‘The dialogue between a cat and a mouse’ in Mahābhārata 12.136 and narratives about spiritual liberation (mokṣa) in Ancient Indian literature

Abstract

‘The dialogue between a cat and a mouse’ (Mārjāramūṣakasaṃvāda) is an animal fable used in the Mahābhārata to provide instruction in statecraft (nīti). This article argues that the Mahābhārata version of this tale must be based on an earlier soteriological allegory about a brahmin who provides spiritual liberation to a king in exchange for protection. The Mārjāramūṣakasaṃvāda abounds in terms and phrases that, in addition to their everyday meanings, have a technical or typical usage in the ascetic traditions of Ancient India. Moreover, the conversation between the cat and the mouse resembles that between a teacher and a disciple, rather than a discussion of a possible alliance between two kings. The hunter of the Mārjāramūṣakasaṃvāda can be identified with the Buddhist Māra. To support the plausibility of this soteriological reading, the article includes a discussion of Buddhist jātakas with a similar plot.

‘The dialogue between a cat¹ and a mouse’ (Mārjāramūṣakasaṃvāda) is one of the animal fables that is used in the Āpaddharmaparvan subsection of the Śāntiparvan of the Mahābhārata to provide instruction in statecraft (nīti).² This story is used to illustrate a nīti teaching about a weak king (the mouse) who should make a temporary alliance with a strong former enemy (the cat) when surrounded by several hostile kings (a mongoose and an owl) and facing certain defeat. A further theme that is explored in the second part of the dialogue is the danger that may arise for the weaker king (the mouse) if his alliance with the stronger king (the cat) continues longer than necessary: the cat may eat the mouse if his survival no longer depends on the mouse. Adam Bowles has already explored these themes and shown how this episode fits in to the broader context of the Āpaddharmaparvan.³

The Mārjāramūṣakasaṃvāda is the second of three animal fables that are told consecutively in the Āpaddharmaparvan. Rudolf Franke has noted that the first⁴ and third⁵ of these episodes have parallels in the Pali jātakas, but he also mentions the Mārjāramūṣakasaṃvāda in relation to some phrases that it shares with other Pali jātakas.⁶ I believe that the third of these fables, Mahābhārata 12.137, shows signs of being adapted from a Buddhist or ascetic source. At any rate, it is clear that it was not composed for the purposes of instruction in nīti because it involves a discussion of philosophical and religious matters. A detailed demonstration of this claim requires another article. The placing of the Mārjāramūṣakasaṃvāda between these two tales increases the possibility that the Mārjāramūṣakasaṃvāda was adapted from Buddhist-influenced or ascetic materials.

In this article I will argue that the Mārjāramūṣakasaṃvāda, as found in the Critical Edition (henceforth: CE) of the Mahābhārata,⁷ must be based on a story that had a different allegorical meaning. The Mārjāramūṣakasaṃvāda uses terminology and imagery commonly found in discourses about renunciation and spiritual liberation (mokṣa), and the cat treats the mouse as a teacher (guru) rather than as an allied king. These features of the story, as well as some contradictions and repetitions in the narrative, can be explained by positing an earlier version of the story, which was meant to illustrate how a brahmin (mouse) provides spiritual liberation to a king (cat), while the king protects the brahmin from hostile powers or from other classes (varṇas) in the society.

After a few more introductory remarks, a short summary of the plot of the Mārjāramūṣakasaṃvāda will be provided and an explanation for a contradiction in the plot will be offered (in the second to fourth sections). The fifth section will introduce some of the key arguments that support a soteriological reading of the tale by noting several words and phrases that can be understood in two different ways—secular and soteriological. The sixth section will discuss two passages in which the mouse is depicted as a (possible) minister to the king, and the seventh will provide evidence for understanding the four animals of the story as representing the four classes (varṇas) of the society. In the eighth section we will turn to the hunter, the only human character, arguing that he represents both death and spiritual bondage. This will lead to a discussion of stories about escaping death and Buddhist jātakas with a comparable plot in the ninth to eleventh sections. In particular, I will emphasise that structurally similar stories about escape, especially from a fowler, are interpreted in a soteriological way as an escape from Māra in the jātakas of the Mahāvastu, while the Pali tradition avoids such an interpretation of its versions of the same jātakas. Some speculations about this difference will be offered. Thus, while the main focus on this article is on the Mārjāramūṣakasaṃvāda, it will deal in some detail with jātaka material. This is indispensable for evaluating the extent to which the proposed soteriological interpretation of the Mahābhārata narrative is plausible. The twelfth and thirteenth sections will discuss some features of the story that are of secondary importance to the narrative, but that can nevertheless be given a soteriological interpretation in order to read the entire⁸ Mārjāramūṣakasaṃvāda as a soteriological allegory.

Narrative and didactic elements of the story

Animal fables in the Mahābhārata are unique in that they are written entirely in verse, as opposed to the Pali jātakas ⁹ and fables in the Hitopadeśa and later Pañcatantra, which are composed of a fixed or didactic versified part and a unique prose version of the story.¹⁰ When one compares, for example, the immediately following Mahābhārata story (12.137) with Kuntanijātaka (no. 343), it is easy to observe that all four verses of the jātaka are found (with minor variations) in Mahābhārata 12.137.30–33. This suggests that at least these four verses were transmitted as a more or less fixed group, around which different narratives could be constructed in prose or verse. Likewise, different Buddhist traditions elaborate different narratives around similar jātaka verse material, as will be exemplified below. When we come to the Mārjāramūṣakasaṃvāda, there is no a priori means for knowing which verses, if any, belong to a hypothesised shorter or ‘core’ version of the story.

It is possible, however, to introduce a different distinction—between narrative and didactic material. Narrative passages are those that are relevant for the story proper, being descriptions of the characters and the events that take place. These can be contrasted with verses or pādas (quarters) of verses that are purely didactic in nature and could easily have circulated as general nīti knowledge, not tied to the particular narrative. Apart from the narrator Bhīṣma's introduction, in the first part of the narrative (up to verse 117) didactic materials are found in verses 39, 44 (pādas c to f), 45 (pādas a and b), 57, 58 (pādas c and d), 89, and 102–106. I propose that these didactic materials were added when the narrative was adapted for nīti purposes. The versified format of the story obscures the fact that at least some of these didactic verses are most likely quotations. Apart from the didactic verses listed above, there are a couple of narrative verses that support the nīti interpretation of the story advocated in the Mahābhārata itself. Verse 42 mentions kṣatravidyā (‘royal knowledge’, a synonym of nīti) and verse 86 refers to the mouse as amitraghna, ‘slayer of enemies’. For my hypothesis to be viable, these must also belong to the nīti reworking of the mokṣa narrative. I will work on the assumption that most of the verses that belong to the story proper predate the adaptation of the story for the purpose of instruction in nīti. Nevertheless, I will not attempt to reconstruct a definite shorter version of the story.

Despite the evidence being inconclusive, I believe the hypothesis of an earlier mokṣa version of the tale is worth exploring due to the cumulative evidence and its inherent plausibility. Moreover, although it does not provide allegorical meaning to every single detail of the story as it now stands, such a mokṣa reading gives a much richer meaning to the details of the story than does its nīti interpretation in the Mahābhārata. By assuming a higher level of sophistication and literary skill than is obvious, it is thus generous to the tradition that produced it.

Plot summary

To make the argument easier to follow, a summary of the story is now provided, which notes some contradictions in the plot. For the purposes of this article, the plot is divided into three parts.

In the first part, a hunter sets a trap into which a cat falls. Then a mouse starts eating the bait attached to the trap. While the mouse is eating, two predators appear. If the mouse climbs down to the ground he will be eaten by a mongoose. If the mouse remains where he is, he may be eaten by an owl who hovers over him. The mouse comes up with a plan to make an alliance with the cat, suggesting to him that he will set the cat free if the cat spares his life and protects him from the other predators. The cat agrees. As soon as the mouse enters the trap, the owl and mongoose leave. (It is conceivable that a shorter version of the story ended here.)

In the second part of the story, the mouse takes his time to release the cat. The reason given for this is that the cat will surely eat the mouse once there is no longer any danger. The mouse only releases the cat when the hunter approaches in the morning so that the cat has no time to eat the mouse. When the hunter arrives, however, the mongoose and the owl are said to leave for a second time, so here a contradiction is introduced. At this point all the characters return to their homes, which looks like an obvious ending.

The story continues, however, with a third part, in which the cat approaches the mouse at his home, asking to continue their friendship. The mouse refuses to come out, observing that the cat could now eat him, and the story ends here.

