Democracy in Ancient India

by Steve Muhlberger, Associate Professor of History, Nipissing University.

Note on this article. I must state right out front that I read no Indian languages, which may lead some readers to dismiss entirely my work in this difficult field. For the more tolerant, let me explain that an earlier version of this article has been read and commented on by several academic readers, whose comments and corrections have been taken into account. The editors of the Journal of World History liked it well enough to ask me to write a broader treatment of democracy's prehistory. This resulted in Phil Paine and I writing "Democracy's Place in World History," which appeared in that journal in 1993. This article, however, never found a home of its own -- in part because I myself could think of few journals that would be interested in an article that concentrates on specialized material yet draws broad conclusions from it.

Returning to it now, in 1998, I find I still believe in my interpretation of the ancient evidence for Indian democracy, and in its relevance to how we understand the world history of democracy. Rather than let it languish further, I am releasing it electronically, for both general and specialist readers. I will be glad to hear your comments. For the reader who wants to look into the question independently, I have posted a bibliography, and of course there are always the footnotes.

 I should make clear that though this article bears my name alone, I was pointed in the right direction by an unpublished essay on democracy by Phil Paine. I also wish to note that I was aided in my research by the collection of Asian literature at Brock University, St. Catharines, Ontario. My philosopher-colleague at Nipissing University, Dr. Wayne Borody, made some suggestions, but neither he nor anyone else is responsibile for any errors or misinterpretations.

In fact, the supposed differences between "Western" and "non-Western" cultures are in this case, as in so many others, more a matter of ideological faith than of cool, impartial judgment. If we are talking about the history of humanity as a whole, democracy is equally new or equally old everywhere. Fair and effective elections, under adult suffrage and in conditions that allow the free discussion of ideas, are a phenomenon of this century. The history of democracy, properly so called, is just beginning. 

The "prehistory" of democracy, however, is scarcely restricted to Europe and Europeanized America and Australasia. A search of world history finds much worth studying. There are no perfect democracies waiting to be discovered, but there is something else: a long history of "government by discussion," in which groups of people having common interests make decisions that affect their lives through debate, consultation, and voting. The vast majority of such groups, it may be objected, are more properly called oligarchies than democracies. But every democracy has been created by widening what was originally a very narrow franchise. The history of government by discussion, which may be called republicanism for brevity's sake, has a claim to the interest of anyone who takes democracy seriously.2

This article will examine one important case of government by discussion -- the republics of Ancient India. Although they are familiar to Indologists, these republics are hardly known to other historians. They deserve, however, a substantial place in world historiography. The experience of Ancient India with republicanism, if better known, would by itself make democracy seem less of a freakish development, and help dispel the common idea that the very concept of democracy is specifically "Western." 

The present article has two goals. First, it will summarize the history of the ancient Indian republics as it is currently known. This survey is restricted to North India and the period before about 400 A.D., when sovereign republics seem to have become extinct. 

Second, the article will examine the historiographical evaluations of the Indian republican experience, and suggest that most of them have placed it in too narrow a context. Ancient Indian democratic experiments, it will be argued, are more important than they are usually granted to be. It is well known that the sources of ancient Indian history present considerable difficulties. All the indigenous ancient literature from the subcontinent has been preserved as part of a religious tradition, Brahmanical, Buddhist or Jaina. When the subject is political theory and its implementation, the preselected nature of sources is a distinct handicap to the researcher. The largest and most influential Indian literary tradition, the Brahmanical, is distinctly hostile to anything resembling democracy. 

Brahmanical literature gives kingship a central place in political life, and seldom hints that anything else is possible. For moral philosophers and legislators such as Manu (reputed author of the Manu-Smrti between 200 B.C.-A.D. 200), the king was a key figure in a social order based on caste (varna ). Caste divided society into functional classes: the Brahmans had magical powers and priestly duties, the ksatriyas were the rulers and warriors, the vaisyas cultivators, and the sudras the lowest part of society, subservient to the other three. Moral law or dharma depended on the observance of these divisions, and the king was the guarantor of dharma , and in particular the privileges of the Brahmans. 3 Another tradition is best exemplified by the Arthasastra of Kautilya (c. 300 B.C.), which alloted the king a more independent role but likewise emphasized his responsibility for peace, justice and stability.4

Both Kautilya's work and the Manu-Smrti are considered classic expressions of ancient Indian political and social theory. A reader of these or other Brahmanical treatises finds it very easy to visualize ancient Indian society as one where "monarchy was the normal form of the state." 5

Until the end of the last century, the only indication that this might not always have been the case came from Greek and Roman accounts of India, mostly histories of India during and just after Alexander the Great's invasion of India in 327-324 B.C. These works spoke of numerous cities and even larger areas being governed as oligarchies and democracies, but they were not always believed by scholars.6 Yet research into the Buddhist Pali Canon during the nineteenth century confirmed this picture of widespread republicanism. The Pali Canon is the earliest version of the Buddhist scriptures, and reached its final form between 400-300 B.C.7 It contains the story of Buddha's life and teaching and his rules for monastic communities. The rules and teachings are presented in the form of anecdotes, explaining the circumstances that called forth the Buddha's authoritative pronouncement. Thus the Pali Canon provides us with many details of life in ancient India, and specifically of the sixth century (the Buddha's lifetime) in the northeast. In 1903, T.W. Rhys Davids, the leading Pali scholar, pointed out in his book Buddhist India8 that the Canon (and the Jatakas, a series of Buddhist legends set in the same period but composed much later) depicted a country in which there were many clans, dominating extensive and populous territories, who made their public decisions in assemblies, moots, or parliaments. 

Rhys Davids' observation was not made in a vacuum. Throughout the nineteenth century, students of local government in India (many of them British bureaucrats) had been fascinated by popular elements in village life.9 The analysis of village government was part of a continuous debate on the goals and methods of imperial policy, and the future of India as a self-governing country. Rhys-Davids' book made the ancient institutions of India relevant to this debate. His reconstruction of a republican past for India was taken up by nationalistic Indian scholars of the 1910s.10 Later generations of Indian scholars have been somewhat embarrassed by the enthusiasm of their elders for early republics and have sought to treat the republics in a more balanced and dispassionate manner.11 Nevertheless, their work, like that of the pioneering nationalists, has been extremely productive. Not only the classical sources and the Pali Canon, but also Buddhist works in Sanskrit, Panini's Sanskrit grammar (the Astadhyayi ), the Mahabharata, the Jaina Canon, and even Kautilya's Arthasastra have been combed for evidence and insights. Coins and inscriptions have documented the existence of republics and the workings of popular assemblies. 

