronunciation. The Sanskrit word akṣara, which literally means 'imperishable' or 'unalterable,' may refer to both 'letter' and 'syllable,'[13] and the sense of 'inalterability' would seem to refer to both the consistency of pronunciation as well as the 'irreducibility' of the syllable as the smallest possible bearer of meaning, and for most practical purposes indeed identical to the letter. It is clear that in the Lalitavistara, the syllables were both written and sounded out, but in other contexts we may be left guessing whether the visible letter or the audible sound or both might be intended.
Therefore, when the 25,000 Perfection of Insight, in introducing the Arapacana syllabary, describes the syllables as dhāraṇī-mukha, I believe that some other translations used in the past are inadequate. This word mukha may mean 'head' (but here 'face' would be more accurate) or 'mouth,' as others have translated it, but I follow the Tibetan in understanding it to mean 'door' or 'gate,' and a bit more abstractly 'access point,' all these translations being indeed possible for the Sanskrit mukha as well. Meaningful translation of the word dhāraṇī has proven especially difficult, so much so that it is generally left untranslated. It shares the same root {dhṛ} with the word Dharma. Dharma is the usual word for Buddhism as a whole, for scriptures [the Buddha's Word], and for sets of factors that go together to sustain the vicious circle of everyday suffering called saṃsāra, as well as sets of factors that go together to sustain the path to the cessation of suffering called nirvāṇa. In the 19th century it was usual to translate Dharma as 'law.' I think something like 'sustaining factor' could make good sense in many contexts (it also avoids prejudicing the very basic Buddhist principle of impermanence). Similarly dhāraṇī,[14] with the same root, also refers to a kind of 'holding,' but in this case serves as a shorthand for 'holding in memory.' To make it simple, a dhāraṇī is a string of syllables which, either individually or as a whole, induce recollection of:
1. particular dharmas as just described,
2. groups of such dharmas as well as dharmas in their entirety,
3. a scriptural text or passage (also called Dharma),
4. a set of Buddhist concepts (which may also summarize a scriptural text or passage).
One early Great Vehicle scripture, The Teachings by Akṣayamati — its oldest existing Chinese translation made by Dharmarakṣa in 308 — defines dhāraṇī as inextricably bound up with memory: "Dhāraṇī means that, by virtue of recollecting the virtuous roots that have been accumulated in the past, one holds the 84,000-dharma heap, one retains all of it, one does not forget, and one holds it correctly in the memory. That's what dhāraṇī means."[15] It continues by explicitly saying that the Word of Buddhas and Bodhisattvas is what is entirely held in the memory (scriptures were already indicated, in fact, by the words '84,000-dharma heap').
The Suvarṇaprabhāsa Sūtra (first translated into Chinese in the early fifth century), in its chapter on bathing rites, not only tells us what dharaṇīs do, but links them with the scripture reciters:
The goddess Sarasvat[ī], covering one of her shoulders with her outer garment, and placing her right knee on the ground, with folded hands asked the Lord's permission to wind up the net of illusions, spread round the chanter of the sutra (Dharmabhāṇaka), to grant him the Dhāraṇī, and to show him the light of true knowledge. "I shall," said she, "restore the words or consonants that may have dropped from the great Sūtra. I shall grant him the Dhāraṇī that his memory may not fail. I shall teach him the mode of holy bathing which will enable the Sūtra to endure for a long time on earth, sowing the seeds of immense good, which will enable numberless creatures to cultivate their intellect, to learn various Śāstras, and to acquire immense merit."[16]
Notice that the word dhāraṇī is often found paired with another word, pratibhāṇa. Jens Braarvig of Oslo has done a study persuasively showing how in its many contexts this pair corresponds quite nicely with western ideas from Greek and Roman times on, about two of the principle parts of rhetoric: memory and eloquence.[17] The Tibetan (spobs-pa) and Sanskrit words which Braarvig translates as 'eloquence' are used in contexts that suggest a sense of outstanding ability, fluency, freeness, boldness, and in some contexts, more specifically, the ability to keep speaking without running out of things to say. The Teachings by Akṣayamati itself associates 'eloquence' with continuity, rapidity, lack of confusion, happiness, sharpness and the like. In short, the word pratibhāṇa does contain all the elements we normally associate with our idea of eloquence. This close pairing of the two concepts appears in many other Great Vehicle scriptures, among them the 25,000 Perfection of Insight Sūtra.[18]
Of course, beyond and apparently quite apart from these usages of the word dhāraṇī, there is a particular class of Buddhist scriptures that emerged early on, and gradually gained autonomy from about the end of the 3rd century, called dhāraṇī-sūtras, on which we should spare a few words.
Part Two: Covenantal.
Dhāraṇī-sūtra titles are the most numerous among the several classes of scriptures found in the Tibetan scriptural canon. Almost always extremely brief, they are very often, but not exclusively, devoted to worldly fears and other rather secular concerns. For example, there are dhāraṇīs against snakebite, against backbiting and slander, against highway robbers, and so forth. There is even one against hemorrhoids.[19] In general, they take the form of a short story. For example, the Buddha's disciple Ananda has been traveling and is terrified of highway robbers. The Buddha tells him a story about how highway robbers were once stopped by saying a string of syllables, a dhāraṇī.[20] What I believe is going on here is, above all, the Word of the Buddha has power and truth. The dhāraṇī recalls the original incident in the life of the Buddha, together with the Buddha's promise that the repetition of the words will have the required effect. As far as the believer is concerned, the effectiveness is based on something we might almost call a contract which, once made, remains binding for all time. Well, at least it would remain binding for those who believe in the power and truth of the Buddha's Word.
Other dhāraṇīs are meant to be stamped in clay or inscribed on various materials, and then inserted into holy objects, into stūpa-monuments which may serve as tombs for holy persons, and into such things as images — even paintings and books. Very probably, the use of dhāraṇīs for such purpose began as a substitute for an older practice of inserting holy relics into the same objects, but particularly in Tibet both dhāraṇīs and relics are likely to be used together. The Tibetan term used for sacred deposits of all kinds, means "dhāraṇī-insertion."[21]
Besides the 'dhāraṇī doors' already mentioned, there are, interestingly enough, 'door dhāraṇīs' made on paper or cloth and placed above doorways, bearing a scriptural text which may say, "Just walking under this once can purify a thousand aeons of sins." I suggest that all these usages of dhāraṇī-sūtras employ a kind of 'covenantal' Buddhology, forming part of a more general devotion toward the Buddha and His Word. The placement of door dhāraṇīs might be found reminiscent of the mezuzah, which contains a sacral deposit of the very words of scripture that justify the practice, using scripture as 'empowerment' for practice just as the door dhāraṇīs do. Similarly, in the Jerusalem temple, the ark which in the view of many scholars served as the footstool for the invisible throne of the divine presence in the Holy of Holies is often believed to have contained the original covenant.[22] Tibetans in particular would more or less immediately recognize both this and the mezuzah as instances of dhāraṇī-insertion. However, in the Buddhist case, there is not just a single covenant, like the one made at Mt. Sinai, but as many different covenants as there are dhāraṇī-sūtras.
By the sixth century or so, a discussion emerged about the effectiveness of dhāraṇīs. It may not be clear who brought up the argument (although they were surely Buddhists), but we do have the Buddhist philosopher Bhavya's response. This passage has been studied and translated twice,[23] but I have also located several later passages by various Indian authors that are rather similar, dating to the ninth through twelfth centuries. Most of these passages, but not the one by Bhavya, make use of a four-fold subclassification of dhāraṇīs which probably has its immediate origins in a fifth-century work by Asaṅga.[24] The four types of dhāraṇī are: [1] dhāraṇī for recollecting dharma[s], [2] for recollecting meanings, [3] for use as mantras, and [4] for withstanding the experience of the ultimate Buddhist teachings like nonproduction and voidness. The passage by Bhavya starts with the argument, which basically says, 'What place do these unintelligible words in barbaric language, or in the Vedas of the "other" religion, hold within the Great Vehicle? They do not lead to the cessation of sin, or to the ending of even the least fault.' Bhavya responds by quoting several scriptural passages, and I will not go into it further, although it is interesting that at this early date he was quite aware of the woman Bodhisattva Tārā and Her mantra. It is evident that the use of mantras in religious practice was already in place in Buddhism by the 6th century.