On the parts of the story

In this article I will propose a mokṣa reading of the first two parts of the story only, as I believe that the third part is an addition to a shorter version of the same story. That the third part is an addition to an already complete narrative is obvious from the structure of the story. Furthermore, two different lessons are learned from the first, or first+second, and third parts. While the first two parts in their nīti interpretation teach mainly about making an alliance with a former enemy, the third part is concerned with the dangers involved in maintaining an alliance with an inherently dangerous companion. Moreover, in the Pañcatantra version, the equivalent of the third part is a separate episode, and the dangerous animal who wants to form a friendship with the mouse is not the same animal that was saved in the first part.

Why do the predators leave twice?

As noted in the summary above, the predators are said to leave twice—first, after the forming of the alliance between the mouse and the cat, and, second, when the hunter arrives. In the first instance, ‘the mongoose and the owl went home without hope having seen them two [i.e. the cat and the mouse who had just become allies]’ (tau dṛṣṭvā nakulolūkau nirāśau jagmatur gṛhān; 12.136.82). This is followed in some northern manuscripts (K4,5 V1 B0-2,5 Da Dn1,n3 D2,3,8) by a further explanation in three verses which are provided as a starred passage in the CE. It clarifies that the predators were ‘unable to attack’ (aśaktau … saṃpradharṣayituṃ; *311, line 4) due to the successful alliance between the mouse and the cat. This explanation of the predators’ departure fits the logic of the nīti framework well. In the second instance of departure, however, the details are slightly different:

Now the mongoose and the owl immediately lost hope, being very frightened upon seeing that terrible-looking [hunter].

atha cāpi susaṃtrastau taṃ dṛṣṭvā ghoradarśanam

kṣaṇena nakulolūkau nairāśyaṃ jagmatus tadā (12.136.112)

This is followed by two verses (113 and 114) that repeat, verbatim, part of the explanation that was relegated in the first instance to a star passage after verse 82.¹¹ I cannot see how a later departure of the predators could fit the logic of the nīti framework. Moreover, the later passage (112–114) appears to be a combination of two distinct ideas. If the cause of the predators’ departure is the arrival of the hunter, it is hardly relevant that they are unable to attack the mouse because of its excellent alliance with the cat.

Thus, the manuscript transmission of this detail of the story is problematic. It is possible to offer a speculative explanation of how this might have come about. In the mokṣa version of the story, the predators left when the hunter arrived. When the nīti version was made, their departure was moved to an earlier point in the story in order to illustrate the potency of the alliance. For some reason, the manuscript tradition retained traces from the earlier version so that all the manuscripts used for the CE retain the contradiction of the predators leaving twice. Confusion arose as to the proper place of the verses 113–114, which were introduced when creating the nīti version, and were kept by some manuscripts in their proper place after 82, but ended up in the CE text in the wrong place, where they form, at least, a possible logical contradiction with verse 112.

The alliance between the cat and the mouse: ordinary and soteriological language

In this section I will list a number of features of the conversation between the cat and the mouse which lend support to a mokṣa interpretation of the story. The Mārjāramūṣakasaṃvāda uses a number of words and expressions that could be interpreted in a secular as well as a soteriological way. To begin with, mokṣa, which in the context of the story most readily means ‘release from the trap’, is one of the most frequently used terms for ‘release from saṃsāra’. In the next sections we will discuss a number of further instances of ambiguous language.

On being nirvṛta

In verse 107 the mouse Palita addresses the cat Lomaśa using another term with two meanings, namely, nirvṛta:

The majority of threads have been cut. Just a single thread remains. I will quickly cut that one also. Become calm (nirvṛta) Lomaśa.

chinnaṃ tu tantubāhulyaṃ tantur eko 'vaśeṣitaḥ

chetsyāmy ahaṃ tad apy āśu nirvṛto bhava lomaśa (12.136.107)

To become nirvṛta, however, also means to become spiritually liberated or to reach nirvāṇa, the latter term being derived from the same verbal base nir+vṛ. Thus, a second meaning emerges—the mouse is promising the cat that his liberation is imminent. The corresponding Pali word nibbuta is commonly used to describe a spiritually liberated person. Nirvṛta characterises a sage at least once in the Śāntiparvan of the Mahābhārata (12.171.59), but that passage has been demonstrated to show ‘remarkable influence from Buddhist and Jain sources’.¹²

Crossing over

When Palita proposed the alliance to the cat, he uses a simile of crossing a river:

Whoever uses a piece of log to cross a very deep and great river, he makes that log cross over and he is carried over by the log. Such shall be our alliance of good escape (sunistara). I will make you cross over and you will make me cross over.

kaś cit tarati kāṣṭhena sugambhīrāṃ mahānadīm

sa tārayati tat kāṣṭhaṃ sa ca kāṣṭhena tāryate

īdṛśo nau samāyogo bhaviṣyati sunistaraḥ ¹³

ahaṃ tvāṃ tārayiṣyāmi tvaṃ ca māṃ tārayiṣyasi (12.136.60–61)

The parable of crossing a river is well known, especially in Buddhist sources, where it stands for spiritual liberation. For example, in the Nāvāsutta (Suttanipāta 316–323¹⁴) a teacher provides knowledge for others as a means to cross over; he is the compassionate boatman. Very similar language is used in Mahābhārata 12.313.22–23. Crossing to the other shore, with or without mention of a teacher, appears in a number of other Mahābhārata passages (for example, 12.227.17; 12.228.2; 12.309.17), including Bhagavadgītā 4.36.

Of course, similar language about getting to the other side could also be used with regard to overcoming more mundane kinds of hardship. For example, Mahābhārata 12.128.49 says that ‘[the king] who possesses a treasury overcomes (tarati) everything (sarvaṃ tarati kośavān)’. Nevertheless, I find that the particular way the mouse expresses himself must be intended to evoke the soteriological usages of ‘crossing over’.

In contrast to other usages of this simile, such as in the Buddhist Nāvāsutta and Mahābhārata's Janakaśukasaṃvāda, the present version claims that the teacher gets himself over. This might have been intended to be satirical, portraying the mouse as acting in his own best interests rather than helping the cat out of compassion.¹⁵

Student and teacher

The dialogue between the cat and the mouse, if didactic nīti verses are removed, does not read like a conversation between two kings. Rather, the relationship between the cat and the mouse is that between a master and a student or devotee.

As they make an alliance, the cat calls himself a student (śiṣya) and a devotee (bhakta) of the mouse (12.136.69), and the cat says that he has sought refuge (śaraṇaṃ gataḥ) from the mouse. Words related to bhakta are commonly used in the Buddhist literature to signify the relationship one should have to a kalyāṇamitra—a ‘good friend’¹⁶ who provides assistance on the spiritual path, or simply to a wise person (paṇḍita).¹⁷ I believe that such usages of the term also inform the present passage.

Of equal significance is the phrase śaraṇaṃ gataḥ, ‘gone for refuge’, which is found in the same verse (69). While it can mean simply ‘to seek protection’, it is also a standard phrase in Buddhism, where ‘going for refuge’ to the Buddha effectively means conversion to Buddhism.

I find it difficult to see why such statements would be made by the cat if this was meant to be no more than an alliance between two kings who are in equal danger.

A possible further parallel with the Buddhist usage of bhakti is related to Theragāthā 370, which says that one who is ‘devoted’ (bhaktimān) to wise people becomes wise (paṇḍita) himself. This recalls the mouse's question: ‘Will this enemy now become wise (paṇḍita) through [our] alliance/association (saṃgatyā)?’ (apīdānīm ayaṃ śatruḥ saṃgatyā paṇḍito bhavet; 12.136.46). This further supports the mouse's position as a possible teacher, and not simply a weak king who draws on his nīti wisdom to survive.

On prasāda

At one point the cat says that he will obtain life by the ‘favour’ (prasāda) of the wise (prājña) mouse (12.136.76). Prasāda is another term that has a technical meaning in the soteriological traditions of Ancient India, apart from its secular meaning of ‘favour’ or ‘help’. Moreover, this technical usage is often related to the teacher-student relationship, which has already been discussed. In a soteriological context, the word prasāda refers more specifically to the liberation-granting favour that the guru, or sometimes a deity, grants to his (or her) disciple. Although such a path to spiritual liberation is not common in the Mahābhārata, it is found in its Janakaśukasaṃvāda (12.313), which claims that the prasāda of his guru Vyāsa has given divine knowledge (jñānaṃ divyam) to King Janaka and has contributed to or caused the perfection of Vyāsa's son Śuka (12.313.42–43).¹⁸ Incidentally, this verse occurs in a conversation that has already been referenced above because it employs a simile of crossing over with the help of a teacher. This suggests that the author of the mokṣa version of the Mārjāramūṣakasaṃvāda may have been familiar with the religious developments recorded in the Janakaśukasaṃvāda.

