The Water Nation and the Fire Nation

I have been listening to the audiobook version of the new history, The Dawn of Everything, by David Graeber and David Wengrow. This is the most important non-fiction book I have ever read. I urge everyone who has any interest in history or politics to read it. The basic argument is that human polities were much more pluralistic and changeable in the past and that the present global political monoculture of class-based hierarchies enforced by bureaucratised violence is, in fact, an oddity rather than a forgone conclusion.

In Chapter 8, the authors briefly discuss the Indus Valley Civilisation. Listening to Graeber & Wengrow describe the place of the water tank in Indus civil architecture I was struck by parallels I had seen with my own eyes in India, but also have encountered many times in ancient Indian texts. Indeed, this is one of the most iconic images associated with India: a person who walks down into the river, performs ritual ablutions and prayers, and emerges a new person. I emphasise ritual here, because waterways in India are now so filthy, fetid, and toxic that one cannot imagine emerging from immersion cleaner than before. Students of Indian history will know, however, that while water is important in Vedic culture, fire is the element they worship more than any other. Where does river worship come from? Well, there was a river based civilisation in India long before the Vedic tribes started arriving. I'm certainly not the first to observe this, but I'm riffing on an idea here and won't refer to others much.

The Water Nation

The Indus civilisation (ca 3300 – 900 BCE) sprawled over the floodplain of the Indus and Sarasvatī Rivers, which consisted of numerous settlements including some large cities. Their economy was varied. They grew numerous crops including wheat, barley, peas, and sesame—as well as raising domesticated animals such as buffalo, zebu cattle, sheep, and goats. They mass produced stone tools and items such as pottery and beads. Moreover, they traded far afield, by sea, with Indus pottery being found in Mesopotamian and Arabian archaeological sites.

The two largest cities, Harappa and Mohenjo-daro, were built to a similar plan. And I say "plan", advisedly, because these cities were clearly designed. Roads were typically wide and straight and had built-in drainage. Indeed, drainage was an important feature of the cities, presumably to allow annual monsoon rain waters to escape the town (and at least in some places to channel them into storage tanks).

The cities have a raised central area that 19th Century archaeologists called "the citadel" that is surrounded by an urban sprawl of mixed-use housing and craft-based light industry (bead making, metal working, etc). Yet there are few other signs of social stratification. Finds of high status objects are evenly distributed, rather than being centered on the citadel (it was apparently not a place where material wealth accumulated). The citadels were walled, but these seem not to have been defensive walls (unlike the walled cities of the second urbanisation where warfare was a constant threat).

At the center of the big Indus cities we do find large scale public architecture: sizeable meeting halls and accommodation for the people who might meet in them, large buildings that appear to be granaries, and so on. Notably at Mohenjo-daro we find a large water tank at the centre, called "the Great Bath" by archaeologists. The water tank was constructed from well-made bricks and sealed with clay and bitumen. The tank is about 12 m x 7 m, and has a maximum depth of 2.4 m and has steps allowing access at either end.

"At Mohenjo-daro, it seems, the focus of civic life was not a palace or a cenotaph, but a public facility for purifying the body." (Graeber and Wengrow: 316).

And this is is significant to Graeber and Wengrow's argument that civilisation does not necessarily entail coercion and control. As I say, for me this evoked images and memories of watching Indian people perform their ritual ablutions in tanks built for the purpose, as well as in rivers.

I'm concerned that no parallel structures have been found elsewhere. It's a definite weak point in the thesis I'm outlining. The only consolation is that archaeology of the Indus Civilisation has barely scratched the surface.

Like the dozens of other Neolithic (recent "Stone Age") or Bronze Age settlements that mark the beginnings of civilisation, the Indus cities were sited on rivers. This is not simply to provide for potable water. In fact, each home in the Indus cities was supplied with a well for this purpose (and drains to allow waste water to escape). One of the advantages of living next to a river is that the water table around it is easy to get at. Well water is filtered through river sediments, mostly clay, and provides reliably clean water. The main advantage of riparian life, however, was transportation. Rivers were the highways of the ancient world.

It now seems very likely that before about 2000 BCE, another river paralleled the Indus but has since dried up due to a drier climate. In Sanskrit, this river is called Sarasvatī. At this time, prior to rapid weakening of the monsoon, the Yamuna flowed west into the Indus Valley and may have been a tributary to the Sarasvatī. Scientists have also made links to the Ghaggar-Hakra system which now flows into the Thar Desert and simply dries up (however, it was like this by the time of the composition of the Ṛgveda, ca 1500-1200 BCE).

We don't know, but can easily imagine that these water-loving people made gods out of the rivers that sustained their way of life. Did they worship the rivers we now call Indus and Sarasvatī? This seems highly likely to me. Michael Witzel has pointed out that, in the Ṛgveda, Sarasvatī is a river but also appears without connection to the eponymous river. At times Sarasvatī appears to represent the milky way (the galaxy that our star, Sūrya, resides in). This mixing of symbolism could be suggestive of two cultures hybridising.

Living next to a river that floods annually due to the monsoons is not always easy. I vividly recall driving over the long Mahatma Gandhi Setu (bridge) in Patna on our way to Bodhgaya. At that point in the dry season the Falgu River (the ancient Narañjana) had completely dried up. The Falgu has a bed that is over 500m across. The Ganga in Patna was about 1 km across, although the river bed is 2.5km wide at that point. The problem is that the flood plain is much wider so the bridge actually spans 5.7 km.

Alluvial soils are extremely fertile and allow for intensive agriculture. Some Indus cities had extensive earthworks designed to move water from annual monsoon floods into storage tanks (e.g. at Dholavira).

Towards the end, the Indus Culture was beset by drought as the climate changed, resulting in a weaker monsoon. If the monsoon fails, India starves. The Indus citizens moved away from cities altogether. This has been seen as a rout, with the citizens fleeing the cities and probably dying in great numbers, leaving chaos, but the Sarasvatī gradually dried up over some centuries, so this process was not sudden from the point of view of those living it. Reading Graeber and Wengrow it seems just as likely to me that, following some very disappointing results, the Indus people just decided to change the way they lived. They moved northwards towards abundant water in the river valleys closer to the Himalayas, what we now call Punjab (i.e. pañca āpas = five rivers). But they settled in village-style communities with a more distributed polity and no centralisation (which emerged later, partly in response to the challenge of Buddhism). There is now no reason to think that this was not a conscious move away from an unsustainable lifestyle towards a more sustainable one, given changing conditions over many generations.

In other words we must not think of the people who lived in massive planned cities as idiots, as naive, or as passive. These were active people, who built great cities, who tamed the monsoon floods and channeled them into agriculture. We have to assume that they discussed what was happening at length and consciously decided how to survive the extended drought and, just as when they built the cities, put a plan into action.

In any case, according to Romila Thapar "The Ṛgveda lacks a sense of civic life founded on the functioning of planned and fortified cities." (2002: 110). Nor do we see elements such as elaborate drainage systems, or the use of fired bricks. It seems clear that the authors of the Ṛgveda did not know of the urbanised Indus Culture at all. What they met in India was a culture of distributed autonomous villages. They settled in this area, which they called the "Āryavrata", the area in which Ārya customs prevail) or the Kurukṣetra (Field of the Kuru), the Kuru being one main faction of the myths in the Mahābhārata, the other being the Paṇḍava.

The Fire Nation

During this same period we have another culture emerging on the other side of the Hindu Kush. Archaeologists have called this the Bactria–Margiana Archaeological Complex (or BMAC) or Oxus Civilization, since it is associated with the Amu Darya (which imperialists named the Oxus River following Greek historians) which flows north into the Aral Sea. Recent work dates occupation of the area to c. 2250–1700 BC.

