Where did Mahāvīra die?

The suttas say that Mahāvīra died at Pāvā. Jain texts also say he dies at Pāvā. Case closed!

I’m sorry, have you met me? Not so fast!

Most sources will tell you that Jain tradition places Pāvā in Magadha near Nalanda. And indeed that is where modern Jains go to honor Mahāvīra.

The suttas, on the other hand, place Pāvā in the Mallian country, i.e. north of the Ganges, about 150 kms or 4 day’s walk from Pawapuri.

It’s here that things start to get complicated, so buckle up or bail out! Also, I only have inadequate sources, so if anyone can help out with better sources I’d be grateful. Please bear this in mind throughout, this may well all be wrong!

where do Buddhist sources locate Pāvā?

In Malla. Except they don’t. The well-known Pāvā is in Malla, and is securely located along a journey in multiple passages. But when speaking of Mahāvīra’s death, they don’t actually specify that it is that Pāvā. Now the obvious inference is that it is the Mallian Pāvā, but it’s not actually stated. There is therefore no explicit contradiction if we assume it was in Magadha.

where do Jain sources locate Pāvā?

In Magadha. Except they don’t. The source typically cited here, when anyone gets more accurate than just “Jain tradition”, is the Kalpa sutra. I haven’t been able to locate an original text for this online, but the translation is here.

The Jain accounts are terribly interested in specifying the astrological conjunctions, but rather less interested in saying where anything happened. The only clarifying detail I can find is this:

In that night in which the Venerable Ascetic Mahāvīra died, &c. (all down to) freed from all pains, the eighteen confederate kings of Kāśī and Kośala, the nine Mallakis and nine Licchavis, on the day of new moon, instituted an illumination on the Poṣadha,

Here we have mention of the ruling clans who celebrated Mahāvīra’s passing. Notably, it includes the Mallas, a minor republic, while omitting Magadha. Surely this indicates that for this text, Pāvā is more likely in Malla than Magadha.

This text names a local king as Hastipāla, but this doesn’t help as he is not mentioned elsewhere.

A much later text, Triṣaṣṭiśalākāpuruṣacaritra, says he went to Apāpa (on this below), without specifying the location. The only detail I can find is that afterwards his disciple went to Rajagaha, but that’s hardly decisive (the Buddha’s disciples also went to Rājagaha after his passing).

There must be some Jain sources that locate Pāvā in Magadha, but I can’t find them. Maybe it is a modern tradition?

is Pāvā in fact Pāvā?

The name is a bit of a mystery, and I’m not sure there is an explanation for it. Probably it’s dialectical.

In Pali it is always called Pāvā. The Jain texts are in a variety of languages which rather confuses the situation. Often it is cited as Pāpā. Which is inconvenient as pāpa means “evil”. This is probably why some texts then cite it as Apāpā, “non-evil”.

To further muddy the waters, in the most distinctive Jain language, Ardhamagadhi, “evil” is spelled pāva. So in Buddhism, evil is pāpa and the place is pāvā, while in Jainism, evil is pāva and the place is pāpā.

Confused yet? We’re just getting started!

who to believe?

Normally in a case like this we should obviously defer to the Jain tradition. The passing of their great teacher is a matter dear to their hearts, whereas to Buddhists it is not.

But in this case, the Buddhist sources are much earlier, maybe a thousand years earlier. There are earlier Jain texts, but so far as I know they say nothing of this. And even a passing glance at the relevant Jain texts shows they are far more concerned with astrology and miracles than with historical accuracy. Compared with them, the Mahaparinibbanasutta is a dry litany of journalistic facts. And as we have seen, there is no actual location for the Jain Pāvā anyway.

So in this case there is good reason to trust the Buddhist account over the Jain.

is Pāvā Pāvārika?

The suttas know of a place called “Pāvārika’s mango grove”, which is outside of Nalanda. In both MN 56 and SN 42.8 this is the location for a discussion concerning Jains; Mahāvīra stayed nearby.

Is it possible that Pāvārika is the Jain Pāvā?

Pāvārika is said in the commentary to the the name of a wealthy local man who owned the mango grove. Note, however, that the word means “quilter”, someone who manufactured and traded in blankets and suchlike.

Now, it is a common practice that villages have a primary economic activity, after which the village is named. Thus when we see names like Hatthigāma (elephant village) or Ambagāma (mango village) we need no special pleading to assume these were villages of elephant trainers and mango growers respectively. And it’s likely that Pāvārika follows the same pattern, being a village of blanket weavers. This, of course, does not contradict the assumption that it was named for a wealthy merchant, since it would be normal that the wealthiest villager would be the one who trades in their goods.

Thus “Pāvārika’s mango grove” is probably “the mango grove of the (head of the) blanket-weavers”.

The root term is pāvāra, (“blanket”), and it is easy to see that the ending could be left off, a normal feature of dialectical change in Indian names.

If we are on the right track, then, the Pāvā in Magadha is none other than the village associated with “Pāvārika’s mango grove”, which was a haunt of the Jains.

We are now left in the rather awkward position that there is contemporary evidence for “Pāvā” in Magadha in Buddhist texts, while Jain texts suggest it was in Malla.

who’s confused now?