Becoming a minister?

The alliance between the cat and the mouse is described using a phrase that is difficult to interpret in its given nīti context. When the mouse convinces the cat to accept his proposal, the mouse ‘being very confident entered [the trap where the cat was] and properly undertook arthas’ (praviveśa suvisrabdhaḥ samyag arthāṃś cacāra ha; 12.136.80). I will explore two possible understandings of the statement, but, unfortunately, neither of them can account for the usage of the plural of artha.¹⁹ One way of translating artha with a motion verb would be to say that the mouse undertook his ‘work’ (artha), as specified in their deal. A problem with this interpretation is that the mouse by no means plays his part properly (samyak), at least from the cat's point of view, because he postpones the cat’s release. (The postponement will be discussed below.) An alternative interpretation would be that the mouse undertook artha in the sense of arthaśāstra, namely, became a minister for the king. A verb derived from the same root car appears with the puruṣārthas dharma, artha, and kāma in Mahābhārata 12.72.3. Again, a satirical interpretation may be suggested. While the verse says that the mouse properly undertook its duty, which was to release the king, it did nothing of the kind, instead becoming a minister of the king.

A very certain statement about the mouse becoming the cat's minister is found later in the story, when the hunter has departed and the mouse is safely back home:

Please become a lord of my body and of my house. Become my teacher regarding all matters. Become my minister, oh wise one, teach me like a father. You have no danger from me, I swear by my life. In wisdom you are like Uśanas himself, whereas I have the strength. Joined with the power of your counsel (or spells: mantra) one would find only triumph.

īśvaro me bhavān astu śarīrasya gṛhasya ca

arthānāṃ caiva sarveṣām anuśāstā ca me bhava

amātyo me bhava prājña piteva hi praśādhi mām

na te 'sti bhayam asmatto jīvitenātmanaḥ śape

buddhyā tvam uśanāḥ sākṣād bale tv adhikṛtā vayam

tvanmantrabalayukto hi vindeta jayam eva ha (12.136.125–127)

Nowhere are the varṇa identities of the cat and the mouse as kṣatriya and brahmin made as clear as in this statement. The cat claims to have physical strength and ambitions of military conquest, while he compares the mouse to the legendary brahmin sage Uśanas as he invites the mouse to become his minister (amātya). Uśanas, however, was not an amātya but the purohita (chief adviser or chaplain) of the asuras (demons), a position normally reserved for brahmins. The reverential attitude of the cat towards the mouse, both in this passage and in verse 36, discussed above, is reminiscent of how a king should treat his purohita according to Arthaśāstra 1.9.10: ‘[The king] should regard [the purohita] like a student regards his master, like a son regards his father, like a servant regards his master (tam ācāryaṃ śiṣyaḥ pitaraṃ putro bhṛtyaḥ svāminam iva cānuvarteta).’²⁰

As noted above, there are reasons to believe that the final part of the narrative was originally separate. Regardless of whether I am right about the original independence of the first part, the cat's speech discussed above shows that at some point in the redaction or transmission of this story, the cat and mouse were perceived as a king and a brahmin.

The four animals

There are four animal characters in the story—a mouse, a cat, a mongoose, and an owl. All four animals represent kings in the nīti version presented in the Mahābhārata. There are several indications that the four animals were originally intended to represent the four classes (varṇas) of the society.²¹ The strongest evidence for identifying the mouse as a brahmin and the cat as a kṣatriya comes from the narrative itself and from their conversation, as discussed in the previous section. In this section I will discuss the names and the adjectives that describe the four animals in relation to the association of specific colours (varṇas) and traits with the four classes (varṇas) of the society. Some of the adjectives that describe the animals, however, seem to be purely physical descriptions of the species (for example, tīkṣṇatuṇḍa, ‘having a sharp beak’, for the owl in verse 32). If an allegorical or double (śleṣa) reading of such adjectives is possible, it escapes me, and I will therefore not discuss them.

The colours of the four classes are identified in Mahābhārata 12.181.5, which states:

The colour of brahmins is white, but the colour of kṣatriyas is red. The colour of vaiśyas is yellow, but the colour of śūdras is black.

brāhmaṇānāṃ sito varṇaḥ kṣatriyāṇāṃ tu lohitaḥ

vaiśyānāṃ pītako varṇaḥ śūdrāṇām asitas tathā (12.181.5)

Animal fables in Indian literature (and beyond) normally illustrate a universal virtue or truth. The animals who act in such fables are associated with certain character traits, and the fables about such animals tell a universal truth about all people who have such characteristics. It is well known that ‘[f]requently it is the names of the participants in fables which carry the moral of the fable’.²² This is true, for example, of the story that directly precedes the Mārjāramūṣakasaṃvāda which tells how Procrastinator (Dīrghasūtra) perished, while Far-Sighted (Dīrghadarśin) escaped from danger (Mahābhārata 12.135). Generally, it appears that when fables need to talk about kings or brahmins in particular they introduce such human characters in the story itself, where they often act alongside animals. Alternatively, a character may be designated as the king of a certain species.

In this case, however, the names of the four animals do not denote virtues, vices, or other characteristics. This convention of the fable genre is probably broken because the story, in its mokṣa version, is more of an allegory than a regular fable. In the mokṣa reading the four animals do not represent character traits, but rather the four classes of the society.

The mouse

The name of the mouse is Palita. This would usually be translated as ‘grey’, but palita can also mean ‘old’. Fitzgerald combines both meanings, here translating Palita as ‘Old Gray’.²³ According to the colour scheme cited above, a brahmin would have to be white rather than grey. If we understand, however, that Palita refers to the hair colour of an old person, it no longer necessarily needs to mean ‘grey’. Some people's hair turns white rather than grey when they become old. Thus, the Pali-English Dictionary (henceforth, PED²⁴) suggests that the compound palitakesa (‘one whose hair is palita’) means ‘with grey (that is, white) hair’ (s.v. palita).²⁵

Palita is first introduced as ‘very wise’ (mahāprājña) in verse 21. The Pali equivalent mahāpañña is usually applied to the Buddha himself or to his chief disciples, and, moreover, in repetitive passages it is often repeated after paṇḍita (learned), an adjective normally reserved for brahmins or śramaṇa teachers. Palita is later qualified by several similar adjectives that testify to his learning and wisdom, such as ‘knowing the highest meaning’ (paramārthajña) in verse 128 or ‘knowing the essence of dharma’ (dharmatattvajña) in verse 181. These would certainly be appropriate characteristics for a brahmin, although an objection could be made that such epithets are not exclusive to brahmins in the Mahābhārata.

The cat

I suggest that the cat represents a kṣatriya in the mokṣa reading of the story. He is the strongest of the four animals. Although this point is made explicit only in the Kathāsaritsāgara version,²⁶ it is presupposed by the Mahābhārata version as well since the owl and mongoose do not even contemplate an approach once the mouse is under the protection of the cat. The cat's name in the critically constituted text is Lomaśa (‘Hairy’), which has no kṣatriya connotations that are obvious to me. When the cat is introduced in verse 22, manuscript T2 reads his name as Lohita (‘Red’), which is the colour of the kṣatriya, as seen above. It must be noted, though, that a change from Lomaśa to Lohita seems unlikely, making the latter the lectio difficillior. Admittedly, the support of a single manuscript is very weak evidence for claiming that this must have been cat's original name. Yet it should be noted that a relatively small number of manuscripts were used to create the CE of any section of the Mahābhārata, and there is not enough data to dismiss this as an isolated scribal mistake or some other corruption. Unfortunately, only two Telugu (T) manuscripts were used by the editors, and it is therefore impossible to determine how frequent the reading ‘Lomaśa’ appears in the Telugu recension. Furthermore, T2 transmits many readings that it does not share with any of the other manuscripts used for the edition, so that one may wish to consider T2 as a witness to a separate recension altogether.²⁷