In the later similar passages, first the ninth century passage by Daṃṣṭrasena, it is the Arapacana syllabary that is explicitly singled out as being phonemes that provide access to the full knowledge of dharmas. The same connection appears in passages by Ratnākaraśānti in the tenth and Abhayākaragupta in the early twelfth centuries. It also seems to be implicit in a passage by Jaggadalavihāra, within a work dated to 998 ce, although he only mentions the first letter 'A'. He says, "The letter 'A' is access point of all dharmas on account of nonproduction."[25]
The passage on the four kinds of dhāraṇīs by Gro-lung-pa is certainly not the earliest, and differs somewhat from the others, still it is a little more intelligibly expressed and therefore translatable:
1. One who has the Dhāraṇī of Words obtains the strength of insight and the memory which can hold for a limitless time immeasurable letters/phonemes which are composed and written, just by hearing — without practicing and without reciting — any particular teaching they have learned.
2. This one [the Dhāraṇī of Meaning] is like the first, but with the following difference: They can hold without limit the meanings of those teachings — without practicing and without mental cultivation — for an immeasurable period of time.
3. [Dhāraṇī of Secret Mantra] means to obtain the power over the samādhi which achieves the blessing power to pacify such things as epidemic diseases.
4. The fourth [Dhāraṇī for Obtaining Forbearance] has as its cause (or, its basis) that the one who has insight dwells in solitude and doesn't say a word; encounters noone, eats appropriate food thinking little about it, and sleeps briefly during the night. That is what the Teacher [the Buddha] means by Mantra of Obtaining Forbearance.[26]
Before moving into the third and final phase under consideration here, the yogic phase, which I consider to be quite distinct even if in some ways interrelated (or at the very least conscious of precedent), I would like not only to summarize, but to add some more further elements and speculate about a more general picture. Even though the early Perfection of Insight sūtras speak so much about 'reading' and 'writing' that we have to think that they were written down from the very beginning, they still frequently mention the Dharmabhāṇaka,[27] the reciter of the scriptures. The reciter was considered a special class within the Buddhist community, and we know from early inscriptions that women could and did serve in this role.[28] It appears that a good memory was the primary job qualification. Anyone who has read the book or viewed the cinematic version of Ray Bradbury's Fahrenheit 451, could imagine that his portrayal of the post-print culture method of preserving literature was inspired somehow by the role of the reciter in early Buddhism before scriptures were commonly committed to writing. I'll state my general speculative theory as simply as possible, in the meantime introducing a little more evidence that may help to support it.
My theory is that there was a code for aiding the memory of the scripture reciters, that the Arapacana syllabary is an example of it, although the passage it was meant to preserve has not been identified; also, that the earliest Great Vehicle scriptures not only preserve a memory of such codes, they most probably had their own system of memory using key-syllables. The memory system, whatever its exact details might have been, itself helps to explain the well-known formulaic and repetitive nature not only of the scriptures in the Pāli canon,[29] but of the Perfection of Insight sūtras as well, particularly in the more lengthy versions. There is a small class of Tibetan literature, that has yet to be touched by scholarship of any kind, which explains to us how the larger Perfection of Insight sūtras can be generated through a process they call 'gres-rkang.[30] The word 'gres is employed in the Tibetan translations of works by Daṃṣṭrasena and Jaggadalavihāra already mentioned, although I have not yet determined what the corresponding Sanskrit word would have been. 'Gres-rkang refers to a repeated piece of text, into which a long list of items are to be inserted. The items to be inserted are the two types of dharmas that were mentioned earlier, the saṃsāric dharmas and the nirvāṇic dharmas.
To give a simple example, instead of saying "All dharmas lack self-nature," we would say, "The sense of seeing lacks self-nature. The sense of hearing lacks self-nature. The sense of smell lacks self-nature." and so on and so on, slotting in perhaps over a hundred terms into the same repeated statement. The early pre-Great Vehicle school known as the Dharmaguptaka had its own version of the monastic code which has been preserved in Chinese. In this text we find exactly this type of repeated sentence formula, "The sense of seeing is impermanent, the sense of hearing is impermanent" etc. This occurs as part of a discussion of chanted recitations in which the chant leader or the monks in general may start the first syllable of the phrase, after which the community, or the laypeople in particular, will join in. Furthermore, it gives the Arapacana itself as an example of a group chanting event, in which monks and laypersons would chant together, whether in unison or in response.[31]
It may be surprising to learn that the earliest Tibetan 'gres-rkang texts, composed in the 11th century, belong to the Tibetan Bon religion. Since a discussion would lead us too far astray, I will just say that Bon is believed to be the original pre-Buddhist religion of Tibet, and it is often accused (by others, not by me) of stealing its scriptures from the Buddhists by changing a word here and there. I recently edited an 800 page catalogue of the Bon scripture collections.[32] Several years ago in Oslo, while working together with a committee on the catalogue, I noticed a very interesting thing about a ten-volume scripture that everyone agrees in some way or another corresponds to the Buddhist Perfection of Insight Sūtras. It exists not only in the ten-volume version, but in a one-volume version as well. Within the latter is a chapter on a dhāraṇī, in which each syllable of the dhāraṇī corresponds to a repeated textual passage — allowing us to expand the one-volume into the ten-volume version — but at the same time corresponding to one of the thirty-two major and eighty minor marks of the Buddha,[33] who in this case is Lord Shenrab, the founder of Bon. (Lord Shenrab is often called by the Tibetan word for Buddha.) In a rather startling way, this brings together devotion to the physical form of the Buddha with devotion to His Word. It reminds us of the Buddha's own advice, "to see me in the corpus of my teachings."[34] It reminds us, too, of the episode in the Lalitavistara Sūtra, among other places, in which the sage Asita came to see the infant Siddhārtha and examined His bodily signs, finding the thirty-two and eighty major and minor marks, which indicated that He would be either a universal monarch or an Enlightened One. Asita exclaimed, "Truly a great wonder has appeared in the world."[35] Just as, or to the degree that, the future Buddhahood is predicted through the marks, later on the marks would allow us to recognize the wonder that was or is the Buddha or His image.
Seeing this principle of text/image identification at work in the Bon scripture set off an alarm in my head. First of all, the chapter in the Bon text on the major and minor marks is located in about the same position as the chapter in the much better known Buddhist texts. In the 25,000 Perfection of Insight Sūtra, this chapter covers three subjects: major marks, minor marks, and letters. It is quite mysterious what the letters have to do with the marks, and why they should make an appearance immediately after them.[36] And the passage on letters is itself mysterious, recommending that one should be skilled in the forty-two letters, and meditate on the forty-two letters as contained in one letter, and on one letter as contained in forty-two letters.[37] Since in the Sanskrit alphabet generally forty-nine or fifty letters are counted, the forty-two letters simply must mean the Arapacana syllabary. Perhaps the Arapacana is, after all, the secret reciters' memory code for the Perfection of Insight Sūtras, but if so, it has not proven possible to know the specific way in which it would have been applied. It seems more likely that these sūtras are playing with a mnemonic system that was already well known — most likely one in use by the Dharmaguptakas for scriptures they were in the practice of reciting, something like the Arthavarga or the Udānavarga[38] — while employing a different system of its own. That there was such a memory code or codes seems certain.
Part Three: Yogic.
Now we enter a seemingly alien world, the yogic world, in which the alphabet occupies a different position. To make a difficult history simple, elements that make up the Vajra Vehicle (Vajrayāna) emerged in around the 4th century and slowly coalesced in various ways until emerging in a fairly full form, with distinctive texts called 'tantras' in about the 8th century (although some would push this back to somewhere in the three preceding centuries). As such the Vajra Vehicle can only be defined as a complex of ideas and practices. It has no single defining characteristic. It employs powerful words called mantras, powerful gestures called mudrās, and special meditation practices that involve intricate visualization processes, and which always employ phonemes and visual syllables for a number of purposes.[39] For present purposes we should try to limit ourselves to underlining the new emphasis these yogic Buddhists placed on the human body, and on the bodily practices of breathing and posture which we normally associate with yoga, and the possibility of attaining Buddhist enlightenment in, as they say, 'one body, one lifetime.' In the Great Vehicle, accomplishing Buddhahood generally is said to take three incalculably long aeons (kalpa). The Vajra Vehicle sees itself as just a more effective way of bringing about the Great Vehicle's aim of Complete Enlightenment. It was and remains, and this is a point on which Tibetans certainly insist, a part of the Great Vehicle.
