Sumit Sarkar

Writing social history / Sumit
Sarkar; New Delhi: Oxford

University Press, 1998 (1-49 p.)
PART ONE
1
The Many Worlds of
Indian History
ntrospection about their own location in society has not been

too common among Indian historians. Our historiographical
essays, tend to become bibliographies, surveys of trends or ^movements within the academic guild. They turn around debates about
assumptions, methods, ideological positions. Through these, historians get pigeon-holed into slots: Neo-cotonial, Nationalist, Communal, Marxist, Subaltern. The existence of not one but many
levels of historical awareness attracts much less attention. But outside
the world of metropolitan centres of learning and research there
are provincial universities and colleges, schoolteachers, an immensely
varied student population, and, beyond these, vast numbers more
or Jess untouched by formal courses, yet with notions about history
and remembrances of things past, the nature and origins of which
it could be interesting to explore. What is neglected is the whole
question of the conditions of production and reception of academic
knowledge, its relationships with different kinds of common sense.1
We tack, in other words, a social history of historiography.
This problem of levels has become exceptionally acute in India
in recent years, with the growth of right-wing Hindu communal
forces, and the multiple responses to the Mandal proposals for
affirmative action in favour of 'backward' castes. In very different
i Which, as Gramsci reminded us, must be understood as a 'collective noun',
and as 'a product of history and a part of the historical process
"Common
sense" is the folklore of philosophy, and is always halfway between folklore
properly speaking and the philosophy, science and economics of the specialists/
Antonio Gramsci, Selections Jnm the Prison Notebooks, ed. Qoentin Hoare and
Geoffrey Nowell Smith (New York, 1971), pp. 325-6.
WRITING SOCIAL HISTORY
ways, both these sets of developments have in effect projected views

of Indian history at variance with what generally holds sway in
today's high-academic circles. More specifically, I have in mind the
debate around the Ramjanmabhumi issue, where well-established
academic knowledge has had to confront, not too effectively, one
kind of organized and largely manufactured common sense. Secular
historians refuted, with ample data and unimpeachable logic, the
justifications put forward by the Hindu Right, for its eventually
successful campaign to demolish a four-hundred-year-old masj id at
Ayodhya. They undeniably had the better of the intellectual and
human argument. Yet for a decisive year or two the views of the
leading historians of the country, most notably scholars at the Centre
for Historical Studies, Jawaharlal Nehru University, had less impact
than pamphlets of the order of Ramjanmabhumi ka Rakta-ranjita
Hihas (Bloody History of the Birthplace of Ram). This, however, was
very far from being a simple triumph of age-old popular faith over
the alienated rationalism of secular intellectuals. Viswa Hindu Parishad (VHP) pamphlets, and audio and video cassettes, systematically
combined an ultimate appeal to faith with a battery of their own
kind of historical facts: quotations from (real or spurious) documents, a certain amount of evidence fielded by archaeologists of
some stature, a parade of alleged facts and dates about precisely
seventy-six battles fought by Hindus to liberate the birthplace of
Ram from the evil 'descendants of Babar'. 'Faith' was deployed as
the final weapon usually only when such 'historical,' arguments were
seen to be in danger of total refutation.2
What this VHP quest for historical facticity revealed was that
history of one kind or another hascome to occupy a position of
exceptional importance in a variety of Indian discourses, but that
in the moulding of many such histories the best scholars often have
a very limited role. Historical consciousness, even when fairly organized, systematic, and far from spontaneous, evidently^ cannot be
equated with the thinking of professional historians alone, still less
with that of its highest echelons. Both the importance of history
and its multiple levels require further probing.
Some 'presentist' explanations, relating to the conditions of
2 For analysis of the 'historical' literature of the Ramjanmabhumi movement,
see Neeladri Bhattacharya, 'Myth, History and the Politics of Ramjanmabhumi't
in S. Gopal (ed.), Anatomy of a Confrontation (New Delhi, 1991), and Pradip
Kumar Datta^*VHFs Ram: The Hindutva Movement in Ayodhya', in G. Pandey
(ed.), Hindus and Others (New Delhi, 1993).
THE MANY WORLDS OF INDIAN HISTORY
production and dissemination of historical awareness in today's