The cat is further introduced as pakṣisattvāvasādaka, ‘destroyer of winged creatures’, in verse 22. Fitzgerald understands the word sattva in this compound differently and translates it as ‘a nemesis of the birds’.²⁸ The entire Southern recension and D7, however, read sarva (‘all’) for pakṣi (‘bird’). The resulting compound sarvasattva is commonly used in the sense of ‘all beings’. I believe that sarvasattva is more likely to be the original reading and that the compound sarvasattvāvasādaka, ‘he who lets down (or hurts) all beings’, might have been used to mark the cat's identity as a kṣatriya.²⁹ A criticism of kṣatriyas as violent and unreliable is one of a variety of views on kings expressed in the Mahābhārata as well as early Buddhist literature. Notably, such a view is especially prevalent in fables and other narrative literature when describing interactions between a king and an ascetic or another wise person. One needs to look no further than the immediately following Mahābhārata episode, where it is observed that ‘one should not trust kṣatriyas who hurt everyone’ (kṣatriyeṣu na viśvāsaḥ kāryaḥ sarvopaghātiṣu; 12.137.14) and who ‘are friendly (saṃbhajanti) as long as they need something but neglect [their friends] when their purpose has been accomplished’ (kāraṇe saṃbhajantīha kṛtārthāḥ saṃtyajanti ca; 12.137.13). Another notable example from narrative literature is the speech of an ascetic to a king in Mahābhārata 12.83.23–36, which explains that it would be too dangerous for the ascetic to serve the king. For an example of a claim about the fundamentally violent and unreliable nature of the kṣatriya from the Buddhist literature, see Cullahaṃsajātaka, p. 345. Thus, there appears to be a convention in some fables and narratives in the Mahābhārata and Buddhist literature that kings are dangerous and not trustworthy. Therefore, sarvasattvāvasādaka would fit in well with a pattern of describing kṣatriyas.

In the context of this trend it may be significant that the compound pakṣisattvāvasādaka, (or the, in my opinion, preferable sarvasattvāvasādaka) ends in a causative of ava+sad. This is the opposite of ud-hṛ.³⁰ Forms based on ud-hṛ are commonly used to talk about help and saving in fable literature, including the Pañcatantra, Hitopadeśa, Kathāsaritsāgara, and the jātakas. Thus Palita also says to Lohita: ‘I will save you’ (ahaṃ tvām uddhariṣyāmi; 12.136.49). Probably here avasādaka is intended more specifically to indicate the ungrateful and unpredictable nature of the cat, which in turn could suggest that the cat is a kṣatriya in line with the stereotypes discussed in the previous paragraph.

The mongoose

The name of the mongoose is Harita (‘Yellow’), which suggests a vaiśya identity, according to the colour scheme given above.

The owl

If the four animals indeed represent the four varṇas, the owl must stand for a śūdra. His name in the CE text, however, looks more noble than that. He is called Candraka, ‘Moon’. This would be an appropriate name for a flying nocturnal predator. Several manuscripts from both Southern and Northern recensions read his name as ‘Caṇḍaka’ in verse 32 where it first appears (K1,2,4 D1 T2 G1). Moreover, Caṇḍaka, rather than Candraka, seems to be behind the unique reading Daṇḍaka found in Ś1. It may be noted that almost always such agreement between the main Kashmirian/Śāradā manuscripts and even a few Southern manuscripts are taken as sufficient for accepting the reading in the reconstituted text. Moreover, Caṇḍaka is a lectio difficilior. This word is not recorded in the major Sanskrit dictionaries (those of Monier Williams³¹ and Böhtlingk and Roth³²), but is recorded by PED as having the same meaning as caṇḍa which in turn means ‘violent, fierce, angry’ in both Sanskrit and Pali.

In verse 73 the owl is described as kṣudra (‘vile, lowly’). This sounds somewhat like śūdra,³³ but I am not the only one to make a connection between the two words. The nirukta (‘folk etymology’) analysis of the Pali word sudda (= Sanskrit śūdra) in the Aggaññasutta of the Dīghanikāya states: ‘Their conduct is fierce/greedy (ludda = Sanskrit rudra/raudra/lubdha), their conduct is vile (khudda = Sanskrit kṣudra), [therefore], oh Vāseṭṭha, they came to be called suddas’ (‘Luddācārā khuddācārā ti’ kho Vāseṭṭha Suddā, Suddā tv eva akkharaṃ upanibbattaṃ; Dīghanikāya 27.26³⁴). Moreover, the other purported characteristic of the original śūdras, namely being ‘fierce’ (rudra/raudra), is a synonym of Caṇḍaka, which I believe to be the original name of the owl. Most likely both lubdha and rudra/raudra were intended to be covered by the Pali word ludda.

Perhaps such word play between kṣudra and śūdra requires a Middle Indic antecedent of the Mārjāramūṣakasaṃvāda to really work. A similar antecedent is suggested by the variations of the owl's name. It would be most elegant if the owl's name was originally Prakrit Caṃdaka, which corresponds at the same time to Sanskrit Candraka and Caṇḍaka. Moreover, as I have already mentioned, the variants of the mouse's name, Palita/Phalita, could also be explained as a remnant of a Middle Indic spelling, Phalita. Finally, the story contains what appears to be an accusative absolute construction prabhātām śarvarīm in verse 24: ‘Having gone home [the hunter] sleeps happily and goes [back to the trap] when the night becomes light [that is, in the morning]’ (gṛhaṃ gatvā sukhaṃ śete prabhātām eti śarvarīm). Such a construction has been attested in some forms of Middle Indic, including Pali,³⁵ but it is not even mentioned in the grammars of Sanskrit³⁶ or epic Sanskrit.³⁷ All of this suggests that the Mārjāramūṣakasaṃvāda might ultimately be derived from a Middle Indic version, although the evidence is far from conclusive.

While the Śāntiparvan of the Mahābhārata usually emphasises the servile nature of the śūdras rather than suggesting that they are inherently wicked, one also finds verses that link śūdras with bad qualities that are generally in line with those attributed to them in the Aggaññasutta:

Those brahmins who delight in violence and falsehood, who are greedy (lubdha), who subsist by any trade, who are black and who have fallen from purity, have become śūdras.

hiṃsānṛtapriyā lubdhāḥ sarvakarmopajīvinaḥ

kṛṣṇāḥ śaucaparibhraṣṭās te dvijāḥ śūdratāṃ gatāḥ (12.181.13)

Another epithet applied to the owl is ‘moving at night’ (kṣapācara; 12.136.32). This further suggests a low class identity; perhaps the night is expected to evoke the black colour, which is the colour of śūdras in the colour scheme quoted above.

While some of the arguments advanced in this section are problematic, especially in so far as they rely on variant readings of the manuscript from different recensions, taken together they provide a strong case for identifying the four animals of the Mārjāramūṣakasaṃvāda with the four classes of society. A soteriological reading of the Mārjāramūṣakasaṃvāda does not necessarily require that this identification is correct. Perhaps only the mouse (teacher) and the cat (student) are significant for the allegory, although it would be more elegant if the other two animals also had a meaningful role to play.

The hunter

The description of the hunter

After discussing the identities of the four animals and some features of the story that invite a soteriological interpretation, we now return to the last remaining character of the story—the hunter.

The hunter appears at the beginning of the story to set a trap, into which the cat falls, and returns to it in the morning to find that the cat has just escaped. The hunter has no obvious allegorical meaning in the nīti version of the story. His possible identity for the mokṣa version is indicated by his name, description, and activity. I will suggest that the hunter represents death and spiritual bondage.

This is how the hunter is described:

In the morning the hunter (caṇḍāla) called ‘Obstacle’ (Parigha) appeared with a weapon in hand. He was ugly, blackish yellow (or black and yellow: kṛṣṇapiṅgala), *bald*,³⁸ *grey*, filthy, and of terrible sight. He had large buttocks, pointy ears, a large mouth, and he was surrounded by a group of dogs.

tataḥ prabhātasamaye vikṛtaḥ kṛṣṇapiṅgalaḥ

sthūlasphig vikaco rūkṣaḥ śvacakraparivāritaḥ

śaṅkukarṇo mahāvaktraḥ palito ghoradarśanaḥ

parigho nāma caṇḍālaḥ śastrapāṇir adṛśyata (12.136.9–10)

The following verse further specifies that the hunter ‘looked like a messenger of the [death god] Yama’ (yamadūtābha; 12.136.11).

This description is not of an ordinary hunter but of a supernatural being, such as a yakṣa or rākṣasa (‘demon’), and many of his visual features are reminiscent of the descriptions of other frightening characters. Several of his epithets bring to mind the descriptions of Ghaṭotkaca, the son of the hero Bhīma, and the rākṣasa (‘demon’) woman Hiḍimbā. Ghaṭotkaca is also described as mahāvaktra, ‘having a large mouth’, and śaṅkukarṇa, ‘with pointy ears’, in Mahābhārata 1.143.28 and as vikaca later in the same passage in 1.143.33.

