U.singh.ieshr. Politics, Violence. Kamandaka's Nitisara 4r5q2x

Indian Economic & Social History Review http://ier.sagepub.com/

Politics, violence and war in Kamandaka's Nitisara Upinder Singh Indian Economic Social History Review 2010 47: 29 DOI: 10.1177/001946460904700102 The online version of this article can be found at: http://ier.sagepub.com/content/47/1/29

Published by: http://www.sagepublications.com

Additional services and information for Indian Economic & Social History Review can be found at: Email Alerts: http://ier.sagepub.com/cgi/alerts Subscriptions: http://ier.sagepub.com/subscriptions Reprints: http://www.sagepub.com/journalsReprints.nav Permissions: http://www.sagepub.com/journalsPermissions.nav Citations: http://ier.sagepub.com/content/47/1/29.refs.html

ed from ier.sagepub.com by Supriya Varma on September 8, 2010

Politics, violence and war in Ka-mandaka’s N tisa-ra Upinder Singh University of Delhi Arguing for a need to move beyond the dominant concerns of the historiography of early medieval India, this article emphasises the importance of historicising and incorporating the ideas of political theorists into the historical construct of the polity of this period. Its focus is a close textual analysis of K€mandaka’s N…tis€ra, an influential political treatise composed at the advent of the early medieval, which offers a graphic morphology of the monarchical power politics of its time, one that was rooted in the genre and scholarly tradition within which the text situated itself, the historical context in which it was produced and the ideas and perspectives of its author. The text maps a political world and reveals a political theorist’s engagement with the problems of the limits and control of monarchical power and violence. In crucial respects, the N…tis€ra’s arguments on these issues are significantly different from those of its more famous predecessor, the Artha€stra. While the Artha€stra reflects a model of an arrogant, absolutist state, the N…tis€ra represents a later, less exultant reflection on political power, one in which non-violence has significantly tempered the discussion of violence, especially with regard to punishment, the royal hunt and war.

Politics, Violence and War in K€mandaka’s N…tis€ra The historiography of early medieval India (c. 600–1300 CE) has been dominated by the question of whether this period should be understood within the frameworks of the integrative, feudal, or segmentary state models.1 Initially enlightening, the 1 The literature on these frameworks is well known and enormous, and it is therefore neither possible nor necessary to give exhaustive references. A good sample of the various views is on display in Kulke ed. The State in India 1000–1700.

Acknowledgements: I would like to thank the anonymous referee for many extremely valuable suggestions which have been incorporated into the final version of this article. I would also like to thank Vijay Tankha, Dilip Simeon, Nayanjot Lahiri, Suryanarayana Nanda and Seema Alavi for various kinds of help rendered.

The Indian Economic and Social History Review, 47, 1 (2010): 29–62 SAGE Los Angeles/London/New Delhi/Singapore/Washington DC DOI: 10.1177/001946460904700102


30 / UPINDER SINGH

half century or so of this debate has reached an ime, and the debate itself has become an obstacle to fresh thinking. Clearly, it is time to frame new questions and re-think the ways in which we can think about the early medieval. A comprehensive, historically grounded intellectual history of this age does not exist, a fact that is especially surprising considering that these centuries were marked by exceptional intellectual vitality. One of the many issues that have received inadequate scholarly attention is the reciprocal relationship between early medieval political processes and the intellectual engagement with these processes in texts of the time. The most important intervention against this indifference has been made by Sheldon Pollock, who has emphasised the fact that the cognitive production of political orders is a significant and integral constituent of these orders and that it is, therefore, essential to explore the ‘political imagination,’ which includes ideas and aspirations of rule.2 Ronald Inden has analysed the representations of the concepts of mastery, lordship and political hierarchy in early medieval India in literary and epigraphic sources.3 And Daud Ali has offered a very thought-provoking, though homogenised, analysis of the representations of courtly culture in texts, inscriptions and art of the first millennium.4 While drawing on the insights of these scholars, my own perspective differs in several respects. While I am interested in analysing the political discourse (I prefer this term to Pollock’s ‘political imagination’) represented in texts, I think it is important that such an analysis must be very carefully calibrated with respect to chronology and spatial context and should not end up presenting an overhomogenised picture of either the discourse or the politics of the time. Further, while identifying shared ideas, and those with an exceptional longevity, it must be equally sensitive to differences in perspective, emphasis and argument within texts of a particular genre and across texts belonging to different genres produced at different points of time. The focus of this article—a close analysis of the N…tis€ra (‘The Essence of Politics’) of K€mandaka—is part of a larger study of political ideas. Comparisons with the Artha€stra provide a useful basis for identifying this text’s perspective. Apart from N…tis€ra’s representation of the morphology of monarchical power, I also examine how the text engaged with an important political problem, namely the interface between kingship and violence, with special reference to punishment, hunting and war. The issue of violence in Indian intellectual traditions and history has many facets, and there are some illuminating works on the subject.5 Violence and non-violence have especially been discussed in the context of sacrifice, religion (Vedic religion, Hinduism, Buddhism and Jainism), asceticism, vegetarianism and Pollock, The Language of the Gods in the World of Men. Inden, Text and Practice, pp. 129–78. 4 Ali, Courtly Culture and Political Life in Early Medieval India. 5 See for instance, Houben and van Kooij eds, Violence Denied. 2 3

The Indian Economic and Social History Review, 47, 1 (2010): 29–62


Politics, violence and war in K€mandaka’s N…tis€ra / 31

Buddhist and Jaina environmental ethics. War has often been discussed by scholars in the context of the Buddhist tradition and works such as the Bhagavad G…t€ and Mah€bh€rata. More off the beaten track is Daud Ali’s analysis of the social meanings of violence, gastronomy and war in the Kali‰kattupparaŠi, a riveting twelfth century text composed in the Co¯la court.6 There is, nevertheless, a need for a more comprehensive, diachronic study of the ways in which the issue of violence was dealt with in ancient and early medieval India. The focus of such a study must not only be on understanding representations of violence, but also on arguments and attitudes towards its various forms. Historicising normative texts raises a fundamental question about the relationship between theory and practice. Pollock has pointed out that in the Indian intellectual tradition, €stra (‘theory’) is generally held to precede and govern prayoga (‘practical activity’), and suggests that the ideas that came to be associated with the nature of €stra may be connected with the belief in the transcendent character of the Vedas.7 However, he himself points to the fact that there were exceptions to this position, significantly enough in works on politics and medicine. A further necessary caveat to this argument is that €stric self-representation should not be conflated with the way in which €stric knowledge was actually produced in early India. The discipline of history assumes that the creation of a textual tradition involved an interface with its historical context, and it can be demonstrated that historical reality intruded into many a ‘normative’ text. However, the biggest challenge in historicising ancient and early medieval political treatises is to meaningfully anchor their political discourses (the plural is deliberately used to underline their diversity in type and perspective) in the peculiarities and demands of their genre, and in their evolving and changing political contexts, without slipping into the error of presenting these discourses as either insulated from or direct reflections of those contexts. K€mandaka’s N…tis€ra and Kau˜ilya’s Artha€stra K€mandaka’s N…tis€ra is a treatise on politics written in Sanskrit verse, consisting of 20 sargas (cantos) subdivided into 36 prakaraŠas (sections).8 It discusses the Ali, Violence, Gastronomy and the Meanings of War in Medieval South India. Pollock, ‘The Theory of Practice and the Practice of Theory in Indian Intellectual History’. He adds that this stance is diametrically opposite to that found in the West, and that adherence to this postulate had profound implications for the production of knowledge in Indian civilisation. 8 The author’s name has been read as K€mandaki by some scholars. The text used for this article is Sisir Kumar Mitra’s revised edition and translation, which is based on Rajendralala Mitra’s The N…tis€ra, or The Elements of Polity by K€mandaki, which was published between 1849 and 1884. This uses an anonymous commentary called the Up€dhy€yanirapek€. Although I have drawn on this edition, I have relied on my own translation of the text. In references hereafter, N…tis€ra has been abbreviated to NS. 6 7



32 / UPINDER SINGH

principles according to which a king should rule his kingdom and how he could attain political paramountcy and prosperity for himself and his subjects. As is the case with many early Indian texts, it is difficult to ascertain when and where the N…tis€ra was written. Estimates of its age generally range between the first and seventh centuries CE.9 The evidence recently cited to suggest the contemporaneity of the N…tis€ra with the Gupta emperor Candragupta II (c. 375–415 CE) is not convincing.10 A more cautious approach, placing the text between c. 500–700 CE is better, and the N…tis€ra can thus be situated at the threshold or the advent of the early medieval. Among the ancient Indian political treatises, it is Kau˜ilya’s Artha€stra that has naturally attracted the maximum attention of scholars, being the first surviving text on the subject and also because of its masterly coverage of an enormous range of issues related to statecraft. The Artha€stra has often been treated simplistically by historians as a direct description of the Maurya state and istration. Such a treatment is problematic because it is a theoretical treatise, not a descriptive work, and although its core probably dates to the Maurya period, it has interpolations belonging to later centuries.11 As for the N…tis€ra, historians have cited stray references from the text to illustrate aspects of Gupta or post-Gupta polity, istration and revenue systems, but the text as a whole has not received the attention it deserves. This, in spite of the fact that it acquired an authoritative reputation, being cited in many later Indian works and also travelling to southeast Asia.12 The scholarly neglect of the work may have been in part because of problems of dating, but the most important reason why it has not been taken seriously enough is that K€mandaka has been viewed as a derivative, unoriginal thinker who tried to simply parrot Kau˜ilya’s

Saletore, Ancient Indian Political Thought and Institutions, p. 9. Willis, The Archaeology of Hindu Ritual, pp. 62–63. Willis’s hypothesis that the ‘Deva’ mentioned in the first verse of the N…tis€ra is none other than Candragupta II, on the grounds that the latter is referred to as ‘Deva’ or ‘Devagupta’ in inscriptions, seems weak. So does his assertion that since K€mandaka describes himself as a disciple of ViŠugupta, alias Kau˜ilya, a generation, or 30–40 years, must separate the two political theorists. 11 The date of the Artha€stra is an issue of continuing debate. Many Indian historians accept R.P. Kangle’s argument that the core of the text was composed in the early Maurya period during the last quarter of the fourth century BCE (Kangle, The Kau˜il…ya Artha€stra, Part III, pp. 59–115); of course additions, interpolations and recasting may have extended into the early centuries CE. Western scholars, on the other hand, are persuaded by the results of Thomas R. Trautmann’s statistical analysis of word frequencies in the Artha€stra (Trautmann, Kau˜ilya and the Artha€stra), on the basis of which he has made a case for several different authors. He suggests that Book 2 (which deals with internal istration) may have been completed by c. 150 CE, and the final compilation of the entire text by c. 250 CE. References to the Artha€stra (abbreviated to AS) in this essay are to Kangle’s critical edition (Kangle, The Kau˜il…ya Artha€stra, Part I). 12 Ghoshal, A History of Indian Political Ideas, p. 395, n. 1. 9