The work of twentieth century scholars has made possible a much different view of ancient political life in India. It has shown us a landscape with kings a-plenty, a culture where the terminology of rule is in the majority of sources relentlessly monarchical, but where, at the same time, the realities of politics are so complex that simply to call them "monarchical" is a grave distortion. Indeed, in ancient India, monarchical thinking was constantly battling with another vision, of self-rule by members of a guild, a village, or an extended kin-group, in other words, any group of equals with a common set of interests. This vision of cooperative self-government often produced republicanism and even democracy comparable to classical Greek democracy. 

Though evidence for non-monarchical government goes back to the Vedas, 12 republican polities were most common and vigorous in the Buddhist period, 600 B.C.-A.D. 200. At this time, India was in the throes of urbanization. The Pali Canon gives a picturesque description of the city of Vesali in the fifth century B.C. as possessing 7707 storied buildings, 7707 pinnacled buildings, 7707 parks and lotus ponds, and a multitude of people, including the famous courtesan Ambapali, whose beauty and artistic achievements contributed mightily to the city's prosperity and reputation. The cities of Kapilavatthu and Kusavati were likewise full of traffic and noise.13 Moving between these cities were great trading caravans of 500 or 1000 carts -- figures that convey no precise measurement, but give a true feeling of scale: caravans that stopped for more than four months in a single place, as they often did because of the rainy season, were described as villages.14 Religion, too, was taking to the road. The hereditary Brahman who was also a householder, as in later Vedic tradition, saw his teachings, authority and perquisites threatened by wandering holy men and self-appointed teachers.15

There were warlord-kings who sought to control this fluid society, some with a measure of success. But the literature, Pali and Sanskrit, Buddhist and Brahmanical, shows that non-monarchical forms of government were omnipresent. There was a complex vocabulary to describe the different types of groups that ran their own affairs.16 Some of these were obviously warrior bands; 17 others more peaceful groups with economic goals; some religious brotherhoods. Such an organization, of whatever type, could be designated, almost indifferently, as a gana or a sangha; and similar though less important bodies were labeled with the terms sreni, puga, or vrata. Gana and sangha, the most important of these terms, originally meant "multitude." By the sixth century B.C., these words meant both a self-governing multitude, in which decisions were made by the members working in common, and the style of government characteristic of such groups. In the case of the strongest of such groups, which acted as sovereign governments, the words are best translated as "republic." 

That there were many sovereign republics in India is easily demonstrated from a number of sources. Perhaps it is best to begin with the Greek evidence, even though it is not the earliest, simply because the Greek writers spoke in a political language that is familiar. 

Perhaps the most useful Greek account of India is Arrian's Anabasis of Alexander , which describes the Macedonian conqueror's campaigns in great detail. The Anabasis, which is derived from the eyewitness accounts of Alexander's companions, 18 portrays him as meeting "free and independent" Indian communities at every turn. What "free and independent" meant is illustrated from the case of Nysa, a city on the border of modern Afghanistan and Pakistan that was ruled by a president named Aculphis and a council of 300. After surrendering to Alexander, Aculphis used the city's supposed connection with the god Dionysus to seek lenient terms from the king: 

"The Nysaeans beseech thee, O king out of respect for Dionysus, to allow them to remain free and independent; for when Dionysus had subjugated the nation of the Indians...he founded this city from the soldiers who had become unfit for military service ...From that time we inhabit Nysa, a free city, and we ourselves are independent, conducting our government with constitutional order." 19

At last, however, after many years had gone, most of the cities adopted the democratic form of government, though some retained the kingly until the invasion of the country by Alexander.24

If we turn to the Indian sources, we find that there is nothing far-fetched about this idea. The most useful sources for mapping north India are three: The Pali Canon, which shows us northeastern India between the Himalayas and the Ganges in the sixth and fifth centuries B.C.; the grammar of Panini, which discusses all of North India, with a focus on the northwest, during the fifth century; and Kautilya's Arthasastra, which is a product of the fourth century, roughly contemporaneous with Megasthenes. All three sources enable us to identify numerous sanghas and ganas, some very minor, others large and powerful.27

What were these republican polities like? According to Panini, all the states and regions (janapadas ) of northern India during his time were based on the settlement or conquest of a given area by an identifiable warrior people who still dominated the political life of that area. Some of these peoples (in Panini's terms janapadins ) were subject to a king, who was at least in theory of their own blood and was perhaps dependent on their special support.28 Elsewhere, the janapadins ran their affairs in a republican manner. Thus in both kinds of state, the government was dominated by people classified as ksatriyas, or, as later ages would put it, members of the warrior caste. 

But in many states, perhaps most, political participation was restricted to a subset of all the ksatriyas . One needed to be not just a warrior, but a member of a specific royal clan, the rajanya.29 Evidence from a number of sources shows that the enfranchised members of many republics, including the Buddha's own Sakyas and the Licchavis with whom he was very familiar, considered themselves to be of royal descent, even brother-kings. The term raja, which in a monarchy certainly meant king, in a state with gana or sangha constitution could designate someone who held a share in sovereignty. In such places, it seems likely that political power was restricted to the heads of a restricted number of "royal families" (rajakulas) among the ruling clans. The heads of these families were consecrated as kings, and thereafter took part in deliberations of state. 

Our Indian republics are beginning to sound extremely undemocratic by our modern standards, with real power concentrated in the hands of a few patriarchs representing the leading lineages of one privileged section of the warrior caste. A reader who has formed this impression is not entirely mistaken. No doubt the rulers of most republics thought of their gana as a closed club -- as did the citizens of Athens, who also defined themselves as a hereditarily privileged group. But, as in ancient Athens, there are other factors which modify the picture, and make it an interesting one for students of democracy. 

