Like many students, I assume that the founder of Buddhism was himself a historical figure (how else does one explain the existence of such a founder-focused religion?), but also that very little is historical in the traditional biography, even as it is presented in the earliest sources. What follows is a paper about history, but about the history of the narratives attached to the Buddha, not about real events in the life of the Founder. My argument will be that much of the narrative material in the scriptures proper (the suttas and vinayas) derives from the Indo-European epic tradition, of which comparativists are gradually beginning to gain a clearer view.
Though it is the nitty-gritty of comparison that I most enjoy, I need to start with issues of method and approach. This paper arises from several others that I have published over the last five years which have explored the similarities between the adventures of Arjuna and those of Odysseus, the hero of Homer’s Odyssey. These similarities, I argue, are best explained by postulating a “proto-hero” whose adventures were told in a “proto-narrative”, from which both the Sanskrit and Greek epics descend. My terminology of course derives from comparative linguistics, and perhaps the narrative was indeed told in proto-Indo-European. However, the language, date, location and literary form of the proto-narrative have not been my main concern. The first question to ask is whether the similarities are sufficiently detailed and systematic to justify the idea of a proto-narrative. For convenience I speak of the proto-narrative in the singular, although of course oral tradition normally exists in variant versions (much as the proto-Indo-European language had variant dialects).
Although I have mentioned two heroes who come specifically from epic, I prefer the more general term “narrative”, having also been interested as a comparativist in the prose narratives of Roman pseudo-history and of the Irish Ulster cycle. I have great respect for the tradition of work on Indogermanische Dichtersprache, recently represented by the large tome by Watkins, but there is also a great deal to be done simply on narrative motifs and structures, abstracted from the form in which they are expressed.
Nobody who works on Indo-European cultural comparativism can nowadays avoid reference to Dumézil and his notion of the three functions. The functions will indeed force themselves on our attention here, but only towards the end, and I defer introducing them until that point.
Two years ago I had no special interest in the Buddha’s biography. The possibilities I glimpse for Mahābhārata-Homer comparison seem rich enough to occupy me for years to come, and it is perhaps stupid to take up a new research topic (albeit a related one) while another, no less exciting, remains uncompleted. Anyway, here is what happened. I was drafting quite a circumscribed paper dealing with varŠas, colours and functions, asking whether Indo-European colour symbolism was based on three colours or four, and I noted that in his discussion of the Licchavi nobles with their various colours Dumézil cited the work of Bareau. Following this up, and scanning the contents page of one of Bareau’s volumes, I noticed a grouping of themes familiar from my Mahābhārata-Homer comparisons. More precisely, I was reminded of the comparison that linked Odysseus’ journey from Calypso’s island to Scheria with Arjuna’s journey from forest exile to Indra’s heaven. The temptation proved irresistible to see how far the new comparison could be carried, and the present paper condenses some of the results.
This anecdotal account serves several purposes. Firstly, I need to make clear that this paper is not by a seasoned Buddhologist. What I try to practise is Indo-European cultural comparativism, hoping thereby to contribute insights of use to those whose interests are more regionally or topically specialised.
Secondly, I need to explain the restricted domain of epic that I address. Reflexes of the Indo-European proto-narrative may well exist in several branches of Indo-European but the branches of which I am least ignorant are the ancient Greek and the Sanskrit. Of these I concentrate here on the Greek, focusing on the comparison between Odysseus and the Buddha. I thus omit one side of what should be a triangular comparison. If Odysseus is indeed cognate both with Arjuna and with the non-historical aspects of the Buddha, then to complete the triangle we need a Buddha-Arjuna comparison. But that remains for the future. Each two-way comparison implies something about the proto-narrative, so one glimpses the potential complexity of the topic.
Thirdly, I need to be clear about aims. I have had to be selective. What I have tried to do is to pick out comparisons or arguments that are not only brief and persuasive but also present interesting problems of method. The central question is the viability of the scripture-epic comparison.
