The Pun in the Root Tantra
Did the Seventeen Dzogchen Tantras Speak the Language of the Great Seal Before “Mahāmudrā” Was a School?
I. A Question Worth Exploring
Somewhere in the root tantra of the Seventeen Tantras of the Menngagde — the sGra thal 'gyur, the “Reverberation of Sound," one of the most-cited scriptures in the entire corpus — a short prophetic passage names the teacher who will hold the lineage after Jñānasūtra. It calls him, in Tibetan, phyag rgya rgya chen. Read as ordinary words, this says: “the great seal." Read as the text almost certainly intends it, it is something more playful and more revealing: a phonetic pun. Phyag rgya renders the Sanskrit mudrā; rgya chen renders “vast" or “great." Reversed into Sanskrit word order, the phrase becomes something very close to Vipulamudrā — which, said aloud, sits a hair's breadth from Vimalamitra, the name of the Indian master to whom this entire tantric cycle is traditionally attributed.
So: is this nothing more than a clever bit of scribal wordplay, using a phrase that simply happened to be lying around in the wider Buddhist vocabulary of the time? Or does the fact that this particular string of syllables was chosen — out of the enormous available space of Sanskrit-Tibetan homophony — to identify the figure at the very root of the Great Perfection's own lineage tell us something about how saturated that phrase already was, in the linguistic and conceptual world the redactors of these tantras were breathing?
That is the question this essay sits with. It does not answer it. What follows is an attempt to lay out, as honestly and as fully as research allows, what is actually findable in and around the Seventeen Tantras on this point — where the evidence is solid, where it is suggestive, and where it runs out entirely into things that would need someone to sit with the Tibetan of all seventeen root texts and their commentaries, line by line, to settle.
II. What “Great Seal” Meant Before It Had a School
Start with the term on its own side of the comparison, because it turns out to have exactly the kind of pre-institutional history the question requires.
Scholarship on the Kagyü tradition's own formation has noted something worth taking seriously: there is no evidence that Mahāmudrā existed, prior to the Indian tantric literature, as anything with its own lineage and its own transmission. What exists earlier is the word and the image — mahāmudrā, the great seal — circulating freely within the Yoginī-tantra literature and, especially, within the dohā, the vajra-song poetry of the Mahāsiddhas: Saraha, Tilopa, Nāropā, and the wider circle of realised adepts whose songs use mahāmudrā the way one might use “the nature of mind" or “co-emergent wisdom" (sahaja) — as a direct pointer to reality itself, not as the proper name of a school one belongs to. It is only later, as this material was systematised — especially through Gampopa in the twelfth century, drawing the sūtra-based approach of the Kadampa together with the siddha-song material transmitted through Marpa and Milarepa — that Mahāmudrā crystallises into something with its own textual corpus, its own graded stages, its own institutional identity as the teaching of a particular lineage.
This matters enormously for the question at hand, because it means the very possibility the question raises — a term doing genuine doctrinal work before it belongs to anyone — is not a speculative frame invented for this inquiry. It is already how serious scholarship describes Mahāmudrā's own early history on the Indian side. The open question is simply whether the Tibetan literature of the Great Perfection, redacted in almost exactly the same centuries this pre-institutional siddha vocabulary was flooding into Tibet, shows contact with that same undifferentiated pool — before Dzogchen, too, hardened into a self-conscious system with its own commentarial architecture, its own Nyingthig distillations, its own doxographical insistence on its distinctness from everything beneath it.
III. The Pun in the Root Tantra
Return, now, to the sGra thal 'gyur passage with this context in place.
The line comes from a lineage prophecy — a genre these tantras use repeatedly, casting the transmission of the teaching forward in time through a sequence of named and pseudonymous holders. In this particular prophecy, Vimalamitra's immediate place in the lineage is marked not by his ordinary name but by this punning epithet, phyag rgya rgya chen, “great seal, greatly vast" — a designation a recent doctoral dissertation on the Indian and Tibetan sources of the sNying thig identifies explicitly as wordplay on Vimalamitra via the reconstructed Sanskrit Vipulamudrā.
What can honestly be said about this: it is a real, findable, citable occurrence of the exact phrase phyag rgya (rgya) chen sitting inside the root text of the Seventeen Tantras — not a modern gloss, not a later commentarial import, but a string embedded in the tantra's own prophetic verse. What cannot honestly be said, on the strength of this one passage alone, is that the tantra is here teaching Mahāmudrā, or citing it as a known system, or drawing on it as a technical term the way the passage a few lines later draws on bodhisattva on the levels or emanation of Vajrapāṇi as recognisable Mahāyāna categories. On its face, this is naming, not doctrine — a pun used to seal a lineage figure's identity into the text through sound, the way tantric literature elsewhere plays with names through numerology, anagram, and homophony.