Goods found in the archaeology tell us that the two cultures were aware of each other and trade between them took place. But they were genetically distinct with little sexual mixing between cultures.

This civilisation spoke an Indo-European language, and is possibly the place where Indo-Iranian split off and became a distinct branch of the family tree, though evidence for language in the ancient world is very limited. Archaeologists believe that the BMAC culture was the forerunner of the so-called Āryans, who split off from the general Indo-Iranian culture ca 1900-1700 BCE and eventually introduced their language and culture into India. While Indo-European languages came to dominate India, genetics tells us that in fact the Āryans were few and mainly men. Max Müller's "Āryan invasion", in which the Āryans were a large invading army, has long been deprecated, though the idea continues to exercise nationalists in India. Still, Indo-European influences were absent from India until the people who spoke Vedic arrived.

Incidentally, the story of the Anglo-Saxon invasion of Britain has gone the same way: there was migration, but not enough to overwhelm the local population. Most white British people have a genome that is predominantly of the same stock as Celtic peoples in Britain and Europe. What differentiates them is that, following the departure of the Romans after almost 400 years of occupation, the thoroughly Romanised British adopted the language (Old English) and customs of a minority group of immigrants from Northern Germany. Europeans have been coming and going from the British Isles since they could walk across Dogger Bank (end of the last ice age) because the land here is extremely fertile, and the winters are much warmer than in Europe because of the gulf stream.

The BMAC culture had domesticated horses from the steppes and wheeled carts. Most strikingly for us the BMCA architecture is built around temples in which fire was either worshipped or was a ritual medium for the expression of religious ideas. Background fire worship in Iran was absorbed into and became a feature of Zoroastrianism:

The Greeks, too, had a cult of the hearth fire, and although Herodotus (3.16) mentions the great veneration in which the Persians held fire, he does not single them out as being in any remarkable way “fire-worshippers,” nor does he know of temples of any kind among them (1 .131). A Zoroastrian temple cult of fire seems to have been first instituted in the later Achaemenid period [4th century BCE], being probably established by the orthodox as a counter-move to the innovation of a temple-cult, with statues of “Anāhīt.” (Encyclopedia Iranica)

Keep in mind that Herodotus (484 – 425 BCE) was reliable enough that later historian, Plutarch (46 – 119 CE), dubbed him "the father of lies". Since the yasna rituals "are essentially those of the Brahmanic yajña [i.e. fire-based ritual], the yasna evidently goes back to the proto-Indo-Iranian times." (Boyce 1997: 12)

Zoroastrianism also had a tradition of itinerant priests known as aθauruuan or āθravan (Old Iranian θ approximates Old Indic th). The similarity of Avestan āθravan and Sanskrit atharvan is striking. Strabo uses the term āθravan to refer to "fire keepers". The Atharvans (Pāḷi ātabbaṇa) seem to be distinct from the Brahmins and had their own sacred text, the Atharvaveda (though it has considerable crossover with the Ṛgveda). But the Atharvans are seen quite negatively; for example, early Buddhist texts paint them as evil and depraved magicians. Herodotus also mentions the Magi ("may guy"), one of six Iranian tribes, who "specialized in hereditary priestly duties and who assumed the duties of the athravan." (Zoroastrian heritage). My suspicion is that the Atharvans were the Āθravan.

The dates of Zarathustra are uncertain but likely to be around 1000 BCE ± 200 years. Records of the Zoroastrian religion are preserved in the ancient Avesta, which must have been roughly contemporaneous with the Ṛgveda (ca.1500 – 1200 BCE), which just fits within one end of the proposed date range for Zarathustra. Meanwhile, in India, fire worship morphed into Brahmanism and gave rise to the Ṛgveda. It appears that the ṛk "verses" were composed by men sitting around a fire taking a psychoactive compound that made them feel imaginative and heroic (the most likely candidate being an extract of Ephedra sinica), telling fantastic stories in metred verse about the Gods, especially Indra, Varuṇa, and Mitra (Avestan Miθra). The central concern of this stories is the maintenance of ṛta conceived of as a kind of cosmic harmony or balance operated by the gods who could be compelled by ritual to tip things in favour of "us" (whoever "us" might be). This theme of cosmic harmony is notably absent from European myth though it is present in Chinese myth (I make no claim here, I merely note the similarity of ṛta and dào 道).

In Parthian Iran (247 BCE – 224 CE), fire temples were constructed with a square base surmounted by a dome, the result looks uncannily like an old Buddhist stupa.

Bazeh Khur Fire Temple, Khorasan
Parthian era 247 BCE-224 CE.

Buddhist Stupa
Swat Valley, 1st-2nd Century CE.

A Confluence of Cultures

It is quite clear that the slow end of the Indus Civilisation was not the end of the Indus people, per se. They had a radical change of lifestyle and they did move upriver, but they persisted, with worship of rivers and daily ablutions forming a central part of their culture. It seems very likely, almost incontrovertible, that these were the people living in Panjab when Indo-Europeans started settling in the region.

As already noted, an influx of settlers does not constitute an "invasion", nor does the influx have to become a majority in order to effect massive social and political changes. The Romanised Celts of Britain voluntarily adopted the language and culture of a Germanic minority. Nor does it mean that the locals are hapless victims. Trade goods have been moving in both directions along the Khyber Pass for millennia, with trade connections as far afield as the Arabian peninsula. As with early Britain, people moved around a lot more during the Neolithic and Bronze Ages. Still, there are no signs of large scale migrations in the region. There was a small influx of (mainly male) Indo-European genes into India from Iran during the Bronze Age and a century or two later the resulting hybrid culture produced the Ṛgveda.

Analysis of the language of the Ṛgveda suggests early contact with a culture that spoke a language related to Muṇḍa, a branch of the Austro-Asiatic family of languages that includes modern Vietnamese, Khmer, and Mon. Michael Witzel (1999) has suggested that the source of Muṇḍa loan words was the remnant population from the collapse of the Indus cities, which seems entirely plausible to me. There are also Dravidian loan words, but Witzel argues that pattern of loanwords in Ṛgveda points towards initial contact with Muṇḍa and only later with Dravidian.

Linguistic features such as retroflex consonants, absent from Indo-European languages generally, appear in Vedic from our earliest records. Madhav Deshpande has referred to these as "regional" features, that is features of the subcontinent, rather than any particular culture. Indeed, they appear to increase in frequency from the older books of Ṛgveda to the younger.

Despite absorbing external influences, the Ṛgveda is still recognisably Indo-European and Vedic is obviously very closely related to Avestan and other Indo-European languages of the day. It seems that an existing culture was impressed enough by the Vedic-speakers to have adopted their language and customs creating a hybrid. We have to assume, following Graeber and Wengrow, and as appears to have happened, that this was a conscious decision.

The confluence of the water nation and the fire nation gave rise to a unique culture. the mature form of which is associated with Painted Grey Ware pottery in the late Iron Age (1200 - 600 BCE). The PGW culture produced a distinctive grey pottery and is associated with permanent village settlements, domesticated horses, and metallurgy (gold, silver, copper, iron). Note that Atharvaveda mentions iron, but Ṛgveda does not (and we need to be wary of arguments from absence as they are often weak unless there is an overriding expectation of presence).

We should also note that as far as Buddhist texts were concerned there were two main kinds of Brahmins: householder Brahmins who often attracted large groups of followers; and ascetic Brahmins, who often adopted jaṭila or dreadlocks (as they still do). Pāli suttas record kings granting land to Brahmins, whole villages it seems, which suggest that they were actively recruiting Brahmins to migrate to their kingdoms. Brahmins were often valued political advisors and royal ritualists (purohita). In particular, Brahmins had a panoply of lavish large-scale public rituals aimed at legitimising the power of the king (and all rulers who wield power over other humans need some justification for doing so, because we all instinctively hate being given orders).