Some of his epithets are elsewhere applied to more terrible characters. The Rod of Punishment (daṇḍa), also personified, is described as śaṅkukarṇa in 12.121.14. The colour ‘dark-yellow’ or colour combination ‘black and yellow’ (kṛṣṇapiṅgala) is also associated with a number of inauspicious or terrible personifications of cosmic forces. It appears in the description of the goddess Brahmahā (‘brahmin-slayer’) in Mahābhārata 12.273.11 and characterises the eyes of Rudra in 12.160.47. The same kṛṣṇapiṅgala, alongside vikaca, describes Time (kāla), personified in Mahābhārata 16.3.2. Colours are important in descriptions of Yama as well. In 3.281.8–9 he is said to be ‘pure black’ (śyāmāvadāta), ‘wear yellow clothing’ (pītavāsasa), and ‘have red eyes’ (raktākṣa). In later literature Yama's messengers are also said to be kṛṣṇapiṅgala (e.g. Garuda Purāṇa 2.18.22³⁹).

Although the story itself says that the hunter looks like a messenger of Yama, some of his features can be associated with Yama himself. These are the weapon (śastra) he carries, the dogs that accompany him, and the already mentioned ‘noose’ (pāśa). The highly ominous figure of Kali, who represents an unlucky throw of the dice, and Kaliyuga, the most miserable of the world ages, is also described as surrounded by dogs (śvabhiḥ parivṛtaḥ) in the additional passage number 9, found in some manuscripts after Mahābhārata 3.55.1.

Thus, the hunter shares epithets with minor demons (yakṣas and rākṣasas) as well as some of the most ominous forces mentioned in the Mahābhārata, such as Brahmahā, Kali, Daṇḍa, Rudra, and Yama himself. There is little doubt that he represents such forces, especially the force of death, and the story explicitly links him with Yama.⁴⁰ This aspect of the hunter's personality is not employed in the nīti version of the story in any way, and I propose that it is a remnant of the mokṣa version of the same story.

There is one more odd detail of the story which further supports the identification of the hunter with death. It is said that ‘the cat was bound in the trap although he was careful’ (mārjāras tv apramatto 'py abadhyata; 12.136.25). A typical feature of the pāśa of Yama and Māra is precisely that it cannot be escaped under normal circumstances.⁴¹

Death is personified in Ancient Indian literature not only as Yama, but also as the Buddhist Māra. Several studies have discussed the continuity between these figures as well as their differences.⁴² A common feature of relevance here is that both Māra and Yama bind people with a noose (pāśa ⁴³), which is the term used throughout the Mārjāramūṣakasaṃvāda to refer to the device in which the cat is caught (verses 37; 55; 73; 83; 84; 86; 91).

While both Yama and Māra bind their ‘victims’ with a noose (pāśa), only Māra's noose is usually equated with desire and spiritual bondage. Given that a number of features of the Mārjāramūṣakasaṃvāda invite a soteriological interpretation, I believe it would be more appropriate to understand the hunter as representative of Māra rather than Yama. Perhaps the hunter fuses aspects of both figures, as will be discussed below. Māra's noose is often used in Buddhist ascetic poetry, where it is usually equated with desire, which is at the root of spiritual bondage. Attributing the functions usually associated with the Buddhist Māra to the hunter, it becomes possible to meaningfully account for the bait and the trap he sets as well as for the subsequent escape from that trap.

On āmiṣa

The hunter of the Mārjāramūṣakasaṃvāda uses bait (āmiṣa) to attract animals to the trap. As noted by Boyd, it is typical of Māra to use āmiṣa as bait.⁴⁴ Indeed, the word āmiṣa, besides meaning ‘bait’, has strong connotations of the ascetic traditions of Ancient India, and I suggest that this word should be understood in both its ordinary and ascetic senses here.

Āmiṣa, which also means ‘flesh’ or any object of enjoyment, is often mentioned in the Mahābhārata, Buddhist, and Jaina texts as a fundamental driving force behind saṃsāra. In this usage, the term āmiṣa plays a similar role to terms like kāma, tṛṣṇā, sneha, and so on, and should perhaps be translated as ‘desire’. In the same textual sources, the adjective nirāmiṣa (without āmiṣa) is used to describe sages or the mental states of sages who have overcome worldly enjoyments. The PED meaning for nirāmiṣa is ‘having no meat or prey; free from sensual desires, disinterested, not material’.

Hunter's name

The hunter's name Parigha, ‘Obstacle’, would be fitting for Māra, since one of Māra's key activities is ‘to set up obstacles (āvaraṇa) and bring about interruption (antarāyam upasaṃharati)’.⁴⁵ Fitzgerald translates the hunter's name as ‘Cross-bar’,⁴⁶ which is the standard meaning of the word parigha in the Mahābhārata. The figurative meaning of the Pali equivalent paligha as ‘obstacle’, however, is only recorded by the PED, whereas the Sanskrit dictionary of Monier-Williams notes the figurative meaning for parigha as ‘obstacle’, beginning with the Raghuvaṃśa (circa fifth century ce).

Escaping death

Both Brahmanical and Buddhist literatures present narratives about escape from Yama and Māra respectively. Nichols has helpfully discussed such stories from the Brahmanical tradition, comparing them with Buddhist understandings of Māra.⁴⁷

There are a number of stories in Indian literature that depict escape from death personified. The most famous of these stories is the Sāvitrī story in the Mahābhārata. A similar motive is also explored in the Buddhist jātakas—the Cullahaṃsajātaka of the Pali tradition (no. 533) and the Mṛgarājajātaka ⁴⁸ of the Mahāvastu.⁴⁹ It is not possible to provide a detailed comparison of the stories, interesting as that may be. In each of them a character caught with the noose of Yama (in the Mahābhārata) or a hunter (in the jātaka versions) is saved when another character displays exceptional love or devotion, as well as mastery of dharma, greatly impressing Yama or the hunter, which results in the character being released. The primary message of these stories does not appear to be soteriological, not even in the Buddhist versions. The obvious virtue illustrated by the Sāvitrī story is her being pativratā, ‘devoted to her husband’. The same applies to the Mahāvastu version, whereas in the Cullahaṃsajātaka version bhakti, ‘devotion’, is displayed by a general towards his king. Additionally, the focus of the stories is on the remarkable understanding of dharma on the part of the saviour characters.

In addition to these, there are stories in which escape from Yama is attained through the grace of Viṣṇu or Śiva. Nichols notes two such episodes in Purāṇic literature.⁵⁰ I would add that there is also one episode in the Mahābhārata in which a dead son is brought back to life through the grace (prasāda) of Śiva (12.149.114). It is doubtful whether escape from death in the Purāṇic accounts related by Nichols should be understood as escape from saṃsāra—almost certainly it should not be so understood in the Mahābhārata episode. The characters of these stories return to life but not as spiritually liberated sages. The dead son in Mahābhārata 12.149 does not obtain a permanent state of immortality, but a full lifespan of a hundred years (12.149.109). Such stories illustrate the power of bhakti, devotion to a deity, and the miraculous power of Viṣṇu or Śiva.

Although neither of the stories so far mentioned in this section seems to imply that escape from death means escape from saṃsāra, it would be natural to identify the two. One of the epithets of the highest reality (brahman) in the Mahābhārata is ‘the immortal’ (amṛta) or the immortal state/place (amṛta pada) (for example, 12.199.32, 12.208.26; 12.209.19). The same epithet, amṛta pada, is also found in the Buddhist tradition, which uses the term amṛta to refer to the state in which there is no longer birth and death, namely, nirvāṇa.