Well I am for a start. Are the Buddhist texts conflating “Pāvārika’s mango grove” with “Pāvā”, thus implicitly locating Mahāvīra’s death in Magadha? And meanwhile the Jain texts are implicitly locating it in Malla?

Or did later Jains confuse “Pāvārika’s mango grove” with the by-then forgotten town of Pāvā?


The Arthaśāstra names the regions that horses were imported into the rest of mainland India from —

  1. prayogyānām uttamāḥ kāmboja-saindhava-āraṭṭa-vanāyujāḥ - “The best horses come from Kamboja (Persia), Sindhu (Southern Sindh), Araṭṭa (parts of Punjab?) and Vanāyu (Arabia)”
  2. madhyamā bāhlīka-pāpeyaka-sauvīraka-taitalāḥ - “The average horses come from Bahlīka (Bactria), Pāpā (Pāli spelling: Pāvā), Sauvīra (Northern Sindh) & Titala??”
  3. śeṣāḥ pratyavarāḥ - “The rest are low quality.”

Most scholars consider Pāpā and Titala to be towns/regions located in the north-west (with possible connections to Central Asia along the silk road) since North-Eastern India has no history of horse breeding, much less exporting them to other locations.

Alexander’s war against the Mallas happened in the Punjab (between the Hydaspes i.e. Vitastā river and the Acesines i.e. Asiknī) - and it appears that the Oxydraci (i.e. Kṣudrakas) allied with the Mallas against Alexander. Perhaps the western city of Pāpā was in this territory - as the Pāli canon also implies that the Pāveyyaka bhikkhus are westerners (and who are contrasted with the pācīnaka i.e. “eastern” bhikkhus – “adhammavādino pācīnakā bhikkhū, dhammavādino pāveyyakā bhikkhū”ti and again at sammatā saṅghena cattāro pācīnakā bhikkhū, cattāro pāveyyakā bhikkhū)

Moreover most Jains (and their oldest and holiest pilgrimage sites) have been concentrated in Western India since antiquity. Here is a map of their current population-density. Jainism in India - Wikipedia - which I also shared earlier. If Pāvā was the place where Mahāvīra (i.e. Nirgrantha Jñātiputra) died, it would make sense to assume that he lived and died in a region where most of his followers were located i.e. Western or North-Western India.

Interesting, I didn’t know that. Do you have any idea what the derivation of the word is?

This one is tricky, as in all relevant cases there is a variant reading pāṭheyyaka. Not really sure what that might mean.

Pāvā is to the west of Vesali, so it’s not impossible, but at the same time, Pāvā itself doesn’t really feature in the narrative and there’s no real reason why those monks should be called “of Pāvā”. Perhaps just because they came to Vesali “through Pava”? But elsewhere Pāveyyaka is used of the Mallians who live there.

I have suggested in the past that they are called “of Pava” due to a trope where monks “of Pava” are presented as an ascetic ideal, by implication associating them with Mahavira. But this could easily be a secondary association.

Some of them we know are not “of Pāvā”: Revata is in Soreyya, Sanavasin is from Mathura.

It’d be more sensible to assume the word meant “from the (north)-west”. If it were pātheyyaka I’d assume it was abbreviated from uttarāpātheyyaka.

Now this is one of those odd things that I’m not sure is relevant. But the Vinaya speaks of horse-dealers from Uttarāpatha at Verañja (pli-tv-bu-vb-pj1:2.1.2). This is quite close to Soreyya, where Revata was staying before the Second Council. In fact Verañja is the next stop north along the northern route, and the route travelled by the Buddha there is the same as that of Revata before the Second Council (Soreyya, Saṅkassa, and Kaṇṇakujja).

Since this sets up the beginning of the Vinaya, while the Second Council is the end, it’s likely that these two routes were deliberately specified so as to echo each other, placing the Buddha in the same region as the “good monks” of the Second Council. (This kind of “ring composition” feature is used often to unify these sprawling texts).

So based on the Pali texts alone, I’m inclined to favor the reading pāṭheyyaka in the sense “from the (north-western) trade route”, although I can’t justify the spelling.

If we are to stick with pāṭheyyaka, perhaps it has nothing to do with their location, but means “literalists”, i.e. those who stick to the letter of the texts—which is exactly what they did at the Council.

Worth noting though that there are many accounts of the Second Council in other Vinayas, so they should be consulted before reaching any conclusions. The Mahisasaka Vinaya has 波旬, which normally stands for pāpa, so supports pāvā.

On the other hand, if we accept the Arthasastra as indicating the presence of a Pāvā to the west, perhaps this is meant. Their distribution in modern times doesn’t mean anything, it was 2,000 years ago. Look at where Christians, Buddhists, etc. live today.

Anyway, all the early sources locate the Jains in Vesali and Magadha, so a western location for Mahavira’s death seems unlikely.

pāveyyaka & pāṭheyyaka (see image below) look very similar in Kharoṣṭhī, hence probably same issue as I’ve described here


1 Like

I may be off base here but I know of currently existing Pāvāgadh (fort of pāvā), a popular tourist and pilgrimage destination because of Hindu and Jain temples and mosques. It is in Gujarat, Champaner district. I am attaching two wiki links, which should have been titled pāvā not pava. Anyway, it might be relevant.


Champāner-Pāvāgadh Archeological Park

1 Like