First, the closed nature of the ruling class is easy to exaggerate. Republics where only descendants of certain families held power were common; but there was another type in which power was shared by all ksatriya families.31 This may not sound like much of a difference, since the restriction to the warrior caste seems to remain. But this is an anachronistic view of the social conditions of the time. The varnas of pre-Christian-era India were not the castes of later periods, with their prohibitions on intermarriage and commensality with other groups.32 Rather, they were the constructs of theorists, much like the division of three orders (priests, warriors and workers) beloved by European writers of the Early Middle Ages.33 Such a classification was useful for debating purposes, but was not a fact of daily existence. Those republics that threw open the political process to all ksatriyas were not extending the franchise from one clearly defined group to another, albeit a larger one, but to all those who could claim, and justify the claim, to be capable of ruling and fighting. 

Other evidence suggests that in some states the enfranchised group was even wider. Such a development is hinted at in Kautilya: according to him, there were two kinds of janapadas, ayudhiya-praya, those made up mostly of soldiers, and sreni-praya , those comprising guilds of craftsmen, traders, and agriculturalists.34 The first were political entities where military tradition alone defined those worthy of power, while the second would seem to be communities where wealth derived from peaceful economic activity gave some access to the political process. This interpretation is supported by the fact that sreni or guilds based on an economic interest were often both part of the armed force of a state and recognized as having jurisdiction over their own members.35 In the Indian republics, as in the Greek poleis or the European cities of the High Middle Ages, economic expansion enabled new groups to take up arms and eventually demand a share in sovereignty36 If it was not granted, one could always form one's own mini-state. Panini's picture of stable, long-established janapadas is certainly the illusion of a systematizing grammarian. As Panini's most thorough modern student has put it, there was "a craze for constituting new republics" which "had reached its climax in the Vahika country and north-west India where clans constituting of as many as one hundred families only organized themselves as Ganas."37 Furthermore, power in some republics was vested in a large number of individuals. In a well-known Jataka tale we are told that in the Licchavi capital of Vesali, there were 7707 kings (rajas), 7707 viceroys, 7707 generals, and 7707 treasurers.38 These figures, since they come from about half a millenium after the period they describe, have little evidentiary value, despite the ingenious efforts of scholars to find a core of hard fact. The tale does not give us the number of Licchavi ruling families (rajakulas), the size of the Licchavi assembly, or any real clues as to the population of Vesali.39 Yet the Jataka does retain the memory of an undisputed feature of Indian republicanism: the rulers were many.40 The same memory can be found in other sources, especially in those critical of republicanism. The Lalitavistara, in an obvious satirical jab, depicts Vesali as being full of Licchavi rajans , each one thinking, "I am king, I am king," and thus a place where piety, age and rank were ignored.41 The Santi Parva section of the Mahabharata shows the participation of too many people in the affairs of state as being a great flaw in the republican polity: 

The gana leaders should be respected as the worldly affairs (of the ganas) depend to a great extent upon them...the spy (department) and the secrecy of counsel (should be left) to the chiefs, for it is not fit that the entire body of the gana should hear those secret matters. The chiefs of gana should carry out together, in secret, works leading to the prosperity of the gana , otherwise the wealth of the gana decays and it meets with danger.42

The numerous members of a sovereign gana or sangha interacted with each other as members of an assembly. Details of the working of such assemblies can be found both in Brahmanical and Buddhist literature. By the time of Panini (fifth century B.C.), there was a terminology for the process of corporate decision-making. Panini gives us the terms for vote, decisions reached by voting, and the completion of a quorum. Another cluster of words indicates that the division of assemblies into political parties was well known. Further, Panini and his commentators show that sometimes a smaller select group within a sangha had special functions -- acting as an executive, or perhaps as a committees for defined purposes.44

The Pali Canon gives a much fuller, if somewhat indirect, depiction of democratic institutions in India, confirming and extending the picture found in Panini. This is found in three of the earliest and most revered parts of the canon, the Maha-parinibbana-suttanta, the Mahavagga, and the Kullavagga.45 These works, taken together, preserve the Buddha's instructions for the proper running of the Buddhist monastic brotherhood -- the sangha -- after his death. They are the best source for voting procedures in a corporate body in the earliest part of the Buddhist period. They also give some insight into the development of democratic ideology. 

The rules for conducting the Buddhist sangha were, according to the first chapter of the Maha-parinibbana-suttanta, based in principle on those commonly found in political sanghas or ganas. In the case of the Buddhist sangha, the key organizational virtue was the full participation of all the monks in the ritual and disciplinary acts of their group. To assure that this would be remembered, detailed rules concerning the voting in monastic assemblies, their membership, and their quorums, were set down in the Mahavagga and the Kullavagga . 

Business could only be transacted legitimately in a full assembly, by a vote of all the members. If, for example, a candidate wanted the upasampada ordination, the question (ñatti) was put to the sangha by a learned and competent member, and the other members asked three times to indicate dissent. If there was none, the sangha was taken to be in agreement with the ñatti. The decision was finalized by the proclamation of the decision of the sangha.46

In many cases, as in the granting of upasampada ordination, unanimity of a full assembly was required.47 Of course, unanimity was not always possible. The Kullavagga provides other techniques that were used in disputes especially dangerous to the unity of the sangha, those which concerned interpretation of the monastic rule itself. If such a dispute had degenerated into bitter and confused debate, it could be decided by majority vote, or referred to a jury or committee specially elected by the sangha to treat the matter at hand.48

It is here that we see a curious combination of well-developed democratic procedure and fear of democracy. The rules for taking votes sanctioned the disallowance by the vote-taker of results that threatened the essential law of the sangha or its unity.49 Yet, if the voting procedure is less than free, the idea that only a free vote could decide contentious issues is strongly present. No decision could be made until some semblance of agreement had been reached.50 Such manipulations of voting were introduced because Buddhist elders were very concerned about the survival of the religious enterprise: disunity of the membership was the great fear of all Indian republics and corporations.51 Yet the idea of a free vote could not be repudiated. The Kullavagga illustrates a conflict within the Buddhist sangha during its earliest centuries between democratic principles and a philosophy that was willing in the name of unity to sacrifice them. 