And yet the questioning move worth making here — without pushing it further than the evidence allows — is this: puns are not made from arbitrary material. A redactor reaching for a phrase that would evoke Vimalamitra's name through sound reached, apparently without strain, for phyag rgya chen po. That the phrase was ready to hand, natural enough to use as raw material for wordplay in a text about the deepest transmission of the Great Perfection's own root lineage, is itself a small but real piece of evidence about how available, how unmarked, how un-sectarian that phrase still was in this textual environment — available to be played with rather than needing to be carefully bracketed off as belonging to a different, rival system. Whether that availability reflects genuine doctrinal kinship or simply a common fund of Buddhist Sanskrit vocabulary that any redactor educated in the period would have had at their fingertips is exactly the kind of question a single passage cannot settle on its own.
IV. A Shared Vocabulary at the Source
The pun is not the only thread. Independent lines of philological research on this exact literature converge on a broader and, in some ways, more persuasive point: that the core technical vocabulary of the sNying thig itself — not just an isolated pun, but its central term for realisation — has a demonstrable path back into the same tenth- and eleventh-century Indian milieu that produced the Mahāsiddha corpus.
The term is rig pa, and its reflexive form rang rig — “awareness" and “self-awareness," the single most load-bearing word in the entire Dzogchen vocabulary. Careful comparative work on the Sanskrit and Apabhraṃśa sources has traced this usage not to the expected Sanskrit vidyā (the term rig pa ordinarily renders), but to saṃvit and svasaṃvedana — “consciousness," “self-reflexive knowing" — terms that appear, in almost exactly this sense, in Śaiva non-dual literature of the same period and, crucially, in the dohā literature of Saraha and Tilopa themselves, where svasaṃvedana names the innate, self-experiencing quality of co-emergent (sahaja) awakened mind — the very awareness Mahāmudrā songs are pointing at. On this reading, rig pa and rang rig begin appearing in Dzogchen tantras in this specific, technical sense around the eleventh century — which is to say, at essentially the same moment, and drawing on essentially the same Indian textual current, that the term mahāmudrā was itself being carried into Tibet through the same siddha-song literature.
This is a different, and in some ways stronger, kind of evidence than the single pun, because it does not rest on one clever wordplay but on a documented pattern of vocabulary transfer at the level of the tradition's most central term. If rig pa itself — the word Dzogchen uses to name what is finally to be recognised — carries a traceable debt to the same Mahāsiddha and Śaiva vocabulary pool that gave the wider Tantric world mahāmudrā, then the occasional appearance of mahāmudrā's own vocabulary inside these tantras looks less like an isolated curiosity and more like a symptom of a genuinely shared linguistic environment.
Other threads point the same direction without, on their own, closing the case. The early Mind Series material — the stratum of Dzogchen literature that precedes the Seventeen Tantras by a century or more — already describes the primordial bodhicitta as effortless and identifies it explicitly with bde ba chen po, “great bliss" (mahāsukha), the culminating co-emergent bliss of Yoginī-tantra and dohā literature alike, declaring that with respect to this great bliss, all conceptual extremes are non-dual. And Nubchen Sangyé Yeshé, one of the earliest indigenous Tibetan systematisers of Dzogchen doxography, writing in the ninth or tenth century, felt compelled to argue at length for Dzogchen's superiority over Chan precisely because — as he himself acknowledges — so much of their vocabulary overlapped. If a Tibetan scholar of that stature had to work this hard to distinguish Dzogchen's terminology from one neighbouring tradition on record, it is not unreasonable to ask, without yet answering, whether a comparable porousness held between Dzogchen's language and the Mahāsiddha material entering Tibet at almost exactly the moment the Seventeen Tantras were taking their present shape.
V. A Chronological Coincidence That May Not Be a Coincidence
This last point deserves to be drawn out on its own, because it is the piece of the puzzle most likely to be underappreciated, and it comes directly from recent, careful dating work on this exact textual corpus.
The scholarly reconstruction of the sNying thig's formation places its key redactors — lCe btsun Seng ge dbang phyug, lCe sgom nag po, and especially Zhang ston bKra shis rdo rje, who is credited with the decisive discovery and redaction of the Seventeen Tantras themselves — in Central Tibet across the eleventh and into the twelfth century, with Zhang ston's own work on the material datable to around 1118–1125. This is not a vague “sometime in the early medieval period" — it is a tightly reconstructed window, built from internal prophecy-chronologies cross-checked against independent biographical sources.
