Wander Alone Like a Rhinocerous: Snp 1.3

With the third discourse, the famous Khaggavisāṇasutta ([snp1.3]), the mood changes once more. Here we find neither philosophical paradoxes nor dramatic dialogue, but a heartfelt and pragmatic plea for the solitary life of a wandering ascetic.

The poem strikes an uncompromising tone, a principled rejection of worldly attachments regardless of the cost, and an equally unsentimental acknowledgement of the tribulations of a solitary life in the forest. Perhaps we could read the three opening Suttas as respective responses to brahmanical philosophy, sensual enjoyment of the lay life, and asceticism.

The Khaggavisāṇasutta is universally accepted as an early text, due to its subject matter, style, and prevalence of archaic linguistic forms. While not wishing to dispute that fact that the Khaggavisāṇasutta is early, there are a range of considerations that suggest that in its current form it has been subject to considerable development. I will discuss these points further before drawing my conclusions.

It is commented on in full in the Niddesa, and is also quoted in the Apadāna. Twelve of the verses are found in Mahāvastu I.357. This is a text of the Lokuttaravāda-Mahāsaṅghikas, who stem from the other party in the first schism, dating some time after Ashoka. It contains twelve verses as compared to forty-one in the Pali, of which seven closely correspond. In addition, a verse is found in the Divyāvadāna of the Mūlasarvastivādins (182.014), corresponding to the second verse in the Pali. Finally, a version in forty verses has been found in a Gandhārī manuscript (Richard Salomon, A Gāndhārī Version of the Rhinoceros Sūtra, University of Washintgton, 2000, pp. 115–201). While the number of verses is the same as the Pali, several of the verses vary, as do their sequence.

The Khaggavisāṇasutta and its central image of “wandering alone” is a Buddhist expression of a broader Indian literature extolling the virtues of the solitary life. This has been explored in a recent article by Kristoffer af Edholm, ‘Wander Alone Like the Rhinoceros!’: The Solitary, Itinerant Renouncer in Ancient Indian Gāthā-Poetry (In: Larsson, S. and af Edholm, K. (eds.) Songs on the Road: Wandering Religious Poets in India, Tibet, and Japan. Pp. 35–66. Stockholm: Stockholm University Press). He presents the Aitareyabrāhmaṇa’s story of Rohita, an exiled prince in the forest who is exhorted by Indra (in disguise as a brahmin) in the virtues of the solitary forest life, wandering pre-eminent and untiring as the Sun. He shows that the idea has ancient roots, and that Indra himself is the archetypal solitary hero, always wandering alone. Many texts exhort the brahmin students and renunciants to wander alone, thus following the example of Indra and the Sun. Jain literature likewise extolls the solitary wanderer, who crushes his defilements like a war-elephant crushing his foe, like a lion, like the Sun, like a bird free in the sky. And the Jains even use the same term khaggivisāṇa.

One thing that is common to all these cases is that what wanders alone is a being, not a thing. (The Sun was regarded, at least poetically, as a divinity.) Nonetheless, there has been a long discussion in the Indological community about whether the term khaggivisāṇa means “that which has a sword for its horn” (i.e. a “rhinocerous”), or “the horn of a rhinocerous”. The most recent discussion on this is by Bhikkhu Bodhi, who supports Norman in reading “the horn of the rhinoceros” against both those who would read it as “a rhinocerous” and those who treat it as deliberately ambiguous. His main point is that “if khagga is the rhinoceros, then khaggavisāṇa is the horn of that animal. By analogy, however, if “rhino” is the animal, then “rhinoceros” must be the horn (“ceros”) of that animal. Dhivan Thomas Jones (Like the Rhinoceros, or Like Its Horn? The Problem of Khaggavisāṇa Revisited, Buddhist Studies Review, 31.2 (2014) 165–178) gives some similar cases of abbreviation in Sanskrit.

It remains the case that throughout the literature, the subject is said to “wander alone” like something else that wanders alone, not to “wander alone” like something else that is alone. This includes the Khaggavisāṇasutta itself, where the mendicant is urged to wander alone like an tusker, or like a king who flees his realm, or like a wild deer. Nowhere else do we split “wander” from “alone”, as is required if we are to read “wander alone like the rhinoceros’ horn [is alone]”.

I wonder, however, whether khagga and khaggavisāṇa are fully identical. When an elephant is used in this context, it is not just any elephant, but a nāga, a fully grown tusker or bull elephant. It seems that the male Indian rhinoceros has similar habits, in that he will associate with a breeding herd for some time, but is often found alone. I suspect that khaggavisāṇa might likewise be a male “horned rhinoceros”. Against this is that fact that female Indian rhinos also have horns. It is also the case, however, that in ancient India there were two species of rhinoceros. The Indian Javan Rhinceros, which became extinct in the 1920s, is unusual in that the cow has no horn.

This may not be a decisive problem, though, as female elephants have horns, yet we still use “tusker” for a bull. Obviously the phallic imagery plays a part here. It’s also true that females, both elephants and rhinos, may sometimes “wander alone”. The point is not that these things are categorical truths, but that they are evocative ideas employed by a poet to paint a mood. This is all very tenuous, but I am inclined to think that khaggavisāṇa means “bull rhino”.