To demonstrate briefly the connection of yogic physiology with sound, phoneme, letter and alphabet, I would like to draw attention to a particular set of five couplets, of the type of song called dohā ('couplet'). The story is told how the author, Vīṇāpāda, born in a royal family, soon evinced a total disinterest in affairs of state. This disturbed his father the king, who was naturally yearning for an heir and successor. Vīṇāpāda was, to use an ugly modern functionalist term, 'dysfunctional.'[40] All he wanted to do was play music. Then one day he happened to meet a spiritual teacher who assigned him the task of avoiding conceptualizations about the music, realizing the sameness of sounds he made with his vīṇā and his experience of those same sounds.
This might, and perhaps with slight reason, seem somehow similar to the Buddha's Parable of the Lute (vīṇā) contained in the Samyutta Nikāya. The Buddha said, suppose there was a king or a minister who had never heard the lute's sound. One day he hears it for the first time and finds it quite entrancing. He orders his aides to locate the source of the sound, and they return with a lute. The king (or minister) insists that it is not the wooden instrument he wants, but only the sound. His helpers try to explain to him that there are a great number of parts (which they name), which must all go together when the player makes music with it. Then the king, in frantic search for the sound, splits the lute into splinters, burns the splinters in a fire, and releases the ashes to the wind or throws them in a swift river. Of course, he does not find the sound at any of these levels of destruction. The irony of the parable is surely intended. The deconstructive foolishness of the king is like the wisdom of the Buddhist meditator when engaged in analytical meditation, searching for the 'I' and not finding it in one place after another.[41] The two different stories of Vīṇāpāda and the foolish king are at the very least similar in connecting the music of the plucked string instrument with some kind of meditation practice.
I do not think we have to take the Vīṇāpāda story too seriously as history, since as often happens in India, it may well have been a reading of the song that inspired the biographical account. It, also, may be a parable, like that of the foolish king, only one parading as biography. But I would especially like to draw attention to the 'unstruck sound' as a well-known phenomenon in yogic meditation practice. It is a kind of roaring — I would like to think of it a sort of phonemic soup — somehow prior to phonemes and inherent in the body. It is heard only after withdrawing the senses from their sense objects. I think it's entirely possible to hear it without doing yoga, through sensory deprivation or simply by sitting quietly in a snow storm when all other sounds are stopped or absorbed by the snow. It is a sonic experience of the internal body, just as the channels described in yoga literature represent an internal meditative experience (whether sensed through some kind of internal sense of touch or through vision) of the body's energetic currents, which may then be influenced or altered by various physical, verbal and mental exercises. At least I think that sums up part of the main message that ought to be heard in Vīṇāpāda's song, which is both densely phrased, so much so that it demands interpretation, and richly suggestive, even without plunging into the thicket of footnotes that nearly every word would seem to require from philological, musicological, Buddhological and yogic perspectives.
Vīṇāpāda's Dohā Song
The solar gourd is joined to the lunar strings,
while the unstruck sound is the [vīṇā's] neck, the avadhūti* the bridge. \1/
*The central vein/channel in yogic physiology is called the avadhūti ('the shaker'). The solar and lunar in the first line would then be the two side channels: the rasanā ('the taster,' 'the rope,' 'the bridle' or 'the tongue') on the right, and the lasanā ('the tongue' or 'the lolling of the tongue') on the left. The terms rasanā and lalanā used by Buddhists correspond to the better known piṅgalā and iḍā in Hindu tantras. See the detailed discussion in Bagchi 1975. In the most elaborate accounts, there may be as many as 72,000 channels in the subtle body, although the three just mentioned are the main ones.
Oh friend, the vīṇā of Heruka* sounds.
The sounds of the strings play themselves [wail] in sympathy [in play]. \2/
*The Tibetan for Heruka (Khrag-'thung) is interpreted most literally as 'Blood-drinker,' a wrathful manifestation of Buddha, including such 'divine forms of high aspiration' as Cakrasamvara and Hevajra.
The continuous sequence [of notes] equals the vowels and consonants.
The 'best of elephants'* first calculates the precise intervals [that create] an even tone [equal flavor]. \3/
*The superior musician, perhaps, or the yogin who has overcome duality as suggested in Munidatta's commentary. Dasgupta (1976, 98) translates this verse very differently: "On hearing the tune of the Āli Kāli, the mighty elephant has entered Samarasa ['equal flavor']."
and then, when [the musician/yogin] presses the thirty-two strings* down on the wood and frets,**
that's when their sounds pervade the whole [instrument]. \4/
*Munidatta explicitly identifies the strings as the thirty-two principle channels in the yogic body, and the divine forms integrated in the body-maṇḍala (compare the Hevajra Tantra; Snellgrove 1980, I 49, 73-74). **The early vīṇā may have been unfretted, and the word translated as 'fret' literally means 'small piece of wood' (it could conceivably refer to the tuning pegs, or even the plectrum, which could be made of wood).
The king* does his dance, the goddess[es?] sings her song.
This Buddha dance is particularly difficult to do. \5/
*There are two possible readings of the Apabhramśa original, one meaning 'king' (adopted in the Tibetan translation) and the other meaning 'the one who has the Vajra.' In either case Munidatta believes it refers to the author Vīṇāpāda, who is celebrating his enlightenment through dance. It is true that the last lines of other dohā songs usually have an explicit reference to the author.
There is one major scriptural text (even if not included in the Tibetan canon) called the Vowels & Consonants Tantra,[42] which is the one with which we will concern ourselves here, and another smaller scripture especially interesting for alphabet usage.[43] Both were preserved in two distinct esoteric lineages from the South Indian teacher Phadampa, who died in Tibet in 1117 ce. They, like three other brief texts composed by obscure authors preserved in translation in the Tibetan Tanjur,[44] take the Sanskrit alphabet as the primary locus in which complex microcosmic-macrocosmic correspondences take place. It ought to be emphasized that, while it is true that the alphabet plays a role in nearly every Buddhist tantra, while spiritual or meditational and magical uses of letters inevitably show up in every single one of them, the particular texts just mentioned grant it an exceptional centrality. I believe that these texts belong to a general hāṭha-yoga atmosphere, such as that which was associated with the Nāth Yogis of Hinduism as well as the Great Siddhas of the Buddhists' Vajra Vehicle. It has long been known that the lists of their human teachers overlap.[45]
The Vowels & Consonants Tantra starts out like most Buddhist scriptures with long scene-setting chapters describing the Teacher and the audience, but it very soon betrays its phonemic approach which is sustained through its twenty-two chapters.
The point of departure is the sacred Volume. The Teacher, the Buddha, says, "Whoever sees the Volume, their faults are purified, their qualities completed."
The audience asks, "What is this Volume?"
Part of His answer: "It is the Volume of mind inscribed with the drawn [letters] of memory," and later on the Teacher declares, "All dharmas are vowels and consonants," and then, "Their substance is transcendent insight (its dawning door being the letter 'A'), their nature is unimpeded (the dawning door being the 'O»'), and their identifying mark is nonduality (the dawning door being the dhāraṇī)."
Then the Teacher announces the five aspects of the phoneme and/or letter:
1. "That of shape that is drawn." (Both visible letters and the visible shapes in the world.)
2. "That of material that is amazing." (Both the writing media and the elements of the world).
3. "That of meaning that is understood/realized." (Ordinary communication as well as the teachings on the Path to Liberation.)
4. "That of words that are illustrative/indicative." (Mantras, primarily, are at issue here.)
5. "That of figures of speech (or similes) that are appropriate."
After saying this, the Teacher smiled and did not say a word.