Finally, Bareau’s volume was not only my starting point, but it still sets some of my parameters. As his title states, Bareau limits himself to the biography as presented in the early Sutrapi˜akas and Vinayapi˜akas, and I too ignore later sources. In the Hindu case proto-narrative material often bypasses the earliest sources to surface in later ones, but I leave open whether that applies also in Buddhist tradition. Moreover, as Bareau’s subtitle states, he also limits himself to the stretch of biography running from the Quest for Enlightenment to the Conversion of ®āriputra and Maudgalyāyana, and for reasons of space I shall not even be able to cover the whole of that stretch – I barely allude to the conversion of Yaśas and Kāśyapa, and I stop short of the final journey to Rājagŗha. Even within this stretch the paper offers only a preliminary and compressed treatment.
Bareau aligns and translates not only the relevant Pali scriptures but also the Chinese versions as used by the Mahāśāsika and Dharmaguptaka sects, and on all of this I gladly draw. On the other hand, his commentary, valuable though it is, is essentially a quest for the historical Buddha, and it is little concerned with the origin of the narrative elements identified as non-historical, many of which it ascribes to the inventiveness of the tour guides serving Buddhist pilgrims. Its aims and conclusions differ so much from mine that I shall seldom allude to it.
I conclude this introduction with certain issues of terminology and method. I shall refer to the Buddha throughout, although strictly speaking the title is only gained at the moment of Enlightenment. When I need a term to refer either synchronically to the Buddha and/or Odysseus, or diachronically to the proto-figure lying behind both of them, I talk of “the protagonist”. The crucial question of method is of course what constitutes a significant similarity, one sufficient to justify the assumption of a historical common origin. There will always be an element of subjective judgement here – no mathematical equation can determine the relative weight of similarities and differences. I would like to think that experience in this field of study is helpful, but cannot be sure even of that. The only cardinal rule is that convincing arguments usually depend on sets of similarities rather than individual ones.
Overview. We must start with a summary of the two stretches of narrative in question.
The Buddha leaves the luxurious life of a prince in Kapilavastu, and travels south-east to seek spiritual perfection. He studies successively with two ascetics, but is dissatisfied with their teaching. Arriving at Uruvilvā near Bodh Gayā he undertakes extreme austerities for six years. When he is near death the gods intervene and he abandons this method. Thereafter he achieves enlightenment under the Bodhi tree, and gradually resumes human contacts. After some hesitation he moves to Benares, 500 km to the west, and begins his teaching career by converting five monks in the Deer Park close to the city. Further conversions follow before he moves back to Uruvilvā and, after various miracles, converts Kāśyapa plus his brothers and the disciples of all three. Then he moves a little north-east, to Rājagŗha, where he is welcomed by King Bimbisāra.
The Odyssey starts with its hero effectively imprisoned on Calypso’s island. Almost twenty years earlier he had left his wife Penelope in Ithaca. Ten years earlier, after the fall of Troy, he had set off for home, but his return had been much delayed. On Athena’s initiative, Zeus starts the hero moving again, and Odysseus sets off alone on a raft, aiming for Scheria, land of the Phaeacians. After an easy start, he is spotted by the hostile god Poseidon, who raises a storm to wreck the raft. Three days later, naked, battered and exhausted, the hero staggers ashore, helped by a kindly river god. He shelters in a thicket on a hillock and enjoys a healing sleep.
Meanwhile Athena, who intervened to help him during the storm, arranges for the nubile princess Nausicaa to drive in her mule cart to the laundry place near where the hero is sleeping. After washing their clothes, she and her maidens dance and play, until their shout wakes Odysseus. Hungry and still naked, he emerges from the thicket and begs help. Fed, clothed and guided by Nausicaa, he travels to the palace of her father Alcinous. The king treats him well and sends him back to Ithaca, where he kills his wife’s suitors and resumes his family life and throne.
The main focus here is on the stretch of story running from the storm that wrecks the raft up to the hero’s departure from Scheria. But we shall also have to look back to the encounters immediately preceding that with Calypso, namely those with Circe and with the various female monsters around the Straits – the Sirens, Scylla and Charybdis. In terms of length we are dealing with about four books’ worth, out of the two sets of 24 books that make up the Homeric corpus, so it cannot be objected that my comparisons are plucked from any old place in a long and varied narrative.