Since the rules of the Buddhist sangha are by far the best known from the period we have been discussing, it is tempting to identify them with the rules of political ganas, particularly those of the Licchavis (or Vajjians), since the Buddha made a clear connection between the principles applicable to the Licchavi polity and those of his sangha.52 But from early on, scholars have recognized that the Buddhist constitution was not an exact imitation of any other: for instance, sovereign republics had a small, elected executive committee to manage the affairs of the gana when the whole membership of the gana was unable to be assembled.53 But neither did the Buddha or his earliest followers invent their complex and carefully formulated parliamentary procedures out of whole cloth. R.C. Majumdar's conclusion, first formulated in 1918, still seems valid: the techniques seen in the Buddhist sangha reflect a sophisticated and widespread political culture based on the popular assembly.54

Similarly, the value placed on full participation of members in the affairs of their sangha must reflect the ideology of those who believed in the sangha-gana form of government in the political sphere. The Buddha's commitment to republicanism (or at least the ideal republican virtues) was a strong one, if we are to believe the Maha-parinibbana-suttanta, among the oldest of Buddhist texts.55 As is common in the Buddhist scriptures, a precept is illustrated by a story. Here Ajatasastru, the King of Maghada, wishes to destroy the Vajjian confederacy (here = the Licchavis) 56 and sends a minister, Vassakara the Brahman, to the Buddha to ask his advice. Will his attack be a success? Rather than answer directly, the Buddha speaks to Ananda, his closest disciples: 

"Have you heard, Ananda, that the Vajjians hold full and frequent public assemblies?" 

"Lord, so I have heard," replied he. 

"So long, Ananda," rejoined the Blessed One, "as the Vajjians hold these full and frequent public assemblies; so long may they be expected not to decline, but to prosper...

"So long, Ananda, as the Vajjians meet together in concord, and rise in concord, and carry out their undertakings in concord...so long as they enact nothing not already established, abrogate nothing that has been already enacted, and act in accordance with the ancient institutions of the Vajjians as established in former days...so long as they honor and esteem and revere and support the Vajjian elders, and hold it a point of duty to hearken to their words...so long as no women or girls belonging to their clans are detained among them by force or abduction...so long as they honor and esteem and revere and support the Vajjian shrines in town or country, and allow not the proper offerings and rites, as formerly given and performed, to fall into desuetude...so long as the rightful protection, defense, and support shall be fully provided for the Arahats among them, so that Arahats from a distance may enter the realm, and the Arahats therein may live at ease -- so long may the Vajjians be expected not to decline, but to prosper." 

Then the Blessed One addressed Vassakara the Brahman, and said, "When I was once staying, O Brahman, at Vesali at the Sarandada Temple, I taught the Vajjians these conditions of welfare; and so long as those conditions shall continue to exist among the Vajjians, so long as the Vajjians shall be well instructed in those conditions, so long may we expect them not to decline, but to prosper."

The same story tells us that once the king's envoy had departed, the Buddha and Ananda went to meet the assembly of monks. Buddha told the monks that they too must observe seven conditions if they were to prosper: Full and frequent assemblies, concord, preserving and not abrogating established institutions, honoring elders, falling "not under the influence of that craving which, springing up within them, would give rise to renewed existence," delighting in a life of solitude, and training "their minds that good and holy men shall come to them, and those who have come shall dwell at ease." 57 These precepts, and others that follow in sets of seven, were the main point for the monks who have transmitted the Maha-parinibbana-suttanta to us. We, however, may wish to emphasize another point: the Buddha saw the virtues necessary for a righteous and prosperous community, whether secular or monastic, as being much the same. Foremost among those virtues was the holding of "full and frequent assemblies." In this, the Buddha spoke not only for himself, and not only out of his personal view of justice and virtue. He based himself on what may be called the democratic tradition in ancient Indian politics -- democratic in that it argued for a wide rather than narrow distribution of political rights, and government by discussion rather than by command and submission.58

The Pali Canon gives us our earliest, and perhaps our best, detailed look at Indian republicanism, its workings, and its political philosophy. About no other republics do we know as much as we do about the Buddhist sangha and the Licchavis in the time of Buddha -- even though we do know that republics survived and were a significant factor until perhaps the fourth century A.D., a period of over 800 years. Scattered inscriptions, a great number of coins, and the occasional notice in Greek sources, the Jatakas or other Indian literature give us a few facts. But any history of Indian republicanism is necessarily a rather schematic one. 

The theme that has most attracted the attention of scholars is the constant danger to republicanism, and its ultimate failure. Much of what we know about the sovereign ganas of India derives from stories of attacks upon them by various conquerors. Yet it is remarkable that for several centuries, the conspicuous successes of monarchs, even the greatest, had only a temporary effect on the sovereign republics and very little effect indeed on the corporate organization of guilds, religious bodies, and villages. The reason is, of course, that Indian kings have seldom been as mighty as they wished to be, or wished to be presented. Conquerors were not in a position to restructure society, to create states as we visualize them today. Rather they were usually content to gain the submission of their neighbors, whether they were other kings or republics.59 These defeated rivals were often left in control of their own affairs, merely required to pay tribute and provide troops for the conquerors next war. The great emperors of ancient India, including Chandragupta Maurya and Asoka, ran rather precarious realms. Once the center weakened, these unraveled very quickly, and society returned to its preceding complexity. Rival dynasties revived, as did defeated republics.60

As Altekar recognized, the mere existence of warlords was not fatal to the republican tradition of politics. Far more important was the slow abandonment of republican ideals by republicans themselves. We have seen that many republics were content even in the earliest days with a very exclusive definition of the political community. In some, ideas of wider participation gained currency and even implementation. But the contrary movement is easier to document. By the third and fourth centuries A.D., states known to be republics in earlier times were subject to hereditary executives. Eventually such republics became monarchies.61

An evolution away from republicanism is clearly seen in the literature of politics and religion. If we grant that the society depicted by the Pali Canon is the beginning of a new era, one with an economy and culture quite distinct from the Vedic period, it immediately becomes obvious that the most democratic ideals are the earliest. The Pali Canon, and to some extent the Jaina Canon, show us energetic movements that rejected the hierarchialism and caste ideology seen in the Vedas and Brahmanas in favor of more egalitarian values. Buddhism and Jainism were scarcely exceptional: they are merely the most successful of many contemporary religious movements, and left us records. It is clear from Panini that egalitarianism was an important element in the fifth century B.C.: he preserves a special term for the gana where "there was no distinction between high and low." 62

Such Brahmanical classics as the Mahabharata, the writings of Kautilya and the Manu-Smrti, works that promoted hierarchy, are manifestations of a later movement (300 B.C.-200 A.D.) away from the degree of egalitarianism that had been achieved. Kautilya, who is traditionally identified with the chief minister of the Mauryan conqueror Chandragupta Maurya (fl. after 300 B.C.), is famous for his advice to monarchs on the best way to tame or destroy ganas through subterfuge; perhaps a more important part of his achievement was to formulate a political science in which royalty was normal, even though his own text shows that ganas were very important factors in the politics of his time.63 Similarly, the accomplishment of the Manu-Smrti was to formulate a view of society where human equality was non-existent and unthinkable. 