10




ideas, sometimes incorrectly.13 This essay seeks to prove that a close reading of the N…tis€ra does not such an assessment. It is also emphasised that political treatises of this kind have to be recognised as important sources for and, in fact, as important constituent elements of, ancient and early medieval polities. Because of its €stric nature, the N…tis€ra should certainly not be read as a direct description of how states were actually governed or royal policies formulated during the time of the Gupta and V€k€ãka empires and their immediate aftermath. And yet, it offers a perceptive, graphic morphology—often abstract rather than literal—of the structure and relationships of monarchical power politics of its time. This morphology was rooted in various things: the genre and scholarly tradition within which the text situated itself, the specific historical and political context in which it was produced and the ideas and perspective of the author, including his philosophical moorings. Beyond the question of what such texts can tell us about the times in which they were written, it is also necessary to recognise their great influence. The authors of such treatises were learned Br€hmaŠas, at least some of whom were closely associated with royal courts. The presumed audience—and also, in large part, the subject—of these works was the ‘political class’: people associated with the exercise of political power in various ways. This included kings (more specifically the vijig…u—the king desirous of extensive conquest), royal officials, counselors, courtiers,14 military commanders, ambassadors and others. Texts like the N…tis€ra are, therefore, not only representative of a political discourse rooted in the political realities of their time, but also made an impact on those realities. It should also be noted that the ideas of the political theorists were known, absorbed and expressed in poetry, drama, didactic stories and sayings, and reached wider audiences through written, oral and performative traditions. Tradition, Authority and Debate Before entering into an analysis of the political discourse of the N…tis€ra, it is necessary to look carefully at certain general issues related to the production of knowledge in ancient and early medieval India. Ancient €stric discourse on politics was part of an intellectual milieu marked by continuous and wide-ranging debate, a fact often masked by the constant invoking of tradition. Texts that came to be considered authoritative often claimed to be abridged (saˆkipta) versions of

13 See, for instance, Saletore, Ancient India Political Thought and Institutions, pp. 54, 289, 340; Ghoshal, A History of Indian Political Ideas, p. 383; Aiyangar, Some Aspects of Ancient Indian Polity, pp. 24–25. 14 While the term anuj…vin literally carries the connotations of dependence and has, therefore, been often translated as ‘dependant,’ in the context of the court, it is better to translate it as ‘courtier.’



34 / UPINDER SINGH

earlier works.15 New scholarship presented itself as part of a venerable tradition of long standing, absorbing the latter’s weight and authority through association, even if it disagreed with that tradition in radical ways. In fact, it can be argued— and this point is borne out by the Artha€stra—that in spite of the great placed on tradition, disagreement with earlier authorities ultimately contributed to a scholar’s reputation.16 Citations were an important way of positioning a new text and author in relation to older scholarly traditions. In the N…tis€ra, there are references to the collective wisdom of the vddha (elderly), vidy€vddha (those mature in wisdom), vidu (learned ones) and those described as maŠalajña (experts in the science of the maŠalas), arthajña (experts in artha), €str€rthacintaka (those who think about the meaning of the €stras) and p™rva munis (sages of the olden days). The text also refers in two places to popular wisdom embodied in laukika verses. 17 K€mandaka cites various specific schools and authorities—the M€navas, Indra, Maya, Bhaspati, Uanas, ViŠugupta, Puloma, Súkra, Vi€l€ka, Par€ara and Bharadv€ja—expressing his agreement or disagreement with them.18 These are cited frequently in other ancient texts (including, for instance, the Artha€stra and the Mah€bh€rata) as authorities on the science of politics and on dharma. The fact that the maximum number of differing views are cited in the section on maŠalayoni (‘the nave of the circle of kings’, NS 8.12) suggests that the conduct of inter-state relations was a topic of especially heated debate. This is being emphasised here because the authors of ancient Indian texts have often been considered as amorphous ‘types,’ completely smothered by tradition and the convention of their genre, with little scope for expressing their distinctive points of view. While Bhaspati is the most frequently cited authority in the N…tis€ra, it is ViŠugupta, alias Kau˜ilya, the author of the Artha€stra—referred to on two occasions as ‘our guru’19—who holds the preeminent position for K€mandaka. The text opens with a salutation to the god GaŠea, the king,20 and ViŠugupta, in 15 Apart from the N…tis€ra this also applies, for instance, to the Artha€stra and the K€mas™tra. For a detailed discussion of this point, see Pollock, ‘The Theory of Practice and the Practice of Theory in Indian Intellectual History’, pp. 512–13. 16 Ali’s (Courtly Culture and Political Life in Early Medieval India, p. 276) argument that the Artha€stra marks a significant shift in the language of lordship, from agonistic to irenic kingship is thought-provoking. However, the shift does not seem to have been a complete one. 17 NS 11.16.32a refers to a laukika g€th€, according to which, a living man is likely to secure happiness even after the lapse of a hundred years (this refers to the cyclical nature of fortune/happiness and misfortune/sorrow). NS 14.21.33 refers to a popular saying (s€rvalaukika prav€da) that the treasury (koa) is the root (m™la) of the king. 18 While some of the schools and thinkers cited were clearly historical, the god Indra and the demon Maya fall within the category of fictive authorities. 19 NS 2.3.6 and NS 11.16.42. 20 According to some scholars, the reference is to the god ViŠu, but the verse seems to refer to a generic king.




that order. The eulogy of ViŠugupta (NS 1.1.2–6) describes him as one who was born in a great lineage with descendants famous all over the world for their ilike conduct in not accepting gifts of any kind; who was as effulgent as the sacrificial fire; who was so well-versed in the Vedas that he had mastered through his intellect all four of them as though they were one; who through his powers, as irresistible as furious thunder, had uprooted the great and powerful Nandas; who, like the god Sáktidhara (K€rttikeya), through the exercise of his mantraakti (power of counsel) had single-handedly secured the world for Candragupta, the moon among men; the learned one, who had produced the nectar of n…ti€stra out of the mighty ocean of artha€stra. The precise identity and background of the author or authors of many ancient Indian texts is often elusive. But this description of ViŠugupta can be read as a portrait of the political Br€hmaŠa—the kind of advisor considered by K€mandaka to be most suited (and most likely) to deliver the teaching on politics. This may well have been a self-portrait of K€mandaka himself. The connection with ViŠugupta was also important for establishing the bona fides and boundaries of the discipline that the N…tis€ra dealt with, and that is probably why K€mandaka ascribes the invention of n…ti€stra to his famous predecessor. The N…tis€ra describes its subject of inquiry as n…ti (explained as derived from nayana, leading or istering) (NS 2.3.15),21 daŠan…ti and r€javidy€. The Artha€stra and N…tis€ra have major overlaps in content and concerns and share a common political vocabulary, including the key ideas and theories of the sapt€‰ga r€jya (the seven-limbed state), r€jamaŠala (the circle of kings) and €guŠya (the six measures in the context of inter-state relations). But there is also much that is different, not only in style—the Artha€stra is in the aphoristic s™tra style interspersed with a few verses, while the N…tis€ra is entirely in verse— but also in specific details and over-all tenor. For instance, a detailed discussion of internal istration and civil and criminal law are missing in the N…tis€ra, as is the advocacy of strict state control over various aspects of the economy, often considered the hall-mark of the Artha€stra. This reflects the narrower scope of n…ti€stra as compared with artha€stra, as well as differences in views about the potential state held by Kau˜ilya and K€mandaka. Both were concerned with political expansion and consolidation, but the N…tis€ra does not share the Artha€stra’s grandiose and overwhelming vision of state power. Neither does its author share Kau˜ilya’s faith in the efficacy of black magic as a political and military tool, a detailed discussion of which is absent in the text. And, as we shall see later, K€mandaka also disagreed with Kau˜ilya on various specific issues related 21 The word ‘n…ti’ comes from the root ‘n…’, which means to take away, guide, conduct, lead, etc.. In its broadest sense, it can refer to policy, rules, or precepts governing any aspect of life. But in the context of the text and branch of knowledge being discussed here, n…ti has a more specific sense related to polity, statecraft and governance.



36 / UPINDER SINGH

to the interface between kingship and violence. It is necessary to emphasise this point, because as mentioned above, many scholars have mistakenly described the N…tis€ra as an unoriginal derivative text which repeats, often poorly, the ideas of the Artha€stra. In the many-faceted and vibrant intellectual milieu of ancient and early medieval India, disciplinary boundaries were understood and the political treatises selfconsciously situate themselves within a larger knowledge universe. But knowledge and ideas also readily flowed across disciplinary boundaries. The texts on polity share with the Dharma€stra and philosophical texts ideas related to karma (the consequences of action), rebirth, caturvarga/trivarga (the four or three goals of human life) and varŠ€rama dharma (dharma based on varŠa and €rama). In fact, V.P. Varma points out that the metaphysical foundations of ancient Indian political thought had important implications for the kinds of questions that the latter asked and did not ask.22 The texts on polity and Dharma€stra share a concern about ethics, and the political theorists in fact constructed a sub-specie of dharma, one that was geared towards the realisation of the political goals of the king. The four vidy€s (branches of knowledge) described by K€mandaka as essential for the preservation of the world, are anv…kik… (philosophy), tray… (the three Vedas), v€rt€ (economics) and daŠan…ti (the science of politics). The N…tis€ra is unequivocal in asserting the preeminence of daŠan…ti among all the branches of knowledge, on the grounds that the latter are useless in its absence (NS 2.3.9).23 Nevertheless, as we shall see, its ideas about politics are imbued with philosophical, ethical and metaphysical presuppositions. The inter-disciplinary dialogue of which the political treatises were a part extended to other realms of specialised scholarship as well. In fact, the organic theory of the sapt€‰ga r€jya may have more connections with medical knowledge than hitherto suspected. It seems to be more than a coincidence that €yurveda talks of seven elements of the body and the political treatises talk of the seven elements or limbs of the state.24 The discussion of the vyasanas (calamities) and concerns 22 Varma (Studies in Hindu Political Thought and its Metaphysical Foundations, pp. 215–17). He argues that the metaphysics of karma and dharma ruled out the raising of problems such as the individual versus the state, politics versus ethics and the political ability of the king to the people. 23 Contrast this with the Artha€stra’s (1.2.12) description of anv…kik…—consisting of s€ˆkhya, yoga and lok€yata—as the lamp of all sciences (prad…paƒ sarvavidy€n€m). 24 The seven elements are chyle, blood, flesh, fat, bone, marrow and semen. The term doa also occurs in €yurveda, but has a different meaning than that in the political treatises. The three doas or humours are substances which circulate within the body—v€ta (wind), pitta (choler) and kapha or sleman (phlegm). Disease arises when a humour collects in the wrong area of the body and becomes inflamed (Wujastyk, The Roots of Ayurveda, pp. 4, 31). The guŠa and doa as indicative of positive qualities and faults occur in many different kinds of discourse, including as features of language in Indian literary theory.




