Query about the three wheels of doctrine

John Powers's picture

A question: does anyone know of tantric texts that refer to Vajrayāna as a third or fourth wheel of dharma? I remember seeing such references, but can't think of any specifics.

Categories: Query

Dear John,


My short answer to your question is "no." I too have come across this in English-language popular "Dharma" literature, but not in pre-modern Indic or Tibetan texts as far as I can remember. But Google tells me that Lama Surya Das and the Shambhala folks talk of a Fourth Turning, so perhaps I just haven't been reading the right books.


Though not directly relevant to the question of a fourth or fifth dharmacakra, the Kālacakra-inflected Hevajra commentary Ṣaṭsāhasrikā-hevajra-ṭīkā of Bodhisattva Vajragarbha (Malati J. Shendge ed. [Delhi 2004] p. 10) has the following passage in its introduction:


yānatritiyaniryāta ekayānaphale sthitaḥ |

śrāvakaṃ pratyekañ câtra mahāyānaṃ tṛtīyakam || 39


caturthaṃ nâsti bauddhānāṃ paṃcamañ ca mataṃ muneḥ | 40ab


The first two pādas are a silent quotation of [Mañjuśrī-] Nāmasaṃgīti135cd [Ronald M. Davidson ed. MCB 20 (1981), p. 59].


Again not directly addressing the question, but very interesting nonetheless, are Dol po pa Shes rab rgyal mtshan's ideas about a "Fourth Council" (bka' bsdu bzhi pa). As far as I can understand it, Dol po pa asserts only three dharmacakras, but his way of explaining them is unusual to say the least (Cyrus Stearns, The Buddha from Dolpo[Albany, NY: 1999] pp. 123 ff.).


Best wishes,


John Newman

Dear John and John,

no, it is not fictive. There must have once existed the view that Mantra is a fourth wheel. In a Tibetan text composed in the late 12th or early 13th century, 'Jig rten mgon po says:

'o na theg pa chen po gsang sngags rdo rje theg pa 'di sde snod gsum gang du 'dus 'ji ltar yin snyam na/ kha cig gis gsang sngags sde snod bzhi par 'dod pa lags skad/

Sde snod gsum gyi nyams len bsil byed tsan dan gyi phreng ba. Jikten Sumgön’s Works 1:173–213, p. 176.
Jikten Sumgön’s Works: Khams gsum chos kyi rgyal po thub dbang ratna shrī’i bka’ ’bum nor bu’i bang mdzod. Collected Works of ’Jig rten gsum mgon. Edited by H. H. Drikung Kyabgon Chetsang (Konchog Tenzin Kunzang Thinley Lhundup). Dehradun: Drikung Kagyu Institute, 2001.

But I too have not been able to identify the proponent.

Best wishes,


RUB Bochum


Tibetan Rnying-ma-pa sources do regularly speak of a "gsang sngags chos 'khor," *Guhyamantradharmacakra," "dharma wheel of secret mantra." But I am not aware of this being referred to as a fourth wheel of dharma in traditional sources. My impression has been that it is implicitly regarded as belonging to the thrid turning of the wheel. In fact, much the same may be said of the Dge-lugs-pa approach, as set out by Mkhas-grub-rje; see. Lessing & Wayman, Mkhas grub rje's Fundamentals of the Buddhist Tantras (Mouton 1968), which also speaks of a "gsang sngags chos 'khor," but clearly as part of the third wheel.

However, it is possible that some contemporary Western teachers have taken talk of a "mantra wheel" to be in effect a fourth.

On the Jo-nang-pa master Dol-po-pa's view of the three wheels, you may wish to look at my The Tibetan Assimilation of Buddhist (OUP 2000), pp. 106-119. It seems clear that, although he spoke of "four councils" and "four ages" of the doctrine, when it came to wheels he referred to the canonical three as taught in the Sandhinirmocana.

best to all,

Matthew Kapstein, EPHE, Paris, and the University of Chicago

Dear Jan,

Your text speaks of a fourth piTaka (sde snod) not a fourth wheel. Rnying ma sources do often speak of a fourth vidyAdharapiTaka, and this is a term indeed known from Indic materials. But it is not the same as a fourth wheel.

all best,

Matthew Kapstein
EPHE, Paris

Dear Matthew,

I though so too, but at least in the eyes of 'Jig rten mgon po it is the same thing. The sixth vajra statement of the first chapter of rDo rje shes rab’s commentary on the dGongs gcig says:

'dod pa'i khams dang gzugs med du/ /sangs rgyas rnams ni sangs mi rgya/ /gzugs kyi khams kyi 'og min du/ /yang dag sangs rgyas der sangs rgya/ /sprul pa po zhig 'dir sangs rgyas/ zhes gsungs pas/ 'di ni rig pa 'dzin pa sngags kyi sde snod bzhi pa dang / sngags chos kyi 'khor lo bzhi par 'dod pa'i lugs de yin gsungs//

Our colleague in Hamburg, Dorji Wangchuk, seems to have given a paper called “Overt and Covert Indian Sources on the Vidyādharapiṭaka” focusing on the topic of the sde snod bzhi pa.