The imagery is tough and masculine: rejecting affection, living independent like a lion, suspicious of others’ motives. An overly harsh attitude to seclusion can lead to a certain hard-heartedness, so the poem does not neglect to mention the development of love and compassion. Indeed, when the Sutta is quoted in the Apadāna, there is an extra verse right at the start, which begins the whole discourse speaking of love and compassion ([tha-ap2:9.3]). It gives rather a different framing to the poem.

A closer look at this verse shows that three of the lines are the same as the verse that follow, meaning it adds only one line to the Suttanipāta version: mettena cittena hitānukampī. In fact, this verse appears four times with only the third line changed—once in the Suttanipāta, the extra verse in the Apadāna, in the Mahāvastu, and in the Gandhārī. The verse opens with the same two lines:

When you’ve laid down arms toward all creatures,
not harming even a single one,

Then the Suttanipāta rather abruptly changes course: “don’t wish for a child, let alone a companion.” But the other three versions all continue on the theme of love. The first verse of the Apadāna has here, “live full of compassion with a loving heart”. It turns out that the Gandhārī version starts with the very same verse including the same third line (metreṇa citiṇa hitaṇ‍ukaṁpi). The Mahāvastu has different line on a similar theme, “having laid down arms against creatures firm and frail” (this line in Pali at [sn6.3:7.3]).

It is also noteworthy that the “lost line” of the Apadāna and the Gandhārī is also found in the Mahāvastu: maitreṇa cittena hitānukaṃpī. But there it is in a different verse along with compassion, joy, and equanimity. Thus we find verses added or subtracted, and lines swapped in different places more or less aptly. The “lost line” of the Apadāna would seem to be a case where a later text preserves a verse lost to the generally earlier texts.

The fact that this is found in the Mahāvastu, however, does not mean that the Mahāvastu version as a whole is earlier. For five of its twelve verses expand on the topic of “affection” (sneha), adding minor variations to a topic dealt with in only one verse of the Pali and the Gandhārī.

All of our current versions have clearly been subject to a degree of revision and flexibility. This is easily done in such a format, where there is no logical sequence of ideas, merely verses on a related theme.

The tradition—including the Niddesa, the Apadāna, the commentary, the Mahāvastu, and the Divyāvadāna—is unanimous in attributing these verses to the paccekabuddhas. The association is so close that the Divyāvadāna uses “one who is like a rhinoceros” as an epithet for a Paccekabuddha, even outside the context of the poem (490.3). Paccekabuddhas were a mysterious class of enlightened sages who lived a solitary life of meditation without establishing a dispensation. They are not prominent in the suttas, featuring in one late sutta of the Majjhima ([mn116]), and in four passing mentions in the Aṅguttara. But the later texts of the Khuddaka show the growing popularity of the idea, mentioning them fifty-seven times. Clearly, if they were part of early Buddhism at all, it was in a very minor role. Given that the Khaggavisāṇasutta itself says nothing of the Paccekabuddhas, it seems unlikely that this was the original intent. It would seem that the idea of the Paccekabuddha was part of the pan-sectarian development of what one might call “Buddhology” in the post-Ashokan period, along with the Jātakas, the pāramī (“perfections”), and the ideal of the Bodhisattva.

Rejecting the traditional Paccekabuddha connection, scholars such as Jayawickrama and Saddhātissa have argued that the sutta represents an early period of ascetic Buddhism before settled monastic life. Bhikkhu Bodhi, on the other hand, points out that the Vinaya Mahāvagga and its parallels depict settled monastic life developing soon after the Buddha began teaching. As an alternative theory, he suggests the Khaggavisāṇasutta may have been meant for a class of forest-dwelling, reclusive mendicants.

The point about the early introduction of building is well-taken. Let alone the Vinaya Mahāvagga, this is found even in the oldest portion of the Vinaya, namely the Pātimokkha, which is common to all the schools and is universally accepted as an early document. There we find not just mention of buildings, but requirements as to building methods and materials, safety regulations, restrictions on the size of huts, environmental regulations governing prospective sites, procedures for communal agreement on new construction, and acknowledgement of the role of sponsors. This speaks to a developed and sophisticated approach to communal buildings in the lifetime of the Buddha.

It thus seems unlikely that these verses predate buildings for monastics, but does this mean the verses were addressed to a distinct class of reclusive mendicants? There are a couple of verses in the Khaggavisāṇasutta that, to my mind, suggest a quite different origin. If we follow these verses back to their source, they originate not from an early period or from a special class of renunciants, but as a call for those who have allowed settled monastic life to dull their inspiration.

These verses are numbers 11 and 12. They nuance the virtue of solitariness: if you can find a good companion, that is best, but if not, live alone. While this is more conciliatory than the sterner stuff typical of the Khaggavisāṇasutta, we have already seen that the messages of love and compassion are also central to the text.