for purification (uddhi) of the various elements of the body politic (for instance, of the maŠalas) in texts such as the N…tis€ra also resonate with issues of health, disease and cure with which the medical treatises grappled, albeit in the context of the body politic rather than the human body.25 The political and medical treatises also share a great concern with poison—the different types, their symptoms, antidotes and how to identify a poisoner. In this particular sphere, the scholars of artha€stra and n…ti€stra must have benefitted from the knowledge and experience of the medical experts. It is also interesting to note that in two places in the N…tis€ra, there are specific surgical analogies. N…tis€r€ 5.8.8 states that the king punishes corrupt royal officers who accumulate wealth in the same way as a swollen abscess is operated on and bled out (€sr€vayed upacit€n) (by a surgeon). And in N…tis€r€ 6.9.13, the king is told to act like a surgeon (alya) in eliminating those who trouble the people. The Vijig…u’s Quest for Power The prime subject as well as audience for the N…tis€ra was the king (r€jan), whose various epithets announce him as lord of the earth, of all men, and of the maŠalas (mahipati, p€rthiva, pthv…pati, mahibhuja, bh™pati, kitibhuja, npa, narapati, narevara, maŠal€dhipa). Monarchy (r€jya) is the only kind of state mentioned by K€mandaka, as his work post-dates the annihilation of the major oligarchies (gaŠas and saˆghas) by the Gupta emperor Samudragupta (c. 350–70). This is in contrast to the Artha€stra, which discusses oligarchies, although it too considers monarchy as the norm and addresses its teaching to the king. The king is not defined by what he is but what he aspires to become. The kind of monarch that the N…tis€ra (and the Artha€stra) has in mind is one who is ambitious and upwardly-mobile, a vijig…u—a king desirous of attaining political paramountcy, one who seeks dominion over the whole earth washed by the ocean (samudraprak€lit€ dh€ritr…) (NS 16.24.35). The graphic image—one which endures in texts and inscriptions over several centuries—of the paramount king is that of one who plants his foot on the heads of enemies adorned with excellent helmets and bejeweled crowns (NS 14.20.12). While the N…tis€ra has a strong sense of the past, the past it invokes is not one that modern historians would regard as ‘historical.’ In fact, the deliberate avoidance or erasure of the latter is because the text’s discourse (like that of other €stras) speaks of universals, not particulars.26 It is not surprising that a work that claimed 25 For an interesting discussion of the connections between medicine and poetry, see Zimmermann, ‘¬tus€tmya: the seasonal cycle and the principle of appropriateness’. 26 Pollock (‘M…m€ˆs€ and the Problem of History in Traditional India’) has expressed this eloquently in a broader context, arguing that in Sanskritic India, history was not unknown but denied. His hypothesis, which merits careful consideration, is that this was because the authority of the Veda was considered by M…m€ˆs€ as based on its timelessness, and that this became a model for all forms of knowledge.



38 / UPINDER SINGH

to lay down the principles of polity for all time to come ignored the inconsequential kings of mundane petty power politics and drew its illustrations from the gods, demons and men of the epic-Pur€Šic tradition, whose fame or notoriety transcended time and space. Apart from analogies between the king and the elements of nature,27 the N…tis€ra frequently compares him with the gods, especially Indra, Yama and Praj€pati. The text abounds in references to Paraur€ma, Ambar…a, Yudhi˜hira, Bh…ma, Nala, Janamejaya and R€ma, leaving no doubt that the Mah€bh€rata and R€m€yaŠa were pivotal to K€mandaka’s political discourse, in fact more so than to that of the Artha€stra.28 The N…tis€ra, like many texts of ancient and early medieval India, talks of the intimate connection between kingship—r€jat€ (NS 20.34.16)—and the prosperity of the king, his realm and his subjects. S´r… and Lakm… represent fortune and prosperity as well as goddesses personifying these things, and many of the references to them are clustered in the Uts€hapraaˆs€ (‘in praise of energy’) prakaraŠa (NS 14.20). The feminine deification of fortune lent itself well to the use of gendered imagery, one that is overtly sexual, even violent; this is found frequently not only in the political treatises but also in Sanskrit k€vya. S´r… and Lakm… are fickle, but cling to the righteous king like his shadow.29 A ruler who is intelligent but inactive due to his addiction to vyasanas (vices), loses the favours of S´r…, just as an impotent man is discarded by women (stribhiƒ aŠa iva; NS 14.20.8). A ruler desirous of enjoying S´r… should not behave like an impotent man; he must always strive to control her with his manly powers, just as a husband controls his wicked wife (NS 14.20.10). An ever energetic ruler gains possession of S´r… by adopting the conduct of the lion (siˆhavtti), just as a man deals with his wicked wife by dragging her by the hair (NS 14.20.11). The reference to Lakm…s (in the plural) entering the king endowed with various virtues, like streams enter the ocean (NS 14.20.7) and the occasional mention of S´r…, too, in the plural, has a distinct polygynist tinge. The Morphology of Monarchical Power The emphasis on the power and ambitions of the vijig…u should not obscure the fact that the polity of ancient Indian political treatises such as the Artha€stra and N…tis€ra is an organic one, where the king is embedded in a web of complex, reciprocal relationships with the other praktis (elements of the state), listed by 27 He is compared with the sun, moon, Vindhyas, rain-bearing clouds, earth, blazing fire, lion, elephant and snake. He is also compared with the kalpavka (the wish-granting tree). 28 The Sanskrit epics were pivotal to many other discourses on politics and kingship in ancient and early medieval India. The full extent of this importance—reflected in texts, inscriptions and sculpture from many parts of the subcontinent and beyond—is something which has not yet been fully gauged. 29 NS 5.8.91 A (this loka is marked kroa, i.e., supplementary in the commentary) and NS 14.20.6.




K€mandaka as sv€min (lord, king), am€tya (counselor), r€˜ra (domain), durga (fort), koa (treasury), bala (military might) and suƒrt (ally).30 This interconnectedness should not be lost sight of in discussions of the structure of the polities of early medieval India. Ali has rightly pointed out that historians have tended to focus on the figure of the king, ignoring those around him, including the royal household and other of the courtly elite.31 This is in spite of the fact that the ancient and early medieval texts have plenty to say about the importance of these groups. The N…tis€ra underlines the fragility of power and the inherent instability of all political and personal relationships and offers a graphic morphology of the political world of its time. In its basic respects, this morphology is similar to that offered by several other texts, and one that has been described by Ali.32 However, this morphology is being spelt out here for three reasons. First, it is an essential part of the N…tis€ra’s mapping of politics. Second, Ali’s analysis conflates features of texts of different kinds (including k€vya and treatises on k€ma, polity and architecture) and centuries (ranging from c. 300–1200 CE), and while these shared a substratum of common ideas, there were also differences of perspective, emphasis and detail. Therefore, it becomes necessary to identify and isolate the N…tis€ra’s specific perspective. And third, the focus in the discussion here is not only on courtly culture but on something larger than that, a sphere that can be called the political. For it must be noted that the king and his court were the epicentre of a complex and far-reaching web of political relationships, manouevres and intrigues, one which extended far beyond the court into the domains and courts of neighbouring and distant kings and chieftains, as well as into the forest. The fundamental premise that underlies K€mandaka’s entire discussion of politics is the view that human nature is essentially selfish and that all people are arth€rthin, i.e., seek to further their personal interest. The challenge for the vijig…u was to harness other people’s desire to further their self-interest to ensure the satisfaction of his own ambition and desire for political aggrandizement. From the point of view of the king, the world was a treacherous but challenging place. Fate was definitely a factor (NS 12.17.20) to be reckoned with, but political success required many inherent and cultivated positive qualities (guŠas) and a great deal of deliberate effort. K€mandaka’s long list of guŠas that the king should possess (there are similarities with the Artha€stra in this discussion) reveals an

30 The seven elements of the state in the Artha€stra are sv€min, am€tya, janapada, durga, koa, daŠa and mitra. The N…tis€ra (NS 4.7.1) makes a slight modification in terminology, replacing janapada with r€˜ra, daŠa with bala, and mitra with suht. It is interesting to note that Bhaspati rather than Kau˜ilya, is cited as the authority on the seven elements of the kingdom (NS 8.12.5). 31 Ali, Courtly Culture and Political Life in Early Medieval India, p. 5. 32 Ali, Courtly Culture and Political Life in Early Medieval India, pp. 103–40.



40 / UPINDER SINGH

important aim of the political theorists—to temper brute power with virtue.33 According to the N…tis€ra, the many guŠas necessary to become a successful vijig…u included nobility of ancestry, intelligence, truthfulness and powers of endurance. The most important quality, however, was prat€pa (prowess). Energy (uts€ha) and constant vigilance were also required to safeguard and extend political power. The maintenance and extension of this power did not only involve coercive power and conquest. It was essential to skillfully use force (daŠa) along with the other political expedients (up€yas), namely s€ma (pacification), d€na (giving gifts) and bheda (creating dissension), in order to generate and maintain confidence (viv€sa) in the various praktis. Confidence, in turn, was an essential prerequisite for eliciting loyalty and love (anur€ga) from subjects, soldiers and allies—the kind of loyalty and love that would extend over many generations. There is no separate word for loyalty in the text, but is subsumed in other such as sev€ (service) and bhakti (devotion), and is emphasised by assertions that the king should not be forsaken by his courtiers. The political importance of the royal household in the politics of the monarchical states of the time is amply evident in the N…tis€ra’s detailed discussion of princes (NS 7.10) and the harem (NS 7.11). of the harem (antaƒpura, avarodhana) were the abhyantara jana (inner people) and included the king’s mother, his wives, courtesans (r™paj…v€s) and the many attendants who waited on all these women. The harem was a place of pleasure and sensual indulgence. It was also the locus of the serious business of producing heirs—an issue that the N…tis€ra does not see as a problem, the virility of the king perhaps never being in doubt. And yet, as for Kau˜ilya, it was also a place of danger and intrigue, one where the king was strongly advised never to completely lose his head in the pursuit of sexual pleasure or emotional engagement. In the N…tis€ra (16.24.28), there are seven types of people associated with the king—his own men, those of his allies, those who have taken refuge with him, those related to him, those associated with him for some specific purpose or action, his servants and those won over by various services and gifts. The category of the king’s own men (nija) included his courtiers (anuj…v…s), to whom he was tied with complex ties of reciprocity. An entire sarga (NS 5.8) is devoted to the relationship between the sv€min and his anuj…v…s. The latter are mentioned in the same breath as the bandhu (kinsmen) and mitra (friends), and the summary of their duties includes giving the king good counsel, dissuading him from inappropriate acts and implementing his desires (NS 5.8.50).34 High-ranking royal officials such as am€tyas, mantr…s, sacivas, adhyakas, d™tas, mah€m€tras, the purohita, the 33 In the context in which the word guŠa occurs here, it does not (as suggested by Ali, ibid., 90–91) have any connection with sattva, rajas and tamas. 34 The text makes a distinction between servants (denoted by bhtya and sevaka) and dependants/ courtiers of a higher socio-political standing (denoted by anuj…vin).




sen€pati and astrologers appear to be included in the category of courtiers. There is a long list of the ideal qualities of an anuj…vin, but it is interesting to note that the synopsis of the most essential qualities leave out high birth and proficiency in the Vedas (NS 5.8.12–15). The discussion of the relationship between the king and his courtiers was no doubt rooted in the contemporary political context— specifically in the need for the king to create around him a group of capable and loyal courtiers, and the latter’s desire and ambition to move up in the courtly hierarchy. The relative status of an individual in this hierarchy was determined by the level of proximity, physical as well as affective, to the king. The king was advised to show his courtiers respect, to pay them in accordance with services rendered, never to discontinue payments they had been enjoying, to offer suitable gifts and rewards, to promote them and to maintain their internal hierarchies. The goal was to have obedient (anugata), loyal and satisfied courtiers. From the point of view of the courtiers, a king is described as worthy of being served by virtue of his possession of good qualities of character, devotion to duty, prosperity and generosity. The last of these was important, for what could not be attained by a courtier who was favoured by a prosperous and generous king? On the other hand, a king who could not provide materially for his people (including his courtiers) would be forsaken by all, like a dried-up tree is forsaken by birds and a milk-less cow is forsaken by its calf.35 The most evocative and most entertaining part of K€mandaka’s treatment of the king’s relationship with his courtiers is his lengthy and very specific discussion of the protocol and decorum of the r€jasabh€ or saˆsad (court) (NS 5.8.17–34). The challenge for the courtier was to be counted among those who were cultured (sabhya, €rya). He was advised to meticulously tailor his deportment and behaviour to prevailing court protocol and propriety in accordance with his rank and position. And yet, he had to simultaneously strive to rise above others in the court milieu by making a strong, distinctive impression, especially on the king. K€mandaka’s discussion of court protocol, broadly similar to that offered by Kau˜ilya (AS 5.4), is general enough to be considered as fairly close to the basics of protocols actually existing in various royal courts of the subcontinent during the time. The intelligent (medh€vin) courtier had to be formally itted into the royal assembly-hall. He was advised to be properly attired, to take his allotted seat and to wait patiently for his turn to pay respect to the king with due humility. According to K€mandaka, the anuj…vin should be conversant with what was appropriate to place and time (deak€lajña) and should be an expert in interpreting the king’s gestures, appearance and movements (i‰git€k€ratattvavid). He should be very careful about where and how he looked and how he spoke, and was advised to gaze intently at the king’s face in order to observe his reactions and to listen very attentively to whatever the lord said. 35

NS 5.8.59; 5.8.63.



