Story and History

particular Nanda, was of questionable birth. A further item not mentioned in the excerpt but found in many sources is that Candragupta was in fact a scion of the Nanda dynasty, though (again) of lowly birth.

The extant stories about the epoch preceding Candragupta’s coup d’état tell about a time when the idea of a world-encompassing empire became current, along with “a new type of unscrupulous and ambitious ruler” (K^ULKE &R^OTHERMUND 2002:56) who balks at nothing to achieve dominance. Parricide was the order of the day, or at least that is what later narratives would have us believe: the bastard Mahāpadma Nanda allegedly assassi-nated the previous king who may have been his father, then elimiassassi-nated the legitimate heirs to the throne. Before the Nandas, four kings of the Śaiśunāga dynasty are said to have attained the throne by killing their fathers (ibid.). After the Nandas, we have Candragupta:

another bastard who took his father’s throne after disposing of his rivals in inheritance;

and another few generations later, Puṣyamitra assassinated the last Maurya ruler and founded the Śuṅga dynasty.

Attempts have of course been made to arrive at some sort of “objective” historical truth by analysing where various accounts agree and where they are in conflict, but the issue I intend to explore here is the interrelation of the narratives themselves, rather than what actually happened. Indians (in a generalised, Orientalistic sense) have been accused of making up stories in place of factual history for at least about a millennium.² Indeed, one of the basic Sanskrit words for “history,” itihāsa³ (literally, “indeed it was thus”—a phrase to open or conclude a narrative) is in fact a technical term for a literary category:

that of the great epic the Mahābhārata (and, more equivocally, the Rāmāyaṇa). It is very often paired with another genre name, purāṇa (literally, “ancient”), meaning a corpus of mythological texts, bardic in origin and redacted into written works from about the mid-dle of the first millennium CE.⁴

Sheldon Pollock has proposed that “the general absence of historical referential-ity in traditional Sanskritic culture” (P^OLLOCK 1989:607) may be connected to the ratiocina-tion that the reason why the Vedas must be accepted as timeless truth is that they have no author and no historical referentiality. Thus, by analogy, any knowledge claiming to be true must be similarly detached from the factual world. History, Pollock concludes, “is not simply absent from or unknown to Sanskritic India; rather it is denied in favor of a model of ‘truth’ that accorded history no epistemological value or social significance” (ibid. 610).

It is also P^OLLOCK (2003:57) who points out that historical referentiality is by no means unknown in Sanskrit literature, and some literary theorists, such as Bhāmaha (7^th century CE), do make a distinction between factual and made-up stories. What was,

2 In the words of al-Bīrūnī (11^th century CE): “Unfortunately the Hindus do not pay much attention to the historical order of things … and when they are … at a loss, not knowing what to say, they invariably take to tale-telling” (SACHAU 1888:10–11).

3 The exact same word, इितहास, is not just one but the basic word for “history” in modern standard Hindi.

4 See also page 97.

ever, more important to (at least a majority of) Sanskrit authors and theorists, was to com-pose aesthetically pleasing and morally uplifting stories. Thus, Bhoja (11^th century) offers the following advice:

If one were to compose a literary work on the basis of a story just as it is found to exist in narratives of the way things were [itihāsa], it could come about that one character, though acting with all due propriety, might not only fail to attain the desired result but might attain precisely the result he does not desire; whereas an-other character, though acting improperly, might attain the result he does desire. In these cases, emendation must be made in such a way that the character acting properly is not denied the result he seeks, whereas the other not only should fail to attain his desire but should also attain what he does not want.⁵

This instruction appears particularly relevant to the Mudrārākṣasa, in which a character acting improperly—Rākṣasa, who from a misguided sense of loyalty deserts Brāhmaṇical civilisation and dedicates his talents to the service of barbarians threatening the capital and the legitimate (if barely so) ruler—attains not his desire, but something he does not want: reintegration into the orthodox fold. Whether or not Viśākhadatta had such clearly enunciated guidelines in mind while composing his drama, it is in my opinion beyond question that for his purposes history was malleable. If he was at all interested in

“what actually happened” in the 320s BCE, he was certainly more interested in making his own point and weaving his own tapestry, incorporating existing yarns wherever they suited his purpose, but changing known stories wherever needed without a second thought. On the following pages, therefore, I shall investigate the stories of Candragupta’s accession to the Nanda throne without regard to their factuality. To quote P^OLLOCK (2007:379) yet again, “Perhaps, instead of assessing whether Indian texts are history or myth, we might ask whether the texts themselves invite us to transcend this very dichot-omy.”