Members of ganas were encouraged to fit themselves into a hierarchical, monarchical framework by a number of factors. Kings were not the only enemies of the ganas . The relationships between competing ganas must have been a constant political problem. Ganas that claimed sovereignty over certain territory were always faced by the competing claims of other corporate groups.64 How were these claims to be sorted out, other than by force? The king had an answer to this question: if he were acknowledged as "the only monarch [i.e. raja, chief executive] of all the corporations," 65 he would commit himself to preserving the legitimate privileges of each of them, and even protect the lesser members of each gana from abuse of power by their leaders. It was a tempting offer, and since the alternative was constant battle, it was slowly accepted, sometimes freely, sometimes under compulsion. The end result was the acceptance of a social order in which many ganas and sanghas existed, but none were sovereign and none were committed to any general egalitarian view of society. They were committed instead to a hierarchy in which they were promised a secure place.66 Such a notional hierarchy seems to have been constructed in North India by the fifth century A.D. Even the Buddhist sangha accommodated itself to it -- which led eventually to the complete victory of the rival Brahmans. 

This was not quite the end of republicanism, because "government by discussion" continued within many ganas and sanghas ; but the idea of hierarchy and inequality, of caste, was increasingly dominant. The degree of corporate autonomy in later Indian society, which is considerable and in itself a very important fact, is in this sense a different topic that the one we have been following. A corporation that accepts itself as a subcaste in a great divine hierarchy is different from the more pugnacious ganas and sanghas of the Pali Canon, Kautilya or even the Jataka stories. 

What have modern historians made of what we might call the golden age of Indian republicanism? We have already distinguished above between two eras of scholarship on the topic. In the first, patriotic enthusiasm and the simple thrill of discovery of unsuspected material characterized scholars' reactions. The former attitude was especially seen in K.P. Jayaswal's Hindu Polity . Published first in article form in 1911-1913, then as a book in 1924, Jayaswal's work was avowedly aimed to show that his countrymen were worthy of independence from Britain. The history of "Hindu" institutions demonstrated an ancient talent for politics: 

The test of a polity is its capacity to live and develop, and its contribution to the culture and happiness of humanity. Hindu polity judged by this test will come out very successfully...The Golden Age of [the Hindu's] polity lies not in the Past but in the Future... Constitutional or social advancement is not a monopoly of any particular race.67

In the second period of scholarship, in the years since independence, a more restrained attitude has been adopted by younger scholars who feel they have nothing to prove. Among these scholars the general tendency has been to emphasize that the republics were not real republics, in the modern usage that implies a universal adult suffrage. The clan-basis and the exclusiveness of the ruling class are much discussed. Sometimes writers have bent over backwards to divorce the Indian republican experience from the history of democracy: 70 thus A.K. Majumdar's judgement that because in a gana-rajya "all inhabitants other than the members of the raja-kulas [had] no rights [and] were treated as inferior citizens," people were actually better off in the monarchies, where "if not the general mass, at least the intellectuals and the commercial community enjoyed freedom in a monarchy, which seems to have been lacking in a gana-rajya." 71 The contrast drawn here is not backed up by any real argument, and makes one wonder about the how the author defines "freedom." 

The reaction has perhaps gone too far.72 One feels that modern scholars have still not come to grips with the existence of widespread republicanism in a region so long thought to be the home par excellence of "Oriental Despotism." 73 Republicanism now has a place in every worthwhile book about ancient India, but it tends to be brushed aside so that one can get back to the main story, which is the development of the surviving Hindu tradition.74 Historians, in India as elsewhere, seem to feel that anything which could be so thoroughly forgotten must have had grievous flaws to begin with.75 Most historians still cannot discuss these republics without qualifying using the qualifiers "tribal" or "clan."76 Long ago Jayaswal rightly protested against the use of these terms: "The evidence does not warrant our calling [republics] 'clans.' Indian republics of the seventh [sic] and sixth centuries B.C...had long passed the tribal stage of society. They were states, Ganas and Samghas, though many of them likely had a national or tribal basis, as every state, ancient or modern, must necessarily have." 77 He was equally correct when he pointed out that "Every state in ancient Rome and Greece was 'tribal' in the last analysis, but no constitutional historian would think of calling the republics of Rome and Greece mere tribal organizations." 78

Yet the phrases "clan-" and "tribal-republic" are still routinely used today in the Indian context, and it is difficult to avoid the conclusion that they are being used perjoratively. In both common and scholarly usage, to label a people's institutions or culture as tribal is to dismiss them from serious consideration. "Tribespeople" are historical dead-ends, and their suppression or absorption by more advanced cultures (usually those ruled by centralizing governments) is taken for granted.79 The terminology of even Indian historians demonstrates the survival of an ancient but inappropriate prejudice in the general evaluation of Indian republicanism. 

Once that prejudice is overcome, Indian republicanism gains a strong claim on the attention of historians, especially those with an interest in comparative or world history. 

It is especially remarkable that, during the near-millenium between 500 B.C. and 400 A.D., we find republics almost anywhere in India that our sources allow us to examine society in any detail. Unless those sources, not least our Greek sources, are extremely deceptive, the republics of India were very likely more extensive and populous than the poleis of the Greeks.80 One cannot help wondering how in many other parts of Eurasia republican and democratic states may have co-existed with the royal dynasties that are a staple of both ancient and modern chronology and conceptualization. This may well be an unanswerable question, but so far no one has even tried to investigate it. If an investigation is made, we may discover things that are as surprising to us as the republics of India originally were. 

The existence of Indian republicanism is a discovery of the twentieth century. The implications of this phenomenon have yet to be fully digested, because historians of the past century have been inordinately in love with the virtues of centralized authority and government by experts, and adhered to an evolutionary historicism that has little good to say about either direct or representative democracy. Perhaps the love affair is fading. If so, historians will find, in the Indian past as elsewhere, plenty of raw material for a new history of the development of human government. 

Notes for "Democracy in Ancient India"

Also, references to Indian primary materials will be made to English translations (where available). Nearly all the secondary literature on the topic is in English. 