Chapter Six opens with the Teacher simply pronouncing, fully amplified "with a lion's roar," the fifty letters of the alphabet. The Buddhas in the ten directions of space respond, each echoing back a different string of letters. Just to give an example, the Buddhas in the eastern direction pronounce the letters ka, ca, ṭa, ta, pa, ya, śa, i, ī, and ṛ. Later on, the ten blue letters of the air-element are correlated with, among bodily organs, the intestines and lungs; in the external world, the birds and Asuras; deities of the Vajra Family; the ritual goal of accomplishing peaceful results. Still later, the air phonemes that came from the east are qualified as being fine and stopped up...[46]
In general, to simplify a large and complex picture, the points of articulation are conceived as being spatially distributed in the mouth and throat cavities in a kind of maṇḍala pattern which, in fact, if we think about it, might well be the case. Just like the positions of the eyes in various yogic gazes, which have a role secondary only to the phonemes in this text, the pronunciation of phonemes by the yogin or yoginī has direct effect on the flows and configurations of winds in the yogic body. (To put it in a poetic and perhaps justifiably convoluted way, the flow of the internal motions — thinking, feeling, emotion, motivation and the like — outward into facial or verbal expression can be made to flow back in the opposite direction and reconfigure the yogin's body in the form of psychosomatic responses.[47] Conscious control of what would under ordinary non-yogic circumstances be externalizing expression, including elements of sound such as tone, musicality, repetitions of phonemes, etc., can effect transformations, stoppages or proliferations of those internal motions in the yogin. [Audience response] theory and theater are both enacted on the single stage of the body, and the actor has fully assumed the role, is completely 'in character.') Later, in Chapter 16, each phoneme is correlated with a specific yogic gaze. After all, to point out something so simple and obvious that few people have ever consciously considered it, the eyes and vocal organs are the two semi-autonomous areas of the head that move and express things through their movement... (expression through hand movement is not prominent in this particular text, although nearly universal to Buddhist tantra, tantric text and ritual alike, in the form of gestures called mudrās).[48] Semi-explicit in the text is a theory of why the pronunciation of mantras by the yogin ought to be effective both in the body and in the world outside.
We should end with a few historically structured reflections, since the very word 'conclusion' has a frightening ring. My own thoughts keep going back and forth on this and that detail, simply refusing to settle down on any pat conclusion. The early Great Vehicle scriptures had a basically two-pronged approach to the letters of the alphabet. One of these makes use of that most rudimentary pedagogical device, the Abecedarium, still used for teaching children the alphabet in many cultures.[49] The beginning of learning, and particularly of the education of an, in some sense, all-knowing Buddha, is after all a very weighty matter that was not, and so should not be, taken lightly. The other connects the letters with the scriptures using a conscious mnemonic technique, which furthermore surely was already in early times conceived of as supplying mysterious meditative access points to Buddhist truths. In these approaches we can see strong connections with the sacred biography of the Buddha and with the preservation of Buddha's Word for the purposes of oral transmission, ritual recitation, and, finally, the sacred Volume which formed a cult object from the very beginnings of the Great Vehicle.[50] Already at this stage we may note connections made between the letters or syllables and the elements of the Buddhist universe called dharmas, and at the same time a less explicit connection of letters with the physical form of the Buddha, specifically the sculpted image of the Buddha as focus of cult that was emerging during the same period in both Gandhāra and Mathurā.[51] This basically devotional use of letters, I believe, led very directly and swiftly into the more contractual form of the dhāraṇī-sūtras where strings of syllables perform their work mainly because of a promise of their effectiveness made by the Buddha in these same scriptures.
The step from the covenantal to the yogic may seem a simple one to take, since practically every Buddhologist believes in a rough and general way that the dhāraṇī was necessary precursor to the Buddhist use of mantras.[52] However, I believe the crude historical picture might be finessed and developed in interesting new ways in the future. I do not want to imply that Buddhists were living in a vacuum, although for simplicity's sake we have at times been speaking as if they were. Surely they received from, as well as contributed to, a broader yogic movement taking place in India in the last centuries of the first millennium and the first centuries of the second. The letter-based yogic speculations of the Vowels & Consonants Tantra are clearly similar in kind with those found in the Śaivite Hindu tantra called the Mālinīvijayottara, for example.[53] Still, chronological uncertainties prevent us from simple conclusions about priority. I would say that in the yogic phase of our historical picture, we may notice some aspects of earlier alphabet usage newly encased in a yogic context. To give examples, in the Vowels & Consonants Tantra when the Teacher recites the Sanskrit alphabet, it probably is with good reason that we are taken back for a moment to the story of the young Siddhārtha's visit to school. And when the Tantra speaks of the dawning of dhāraṇī access points, or states that "All dharmas are vowels and consonants," these seem to be conscious re-articulations of ideas from the Perfection of Insight sūtras.[54] What is freshly expressed is a quite sophisticated linkage of the letters with elements of the external world and of the letters with the internal, yogically experienced body. And all in the service of transforming our instruments of engagement with the world, not just the body, but also speech and mind.
Bibliography
Allon, Mark
1997. "The Oral Composition and Transmission of Early Buddhist Texts." In Indian Insights: Buddhism, Brahmanism and Bhakti, eds. Peter Connolly & Sue Hamilton, 39-61. London: Luzac Oriental.
Alper, Harvey P.
1991. Understanding Mantras. Delhi: Motilal Banarsidass.
Bagchi, Prabodh Chandra
1975. "Some Technical Terms of the Tantras: Candra-Sūrya." In Studies in the Tantras, Part 1, ed. P. C. Bagchi, 61-73. Calcutta: University of Calcutta.
Bayes, Gwendolyn
1983. The Voice of the Buddha: The Beauty of Compassion. Berkeley: Dharma Publishing.
Bentor, Yael
1995. "On the Indian Origin of the Tibetan Practice of Depositing Relics and Dhâranîs in Stûpas and Images." Journal of the American Oriental Society 115.2:248-261.
1996. Consecration of Images and Stūpas in Indo-Tibetan Tantric Buddhism. Leiden: E. J. Brill.
Bernhard, Franz
1965. Udānavarga [Sanskrittexte aus den Turfanfunden 10]. Göttingen: Vandenhoeck & Ruprecht.
Bhattacharyya, D.C.
1987. "Dramatic Content of Tantric Buddhist Art and Rituals." In Indological Studies: Essays in Memory of Shri S.P. Singhal, ed. Devendra Handa, 93-104. Delhi: Caxton Publications.
Bodhi, Bhikkhu
2000. The Connected Discourses of the Buddha: A New Translation of the Saṃyutta Nikāya. Boston: Wisdom Publications.
Boucher, Daniel
1991. "The Pratītyasamutpādagāthā and Its Role in the Medieval Cult of the Relics." Journal of the International Association of Buddhist Studies 14.1:1-27.
Braarvig, Jens
n.d. "Bhavya on Mantras: Apologetic Endeavours on Behalf of the Mahāyāna." Pre-published draft courtesy of the author; the published version, with identical title, is not presently available to me: 1997. Aspects of Buddhism [Studia Indologiczne no. 4], Oriental Institute (Warsaw), ed. Agata Bareja-Starzynska & Marek Mejor, 31-40.
1985. "Dhāraṇī and Pratibhāna: Memory and Eloquence of the Bodhisattvas." Journal of the International Association of Buddhist Studies 8.1:17-29.
1993. Akṣayamatinirdeśasūtra: Edition of Extant Manuscripts with an Index. Oslo: Solum Forlag.
Brough, John
1977. "The Arapacana Syllabary in the Old Lalita-vistara." Bulletin of the School of Oriental and African Studies 40:85-95.
Cohn, Jim
2002. "Mahamudra Disability: Ancient India-Tibetan Social Theory." Disability Studies Quarterly 22.2:175-81.
Conze, Edward
1972. Buddhist Wisdom Books: Containing The Diamond Sutra and The Heart Sutra. New York: Harper Torchbooks.
1973. Perfect Wisdom: The Short Prajñāpāramitā Texts. London: Luzac & Co.
1975. The Perfection of Wisdom in Eight Thousand Lines and Its Verse Summary. Bolinas, California: Four Seasons Foundation.
1984. The Large Sutra on Perfect Wisdom, with the Divisions of the Abhisamayālaṅkāra. Berkeley, California: University of California Press.
Coomaraswamy, Ananda K.
1930. "The Parts of a Vīṇā." Journal of the American Oriental Society 50:244-53.
1931a. "The Old Indian Vīṇā." Journal of the American Oriental Society 51.1:47-50.
1931b. "Upavīṇa." Journal of the American Oriental Society 51.3:284-5.
1937. "The Parts of a Vīṇā." Journal of the American Oriental Society 57.1:101-3.
Cousins, L. S.
1997. "Aspects of Esoteric Southern Buddhism." In Indian Insights: Buddhism, Brahmanism and Bhakti, eds. Peter Connolly & Sue Hamilton, 185-207. London: Luzac Oriental.