There is, however, one well-known passage in Brahmanical literature which links Yama to aspirations for immortality, understood as spiritual liberation. In the Kaṭha Upaniṣad a brahmin youth Naciketas entertains a dialogue with Yama, during which Yama tells Naciketas how to reach complete peace (śāntim atyantam eti), crossing over both birth and death (tarati janmamṛtyū; 1.17⁵¹). As noted by Nichols, elements of this Upaniṣad can be interpreted as ‘reinforcing orthodox sacrificial practices, while incorporating philosophical notions about transmigration advanced by renouncer groups’,⁵² and the portrayal of Yama in this text incorporates standard Brahmanical tropes with features of the Buddhist Māra. Of particular interest is Nichols’ observation that in the Kaṭha Upaniṣad Yama undertakes a typical activity of Māra when he attempts to tempt Naciketas with worldly goods such as wealth, power, and beautiful girls.⁵³

The example of Kaṭha Upaniṣad demonstrates that the opposition between Brahmanical Yama and Buddhist Māra is not always clear-cut in Ancient Indian literature. I believe that the Mārjāramūṣakasaṃvāda offers a similar fusion of typical aspects of the two deities.⁵⁴

Māra and saviour animals in Buddhist jātakas

The earlier ‘mokṣa’ version of the Mārjāramūṣakasaṃvāda, which is proposed in this article, requires, among other things, that the hunter represents the agency of Māra and that an escape from his pāśa represents spiritual liberation. Some evidence in support of such an interpretation of the story has already been provided. The likelihood of the existence of such a mokṣa version of the story is increased by the existence of similar Buddhist narratives that depict an escape from a hunter or a fowler who is traditionally identified as Māra.

If one looks at the Pali jātakas, there is little evidence to support a hypothesis that the early Buddhist communities would have understood the Mārjāramūṣakasaṃvāda or similar narratives as soteriological allegories. I have already mentioned, for example, the Cullahaṃsajātaka which, despite resembling the Mārjāramūṣakasaṃvāda in several ways, is not given a soteriological interpretation by the tradition.

This section will provide an overview of the Buddhist jātakas that involve animals and escape. While not all of these jātakas merit a soteriological interpretation or are given such a reading by the tradition, I will note that some jātakas found in the Mahāvastu reinforce my reading of the Mārjāramūṣakasaṃvāda.

The Buddhist jātaka collections include a number of stories where one animal saves another animal or human. When encountering such a motive in a Buddhist text, the possibility of interpreting it in a soteriological way naturally presents itself, especially if a character is saved from something that is frequently likened to saṃsāra in ascetic literature, such as a trap made from pāśa, as in the Mārjāramūṣakasaṃvāda, or, for example, a violent river, as in the Rurujātaka (no. 482). On the other hand, helping others is basic to human life, and stories about helping or saving others provide a good basis for talking about virtues and vices generally.

The role of animals in early Buddhist literature has been explored in a monograph by Reiko Ohnuma.⁵⁵ Helpfully, this book also includes a chapter on ‘Animal Saviours’, which deals with stories that are comparable to the Mārjāramūṣakasaṃvāda.

One of the key trends discovered by Ohnuma is that in a number of jātaka stories a human or an animal saved by another animal displays ingratitude towards its saviour.⁵⁶ The Mārjāramūṣakasaṃvāda is an interesting answer to this trend, since in the second part of the story the mouse is suspicious of the cat and refuses to become his minister after the cat is saved and the predators are gone. The final part of the Mārjāramūṣakasaṃvāda reflects the same discourse, using some key vocabulary that often appears in these stories, such as kṛtajña ‘recognising a favour, grateful’ and ud-hṛ ‘to save’. It contrasts with the Buddhist jātakas, since the cat is not even given a chance to display his ingratitude to the cautious mouse.

Another key finding of Ohnuma is that several

jātakas also [in addition to the Nigrodhamigajātaka] involve a saviour animal who converts the king through his exaggerated virtue and then uses this leverage to win the safety of all other animals—including the deer of the Nandiyamiga Jātaka (No. 385), the deer of the Ruru Jātaka (No. 482), the crow of the Supatta Jātaka (No. 292), the crow of the Kāka Jātaka (No. 140), and the dog of the Kukkara Jātaka (No. 22).⁵⁷

The Mārjāramūṣakasaṃvāda appears to be in conversation with this trend as well, as long as it is accepted that the cat is identified with a potentially dangerous king. In the Mārjāramūṣakasaṃvāda, however, there is no ‘conversion’ of the king, or winning the safety of others. The mouse manages to keep him in check by making cat's release dependent on the mouse's help.

It seems quite clear that both types of ‘animal saviour’ jātaka identified by Ohnuma are primarily concerned with topics other than spiritual liberation. There are, nevertheless, some jātaka stories that involve a saviour animal, but do not present any aftermath to the escape and therefore do not conform to either of the narrative themes identified by Ohnuma. It would be worth exploring whether such stories can be interpreted in a soteriological way and whether they have been given such an interpretation by Buddhist tradition(s). Within the Pali jātaka collection, such a story is, for example, the Mitacintijātaka (no. 114) and the Kakkarajātaka (no. 209), which finds a counterpart in the Śakuntakajātaka number 2 of the Mahāvastu. The Pali tradition offers no soteriological interpretation to such jātakas. Moreover, Māra plays virtually no role in the entire Pali jātaka collection, being identified with a character in just one among 547 stories. When a mean or harmful character appears in a Pali jātaka, he is usually identified as a previous incarnation of Buddha's unfaithful disciple Devadatta.⁵⁸ In other words, the Pali jātaka collection offers little support for the interpretation of the Mārjāramūṣakasaṃvāda that I am proposing.

The situation is reversed, however, in the Mahāvastu. There Devadatta appears in just one story, while Māra is identified with the villain in all the others. The Mahāvastu contains a sequence of jātakas which illustrate how Māra unsuccessfully tried to overcome the bodhisattva in his previous lives when the latter was born in animal form. These are the Śakuntaka jātakas number 1 and 2, Kacchapajātaka, and Markaṭajātaka.⁵⁹ There are three other stories in the Mahāvastu that relate how the bodhisattva overcame Māra as an animal in a previous life. These are the Vṛṣabhajātaka, and Vānara jātakas numbers 1 and 2. Some of these jātakas are also found in the Pali tradition, where they are usually linked with Devadatta rather than Māra.

While all of these jātakas depict an escape from danger presented by Māra, both Śakuntaka jātakas in particular should be compared to the Mārjāramūṣakasaṃvāda because in these two jātakas Māra is a fowler who catches his victims with a pāśa.

The villain of the Śakuntakajātaka 1 is a fowler (śākuntika) who catches birds with snares (jāla) and traps (pāśa), feeds them until they are fat, and then sells them for slaughter. The Buddha, who narrates the story, identifies the fowler with Māra. The bodhisattva himself is a bird called Paṇḍitajātiko who is caught by the fowler but observes what becomes of other birds and begins fasting. This makes him undesirable to potential buyers and eventually he is allowed to escape back to the forest.

This story employs the double meaning of the word mokṣa, just as the Mārjāramūṣakasaṃvāda does. In the introduction the Buddha makes a statement about mokṣa as spiritual liberation in relation to the time he spent in the ‘forest of austerities’ (tapovana), practising physical asceticism before discovering the uniquely Buddhist dharma:⁶⁰ ‘It was not just at that time that I practiced austerities for the sake of liberation (mokṣa)’ (na bhikṣavo etarhi eva maye mokṣābhiprāyeṇa duṣkaraṃ cīrṇaṃ; 2.241). In the conclusion he says: ‘At that time also I carried out austerities for the sake of escape (mokṣa) from the cage of that same Māra, who was the fowler’ (tadāpi mayā etasya mārasya śākuntikasya paṃjarāto mokṣābhiprāyeṇa duṣkaraṃ cīrṇaṃ; 2.243).

In the second Śakuntakajātaka of the Mahāvastu Māra is again a fowler and the bodhisattva is again a bird, but this time he is a leader of a pack of birds. The bodhisattva escapes the fowler and saves his flight of birds by recognising the fowler's disguise as a tree.

Thus, there are two episodes in the Mahāvastu in which escaping from a pāśa set by a fowler is identified with overcoming Māra, and in the second of them, other beings are saved from pāśa as well. This demonstrates that at least some Buddhist communities employed the possible allegorical meaning of stories about escape from hunters. Such stories employ vocabulary that is used in Buddhist ascetic poetry to talk about the activity of Māra, bondage, and escape from saṃsāra. The presence of such stories in a Buddhist tradition increases the likelihood that the Mārjāramūṣakasaṃvāda, too, was originally understood in a soteriological context.

I do not believe that these stories about escape were necessarily composed with a soteriological meaning in mind, although some might have been. Most of the stories mentioned above, including the Śakuntaka stories, extol wisdom or cleverness, termed variously as cintā, buddhi, kuśalatā, or prajñā. The name of the bodhisattva in the Śakuntaka stories is Paṇḍitajātiko, ‘Wise by Nature’. It seems quite certain that some, if not all, of these stories were composed to illustrate the virtue of wisdom, most probably in a secular context. Different Buddhist communities used these stories for their own purposes. While the Pali tradition prefers to dwell on the Devadatta as the villain (where one appears), the Mahāvastu, associated with the Mahāsāṃghikas, prefers to identify the villain with Māra, giving such stories a soteriological edge.