India, are fairly obvious, and helpful up to a point The leading
members of the historians' guild write and teach mainly in English
for easy inter-regional and international communication. The majority of universities and colleges, however, have switched over to Hindi
or regional languages, translations are far from abundant, and the
historical common sense of the bulk of students and teachers is
determined much more by textbooks of very poor quality, or media
influences. After independence, history, and particularly narratives
of the 'freedom struggle' or the 'national movement', became a major
means of legitimizing ruling groups in the postcolonial nation-state
through claims of continuity with a glorious past. A very eclectic
range of 'national heroes' therefore had to be projected as knights
in shining armour, abstracted from real-life contradictions and
contextual pressures. Through the media and the majority of schools,
the message that has been constantly broadcast is that history is
valuable because it stimulates pride in one's country. The other
meaning of history, in these days of'objective' tests and proliferating
quiz culture, is of random facts and dates that have to be efficiently
memorized. Patriotism and quiz culture combine to ensure a very
low priority, in the bulk of history-teaching, to techniques of critical
evaluation of narratives about the past and the development of
questioning attitudes. History, in other words, tends to become
hagiography, and this opens the way towards giving hagiography the
present-day status and aura of history. Sometimes the links with
current chauvinistic developments are even more direct, most notably
through the enormously popular state television screening of the
Ramayana, just before the Ayodhya movement got into its stride.
The epic heroes were presented there as national figures, and Rain
returned to Ayodhya in triumph amidst wildly anachronistic twentieth-century slogans of 'Long live Mother India'.
But a merely presentist explanation will not take us very far. The
centrality of history today, as well as its markedly multi-level features,
are not universal or natural phenomena. They are evidently related
to the ways in which history came to be taught, written, and
exceptionally valorized under colonial and then postcolonial conditions. My essay, then, will have to go back to the nineteenth century,
when specifically modern ways of thinking about history are generally
supposed to have begun in India. My own area of competence, as
well as the early location of colonial power and cultural influence,
justify a primary focus on Bengal material.
WHITING SOCIAL HISTORY
There is a second reason, too, why a retrospect, at once historiographical and social-historical, is relevant for my argument,
and once again the polemics around Ramjanmabhumi provide a
point of entry. The importance given to apparently scientific history,
complete with facts, dates and evidence, as well tt the central
assumption of Hindus and Muslims as homogenized blocs existing
fundamentally unchanged across a thousand years, expose the
'tradition' deployed by the Sangh Parivar as overwhelmingly invented, moulded by colonial and postcolonial conditions and
influences. No great effort is required to recognize it to be as
'modern' as its secular-rationalistic Other. Our glance back at
colonial Indian historiography will incidentally confirm that the
'history* used by the Ramjanmabhumi movement was not any
spontaneous welling-up of folk or papular memories, but made up
of bits and pieces from the academic wisdom of an earlier generation
of nationalist historians, as orchestrated by a very modern political
machine. A generalized critique of post-Enlightenment modernity
and Orientalizing colonial discourse, therefore, might seem to offer
an effective ground for the rebuttal of Hindutva's claims to indigenous authenticity. More generally, the moods stimulated by
Edward Said's Orientalism which have been transplanted to South
Asia by Partha Chatterjee and the later volumes of Subaltern Studies
have provided tor many intellectuals an overall framework that
combines the virtues of apparent radicalism with a satisfactory
distance from the Marxism of yesteryear, now widely assumed to
be finally and deservedly dead.
Many of the essays in this volume express my sense of disquiet
with this current turn in South Asian scholarship. Very briefly, at
this point: what had started as an understandable dissatisfaction
with the economistic reductionisms of much 'official' Marxism is
now contributing to another kind of narrowing of horizons, one
that conflates colonial exploitation with Western cultural domination. Colonial discourse analysis abstracts itself, except in the most
general terms, from histories of production and social relationships.
A 'culturalism' now further attenuated into readings of isolable
texts has become, after the presumed demise of Marxism, extremely
nervous of all 'material* histories: the spectre of economic reductionism looms everywhere. Colonial-Western cultural hegemony,
secondly, tends to get homogenized, abstracted from internal tensions, and presented as aU-pervasive, virtually irresistible within its
own domain those touched by it become capable of only
'derivative discourses'. A total rupture then has to be presumed

between pre-colonial and colonial, and a temptation sometimes
develops to make the former a world of attractive ^traditions, of
innocence confronting Western power-knowledge.3 A more fundamental methodological problem is the abandonment in practice of
any quest for immanent critique through the elision of possibilities
of mutually conflicting groups taking over and using in diverse,
partially- autonomous ways, elements from dominant structures and
discourses. What is ignored, in other words, is precisely that which
had been central to Marxist analysis: the dialectical search for contradictions within structures. If modern power is total and irresistible
within its own domain, autonomy or resistance can be located only
in grounds outside its reach: in a 'community-consciousness* that
is pre-colonial or somehow untainted by post-Enlightenment powerknowledge, or in fleeting, random moments of fragmentary resistance. These become the only valid counterpoints against the
ultimate repository of that power-JcnowJedge the colonial 01
postcolonial 'nation-state'. We have moved, then, from perspectives
in which relationships between capitalist imperialism and multiple
strands within anti-colonial movements had constituted the basic
framework, to one where the post-Enlightenment modern state is
counterposed to community. Questions of exploitation and power
have been collapsed into a unitary vision of the modern bureaucratic
state as the sole source of oppression.
These caveats, summarized in a telegraphic manner, represent
some of the problems I have been encountering in my own
thinking and research, and I intend to elaborate them in specific
historical contexts, as, hopefully, invitation to dialogue rather
than confrontational polemic. An exploration of colonial and
postcolonial historical consciousness, vital for understanding today's many worlds of history, can be useful also as the fust of
these contexts. A framework grounded in the assumption of
pervasive colonial cultural domination has naturally paid considerable attention to the development of 'modern1 attitudes towards
history, and in fact this provides at first sight exceptionally
3 The 'derivative discourse' argument was elaborated by Partha Chatterjee in
Nationalist Thought and the Colonial World: A Derivative Discourse? (Delhi, 1986).
In a seminal text published three years before, Ashis Nandy had sought to
'justify and defend the innocence which confronted modern Western colonialism and its various psychological offshoots in India.' The Intimate Enemy
(Delhi, 1983), p. ix.

4
suitable material for a colonial discourse approach, I hope to

show that the assumptions I have just catalogued tend to be
restrictive, even when deployed on such a favourable site. In
part, then, I will have to go over already familiar material, but
hopefully making new points and sometimes arguing the case
for fundamental departures. Three temporal cross-sections appear
particularly relevant: the apparently total early-nineteenth-century
rupture with which 'modern' Indian historiography began; latenineteenth and early twentieth-century crystallizations of 'nationalist' and 'communalist' historical assumptions and methods; and
today's predicaments.
II
Pre-colonial India, with its very long traditions of written culture,
produced numerous texts of recognizable historical intent or value:
Puranic king-lists, dynastic chronicles, histories of castes and religious
sects, biographies of holy men, genealogies of prominent families.
As elsewhere, there were evident links between the quantum of such
texts or documents and levels of organized, bureaucratic power. Thus,
ancient Indian historiography, not surprisingly, never attained the
stature of that of China with its unique bureaucratic continuity, and
historical accounts became much more numerous under the Delhi
sultanate and the Mughal empire. (Islam, with its single Hizrat era,
also brought in a new chronological certitude.)
Yet it remains undeniable that the impact and imposition of
Western historiographical models through English education and
British Indian scholarship created a widespread sense of a tabula rasa.
Pre-colonial texts, since then, have always figured as "sources' to be
evaluated by modern Western canons, not as methodological influences. In 1958 a competent survey of history-writing in nineteenth and
early twentieth-century Bengal could assume that 'we had to start
from scratch'.5
A convenient and much-used initial benchmark for examining
this rupture is provided by Mrityunjoy Vidyalankar's Rajabali
4
Ranajit Guha, An Indian Historiography of India: A Nineteenth-Century Agenda

and Its Implications (Calcutta, 1988); Partha Chatterji, The Nation and Its Fragments:
Colonial and Post-Colonial Histories (Princeton and Delhi, 1994), Chapters iv, V.
5
B.P. Mukherji, History, in a collection entitled, significantly, for my argument, Studies in the Bengal Renaissance, ed. A. Gupta (Calcutta, 1958).
(Chronicle of Kings, Serampur, 1808). Its author was an orthodox