1. See for example Herodotus, The Histories 7. 135, trans. Aubrey de Sélincourt, rev. ed. (Harmondsworth, 1972), p. 485: the famous reply of the Spartan emissaries to the Persian general Hydarnes. Back to text.

2. For more on this, see Steven Muhlberger and Phil Paine, "Democracy's Place in World History," Journal of World History 4 (1993): 23-45 and the World History of Democracy site, especially Chapter Two -- Democracy at the Basic Level: Government by consent in small communities. Back to text.

3. A.S. Altekar, State and Government in Ancient India, 3rd edn. rev. and enlarged (Delhi, 1958; first ed. 1949), p. 1; the Manu-Smrti translated by G. Bühler as The Laws of Manu, vol. 25 of Sacred Books of the East, hereafter SBE] ed. F. Max Müller (Oxford, 1886). Back to text.

4. Kautilya's Arthasastra, trans. by R. Shamasastry, 4th ed. (Mysore, 1951; first ed. 1915). Back to text.

5. Altekar, State and Government in Ancient India, p. 1 (hereafter State and Government ); but see the same work, p. 109, where the statement is qualified as a prelude to discussing republics. Back to text.

6. Altekar, State and Government, pp. 110-111; K.P. Jayaswal, Hindu Polity: A Constitutional History of India in Hindu Times, 2nd. and enl. ed. (Bangalore, 1943), p. 58. Back to text.

7. An introduction to the Pali Canon may be found in R.C. Majumdar, The History and Culture of the Indian People, vol. 2, The Age of Imperial Unity, (Bombay, 1951), pp. 396-411. Back to text.

8. (London, 1903). Back to text.

9. See, for instance, Sir Henry Sumner Maine, Village Communities in the East and West (1889; reprint edn. New York, 1974). Back to text.

10. K.P. Jayaswal, Hindu Polity: A Constitutional History of India in Hindu Times 2nd and enl. edn. (Bangalore, 1943), published first in article form in 1911-13; D.R. Bhandarkar, Lectures on the Ancient History of India on the Period form 650 to 325 B.C., The Carmichael Lectures, 1918 (Calcutta, 1919); R.C. Majumdar. Corporate Life in Ancient India, (orig. written in 1918; cited here from the 3rd ed., Calcutta, 1969, as Corporate Life). Back to text.

11. E.g. Altekar (n. 6); J.P. Sharma, Republics in Ancient India, c. 1500 B.C.-500 B.C. (Leiden, 1968) [hereafter Republics]; U.N. Ghoshal, A History of Indian Public Life, vol. 2, The Pre-Maurya and Maurya Period (Oxford, 1966). For the embarrassment, see Sharma, Republics, pp. 2-3. Back to text.

12. Sharma, Republics, pp. 15-62, 237. Back to text.

13. Narendra Wagle, Society at the Time of the Buddha (Bombay: 1966), pp. 27-28. Back to text.

14. Wagle, Society at the Time of the Buddha, pp. 147-148. Back to text.

15. Sukumar Dutt, Buddhist Monks and Monasteries of India (London, 1957), pp. 35-44. Back to text.

16. V.S. Agrawala, India as Known to Panini: A study of the cultural material in the Ashatadhyayi, 2nd edn. rev. and enl. (Varanasi, 1963), pp. 426-444 [hereafter, Panini]; Sharma, Republics, pp. 8-14. A.K. Majumdar, Concise History of Ancient History, vol. 2: Political Theory, Administration, and Economic Life (New Delhi, 1980), p. 131 [hereafter, Concise History]. Back to text.

17. It is often assumed in the literature that mercenary bands or wild tribes must be clearly distinguished from true political communities. A reading of Xenophon's Anabasis (trans. by W.H.D. Rouse as The March Up Country (New York and East Lansing, 1959)) would give food for thought about this distinction. The army Xenophon was part of and led for a time is perhaps the best documented example of the day-to-day political life of a Greek community that we have. Back to text.

18. See "Arrianus, Flavius" Oxford Classical Dictionary, 2nd edn. (Oxford, 1970), pp. 122-123. Back to text.

19.. Arrian 5.1-2; all translations from the Greek sources are taken from R.C. Majumdar's compilation, The Classical Accounts of India (Calcutta, 1960) [hereafter Classical Accounts] -- in this case, p. 20. However, those who don't have access to that handy work can find these authors, whose books are all well-known classical works, in standard editions and translations. Back to text.

20. Arrian, 5.22, 5.25-6.14, Classical Accounts, pp. 47, 64-75. Back to text.

21. Q. Curtius Rufus, History of Alexander the Great 9.8, Classical Accounts, p. 151; Diodorus Siculus, Bibliotheca Historica 17.104, Classical Accounts, p. 180. Back to text.

22. Ibid. Back to text.

23. Altekar, State and Government, p. 111. Back to text.

24. Diodorus Siculus 2.39, Classical Accounts, p. 236; cf. Arrian's Indika 9, Classical Accounts, p. 223, which seems to derive from the same source, i.e. Megasthenes, for whom see below. Back to text.

25. Otto Stein, "Megasthenes (2)," Real-Encyclopädie der classischen Altertumwissenschaft, ed. A. von Pauly, G. Wissowa, et. al. (Stuttgart, 1893-) vol. 15, pt. 1, col. 232-3. Back to text.

26. R.C. Majumdar, Classical Accounts, Appendix I, pp. 461-473, throws doubt on the authority of this whole section of Diodorus (2.35-42, called "the Epitome of Megasthenes,"), but classicists do not share his doubts, though they grant that the original material may have been handled roughly by later epitomizers. See Otto Stein, "Megasthenes (2)," col. 255; Barbara C.J. Timmer, Megasthenes en de Indische Maatschaapij (Amsterdam, 1930); Diodorus of Sicily, trans. by C.H. Oldfather Cambridge, Mass. and London, 1935), vol. 2, p. vii. Back to text.

27. Kautilya, 11.1; Agrawala, Panini, pp. 445-457; see the short history of known republics in Altekar, State and Government pp. 118-123. See Joseph E. Schwartzenberg, ed., A Historical Atlas of South Asia (Chicago and London, 1978), p. 16 (Plate III.B.2). Back to text.

28. Agrawala, Panini, pp. 426-428; Benoychandra Sen, Studies in the Buddhist Jatakas: Tradition and Polity (Calcutta, 1974), pp. 157-159. Back to text.