Signe Cohen has conjectured that part of what drove the composition of the Upaniṣads was a conflict between conservative Brahmins from the Āryavrata and progressive Brahmins who had begun migrating east (into the world of the early Buddhists). The word āryavrata can be compared to the term Danelaw, it is the region where the customs (vrata) of the Ārya are upheld. Brahmins consciously referred to themselves as ārya (a word cognate with modern name for Persia, i.e. Iran or Īrān).

We tend to massively over simplify Indian history because we have just two literary sources: Vedic and Buddhist. Jains have been around as long, but their extant literary tradition dates from a later period. Jain tradition says that the first generation of texts was lost. Loads of other people were around, they just left no literature. We have to look to the archaeology and what it can tell us about this period.

Archaeology and linguistic evidence point to multiple communities speaking languages from at least three distinct language families: Muṇḍa, Dravidian, Vedic, along with an unknown number of minority languages. Note that some tribes in central India, until recently living a hunter gatherer lifestyle, speak in languages unrelated to any known language. This suggests a deep antiquity for these cultures, perhaps on the the scale of Australian Aboriginals, i.e. tens of thousands of years.

The hegemonic Vedic culture in India is clearly an amalgam of Indo-European and indigenous influences. A mix of Water Nation and Fire Nation.

~~oOo~~~

Bibliography

Attwood, J. (2015) "Who Were the Artharvans?" Jayarava's Raves 10 July 2015.

Boyce , M. (1997). "" In Companion Encyclopedia of Asian Philosophy, edited by Brian Carr and Indira Mahalingam, 4-20.

Clift, P. D., Carter, A., Giosan, L et al. (2012) "U-Pb zircon dating evidence for a Pleistocene Sarasvati River and capture of the Yamuna River." Geology 40 (3): 211–214. doi: https://doi.org/10.1130/G32840.1

Graeber, D. & Wengrow, D. (2012) The Dawn of Everything: A New History of Humanity. UK: Allen Lane.

Khonde, N., Singh, S.K., Maurya, D.M. et al. (2017) "Tracing the Vedic Saraswati River in the Great Rann of Kachchh". Nature: Scientific Reports 7, 5476 . https://doi.org/10.1038/s41598-017-05745-8

Mahoney, William K. (1998). The Artful Universe: An Introduction to the Vedic Religious Imagination. State University of New York Press.

Thapar, Romila. (2002). The Penguin History of Early India: From Origins to AD 1300. Penguin Books.

Witzel, Michael. (1999) Substrate Languages in Old Indo-Aryan (Ṛgvedic, Middle and Late Vedic). Electronic Journal of Vedic Studies (EJVS) 5(1): 1–67.

Yama and Hell

Japanese Yama (Enma)
as a Confucian administrator.

Yama is a fascinating figure. He rules over the afterlife, but is not one of the devas. Vedic myth names him as the first man to find his way to the realm of the ancestors (pitṛloka). He is thus a culture hero who opens the possibility of rebirth for Brahmin ritualists. The realm of ancestors starts off on the same level of the devas, and is progressively demoted until it becomes a place of torment and punishment. In parallel, the departed (preta) are transformed from the fortunate ones going to their ancestors, to a tortured group of ghosts stuck in limbo.

As we saw in an earlier essay, Yama has a twin sister Yamī. In fact the most likely meaning of the name Yama is 'twin'. Yama has a counterpart in Iranian myth called Yima and, in Avestan myth, the incest of the twins helps to found the human race. In the Ṛgveda the brother resists incest with his sister. I've written about the curious fact that the Pali suttas record that the Śākyas claim descent from a sibling incest mating, which I take to be evidence of their connection to Iran (see Possible Iranian Origins for the Śākyas and Aspects of Buddhism). Brother-sister incest was common amongst ancient Iranian royalty, a practice I believe them to have adopted on the Egyptian model. Some scholars have tried to link Yama to the Norse Ymir, but this is disputed.

Yama in RV 10.14 has two messengers which are brindle-coloured, four-eyed dogs (sārameyaú śuvā́nau caturakṣaú śabálau) with flared nostrils (urūṇasā́v). They wander among men, satisfying themselves on the breath of life (asu). However they are also keepers of the path (pathirákṣī)  and watch over men (nṛcákṣasau). Note that some authorities think that śabala (brindle) is cognate with Greek ḱerberos (spotted), the name of the Hades's 3-headed watchdog. Hades named his dog "Spot". The Buddhist Yama also two messengers though their form as dogs seems not to be mentioned.

Yama as we know him in early Buddhist texts is the ruler (rājan) of the rebirth destinations known as Niraya (Pali) or Naraka (Pāḷi & Sanskrit). PED derives niraya from nis+√i 'to go down' (nis- followed by a vowel become nir-). PED also cites a parallel in Buddhist Hybrid Sanskrit, though this is not listed in Edgerton's BHSD. However relating this to Classical and Vedic naraka is not straight forward. The vowels a and i are not interchangeable and the prefix nis or nir cannot simply become nar. It might work if there was an ancestral term such as nṛ or nara a real word meaning 'man, hero, person'. In secondary formations the vowel ṛ is strengthened to ra or ar. The word nṛ derives from an Indo-European root *ner and via Greek (a-nēr > andr) is the source of words such as androgynous, polyandrous, and philander. It comes into Sanskrit again as √nṛt 'to dance' (from the connotation of vigour). A naraka would then be something belonging to men or people, or heroic. And we can imagine Prakrit representing this as niraka. The substitution of -ya for -ka is conceivable as both can be adjectival. But this doesn't explain the nature of niraya/naraka. PED lists the etymology of naraka as "doubtful". I'll come back to this question after surveying the literature on Yama and Hell.


Yama in the Ṛgveda.

As a place of extreme suffering, the levels of Naraka are often referred to as "hell realms". One of the key early sources for the story of Yama as king of the afterlife is Ṛgveda 10.14.2:

yamó no gātúm prathamó viveda 
naíṣā́ gávyūtir ápabhartavā́ u |
yátrā naḥ pū́rve pitáraḥ pareyúr
enā́ jajñānā́ḥ pathíyā ánu svā́ḥ || 10.14.2 ||

Yama was first to discover this pasture that cannot be taken away.
Where our ancestor crossed over, all the born follow, by their own path.

As described here Yama seems to have been a man (or perhaps an earthly king) who was the first to discover the pitṛloka and be reborn (in heaven) along with his ancestors. Later in the Upaniṣads this is described as 'the world won by the ancestors' (pitṝṇāṃ jitaloka BU 4.3.33). Whether we should take this literally as representing the introduction of the idea of rebirth into Vedic cosmology or as a cosmogonical myth is not clear. Rebirth, though not absent as previously thought, is far from prominent in the Ṛgveda. Since rebirth is not a feature of Indo-European eschatology generally, it may be that as Indic speakers moved into the sub-continent they adopted a rebirth eschatology based on indigenous models. Rebirth does seem to be a regional feature of India thought. So taking this as a myth based on historical events is not entirely far-fetched.

There is a description of Yama's realm in a hymn to Soma (Ṛgveda 9.113.7-11). There an inextinguishable light (jyótir ájasraṃ) shines. It is a realm that is deathless and imperishable (amŕ̥te loké ákṣita). There heaven or the sky is bounded (avaródhanaṃ diváḥ) or perhaps "the inner apartment". It is the place where the dead are satisfied with sacrificial offerings (svadhā́ ca yátra tŕ̥ptiś ca). The refrain prayer of the Kavi in the deathless realm (amŕ̥te loké) is mā́m amŕ̥taṃ kr̥dhi "make me deathless". Which seems to be a prayer to be allowed to stay in Yama's realm instead of being reborn. As we will see in a subsequent essay the Ṛgveda is ambiguous on the question of the afterlife. This description is consistent with Vedic conceptions of heaven more generally. Thus the ancestors (pū́rve pitáraḥ), in this sūkta, seem to live in heaven.