Brahman commissioned by the British authorities to write th
first overall historical survey in Bengali prose, to serve as a
language text for Company officials being trained at Fort William
College. The text began by expounding, in a manner totally
unselfconscious and free of defensive apologetics, the standard
Brahmanical concept of time as cyclical, with Satya, Treta, Dwapar
and Kaliyuga endlessly succeeding each other. The moral trajectory
across the four-yuga cycle was always imagined as inevitably
retrogressive, and the present (invariably, in these texts, the Kaliyuga) was the worst of times, characterized by overmighty Shudras
and insubordinate women. Time, in other words, was never abstract,
empty duration: it was relevant primarily for moral qualities
assumed to be inseparable from its cyclical phases. The principal
role of the yuga-cycle in Brahmanical discourses, from the Maba\ bharata down to Mrityunjoy, was to suggest through dystopia
the indispensability of right caste and gender hierarchy. The two
have been necessarily imagined as interdependent, for purity of
caste lineage is vitally related to male control over the reproductive
capacities of women, ensured through marriage, within the permitted boundaries.
For the rest, Rajabali was a compendium of king-lists, many of
them soon to be discarded as mythical by modern historians. The
striking feature, for anyone trained in Indian history in the ways that
became standard from around the 1820s, is really a notable absence.
Mrityunjoy displayed no awareness at all of any breaks between
'Hindu', 'Muslim', and 'British' periods, but remained content with
awarding good or bad marks to kings with a fine indifference to religious identities. The recurrent criteria, incidentally, for immoral behaviour are strayinata (subordination to women) and nimakbarami (being
'untrue to one's salt1, i.e. violating obligations of loyalty and obedience,
and thus implicitly weakening proper hierarchical relationships). The
link with standard Kaliyuga notions of disorder is fairly obvious.
As in the bulk of pre-colonial history-writing, the predominant
note in Rajabali was didactic, with exploration of the uniqueness
of historical situations less important than teaching obedience and
morality through archetypes. A ninth-century Jaina text had described 'Itihasa' (history) as 'a very desirable subject... it prescribes
dharma [right conduct]', and even Kalhana, much praised by modern
b
It has been used in that way recently by both Ranajit Guha, and Partha
Chatterjec, op. cit.
historians for his unusually critical treatment of sources and striving

for objectivity, claimed that his Rajatarangini would be 'useful for
kings as a stimulant or a sedative, like a physic, according to time
and place.*7
The contrasts with histories that English-educated Indians started
writing after around the mid nineteenth century are obvious enough.
British rule brought with it clocks and a notion of time as linear,
abstract, measurable in entirely non-qualitative units, an independent
framework within which events happen.8 The other major change
was the imposition of the ancient/medieval/modern schema which
had become standard in the post-Renaissance West. James Mill
transplanted this into India by dividing the subcontinent's history
into Hindu, Muslim and British periods. By the time of Nilmoni
Basak's Bbaratbanbcr Itihas (History of India, Calcutta, 1857), which
may serve as our second benchmark, the yuga cycle is mentioned
only in a brief, defensive preface, after which the author quickly
passes on to a periodization that distinguished the 'Age of Hindu
Empires' from 'Muslim Kingdoms'.
Yet I think it is important to resist bland, homogenized
presentations, both of pre-colonial notions of time and history
as well as of the colonial rupture. It is now generally recognized
that the cyclical/linear binary is not absolute, for duration, or
sequentiality, is common to both.9 High-Hindu cyclical time, for
7 R.C Majumdar, Ideas of History in Sanskrit Literatim, in CH. Phillips (ed.),
Historians of India, Pakistan, and Ceylon (London, 1961), p. 21.
* I have found very helpful Moishe Postone's recent suggestion that a concrete/abstract distinction is more relevant than the conventional cyclical/linear
binary. Elaborating a suggestion of E.P. Thompson (Time, Work-Discipline
and Industrial Capitalism', Past and Present, 38, 1967), Postone argues that the
concrete time of a pre-capitalist societies 'was not an autonomous category,
independent of events, it could be determined qualitatively, as good or bad.
sacred or profane.' It 'is characterized less by its direction than the fact that it
is a dependent variable.' Moishe Postone, Time, Labour and Social Domination:
A Reinterpretation of Marx's Critical Theory (Cambridge, 1993), pp. 200-16, passim.
9
Combinations of the two have been noticed in places and times as far
apart as present-day Bali and pre-colonial Yucutan. Balinese notions of duration
exhibit features of both cyclicity and linearity. The Cbilam Balam texts of
Yucutan described endless cycles marked out by specific moral qualities, but
the Mayans also had chronicles of ruling dynasties which were entirely linear.
LEA. Howe, 'The Social Determination of Knowledge: Maurice Bloch and
Balinese Time', Man, New Series, 19, 1981; Nancy M. Farms, 'Remembering
the Future, Anticipating the Past History, Time and Cosmology among the
Maya of Yucutan', Comparative Studies in Society and HistoryT 29, iii, July 1987.
instance, encompassed an element of linearity, for within a mabayuga the successive downward movement of Satya through Treta
and Dwapar to Kali was taken to be irreversible. The polarity
between concrete and abstract time is perhaps more fundamental,
but here again further distinctions within both might become
necessary. Thus, generalizations of the order attempted by Mircea
Eliade about a 'myth of the eternal return', supposedly characteristic
of an undifferentiated 'Hindu' world or even of 'traditional
civilization',10 really rest upon an unproven assumption of homogeneity. That we know little - virtually nothing, in feet of
notions about time among pre-colonial peasants or lower-caste
people is surely no ground for assuming that they must have
invariably internalized the highly Brahmanical and hierarchized
values which are inseparable from formulations of the four-yuga
cycle that have come down to us. One needs to be open, rather,
to the possibility of work or task-oriented times, which could
vary greatly in precision according to specific requirements. The
purohit and astrologer needed a precise fix on certain 'time points'
to determine auspicious ritual moments or make predictions.11
Rural labour processes, in contrast, demanded little more than
a grasp over general seasonal and- daily rhythms.
Even at more philosophical or speculative levels, some scholars
feel that Brahmanical texts indicate the presence of not one but
several layers in pre-Islamic Indian notions of time Raymondo
Panikkar, for instance, refers to a tradition called Kalarada where
time is placed above all gods and identified with images of death
as supreme leveller of all distinctions, human and even divine. He
considers this to have been 'a widely-held popular view, belonging
probably to the less Brahmanic stratum of Indian tradition'.12
1 Mircea Eliade, The Myth of the Eternal Return (Paris, 1949, trans., London,
1955), Chapters i, n, iv.
11
The concrete time of medieval European monasteries, Postone points out,
had its own, specific notion of time-discipline, through 'a series of time points,
which marked when various activities were to be done.' This, however, is quite
distinct from capitalist forms of time-discipline, where 'commensurable, interchangeable, and invariable' time-units become 'the measure of activity.' Once
again, it is the concrete/abstract distinction which is crucial. Postone, pp. 203,
209.
12 Raymondo Panikkar, Time and History in the Tradition of India: Kala
and Karma', in L Gardet, et aL (ed.), Cultures and rime (Paris, 1976). The
Mababbamta, interestingly, associates this view with Bali, a demon (Asura) chief
defeated by Indra, the king of the gods: 'In Time's course many thousands of
10
Orthodox Brahmanical denunciations of Kalavada often associated