It is difficult to say how such stories were understood when they first became a part of Buddhist literature and whether there was even a single main interpretation. The emphasis on Devadatta in the Pali tradition strikes me as odd, while the identification of at least the fowlers with Māra seems more natural. We have already seen that identifying hunters with death is a trend in Indian narrative literature, and their use of bait and snares (pāśa) further suggests a link with Māra.

The Kakkarajātaka of the Pali canon and the Śakuntakajātaka no. 2 of the Mahāvastu

It may be worth briefly looking at the Kakkarajātaka (no. 209) which is the Pali counterpart of the second Śakuntakajātaka of the Mahāvastu. The verse element of both jātakas is related, although it is not identical, while the surrounding narrative is somewhat different. Both stories describe how a bird recognises a fowler disguised as a tree and thereby escapes from him.

Although a soteriological interpretation of this story is possible and given to the Mahāvastu version, as discussed above, no such explanation is offered by the Pali tradition. While in the Mahāvastu the context for relating the group of Māra stories was Māra's attempts to tempt the ascetic bodhisattva, in the Pali canon the Kakkarajātaka is told in relation to a monk who knows what food is good for him.

Following the trend noted above, the fowler (sakuṇaluddaka) is identified as Devadatta. Surprisingly, the bodhisattva is not identified as the bird who escapes the fowler, but as a tree-dwelling spirit (yakṣa) who does nothing whatsoever but simply observes the scene. Such inactivity on the part of the bodhisattva does not allow the reader to learn about the virtues of the bodhisattva himself and, moreover, it disturbs the characteristic Devadatta-bodhisattva antagonism.

The jātaka only has two verses. The second verse makes it clear that the story was originally primarily about the bird and the hunter, and its imagery invites a soteriological interpretation. That the verse does not exactly match the narrative is something that cannot be explained at the moment.

This old skilful partridge has come having broken the net of horse-tail snares (vālapāsa). He departs and speaks.

purāṇakakkaro ayaṃ bhetvā pañjaram āgato,

kusalo vālapāsānaṃ apakkamati bhāsatīti

It hardly needs stating that breaking the net or cage (pañjara) suggests spiritual liberation, but just to be sure, we note that the same phrase appears in Buddhavamsa 23.2:

Having stopped all existence, having reached the perfection of practice, like a lion having broken a cage, he obtained unsurpassed awakening.

ugghāṭetvā sabbabhavaṃ cariyāpāramiṅ gato

sīho va pañjaraṃ bhetvā patto sambodhim uttamaṃ ⁶¹

The second verse of the Kakkarajātaka is also worth quoting, because it contains possible parallels with the description of the hunter in the Mārjāramūṣakasaṃvāda. Here is what the bird says, apparently addressing the fowler disguised as a tree:

I have seen assakanna and vibhitaka trees in the forest, but they are not capable to do as you, tree, are.

diṭṭhā mayā vane rukkhā assakaṇṇavibhīṭakā,

na tāni evaṃ sakkanti yathā tvaṃ rukkha sakkasīti

What I have just provided is a conventional translation of the verse. After reading the description of the hunter in the Mārjāramūṣakasaṃvāda, a different reading of the verse can be made, according to which the bird recognises the fowler to be a demonic character, not unlike the Parigha of the Mārjāramūṣakasaṃvāda. The word rukkha could correspond not only to Sanskrit vṛkṣa (‘tree’), but also to Sanskrit rūkṣa (‘dirty, unpleasant’), which is one of the adjectives describing the hunter in the Mārjāramūṣakasaṃvāda. Moreover, the choice of the species of trees that are named does not seem arbitrary. The first part of assakaṇṇa (Sanskrit aśvakarṇa, ‘horse-ears’) could also be related to Sanskrit aśri, ‘sharp’. In that case assakaṇṇa would be a synonym of śaṅkukarṇa (‘pointy-eared’), which also describes Parigha. The name of the second tree, vibhīṭaka (Sanskrit vibhītaka), means ‘frightening’, which is also in line with the description of Parigha.⁶²

If both Mārjāramūṣakasaṃvāda and Kakkarajātaka refer to a stereotypical description of a supernatural hunter/fowler, in the context of escaping from the pāśa of the hunter, there is some reason to suppose that both episodes might be related, albeit distantly. Moreover, both of them delight in puns and playful language.

One gets the impression that the Pali tradition deliberately avoids a soteriological interpretation of these jātaka verses. It is not difficult to imagine the reasons for this. The jātakas tell us about the lives of the Buddha before he became a buddha, and therefore it would be inappropriate for a jātaka story to illustrate how the Buddha escaped Māra in the past. This appears not to have been a problem for the communities that transmitted the Mahāvastu. There are reasons to believe that the shared ancestor of the Pali Kakkarajātaka and the second Śakuntakajātaka of the Mahāvastu did address the topic of spiritual liberation. Such an interpretation is invited by the verse material and remains explicit in the Mahāvastu version.

This hypothesis is supported by the observation that there is no hesitation in identifying a hunter with Māra in the Makkaṭasutta of the Saṃyuttanikāya,⁶³ which relates a tale about a hunter who sets bait and manages to catch some monkeys, while others escape. The monkeys who manage to escape Māra's bait are identified with monks who practise mindfulness and do not engage with the sensual realm, thus not giving Māra an opportunity. Furthermore, the example of the Makkaṭasutta demonstrates that the possibility of identifying a hunter in an escape story with Māra seemed natural to the Pali tradition as well, as long as the story is not labelled as a jātaka.

A survey of Buddhist narratives about escaping hunters or fowlers reveals that there is a tendency to identify them with Māra in the jātakas of the Mahāvastu as well as in the Makkaṭasutta of the Pali canon. Nevertheless, this possibility of interpretation is always avoided in the jātakas of the Pali canon.

I now return to the Mārjāramūṣakasaṃvāda to comment on two aspects of the story that can be interpreted as belonging to the soteriological allegory. The discussion of these points is secondary to the main argument of the article. In other words, I believe that the identification of the mouse, cat, and hunter with a teacher, student, and death/Māra, respectively, holds, regardless of whether the proposed interpretation of these additional details is correct or not.

A delay in liberation

A peculiar feature of the Mārjāramūṣakasaṃvāda is the hesitation on the part of the mouse when it comes to releasing the cat from the trap. In the nīti version of the story, it is explained that the mouse hesitates in order to wait for the hunter, so that the cat is released just before he arrives, leaving no time for the cat to eat the mouse.

It is possible that this episode of the story was added at a later stage and has no bearing on the mokṣa version of the story. It should be noted, however, that the postponement of the cat's mokṣa could be read in light of the doctrine that a king's primary duty is to protect his subjects, especially brahmins, and he is only allowed to pursue mokṣa in old age, as it were, in retirement. This idea is put forward, for example, in Mahābhārata 12.63.18–22. The approach of the hunter would signify that the cat is becoming old.

Buddhist narratives of Brahmanical decline

A number of reasons have been put forward to argue that the Mārjāramūṣakasaṃvāda is based on an earlier tale about a brahmin who liberates a king in exchange for protection and patronage. Supposing that this hypothesis is found convincing, further questions arise regarding the authorship of the story, the intention of the authors, their possible religious affiliation, if any, and so forth. It is not possible to treat all of these questions here, but there is one, possibly significant, element of the story that has not yet been touched upon and which can provide a basis for speculating on some of these issues.