We do find hell in the White or Śukla Yajurveda (30.5).

bráhmaṇe brāhmaṇáṃ kṣatrā́ya rājanyàṃ marúdbhyo váiśyaṃ tápase śūdráṃ támase táskaraṃ nārakā́ya vīraháṇaṃ pāpmáne klībám ākrayā́yā ayogū́ṃ kā́māya pum̐ścalū́m átikruṣṭāya māgadhám ||

For Brahman (Priesthood) he binds a Brahman to the stake; for Kshatra (Royalty) a Râjanya; for the Maruts a Vaisya; for Penance a Sûdra; for Darkness a robber; for Hell a homicide or a man who has lost his consecrated fire; for Misfortune a eunuch; for Venality an Ayogû; for Kâma a harlot; for Excessive Noise a Mâgadha. The Texts of the White Yajurveda, tr. Ralph T.H. Griffith, [1899], at sacred-texts.com

In the Śatapatha Brāhmaṇa (8th-6th century BCE?), as in the Bṛhadāraṇyaka Upaniṣad (sometimes considered to be an extra chapter of the ŚB), the dead are rewarded or punished according to their performance of the rituals. (Cuevas 271). By the time of the early Upaniṣads however the performance of rituals was seen as inferior to the performance of seeking ātman in one's heart (sometimes referred to as an internalised ritual). Ritual only leads to continued rebirth, whereas realisation of identity with ātman/brahman allowed the practitioner to escape birth and death all together. However there is still no sign of an afterlife destination in which wrong-doers are punished.

Yama in the Garuḍa Purāṇa (4th century CE?) is more like the Buddhist king of Hell as we find him in the Buddhist texts. The dead person is taken by the "High Way" and assumes a body formed from the funeral offerings (piṇḍa) and "feels hungry by day and night". The messengers of Yama are now torturers (Cuevas 271).

Is there Hell in the Ṛgveda Veda? 

Accounts of the afterlife in the Ṛgveda are far from unambiguous. Scholars have identified five Ṛgveda passages that might be a reference to hell: 2.29.6, 7.104.3, 9.73.8-9. 10.14.10-11, and 10.152.4 (Stausberg 2000: 219). The most suggestive passage is in sūkta 7.104 which calls on Indra and Soma to destroy an evil sorcerer (yātu) also called demon (rakṣa). In gāthā 3 the poet called on Indra and Soma:

índrāsomā duṣkŕ̥to vavré antár
anārambhaṇé támasi prá vidhyatam |
yáthā nā́taḥ púnar ékaś canódáyat
tád vām astu sáhase manyumác chávaḥ || 7.104.3 ||

O Indra and Soma, the evil doers were hurled into a pit which is beginningless darkness.
Not one returns from there, may your rage overpower them. [My translation]

Understanding this requires us to look at the context (a series of curses wishing harm and ill on an enemy) and the grammar of the sequence vavré antár anārambhaṇé támasi prá vidhyatam. The various translators produce similar translations:

• Stausberg "... throw them forth the evil doers into the enclosure, into the anchorless darkness."
• Doniger "... pierce the evil-doers and hurl them into the pit, the bottomless darkness."
• Griffiths " plunge the wicked in the depth, yea, cast them into darkness that hath no support,"

It seems Stausberg is struggling with the vocab: 'anchorless' as a reading of anārambhaṇa is peculiar. Ārambhaṇa means 'take hold of, seize; beginning, commencement'. As an adjective anārambhaṇa must mean something like 'beginningless', or as we would say "bottomless". Also vavra is a place of hiding or concealment, a cavern, cave or hole (from √vṛ 'to conceal') so enclosure also seems peculiar. Doniger is trying too hard here, she elects to use both meanings of pra√vyadh, i.e. 'pierce' and 'hurl' (293), where 'hurl' seems sufficient. Griffiths seems to grasp the phrase, but his pseudo-Biblical language is anachronistic. If we step through the structure of padas a & b:

índrāsomā duṣkŕ̥to vavré antár anārambhaṇé támasi prá vidhyatam |

The verbal form, prá vidhyata, is a passive past participle. Note that in Vedic the pre-verb is not always directly connected to the root. In Classical Sanskrit this would be pravidhyata. Indra and Soma are addressed using the vocative case. They are being asked to do the action of hurling (pra√vyadh) [verbal form] the patient, i.e. evil doers (duṣkṛta), into (antar) a hole/pit (vivra) which is darkness (tamas). It is ambiguous on the face of it whether it is a pit which is bottomless or the darkness which is beginningless (and presumably endless). However in RV 1.182.6 (below) we find anārambhaṇé támasi and 'pit' substituted by waters (apsu) suggesting that 'beginningless darkness' was intended.

I don't see why any translators might have chosen to refer to vivra with the definite article. Why "the pit"? It makes this seem like a reference to a known entity. Which pit is the text referring to? In fact no such pit exists in the text. It makes a great deal more sense, given that we have no definite referent, to use the indefinite article 'a pit'.

So the poet is simply asking his gods to bury his enemies in a dark bottomless hole so that they cannot return. This perhaps leaves open the possibility that this poet believed in rebirth and he wanted his gods not only to kill his enemies, but to prevent them from being reborn (a more comprehensive curse! He also requests that the gods burn, crush, shatter, scorch, kill, exile, cut down the same enemies. This does not seem to be a reference to Hell, the poet wishes the gods to punish his enemy in the here and now rather than in the afterlife; if anything he wants to deny them an afterlife. The poet is saying "O Lord, smite my enemies." It's a common theme in these ancient tribal scriptures. We find similar curses in the Old Testament of the Bible and the Avestan Hymn to Mithra.

In his discussion Stausberg highlights RV 1.182.6 which uses some of the same terminology:

ávaviddhaṃ taugriyám apsú antár anārambhaṇé támasi práviddham |
cátasro nā́vo jáṭhalasya júṣṭā úd aśvíbhyām iṣitā́ḥ pārayanti || 

Four ships most welcome in the midst of ocean,
Urged by the Asvins, save the son of Tugra,
Him who was cast down headlong in the waters,
Plunged in the thick inevitable darkness. [Griffiths]

However the context is very different. Tugra is rescued after being "cast into the bottomless darkness of the waters" (apsú antár anārambhaṇé támasi práviddham). Our conclusion is the complete opposite of Stausberg's. The two passages are linguistically similar in describing a hole and the deep ocean as bottomless and dark, but there's still no hint of a post-mortem destination.

RV 10.14 is a key sūkta for Yama and also contains some references that have been read as referring to Hell. However they don't mention any of the usual ideas associated with Hell. Indeed the suggested passages end with "grant him good-fortune and health, O King." (rājan svastí cāsmā anamīváṃ ca dhehi) Which doesn't sound much like Hell.

RV 2.29.6 makes a request to several pairs of gods—the twin Ādityas, Varuṇa & Mitra, Indra & Maruts—to be forgiven failings and to be saved the destruction of wolves (nijúro vŕ̥kasya), and from a pit (kartā́d) and from falling (avapada). The later two don't seem to be construed together, the request is phrases as "from a pit" (in the ablative singular). Note that this is a different word for a pit and it has absolutely no context that might relate it to hell.

RV 9.73.8-9 looks more promising. In this sūkta Varuṇa, guardian of the cosmic order, (r̥tásya gopā́) is asked to drive the hated ones, who don't perform the rites, into a pit (ávā́juṣṭān vidhyati karté avratā́n 8d) and those who are incompetent with fall into a pit (átrā kartám áva padāti áprabhuḥ 9d), unlike the wise (dhī́rāś). The word for pit is karta as in RV 2.29. However is the pit anything supernatural here, or is it a pit? 