it with materialism, nastika views, and an epistemological position
called pratyahkyavada, according to which sense impressions constituted the sole criterion of valid knowledge or proof.13 Such views
were lumped together in dominant philosophical discourse as the
tradition of Lokayata, said to have been founded by Carvaka. (One,
evidently pejorative, meaning of Lokayata, incidentally, is 'that
which was prevalent among the common people'.) Quite remarkably,
Mrityunjoy's Rajabali at one point interrupts its placid chronicle
of kings with a two-page diatribe against nastika views which once
again associated the extreme empiricism of pratyakshyavada with
Kalavada: 'like trees in a mighty forest, the world appears and
disappears by itself, subject only to time.'14
Kaliyuga, we are always told, will end in an apocalyptic manner,
with universal destruction (yuga-pralay), or, alternatively, the coming
of Kalki-avatar (the last incarnation of the high-god Vishnu), after
which another identical cycle will commence. In yet another interesting shift or variation, however, the apocalypse, and in some ways
the entire framework of four-yuga cycles, seems to have become
somewhat downgraded over time. Yuga-pralay was pushed for out
into the future, and so in practical terms the key message became
one of enduring the inevitable evils of 432,000 years of Kaliyuga
through tightened-up rules of caste and patriarchal discipline. The
endless cycles themselves came to be considered part of the world
of maya (illusion, or, more exactly, inferior order of reality) in the
increasingly dominant philosophy of Vedanta.15
Indras and deities have been swept off yuga after ruga. . . . Time has no
master . . . wealth, comforts, rank, prosperity, all fall a prey to time . . . All
things that proudly raise their heads high are destined to fall down.' Mababharata, English translation by Pratap C. Roy, ed. Hiralal Haldar (12 vols,
Calcutta, n.d.), vol. a, pp. 140-56.
13 Nastika, often taken today as the equivalent of an atheistic position, meant
more precisely in Indian philosophical traditions the denial that revealed texts
(sntti, i.e. the Vedas) have the status of valid proof (pramana) in philosophical
arguments. The extreme empiricism of pratakshyavada entailed such a denial,
and could lead also to the rejection of belief in gods or the immortality of
souls. Debiprasada Chattopadhyay, Lokayata: A Study in Ancient Indian Materialism (New Delhi, 1959, 1978), Chapter I. Surendranath Dasgupta, A History of
Indian Philosophy, vol. Ill (London, 1961), pp. 512-50.
M Mrityunjoy Vidyalankar, Rajabali (Serampur, 1808; Calcutta, 1905), pp. 1113. Mrityunjoy attributed the origin of these views to an Asura king.
!5 For a more detailed account of Kaliyuga and its variations, see Chapter 8
below.
11
The relevant points, in considering the importance of these varied

notions of time for perceptions of history, are that none of the
alternative frameworks hindered the construction of narratives about
the past connected to specific purposes, such as the glorification of
dynasties, families or religious traditions; but neither did they require,
or stimulate, a sense of overall social process or interest in its possible
causes.16 The central high-Hindu ideal was the individual breaking
out of the bondage of karma (endless rebirth, in which merits and
sins accumulated in previous lives rigidly determined one's status in
life, and therefore one's dharma, in the sense of appropriate rituals
and duties). Unlike Christianity or Islam, with their notions of a
day of judgement common to all, the idea of salvation here was not
community-based, and so the conception of universal causality
implicit, in a way, in the doctrine of karma, applied only to
individuals. It did not lead to any interest in the causes of aggregate
phenomena. The yuga framework did involve a sense of a moral
texture common to an era, but then its lineaments and causes were
already known: being fore-ordained, divinely determined, or related
to the quality of kings - i.e. to the Ma (game, or play), as it was
said, of time, gods or kings. The purpose of history remained,
therefore, a combination of royal propaganda and the teaching of.
dharma with examples, and its place in education, we shall see, seems
to have been negligible or non-existent.
Historical texts became much more abundant under the Delhi
sultanate, the Mughals, and their successor states, but it is doubtful
whether there was a fundamental break with regard to the aims and
presuppositions of history-writing. The Persian narratives are often
impressive in the careful attention they pay to specific usually
military-administrative events, and even to their secular, 'secondary'
causes. Yet the overall aim remained, as earlier, a combination of
exalting the rulers with religious and moral teaching via examples.
'Interest concentrated on how far a man conformed to an ideal
prototype, not how far he diverged.'17 The dominant Islamic conception of time as 'piecemeal vision of . . . a sequence of instants
. . . which are the signs and spaces of God's intervention' also did
16
Pre-colonial histories, Ranajit Guha has argued, consequently tended to