Global similarities. Even at this preliminary level certain similarities exist between Greek and Buddhist stories. The protagonist is a prince who leaves wife and infant son at home and sets off to travel, sometimes alone, sometimes in company. At one point he undergoes a lone physical ordeal away from human habitation. With divine help he survives, and travels to a capital where he receives hospitality and gifts from a king.
These similarities must of course be weighed against the numerous differences. For instance, the Buddha’s early life lacks any parallel for the Trojan War, and unlike the Greek hero he never travels by sea. More fundamentally, the whole orientation is different. The Buddha, founder of a religion, definitively renounces a throne, family life and sex when he leaves Kapilavastu. At his Enlightenment he defers full entry to NirvāŠa until his community is well established, and his main relationship to others is that of a teacher who converts them to the condition of Buddhist layman or monk. The aims of Odysseus, an epic hero, are mundane. During the period that concerns us his main aim is to return to wife and throne. He does not set out to be a spiritual leader and has scarcely ever been regarded as such.
Pervasive though it is, this difference in orientation does not invalidate the comparison. A priori those who developed the Buddha’s biography could well have drawn on heroic biographies for themes and motifs which they then spiritualised. If Odysseus mobilises the Phaeacians to aid his return, the Buddha could mobilise people to develop his new sect or religion. If Odysseus was a king, the Buddha could reject his royal birth and his potential to become acakravartin, and instead set in motion the dharmacakra, the wheel of the law.
But all this is rather general, and the argument must turn on details. For a systematic approach one has two options – to work through Bareau looking for Greek parallels, or conversely to work through Homer and look for Buddhist parallels. Here, trying to combine brevity and cogency, I sacrifice consistency and mix the two procedures. Incidentally, we shall often find that a single theme in the Greek has more than one parallel in the Buddhist tradition, and occasionally vice versa. I start with the Buddha.
A pleasant spot. When he leaves his teachers, the Buddha wanders by himself seeking a place of perfect peace. He finds what he wants near Uruvilvā. It is “a charming spot”, with a pretty but dense mountain grove, a rapidly flowing river (the Nirañjanā), and soft grass. It is here that he takes up his position under the Bodhi tree.
In the whole of Homer, the only comparable landscape I can think of is the one where Odysseus finds himself when he reaches the shore of Scheria. It is described as “an excellent place”. After landing there, the hero climbs a slope to settle himself in a dense thicket. The nearby river in which Nausicaa’s party wash their clothes has abundant clear water and is lined with grassy meadows. Altogether, the girls’ picnic spot is presented as idyllic.
In short we have here seven discrete similarities: solo arrival of protagonist; general attractiveness; elevation; dense thicket; place to stay; clear river; grass. Of course there are also differences. For instance the Buddha’s spot is chosen voluntarily and for long-term spiritual ends; Odysseus has much less choice and merely wants to survive the night. The Buddha notes nearby villages where he can find food, while Homer mentions no such settlements. But the initial set of seven similarities can be expanded if we look more closely at the Greek.
When he lands after his long ordeal Odysseus collapses. On recovery he sinks down in the reeds to consider his options. If he stays in the river bed he risks dying of cold; if he climbs to the thicket he risks wild beasts. Choosing the second course of action, he takes shelter under a thorn tree and olive tree which grow from a single spot and intertwine so densely as to provide excellent shelter. Under them, to his joy, is a pile of fallen leaves, which he heaps together over himself so as to form a bed. He is like someone on a remote farm who buries a firebrand beneath the ashes so as to keep a fire alight overnight. Athena gives him a good night of healing sleep.
We can now return to Uruvilvā.
Dead leaves and joy. Reflecting on the lonely life of a forest hermit, the Buddha resolves to overcome the fears produced by forest noises, and among these noises he lists the wind stirring dead leaves fallen on the ground. The parallel Chinese text ignores the leaves but mentions no less than eight times the joy experienced by the Buddha when he reflects on his fitness for the life of a hermit. In the Greek Odysseus simply rejoices at the sight of the dead leaves.