The History of a Story

The Mudrārākṣasa begins in medias res as far as the story of the Nandas and Candra-gupta is concerned. While Cāṇakya’s soliloquy at the start of the play does reveal numer-ous details, the spectators are left to gather most of what had happened prior to the action from a scattering of small hints dropped throughout the play. Many of these hints are in fact so brief that I cannot but assume that Viśākhadatta must have known and, what is more, must have expected his audience to know, a fairly detailed story about the end of

5 ŚP cited and translated in POLLOCK 2003:58.

the Nanda reign.⁶ Whatever he did say about the antecedents would have served as point-ers to help the audience put the story in context, while whatever he left unsaid would have been supplied from their contextual knowledge.

The play indicates beyond doubt that a dynasty called Nanda had reigned in Pāṭaliputra; that at least some of these rulers did not perform their duties satisfactorily;

that they had insulted Cāṇakya by publicly dragging him down from a seat of honour, whereupon he untied his topknot and vowed to exterminate them; that Cāṇakya decided to enthrone Candragupta, who was a Nanda descendant in some sense, yet not a member of the dynasty; and that to conquer the city they had made alliance with the barbarian king Parvataka, promising him half the kingdom in return for his military aid.

The search for a particular story behind the Mudrārākṣasa is far from new. The very earliest dateable work that refers to it by title—the Daśarūpāvaloka of Dhanika, com-posed in the late 10^th century—makes an attempt to ascribe a source to it. The apropos of this is an instruction in the Daśarūpa that a prospective author should first study texts such as the Rāmāyaṇa and the Bṛhatkathā, and then proceed to compose an interesting story of his own. The implication, at least as Dhanika understands the Avaloka, is that the core of the plot may be derived from an existing literary (or quasi-historical) work, embellished with the author’s fancy, just as Bhoja opined in the passage quoted above.

Dhanika claims that our play is based on the Bṛhatkathā,⁷ and Parab’s edition of the Daśarūpa even has a purported Bṛhatkathā citation⁸ at this point, thought to have served as the kernel which Viśākhadatta elaborated his play. The stanzas cited, however, come from the Bṛhatkathāmañjarī of Kṣemendra, ⁹ a retelling of the Bṛhatkathā that in fact postdates the Daśarūpāvaloka. Therefore, as a number of scholars¹⁰ have pointed out, the citation must be a later interpolation in the Avaloka. As a matter of fact, it is only found in one manuscript of the Daśarūpāvaloka,¹¹ and it has even been proposed that the entire pas-sage surrounding the citation may be spurious.¹²

6 I agree with TRAUTMANN (1971:42) in that “The Mudrārākṣasa was composed for a limited and highly

sophisticated audience whose members we must suppose to have been thoroughly familiar with arthaśāstra through their education and with intrigue through experience.” They must have been likewise well acquainted with stories about the kings and wars of yore.

7 Avaloka ad. Daśarūpa 1.68 (p. 34): tatra bṛhatkathāmūlaṃ Mudrārākṣasam.

8 cāṇakyanāmnā tenātha śakaṭālagṛhe rahaḥ| kṛtyāṃ vidhāya sahasā saputro nihato nṛpaḥ|| yoganandayaśaḥśeṣe pūrvanandasutas tataḥ| candraguptaḥ kṛto rājā cāṇakyena mahaujasā|| iti bṛhatkathāyāṃ sūcitam.