be 'made up of discrete moments, recovered synchronically as the occasion
required.' Ranajit Guha, An Indian Historiography of India: A Nineteenth Century
Agenda and Its Implications (Calcutta, 1988).
l ? P. Hardy, Historians of Medieval India (London, 1960, 1966), pp. 113, 118.
12
13
WRTTING SOCIAL HISTORY
not encourage, on the whole, any total view of history as diachronic

social process."
We may now be in a better position to evaluate the precise extent
and lineaments of the 'break' brought about in the ways of constructing die past by colonial rule and English education. In the making
of British Indian historiography, and 'colonial knowledge' in general,
an important dimension of genuine curiosity and excitement, as a
vast and varied subcontinent was opened up to the Western gaze,
went along with fairly obvious links with the logic of colonial power.
The British, as utterly alien rulers, needed to know something about
the traditions and 'prejudices' of their subjects; extraction of revenue,
dispensation of justice, and maintenance of order all demanded knowledge of past administrative practices; meticulous enquiry into possible
causes became standard practice after every rebellion. The powerknowledge theme is very self-consciously present in much official
writing: Risley's Tribes and Castes of Bengal (1891), for instance, claimed
that an ethnographic recording of the customs of people was 'as
necessary an incident of good administration as a cadastral survey of
the land and a record of rights of its tenants.'19
provoked a search for sustenance in past glories, real or imagined,

as well as efforts to use history to probe the causes of present-day
misfortunes. History thus came to acquire a new centrality in the
concerns of the Indian intelligentsia, for it became a principal 'way
of talking about the collective self, and bringing it into existence.'20
This much is well known, and hardly in need of restatement. Very
much less explored, but perhaps more significant, is the vastly
extended reach of history. While the transition from manuscript to
print obviously enlarged the potential readership of history books,
the subject itself came to acquire a totally new position in structures
of formal education. These structures themselves simultaneously became more crucial, for the links between formal education and
respectable jobs, professions, and careers were now tighter than ever
before. It was no longer possible, for instance, under colonial 'law
and order' for military adventurers to carve out kingdoms for themselves, while recruitment to administrative posts became dependent
on examinations. It had been possible for an Akbar to remain virtually
illiterate: not so for the meanest colonial Indian official or clerk. The
restriction of alternative opportunities (of independent business enterprise, for instance, particularly in Bengal, with its overwhelming
colonial economic presence) further enhanced the centrality of formal
'liberal' education. This became indispensable for a self-consciously
'middle-class' existence, one which combined the material and cultural
resources for entering high schools and colleges with a need for
income from jobs or professions. Such need was less acute for the
really big zamindars or businessmen, and impossible to satisfy for
the vast majority of peasants.
Notions of time, now assumed to be linear and abstract (i.e. no

longer primarily perceived in terms of moral quality, as Kaliyuga had
beenX methods of collecting data and assessing its reliability, and,
above all, levels of efficiency in the processes of accumulating knowledge had altered dramatically through the incursion of the postEnlightenment West with its novel and expanding resources of
bureaucratic power, less so, possibly, the basic motivation* of more
effective governance and legitimation of authority, which may not
have seemed very novel to, say, Abul Fazl. Power-knowledge far
antedates the modern West, as Umberto Eco so delightfully reminds
us in his the Nome of the Rose, -set in a medieval monastic library
where control over dangerous knowledge is defended through murder.
What was new was the unprecedented importance and reach that
history quickly acquired in colonial times. History became the
principal instrument for inculcating the stereotypical dichotomy
between the backward, immobile Orient as contrasted with the
dynamic, Christian and/or scientific West, thus simultaneously buttressing British self-confidence and reminding Indians of their lowly
place in the world's scheme of things. Foreign rule, conversely, soon
is L Gardet, p. 201.
H.H. Rislcy, Tribes and Castes of Bengal Ethnographic Glossary, vol. I (Calcutta,
1891, 1981), p. vii.
The place of history in the new educational system thus needs

some explication. Derozio, we know, taught History to Hindu
College students, and is said to have inspired 'Young Bengal' through
'examples from ancient history of the love of justice, patriotism,
philosophy and self-abnegation.'21 But surprisingly little is known
about the mundane details of college courses and texts, and there is
a need to explore the patterns of change and continuity involved in
the displacement of traditional patbsbaias, tols and madrasas by the
20 Sudipta Kaviraj, The Unhappy Consciousness: Bankimcbandra Chattopadbyay
and the formation of Nationalist Discourse in India (Delhi, 1995), p. 108.
21 ?yv'\&zn&M\aA, Biographical Sketch of David Hare (Calcutta, 1877), p. 27,
quoted in my 'Complexities of Young Bengal', Nineteenth Century Studies,
October, 1973; reprinted in Sumit Szikai.A Critique of Colonial India (Calcutta,
1985), pp. 20-1.
14
15
modern classroom ordering of space, time, and methods of teaching.22 The new interest in Western education stimulated by colonial
discourse analysis remains content, far too often, with yet another
critique of Macau lay's Minute.
For elementary education, however, Kazi Shahidullah's Patbshahs
into Schools does provide some illuminating details, based on a
diligent study of the surveys of Francis Buchanan and William Adam.
The traditional pathshala had concentrated on providing an eminently practical training in language, arithmetic and accountancy, while
the few written texts (manuscripts) were of a religious, moral, or
grammatical kind: history of any sort seems to have been conspicuously absent. The ear/y Bengali printed textbooks provided free
to pathshalas by the School Book Society from 1817 onwards, in
contrast, 'covered a variety of subjects like History, Geography and
Astronomy*. The publications of the Serampur missionaries meant
for schools, similarly, included Dig Danban, 'a miscellaneous collection of Truths and facts covering history, science and ethics', ind
so often made that the culture imposed through Western education