42 / UPINDER SINGH

Just as important as creating the right impression was avoiding creating an adverse one. There were many things that the anuj…vin was advised not to do under any circumstances (NS 5.8.23–33). He should never display any arrogance, never imitate the king’s attire or manner of speech, and never occupy someone else’s place or seat. Unseemly behaviour liable to attract adverse attention and, therefore, to be strictly avoided in the presence of the king, included talking or laughing loudly, coughing, spitting, sitting in an ungainly posture, yawning, stretching limbs and making sounds by twisting the fingers. In times of emergency, the anuj…vin could be daring and take the initiative in addressing the king, but even then only with due politeness. While attempting to move closer to the king and up the ladder of success (these two things were synonymous), the ambitious courtier had to reckon with those who shared these ambitions and those who had already achieved them. Apart from being deferential towards the king, he was advised to be respectful towards superiors and the king’s sons, friends, companions and favourites (vallabhas). Such people, it is pointed out, could pierce the heart of the courtier in the saˆsad by their ridicule (NS 5.8.19–20). The courtier was also advised to avoid meeting or interacting with women (presumably of the harem) and their supervisors, habitual sinners, messengers from hostile chiefs, and those who had been dismissed by the master (NS 5.8.32). In its description of the overlap between the personal and the political, the N…tis€ra offers an important insight into the polities of mid-first millennium India. Some insights can also be gleaned by noting what the text does not mention or highlight. The political elites of the time must have been internally divided into factions, and the inter-relationships among these factions must have had many complex strands, including alliance, competition, rivalry and hostility. The N…tis€ra is intriguingly reticent on this important issue and this reticence may have been deliberate.36 K€mandaka’s aim may have been to deliberately mask elements of factionalism and conflict within the political elite and to emphasise and thereby try to inculcate within that elite a certain cohesion that did not exist in actuality. This may be why the text describes the courtiers as united in their service to the king, repeatedly emphasising the personal benefits that could be reaped through such service. The composers of k€vyas and gnomic literature may well have brooded over the fact that a loss of autonomy was the corollary of the courtier’s dependence,37 but political treatises such as K€mandaka did not waste their time reflecting on this issue. 36 There is ing reference to hierarchies within the court circle, for instance, in the mention of the honorific paraphernalia such as umbrella, fly-whisk, coachman and grooms that the mah€m€tras and certain other high-ranking officials enjoyed (NS 13.19.44). The dependant is advised to observe decorum in deference to other courtiers’ rank and there is mention of his rivals (vipaka; 5.8.40). But there is no direct or detailed discussion of court factions and rivalries. 37 Ali, Courtly Culture and Political Life in Early Medieval India, pp. 246–47.




Emotions were an important part of K€mandaka’s political discourse, and political success was considered to be considerably dependent on the ability to create in oneself and in others certain desirable emotional states and dispositions.38 In his interactions with his courtiers, the king’s aim was to secure their devotion (bhakti), loyalty and affection. For courtiers, a crucial objective was to obtain royal affection (anur€ga), and to regain it if it was lost for some reason or another. There was also the more pragmatic aim of acquiring permanence (sth€ne sthairya) of position (NS 5.8.5), something no doubt difficult in a polity where everything was always in a state of flux. In broadly similar to those of the Artha€stra (AS 5.5), but in greater detail, the N…tis€ra explains how the anuj…vin could gauge the success or failure of his attempts to worm his way into his master’s affections. K€mandaka lists the visible signs of a king who was attached (anurakta) and one who was indifferent or hostile (virakta) (NS 5.8.35–38). A king who regarded a courtier with favour would express delight on meeting him, accept his views with appreciation, offer him a seat close to his own and make solicitous enquiries about his welfare. He would not hesitate to meet the courtier alone in an inner chamber and would entrust confidential tasks to him. On hearing other people praise the anuj…vin’s work, the king would applaud and congratulate him. He would mention the courtier’s name and praise his qualities in the course of his discussions with others. Otherwise unpalatable words would be tolerated from a favoured courtier, and the king would act on his advice. The description of the signs of a king displeased with his anuj…vin are more graphic and are also dilated upon (NS 5.8.39–46). While the courtier was still talking about an undertaking and its results, the king might suddenly break out in laughter, and moving towards him with a frowning face, suddenly walk out of the room. However appropriate the observations made by such an employee might be, the king would pay no heed to them and, stopping him in the midst of his speech by expressing dissatisfaction, he would instead accept a contrary opinion. Feigning was also a strategy of expressing disfavor. If the king was lying in bed and was entreated for a favour by the courtier, he would pretend he was still asleep. If the courtier tried to wake him up, he would pretend to continue to be asleep. The good qualities of an out-of-favour courtier would never be recognised and he would be the object of frequent cutting remarks, censure (pariv€da) and derision. His services would be dispensed with for imaginary faults. Promised rewards would be withheld and other people would receive the credit for his meritorious actions. The king would be indifferent to such an anuj…vin’s fortune or ruin and would incite his rivals (vipaka) against him.39 These, according to K€mandaka, 38 Ali has very effectively demonstrated (2006: 183) the coincidence of the terminology of emotions and political dispositions in a variety of courtly literature of the first millennium. 39 As mentioned, earlier, this is one of the few hints about fractures and factions within the courtly circle.



44 / UPINDER SINGH

are the signs of a master who is virakta. Kau˜ilya gives a similar, but more concise , but adds that a king’s disfavour could also be understood through intimations in the form of a change in behaviour of animals and birds, for instance in the barking of a dog or a heron flying towards the left (AS 5.5.10). The N…tis€ra (like the Artha€stra) indicates the great importance of the king’s kinsfolk in the world of political power. Apart from their place within the royal household, allies included those related by blood (aurasam) (NS 4.7.74). But in a situation where feigning and double-dealing were intrinsic parts of political culture, the king could never rely on their loyalty or allegiance, and the danger of betrayal was not a possibility, it was a very likely probability. A king ed by his brothers was invincible (NS 9.14.46), but a king could not trust even his father for seeking shelter (NS 11.16.35), let alone his sons and wives. The latter were, in fact, among the greatest sources of danger to him. Personal friendship too had a place in political discourse and in the political world.40 Kings had to win over friends through kindness (NS 3.6.33) and they were the king’s companions during his leisure-time pursuits (NS 7.11.34). The killing of friends was one of the various possible causes of war (NS 10.15.4), along with other causes including dishonor, the killing of kin and the abduction of women. The king’s friends seem to be distinct from the vallabhas or favourites, although one can imagine that there must have been some overlap between the two categories. The vallabhas seem to have been considered especially problematic characters by K€mandaka. In fact, they are the only group singled out for specific mention in the kaŠãkaodhana (‘removal of thorns’) section (NS 6.9.9) and are also mentioned as one of several sources of fear to the subjects (NS 5.8.82 ). It may be noted that the phrase vallabha-durlabha (‘not to be entered by royal favourites’) in numerous land grant inscriptions of the early medieval period suggests that the vallabhas were identified as potential trouble-makers for the donees, villagers, or both. The denoting friendship and cordiality had many different nuances, and the vocabulary of kinship and friendship extended to many political relationships. The familial term pariv€ra is used synonymously for sacivas (counselors) (NS 4.7.10). The ally is referred to as mitra or suht. Cordiality had to be cultivated with the maŠalikas (governors) of distant regions and with governors of forts. Attachment, estrangement, love, loyalty, confidence and friendship are sentiments that are invoked to describe relations between king, courtiers, subjects and other rulers. Disposition and sentiment were clearly important ingredients in the N…tis€ra’s political discourse and, presumably, in the monarchical power politics of its time. Of course, revealing as the basic morphology of monarchical power 40 Since the suƒrt and mitra are used in the text to refer to personal friends as well as political allies, in some cases it is not clear which of these meanings is intended.




and courtly life that is represented in the N…tis€ra is, it is essential to keep in mind that it is an idealised and aestheticised morphology.41 Dangers to King and Kingdom Like his counterpart of the Artha€stra, the king of the N…tis€ra too inhabits a very dangerous world, and his foremost challenge (and indeed duty) is to protect himself. The detailed description of the king as a figure assailed at all times and from all sides by the threat of assassination, especially through poison, may have been realistic. And even if it is exaggerated, it suggests that this was considered a very real threat and a source of anxiety for kings and political theorists alike. It was because of the ever-present danger to his person that the king was advised to be well protected, ever-vigilant, and to sleep lightly like a yogin (NS 16.24.44). Going by the lengthiness of the discussion, the most dangerous place for the king was the harem (NS 7.11.41–50). This was a space where there was much coming and going, and all these movements required careful regulation. of the antaƒpura were to be watched over by officers known as antaƒpur€m€tyas. Spies in various disguises were also to keep a strict watch over everyone.42 There is something rather bizarre about the scene painted by K€mandaka—he suggests that the king move about the antaƒpura escorted by daring, mailed and turbaned hunch backs (kubjas), hunters (kir€tas) and dwarfs (v€manas). Within the harem, . the king should always be protected by armed palace guards (antaravam ika sainya). Men of (over?) 80 years and women of 50 years and eunuchs should be appointed as attendants to of the harem. Even when going to meet his mother, the king should first ensure the purification of undesirable elements from the harem. While entering his mother’s apartment, he should be escorted by trustworthy armed followers and should not linger in narrow ages or deep alleys, lest he be attacked by murderous assailants. The antaƒpura was a place of pleasure where the king engaged in sexual activity with wives and courtesans (r™paj…v€s), but K€mandaka seems to recommend that the king should not sleep there, as no matter how beloved she might be, too much confidence must never be placed in a woman. To hammer home the point, numerous examples are given of treacherous queens who had killed their husbands. Sons too were a source of serious trouble, and had to be both protected and protected 41 One of the problems in Ali’s discussion is that he occasionally slips from his stance of viewing the various discourses on courtly culture as representations to one where these representations are presented as descriptions of courtly life. He does, however, point out that the texts themselves mention the breach of court protocol. 42 NS 13.19.42–43 suggests that spies pretending to be idiots, dumb, blind and deaf persons, eunuchs, kir€tas (hunters), dwarfs and hunchbacks, petty craftsmen, monks/mendicants, minstrels, slave women, garland-makers and artists should keep a watch on of the harem.