Dear colleagues,


I look forward to John Powers' nītārtha explanation of all this, but as I understand it trying to fit vajrayāna into one or the other of the last two dharmacakras (the first dharmacakra is clearly irrelevant) would be a category mistake. The Saṃdhinirmocana dharmacakra scheme lays out a hierarchical categorization of Śākyamuni Buddha's statements about ontology. Since vajrayāna/mantrayāna is merely the mantranaya subdivision of the mahāyāna (the other being the pāramitānaya), placing vajrayāna/mantrayāna as a whole exclusively in one or the other of the last two dharmacakras would be as artificial as placing the mahāyāna as a whole within one or the other of the last two dharmacakras. Of course specific statements in the tantras pertaining to ontology could be categorized and argued about as nītārtha or neyārtha within the Saṃdhinirmocana scheme, but that is another matter altogether.

            I wonder if the impulse to "elevate" Vajrayāna to the status of a Fourth Dharmacakra is not similar to the modern inventions "Sahajayāna" and "Kālacakrayāna" which, as far as I know, are not found in pre-modern Indic or Tibetan sources? In any case, it is indeed interesting to note that this idea can be found in pre-modern Tibetan sources, and it would be very interesting to trace its Indic roots, if any.

            Following up on Dol po pa's ideas about the dharmacakras, he says: "I bow to you who teach that the ultimate Dharma Wheel is the Final Wheel, the ultimate vehicle is the Mahāyāna. The ultimate Mahāyāna is the vehicle of the Buddha-nature, and the ultimate Buddha-nature is great bliss" etc. (Stearns trans. p. 119, op. cit. in previous post). As Stearns notes, Dol po pa holds that "the Buddha-nature...is pure, self (ātman), blissful, and permanent (gtsang bdag bde dang rtag pa)" (p. 241, n. 37). As previously mentioned in this thread, Dol po pa posits the standard set of three dharmacakras, but he radically "redefines" their meaning and content (Stearns pp. 86 ff.) in line with his interpretation of the Saṃdhinirmocana trisvabhāva doctrine and other ideas.



Dear John,

While this does not refer to exclusively tantric texts, here is how at least two thinkers tried to fit Tantra into the categories the dharmacakras, etc. Kongtrül Lodrö Tayé in his Shes bya kun khyab discusses two types of dharmacakras—common and uncommon—and then divides the former into the three dharmacakras and the latter into different tantric teachings of Kriyā, Caryā, etc. (mi rigs dpe skrun khang, 1982, stod cha, 359ff.). While on the surface it looks like he does not include Tantra into the three dharmacakras, this impression might be somewhat superficial (after all, he also discusses Vajrayāna separately from Mahāyāna, but this does not entail that for him the former is not included into the latter). Anyhow, Shakya Chokden, on the other hand, clearly treats Tantra as a subdivision of the third dharamākra. He talks about the final (i.e., the third) pronouncements in terms of Pāramitāyāna and Mantrayāna (see my Radiant Emptiness, 105) and also divides the last dharmacakra into Pāramitā and Mantra, with the former further divided into Sūtras and Abhidharma (239).

As a side note, it is interesting to mention that while for Tibetan thinkers (Bsam gtan mig sgon, etc., aside) the question was usually not which teachings—sūtric or tantric—where higher, but rather how high the tantric teachings were (whether they should be included into the third dharmacakra or treated separately as the fourth dharmacakra, etc.), Japanese thinkers disagreed on whether Tantra as such (still as a part of Mahāyāna) should be elevated above all sūtric teachings (as Kūkai did) or be incorporated within the framework of the Lotus Sūtra teachings as interpreted in Tendai (according to Saichō). Hence the two different takes (Tōmitsu vs. Taimitsu) followed by Shingon and Tendai traditions ( see J. Stone, Original Enlightenment and the Transformation of Medieval Japanese Buddhism, 20-21).

While I agree with the general trend of these remarks (i.e., Vajrayāna was not considered a "third wheel," at least in the Saṃdhinirmocana sense). However, there are reasons to think that some may have had some ideas trending along these lines. For instance, the Ninth Chapter of Āryadeva's Caryāmelāpakapradīpa (when he takes up the edgy notion of sensual practices) reads:

“First of all, the Lord, in the condition of a bodhisattva in his last existence, having surveyed the continent [on which he was to be born] and so on, having descended from residence [in] Tuṣita [Heaven]...having manifested himself in a form free of passion, undertook the realization of the Four Noble Truths and the passion-free practices for those of inferior ambition (hīnādhimuktikānāṃ). Further, for those who adhere to the Mahāyāna,  [He] undertook the realization of objective selflessness [such as that] of the eight consciousnesses, {the bodies,} etc., and [undertook] the practices of the stages and perfections, and so on. Further, emanating in the form of a universal monarch for those who aspire to the profound [Adamantine Way, He] undertook the realization of the nonduality of the two realities and the practices of the objects of passion." (adapted from Wedemeyer 2007, pp. 281–282)

This is not too far afield from the idea of the Buddha's teaching career being divisible into three phases of realism, anti-realism, and a combination of the two, but Āryadeva does not explicitly identify these as "turnings of the wheel."