was always rationalist and invariably dominated by 'post-Enlightenment' values seems in need of some questioning. An earlier historiography had hailed such rationalism as harbinger of a Bengal (sometimes Indian) 'renaissance' or 'awakening'. The critique of colonial
discourse which today has largely displaced that old consensus inverts
the value judgement but otherwise maintains a basic continuity
through its assertion of a total rupture.24 There was nothing particularly rational (or secular), surely, about the oft-repeated formula
of British rule in India being an act of divine providence. Again,
the missionaries used modern Western science to undermine the
'superstitions' of the Hindus, but their overall aim was conversion
to another, not noticeably more rational, religion. Here I must add
that I find Dipesh Chakrabarty's assertion in a recent article that
'missionaries did not perceive much contradiction between 'rationalism and the precepts of Christianity' - difficult to understand.
Chakrabarty refers in particular to Alexander Duff, and cites M A
Laird (1972) as the source for his reading of the Scottish missionary.25
Duffs own India and India Missions (Edinburgh, 1839), in striking
contrast, recounts how much effort he had had to make to persuade
the pupils of Derozio to accept the Reformation as their model in
place of 'the terrible issue of French illumination and reform in the
last century', and how happy he was when Krishnamohan Banerji's
conversion to Christianity indicated that 'avowed atheism' was on
the decline.26
Historical Anecdota took its place next to Aesop's Fabks as reading
lessons 'illustrative of justice, fidelity, probity and humanity*. To

take a final example, the hundred-odd vernacular village schools set
up in some Bengal districts by Hardinge's orders in 1844 were to
have a curriculum of 'vernacular reading, writing, arithmetic, geography, and history of India and Bengal.'23 The missionary juxtaposition of Aesop with historical anecdotes is an important indication
that history of one sort or another was now being given an exalted
place in general education, even outside its position as a distinct
subject. The pattern continues, and quite emphatically so: school
textbooks for teaching Hindi today, for instance, tend to contain
an enormous amount of crude 'historical1 tales for inculcating a
patriotism often difficult to disentangle from Hindu-communal
assumptions and values.
History, then, acquired a new and vast pedagogical and intellectual
domain in the nineteenth century. Content-wise, however, it would
be misleading to assume a simple, unambiguous or complete rupture,
a leap from 'myth* to positivistic objectivity'. In fact the assumption
22 Work like Barbara Metcalf s study of the transition from madras* t o the
Deoband seminary {Islamic Revival in British India: Deoband, 1860-1900, Princeton, 1982) needs to be followed up. See now, however, the useful collection
of essays, Nigel Crook (ed.), The Transmission of Knowledge in South Asia (Delhi,
O U P , 1996).
23 Kazi Shahidullah, Pathshalas into Schools (Calcutta, 1987), pp. 15, 25, 29, 33.
The myth of Kahyuga vanished quickly from formal Indian

historical writings or textbooks, but it continued to enjoy a vigorous
if interestingly modulated life in other texts and contexts right down
to the early twentieth century- Kaliyuga, I have argued elsewhere,
became a whole language for expressing resentments about the new
discipline of time being imposed under colonial rule in clerical
office-work (chakri). It thus provides an important entry point into
24
A similar continuity is noticeable in^he focus, i n 'renaissance' and 'Saidian'

writing alike, o n the high literati alone. For elaborations of this argument, as
well as some efforts at developing an alternative approach, see Chapters 5 and
6.
25 Dipesh Chakrabarty, 'Radical Histories and the Question of Enlightenment Rationalism: Some Recent Critiques of Subaltern Studies', Economic and
Political Weekly, xxx, 14, 8 April 1995, p . 752.
, 26 Alexander Duff, India and India Missions (Edinburgh, 1839), pp. 629, 667,
cited in m y 'Complexities of Young Bengal', Critique of Colonial India, pp. 20,
26.
16
a level of colonial middle-class life largely ignored so far by historians.27 A recent thesis has emphasized the intertwining of what,
by strict Enlightenment-rationalist standards, should have been dismissed as myth, with positivistic facticity in much nineteenth-century
Bengal history writing. Nilmoni Basak may have been embarrassed
by notions of cyclical time, but his Nabanari (4th cd., Calcutta, 1865)
could still lump together two historical with seven mythical or
legendary figures in biographies of exemplary women.28 And Sudipta
Kaviraj has drawn our attention to the combination of 'real1 and
'imaginary* histories in the writings of major late-nineteenth-century
figures such as Bankimchandra and Romeshchandra Dutt. Dutt,
India's pioneer economic historian, was also the author of historical
romances about Rajputs and Marathas, while Bankimchandra made
several attempts to reconstruct bits and pieces of Bengal's past on
the basis of carefully sifted evidence. Bankim, Kaviraj suggests, came
to feel that 'the rational discourse of fact-gathering' could provide
inadequate grounds for the kind of itihasa that 'Bengalis' needed if
they were to become 'men*, or in order to constitute themselves into
a collective self, hence there was a shift in Bankings later writings
to the 'mythic discourse' of the historical novel.29
Indian history provides at first sight a particularly telling instance

of a 'derivative discourse' which lives on even today in many Indian
textbooks and syllabi, inadequately concealed by a nomenclatural
change ('medieval' in place of 'Muslim', but still beginning from
the establishment of the Delhi sultanate). That the implications of
this have often been communal is equally obvious. Even at its most
innocuous, the translation of ancient and medieval into Hindu
and Muslim assumed the existence of homogenized entities, supposedly unified by religion, as the basic building blocks of all precolonial history. In addition the stigma, commonly attached to the
middle term in the ancient/medieval/modern schema evolved in
post-Renaissance Europe, deepened when the British transplanted
it into India. Islam had been the great enemy of Christendom, the
British had displaced Mughal emperors, the 1857 revolt and the
Wahabi movement seemed to indicate Muslims to be on the whole
more dangerous than the Hindus, and meanwhile Orientalist scholarship claimed to have discovered a glorious 'classical' age of early
Hinduism (which was embodied in a language generically related
to Greek and Latin). Many of the central propositions of mainstream
nationalism and Hindu communalism (and, with values inverted,
of its Muslim alter ego) can thus be shown to have originated in
colonial discursive patterns.
This collective self, however, was for Bankim almost invariably