Finally 10.152.4 In pada b Griffiths reads ádharaṃ gamayā támaḥ as "Send [him] down to nether darkness" but adharaṃ and tamaḥ are not in the same case. If tamaḥ here is a noun, and the verb is √gam 'to go' then (as in the Life of Brian) the verbs of motion take the accusative: tamam. Here tamaḥ is a nominative singular. "Nether Darkness would translate" adharam tamam, but not adharam tamaḥ. If we take the pada as a whole:

yó asmā́m̐ abhidā́sati ádharaṃ gamayā támaḥ  

He who is dark (yo tamaḥ), treating us as inferior (asmā́m̐ abhidā́sati ádharaṃ) should be made to go (gamayā). 

Thus again the relationship to Hell is less than tenuous. And this sums up all of the evidence for Hell in the Ṛgveda. We can be fairly certain that the Ṛgveda has no conception of a afterlife realm of punishment that corresponds to Hell. We need to look more closely at what kind of afterlife the Ṛgveda does know: i.e. the pitṛloka, discovered by Yama, and the devaloka.

Pitṛloka & Devaloka. 

Initially the pitṛloka and the devaloka were more or less on the same level even when they were distinguished. It seems that the devaloka was not initially thought of as an afterlife destination. Humans were not reborn as gods. This may be a Buddhist innovation. Cuevas notes that the pitṛloka came to be demoted in height and status, becoming associated with the antarīkṣa (for the significance of vertical spatial metaphors see Metaphors and Materialism). By the time of the early Upaniṣads the pitṛloka is associated with "the moon, darkness, sacrificial activity and rebirth" whereas the devaloka is associated with "the sun, light, knowledge and immortality" (Cuevas 272). This is particularly seen in the passages regarding the five fire knowledge (pañcāgnividyā) that describe a number of after-life paths and destinations. By contrast going to the devas becomes the first step on a journey out of saṃsāra that culminates in going to Brahman.

click to embiggen

For a culture which sees the performance of ritual as determining one's afterlife destination there appears to be little or no need for a concept of Hell. The Vedas hint at a bad destination for enemies of the Brahmins, but it's not until the world is ethicised that an afterlife which punishes wrong doing is needed. And by punishment I mean something beyond the withholding of paradise from the inept ritualist. How and when Hell becomes part of Vedic cosmology and eschatology is not entirely clear and I have only a few scattered references to work from. There's not much to indicate that one could return to the human realm having been in Hell.

If we do not see hell as an afterlife destination in the Ṛgveda, then the obvious question is when do we see it in Indian literature? This is not a question I can answer yet.

We can now come back to the question of the meaning of niraya/naraka. In seeking to understand the word, such etymology as there is has sought a connection to Hell. However as we see originally Yama's loka was original not an underworld place of suffering at all. Indeed it was a place in the sky where one experienced (presumably joyful) reuniting with one's ancestors. It became the destination for men (nṛ) who performed the correct rituals. As such a name which was a collective adjective based on nṛ i.e. naraka or nāraka would make sense. We could then explain niraya as a dialectical variation. Against this explanation is the lack of any parallels. All the words starting with nir- in PED are derived from the suffix nis-. This fact suggests that niraya and naraka are two unrelated words. My hunch, however, is that they are related.


Hell in Zoroastrianism

Based on ideas first put forward by Michael Witzel I've speculated that the impetus to escape from a once happy rebirth eschatology was also influenced by Iranian (i.e. Zoroastrian) ideas. The vector for these ideas being an influx of Iranian tribes, including the Śākyas, whose culture gave rise to śrāmaṇa religions. Since we do not see Hell in Vedic, it's possible that the idea of Hell came from this same source. In order for this to be true the Vedic speaking people's had to leave Iran before the advent of Zoroastrianism which is difficult to date, but generally placed at about 1000 BCE.

However Hell is barely mentioned in the oldest Zoroastrian scriptures. As the Encyclopedia Iranica (EI) says:

Hell is not explicitly mentioned in the Gathas. There are only allusions made to it, if not in Yasna 31.20, at least in Yasna 46.11, where it is said that the soul and the daēnā of the wicked arriving at the Činwad Bridge (Av. činuuatō pərətu) will be guests in the “house of falsehood” (Av. drūjō dəmānā-), and in Yasna 51.13.The word hell, literally bad existence (Av. daožaŋᵛha-, Pahl. dušox, Pers. duzaḵ) only occurs in the later Avesta. 

When Hell is mentioned it is a place of torture in recompense for bad thoughts, words and deeds. Unfortunately for my conjecture the time-line is not yet clear, but the indications are that Hell developed at around the same time in Iran as it did in India.

The Iranian twin of Yama is a mythic King called Jamšid aka Yima. He is a culture hero, a king who ruled the world in a Golden Age. "Yima is said to be like the sun to look at among men (huuarə.darəsō maṧiiānąm; Yasna 9.4) and his life is immortal and “sun-filled” (xᵛanuuaṇt, Yasna 9.1)," (EI). As with Yama, the Iranian Yima is the son of a solar figure (Skt. Vivasvant, Av. Vīuuaŋᵛhant, “the one who shines far and wide”, and in this aspect he "made the world immortal",. How Yima bequeathed immortality and why humans are no longer immortal are not told in older texts and several versions of the story exist in later texts. Stories which connect Yima to Hell come rather late in the piece.

There are three references in the narratives above to Yima going to Hell: for his sins, in order to close the door to Hell so that death would be kept out, and in order to bring the paymān(ag) [right measure] out of Hell. (EI)

Paymān "is characteristic of Zoroastrian ethics and is discussed at length in the Middle Persian texts" (EI). So while the connection to Hell is not entirely clear, Yama is a figure common to both Indian and Iranian myth, giving him considerable antiquity. And in both mythic systems he is associated with extending the lives of humans: Yama through rebirth, and Yima through immortality. However both meet with a downfall: Yama becomes the ruler of Hell, and Yima sins and is sent to Hell as punishment. That there should be commonality in the earlier versions of the myth is not unexpected since we already know of parallels between the Ṛgveda and the Avesta, but that that developments of the myth should continue to follow parallel paths is intriguing. 

Yama in Buddhist Texts

Yama is mentioned in only a few texts. In an earlier essay on the history of Kamma I wrote:

Consider the Devadūta Sutta (MN 130, M iii.178) which explains how after death a being who has behaved badly might be reborn in hell (niraya); there they will be seized by the guardians of hell (nirayapālā), dragged before King Yama and cross-examined about their evil conduct of body, speech and mind. Unable to account for themselves, they are then condemned to horrific tortures which are graphically described. It is emphasised that "as long as that evil action is not destroyed, he does not die" (na ca tāva kālaṅkaroti yāva na taṃ pāpakammaṃ byantīhoti).

This is one of the most important occurrences. Another slightly different version of the story is found at AN 3.36, showing once again that the Pāḷi Canon is an incompletely merged anthology drawing on multiple retellings of the source material.

At SN 1.49 those who are stingy or hinder alms gathering are said to be reborn in Hell, as an animal or in Yama's realm (Nirayaṃ tiracchānayoniṃ, yamalokaṃ upapajjare), which is interesting. Recall that the departed (preta) where originally on their way to Yama's realm (yamaloka) to live with their ancestors (pitṛ), but the pretas became a kind of being in purgatory. Thus yamaloka here, as distinguished from niraya, might refer to the pretas. Bodhi also concludes this, but we don't know. Buddhaghosa's commentary is silent at this point.