Comparison involving fire. At the same period of his life the Buddha formulates three comparisons involving a piece of wood and its suitability for lighting a fire. A wet stick picked from the water by a passer-by will not make a successful firestick, and will resemble an imperfect religious seeker. It is much the same if someone else has already thrown the wet stick out of the water. But a really dry stick picked up in the same way will do the job, and will resemble a well-prepared Seeker.
Compare the firebrand simile in Homer. i) Both traditions present the fire not in the main story but as an element within a simile or comparison. ii) Both concern the possibility of lighting or relighting a fire. iii) In both the protagonist is compared to a piece of wood (firestick, firebrand). iv) Both involve exit from water – Odysseus himself has just emerged from the sea).
Shelter from elements and beasts. After his enlightenment the Buddha stays on in Uruvilvā, no doubt in the same thicket, under a sequence of different trees. While he is under a mucalinda tree, a storm-cloud causes cold, windy, rainy weather for a week. Mucalinda, king of the Nāgas, winds himself seven times round the cross-legged Buddha, and extends his hood over him, thus protecting him against heat, cold and wind, against reptiles and against the teeth (of wild beasts).
In the Greek the two trees give Odysseus a shelter so dense that neither sun, wind nor rain can penetrate them. Although Mucalinda, the shelter-provider, is usually thought of as a nāga, as animal not vegetable, the same name is also given to the tree. Also, although Odysseus does not face a storm while he is in the thicket, he has of course only recently escaped from one; and although he encounters no wild animals, he has envisaged them as a danger if he passes the night in the thicket.
Emergence from river. The Nirañjanā at Uruvilvā has already been compared to the nameless river in the Odyssey, but we need a fuller account of the latter.
Desperate to land, Odysseus swims along parallel to the cliffs of Scheria. When at last he finds the attractive estuary, he prays to the River God, as a supplicant trying to escape the sea. The God responds at once by checking his stream and bringing the hero safely to land. The hero entrusts to the river the divine buoyancy aid lent him at sea by the goddess Ino, and the river returns it to her. It is to this same river that Nausicaa comes on her laundry expedition, to its stone-lined tanks built for washing clothes. The girls themselves bathe in the river and after contact is made, so does the hero. When he emerges from it, he is rendered radiant by Athena. He then goes to sit on the sea-shore and is admired by Nausicaa.
In this passage the hero twice emerges from the river: (i) when he exits at the estuary, before recovering strength and climbing to the thicket, and (ii) after his deliberate bath. Buddhist tradition as it were conflates the two occasions: when the Buddha gives up his extreme asceticism, he recovers his strength, bathes in the Nirañjanā and climbs its bank to sit under the Bodhi tree. Here the bathe is deliberate, but it precedes ascent to the grove.
Some further comparisons bearing on the river (in fact the ones that first caught my attention) involve the second stay in Uruvilvā, when the Buddha performs a sequence of miracles to impress Kāśyapa.
Helpful river. After a meal with Kāśyapa the Buddha returns to his grove. He needs water to wash, and spontaneously the river moves its channel closer to the Buddha, so that he can use it more easily. This obliging behaviour echoes the Greek river which checks its current to help Odysseus to land – this is of course the same river as the one in which the hero later washes.
In both traditions, the theme of the helpful river is followed promptly by the theme of washing clothes.
Washing clothes. The Buddha finds a dirty cloth and decides to wash it for use as a robe. Indra digs him a pond, brings a stone on which to do the washing, and another stone on which to dry it. The Buddha then strips and bathes in the pond, and in order to exit he grips the branch offered to him by a tree, which obligingly tilts over towards him for that purpose. The three Buddhist texts vary in details and differ considerably from the Greek, e.g. in having a pond not a river, and in having the protagonist, not a visitor like Nausicaa, as clothes-washer. However the similarities are as follows. (i) Washing and drying of clothes on stone surfaces: the Phaeacian laundry is spread out on sea-washed pebbles. (ii) The involvement of a deity: Athena effectively sends Nausicaa to the river, Indra desires the Buddha to use the pond he has just made. (iii) The branch motif: the naked Buddha gripping a branch to emerge from the pond parallels Odysseus gripping and breaking off a branch to hide his genitals as he emerges naked from the thicket.