Cüppers, Christoph and Per K. Sørensen
1998. A Collection of Tibetan Proverbs and Sayings: Gems of Tibetan Wisdom and Wit [Tibetan and Indo-Tibetan Studies series no. 7]. Stuttgart: Franz Steiner Verlag.
Dasgupta, Shashibhusan
1976. Obscure Religious Cults. Calcutta: Firma KLM Private Ltd.
Dreyfus, Georges B.J.
2003. The Sound of Two Hands Clapping: The Education of a Tibetan Buddhist Monk. Berkeley: University of California Press.
Durt, Hubert
1994. "DA." In Hôbôgirin: Dictionair encyclopédique du Bouddhisme d'après les sources chinoises et japanaises 6:565-572. Tokyo: Maison Franco-Japanaise.
Emmerick, R. E.
1996. The Sūtra of Golden Light: Being a Translation of the Suvarṇaprabhāsottamasūtra. Oxford: The Pali Text Society.
Gethin, Rupert
1993. "The Mātikās: Memorization, Mindfulness, and the List." In In the Mirror of Memory: Reflections on Mindfulness and Remembrance in Indian and Tibetan Buddhism, ed. Janet Gyatso, 149-172. Delhi: Sri Satguru Publications.
Goonesekere, Lakshmi R.
1968. "Bhāṇaka." In Encyclopaedia of Buddhism, ed. G.P. Malalasekera, 2.4:688-690. Ceylon: Government Press.
Gyatso, Janet
1993. "Letter Magic: A Peircean Perspective on the Semiotics of Rdo Grub-chen's Dhāraṇī Memory." In In the Mirror of Memory: Reflections on Mindfulness and Remembrance in Indian and Tibetan Buddhism, ed. Janet Gyatso, 173-213. Delhi: Sri Satguru Publications.
Haran, Menahem
1985. Temples and Temple Service in Ancient Israel. Winona Lake, Indiana: Eisenbrauns.
Harrison, Paul
1993. "Commemoration and Identification in Buddhānusmṛti." In In the Mirror of Memory: Reflections on Mindfulness and Remembrance in Indian and Tibetan Buddhism, ed. Janet Gyatso, 215-238. Delhi: Sri Satguru Publications.
Hirakawa Akira
1990. A History of Indian Buddhism from Śākyamuni to Early Mahāyāna. Trans. Paul Groner. Honolulu: University of Hawaii Press.
Hopkins, Jeffrey
1985. "Reply to Alex Wayman's Review of The Yoga of Tibet." The Tibet Society Journal 5:73-95.
Jackson, David
1996. "The bstan rim ('Stages of the Doctrine') and Similar Graded Expositions of the Bodhisattva's Path." In Tibetan Literature: Studies in Genre, eds. José Cabezón & Roger Jackson, 229-243. Ithaca: Snow Lion.
de Jong, Jan W.
1954. "L'épisode d'Asita dans le Lalitavistara." In Asiatica: Festschrift Friedrich Weller, 312-325. Leipzig: Otto Harrassowitz.
Kapstein, Matthew
2001. Reason's Traces: Identity and Interpretation in Indian and Tibetan Buddhist Thought. Boston: Wisdom Publications.
Khan, Muhammad Ashraf Khan
1993. Gandhara Sculptures in the Swat Museum. Saidu Sharif, Afghanistan: Archaeological Museum (Saidu Sharif).
Khettry, Sarita
2001. "Origin of the Image of Mañjuśrī." Journal of the Asiatic Society [Calcutta] 43.4:22-42.
Kölver, Bernhard
1992. "Das Symbol evaṃ." Studien zur Indologie und Iranistik 16:101-107.
Konow, Sten
1933. "The Arapacana Alphabet and the Sakas." Acta Orientalia 12.1:13-24.
Kvaerne, Per
1986. An Anthology of Buddhist Tantric Songs. Bangkok: White Orchid Press.
Lamotte, Étienne
1988. History of Indian Buddhism from the Origins to the Śaka Era. Trans. Sara Webb-Boin. Louvain-la-Neuve: Institut Orientaliste.
Lévi, Sylvain
1905. "La liste des écritures dans le Lalitavistara, comparée aux textes paralléles en chinois." Bulletin de l'École Français d'Extrême Orient 4:289-290.
1915. "Sur la récitation primitive des textes bouddhiques." Journal Asiatique, 11th series 5:401-447.
Malalasekera, G.P.
1966. Encyclopaedia of Buddhism, Ceylon: Government Press.
Martin, Dan
1987. "Illusion Web: Locating the Guhyagarbha Tantra in Buddhist Intellectual History." In Silver on Lapis: Tibetan Literary Culture and History, ed. Christopher I. Beckwith, 175-220. Bloomington, Indiana: The Tibet Society.
2000. "Comparing Treasuries: Mental States and Other Mdzod-phug Lists and Passages with Parallels in Abhidharma Works by Vasubandhu and Asaṅga, or in Prajñāpāramitā Sūtras, a Progress Report." In New Horizons in Bon Studies [Senri Ethnological Reports, no. 15], eds. Samten G. Karmay & Yasuhiko Nagano, 21-88. Osaka: National Museum of Ethnology.
Martin, Dan, Tseyang Changngoba, Namgyal Nyima Dagkar, Per Kvaerne, Dondrup Lhagyal, Donatella Rossi, Tsering Thar
2003. A Catalogue of the Bon Kanjur [Bon Studies series no. 8], ed. Dan Martin; project coordinator, Per Kvaerne; series editor, Yasuhiko Nagano. Osaka: National Museum of Ethnology.
Meisezahl, R.O.
1968. "Über zwei Mdo-maṅ Redaktionen und ihre Editionen in Tibet und China." Zentralasiatische Studien 2:67-139.
Miller, Roy Andrew
1966. "Buddhist Hybrid Sanskrit Āli Kāli as Grammatical Terms in Tibet." Harvard Journal of Asiatic Studies 26:125-47.
Mitra, Rajendralala
1981. The Sanskrit Buddhist Literature of Nepal. New Delhi: Cosmo Publications. Reprint of 1882 edition.
Mukherjee, B.N.
1999. "Arapacana: A Mystic Buddhist Script." In Tantric Buddhism: Centennial Tribute to Dr. Benoytosh Bhattacharyya, eds. N.N. Bhattacharyya & Amartya Ghosh, 303-317. New Delhi: Manohar.
Muller-Ortega, Paul E.
1992. "Tantric Meditation: Vocalic Beginnings." In Ritual and Speculation in Early Tantrism: Studies in Honor of André Padoux, ed. Teun Goudriaan, 227-245. Albany: State University of New York Press.
Naveh, Joseph
1997. Early History of the Alphabet: An Introduction to West Semitic Epigraphy and Paleography. Jerusalem: Magnes Press.
Onians, Isabelle
2002. "Tantric Buddhism and Aesthetics: Art and Morality." In Impressions of Bhutan and Tibetan Art: Tibetan Studies III, eds. John Ardussi & Henk Blezer, 113-123. Leiden: Brill.
Padoux, André
1990. Vāc: The Concept of the Word in Selected Hindu Tantras, trans. Jacques Gontier. Albany: State University of New York Press.
Phadampa [Pha-dam-pa Sangs-rgyas]
1979. A-li Ka-li Gsang-ba Bsam-gyis-mi-khyab-pa Chu-klung Chen-po'i Rgyud ['The Tantra of the Inconceivable Secret Great Flood of Vowels and Consonants']. In The Tradition of Pha Dampa Sangyas: A Treasured Collection of his Teachings Transmitted by T[h]ug[s]-sras Kun-dga', ed. B. Nimri Aziz, 1:6-114. Thimphu, Bhutan: Kunsang Tobgey.
Repsher, Brian
1998. The Rite of Church Dedication in the Early Medieval Era. Lewiston: The Edwin Mellen Press.
Rijal, Babu Krishna
1979. Archaeological Remains of Kapilavastu, Lumbini and Devadaha. Kathmandu, Nepal: Educational Enterprises.
Salomon, Richard
1990. "New Evidence for a Gāndhārī Origin of the Arapacana Syllabary." Journal of the American Oriental Society 110.2:255-273.
1993. "An Additional Note on Arapacana." Journal of the American Oriental Society 113.2:275-276.
Sarasvati, D. C.
1963. "The Mystic Significance of Letters: Their Application to the Art of Poetic Composition." The Adyar Library Bulletin 27:89-115.