At SN 1.33 we find Yama mentioned in an udāna uttered by a devatā:

Yo dhammaladdhassa dadāti dānaṃ,
Uṭṭhānavīriyādhigatassa jantu;
Atikkamma so vetaraṇiṃ yamassa,
Dibbāni ṭhānāni upeti macco ti.

The one who gives the gift of the received Dharma
Obtained though exertion and devotion
He crosses over Yama's river Vetaraṇī
That mortal one approaches the heavenly regions. 

The river Vetaraṇī is mentioned only one other time in the Suttanipata, Sn 674. It appears to be a river in Hell itself that the evil-doers fall into, and thus not much like the Styx, contra Bodhi in his translations notes on SN (2000: 364-5 n.67).

Finally Yama receives a passing mention: DN 13 (i.246) in a list of Vedic devas. This is not much to go on. Yama is a rāja, who rules over Hell, questions the souls of the dead, and has some messengers. This is broadly speaking the Vedic Yama.

Thus despite his later prominence in Buddhist myth, Yama is actually quite a marginal figure in the Nikāyas. Anālayo notes in his study of the Majjhimanikāya that Yama's role in the Buddhist texts has been reduced from active to passive so as to avoid a conflict with the doctrine of karma (2011: 748 n.303). Most of the later stories and images seem to depend on the Devadūta Sutta. This text was translated into Chinese five separate times (EA 32.4, T 86, T 42, T 43, MA 64) and there are a number of partial parallels ( T 24*, T 25*, T 212.9*, T 741*, DA 30*). The variations are discussed by Anālayo (2011: 747-53). A translation of MĀ 64 can be found in Bingenheimer, Anālayo and Bucknell (2013: 407).

Māra

It's worth saying a few words about Māra here, though he deserves his own essay. In contrast to Yama who presides over Naraka, Māra is an unrelated figure apparently emerging from non-Vedic tradition, along with Yakṣas and a goddess of good fortune known as Sirī (Skt. Śrī). The name derives from the causative form of the verb √mṛ 'die'; present verb mṛyate, causative mārayati. Thus māra is literally 'causing to die', or 'killing'. He's also known by the epithet pāpima 'evil one'. Māra is sometimes said to preside or rule over saṃsāra, and one of his biggest concerns is that people will escape saṃsāra. In this sense he stands for the repeated deaths that one must undergo in saṃsāra and all the associated grief. Māra uses the weapons of saṃsāra (desire, aversion, confusion), often in personified forms (his daughters represent desire and his army aversion). However though it might seem obvious to link Māra with Yama, there seems to be no connection between the two in practice. Yama is not evil in the way that Māra is. However Yama is sometimes written of as a personification of death, where he is called Mṛtyu 'Death'.

~~oOo~~

Bibliography

Ṛgveda texts taken from the metrically restored text by Karen Thomson and Jonathan Slocum. 

Anālayo (2011) A Comparative Study of the Majjhima-nikāya. Volume 2 (Studies of Discourses 91 to 152, Conclusion, Abbreviations, References, Appendix). Dharma Drum Publishing Corporation.

Bingenheimer, Marcus. Anālayo & Bucknell, Roderick S. (eds) (2013) The Madhyama Āgama (Middle Length Discourses) Vol. 1. (Taishō Vol. 2, no. 26). Bukkyo Dendo Kyokai America.

Cuevas, Bryan Jaré. (1996) 'Predecessors and Prototypes: Towards a Conceptual History of the Buddhist Antarabhava.' Numen 43(3): 263-302.

Doniger O'Flaherty, Wendy. (1981) The Rig Veda: An Anthology. Penguin

Jurewicz, Joanna. (2008) 'Rebirth eschatology in the Rgveda. In search for roots of transmigration.'  Indologica Taurinensia: The Journal of the International Association of Sanskrit Studies. 34: 183-

Skjaervo, Prods Oktor. (2012) 'Jamšid i. Myth of Jamšid.' Encyclopaedia Iranica. Vol. XIV, Fasc. 5, pp. 501-522. http://www.iranicaonline.org/articles/jamsid-i

Stausberg, Michael (2000). “Hell in Zoroastrian History.” Numen 56: 217-253. http://michaelstausberg.net/old_site/Texts/Stausberg%20Hell%20Numen%2056.pdf

Sarvāstivāda and the Chinese Sarva Sūtra

The Sarvāstivāda School ostensibly forms part of the background against which the Prajñāpāramitā literature developed. Indeed many see the early Prajñāpāramitā texts as taking the time to refute certain Sarvāstivādin ideas. Nāgārjuna appears to be in a conversation with Sarvāstivādins. This suggestion of a conflict between Prajñāpāramitā and Sarvāstivāda is an important one for understanding the Prajñāpāramitāhṛdaya. This is because, despite the received tradition about when it was composed, and despite various late sectarian commentaries, the Hṛdaya is largely made up of chunks of texts from the Pañcaviṃśatisāhasrikā Prajñāpāramitā Sūtra and thus has more in common with the outlook of early rather than late Prajñāpāramitā thought.

The Sarvāstivādins are also intrinsically interesting in that they seem to have been the dominant Buddhist school in India for a considerable length of time. There seem to be two main reasons we know less about them than we do about the Theravādins. Firstly their texts mainly only survive in Chinese translation and until relatively recently Westerners have not been very interested in these texts (possibly influenced by the linguistic demands of having to know at least Sanskrit, Pali and Classical Chinese in order to study the literature); or in fact interested only where Sanskrit "originals" are known to exist (Cf. Which Mahāyāna Texts?). Secondly, the focus has long been on Theravāda as representative of early Buddhism. But if any Buddhist school "represents" early Indian Buddhism it is the Sarvāstivāda.

A third reason we might not think of them as representative is that early Buddhist was eventually eclipsed first by Mahāyāna Buddhism (though very often on the basis of the Sarvāstivādin Vinaya) and then Tantrism. So that even where monks follow a rule based on the (Mūla)Sarvāstivāda Vinaya and study Abhidharma through commentaries on the Sarvāstivāda Abhidharma (cf. Nietupski 2009), they don't identify themselves as Sarvāstivādins.

In her discussion of the Sarvāstivādin school Professor Collett Cox makes an interesting point about the name of the school. She says:

As encapsulated in the name 'Sarvāstivādin,' the Sarvāstivādins are characterized as maintaining that "everything exists" (sarvam asti). However, the simplicity of this ontological assertion contains the seeds of doctrinal divergence because the referent of the term 'everything' and the manner in which this "everything" is considered to "exist" must be specified. Certain early Abhidharma texts identify the term 'everything' in the declaration that "everything exists" as referring to the twelve sense spheres including the six sense organs and their corresponding object-fields. So also the *Mahāvibhāṣā, in a discussion of the twelve sense spheres, cites a sūtra passage in which the term everything' is defined by the Buddha as "precisely the twelve sense spheres from the form sense sphere (rūpāyatana) up to and including the factor sense sphere (dharmāyatana).

I want to make a short digression to point out that Buddhists very often use the word asti not as a verb (3rd person singular indicative) 'it exists' but as an action noun meaning 'existing, being'. Thus we don't need to syntactically derive sarvāsti in Sarvāstivāda from the sentence sarvam asti 'it all exists', with an awkward silence over why sarvam loses it's case ending; but can treat the word as a karmadhāraya compound sarva-asti 'everything existing'.

The sūtra passage referred to in the Mahāvibhāṣā (Cox's asterisk means the Sanskrit title is reconstructed from Chinese) probably corresponds to the Pāli Sabba Sutta (SN 35.23). This sutta has a Chinese parallel from one of the two Saṃyuktāgama translations (T 99 #319, 91a24-b03) this one being attributed to the Mūlasarvāstivāda School. "T 99 is widely considered to have been translated in the period 435-443 CE from a Sanskrit Saṃyuktāgama brought to China from Sri Lanka." (Bucknell 2011). Since I see this as a seminal text, and as it seems directly related to the Sarvāstivāda School, I will present my rough translation of the Taisho Tripiṭaka version.