Provisioning of protagonist
According to Bareau, the Mucalinda episode is immediately followed by the story of the merchants Trapuùa and Bhallika – the first named characters to feed the Buddha after his Enlightenment, and the first to become his lay followers. Compare Nausicaa, the first person to feed Odysseus after his ordeals and the first Phaeacian to give him help and support. Here is a conflated summary of the Buddhist story.
The merchant pair, differentiated only by name, are part of a large caravan. The unnamed deity of the Bodhi tree (probably male) is a friend or relative of the pair, and tells them that they are near the Buddha and that, for their own spiritual benefit, they should go and take him food and drink. When they do, the Buddha has first to acquire a suitable bowl for their offerings. He then allows them to take the two refuges.
Now back to the Greek.
After putting Odysseus to sleep, Athena goes to the Phaeacian capital and enters Nausicaa’s bed-chamber like a breath of air. She takes the form of Dymas’ daughter, Nausicaa’s intimate friend, and tells the princess to undertake the laundry expedition. In the morning Nausicaa recalls her dream and sets off for the river mouth.
After the laundry and picnic the girls play a ball game. A catch is missed, the ball is lost in the water, and their shout of dismay wakes the hero. Odysseus emerges and accosts Nausicaa, who reassures her frightened companions and issues her instructions. Accordingly, after Odysseus has bathed, they serve him his first meal – no doubt picnic leftovers – which he eats ravenously.
In both stories the protagonist is for the moment static and the provisioner undertakes the journey that leads to the meeting. However, in both cases the provisioner is set in motion by a deity who has previously been with the protagonist. In other words, the unnamed tree deity, who leaves the Buddha in order to mobilise the merchants, parallels Athena, who leaves Odysseus in order to mobilise Nausicaa. The Indian deity is an intimate friend of the merchants, and although Athena is not an intimate friend of Nausicaa, she takes the form of Dymas’ daughter, who is just that.
When Odysseus emerges “to go among the maidens” the servants scatter in fear of the naked stranger. Only the princess stands her ground and later reassures her companions. In Mahāśāsaka tradition at the start of the episode the Buddha ends his samādhi and “goes to walk among men”. Then, to gain the attention of the caravaneers, the godling halts their cart-oxen and the party scatters in fear. The godling issues his instructions from the sky and reassures them. Note the common phenomenon that Greek females parallel Buddhist males – I think that Buddhist tradition has masculinised as well as spiritualised the epic tradition.
The Buddha is initially reluctant to eat for lack of a begging bowl, but then he acquires four stone bowls from the Lokapālas and, to avoid favouring one out of the four, compresses all four bowls into one. We can recognise two consecutive motifs – reluctance and miraculous craftsmanship. In encountering his provisioners, Odysseus too is reluctant, not to eat, but to bathe within sight of the maidens. Moreover, immediately after his bath, Athena transforms him, and her act is compared to that of a divinely instructed craftsman overlaying silver with gold. The transformed Odysseus radiates grace, so that Nausicaa likens him to a god. Compare the Buddha’s six-foot halo which makes him resemble a golden mountain and impresses the merchants when they see him.
Even in this compressed account of the provisioning of the protagonist I count eight distinct similarities: the provisioners’ journey, mobilisation by deity, intimate friend, protagonist enters company, scattering in fear followed by reassurance, protagonist’s reluctance, miraculous craftsmanship, and radiance. There are certainly more to be found in this episode, but at least my style of argument should be becoming clear. Taken singly, some individual rapprochements may seem tenuous, but the set as a whole is not easy to dismiss.
