Atīśa’s Lamp for the Path to Enlightenment: Its Influence on Lamrim and Dzogchen Ngöndro
Few texts in the history of Secret Mantra have exercised an influence so disproportionate to their length as Atīśa Dīpaṃkaraśrījñāna’s Bodhipathapradīpa — some sixty-nine verses, composed at Nethang around 1042 CE at the invitation of the West Tibetan monk-king Jangchub Ö (Bodhiprabha). It is tempting, given its brevity, to treat the text as a modest devotional summary, a courtesy extended to a royal patron. Such a reading would badly undersell the work. The Bodhipathapradīpa is better understood as a compressed architectural drawing — a blueprint from which an enormous portion of the subsequent edifice of Secret Mantra practice literature was built, quite often without full acknowledgment of the debt.
Two claims will be developed here. First, that the entire lamrim (lam rim, “stages of the path”) genre, in every school of Secret Mantra, is Atīśa’s direct descendant — while Atīśa himself was working from an already-existing Indian synthesis, most decisively that of Kamalaśīla’s Bhāvanākramas.
Second, and more provocatively, that the sequence of practices constituting the Nyingma Ngöndro (sngon 'gro, “preliminary practices”) — refuge and prostration, bodhicitta generation, purification, maṇḍala offering, and guru yoga, culminating in receiving the empowerments and readiness for the pointing-out instruction — reproduces, with remarkable structural fidelity, the very sequence Atīśa lays out in his short text. If this is correct, it suggests that even the tradition most closely associated with a “sudden,” non-gradual view of awakening — rDzogs chen — organises its own approach to that view along essentially Atīśa’s architecture. The final section considers what the Bodhipathapradīpa itself, read carefully, has to teach us about the nature of Dzogchen practice as such.
Kamalaśīla’s Influence: The Bhāvanākramas as Deep Structure
Atīśa did not invent the graduated path; he inherited it, and the single most decisive prior articulation of it is Kamalaśīla’s three Bhāvanākrama treatises, composed a quarter-millennium earlier in the aftermath of the so-called Samye debate. Kamalaśīla’s defense of the gradualist position — against the Chinese Chan master Heshang Moheyan’s advocacy of an instantaneous, non-conceptual approach requiring no accumulation of method — produced a template that would prove almost impossible for later Tibetan systematisers to escape, whatever their sectarian affiliation:
Bodhicitta as the indispensable motivational ground; the cultivation of calm abiding (śamatha) as the necessary basis for any subsequent insight; analytical investigation into the emptiness of persons and phenomena; and, crucially, the union of calm abiding and insight, of method and wisdom, as the actual mechanism of liberation rather than either pole alone.
One does not need to search hard for this template inside the Bodhipathapradīpa. Verses 39 through 55 reproduce Kamalaśīla’s sequence almost step for step: the insistence that intuitive knowledge cannot arise without calm abiding (v. 39); the warning that meditative effort without the correct supporting factors is wasted no matter how prolonged (v. 40); the passage from calm abiding to the cultivation of wisdom (v. 42); and above all the axiom, stated with the same rigor Kamalaśīla brings to it, that wisdom without method and method without wisdom are both, symmetrically, forms of bondage (v. 44).
This is not incidental resemblance. Atīśa was a formally trained Madhyamaka master educated within the very monastic university culture — Vikramaśīla, Nālandā — that had absorbed and canonised Kamalaśīla's model as orthodoxy. What Atīśa contributes that is genuinely his own is not the underlying structure but its extraordinary compression, its integration with tantric practice (verses 61–68), and its explicit grounding of the entire edifice in the threefold typology of motivation (vv. 2–5) — the scopes of small, middling, and great capacity that would become the organising device of every subsequent lamrim.
The Lamrim Genre as a Chain of Commentaries on Atīśa
It is not an exaggeration to say that the lamrim genre, across all four major Tibetan schools, exists as a series of expansions upon the Bodhipathapradīpa’s skeleton. Atīśa’s own direct disciples and their successors — Dromtönpa, Potowa, the geshes of Sangphu — treated the text as foundational scripture for the tradition that would eventually bear the name bKa' gdams pa, “those bound by the word.”