29. Agrawala, Panini, pp. 430-432. Back to text.

30. Altekar, State and Government, p. 135; Sharma, Republics, pp. 12-13, 99-108, 112, 175-176. Back to text.

31. Altekar, State and Government, p. 114. Back to text.

32. Wagle, Society at the Time of the Buddha, pp. 132-33, 156-158. Back to text.

33. Georges Duby, The Three Orders: Feudal Society Imagined, tr. Victor Goldhammer (Chicago, 1980); Jacques Le Goff, "Labor, Techniques and Craftsmen in the Value Systems of the Early Middle Ages (Fifth to Tenth Centuries)," in Time, Work, and Culture in the Middle Ages, tr. Victor Goldhammer (Chicago, 1980), pp. 71-86. Back to text.

34. Agrawala, Panini, pp. 436-439. Contra, Ghoshal, A History of Indian Public Life, ii, p. 195, n. 5, who rejects Agrawala's interpretation of the evidence in Panini and Kautilya, and insists on a strict (but anachronistic) division between political, military, and social and economic groups. A fair reading of Kautilya shows that "corporations" of whatever sort could be important political and military factors, whether they were sovereign or not, and whether they "lived by the name of raja" (Kautilya, 11.1, tr. Shamasastry, p. 407) or not. Back to text.

35. See esp. R.C. Majumdar, Corporate Life, pp. 18-29, 60-63; Charles Drekmeier, Kingship and Community in Early India (Stanford, 1962), pp. 275-277. Back to text.

36. W.G. Forrest, The Emergence of Greek Democracy, 800-400 B.C. (New York, 1966), esp. pp. 67-97; J.K. Hyde, Society and Politics in Medieval Italy: The Evolution of the Civil Life, 1000-1350, esp. 48-60, 104-118; John Hine Mundy, Liberty and Political Power in Toulouse 1050-1230 (New York, 1954). Back to text.

37. Agrawala, Panini, p. 432. Again cf. Italy at the beginning of the High Middle Ages, Hyde, Society and Politics in Medieval Italy, pp. 56-57. Back to text.

38. Jataka 149, trans. in The Jataka, or Stories of the Buddha's Former Births, ed. E.B. Cowell, tr. by Various Hands, 6 vols. (1895; reprint, London, 1957), 1: 316. Jataka 301 (Cowell trans., 3: 1) also mentions 7707 kings, "all of them given to argument and disputation." Back to text.

39. Every scholar to approach this material has wrestled with this number, none more diligently than Sharma, Republics, pp. 99-104. It is hard to take any of them very seriously once one has examined Jataka 149 itself. Here, as in many other places, 7077 is used as a large, ideal number. Back to text.

40. Similarly suggestive numbers can be found in Jataka 465 (Cowell trans., 4: 94) where 500 Licchavi kings (not necessarily the entire body of kings) are mentioned; in the Mahavastu, which refers to "twice 84,000 Licchavi rajas residing within the city of Vesali," (Sharma, Republics, p. 99; the Mahavastu is yet untranslated into a European language) and Jataka 547 (Cowell trans., 6: 266), which mentions 60,000 ksatriyas in the Ceta state, all of whom were styled rajano (Agrawala, Panini, p. 432). Back to text.

41. Agrawala, Panini, p. 430; Sharma, Republics, p. 101; A.K. Majumdar, Concise History, 2: 140. No translation of the Lalitavistara into a European language was available to me. Back to text.

42. Mahabharata 12.107, trans. by R.C. Majumdar, Corporate Life, 251. Back to text.

43. A.K. Majumdar, Concise History, 2: 140, referring to Acharangasutra II.3.1.10. The SBE translation of the Acharangasutra (vol. 22 (1884), tr. Hermann Jacobi) of this passage entirely conceals the meaning of gana. This is typical of older translations, and some not so old (e.g. the Roy trans. of the Mahabharata, Santi Parva (Calcutta, 1962), c. 107, where Roy insists that gana here must be understood as denoting an aristocracy of wealth and blood). Back to text.

44. Agrawala, Panini, pp. 433-435. Back to text.

45. The Maha-parinibbana-suttanta: Buddhist Suttas vol. 1, tr. T.W. Rhys Davids, SBE 11 (1881): 1-136. Mahavagga, Kullavagga, and Pattimokkha: Vinaya Texts, tr. T.W. Rhys Davids and H. Oldenberg, SBE vol. 13, 17, 20 (1881, 1882, 1885). Back to text.

46. Mahavagga 1.28, SBE 13: 169-170. Back to text.

47. Note complex rules, e.g. Mahavagga 9.4.7-8, SBE 17: 217-272, establishing who has the right to vote (i.e., in such cases, to object). Back to text.

48. Kullavagga 4.9-14, SBE 20: 24-65. Back to text.

49. Kullavagga 4.10.1, SBE 20: 20-26, where it is stated that taking of votes is invalid "when the taker of votes [an elected official] knows that those whose opinions are not in accordance with the law will be in the majority," or "when he is in doubt whether the voting will result in a schism in the Samgha," or "when they do not vote in accordance with the view that they really hold." Kullavagga 4.14.26, SBE 20: 56-57 shows how the vote-taker was permitted to prevent the will of the majority from being enacted even in a secret vote, by throwing out the results if the winners' opinion went against the law -- or his interpretation of it. Back to text.

50. See Kullavagga 4.14.25-26, SBE 20: 54-57, where the emphasis is on reconciling monks to a decision which they were opposed to. Voting is one method of doing so; manipulation of votes preserves the religious law without splitting the sangha. Back to text.

51. It is commonly accepted by scholars that the regulations we have been discussing are, in the form we have them, the product of a long evolution, though all of them are attributed to the Buddha. See Rhys Davids' and Oldenberg's introduction to the Vinaya Texts, SBE 13: ix-xxxvii, and notes throughout. For the concern with disunity, see the extract from the Maha-parinibbana-suttanta (i.1) below; the Mahabharata, Santi Parva 107, and Kautilya, 11.1 (which despite their monarchist purpose, contain passages of republican thought -- see below, n. 71); Altekar, State and Government, pp. 129-130; A.K. Majumdar, Concise History, 2: 140. Back to text.