The mouse only gets into trouble in special circumstances. When the cat falls in a trap, the mouse comes out fearlessly:

While looking for food he soon saw the bait (āmiṣa), ascended the trap and started eating the bait, inwardly laughing at his bound enemy [i.e., the cat], while himself staying at the top of that [trap]. While intent on the bait he happened to look down and saw another terrible enemy.

bhakṣaṃ vicaramāṇena nacirād dṛṣṭam āmiṣam

sa tam unmātham āruhya tad āmiṣam abhakṣayat

tasyopari sapatnasya baddhasya manasā hasan

āmiṣe tu prasaktaḥ sa kadā cid avalokayan

apaśyad aparaṃ ghoram ātmanaḥ śatrum āgatam (12.136.27cd–29)

The phrase bhakṣaṃ vicaramāṇena (from vi+car) in this passage immediately reminds one of the usage of word bhaikṣa, sometimes spelled bhaikṣya,⁶⁴ with the root car to designate the practice of wandering mendicancy or going on a begging round. Similar meaning is carried by the compound bhaikṣacaryā, sometimes spelled bhaikṣyacaryā, and by the corresponding Pali (bhikkhāya car, bhikkhācariyā, and so on) and Prakrit constructions and compounds. Perhaps the allegorical meaning of this is that the mouse starts off as a wandering ascetic (śramaṇa), and becomes tempted by wealth (āmiṣa). The connotations of āmiṣa in ascetic discourses have been discussed above. As long as the mouse has nothing, he has nothing to fear, but as soon as he acquires wealth, he finds himself in danger, because that wealth can be forcefully taken away by others. The mouse solves the problem by transforming from a śramaṇa into a powerful king-advising brahmin. Reading such a transformation into the present story concurs with a claim found in several Buddhist texts that the ‘original’ brahmins were similar to śramaṇas in lifestyle, while contemporary materialistic brahmins have failed to live up to this ideal. Such a ‘history’ of brahmins is told in the Aggaññasutta of the Dīgha Nikāya as well as in the Brāhmaṇadhammikasutta of the Sutta Nipāta. The latter is especially useful for our present purpose as it describes how the originally modest and virtuous brahmins started coveting the luxuries of a royal life and came up with Vedic hymns and sacrifices to obtain wealth from kings.

The interpretation of Palita‘s life along similar lines is also supported by the usage of the phrase ‘āmiṣe tu prasaktaḥ’ to describe the mouse. This can also mean ‘attached to material things’. The opposite of this—‘he has no attachment to material things’ (nāmiṣeṣu prasaṅgo ’sti; 12.152.21)—is used elsewhere in the Āpaddharmaparvan to describe a virtuous person.

These observations would further support the possibility that the Mārjāramūṣakasaṃvāda was adapted from a mokṣa-themed tale that was satirical—or at least critical of the situation that it alludes to. It appears to describe the ‘fall’ of an imagined ancient ideal brahmin from his natural state of śramaṇahood as he becomes interested in the material world (āmiṣa). Perhaps, through a rich donation, he is transformed from the state of a bhikṣu to the state of a rich landowner or the like. He now needs protection and finds the strongest ally possible—the king. The brahmin promises to provide spiritual liberation (mokṣa) to the king, but first he makes the king live out a lifetime in the material sphere, protecting the brahmin. Meanwhile, the brahmin lives safely under the king's protection: ‘He calmly slept in the bosom of the cat like in the bosom of his parents’ (mārjārorasi visrabdhaḥ suṣvāpa pitṛmātṛvat; 12.136.81).

In contrast to the Buddha, who teaches the dharma out of compassion so as to help others cross the ocean of saṃsāra, the mouse does it for his own benefit, so that he himself also ‘crosses over’. There seems to be no real conversion of the king from his original violent stance to a more peaceful state.

Conclusion

This article has proposed a new interpretation of the Mārjāramūṣakasaṃvāda in the Mahābhārata and discussed stories about escape from death in the Mahābhārata and early Buddhist literature, some of which bear a thematic resemblance to the Mārjāramūṣakasaṃvāda.

The Mārjāramūṣakasaṃvāda of the Mahābhārata abounds in terms and phrases that, in addition to their everyday meanings, have a technical or typical usage in the ascetic traditions of Ancient India. Moreover, the conversation between the cat and the mouse resembles that of a teacher and a disciple rather than a discussion of a possible alliance between two kings. All of this suggests that the Mārjāramūṣakasaṃvāda is based on an earlier version of the same tale which was composed as an allegory about an alliance between a brahmin and a king, in which the brahmin liberates the king in exchange for protection. The details of such an alliance corresponds to religious developments recorded elsewhere in the Mahābhārata. There are some reasons to believe that the other two animals in the story were intended to represent the other two varṇas of the society, namely, vaiśya and śūdra. The hunter of the story can be identified with the Buddhist Māra, who represents both death and desire, which is at the root of spiritual bondage. Some of the arguments made in this article, especially those related to the varṇa identities of the animals, rely on linguistic probability and should be taken with a grain of salt. It is the cumulative evidence of such arguments that makes them compelling to explore.

The nīti interpretation of the Mārjāramūṣakasaṃvāda, which is advanced in the Mahābhārata itself, does not account for some of the story's key details as there are a few contradictions in the plot and descriptions of the characters. The proposed mokṣa interpretation of the same story finds meaning in most of the details of the story and explains how the contradictions might have arisen. It is charitable towards the author(s) of the story as it assumes that the story was carefully crafted to relish in double meanings, word plays, and to have an allegorical meaning as a whole. This would make the Mārjāramūṣakasaṃvāda a unique piece of literature that breaks the conventions of the fable genre to provide a socio-religious commentary by the means of an allegory.

Some features of the story suggest that it was intended to be satirical. While the Buddha is the compassionate boatman who helps others cross over, the mouse gets himself over as well. Dangerous kings are converted to non-violence in the Buddhist jātakas, but no such transformation happens to the cat in the Mārjāramūṣakasaṃvāda. While the Buddha overcomes the temptation (āmiṣa) set by Māra, the mouse is attached (prasakta) to the bait (āmiṣa) set by the hunter and manipulates the cat for protection. It is this calculated self-interest on the part of the mouse, as he manipulates the cat, that provides a link between the mokṣa reading of the story offered in this article and the nīti reading provided in the Mahābhārata.

The findings of the present study hint at a complex text history for the Mārjāramūṣakasaṃvāda which cannot be fully disentangled at present. It assumes at least two versions of the story—an earlier mokṣa layer and a later nīti reworking—but there may have been more editorial steps before its present state was reached. There are some grounds to believe that the tale was translated from a Middle Indic original or influenced by a Middle Indic version, although the evidence is inconclusive. In a number of instances I have emphasised terminological and thematic continuity with the Buddhist literature. Most of these elements, however, such as the metaphor of crossing over, can also be found in ascetic discourses in the Mahābhārata itself. Moreover, the alliance between a king and a liberation-granting teacher seems to be a religious development described primarily in Brahmanical literature.

To assess the plausibility of a soteriological interpretation of the Mārjāramūṣakasaṃvāda I have discussed structurally similar stories about escape. The ascetic poetry of Ancient India employs figurative language to talk about spiritual liberation, using such examples as ‘escape from a net’, ‘cage’, or the like. Sometimes, it mentions the noose (pāśa) of Māra as something to be escaped from. It would seem natural to attribute a soteriological interpretation to other Ancient Indian narratives about an escape from the pāśa of a hunter or from death.

In general, two types of stories about escape from death can be identified in Ancient Indian literature. The first tells how a faithful character saves their friend from the noose of Death or a hunter who represents death. Nevertheless, this escape is only temporary and does not appear to have soteriological significance in such stories. To this group belongs, for example, the famous Sāvitrī story of the Mahābhārata as well as the Pali Cullahaṃsajātaka and the Mṛgarājajātaka of the Mahāvastu.

The second type of story extols the virtue of cleverness which allows the main character to escape death and sometimes to help others escape as well. Only some of these jātakas involve a character who lays out a pāśa. Of this group I have discussed the Śakuntaka jātakas 1 and 2 from the Mahāvastu and the Pali Kakkarajātaka which is related to Śakuntakajātaka 2. All of these jātakas talk about a bird who escapes the pāśa of a fowler, and the description of the fowler in the Kakkarajātaka resembles the description of the hunter in the Mārjāramūṣakasaṃvāda. The fowler is identified with Māra in both of the Śakuntaka jātakas, but not in the Kakkarajātaka.

This brief survey reveals that not every story about escape from death, or even from the pāśa of a hunter or a fowler, has been traditionally understood in a soteriological way. Nevertheless, both the Mahāvastu and the Pali canon include stories in which the escape from a trap set by a hunter or a fowler is understood as a representation of spiritual liberation and an escape from Māra. This increases the likelihood that the cat's escape from the pāśa of a hunter in the Mārjāramūṣakasaṃvāda was also originally understood to represent spiritual liberation (mokṣa).

While Māra often figures as a character in the jātakas of the Mahāvastu, the Pali tradition identifies jātaka villains as the previous births of the Buddha's unfaithful disciple Devadatta. I have suggested that this must be a peculiarity of the Pali jātaka tradition, especially since the Pali tradition does not hesitate to identify a hunter with Māra in the Makkaṭasutta, which is not classified as a jātaka tale.