Scharfe, Hartmut
2003. "Review of Pieter C. Verhagen, A History of Sanskrit Grammatical Literature in Tibet, Volume Two." Indo-Iranian Journal 46:156-160.
Schmithausen, Lambert
1997. Maitrī and Magic: Aspects of the Buddhist Attitude toward the Dangerous in Nature. Vienna: Verlag der Österreichischen Akademie der Wissenschaften.
Skilling, Peter
1992. "The Rakṣā Literature of the Śrāvakayāna." Journal of the Pali Text Society 16:109-82.
1996. "An Arapacana Syllabary in the Bhadrakalpika-sūtra." Journal of the American Oriental Society 116.3:522-523.
Sna-nam Btsun-pa Skal-bzang-chos-kyi-rgya-mtsho
1994. Sangs-rgyas Bcom-ldan-'das-kyi Rnam-par Thar-pa Rmad-du Byung-ba Mdzad-pa 'Khrul-ba Med-par Brjod-pa Bde-bar Gshegs-pa'i Spyod-pa Mchog-gi Gter ['Treasure of the Supreme Life of the Sugata: Unerring Account of the Deeds in His Wondrous Biography'], Xining, Amdo: Mtsho-sngon Mi-rigs Dpe-skrun-khang.
Snellgrove, David L.
1980. The Hevajra Tantra: A Critical Study. London: Oxford University Press.
Taube, Manfred
1968. "Zur Textgeschichte einiger Gzuṅs-bsdus-Ausgaben." Zentralasiatische Studien 2:55-66.
Templeman, David
1989. Tāranātha's Life of Kṛṣṇācārya/Kāṇha. Dharamsala, Himachal Pradesh: Library of Tibetan Works and Archives.
Tucci, Giuseppe
1930. "Animadversiones Indicae." Journal of the Asiatic Society of Bengal, new series 26:125-160.
1999: Tibetan Painted Scrolls. Bangkok: SDI Publications. Reprint of 1949 edition.
Upasak, C. S.
2001. History and Paleography of Kharostī Script. Sarnath: Central Institute of Higher Tibetan Studies.
Verhagen, Pieter Cornelis
2001. A History of Sanskrit Grammatical Literature in Tibet, Volume Two: Assimilation into Indigenous Scholarship. Leiden: Brill.
2002. "Studies in Indo-Tibetan Buddhist Hermeneutics (3): Grammatical Models in Buddhist Formulas." In Religion & Secular Culture in Tibet, ed. Henk Blezer, 143-161. Leiden: Brill.
Wayman, Alex
1957. "Contributions Regarding the Thirty-two Characteristics of the Great Person." Sino-Indian Studies 4.3-4:243-260.
1975. "The Significance of Mantras, from the Veda down to Buddhist Tantric Practice." The Adyar Library Bulletin 39:65-89. Reprinted, and perhaps more accessible, in 1984. Buddhist Insight: Essays by Alex Wayman, ed. George Elder, 413-430. Delhi: Motilal Banarsidass.
* Although the research lasted many years, this paper was put in writing while a member of a research group devoted to Indian poetry, chaired by Yigal Bronner, at the Institute for Advanced Studies, the Hebrew University, Jerusalem. In keeping with the academic — not specifically Indological, Tibetological or Buddhological — setting for which the original paper was intended, Sanskrit and Tibetan terms are kept to a minimum, and often bibliographical references are supplied with persons who do not know those particular languages in mind. An attempted synthesis of previous academic scholarship (I hope that I have not badly misrepresented anyone's views), there is a correspondingly lessened emphasis on my own research into the texts in their original languages. For Tibetan-translated canonical texts, in order to avoid bibliographical complications, I generally make reference to numbers in the Tohoku (Toh.) catalogue of the Derge Canon (Hakuju Ui, et al., A Complete Catalogue of the Tibetan Buddhist Canons, Tohoku University, Tokyo, 1934). Derge Canon (Kanjur and Tanjur) texts are, in a number of cases, available to me in searchable digital format (thanks to the Asian Classics Input Project), although the readings were checked against the 'original' Derge canon (albeit in the form of microfiche supplied by the Institute for the Advanced Study of World Religions, Stony Brook). The entire Derge canon has recently been made available, too, in scanned format (in the form of compact disks from the Tibetan Buddhism Research Center, New York City).
[1] This proverb (or in Tibetan, gtam-dpe, 'pattern [for] speech' or perhaps even 'oral simile') is, in one form or another, known to every Tibetan speaker. Several variants of it are recorded (with no translation provided) in Cüppers & Sørensen (1998, 226, nos. 10218-10222).
[2] There is a considerable literature on the identification of ancient Kapilavastu, much of it published in India and Nepal. For the Nepalese side, identifying it with the extensive (and inadequately excavated) ruins of Tilaurakot, see for example, Rijal 1979.
[3] A late fifteenth century Tibetan-authored biography of the Buddha (Sna-nam Btsun-pa 1994, 43) gives the schoolmaster Viśvāmitra the additional name Srin-bu-go-cha (in Sanskrit, perhaps Kṛmivarman, or 'Bug Armor'). This probably results from combination with the Abhiniṣkramaṇa-sūtra (Toh. no. 301, fols. 17-18), where the Buddha's school teacher is indeed given this other name. A charming green phylite relief, kept in the Swat Museum in Pakistan, depicts the young Buddha and a companion on their way to school riding on two rams, accompanied by two adult guardians, one of them holding an umbrella above the childrens' heads (Khan 1993, 71).
[4] Lalitavistara Sūtra (Toh. no. 95, fol. 108). English readers will have to content themselves with the translation (based on the French of E.P. Foucaux, which is not available to me, but making reference to the Tibetan text) found in Bayes (1983, I 187-195).
[5] See Lévi (1905) for a notice of this list of scripts.
[6] On Kharoṣṭhī script, see Upasak 2001. On the eastern Aramaic scripts, see Naveh (1997, 132-153). Naveh does not seem to be aware of the existence of Kharoṣṭhī as such, although he does briefly mention Aramaic script use in India (on p. 127).
[7] For an excellent list of references on the Arapacana, including some that will not be mentioned here, see Gyatso (1993, 198). Note the more recent study by Verhagen (2002, 143-149), who quite interestingly tends to the conclusion that the Arapacana was a 'real' alphabet of Gāndhārī.
[8] Mañjuśrī is a Bodhisattva, depicted with royal ornaments (and not monastic robes), with a sword lifted as if ready to strike in His right hand, and a book in His left (or balanced on a lotus held in His left hand). Khettry (2001) has traced images identified with Mañjurī holding the book (but without the sword) to the first centuries of the Common Era.
[9] See Dreyfus 2003, 85, with a general discussion of Tibetan monastic memorization practices on pp. 85-97.
[10] The best listing so far of the many sūtras that have the Arapacana in one form or another is the one located in Durt 1994.
[11] Conze 1984, 160; Wayman 1975, 78-79. I used the Derge Kanjur version of the Pañcaviṃśatisāhasrikā-prajñāpāramitā (Toh. no. 9, fol. 344).
[12] The word mātrikā (Pāli mātikā) is also used in Abhidharma texts to refer to lists of the main elements of Buddhist psychological (and other types of) analysis. These are lists of words, not of syllables. See especially Gethin 1993.
[13] Just like the Tibetan word that was used to translate it, yi-ge. See the discussion of this point in Hopkins (1985, 76-77).
[14] In my estimation, the most valuable modern discussion of dhāraṇīs is the one found in Gyatso 1993.
[15] My translation based on Braarvig's careful edition (1993, I 148); compare Braarvig (1985, 18; 1993, II 556-557).
[16] Translation by Mitra (1981, 244), which might be compared to the rather different translation by Emmerick (1996, 44-45).
[17] Braarvig 1985.
[18] Note that the Perfection of Insight sūtras come in many sizes, ranging from the 100,000 in twelve volumes (Toh. no. 8) down to the one on the letter 'A' in a few lines (English translation in Conze [1973, 201]).