Hindu, and pitted usually against Muslims, in language that sometimes turns downright abusive. This is a feature that Kaviraj seems
disinclined to probe, but it does seem to suggest a high degree of
the internalization of the tripartite schema in its most anti-Muslim
form.30 The far-reaching impact of James Mill's periodization of
27 See Chapter 6, 'Renaissance and Kaliyuga: Time, Myth and History in
Colonial Bengal', and Chapter 8, 'Kaliyuga, Chakri and Bhakti: Ramakrishna
and His Times'.
28
Indira Chowdhuiy Sengupta, 'Colonialism and Cultural Identity: T h e

Making o f a H i n d u Discourse', Chapter u (unpublished thesis, School o f
Oriental a n d African Studies, London University, 1993). I am very grateful t o
Ms Sengupta for allowing m e t o read her thesis.
29 Kaviraj, p p . 124, 131.
30
Unlike many o f his contemporaries, however, Bankim did n o t accept an
idealized, golden-age view o f the ancient 'Hindu' period, and went so far as t o
suggest that the conditions of the Shudras might have been worse in independent
Aryan India than under British rule. Paradoxically, this recognition of discordances within the H i n d u fold seems t o have stimulated his passionate search
in t h e 1880s for an imaginary history of Hindu war against a n externalized
Muslim Other. See the critical, yet nuanced, discussion of Bankimchandra in
Tanika Sarkar, 'Bankimchandra a n d t h e Impossibility of a Political Agenda',
(htford Literary Review, XVI, 1-2, 1994.
17
But how much, really, do such origins explain? Marc Bloch warned
historians many years ago of the 'idol of origins', the tendency to
assume that 'a beginning . . . is a complete explanation.'31 If histories
written within the Saidian mould homogenize, they also often tend
to impose closures by suggesting ready answers to issues that could
have developed into interesting inquiries. Even in cases where the
derivation is undoubted, we need to ask further questions as to what
is (and sometimes is not) being accepted or internalized, by precisely
which groups, and why. Western critiques of the conditions of Hindu
women acquired an early resonance in Bengali middle-class circles:
much less so the equally trenchant attacks on caste and high-caste
oppression. The tripartite schema and the related myth of centuries
of Muslim tyranny were very quickly taken over, but not the fairly
common missionary or utilitarian denunciation of all Indian culture
which was embodied, in its most notorious form, in Macaulay.
An incidental reference in Rajat Kanta Ra~y's recent account (in
Bengali) of politics and society around 1757 can provide an example
31 Marc Bloch, The Historian's Craft (1954; Manchester, 1963), pp. 29-35.
18
of the kind of inquiry that has remained foreclosed,32 both in the

earlier 'Bengal Renaissance* historiography and in its current Saidinspired inversions.33 Rajiblochan Mukhopadhyay's Maharaja Krisbnacbandra Raysva Charitra (1805), the second biographical work that
we have in Bengali prose, contains a very early example of the
Muslim tyranny myth that would become near-ubiquitous in so
much nineteenth<entury Bengali Hindu writing. Rajiblochan's hero
Krishnachandra, the powerful zamindar of Krishnanagar (Nadia
district), is shown to be taking a leading part in a conspiracy of
Hindu zamindars, court officials, financiers and Mir Jafor to seek
the aid of the English in overthrowing Nawab Sirajuddoula. A
clearly formulated desire to end 'Yavana' misrule is attributed to
the Hindu plotters (who, curiously, are described as openly displaying their anti-Muslim motivations in front of Mir Jafar). As Rajat
Kanta Ray points out, more contemporary texts, whether in Persian
or English, or the odd Bengali village poems referring to Plassey,
are totally silent about any such self-consciously Hindu conspiracy
to overthrow Muslim tyranny (as distinct from the misrule of a
particular nawab).34 Rajiblochan's text, interestingly, precedes the
organized spread of English education, and for that matter Mill's
tripartite schema of Indian history. It combines adulation for the
Krishnanagar Raj family with flattery of the English as liberators
from Muslim misrule, and seems to demand location in a milieu
of high-caste Hindu literati transiting from zamindari to Company
patronage. Rajiblochan had a family connection with the Krishnanagar Raj, and had then been recruited, along with other Nadia
pandits, by William Carey to work for the Bengali printing press
set up by Baptist missionaries at Serampur from where he followed
his master to Fort William College, established by Wellesley in
1800.35 It may not be irrelevant to note that Maharaja Krishnachandra is remembered as a very major patron of orthodox Brahmanical culture, and that his power base was in a region (Nadia)
notorious in the eighteenth century and beyond for its multitude
of heterodox lower-caste sects: Kartabhaja, Sahebdhani, Balarami,
and many others.36 It is tempting to suggest a link, therefore,
32 Rajat Kanta Ray, Palasir Sbarayantra o Sekaicr Samaj (Calcutta, 1994).
33 See below, Chapter 6, for a discussion of the 'renaissance' debate.
Ibid., p p . 183-7 a n d passim,
35 Brojendranath Banerji, 'Fort William Colleger Pandit', p p . 28-9, in SahityaSadbak-Caritmala, i (Calcutta, 1942).
36 Sudhir Chakrabarti, Sahebdhani Sampraday Tader Can (Calcutta, 1985).
19
between the easy acceptance of some key aspects of Anglo-Indian