52. Maha-parinibbana-suttanta 1.1, SBE 9: 6-7; see below. Back to text.

53. Altekar, pp. 126-127, 132-134; Sharma, Republics, pp. 12, 110-111. Back to text.

54. Corporate Life, pp. 233-234; A.K. Majumdar, Concise History, 2: 137. Back to text.

55. The Maha-parinibbana-suttanta is the story of the "great decease of the Buddha" and as such includes both colorful anecdotes and important last-minute instructions to his followers. Back to text.

56. The Pali Canon uses both the term Vajji (Vriji in Sanskrit) and Licchavi to designate a republican polity based at Vesali. Scholars believe that the Licchavi were the people who lived at Vesali, while Vajji was the name of a confederation that they headed. For a detailed discussion, see Sharma, Republics, pp. 81-84, 93-97. Back to text.

57. Maha-parinibbana-suttanta 1.1, SBE 11: 6-7. Back to text.

58. In this sense R.C. Majumdar was right in calling the Buddha "an apostle of democracy;" Corporate Life, p. 219. Contra, Drekmeier, Kingship and Community in Early India, p. 113. Back to text.

59. Sen, Studies in the Buddhist Jatakas, pp. 60-64. Compare Burton Stein, Peasant State and Society in Medieval South India (Delhi, 1980) for a similar evaluation of South Indian monarchy in a later period. Back to text.

60. Altekar, State and Government, p. 136. Back to text.

61. Altekar, State and Government, pp. 137-138; A.K. Majumdar, Concise History, 2: 144. Back to text.

62. Agrawala, Panini, p. 428. What may be the clearest statement of egalitarian political ideology only comes to us through many intermediaries, as a tantalizing passage in Diodorus Siculus (2.39; Classical Accounts, p. 236) which seems to derive from Megasthenes: "Of several remarkable customs existing among the Indians, there is one prescribed by their [sc. Indian] ancient philosophers which one may regard as truly admirable: for the law ordains that no one among them shall, under any circumstances, be a slave, but that, enjoying freedom, they shall respect the principle of equality in all persons: for those, they thought, who have learned neither to domineer over nor to cringe to others will attain the life best adapted for all vicissitudes of lot: since it is silly to make laws on the basis of equality of all persons and yet to establish inequalities in social intercourse." Megasthenes (who was a contemporary of Kautilya) is often criticized for the good reason that slavery and other forms of inequality did indeed exist among the Indians. But perhaps he correctly presented the views of "their ancient philosophers." Back to text.

63. Kautilya, 11.1, Shamasastry tr. p. 410. The Mahabharata, Santi Parva, a royalist treatise on morality and politics, likewise mentions ganas (in c. 107; cf. c. 81) only to show how a raja who is not yet a true monarch in his state can implement his will -- and as we have seen, eliminating popular participation in government is an essential part of this. It is interesting to note that there are in both works passages that urge the raja to cooperate with the gana and, like the Maha-parinibbana-suttanta, emphasize the dangers to a gana of disunity. R.C. Majumdar (in Ancient India, 7th ed. (Delhi, 1974), p. 159) regarded Mahabharata, Santi Parva 107 as a piece of republican political science reworked for monarchist purposes. Back to text.

64. Altekar, State and Government, p. 124, draws attention to the existence of republican-style local government within the greater republic. Cf. the Italian situation described by Hyde, Society and Politics in Medieval Italy, p. 104: "Government under medieval conditions was always a precarious matter...the Italian cities faced special problems of their own, derived from the fact that the commune was originally no more than one kind of societas in a society that abounded in societates, so that it was an uphill task to assert any special claim to the loyalty and obedience of the citizens." Back to text.

65. Kautilya, 11.1, Shamasastry trans., p. 410. Back to text.

66. See R.C. Majumdar, Corporate Life, pp. 42-59 for the attitude of later Dharmasastra writers to the place of semi-autonomous corporations and kindreds in the monarchical polity of the fifth century A.D. and later. Back to text.

67. Pp. 366-367. Back to text.

68. N.B. the introduction: "To the memory of the Republican Vrishnis, Kathas, Vaisalas, and Sakyas who announced philosophies of freedom from devas, death, cruelty and caste." Back to text.

69. See above, n. 10. Back to text.

70. See esp. Ghoshal's treatment, A History of Indian Public Life, ii, pp. 185-197, which goes almost as far in one direction as Jayaswal went in the other. Cf. Drekmeier, Kingship and Community in Early India, p. 279; A.K. Majumdar, Concise History, ii, pp. 139-144; Burton Stein, "Politics, Peasants and the Deconstruction of Feudalism in Medieval India," Journal of Peasant Studies, xii, no. 2-3 (1985), p. 62 (discussing South India at a later period). Back to text.

71. A.K. Majumdar, Concise History, 2: 143. Back to text.

72. A similar tendency in recent decades to dismiss democratic elements in classical Athens and republican Rome is now being challenged: e.g. Ellen Meiksins Wood, Peasant-Citizens and Slave: The Foundation of Athenian Democracy, corrected paperback edn. (London, 1989) and much more cautiously by John North, "Politics and Aristocracy in the Roman Republic," Classical Philology, 85 (1990): 277-287 and reply to W.V. Harris's criticisms, pp. 297-298; John North, "Democratic Politics in Republican Rome," Past and Present 126 (1990): 3-21. Back to text.

73. Romila Thapar, A History of India, vol. 1 (Harmondsworth, 1966), p. 19; Bhandarkar, Lectures on the Ancient History of India, p. ix (written in 1918): "We have been so much accustomed to read and hear of Monarchy in India being always and invariably unfettered and despotic that the above conclusion [that republics were important in ancient India] is apt to appear incredible to many as it no doubt was to me for a long time." Back to text.

74. A.L. Basham, The Wonder That was India (London, 1954), pp. 96-98. Back to text.

75. In European history, the Anglo-Saxons have often been treated as a failed culture, and the Visigothic kingdom of Spain is seldom approached in any other way. See the opening remarks of Roger Collins, The Arab Conquest of Spain, 710-789 (Oxford, 1989). Back to text.

76. Thapar is one of the few to avoid this usage. Back to text.

77. Jayaswal, Hindu Polity, p. 46. Back to text.

78. Jayaswal, Hindu Polity, p. 116. Back to text.

79. For a general discussion of the concept of "tribalism," see Eric R. Wolf, Europe and the People Without History (Berkeley, 1982). Back to text.

80. Agrawala, Panini, pp. 479-493. Back to text.