Gampopa, uniting Kadam graduated-path methodology with Mahāmudrā instruction inherited through Milarepa, produced the Jewel Ornament of Liberation, whose architecture of persons of three capacities, refuge, karma, renunciation, bodhicitta, and the six perfections is Atīśa’s outline given flesh.
Tsongkhapa’s Lamrim Chenmo, the great synthetic monument of the Geluk tradition, opens by naming the Bodhipathapradīpa explicitly as its root text and structural template — an act of citation rare in its directness, functioning almost as a colophon of intellectual debt.
Sakya and Kagyu lamrim literatures, whatever their doctrinal inflections, retain the same three-scopes chassis. Even where a tradition’s ultimate view diverges sharply from Geluk Madhyamaka-Prāsaṅgika orthodoxy, the sequence by which a practitioner is led toward that view remains recognisably Atīśa’s: motivation before method, ethics before concentration, concentration before wisdom, and — decisively — the refusal to let any one of these stand in for the others.
The Ngöndro Hypothesis
This brings us to the more striking claim. The Nyingma preliminary practices, as they reach their classical codification in works such as Patrul Rinpoche’s nineteenth-century Kunzang Lama'i Shelung (itself transmitting Jigme Lingpa's Longchen Nyingthig cycle), are conventionally read as belonging to a separate, tantric current of transmission — the Nyingma lineage’s own inheritance from the earlier translation period, developed largely independently of the “New Translation” (gsar ma) scholasticism that Atīśa represents. Yet set the standard Ngöndro sequence beside the Bodhipathapradīpa’s own progression and the correspondence is difficult to dismiss as coincidence.
Atīśa opens the practical portion of his text with refuge, taken bodily — kneeling, palms joined, repeated three times (v. 9) — exactly the posture and structure of the Ngöndro’s outer refuge practice, itself paired with prostration. He proceeds immediately to the generation of bodhicitta through contemplation of universal suffering (vv. 10–12), formalised into an actual liturgy of vows (vv. 27–32) — the direct structural ancestor of the Ngöndro’s bodhicitta section. The image, in verses 15–18, of an offering of jewel-filled Buddha-fields “as numerous as the sands of the Gaṅgā,” surpassed instantly by a single sincere turning of the mind toward awakening, is the same rhetorical and devotional logic that animates the Ngöndro’s maṇḍala offering — a practice of symbolically offering the entire universe, precisely in order to demonstrate that the value of the gesture lies not in the material vastness offered but in the completeness of the intention behind it. Atīśa’s insistence on purifying “all actions of body, speech, and mind” (v. 32) anticipates, structurally, the purificatory logic of Vajrasattva practice, even though the specific tantric method differs. And Atīśa’s extended treatment of the qualified teacher and the necessity of pleasing him “through devotion... by offering... and by honouring all his instructions” (vv. 23–25, 63) is precisely the logic that culminates, in the Ngöndro, in guruyoga — the practice in which devotion to the teacher becomes the very door through which the higher instruction, in the Nyingma case the pointing-out of rigpa, is received.
Several careful qualifications are owed here, in the interest of scholarly honesty rather than rhetorical enthusiasm. The Ngöndro as a fixed, five-part liturgical set is a comparatively late historical crystallisation, and its named architects — Jigme Lingpa above all — were not working under any documented instruction to model their sequence on the Bodhipathapradīpa specifically.
The claim being advanced is therefore not one of direct, attested textual dependency in the manner of Tsongkhapa’s explicit citation, but of structural convergence: a deep grammar of graduated preparation — orientation, motivation, purification, accumulation, devotion, then opening — that had, by the time of the later Nyingma systematisers, become so thoroughly naturalised across Tibetan practice through centuries of Kadam, and later Geluk and Sakya, lamrim pedagogy that it could hardly help but shape how any Tibetan tradition, including one asserting a philosophically distinct view, would organise its own preliminaries.