46 / UPINDER SINGH

from (Kau˜ilya too warns of these problems). The first verse of the R€japutrarakaŠa prakaraŠa (NS 7.10) suggests that this section is concerned with the protection of princes. But the subsequent discussion makes it abundantly clear that it is really about protection of the king from princes. Whether descriptive or exaggerated, the entire discussion indicates that these issues were considered central to the safety and survival of the king. Political success did not only involve the king keeping his own house in order, it also required effective management and manipulation of many other relationships. Beyond his own household and court, the king interacted with those of other kings. Allies, neutral parties and enemies could be identified according to certain principles and could be dealt with effectively in various ways. The enemy (ari) was a potent source of danger to the king and to his sovereignty. What complicated matters was the fact that in the circle of kings, relationships were ever-changing— at one stroke, allies could become enemies and vice versa (NS 8.13.72–73). Potential trouble-makers included those only partially integrated into the circle of kings—s€mantas (bordering chiefs or rulers) and €ãvikas (forest dwellers), frequently mentioned in the same breath (NS 14.21.29; 15.23.22). In the N…tis€ra, as in the Artha€stra, the term s€manta does not yet have the connotations of a subordinate feudatory, which it acquired in later times.43 The category of subordinate rulers in fact seems to be represented in the discussion of types of alliances rather than of the s€mantas. For instance, there is a discussion of the various kinds of treaties or agreements that could be concluded with a weaker or defeated power. Among these, the puru€ntara sandhi carried the express obligation that the army chiefs (yodhamukhyas) of the ally would serve the vijig…u’s interests (NS 9.14.13). But there is no detailed description of the elaborate protocols that involved an ostentatious display of the hierarchy of power between paramount and subordinate kings, of the sort that are found in texts of later centuries. Ancient Indian political theorists were tuned in to the dangers of political crisis and collapse. In several places, Kau˜ilya alludes to the danger of an insurrection by disaffected subjects (praktikopa) (e.g., AS 1.19.28). K€mandaka classifies disturbances that could assail the kingdom into two categories—internal (antaƒprakopa) and external (b€hyaprakopa) (NS 16.24.19–21). The former, described as potentially more harmful, included disaffection among the royal purohita, am€tyas, princes, of the royal family, commanders and chiefs of army contingents. B€hyaprakopa included disaffection among provincial governors, frontier guards, forest people and those compelled to surrender.44 43 See Gopal, ‘S€manta—Its Varying Significance in Ancient India’ for a discussion of changes in the connotations of the term s€manta. 44 AS 9.5 has a more general discussion of troubles emanating from the interior and the outer regions.




But the king’s most dangerous enemy was the king himself. The N…tis€ra talks at great length about the problems that a kingdom faces due to the king’s own character and dispositions. These include vyasanas emanating out of vanity (mada), anger (krodha) and attachment to sensual pleasures (k€ma) (NS 15.23). A kingdom in which the king is afflicted by vyasanas is in deep trouble, even if the other praktis are functioning well. The king was also implicated in a number of problems arising out of the fact that the exercise of power invariably involved violence of various kinds, and it is to this that we now turn. The Problem of Political Violence There are many problems in correlating the connotations of the value-loaded term English word ‘violence’ with what is often taken as its closest Indian counterpart— the Sanskrit word hiˆs€. Although violence has been a perennial feature of human history, its definition, the difference between legitimate and illegitimate infliction of injury or use of force, and the grounds of justification for or condemnation of these are very culture-specific.45 War forms a central event in the two great Sanskrit epics and the dilemmas and problems associated with large-scale military conflict are more graphically revealed in the Mah€bh€rata rather than the R€m€yaŠa.46 But it also looms large or lurks on the fringes of many an ancient and early medieval text. In the political treatises, it is centre stage. Without going into the intricacies of semantic issues related to violence, the focus here is on certain specific activities associated with kingship that, according to our contemporary notions of violence, inevitably involved a measure of violence towards humans or animals—the punishment of criminals, war and hunting. The N…tis€ra, like the Indian tradition in general, distinguishes between legitimate and illegitimate force. It is evident that the political theorists were concerned with theorising the limits of force and violence perpetrated by the state in these spheres. The fact that daŠa means both force and justice directs our attention to the insistence that the use of force must never be impulsive or random but must always be tempered by reflection and calculation, involving a judicious compromise between the demands of political expediency and justice. As for Kau˜ilya, for K€mandaka too, the goals of kingship were the attainment of enduring political paramountcy and the prosperity of the king and his subjects. Attaining these goals often involved using violent and what would ordinarily be considered deceitful means, and the political theorists were not squeamish about See, for instance, Houben and van Kooij eds, Violence Denied, pp. 1–3. Much has been written on this theme. For instance, Feller Jatavallabhula (‘RaŠayajña: the Mah€bh€rata war as a sacrifice’) examines the idea of war as sacrifice in the Mah€bh€rata, while Biardeau (‘Ancient Brahmanism, or Impossible Violence’) looks at issues related to violence in the Manu Smti and the Mah€bh€rata, especially at Arjuna’s assertion of the impossibility of avoiding violence. 45 46



48 / UPINDER SINGH

such matters. DaŠa involves suppression (‘damo daŠa iti khy€ta’, NS 2.3.15) and as mentioned above, it means both coercive power as well as justice. The opening verse of the N…tis€ra refers to the king as the wielder of daŠa (daŠadhara, NS 1.1.1). He maintains varŠ€rama dharma through daŠaakti (the power of daŠa) (NS 2.3.34). DaŠa must be exercised to ensure the protection and promotion of the prosperity of the praj€ (subjects), and there was a reciprocal relationship between the prosperity of the praj€ and the r€jan NS 1.1.14). K€mandaka offers various justifications for violence, referred to in one place as sim.havtti (‘the policy of a lion’) (NS 12.17.25). The most important of these is the attainment of desired ends, specifically the expansion and consolidation of political power. Violence is also justified on the grounds of what would result from its absence. In this world, according to K€mandaka (NS 2.3.40), people move about in different directions, trying to pursue their own interests by devouring others, as though out of greed for the latters’ flesh. DaŠa is necessary, otherwise mastyany€ya (‘the law of the fish,’ i.e., the big fish eating the smaller fish), the much favoured, enduring trope for disorder par excellence in many ancient Indian texts, prevails. The discussion of up€ûdaŠa (secret killing) includes advice on the modus operandi to kill adversaries, and the section on m€y€ describes various sly tactics to defeat enemies (NS 6.9.10–13). Enemies can be legitimately killed by the secret istration of poison or by enlisting the services of estranged court physicians (NS 9.14.70). Harshness or violence may also be necessary to deal with dishonest and impious people (NS 6.9.5), those who obstruct the course of dharma, or r€javallabhas who create trouble, individually or collectively (NS 6.9.10). K€mandaka recommends that royal favourites (npavallabhas) should be killed through up€ûdanda if they cause loss of lives and become a source of anxiety to the people (NS 18.27.11). Justice is another important justification for violence. However, the king must be careful to blend the use of coercive power (daŠa) with naya (legal procedure) in order to be praised as a yuktadaŠa (NS 15.23.12). He is urged to use daŠa as firmly as DaŠin (i.e., Yama), but blended with the impartiality of the nature of the earth and comion similar to that shown by the creator Praj€pati towards his own created beings (NS 3.6.1). Coercion must be tempered with justice and a sense of proportion, for excessively harsh punishment terrifies the people, just as leniency makes the king worthy of contempt (NS 6.9.15; p. 131). So far, all this is in conformity with the attitude of many ancient Indian texts. A difference in perspective emerges when K€mandaka speaks of three types of daŠa—capital punishment, fines and rigorous punishment causing bodily and mental suffering. There are two types of execution: open (prak€adaŠa) and secret (up€ûdaŠa). An intelligent ruler desirous of religious merit should not inflict capital punishment on Br€hmaŠas and men of dh€rmika disposition or on antyajas (‘outsiders’ or outcastes); the reason for excluding the latter is not made The Indian Economic and Social History Review, 47, 1 (2010): 29–62



explicit (NS 18.27.13). In fact, according to the N…tis€ra, capital punishment (pr€Š€ntika daŠa) should be avoided even for the gravest offence, with the exception of the most serious one, namely rajy€pah€ra (usurpation) (NS 15.23.16). K€mandaka’s disapproval of capital punishment is in sharp contrast to Kau˜ilya, who recommends the death penalty for several offences apart from those that are treasonable, from robbing the treasury to stealing or killing or inciting someone to steal or kill an animal belonging to the royal herd.47 Embedded in a political discourse that is peppered with disquisitions on violence, there is also mention of the virtue of ahiˆs€ (non-injury). Ahiˆs€, refined speech (sunt€ v€Š…), truthfulness (satya), purity (auca), pity (day€) and forgiveness (kam€) constitute the s€m€nya-dharma, the dharma which is applicable to all people, irrespective of varŠa and gender (NS 2.4.32). The code of conduct of the vijig…u includes avoiding the company of dishonest and unrighteous folk, offering to honest people and observing ahiˆs€ towards all beings (NS 14.21.51). This is in tune with the Artha€stra, which expresses similar views (1.3.13; 1.7.2). But K€mandaka’s distinctive stance on violence, already hinted at in his position on capital punishment, is reflected more clearly in discussions of specific issues such as the royal hunt and war. The King, the Forest and the Hunt In many cultures, in many chronological contexts, the royal hunt has been seen as a natural activity for kings, and in fact as an important expression of the king’s sovereignty. The importance of the royal hunt has been recognised by scholars in the context of the Mughal emperors,48 but not so in the context of earlier rulers of the subcontinent. The hunting expeditions of the Mughals are described in the Persian chronicles and represented in miniature paintings, and the authors of these chronicles did not consider hunting a problematic activity from either a pragmatic or an ethical point of view.49 The ancient Indian political theorists, on the other hand, had much to say on the matter. And one of the most significant aspects of the N…ts€ra is its opposition to the royal hunt, an activity that was considered by Kau˜ilya as integral to the king’s way of life. The forest and forest people loom large in the writings of political theorists. For K€mandaka (as for Kau˜ilya and the Sanskrit poets), the forest was a place associated with renunciants and ascetics. But more importantly, the political theorists recognised it as a place exceptionally rich in economic and military resources— especially elephants, which were greatly prized for their role in war—and where kings build forts (vanadurga). Elephant enclosures (i.e., forests) and regular forests AS 4.11.11–12; 2.5.17; 2.5.20; 2.29.16. See, for instance, Koch (Mughal Art and Imperial Ideology). 49 Divyabhanusinh Chavda, personal communication. 47 48