*Sarva Sūtra (Title not given in Chinese).

如是我聞。一時。佛住舍衛國祇樹給孤獨園。
Thus have I heard, one time, the Buddha was staying in Śravāsti (舍衛國) in the Jeta Grove of Anathapindada.

時。有生聞婆羅門往詣佛所。共相問訊。問訊已。退坐一面。
Then, Jānussoṇī, the brahmin, approached the Buddha, exchanged greetings, and retreated to one side.

白佛言。瞿曇。所謂一切者。云何名一切。　
He said to the Buddha, "Gautama, 'all' (一切 yīqiè = Skt. sarvaṃ) is said, what is 'all'?"

佛告婆羅門。一切者。謂十二入處。眼色．耳聲．鼻香．舌味．身觸．意法。是名一切。
The buddha answered the Brahmin, "'All' is namely the 12 āyatanas: eye & form (colour), ear & sound, nose & smell, tongue & taste, body & touch, mind & dharmas. This is called 'all'.

若復說言此非一切。沙門瞿曇所說一切。
Even if the words were said 'this is not all', Śrāmana Gauatama says this is 'all'.

我今捨。別立餘一切者。
I will explain the rejection. A different 'all' does not stand.

彼但有言說。問已不知。增其疑惑。所以者何。
He who speaks this. Asked, does not know. His doubts increase. So who and where?

非其境界故。
Because this is not the proper domain (Skt. viṣaya 境界 jìngjiè )."

時。生聞婆羅門聞佛所說。歡喜隨喜奉行。
Then, Jānussoṇī. the brahmin having heard the Buddha was delighted and rejoiced.

~o~

Compare the Pāli Text:

Sabba Sutta (S iv.15)

At Sāvatthi: I will teach you the whole, monks. Listen to this. What, monks, is the whole? The eye and forms, the ear and sounds, the nose and smells, the tongue and tastes, the body and touches, the mind and mental phenomena: this, monks, is called ‘the whole’. If anyone says ‘I reject this whole, I will declare another whole’ that would just be hot air. Questioned about it, they wouldn't be able to explain, and would become exasperated. Why is this? Because that, monks, is in the wrong location (avisaya).

~o~

Clearly there is very little different between these two, except in the nidāna or setting. But the existence of these texts begs the question why the Buddha might have wanted to define 'all', 'the whole', or 'everything' (all translations of Skt sarvam, Pāli sabbaṃ)? In fact the idiom is one that derives from Vedic literature. For example in Ṛgveda (RV 8.58.2):

éka evā́gnír bahudhā́ sámiddha
ékaḥ sū́ryo víśvam ánu prábhūtaḥ
ékaivóṣā́ḥ sárvam idáṃ ví bhāti
ékaṃ vā́ idáṃ ví babhūva sárvam

Only one fire kindles many times.
One sun is all penetrating.
Dawns as one, shine on all this.
From this one, unfolds the whole.

I've cited this verse previously in relation to the Fire Sermon. Another key text for understanding the idiom is Bṛhadāraṇyaka Upaniṣad, for example:

brahma vā idam agra āsīt | tad ātmānam evāvet | ahaṃ brahmāsmīti | tasmāt tat sarvam abhavat | (BU 1.4.10)

At first there was only Brahman, and it only knew itself "I am Brahman". From that it became everything (sarvam).

So, sarvam means 'the created world'. Sometimes the idiom is idam sarvam meaning "this whole [world]". For the Ṛgveda it seems to mean Creation in a fairly literal sense. In BU sarvam begins to take on a more mystical sense, it is the manifest aspect of Brahman. And it is through identifying oneself with the world, i.e. with Brahman, that one attains (re)union with Brahman. And this may be why Buddhists called the meditation in which one identifies with all the beings in the world, brahmavihāra 'the dwelling of Brahmā (or Brahman)'.

The Sarva Sūtra almost certainly reacting to the Vedic usage, sets all this aside. The whole from this point of view is the six senses and their objects. In another Pāḷi text, the Dvayam Sutta (SN 35.92) this same formula is referred to as "the pair" (dvayam). But it is not the objects or the senses per se that concern the sutta. The alternate term dvayam reminds us that the two together form the basis for the arising of sense-consciousness. The beginning and end of the interest of the text is the sense objects and sense organs. All experience begins with these, there is no other source of experience. It is through observing experience that we are liberated. Any speculation which lies outside of these, particularly any metaphysical speculation about the nature of the objects of the senses, is out of bounds (avisaya).

Cox (1995: 134 ff.) goes on to outline some of the arguments between Saṅghabhadra and Vasubandhu about the interpretation of this sūtra and the implications each drew from it. The story is far from simple. "The very sutra passage that defines the term 'everything' as the twelve sense spheres is cited by both Vasubandhu and Saṅghabhadra as scriptural justification for their divergent ontological positions." (1995: 134).

But one thing that the text does not say, in either Pāḷi or Chinese, is that the āyatanas "exist". They represent the whole of the Buddhist's field of interest and, citing the Kaccānagotta Sutta, we believe the early Buddhists were not interested in the issue of whether or not they exist. However since the phrase imasmin sati idam hoti "this existing, that exists" is so central to Buddhist we must allow that the sense object and sense organ cannot be non-existent if they are to act as conditions for the arising of viññāna and phassa etc. And this leads into a discussion about how the Sarvāstivāda got its name and its eponymous doctrine which will be covered in a forthcoming essay. 

~~oOo~~

Bibliography

Bastow, David. (1995) 'The First Argument for Sarvāstivāda.' Asian Philosophy. 5(2): 109-125. txt online.

Bucknell, Roderick S. (2011) 'The historical relationship between the two Chinese Saṃyuktāgama translations.' Chung-Hwa Buddhist Journal. 24: 35-70. pdf.

Cox, Collett. (1995) Disputed Dharmas Early Buddhist Theories on Existence: An Annotated Translation of the Section on Factors Dissociated from Thought from Sanghabhadra's Nyāyānusāra. Tokyo The International Institute for Buddhist Studies. pdf.

Nietupski, Paul K.  (2009) 'Guṇaprabha’s Vinayasūtra Corpus: Texts and Contexts.' JIATS, no. 5 (December 2009), 19 pp.

Is Karma responsible for Everything?

Recently a friend asked my opinion about a verse from Vāseṭṭha Sutta in the Sutta Nipāta. It was being cited as a proof text for the idea that karma (ours and other peoples) is the sole source of all our experience. The question of what karma is responsible for is one that seems to come up again and again. Partly because there are so many versions of what karma really is or means. The Sutta Nipāta is generally considered to have been composed quite early - the last two chapters are often said to be the oldest layer of the Pāli Canon. Saddhatissa's translation of the verse was quoted so let's start there, and then I'll work through the verse one phrase at a time:

The world exists because of causal actions,
all things are produced by causal actions
and all beings are governed and bound by causal actions.
they are fixed like the rolling wheel of a cart,
fixed by the pin of its axle shaft. (p.75)

The Pali is

1a | Kammunā vattati loko,
1b | kammunā vattati pajā;
2a | Kammanibandhanā sattā,
2b | rathassāṇīva yāyato. (Sn 654)

Now the Sutta Nipāta is notoriously difficult to translate due to the archaic language. A more literal translation would be:

"The world exists through actions, offspring exist through actions;
Beings are bound to actions, driven like a horse-chariot axle pin."

or perhaps -

"The world is moved by actions, people are moved by actions;
Beings are fettered by actions, driven like a horse-chariot axle pin."