The Nyingma tradition did not need to copy Atīśa; it inherited an ecosystem in which Atīśa’s grammar had become close to the only available grammar for graduated preparation, absorbed through centuries of shared monastic education, cross-lineage study, and the simple prestige of the Kadam reform. To notice this is not to diminish the originality of the Ngöndro; it is to notice how thoroughly one small Indian monk’s compressed outline came to underwrite the practical architecture of a tradition that would, at the level of philosophical view, eventually distinguish itself quite sharply from his own Madhyamaka scholasticism.
What the Bodhipathapradīpa Illuminates in the Dzogchen View
If the structural debt runs this deep, it is worth asking what light Atīśa’s text — composed without any reference to Dzogchen terminology whatsoever — might nonetheless throw on the Great Perfection’s own understanding of practice.
Consider verses 56 through 58. Atīśa’s claim that “becoming itself is born of conceptual thought, and is, in its very nature, nothing other than that conceptual thought,” so that the relinquishing of conceptuality is nirvāṇa rather than merely a technique for approaching it, sits close to the Dzogchen distinction between sems, the conceptually elaborating ordinary mind, and rigpa, naturally occurring awareness prior to and untouched by that elaboration.
The image Atīśa cites — the non-conceptual “shining forth like open sky” once conceptual thought subsides — is functionally identical to the Dzogchen insistence that rigpa is not manufactured by meditation but uncovered, exactly as sky is not created when cloud dissolves. A Geluk scholastic reading this verse will hear a description of the vipaśyanā realising emptiness through sustained analysis; a Nyingma yogin will hear something close to a description of trekchö, “cutting through” to what was already primordially present. The verse does not adjudicate between these readings — and this is precisely the point. It suggests that the result both traditions are pointing toward may be considerably less divergent than their respective methodological vocabularies suggest, even where their accounts of how one arrives there differ sharply.
Equally instructive is Atīśa’s refusal, in verse 44, to let wisdom stand without method or method without wisdom, each alone constituting “bondage.” This axiom has frequently been invoked, historically, as a polemical corrective from lamrim-oriented scholars against a caricature of Dzogchen as a shortcut permitting the neglect of ethical and motivational groundwork. But read charitably, and in light of the above, the axiom is not external to Dzogchen at all — it is precisely why the tradition insists on Ngöndro before the pointing-out instruction is given, and why authentic Dzogchen masters have historically been as insistent as any Geluk master that view without the accumulated ground of compassion and purification collapses into exactly the bondage Atīśa describes: a wisdom, or a claim to wisdom, that has outrun the method meant to carry it.
Atīśa’s text, in this light, functions almost as an internal check within the Dzogchen tradition’s own house — a reminder, arriving from outside its specific vocabulary, of why the “sudden” teaching is nonetheless approached gradually.
Finally, verse 60’s image of passing “in due order” through stages of warmth toward the bodhisattva grounds, arriving finally at a point where buddhahood is “no longer distant,” offers a useful corrective to a common misunderstanding of Dzogchen as wholly without structure or sequence. Even a tradition organized around instantaneous recognition retains, in its account of stabilising that recognition — the sixteen bhūmis of the Yeshe Lama and the various measures of the strength of rigpa — a structured unfolding remarkably continuous with Atīśa’s own vision of gradual ripening. The instantaneousness belongs to the moment of recognition; the maturation of that recognition into full awakening remains, even for the Great Perfection, a path walked in stages, however differently those stages may be named.
What emerges, then, is not a claim that Dzogchen is “really” just Kadam thought in a more tantric dress, nor a diminishment of its distinctive view. It is the more interesting and more defensible observation that Atīśa’s small lamp, kindled from Kamalaśīla’s earlier synthesis and passed down through the entire lamrim tradition, continues to illuminate corners of all Secret Mantra practice — including its most apparently “sudden” and view-centred school — that its author could not have anticipated, and that a careful reading of his text can still help us see.