50 / UPINDER SINGH

were two of the eight sources of income (aãvarga) of the state (NS 5.8.78–79). Forest dwellers (€ãvikas) were by nature adh€rmika (impious), lubdha (greedy), an€rya (uncultured) and satyabhedin (untrustworthy) (NS 19.28.8). Forest troops (€raŠyaka/€ãvika bala) had to be used by the king in his military campaigns, but they were even more unreliable than troops alienated from the enemy camp (NS 19.28.9–10). From the king’s point of view, the forest was, at the end of the day, a lucrative, but problematic space. The forest was also a place where the king hunted, but K€mandaka had strong reservations about this activity. The dangers of the y€navyasana (the calamity of the march), described in the N…tis€ra just before those of mgay€vyasana (the calamity of the hunt), appear to apply to both (NS 15.23.19–22).50 These include the physical strain resulting from prolonged riding, accidental fall or injury, and the loss of horses or chariots. Further, there is the suffering caused by hunger, thirst, exhaustion, severe cold, storm, heat, and wastage of resources. Travelling through areas that are very hot, sandy or thorny, or dense forests infested with prickly creepers and shrubs, or hilly areas prone to falling boulders, or tracks that are uneven due to stones, earthen mounds and ant hills—all this causes much distress. Enemies may be lurking among rocks, rivers, or forests, and there is the possibility of sudden capture or death at the hands of s€mantas, €ãvikas and others. The other dangers specifically arising out of the mgay€vyasana include the possibility of the king being attacked by his own followers or kinsmen, captured by enemies, or mauled by bears, pythons, wild elephants, lions or tigers. He may lose his way in the forest, with the path obscured due to smoke arising from forest fires, and may be reduced to wandering around helplessly (NS 15.23.23–24). According to K€mandaka, these potential dangers can to some extent be neutralised by ensuring that the king rides on the back of a swift but easily controllable animal, by having the outskirts of the forest carefully examined and protected against all dangers, and ensuring that their interiors are well lit and rendered free of ferocious animals (NS 2.3.36). But this does not fully settle matters. For K€mandaka, as for Kau˜ilya, hunting, along with women, drinking and gambling, is a royal vice. But while K€mandaka is willing to accept moderate levels of indulgence in women and drink, gambling and hunting are to be shunned as far as is possible. Apart from the physical dangers it entails for the king, K€mandaka’s objection to hunting is also based on the fact that this activity could lead to the king wasting his time, and also that any kind of addiction weakens character. But his most important argument is that hunting is a great vyasana (mahat vyasana) because of the inherent evils of taking life (do€ƒ pr€Šahar€ƒ) (NS 15.23.23). To make sure that he has covered all the ground concerning this issue, K€mandaka lists the various supposed benefits of hunting (NS 15.23.25). He cites 50

Vyasana can mean vice, calamity or affliction.




the view that hunting provides the king with physical exercise which, in turn, results in his developing endurance, immunity from indigestion, heaviness and susceptibility to catching cold. Another argument proferred by some in favour of hunting is that it develops skill and excellence in hitting stationary or moving targets with arrows. These are, in fact, precisely the arguments made by Kau˜ilya in the Artha€stra (8.3.46). But K€mandaka firmly refutes them by asserting that all these benefits can be obtained through other means. For instance, maladies such as indigestion can be remedied through regular physical exercise, and marksmanship in archery by practising with artificial targets (NS 15.23.27). This is in sharp contrast to Kau˜ilya, who describes hunting as the least harmful and gambling as the most harmful of the vyasanas (AS 8.3). K€mandaka also suggests another interesting alternative (NS 15.23.28–40) to the regular royal hunt. Kau˜ilya too refers to this option, though very briefly. For a king who is unable to give up hunting, an artificial, sanitised game forest should be created (mg€raŠya), where he could hunt for sport (kr…€) alone.51 The features of this game forest are then specified: It should be located just outside the town (presumably the capital city), should be over half a yojan€ in length and breadth, and should be surrounded by a ditch and ramparts so that the animals cannot escape. It should be situated at the foot of a hill or next to a river, and should have plentiful supplies of water and grass. It should not have thorny creepers, shrubs, or poisonous plants. Any crevices in the ground should be filled up with earth and gravel, and the surface should be leveled by removing stumps of trees, mounds of earth and rocks. It should be made attractive with well-known flower-bearing and fruit-bearing trees providing pleasing, thick and cool shade. The pools in this park should be shallow, abounding in flowers and birds of different species, and cleared of ferocious aquatic animals. The park should have beautiful creepers laden with flowers and leaves inside and on the sides of the surrounding ditch. It should be provided with animals such as she-elephants and their young ones, tigers and other big game with their teeth and nails removed, and horned animals whose horns have been broken. A space outside the park should be cleared of trees and pillars, and the ground should be leveled, so that it is inaccessible to enemy forces and enhances the feeling of comfort for the king. The park should be guarded by trustworthy forest people who are resolute, hardy, painstaking, and conversant with the moods of wild animals. The king’s own men, of boundless energy and experienced in hunting, should introduce various wild animals into the park. The king may then enter it for sport, accompanied by a select group of 51 References to such a sanitised forest occur in AS 1.21.23; 2.20.3. AS 2.20.4 suggests that the king should in addition establish another kind of animal park where animals are welcomed as guests, presumably a sanctuary where they are given protection. Elsewhere (AS 2.26.1), Kau˜ilya talks of fines that should be imposed on those who kill animals whose killing has been prohibited or those that inhabit the king’s reserved park.



52 / UPINDER SINGH

trusted attendants, without detriment to his other duties. As he enters, fully armed soldiers should carefully stand guard outside, vigilant for signs of danger to their royal master. In his detailed description of an artificial, sanitised game forest for the royal hunt, K€mandaka offers a compromise between the royal predilection for hunting and the dangers and problems that this activity entailed.52 In spite of describing this option, it is noteworthy that K€mandaka disapproves of hunting not only on pragmatic grounds, i.e., the physical danger to the king and the possibility of it entailing a neglect of royal duties. He also objects to it on two moral grounds— i.e., it harms the king by weakening his character, and it involves violence against animals. The second argument indicates that it is not excessive hunting alone that is considered a problem; hunting itself is problematic because of the fact that it does violence to animals. This stand against an activity conventionally associated with kingship can be further connected with K€mandaka’s stand on the most violent of all political activities—war. The Problem of War In ancient and early medieval India, battles were fought (as they are now as well), for a variety of reasons including the control of land and resources and as an assertion of political hegemony, and given the endemic nature of war, it is not surprising that disquisitions on the subject abound in texts. In the context of ancient India, the Maurya emperor Aoka is considered the foremost exemplar and proponent of the principle of non-violence at the political as well as personal level. His thirteenth rock edict is, in fact, a remarkable document, giving a strong, reasoned critique of war, raising the discourse of kingship and conquest to a completely new level. However, although the political theorists recognised war as a necessary instrument of state policy, they also recognised its dangers and problematic nature. Moreover, attitudes towards war could and did differ. Ali briefly refers to the complex and ambivalent connection between courtly manners and violence, arguing that while military prowess and ritualised and honourable violence were important parts of courtly culture, they came to be tempered by irenic values and an emphasis on comion, kindness and gentility.53 My own perspective on the issue of violence is different. The focus here is not the 52 That this idea was well known is evident in K€lid€sa’s Raghuvaâ (9. 53) where Daaratha is described as going to hunt in a sanitised forest (described in much less detail). It is no coincidence that calamity befalls him there, when he inadvertently kills the son of a blind ascetic, resulting in the curse that leads to the subsequent tragic events. For what could be more graphic than this as an illustration of the calamities that would befall kings who fell prey to the vice of excessive hunting? 53 Ali, Courtly Culture and Political Life in Early Medieval India, pp. 99–102.




relationship between violence and courtly manners and culture, but on violence as a politico-ethical problem, and the ways in which the political theorists addressed this problem. General disquisitions on war (vigraha, yuddha) can be distinguished from the strategies to be adopted in military expeditions (y€na), and K€mandaka, like Kau˜ilya, discusses both. As mentioned earlier, the key player whose interests are central to the N…tis€ra is the vijig…u, and there are detailed discussions of military strategies and formations. In fact, the text culminates in a description of a successful military campaign. However, within all this, it also makes a strong case for the exercise of extreme caution in waging war, and the case it makes is not simply one of expediency. Thus, while there is much that is in common between the Artha€stra’s and N…tis€ra’s discussion of the conduct of inter-state relations, for instance in the idea of twelve elements in the r€jamaŠala (circle of kings) and the six strategies (guŠas), there is also much that is significantly different.54 Various aspects of war are discussed in the Vigrahavikalpa prakaraŠa (NS 10.15) and in other sections as well. The typology of war in the N…tis€ra includes the basic distinction between k™ãyuddha (secret war) and prak€ayuddha (open war) (NS 19.31.54). K™ãyuddha includes duping and enticing the enemy, nocturnal raids and setting up camouflaged encampments. K€mandaka asserts that the king does not transgress dharma by killing the enemy through the tactics of k™ãyuddha. The example given is that of Avatth€man killing the sons of the P€Šavas while they were asleep. K€mandaka also gives a detailed listing of the causes of war (NS 10.15.3–5) such as the usurpation of the kingdom, the abduction of women, the luring away of learned men and soldiers, the killing of friends and political rivalry. He also mentions a third kind of war—mantrayuddha (diplomatic warfare) (19.28.15–17).55 It may be noted that K€mandaka does not use Kau˜ilya’s wellknown typology of the types of conquerors—the dharmavijayin (who conquers for the sake of glory and is satisfied with the mere submission of the defeated king), lobhavijayin (who conquers out of greed and wants to obtain land, money, or both) and asuravijayin (who makes conquests like a demon, seizing the land, money, sons and wives of the conquered king and kills him) (AS 12.1.10). In the N…tis€ra, war is no longer graded according to a hierarchy of honour and propriety. The only relevant issues are its cost and chances of success. What is most significant from the point of view of political theory, especially when seen in the context of the endemic warfare of the time, is the fact that the 54 The six guŠas are sandhi (alliance), vigraha (war), y€na (marching on an expedition), €sana (staying in one place), saˆraya (seeking shelter) and dvaidh…bh€va (a double policy of sandhi with one king and vigraha with another). 55 Kau˜ilya too uses the categories of prak€ayuddha, k™ãyuddha and mantrayuddha, but he adds a fourth one—t™Š…myuddha (silent war), which involves the use of secret practices and instigation through secret agents (AS 7.6.41).