For a more modern take on this issue: The doctrine of the three wheel bodies (sanrinjin) figures prominently in Shinnyo-en (the Garden of Truth), which is a modern Japanese lay reform movement of Shingon esoteric Buddhism. According to Shinnyo-en, the three main images of Sakyamuni Buddha, Kannon (Avalokitesvara) and Fudo myoo (Acala vidyaraja) embody the three wheel bodies of buddhahood. "...the doctrine of Dainichi’s three wheel bodies (sanrinjin) basically categorizes the Buddha’s modes of teaching (lit. “turning of the wheel” of the dharma) into three main body-types. A Buddha’s Truth body (lit. self-nature wheel body jishōrinjin) teaches simply by manifesting the truth or “thusness” of reality. Bodhisattvas like Kannon reincarnate as Law bodies (lit. true dharma wheel body shōbōrinjin) to expound the dharma through speech and other compassionate methods.Fierce wisdom kings like Fudō appear as Teaching bodies (lit. teaching edict wheel body kyōryōrinjin) to teach and dispel stubborn delusions by any expedient means necessary. These three pedagogical and soteriological aspects of Dainichi are ultimately empty of fixed essence, however, for like white light refracting through a prism into three distinct colors, they are effectively indistinct from universal Buddhahood itself."

Pamela D. Winfield (2019) Shinnyo-En and the Formulation of a New Esoteric Iconography, Material Religion, 15:1, 27-53, DOI: 10.1080/17432200.2019.1568756


Dear colleagues,


I hope I am not beating a dead cakra, but I am still perplexed by the notion that mKhas grub rje (or any dGe lugs pa, for that matter) would categorize the Vajrayāna in the third dharmacakra. I went back to look at Matthew Kapstein's original contribution in this thread that suggest this. He says: "In fact, much the same may be said of the Dge-lugs-pa approach, as set out by Mkhas-grub-rje; see Lessing & Wayman, Mkhas grub rje's Fundamentals of the Buddhist Tantras (Mouton 1968), which also speaks of a 'gsang sngags chos 'khor,' but clearly as part of the third wheel."


I suspect Matthew is mistaken. mKhas grub rje does subdivide (Lessing & Wayman Tib. p. 40) the teaching of the Mahāyāna into a pha rol tu phyin pa'i theg pa'i chos 'khor 'khor tshul and a gsang ngags kyi theg pa'i chos 'khor 'khor tshul (Tib. p. 100 ff.), but I have not found a place where he connects this to the Saṃdhinirmocana 3-dharmacakra scheme. Where mKhas grub rje does­ in fact discuss the Saṃdhinirmocana 3-dharmacakra scheme (Tib. p. 52), he presents the standard dGe lugs interpretation that takes the second dharmacakra as teaching Madhyamaka, which is nītārtha, the other two dharmacakras being neyārtha. To state the obvious, it would be bizarre for mKhas grub rje to hold that the Vajrayāna is neyārtha.


I did not reread the entirety of mKhas grub rje's rGyud sde spyi'i rnam gzhag, so perhaps I am missing something. If so, I would be grateful for any clarification others might offer.


With best wishes,



Dear John,

To clarify - I nowhere suggested that Mkhas grub assimilates the tantras to the Sandhinirmocana's scheme. The Dge lugs pa drew on several sources for their three wheels' doctrine, not the SNS alone. Where I was indeed mistaken, however, was in stating that Mkhas grub, like the Rnying ma pa, connects the tantras to the third wheel (under whatever interpretation). He does not. He connects them to the second:
See, in his discussion of the second wheel, Wayman & Lessing, p. 92 (text), 93 (trans.):
sngags kyi rgyud sde thams cad kyi lta ba thal 'gyur ro
"The doctrine (darśana) of all [four] sections of the tantras is Prāsaṅgika."
and, less explicitly, p. 94 (text), 95 (trans.), where, still in the section on the second wheel, he says that he will expand on the tantras later in the book, evidently suggesting that the topic does belong to the second wheel.

In the context of the original query of John Powers, my point was just to note that the traditional sources known to me from Tibet do not accept the notion of a fourth wheel, even if they do affirm a "dharma-wheel of mantra."

I hope that this resolves the puzzlement.


Matthew Kapstein
EPHE, Paris, and the University of Chicago