Five brief comparisons. (i) Towards the end of the Kāśyapa episode, the River Nirañjanā carries away the shorn hair and discarded paraphernalia of the converted eldest brother to set in motion the conversion of the younger ones; compare the Scherian river carrying away Odysseus’ buoyancy aid, the veil or mantilla of Ino, to return it to that goddess. (ii) Kāśyapa himself can be compared to Poseidon, the hero’s somewhat half-hearted enemy, who is ultimately unsuccessful. (iii) The notion of awakening implied by words like buddha and bodhi parallels the awakening of Odysseus by the girls’ shout. (iv) Yaśas travelling from his palace to the Buddha in the Deer Park outside Benares provides another, if less obvious, parallel to Nausicaa’s journey from palace to protagonist. (v) However, when Yaśas’ father follows his son, this journey corresponds (confusingly at first) to the journey in the opposite direction, when Odysseus follows Nausicaa to the palace – both stories include the theme of one-way invisibility.
Two individuals followed by a pentad. Rather than exploring these stories in detail, I now concentrate on the end of the first stay in Uruvilvā and the first sermon in Benares.
The Buddha, realising that much of human kind will fail to appreciate his doctrine, hesitates over whether to promulgate it at all. Persuaded to do so by Brahmā, he has then to decide whom to address first. His first and second choice are respectively his first and second ascetic teachers, Ālāra Kālāma and Udraka Rāmaputra – but both turn out to have died recently. His third choice is the set of five religious seekers, led by KauŠrinya, who were with him at the time of his tapas but left in disgust when he gave up his extreme asceticism, and are now staying near Benares. They are the first converts to the sangha, to be followed soon by Yaśas and his two groups of friends.
Is there anything in the Greek corresponding to the three choices of the Buddha? If we focus on the number of potential hearers for each choice, the Buddhist pattern is one-one-five. Now before reaching Scheria, Odysseus has recently had a number of encounters with females: first with Circe, then around the Straits with the two Sirens plus Scylla and Charybdis, and finally with Calypso. This is a one-four-one sequence and at first sight it offers little help. However, nearly all post-Homeric sources give three Sirens, i.e. five Straits Monsters in all, which gives the sequence one-five-one. To take the matter further, we need at this point to glance at the Mahābhārata, the third corner of the triangle mentioned earlier, and look at the cognate figures identified in my earlier papers. The Sanskrit epic not only has the Straits Monsters unambiguously paralleled by five crocodiles, but these five crocodiles are encountered after the figure that parallels Calypso: the Sanskrit sequence is Ulūpā (corresponding to Circe), Citrā‰gada (corresponding to Calypso), and the crocodiles led by Vargā. This is precisely the one-one-five sequence that we are looking for. In other words, perhaps the two teachers in the Buddhist story are the masculinised and somewhat spiritualised parallels to Circe and Calypso, while the five original monks parallel the monsters. So is there evidence to support this hypothesis?
(i) Pairing. The most striking feature of the teachers is their similarity. Apart from name and doctrine, the main difference is the degree of inducement they offer for their pupil to stay with them: Ālāra Kālāma offers joint leadership of the disciples, Udraka Rāmaputra goes further and simply offers leadership. But Circe and Calypso are also very similar: both are beautiful-haired guileful nymphs referred to as “dread goddess of human speech”, both are weavers living on wooded islands with sparse social contacts, and both swear oaths to the hero and provide for him a following wind when he departs. But of the two it is only the second who seriously tries to induce the hero to remain with her.
(ii) Pair contrasts with pentad. The Buddhist pair of teachers contrasts with the pentad not only in number but also as teachers versus taught. Moreover the Buddha leaves the teachers in disgust, while it is the pentad who leave the Buddha, disgusted at his stopping tapas. Apart from the numbers, the Greek contrast is between lovers and destroyers, and also between habitats – a single wooded island each versus dispersed locations around the Straits.
(iii) Hostility. The pentad see the Buddha approaching and agree in various ways to cold-shoulder him. The Straits Monsters are all in their various ways hostile, even dangerous, to mariners.