9 Bṛhatkathāmañjarī 1.2.216–217, cāṇakyanāmnā tenātha śakaṭālagṛhe rahaḥ| kṛtyāṃ vidhāya saptāhāt saputro nihato nṛpaḥ|| yoganande yaśaḥśeṣe pūrvanandasutas tataḥ| candragupto dhṛto rājye cāṇakyena mahaujasā|| My bold emphasis highlights the differences from the version cited in the Daśarūpāvaloka; for all these, the readings found in the Bṛhatkathāmañjarī are in my opinion superior.

10 Including TELANG 1884:xviii–xix (with reference to Fitzedward Hall’s 1865 edition of the Daśarūpa); KONOW 1914:64; DHRUVA 1923:xxiii, CHATTERJEE 1935:209 and so forth.

11 HAAS 1912:xxxiii.

12 According to Raghavan 1935:491, on the basis of a MS of the Avaloka that omits this passage, and another manuscript with a sub-commentary on the Avaloka that does not comment on this part. Whether or not this particular reference to the Mudrārākṣasa was part of the original Daśarūpāvaloka, Dhanika’s work also refers in other places to our play.

While there is little likelihood to this particular passage of the Bṛhatkathāmañjarī¹³ being the inspiration for the Mudrārākṣasa, the story of “the fall of the house of Nanda”

was evidently a very popular tale in pre-modern India, told and re-told in countless ver-sions including that in the Bṛhatkathāmañjarī. D^HRUVA (1891:31) called attention to a verse in the Pañcatantra that clearly refers to the plot of the Mudrārākṣasa, saying “One should denigrate the prime official of the enemy faction with forged documents and donations of wealth, as Viṣṇugupta did to Rākṣasa.”¹⁴ Dhruva does not ignore the possibility that this reference may not be to our play, but to a lost text with the same storyline, on which Viśākhadatta too may have based his opus. He observes, however, that if the “forged doc-uments” in this stanza had played a role in such a hypothetical text, then Viśākhadatta would have had very little claim to originality,¹⁵ so it is more likely that the composer of this verse had the Mudrārākṣasa in mind. Even so, the stanza is unlikely to be early.¹⁶ It is absent from earlier Pañcatantra versions such as the Tantrākhyāyikā, the Southern Pañcatantra and the late 6^th-century Syriac translation,¹⁷ so its presence in the textus simplicior—no earlier than the mid 9^th century and no later than the late 12^th century ¹⁸— hardly narrows the age bracket available for Viśākhadatta. It does, however, show that at the time this text of the Pañcatantra was redacted, the Mudrārākṣasa (or at least its story) enjoyed wide enough popularity to be referred to in a proverbial style.

The existence of some of the varied (and often contradictory) tales about this topic was pointed out among others by K^ONOW (1914:64). Interest in these stories was partly fuelled by the obsessive concern of the West for “what actually happened,” but some scholars have also done meticulous work on the stories themselves.¹⁹ In this part of my dissertation I will attempt to trace the history of the Mudrārākṣasa story. In the next five chapters I introduce the various traditions that preserve accounts—both purportedly historical and admittedly fictional—of the transfer of power from the Nandas to the Mau-ryas. In doing so I follow a rough chronological order not of the textual sources but of what I believe to be stages in the evolution of the story. The penultimate chapter of this part (page 136ff.) further analyses some characters and events that recur over and over again in these stories.

13 See page 119 below for the story the citation belongs to.

14 Pañcatantra (textus simplicior) 3.138, kūṭalekhyair dhanotsargair dūṣayec chatrupakṣajam| pradhānapuruṣaṃ yadvad viṣṇuguptena rākṣasaḥ||

15 Moreover, he probably would not have featured the seal, the instrument of forging the crucial letter in the title, “a play about Rākṣasa made unique by a seal” (see page 10).

16 Or more bluntly, “The stanza PT III.138 cannot be dated” (KONOW 1914:67).

17 See the correspondence list of Pañcatantra verses in HERTEL 1904:59.

18 TAYLOR 2007:22.

19 Particularly: RUBEN 1956:150–202 and RAGHAVAN 1973:1–68. TRAUTMANN 1971:10–67 has ulterior motives (exploring the historicity of Cāṇakya), but also presents a profound and thought-provoking analysis of the interrelations of the story versions.