These verses also appear in the Gandhārī, although in a different position. Nonetheless, they are clearly an interpolation. They have a distinct form and are act a pair, as the first verse lacks the usual ending tag line and instead leads on to the second. What is more, these verses occur independently in a quite different context, which is found in the Vinaya ([pli-tv-kd10:3.1.32]–39), as well as [mn128:6.30]–37 and [ja428:8.1]–9.4 (They are also found at [dhp328]–9, where there is no context). While it is therefore clear that they have an independent origin, it seems that the redactors of old considered their message and style congruent enough to justify their inclusion in the Khaggavisāṇasutta. This may have predated the split between the Pali and Gandhārī versions, although the similarities of theme could have easily prompted them to be included separately.

Now, the background to these verses concerns the so-called incident at Kosambi, which is related in detail in the Vinaya. There, monks had a rather disspiriting falling-out, allowing a trivial difference of opinion to spiral into an acrimonious dispute. The Buddha, unable to reconcile them, gave up and retreated alone in the forest, where he spoke these verses. This famous story is overlooked entirely in the commentary to the Khaggavisāṇasutta. But surely it gives us a clue to the role such verses play in the community.

Each verse of the Khaggavisāṇasutta is an ought. When a spiritual teacher sees that a community is out of balance, they teach what people should do, not what they are already doing. The narrative background shows that these two verses were a reaction to increasing settledness and institutionalization. And if that is true of these verses, it is likely the case for the Khaggavisāṇasutta as a whole.

This highlights a further fact, one that we know with far more certainty than anything about the origins of the text. The Khaggavisāṇasutta has been maintained and passed down for thousnads of years in monastic communities. It was widely popular across the traditions, and the overwhelming majority of those who learned and recited it, and later, those who read and studied it, have been settled monastics. Why would we think this was not the original audience? Spiritual teachings do not act in a unilateral way, where the one thing is addressed to one group of people who are already doing that thing. No: the whole point is to instill a yearning for something else.

The monastic vocation, to this day, has a tension between a life of institutionalized security and one of wandering freedom. This is part of the core dynamic of our communities. It cannot be resolved by a decree; all mendicants feel the call of the solitary life. To live alone, aloof from the dramas and desires of other people, the open road and the creatures and the trees your only companions; who has not felt the yearning for such a life?

While it does not affect the above argument, I should note that there is a confusion in these verses that needs clearing up. The reading of the two verses differs in the editions. The Pali Text Society edition abbreviates the tag line with just eko care, implying that it should be expanded as the usual “wander alone like a rhinoceros” (eko care khaggavisāṇakappo). The Buddha Jayanthi Tipitaka edition spells out the “rhinoceros” tag here, supporting the PTS reading. This has been followed by Norman and Bodhi, who both note that this line differs from its parallels. If this reading is correct, then these verses have been adapted to this context by giving them the “rhinoceros” tag. In the Mahāsaṅgīti edition, however, these verses are essentially identical with their parallels, ending with the tag “wander alone like a tusker in the wilds” (eko care mātaṅgaraññeva nāgo). That this is no oversight is clear from the fact that MS acknowledges the PTS and BJT variant. The commentary is no help, as it is silent on this line. In the Apadāna version, the “elephant” tag is found in all three editions. The Mahāsaṅgīti edition, however, has the “rhinoceros” version in the Niddesa ([cnd23:140.4]), as does the BJT and apparently also the PTS, though once again it is abbreviated. The Mahāsaṅgīti readings throughout are the same as the VRI edition on which it is based, representing the Burmese edition of the Sixth Council. The Gandhārī has “rhinceros”.

Clearly the last line of the verse originally referred to an “elephant” and it was changed to “rhinoceros” to fit the context of the Khaggavisāṇasutta. Editors have evidently been swayed in both ways, either to keeping consistency with the original context, or creating a new consistency with the new context. Given the support of the Gandhārī, I am inclined to think the change was a deliberate choice of the composer or compiler at an early date.

Returning to the question of dating, we have seen that the Khaggavisāṇasutta is subject to many of the same issues found throughout the verse collections. Verses are added or lost, lines are swapped around, exact terms and phrases are varied. The interpretation of the Niddesa, the earliest commentary, is not infallible, and it employs a doctrinal framework, the Paccekabuddha, which is later than the original context. Note that it is the last text of the Niddesa, following the Pārāyanavagga, which also includes late passages. Verses have been multiplied in a formulaic way, there are repetitions, imports from elsewhere, and awkward transitions. The whole is, like the chapters of the Dhammapada, a compilation, a form that readily lends itself to expansion and reshuffling. Comparison with Gandhārī and Sanskrit versions reveals no less variation than we might find in any other verse collections. Attempts to identify the text with an early period of renunciant life are not persuasive. Where we can establish context, it suggests that the text began as it continued and as it is today: a call for renewal praising the solitary wandering life primarily for those who are not living that life.


I have updated this essay, and my apologies for anyone who read it earlier! i may yet do so again.

In particular, see the incendiary final paragraph, where I overturn a century’s consensus!