(iv) Overwhelming psychological force. The pentad in Benares simply find themselves unable to stick to their agreement, and Odysseus, although he knows that he should not, and physically he cannot, succumb to the lure of the Sirens, desperately desires to. Note that the motif of “intentions overwhelmed” applies in the one case to the protagonist’s opponents, in the other to himself.
(v) Ontological transformation. The pentad are converted and become Arhats. In non-Homeric sources such as Apollodorus and vase paintings, the Sirens are transformed by the passage of Odysseus. As had been prophesied, they die by casting themselves into the sea and at least one of them then becomes a godling.
(vi) Meal for six. After the begging rounds of the first monks the food is explicitly “sufficient to nourish six persons”. Is it accident that Scylla has six heads and that each makes a meal from one of Odysseus’ crew?
I conclude then that, surprising though it may seem, the first five Buddhist monks are cognate with the Greek Straits Monsters. But if so, if the rapprochement bearing on the context of the first sermon is correct, one might wonder about the doctrines expounded on the same occasion, and in particular about the Middle Way. The Middle Way cannot have as parallel the passage of Odysseus between the two cliffs, if only because he passes either on the one side or the other. However, a more abstract parallel may be detectable in his passage between the mundane pleasures represented by the Sirens with their beautiful singing and their fascinating knowledge, and the deadly horrors represented by the two Cliff monsters.
Dumézil and functions
As I said above, anyone working in Indo-European cultural comparativism now has to come to terms with the work of Dumézil, but despite the inspiration his work has given me, I have thought for more than ten years that his trifunctional theory needs to be slightly recast. Traditional Indo-European materials do indeed often show the characteristic pattern of first, second, third function – religion, war, production, or the sacred, force, fertility, or priests, warriors, merchants/farmers – the pattern is manifested in indefinitely many forms. However, often though not always, the triad is set within a larger pattern, which is typically pentadic. We need, I argue, to recognise a fourth function, pertaining to what is outside, other or beyond; but this fourth function is split between positively valued and negatively valued – for instance between the transcendent king (or the like) at the top, and the despised slave at the bottom.
Thus in the Rig Vedic myth of the origin of the varŠas, Puru±a represents the valued fourth function, the brahmans, k±atriya and vaiśya the classical or “core” functions, and the ®ūdra the devalued fourth function. To put it briefly, the proto-IE ideology was trifunctional, but there was more to it. What lay outside the three functions was often part of the same “form of primitive classification” (I deliberately use the language of Durkheim and Mauss).
Logically, a search for rapprochements between narratives and a search for five-element structures conforming to the classification are different undertakings, but in practice they overlap. The student alert to the patterns will find that they impose themselves, and I shall end by looking at two interlinked instances, which might be called Brahmā’s Address and Māra’s Testing. Dumézil himself discusses both of them in a late essay on the hesitation and temptation of the Buddha.
As we noted, the Buddha’s hesitation was overcome by the urging of Brahmā, who concludes with a passage in stanza form:
There formerly appeared among the Magadhans an impure doctrine thought up by polluted beings. Open this door of immortality. Listen to the doctrine conceived by the unpolluted one.
Just as someone standing on a rock on the summit of a mountain can see the people all around, so, O sage, climbing the terrace of the doctrine, you who see all things, look at the people afflicted by grief, dominated by birth and old age, you whose grief has disappeared.
Arise, O hero victorious in battle; O caravan leader freed of debts, travel the world. Preach the doctrine, O Blessed One, there will be knowers of it.
Dumézil notes correctly that the vocatives “sage”, “hero” and “caravan-leader” (sumedha, vāra, satthavāha) constitute a typical trifunctional set – first, second and third function respectively. But he ignores two things. First, he ignores the final vocative, Bhagavā, which addresses the Buddha himself: the Buddha, a unique figure in his own world-age, quite outside the normal run of sages, heroes and merchants, qualifies as valued fourth-functional. Second, he also ignores, in the first stanza, the polluted beings who should listen to the doctrine, who qualify as devalued fourth-functional. As so often, a trifunctional grouping is set within a larger pentadic structure.