[19] Arśapraśamani Sūtra (Toh. no. 621). A Chinese version also exists, its contents described by Ratna Handurukande in Malalasekera (1966: 96). We should avoid falling into the mistaken notion that these types of dhāraṇīs were a Great Vehicle invention. Although called rakṣā ('protection'), rather than dhāraṇī, Elder School (Theravāda) texts that are very much like dhāraṇīs do exist, and it is remarkable that some of those texts have been preserved in the dhāraṇī sections of the Kanjur in Tibetan translations, as Skilling (1992) has demonstrated. Schmithausen (1997) studies several examples including some against snakebite, together with a good discussion of the protective as well as the 'contractual' (or pact of friendship) nature of these types of texts. See also Cousins (1997) for remarkable instances of letter and mantra usages in Southern Buddhism. In Tibet, dhāraṇī collections called Gzungs-'dus and Mdo-mang (see Taube 1968 and Meisezahl 1968; the Bon religion also has its own versions of these collections), in one or two volumes, were quite popular, perhaps the most likely book to be found laying on a home altar. While many of the texts in these collections are found in the canon, some others are not. It could be said that the dhāraṇī-sūtras have been relatively neglected by scholars, but it is also true that collecting the bits and pieces published here and there would result in a very large bibliography, which will not be attempted here.
[20] See, for this example, Coravidhvaṃsana Dhāraṇī (Toh. no. 629). In the Tibetan form of the title, 'Phags-pa Mi-rgod Rnam-par 'Joms-pa zhes bya-ba'i Gzungs, we find the word mi-rgod. While it has the literal meaning 'wild man,' some people enthusiastically endorse the opinion that mi-rgod ought to be a name for the redoubtable Abominable Snowman. In this particular text, it is clear that mi-rgod are neither hairy humanoid beasts nor bestial humans, but something unfortunately much less arcane: highway robbers or bandits.
[21] In Tibetan, gzungs-gzhug, on which, see Bentor 1995.
[22] Haran (1985, 251 & 255) understands this in terms of ancient Middle Eastern practices connected with sacral deposits in general.
[23] Braarvig n.d. and Kapstein 2001, 233-255.
[24] This means the Bodhisattvabhūmi (Toh. no. 4037; see Braarvig 1985, 19 for the full citation). For an especially valuable discussion of these categories see Gyatso (1993, 175-176).
[25] Daṃṣṭrasena's Śatasāhasrikā-pañcaviṃśatisāhasrikā-aṣṭādaśasāhasrikā-prajñāpāramitā-bṛhaṭṭīkā (Toh. no. 3808, fols. 146-147); Ratnākaraśānti's Aṣṭasāhasrikāprajñāpāramitā-pañjikāsārottamā (Toh. no. 3803, fols. 39-40); two works of Abhayākaragupta, the Muṇimatālaṃkāra (Toh. no. 3903, fols. 230-232) and Aṣṭasāhasrikāprajñāpāramitā-vṛtti-marmakaumudī (Toh. no. 3805, fol. 65); and Jaggadalavihāra's Bhagavatyāmnāyānuśāriṇi-nāma-vyākhyā (Toh. no. 3811, fols. 302-303).
[26] The basis for this translation is an Asian Classics Input Project (www.asianclassics.org) digital text, no. SL0070-1, since the work by Gro-lung-pa has not yet been published in any other form (only a very few woodblock prints survive, in Mongolia, St. Petersburg and Patna; I have personally inspected a print in St. Petersburg). Composed in around 1100 ce, the title is: Bde-bar-gshegs-pa'i Bstan-pa Rin-po-che-la 'Jug-pa'i Lam-gyi Rim-pa Rnam-par Bshad-pa, and the passage is located at folios 285-286. For a brief outline and discussion of this work, see Jackson (1996, 230-231). Gro-lung-pa's explanation of the Dhāraṇī for Obtaining Forbearance is quite unique (all the other ninth- through twelfth-century passages we have mentioned make it first, not last in the list, and describe it as the ability to withstand Buddhist truths). 'Obtaining forbearance' in Gro-lung-pa's passage has a technical meaning associated with Great Vehicle ideas about stages in the Path to Enlightenment. It belongs to a higher stage of what is known as the 'Path of Application,' almost immediately preceding the direct vision of the Truth. In the Path of Application, various moderately strict disciplines (such as those mentioned here) are recommended.
[27] In Tibetan, Chos-kyi Smra-ba-po. I believe it is significant that the word bhāṇaka shares with the word for eloquence that we have already discussed the same Sanskrit root bhaṇ. For a general treatment on the bhāṇaka, see Goonesekere (1968).
[28] Hirakawa 1990, 30.
[29] See especially Allon (1997) and literature cited there.
[30] I have briefly discussed, and given bibliographical references to, several of these works in Martin (2000, 66).
[31] Lévi 1915, 439-440; Lamotte 1988, 497-498.
[32] See Martin et al. 2003. The catalogue of the Bdal-'bum volumes, which are the ones discussed here, may be found in the same publication, pages 253-265.
[33] On this subject, see especially Wayman 1957.
[34] Boucher (1991, 2 & 17 note 3) has noted a number of Pāli and Sanskrit versions of this statement. We might add, too, a similar statement in the Vajracchedikā Sūtra (Conze 1972, 63).
[35] See de Jong 1954. For the account in the Lalitavistara, see Bayes (1983, I 150-163).
[36] One of the latest among the Tibetan Dunhuang documents, probably dating not much earlier than the early 11th century, somehow correlates the vowels and consonants (here referred to as a-li ka-li) with the marks and signs (see Verhagen 2001, 30-36 for a long discussion). It may be interesting to consider the following instance in which Buddha's bodily marks are identified with letters. In consecration rituals intended specifically for scriptural books, we find a recent Tibetan manual suggesting that, after visualizing the physical book away, it is replaced by Buddha Amitābha in the form of a book. At the same time, the major and minor marks of the Buddha transform into vowels and consonants which are then imagined to dwell on each and every page of the scripture (Bentor 1996, 295). Among the preparatory rites that precede the consecration proper, we find letters being written on a mirror (which reflects the item that will be consecrated), then rinsed with water which falls on flowers to be used later on, imbuing them with the power of the letters; letters that have already been empowered by transferring holy words through a dhāraṇī-thread (Bentor 1996, 116-117). This ritual power-line is, by the way, used in Paritta ceremonies in Sri Lanka, as well as by Newar Vajrācāryas in Nepal. Different in the similarity of its consecratory function is the Abecedarium rite which often forms a part of Roman church dedications. In it, the Bishop writes the letters of the Greek and Latin alphabet on the floor of the church, using his crosier to draw the letters on small piles of ashes, creating the overall form of a [St. Andrew's] cross (Repsher 1998, 50-52, 57, 82-84). To underline the obvious, the letters of Greek and Latin are the ones that form the holy scriptures, the Septuagint and Vulgate (as I see it, the rite employs the elements of sacred scripture to make something else sacred), while they also signify the "beginnings of faith," just as the alphabet is the beginning of learning. In the Tibetan case, the primary source of the empowering is the repetition of the "Ye Dharmā," which is believed to epitomize the scriptures in a different way (Bentor 1996, 114).
[37] See Conze (1984, 587), for a complete translation of this passage.
[38] Other possibilities that ought to be investigated: There was a list of 42 or 44 mental states and associated factors in the Dharmaskandha, an Abhidharma text that has been dated to the time of Aśoka. Perhaps this or another Abhidharma list of these or other such dharmas are behind the Arapacana. A Sūtra in Forty-two Sections survives in Chinese translation. One problem is that texts such as the Udānavarga have been re-arranged, and there is no guarantee, either, that the forms of the texts as we have them would have been followed in the early recitation practices. The Sanskrit Dharmapāda (itself a form of the Udānavarga) has its first chapter entitled Anitya ('Impermanent'), anitya being the word for the first letter, the letter 'A', in the Sanskrit Abecedarium of the later Lalitavistara, as given above (see Bernhard 1965, 95).