54 / UPINDER SINGH

N…tis€ra contains many different kinds of very specific arguments against war. Objections to war on pragmatic grounds are to be expected in the political treatises. The basic point on which the experts on politics agreed was that it was essential for the vijig…u to carefully assess the likely costs and consequences of war. The potential gains of war are territory (this is the most important), allies and wealth, and the king should embark on war only if there was a clear prospect of attaining these (NS 10.15.31). The N…tis€ra also recommends a long-term perspective on pragmatism, pointing out that political success does not hinge on a single victory. Like Paraur€ma, the king who commands respect from all through his prowess is the one who has to his credit many victories on different battlefields (NS 9.14.51). K€mandaka lists 16 types of war (vigraha) that should not be fought (NS 10.15. 19–23). Although there were those who thought otherwise, for K€mandaka, there was no point in embarking on war if the enemy was much more powerful and the chances of victory bleak.56 There was no justification for fighting a more powerful enemy, for clouds can never move in a direction opposite to that of the wind. Even if the enemy equalled the vijig…u in of resources, war could lead to death and destruction, sometimes of both parties. Both would perish like two unbaked pitchers striking against each other, like the demons SuŠa and UpasuŠa destroyed each other. Other wars that should not be fought included those for the sake of others or for the sake of women, those against venerable Br€hmaŠas, those that promised to be long drawn out, and those undertaken in times when troop movement was difficult. The risks of war were enhanced by the uncertainties it entailed, and there was no point risking what could be seen for unseen gains (NS 16.24.14). This is why a prudent king should avoid war, even when it was thrust on him. ‘As victory in war is always uncertain, it should not be launched without careful deliberation’ (NS 10.15.24). The policy of reeds (vaitas… vrtti) rather than that of snakes (bhauja‰g… vtti) should be followed, that it, it was better to be flexible rather than attack at the slightest provocation (NS 10.15.35–36). Kau˜ilya too briefly refers to the fact that war entails losses, expenses, marches away from home and hindrances (AS 7.2.2), but K€mandaka dwells on this issue in greater detail. Apart from the arguments against war based on expediency and the uncertainty of gains, the N…tis€ra, has many verses dilating on the inevitably disastrous results of war, especially one launched hastily without due consideration and consultation. If a ruler acts in a way contrary to the €stra and suddenly falls on an enemy, it is unlikely that he will be able to get out of this situation without feeling the impact of the enemy’s sword (NS 12.17.6). In the course of war, the king could, in a single instant, suffer the loss of wives, friends, allies, wealth, 56 Others differed on this point. For instance, K€mandaka cites Bharadv€ja as stating that a king should fight against the enemy with all his might, like a lion, and that it was possible for a weaker king to outmanouvre a stronger one through sheer courage (NS 9.14.56).




kingdom, fame and even his own life (NS 19.14.75). In view of the fact that war necessarily entails loss of men and resources, various difficulties and the death of principal officers, an intelligent ruler should not continue war, even if he has to willingly accept hardship, for war has inherently disastrous consequences (doas) (NS 19.14.72). Considering the constant anxiety and mental suffering resulting from war, the intelligent ruler should not indulge in frequent warfare (NS 19.14.74). Thus, K€mandaka argues persuasively, recourse to war, especially frequent war, must be avoided. K€mandaka also points to the fact that wars often served the selfish interests of of the political class other than the king. For instance, mantr…s (ministers) may desire a prolongation of war due to their self-interest, and a ruler who acts on their counsel may simply play into their hands (NS 12.17.41). The idea of setting his house in order before launching on fresh military campaigns is also emphasised when K€mandaka states that the vyasanas of state should be remedied before a ruler launches an attack against the enemy (NS 14.21.18). The text further points out that war was neither the only nor the best expedient (up€ya) that could be used by the vijig…u to achieve his ends. S€ma (conciliation), d€na (gifts), bheda (sowing dissension) were the well-known list of political expedients mentioned by Kau˜ilya. K€mandaka expands this list by adding three more—m€y€ (deceitful tactics), upek€ (indifference) and indraj€la (conjuring tricks) (NS 18.27.3). He argues that conciliatory measures should always be adopted to prevent war. The ancient political treatises refer to three types of power at the command of the king. Of these, K€mandaka describes mantraakti (the power of counsel) as superior to prabhuakti (the power of the lordship, i.e., military might) and uts€haakti (the power of energy) (NS 12.17.7). In this, he is of the same opinion as Kau˜ilya.57 Only by the possession of mantrabala does a ruler, following the track of naya, become capable of subjugating the powerful enemies who are like vicious serpents (NS 12.17.58). Implicit here is the idea that brute force is not the best option for maximising political gain. Kau˜ilya too drives home the point when he asserts that if the vijig…u uses excessive force, the circle of kings may rise against him and he may be destroyed, or that force (daŠa) cannot be used against a multitude of people. The political theorists were obviously keenly aware of the limits of the efficacy of force. The final, culminating prakaraŠa (NS 20.36) of the N…tis€ra deals with prak€ayuddha (the conduct of open war), and everything in the text seems to be leading up to the crisp description of a successful military charge against the enemy. But before getting to this point, K€mandaka has offered his audience abundant and diverse arguments to make his point that war must always be a last resort. 57

The Artha€stra too gives primacy to mantraakti (AS 9.1.29).



56 / UPINDER SINGH

Controlling the Controller While the goal of the N…tis€ra’s teaching is political success, an awareness of the possibilities of political malfunction through excess, imbalance and tyranny is ever-present. In certain situations, for instance if the king is excessively attached to dharma or artha, of if he is mentally ill, the functions of the vijig…u should be discharged by mantr…s possessing the requisite qualities (NS 14.21.60). The ability of the king to achieve his political ambitions hinged on his ability to effectively control the various praktis of the state. The N…tis€ra, like many other texts, recommends that the king cultivate discipline, self control and equanimity in himself, in princes, and among his subjects.58 The very first prakaraŠa of the N…tis€ra, the Indriyajaya prakaraŠa, deals with the topic of the control of the senses.59 The Artha€stra too emphasises the importance of discipline and control of the senses (AS 1.6.3). Restraint of the ions and self-control are among the important qualities (€tmasampad) of the . king (NS 4.7.15–19); in fact, these are his pre-eminent qualities, his €tmasamsk€ra (NS 4.7.4). These qualities were connected with a character trait that was greatly valorised by political theorists—vinaya, which was a cocktail of several things: discipline, good breeding, propriety, humility, modesty, mildness and good behaviour. Vinaya is a quality that a king should possess, one that should be inculcated in princes, and a crucial factor in defeat or victory (NS 1.2.70). Ali argues that that for of the political class, control of the senses was considered a precondition for the enjoyment of sensual and worldly pleasures and self-mastery and equanimity as essential for success in the courtly circle.60 Even if, as he suggests, the emphasis on vinaya in the court context functioned as a powerful worldly idea, a kind of internalised self-regulating mechanism that helped people succeed in getting ahead and also helped maintain the courtly hierarchy, its origins still require explanation. The source for the emphasis on self-control seems to be two-fold. First, as mentioned earlier, the discourse of political theorists was embedded in a larger philosophical discourse. The emphasis on self-control and equanimity in the N…tis€ra seems to have, in part, sprung from the philosophic underpinnings of the political discourse, the desirability of the control of the senses being a fairly widespread idea in ancient Indian philosophical systems.61 Early in the text, K€mandaka defines anv…kik… as that which develops the self-knowledge (€tmavidy€) that looks through happiness and sorrow,

The king, through daŠa, has to restrain his subjects running after vices (viayas) (NS 2.5.43). It should be noted that a very basic level of self-control is being advocated here. For instance, higher than control over the senses is €tmajaya (control over the self). 60 Ali, Courtly Culture and Political Life in Early Medieval India, pp. 138, 241, 245. 61 It features, for instance, in s€ˆkhya, yoga, Upaniadic, Buddhist and Jaina thought. 58 59




and asserts that it is by realising the true nature of joy and sorrow that the king renounces them both (NS 2.3.11). Numerous examples are given to prove the transience of life and its pleasures, and great emphasis is placed on the control of the sense organs (NS 3.6.9). Striking in its Upaniadic ring is the assertion (NS 4.7.78) that just as the antar€tma, residing in the midst of prakti (nature), permeates (samanute) the world consisting of moving and unmoving elements, similarly does the king, in the midst of the praktis (elements of the state) permeate the world consisting of moving and unmoving elements. As important as the text’s d€ranic underpinnings for understanding the emphasis on royal self-control was its politico-historical context. In an age of political aggrandisement, political theorists must have not only been concerned with the question of how the power of the king could be increased but also with how it could be contained. Emphasising the importance of mantraakti (the power of counsel) could only go so far. Ultimately, in ancient monarchical states, the only agent of effective control on the ambitions and transgressions of the king was the king himself. This may have been the second important element explaining K€mandaka’s (and Kau˜ilya’s) emphasis on the king controlling himself and his ions. The contradiction that the N…tis€ra offers—what may in fact be described as an important element in the classical Indian ideal of kingship—is that of a king who aspires to become a world conqueror but who is not moved by the lust for power, or for anything else, for that matter. Renunciation was built into the ideal prototype of the king, and this is reflected in the ideal of the rajari, an ideal which is pervasive in the cultural traditions of ancient and early medieval India.62 Historicising the N…tis€ra Historicising the N…tis€ra involves contextualising it within the genre of ancient and early medieval Indian political treatises and within the realities of monarchical power politics at the advent of the early medieval. The perspective represented in this text is that of a Br€hmaŠa political theorist who was probably closely involved in contemporary politics, addressing of the political class, including the king. While the text is broadly speaking ‘normative’ in nature, within this normative discourse, we can view a morphology of monarchical power politics and we can also see the author grappling with pressing issues of his time, including those related to unbridled and unsatiated royal ambitions and endemic war and violence. The leading political theorists of ancient and early medieval India had a similar socio-political background and shared similar concerns and conceptual vocabulary, ones which extended beyond their circle to other of the intelligentsia, including the poets. They participated in the creation of a basic common stock of 62

AS 1.7.3 in fact describes the life of such a r€jai.



58 / UPINDER SINGH

ideas and metaphors that became part of a relatively stable classical Indian model of kingship which, with regional and chronological variations, spread beyond the confines of the subcontinent into southeast Asia as well.63 Yet, within the parameters of this model, these thinkers had their distinct and distinctive positions and points of view as well. In fact, a close reading of the N…tis€ra reveals that the usual description of this text as an unoriginal versified summary of the teaching of the Artha€stra is incorrect. K€mandaka certainly drew on the Kau˜ilya’s ideas (and those of others as well), but he had his own point of view on several matters. His concerns and opinions can be gauged through a careful analysis of the issues he discusses, his arguments and emphases. Just as interesting as his assertions are his silences. There is only a hint of cleavages within the political elite, and it may be noted that apart from a brief reference to the crime of usurpation, K€mandaka does not directly discuss issues which must have been of pressing practical import such as disputed succession, coups and dynastic changes. The morphology of the state, royal court and household in the N…tis€ra corresponds broadly to those of the Artha€stra. And yet, in spite of shared rhetoric and imagery, K€mandaka lacks Kau˜ilya’s confident, even audacious vision of political power and empire. This must have been at least in part due to the fact that the core of the Artha€stra, was composed at least half a millennium, if not more, earlier, during a period of aggressive empire-building, while the N…tis€ra was composed in a very different political scenario, against the backdrop of imperial decline (of the Guptas and V€k€ãkas). The political battles were now among monarchical states and the oligarchies no longer figured as contenders among the circle of kings. Compared with still later texts, although the ideal of political paramountcy is very important, in the N…tis€ra, the s€mantas still seem to be bordering chiefs, and the discussion of the protocol between paramount and subordinate rulers is not as detailed or elaborate. The N…tis€ra’s ostensible aim was to reveal how the vijig…u could achieve his goal of political paramountcy and the text often has the ring of idealisation and universalisation, especially when it talks of the ideal virtues of the king. But there seems to be something more than banal idealisation or pious platitude here. Although many of the virtues that are described as desirable in a king are presented as inborn, they are actually cultivable, and the idea that is implicit is that there is a difference between a king who becomes king and one who is worthy of being one. 63 The continuities and variations in the classical model of kingship can be seen clearly in studies of the imaging of kingship in inscriptions belonging to different centuries and to different regions and sub-regions. For an analysis of the details and patterns revealed in the inscriptions of ancient and early medieval Orissa, see Singh, Kings, Br€hmaŠas and Temples in Orissa, pp. 82–122.