Dumézil now moves to later sources which replace Brahmā’s Address with Māra’s Testing. He draws on a summary schema of the story presented in a standard handbook, where the Buddha is tested by Māra in three domains: that of warfare (the army of demons); the jural domain (when the Earth bears witness), and the domain of sexuality (Māra’s three seductive daughters). The domains fairly clearly represent the second, first and third function respectively. However the agents in the story are not only the demon soldiers, the trial witness and the representatives of the pleasures of the senses: the narrative structure also includes the triumphant undergoer of the tests and the devalued setter of them. In other words, the full functional pentad is again present, with the Buddha again in the supreme slot, but now with Māra representing the devalued fourth function.
My argument has been that a great deal in the Buddha’s biography has Indo-European roots, but at this stage I cannot offer any clear account of the paths along which this ancient tradition reached Buddhism. Certainly, those who first told the Buddha’s biography were not adapting either the Greek or the Sanskrit epic in anything like the form in which we now read them. For instance, if they had been adapting Homer, why did they suddenly cease to do so when they came to the Buddhas’s choice of audience for his first sermon, and have recourse instead to the one-one-five pattern which appears in the Sanskrit? All I can suggest is the hypothesis of a distinct “religious hero” tradition which at some point diverged from an epic hero tradition. But whether this separation occurred in India a few generations before the advent of writing, or very early in the history of the Indo-European tradition, or somewhere in between, I leave open. In any case, one way or another, the scriptural protagonist and the epic protagonist share a common origin.
 The first was “The hero’s five relationships: a Proto-Indo-European story”, in Julia Leslie (ed), Myth and myth-making: continuous evolution in Indian tradition (London: Curzon, 1996), pp. 1-20. More recently, see “Argus and Hanuman: Odysseus’ dog in the light of the Mahābhārata”, to appear in J. Indo-European Studies, 2000.
 Calvert Watkins, How to kill a dragon: aspects of Indo-European poetics (New York: Oxford University Press, 1995).
 N.J. Allen, “Varnas, colours and functions: expanding Dumézil’s schema”, Zeitschrift für Religionswissenschaft, Vol. 6 (1998), pp. 163-177.
 André Bareau, Recherches sur la biographie du Bouddha dans les Sutrapi˜aka et les Vinayapi˜aka anciens: de la quête de l’éveil à la conversion de ®āriputra et de Maudgalyāyana, (Paris: Ecole française d’Extrême-Orient, 1963).
 N.J. Allen, “The Indo-European prehistory of yoga”, International Journal of Hindu Studies, Vol. 2 (1998), pp. 1-20.
 I conflate the Pali and Chinese passages given by Bareau, op. cit., p. 28. I shall henceforth abbreviate the latter source to B.
 B p. 36-39; 5.486.
 B pp. 42-3.
 B pp. 101-2.
 Synthesised from 5.438-462, 6.85-98, 6.223-245.
 B p. 57.
 B p. 276.
 B p. 282-5, esp. Dharmaguptaka version; 6.94-95, 128-9.
 B pp. 106-123; 6.1-250. Nausicaa can also be compared to Sujātā.
 6.229-235; B p. 107.
 N.J. Allen, “Les crocodiles qui se transforment en nymphes”, Ollodagus, Vol. 13 (1999), pp. 151-167.
 B pp. 13-27; 5.203-213.
 B p. 55.
 B p. 161.
 B p. 184.
 G. Dumézil, La courtisane et les seigneurs colorés, (Paris: Gallimard, 1983), pp. 36-45.
24 B 135-6; Majjhima Nikāya, ed. V. Trenckner (London: G. Cumberledge, 1948), Vol. 1, pp. 168-9.
 L. Renou and J. Filliozat, L’Inde classique, vol. 2, (Paris: Ecole française d’Extrême-Orient, Hanoi, 1953), pp. 475-6.
 A full discussion would have to take into account the rapprochements between the biographies of the Buddha and of Krishna noted long ago by E. Senart, Essai sur la légende du Bouddha: son caractère et ses origines, (2nd ed, Paris: Leroux, 1882).