[39] There is a great deal of Vajra Vehicle letter usage that will not be considered here in any detail. The first word of every Buddhist scripture, Evaṃ, literally meaning 'thus' or 'just so' is understood as combining the feminine 'E' (standing for the whole string of vowels), symbol of Insight (prajñā) with the masculine 'VA»' (standing for the whole string of consonants), symbol of Method (upāya). Kölver (1992) believes this symbolism is based on the form of the letters as found in early inscriptions from Mathurā and elsewhere, in which the 'E' is shaped like a downward pointing triangle, and the 'VA»' like an upward pointing triangle. I have dealt with some of this sort of letter symbolism, the evaṃ in particular, in Martin (1987, 197-198). We might point out, well-known as it may be, that the first words of Buddhist scriptures ('Just so was it heard by me at one time') are not spoken by the Buddha, but probably represent a ritualistic phrase used by the Dharmabhāṇaka before beginning each recitation (which doesn't contradict the idea shared by many that they are the words of Buddha's disciple Ānanda). Seed-syllables (bīja) are used in visualizations to generate divine forms, and it is often the case that these are based on first letters (for example, the lotus [padma] on which the divine form is seated is often generated from the syllable 'PA»', the divine form Tārā from the syllable 'TA»' etc.). In tantras of the Yoga class, such as the Mahāvairocana, there is a great deal of speculation on the letter 'A', which is of cardinal importance for the Shingon Buddhists of Japan, but also for the Bon and Rnying-ma-pa schools of Tibet. Another interesting use of the alphabet is in the 'Mantra Picking' or 'Mantra Extraction' (Mantrodhāra) chapter that one finds in many tantras (see for example Miller 1966, 138-140; the Vowels and Consonants Tantra discussed below also has such a chapter). Here the alphabet is numerically encoded as part of a method for both concealing and preserving the letters of the mantras (I believe the intent was to prevent the corruptions that do all too commonly occur in the scribal reproduction of mantras).
[40] This aspect of Viṇāpāda, as one of several of the eighty-four Great Siddhas who experienced one or another form of disability, has been drawn out in an article by a specialist in 'Disability Services' (Cohn 2002). For the Tibetan and Apabhraṃśa texts of the song and Munidatta's Sanskrit commentary on it, I have relied entirely on Kvaerne (1986, 145-150), with its very valuable philological discussions. This is a free and by no means a 'final' translation. It is based sometimes on the Tibetan and sometimes on the Apabhraṃśa version, and benefits from consultation with Herman Tieken (Leiden), for the Apabhraṃśa vocabulary, and Tom Hunter (Jakarta), for musicological aspects.
[41] For the Parable of the Lute, see Bodhi 2000, II 1253-4. For the parts of the lute, see Coomaraswamy 1930, 1931a, 1931b, 1937. Unfortunately, Coomaraswamy's discussions about the Sanskrit and Pāli vocabulary for parts of the vīṇā were not of much help for understanding those used in the Apabhramśa song which follows.
[42] Phadampa 1979, I 6-114. This work was undoubtedly rare in the past, and this, rather than any doubts about its authenticity (and I know of hardly any such doubt ever being expressed in Tibetan literature; one 15th-century scholar named Bo-dong-pa had brief doubts, but soon realized his error...), would sufficiently explain why it was never included in the canonical Tanjur collection. Another copy has, however, been preserved within the collected works of Bo-dong-pa Phyogs-las-rnam-rgyal (1375-1451), and selections from it were published in the late nineteenth century (in the collection known as the Gdams-ngag Mdzod). It was not entirely unavailable, just difficult to obtain. The words in the title for vowels and consonants, āli and kāli, are discussed in Miller 1966. They appear in the first verse of the main Tibetan grammatical treatise believed to have been written by Thon-mi Sambhoṭa in the first half of the seventh century. Since the same words are used by Vīṇāpāda and a few other Dohā songs, it would appear that they are simply Apabhraṃśa equivalents of Sanskrit ādi and kādi, which mean 'A-series' and 'KA-series' (so it is not the case that āli and kāli are "not known so far from Indian sources" — Scharfe 2003, 157). As yet I know of no Indian instances of āli and kāli that would beyond all doubt predate the tenth century or so (although they do occur in the Hevajra Tantra, which some date to the eighth century), making their use in a seventh-century Tibetan work rather puzzling (see Miller 1966, 138 & 146).
[43] The Samatāvastupradīpa (Toh. no. 2319). The title could be translated 'Lamp of the Expanse of Sameness.' It is one among nine texts known collectively as the 'Nine Lamps' which were transmitted to Tibetans by Phadampa in one of his earlier visits to Tibet. They were translated by an obscure Kashmiri named Jñānaguhya, who is believed to have been a pupil who accompanied Phadampa on the same visit.
[44] The following three brief texts are located together in the Derge Tanjur: Gandha's Ālikālimantrajñāna [or Ālikālimantrakrama?] (Toh. no. 2404), Dhamadhuma's Kālimārgabhāvanā (Toh. no. 2405), and Sāgara's Samvaracakrālikālimahāyogabhāvanā (Toh. no. 2406). It seems probable, given the names of the authors, that these texts are connected to Caryā yogis who belonged to the immediate circle or later followers of the Great Siddha Kāṇha, based in, but not limited to, northeastern India (Dhama and Dhuma were two immediate disciples of his; see Templeman 1989, 53).
[45] Tucci 1930, 138 ff.
[46] It was a paper once delivered by David Shulman (Hebrew University, Jerusalem) that first made me aware that some rather late South Indian theorists of poetics also had ideas about the meanings of particular phonemes, ideas that may presume prior discussions on the use of mantras (see, for example, Sarasvati 1963). Although once widespread in many cultures, theories about the meaningfulness of isolated phonemes have largely fallen into disrepute. Still, arguments occasionally surface in modern discussions about the meanings of language (for example, in English language, how many adjectives used to qualify snakes begin with, or otherwise contain, sibilants?). Their relevance in realms of poetics and religion could remain regardless of what the linguists and grammarians, with their different aims, might have to say (compare Padoux's comments in Alper [1991, 305]). One might with reason be reminded of the wondrous ideas of the Irish poet who went by the name of AE (i.e., George William Russell, in his 1918 book entitled Candle of Vision, not presently available to me), although it is quite conceivable he was inspired or influenced by Indian sources.
[47] In fact, the lunar vein, the rasanā, is explicitly associated with forward or outward flowing movements, while the solar lalanā is described as 'taking in' (or even 'eating'). See Bagchi (1975, 65). The classic work on Indian dramatic sciences, the Nāṭyaśāstra, chapter 8, verses 38-125, has a detailed analysis of dramatic gazes and eye expressions, enumerating thirty-six types. Bharata's interest, as a dramatist, is in eye expressions that convey the actor's emotional states to the audience, while in yogic contexts the same or similar gazes are used to control the yogin's own mental states and subtle physiological movements associated with them.
[48] To my knowledge, the only work which explicitly underlines the continuities between the classic work of Indian drama (also including dance, music and aesthetic theory) by Bharata with the expressive elements of gesture, posture and facial expression as found in the Buddhist Vajra Vehicle is Bhattacharyya (1987), although there are some very significant observations, too, in Onians (2002). In my opinion much more thinking ought to be done along these lines.
[49] Mukherjee (1999, 303) gives an example from contemporary Bengal, in which children are taught the letters by means of an Abecedarium that consists of complete sentences (not just words) beginning in turn with each of the letters in alphabetic order.
[50] On this point it would be worthwhile to recall the story of Sadāprarudita's quest for the Volume of the Perfecton of Insight Sūtra, which after great hardship he finds enshrined and sealed with seven seals (Conze 1975, 277-299). The text in which this story is told is the very same text that is found in the story, sealed with the seals.
[51] This ought to lead us into a consideration of the age old Buddhist practice of 'calling the Buddha to mind' or 'recollecting the Buddha' (Buddhānusmṛti), on which see particularly Harrison 1993. This practice may involve everything from contemplating the Buddha's abstract qualities to actual techniques leading to the visualization of the Buddha's physical form (including the marks and signs mentioned earlier), as well as complexes of various such practices.
[52] Tucci 1999, 224, for example.
[53] This tantra figures very largely in Padoux 1990, an academic work (originally published in French in 1964 as a doctoral dissertation) highly recommended for those interested in investigating or comparing Hindu tantra use of letters and phonemes, as is Muller-Ortega 1992.
[54] And there are a number of excellent reasons for locating Phadampa firmly within the orbit of the Perfection of Insight sūtras. The Vowels and Consonants Tantra is immediately preceded by the Heart Sūtra (translated in Conze 1972; perhaps one the most popular among the shorter Perfection of Insight sūtras, it is the very first text in the five [originally four] volume collection in which this Tantra was preserved). The Tantra is sealed with seven seals, just as is the Volume in the story of Sādaprarudita's quest. Phadampa, although apparently born in the coastal part of present-day Andhra Pradesh in southern India, did his early monastic studies at Vikramaśīla in Bihar, which had a curriculum emphasizing the Perfection of Insight sūtras and their commentaries by Haribhadra (late ninth century).