The entire discussion can be seen as an attempt of a political theorist to emphasise the ethical dimension of political discourse.64 A similar concern for building bridges between ethics and political realities can be seen in K€mandaka’s discussion of various forms of violence associated with kingship—punishment, hunting and war. While he does justify violent means in order to justify certain ends ( justice, the desire for exciting sport, and the goal of territorial expansion respectively), a careful reading of the text suggests a more complex and nuanced perspective. Along with advice, there is a great deal of onition and warning of the calamities that will afflict the kingdom if a king lacks the necessary virtues or abilities, or if the balance of virtue that is necessary for the other human agents in the sapt€‰ga r€jya is disturbed. The long deliberative sections on war, advocating extreme caution, suggest the despondency of a political thinker who disapproved of the frequent destructive warfare that marked his time. That this disapproval was part of K€mandaka’s larger convictions related to violence and non-violence is evident from his view on capital punishment. It is also evident in his diatribe against the royal hunt, which, contrary to Kau˜ilya’s view on the matter, is viewed as the worst of the royal vices, and is disapproved of not only on the grounds of expediency but also on the grounds that it involves moral weakening of the king and death to the hunted animals. These radical points of view have, strangely enough, hitherto gone unnoticed in works on ancient Indian political thought, which have incorrectly presented the N…t€s€ra as a feeble versified echo of the Artha€stra. Further, embedded in a text which seems to be a celebration of royal and political ambitions, is a strong insistence that the king exercise control over his senses. This insistence may have been a reaction to the disastrous results of the wanton, licentious lifestyle of many contemporary kings and/or as a reflection of the most basic form of self-control that was advocated by many d€ranic schools of the time. The Upaniadic ring of N…tis€ra 4.7.78, mentioned earlier in this essay, seems to betray a more specific philosophical orientation. But apart from the philosophical inspiration, the emphasis on self control can also be seen as an attempt of the political theorists to deal with a very central problem: How was the power of the king to be controlled and checked in a polity which lacked any institutional checks? Virtue, caution and power of counsel were emphasised again and again. But theorists such as K€mandaka recognised that ultimately, no external controls could be counted on, and the only real control on the king’s power was the one that he had to be persuaded to exercise over himself. 64 I disagree completely with scholars who argue (e.g., Ghoshal, A History of Indian Political Ideas, p. 385) that the politics of texts such as the Artha€stra and N…tis€ra is devoid of ethics. In fact ethics was central to ancient Indian discourse on politics.



60 / UPINDER SINGH

Incorporating Perspectives on Violence into Political History In spite of problems in dating their work precisely, it is essential to incorporate the ideas of the political theorists into historical writings on political processes in ancient and early medieval India. Such an exercise involves situating these texts within their historical context, a careful reading of their opinions and arguments on various issues, and a comparison between texts (and inscriptions) belonging to different periods of time. Comparison reveals much continuity in of concepts and vocabulary, but also indicates differences in perspective, and shifts in emphasis and nuance. We have seen that a comparison of the Artha€stra and N…tis€ra reflects a refashioning of the political model. While the Artha€stra can be seen as a brilliant exposition of the dizzy heights of power to which a king could aspire, inspired by the vision of an omniscient, omnipotent and omnipresent state, the N…tis€ra’s tone is more cautious and restrained. K€mandaka is certainly concerned with how a king could increase his power and dominion, but he is equally, if not more, concerned with how royal power, war and violence could be contained and controlled. The Artha€stra reflects an earlier model of an arrogant, absolutist state; the N…tis€ra represents a later, less exultant reflection on political power, one in which non-violence has significantly tempered the discussion of violence, especially with regard to punishment, the royal hunt and war. In the contexts of perspectives on war and hunting in ancient and early medieval times, it does not help to necessarily look for evidence of wholesale approval or disapproval of these activities.65 In fact, the assertion that the issue of ahiˆs€ was never raised in connection with the king’s exercising his own functions was never raised in ancient India (Vidal et. al, 2003: 17) is inaccurate.66 These were not questions which involved an either/or choice between the extremes of violence and non-violence (both in general or in their specific manifestations). There was a constant tension between these two poles, and a number of possibilities between them, and the implications of these possibilities were explored in various ways. It is this range of attitudes related to hiˆs€ and ahiˆs€ that require careful and extensive investigation.67 Apart from the Artha€stra and N…tis€ra, there are several other less-known texts on polity, most of them known through fleeting references in secondary 65 Reading between the lines also helps. Doing this, Malamoud (‘Remarks on Dissuasion in Ancient India’) identifies elements of what we call dissuasion in the Artha€stra. 66 Vidal et al., 2003, p. 17. 67 This is the focus of my larger study. For a discussion of different views within the Brahmanical tradition towards the specific issue of violence in sacrifice, within and across various periods of history, see Houben (‘To kill or not to kill the sacrificial animal (yajña-pau)?’). For an extremely interesting discussion of representations of war in the Kali‰kattupparaŠi, a twelfth century Tamil text, see Ali, ‘Violence, Gastronomy and the Meanings of War in Medieval South India’.




literature.68 We have seen how the subject of n…ti emerged from within the larger of discipline of artha€stra. But issues related to rulership and polity were simultaneously being also taken over and discussed in other kinds of works, principally the Dharma€stra texts and the epics, and they were also explored and expressed in poetry, drama, didactic tales and sayings, not only in Sanskrit but in other languages as well. The question of how political violence was conceptualised, defined, justified, delimited, criticised or condemned in texts belonging to different genres, languages and periods needs to be analysed, and the results of such an analysis should be part of the historian’s discourse on political processes in ancient and early medieval India. The sheer pervasiveness of political violence in human history, and the problem it presents in our own time, makes an engagement with this issue especially pertinent. Abbreviations AS = Artha€stra NS = N…tis€ra

References Ali, Daud. ‘Violence, Gastronomy and the Meanings of War in Medieval South India’, The Medieval History Journal, Vol. 3 (2), 2000, pp. 261–89. ———. Courtly Culture and Political Life in Early Medieval India, New Delhi, 2006. Aiyangar, K.V.R. Some Aspects of Ancient Indian Polity, 2nd edn., Patna, [1935] 1998. Biardeau, Madeleine. ‘Ancient Brahmanism, or Impossible Violence’, in Denis Vidal, Gilles Tarabout and Eric Meyer, eds, Violence/Non-violence: Some Hindu Perspectives, New Delhi, 2003, pp. 85–104. Feller Jatavallabhula, Danielle. ‘RaŠayajña: the Mah€bh€rata War as a Sacrifice’, in Jan E.M. Houben and Karel R. van Kooij, eds, Violence Denied: Violence, Non-Violence and the Rationalization of Violence in South Asian Cultural History, Leiden, 1999, pp. 69–103. Ghoshal, U.N. A History of Indian Political Ideas: The Ancient Period and the Period of Transition to the Middle Ages. Reprint edn, Oxford [1959], 1966. Gopal, Lallanji. ‘S€manta—its Varying Significance in Ancient India’, Journal of the Royal Asiatic Society, Vol. 5 (1–2), 1963, pp. 21–37. Houben, Jan E.M. ‘To Kill or Not to Kill the Sacrificial Animal (Yajña-Pau)? Arguments and Perspectives in Brahmanical Ethical Philosophy’, in Jan E.M. Houben and Karel R. van Kooij, eds, Violence Denied: Violence, Non-Violence and the Rationalization of Violence in South Asian Cultural History, Leiden, 1999, pp. 105–83. 68 For instance, Kane mentions several texts on n…ti such as the N…tinirŠ…ti of Yogghama, N…tiprak€a of Janamejaya, N…tiv€ky€mta and the Súkran…tis€ra. Sections of the last-mentioned of these may be as late as the 19th century (History of Dharma€stra, Vol. 1, Part I, pp. 155, 265, 277, 288).



62 / UPINDER SINGH Houben, Jan E.M. and Karel R. van Kooij, eds. Violence Denied: Violence, Non-Violence and the Rationalization of Violence in South Asian Cultural History, Leiden, 1999. Inden, Ronald. Text and Practice: Essays on South Asian History, New Delhi, 2006. Kane, P.V. History of Dharma€stra (Ancient and Mediaeval Religious and Civil Law in India), Vol. 1, Part I. Revised and enlarged 2nd reprint edn, Poona, 1990. Kangle, R.P. The Kau˜il…ya Artha€stra, Part III, Bombay, 1965. ———. The Kau˜il…ya Artha€stra, Part I: A Critical Edition with a Glossary, Bombay, 1970. Koch, Ebba. Mughal Art and Imperial Ideology: Collected Essays, New Delhi, 2001. Kulke, Hermann. ed., The State in India 1000–1700, Oxford in India Readings, Themes in Indian History Series, Delhi, 1997. Malamoud, Charles. ‘Remarks on Dissuasion in Ancient India’, in Denis Vidal, Gilles Tarabout and Eric Meyer, eds, Violence/Non-violence: Some Hindu Perspectives, New Delhi, 2003, pp. 209–18. Mitra, Rajendralala. ed., The N…tis€ra, or The Elements of Polity by K€mandaki, Revised with English Translation by Sisir Kumar Mitra, Calcutta, 1982. Pollock, Sheldon. ‘The Theory of Practice and the Practice of Theory in Indian Intellectual History’, Journal of the American Oriental Society, Vol. 105 (3), Indological Studies Dedicated to Daniel H.H. Ingalls, 1985, pp. 499–519. ———. ‘M…m€m. s€ and the Problem of History in Traditional India’, Journal of the American Oriental Society, Vol. 109 (4), 1989, pp. 603–10. ———. The Language of the Gods in the World of Men: Sanskrit, Culture and Power in Premodern India, Berkeley, 2006. Saletore, Bhasker Anand. Ancient Indian Political Thought and Institutions, New York, 1963. Singh, Upinder. Kings, Br€hmaŠas and Temples in Orissa: An Epigraphic Study, AD 300–1147, New Delhi, 1994. Trautmann, Thomas, R. Kau˜ilya and the Artha€stra: A Statistical Investigation of the Authorship and Evolution of the Text, Leiden, 1971. Varma, Vishwanath Prasad. Studies in Hindu Political Thought and its Metaphysical Foundations, 3rd revised and enlarged ed. Delhi, 1974. Vidal, Denis, Gilles, Tarabout, and Eric Meyer, eds. Violence/Non-violence: Some Hindu Perspectives, New Delhi, 2003. Willis, Michael. The Archaeology of Hindu Ritual: Temples and the Establishment of the Gods, New Delhi, 2009. Wujastyk, Dominik. The Roots of Ayurveda: Selections from Sanskrit Medical Writings, New Delhi, 2001. Zimmermann, Francis. ‘¬tus€tmya: The Seasonal Cycle and the Principle of Appropriateness’, Social Science and Medicine, 14B (2), 1980, pp. 99–106.



U.singh.ieshr. Politics, Violence. Kamandaka's Nitisara 4r5q2x

Overview 1s532p

More details 6h715l

Related Documents 2w1qw

U.singh.ieshr. Politics, Violence. Kamandaka's Nitisara 4r5q2x

Structural Violence g1m62

Collective Violence 82p5m

Workplace Violence-nursing 4t3c3c

Workplace Violence Case Study 6q2c1a

Assignment Violence In Sport 493k3a

More Documents from "Prathama Banerjee" 5h5653

U.singh.ieshr. Politics, Violence. Kamandaka's Nitisara 4r5q2x

Nike Brand Analysis 5g6p10

116360_pedewasan.docx 2u4b5h

Makalah Pengendalian Mikroorganisme 634y4y

Laporan Pendahuluan Ureterolithiasis 646zr

E-squared 65e1l