Now set this beside the other great transmission event of exactly the same Tibetan century: the “later propagation" (phyi dar) of Buddhism from India, the period in which Marpa travels to India and receives, among a great deal else, the dohā and mahāmudrā transmissions from Nāropā and Maitrīpa, which Milarepa and then Gampopa would carry forward into what becomes the Dagpo Kagyü — the very tradition whose founder would, within a generation or two of Zhang ston's own activity, name The Exposition of the Jewel-Lineage as Mahāmudrā's basic text. The same decades that were producing the Seventeen Tantras in Central Tibet were producing the Tibetan reception of the entire Mahāsiddha corpus elsewhere in the same country, often among monastics and translators who moved in overlapping circles, studied in overlapping regions of dBus and gTsang, and drew — as the scholarship on the sNying thig's formation independently argues on entirely separate grounds — on a wider eleventh-century Tibetan religious environment that was, at this specific moment, undergoing a documented process of “re-Tantricization," pulling vocabulary, imagery, and practice back in from the broader Tantric world after an earlier, more austere phase of Dzogchen had deliberately shed it.
If the redactors of the Seventeen Tantras were working, in this specific window, inside a Tibetan religious culture saturated with newly arriving Mahāsiddha material — the same culture in which the word mahāmudrā was, at that very moment, still closer to its old, unsystematised, pointing-instruction sense than to the graded yoga curriculum it would later become in Kagyü scholastic hands — then a pun on Vimalamitra's name built from exactly that vocabulary, and a technical term for awareness demonstrably drawn from exactly that same siddha-song current, stop looking like coincidences requiring explanation and start looking like exactly what one would expect two contemporaneous streams, drawing on the same well, still upstream of their later sectarian separation, to produce.
Whether that constitutes “the same overall stream," in the strong sense or something looser — two traditions independently reaching for a shared fund of available Buddhist vocabulary without ever being, in any meaningful sense, undivided — is exactly the kind of judgment call that a chronological argument of this kind can motivate but cannot, by itself, prove.
VI. The Question, Left Open
Longchenpa's great systematising distillations of the sNying thig — the works that gave the Seventeen Tantras their settled, authoritative place at the summit of the Nyingma path — lie two full centuries after Zhang ston's redaction, and it is only through that later distillation that most practitioners today encounter this material at all: pre-digested, cross-referenced, folded into a curriculum with its own settled vocabulary, its own careful distinctions from every neighboring vehicle including, by then, a fully institutionalised Mahāmudrā with centuries of its own commentarial tradition behind it. What sits underneath that later settlement — in the actual root syllables of the tantras themselves, redacted in the early twelfth century by figures working in the same soil that was, in the same years, receiving an entirely different stream of Indian material under an entirely different name — is something this essay has tried to bring back into view rather than resolve.
Perhaps the honest shape of the answer is that there is no single answer: that in the earliest layer of this material, before either tradition had hardened into a school with borders to defend, phyag rgya chen po was simply one of several available words — alongside rig pa, rang rig, bde ba chen po, lhan cig skyes pa — for pointing at the same recognition, used by different circles of practitioners and redactors who would not, for another century or more, have had any particular reason to sort their vocabulary by lineage at all. If that is so, then the deepest continuity between the Great Seal and the Great Perfection may not be a matter of one influencing the other, or of a shared origin later obscured, but something stranger and perhaps more interesting: two names, arrived at from different directions, for a recognition that was never, in either tradition's own earliest and least self-conscious voice, two different things to begin with.
Whether that is true is not something this piece can settle. It is, I hope, now at least a question worth taking into the Tibetan itself.
A note: the primary textual claim in this essay — the occurrence of phyag rgya rgya chen in the sGra thal 'gyur as a punning epithet for Vimalamitra — is drawn from a 2024 Berkeley dissertation on the origins of the rDzogs chen sNying thig, which cites the Tibetan directly and provides the Sanskrit reconstruction. The account of Mahāmudrā's pre-institutional history, the tracing of rig pa/rang rig to saṃvit/svasaṃvedana in the Mahāsiddha and Śaiva material, the "great bliss" language in the early Mind Series corpus, and the chronological reconstruction of the Seventeen Tantras' redaction draw on direct, independent study of the root texts.