"The world" (loka) is a metaphor for one's inner world of experience - it should be read this way in almost every case. It's an old Vedic metaphor which we share when we say things like - "he lives in a world of his own". I note that the text is a discussion with two Brahmins so it shouldn't be a big surprise to find Vedic overtones in the language. A 'loka' was originally an open space, like a clearing in the forest, in which one could see clearly. So phrase 1a (ie 1st line, 1st phrase) means that the world of experience is driven by kamma - crucially we keep experiencing vedanā because vedanā is the result of previous kamma. The Pāli texts are clear that vedanā is the result (vipaka) of kamma - i.e. only the broad outlines of one's experience, the pleasure and pain, are the result of kamma, not the specific details of what causes the experience. As we know unenlightened experience is dukkha, and can argue in this case that world (loka) and suffering (dukkha) are equivalents. In effect 1a is a less sophisticated way of saying greed, hatred and delusion keep saṃsara going.

The verb vattati (repeated in phrase 1b) can mean "to move, or go; to be, exist; to fare, to do" - i.e. it has the same broad reference of other Pāli verbs for "to be" like bhavati. Note the dynamic aspect that our English "to be" often lacks. Given the image at the end we might also have translated it: "is powered by": as in "experience is powered by kamma". This would also make sense. The Pali English Dictionary (PED) says that in this context it means "keeps up, goes on". Which is more or less the same thing. The idea is that kamma is what "drives" the process. In 2b the verb is yāyati which does mean "to drive". Recall also that the Buddha redefined kamma: cetanāhaṃ, bhikkhave, kammaṃ vadāmi - I say, monks, that intention (cetanā) is kamma (AN vi.63). In this light I'm not convinced that "causal action" is any more helpful as a translation than simply "action" - it might have been better to leave kamma untranslated.

The purpose of phrase 1b is to link rebirth to kamma. This may seem a strange point to make, but in a Vedic context it was important because early Brahminical versions of rebirth did not link it to kamma (see the first chapter of the Bṛhadāranyaka Upaniṣad for instance where is it known as 'redeath' and not linked to kamma, or to ātman!). Calling people "pajā", although a common usage, may well be a reference to Pajāpati (Sanskrit Prajāpati) the Vedic god of creation and father of all beings, i.e. it once again suggests a Vedic flavour in this passage.

In phrase 2a beings (sattā) are bound (bandhanā) by their kamma. That is, having acted you must live with the consequences. This is a distinction from other kamma theories which state that you can burn-up old kamma by experiencing suffering now. That idea is associated in the Pali Canon with the Jains who did severe austerities and self-torture, sometimes as a prelude to starving themselves to death in pursuit of liberation. For the Buddha there are ways (basically general spiritual practices) to lessen the impact of kamma-vipaka, but not to avoid it altogether. (I write about this in my article for the Journal of Buddhist Ethics). Bandhu was the Vedic word for the very important relationships between mirco- and macro-cosm, between earth and heaven. Understanding and manipulating bandhu was the central function of the Brahmin priests. So again this is drawing on Vedic idiom.

Phrase 2b finishes with a image that sums up the verse. The phrase is parsed as ratha assa āṇī va yāyato. It is driven (yāyato) only or like (iva or eva elided to va following a long vowel) a horse-chariot (rathassa) wheel pin (āṇī *). Saddhatissa sees this image as reinforcing the idea that we are bound (bandhanā) to the results of our actions and tries to bring this out in his translation. The past-participle yāyato means "driven" - it's an intensive for of yāti "to go, go on, proceed, to go away". PED gives as an example "yāyena yāyati to drive in a carriage". The image seems to relate 'the bond' to the driving in of a wheel pin - presumably these were wedge shaped and were hammered into place to hold the wheel on chariot. So there is a further sense that kamma is what keeps the wheels on the carriage, that it keeps the whole business of suffering going. The use of yāyato reinforces this as it suggests motive power - kamma being what keeps the wheel of saṃsāra turning. The horse chariot was powerfully associated with the Brahmins in the Vedic period, although I'm not sure if that would hold in the Buddha's time and place. The horse chariot was a war chariot, a symbol of royal power and of conquest; whereas the ox-cart was a more agricultural vehicle. So perhaps here also the Buddha is using Vedic imagery to suit his audience?

I think Saddhatissa has erred by introducing the word "all" into the translation - it makes the text sound too absolute. There is nothing here that could be translated as "all things". Clearly also there is nothing here to suggest that "all events" are, or that "everything" is, the result of kamma.

The idea that everything that happens is a result of kamma is a common enough wrong view to have a name: Pubbekata-hetu-ditthi (literally "the with-past-actions-as-cause view"). For a canonical discussion of this you could try the Devadaha Sutta (Majjhima Nikāya 101). Bhikkhu Thanissaro's version on Access to Insight comes with a useful introduction. A problem for contemporary Buddhists is that this version of karma - i.e. that everything that happens is the result of previous actions - is taught by Tibetan Buddhist teachers. I don't see any way of reconciling these views - they are mutually exclusive. But then my own take on kamma/karma and belief in general is pragmatic: if what you belief about karma makes you a better person (less greedy, angry, ignorant; more generous, loving, wise etc.) and helps you to see that experiences are impermanent, unsatisfying, and insubstantial; then I don't think it's so important what you believe. I've talked about this aspect of belief in a couple of earlier essays: Karma and Rebirth; Beliefs can be Heaven or Hell. For another interesting view on Karma look at Nagapriya's article for Tricycle Magazine: Donating the Future.

One final note on the possible presence of Vedic metaphors and terms. Firstly I may be overstating this connection. But even if I am not, then secondly we need only assume that the Buddha was using language which he knew would be familiar to his audience. Importantly we don't need to assume that the Buddha bought into the Vedic world view. It does suggest that the Buddha might have been familiar with some of the ideas of Brahmins, and there is plenty of other evidence to back this up. He may well have acquired this familiarity in religious debates which were a feature of life in the Buddha's time, rather through having studied the Vedas. Indeed the latter is unlikely if only because the Buddha was not a Brahmin, and therefore excluded from learning Sanskrit and studying the texts.


Note

* The word āṇi has an interesting history. It is attested in the oldest parts of the Ṛgveda (e.g. RV 5.43.8) but appears not to be an Indo-Aryan word, i.e. not from a language related to Vedic or Sanskrit or any of their variants and offshoots. Previously it was thought to be a loan word from the linguistically unrelated Dravidian family, now most commonly associated with South India and represented in the present by Telegu, Malayalam, and Tamil. However Michael Witzel (1998 p.18), has shown that this is unlikely, and in fact scholars of Dravidian languages saw it as a loan-word from Indo-Aryan! Witzel suggests that it is one of about 200-250 loan words from an archaic form of a Munda language. This suggests that Munda was the language originally spoken in the Kurukṣetra which came to be the heartland of the Vedic speaking peoples. Munda is a member of the Austro-Asiatic family of languages which includes Thai, Malay, Cambodian and Vietnamese. The Munda speaking people are often called "tribals" and live on the margins of Indian society, some maintaining a nomadic lifestyle. If any group could be considered aboriginal in India it is them. Their religions are animistic, and I want at some point to see if they share any of the animistic beliefs found in early Buddhism.

Reading

• Saddhatissa, H. 1985. The Sutta-Nipāta. Richmond, Surrey: Curson Press, 1994.
• Devadaha Sutta. MN 101. Access to Insight. Includes a useful introduction to the text by Bhikkhu Thanissaro.
• Nagapriya. 2009. 'Donating the Future.' Tricycle Magazine Website.
  
• Witzel, Michael. 1998. 'The Languages of Harappa : Early linguistic data and the Indus civilization.' in J. Kenoyer (ed.) Proceedings of the conference on the Indus civilization, Madison 1998. provisional. pdf