historiography, projecting the British as saviours of Hindus from
earlier Muslim misrule, and possible late-eighteenth and nineteenthcentury efforts at reasserting high-caste power. These came to be
closely associated with the eventually more secure zamindari and
tenure-holdings of post-Cornwallis Bengal, and the new importance
of the 'liberal' professions entry to which became restricted to
the products of bhadralok-dominated higher education. For such
a strata privileged, benefiting in many ways from foreign rule,
yet increasingly aware of a humiliating colonial dependency - the
thought that British rule was a great improvement on 'medieval
Muslim tyranny' could provide considerable solace as well as a safe
and distant site for locating a largely imaginary history of Hindu
prowess against not British, but Muslim invaders.
That colonial history, developed primarily to sustain and ratify
British rule, quickly became the ground for contradictory and limited
yet powerful assertions of patriotism is a well known and quite
undeniable feature of nineteenth-century Bengali intellectual history.
Yet some minutiae of dating and language indicate that it still puts
the framework of derivative discourse under some strain. Partha
Chatterjee feels impelled to add the phrase 'curiously enough' to the
fact that 'the new Indian literati, while it enthusiastically embraced
the modern rational principles of European historiography, did not
accept the history of India as it was written by British historians*.37
Ranajit Guha pushes the moment of autonomous assertion forward
to Bankimchandra's historical essays of the late 1870s and early
1880$. Chatterjee, through a survey of school textbooks, brings it
back to the 1860s. Neither mention the plenitude of very similar
material in the proceedings of the Derozian Society for Acquisition
of General Knowledge (1838-43), i.e. emerging precisely from a group
often accused of being, quintessential^, denationalized Anglicists,
who in Guha's framework should have displayed 'unquestioning
servility to the ruling power*.38 Pyarichand Mitra's State ofHindoostan
under the Hindus, for instance, combined warm references to the
'Xattries' (Kshatriyas) as 'great warriors', akin to 'the Rajpoots and
Marhatas who are but their descendants', with much celebration of
ancient Hindu cultural glory. Such Hindu nationalistic themes had
been inserted, it goes without saying, into the general tripartite
3? Partha Chatterjee, The Nation and Its Fragments {Princeton, 1993; Delhi,
1994), p. 88.
38 Ranajit Guha, pp. 17-18.
framework of ancient Hindu glory/medieval Muslim tyranny and

decline/ modern reawakening, and other contributors to the SAGK.
proceedings extended the formula to conditions of women and the
history of the Bengali language.39 A decade later Nilmoni Basak,
who had also been a member of that Derozian society,40 likewise
accepted without question Mill's periodization, but then launched
into a bitter attack on British writings on Indian history for denigrating Hindu achievements. A trend-setting feature of Basak's Bbaratvttrsbcr Itihas was the effort to shift the focus within the 'Hindu'
period from politics to culture and religion. Relegating to a closing
section the "brief description' of Hindu kingdoms difficult to
reconstruct, full of rabies, without a firm chronology Basak
embarked upon an enthusiastic account of theories of statecraft and
law, religion, literature and science, even Hindu colonies and cultural
influences allegedly in places as far distant as Bali and Peru.41
These are matters of detail, relevant only for their symptomatic
value: flu more crucial are the constraints and closures late-Subalternism is imposing through its key assumption of statism as the root
of all evil in modernity. The corollary often drawn is that modern
history-writing is necessarily state-centred: it is either narrowly
political in subject matter, or looks at other processes from the
point of view of the making or unmaking of states. Several recent
essays by Gyanendra Pandey, in particular, assume almost as a
matter of course that all post-Enlightenment historiography has
been the 'grami narrative' of the nation-state - till, presumably,
the present moment of liberation achieved through the contemplation of'fragments*.42 For Partha Chatterjee, similarly, the important
thing about late-nineteenth-century Bengali history textbooks is that
in them 'history had become merely the struggle for power'. 'Hindu
nationalism*, it seems, is unacceptable for Chatterjee (and Pandey)
in large part because 'like other modern ideologies, it allows for a

central role of the state in the modernization of society in this
sense, the framework of its reasoning is entirely secular.*43 This
position is close to that of Ashis Nandy, who has been critiquing
Hindutva for a number of years now from a consistently anti-secular
standpoint.
I have many differences with such assertions, which I intend to
elaborate in several of the essays that follow. For the moment, my
concern is only with the homogenizing silences they impose on
colonial Indian, and specifically Bengal, historiography. For, a striking feature of much late-nineteenth and early-twentieth century
Bengali history-writing and thinking about the past was precisely its
persistent critique of state-centred, merely political, histories. This
was a critique, further, that at times deployed arguments uncannily
close to some in common use today, for central to it was the
assumption that statism was a principal instrument of modern
Western cultural domination. Such recurrence appears both significant and worthy of exploration.
The valorization of culture over narratives of kings and wars in
Pyarichand Mitra or Nilmoni Basak had been a response to the
paucity, at that time, of firm, chronologically grounded data about
the ancient or 'Hindu' period, which was consequently then in some
danger of being dismissed by British scholars as having no history
in the sense of worthwhile past politics. An absence of information
about dynasties and wars had become much less of a problem by
the late-nineteenth and early-twentieth century, and yet it was then,
and most notably in and around the Swadeshi years, that samoj
(society, community) came to be regularly counterposed to rasbtra
or rajsbakti (state, the political domain). The real history of India,
it was repeatedly asserted, was located in the first, not the second,
for samaj embodied the distinctive qualities peculiar to the genius,
culture and religion of the Indian people.
In moves the theoretical lineages of which go back to Herder
romantic nationalisms in many parts of the world have often
identified value and authenticity with difference, with what is
supposedly distinctive and unique to a particular language, culture,
or history. A politics of identity grounded in the recognition o
such difference has thus been repeatedly counterposed to that of
equal dignity and universal rights to borrow the terms of a
20
39
Pyarichand Mitra, State of Hindoostan under the Hindoor, Maheshchandra
Deb, A Sketch of the Condition of Hindoo Women; Udaychandra Addya, Bangla
Bhasha Uttamrupe Sbiksbakantner Abashyakata, in Gautam Chattopadhyay,
Awakening in Bengal in Eariy Nineteenth Century: Selected Documents, vol. I (Calcutta, 1965), pp. 94, 131, 156, 178-80, and Appendix, i-ii.
*> See the list of members of the SAGK in Chattopadhyay, pp. lxiv-v.
*i Basak, Preface, and Chapters 2-7. Mrityunjoy's text in contrast had been
entirely about kings.
42
Gyanendra Pandey, 'In Defence of the Fragment Writing about HinduMuslim Riots in India Today', Economic and Political Weekly, Annual Number,
1991, and 'The Prose of Otherness', in Subaltern Studies VIII (Delhi, 1995).
Chatterjee, Nation, pp. 91, 110.
21
22

44
seminal recent analysis by Charles Taylor. The specific contexts

for its flowering in late-colonial Bengal still require investigation,
but there certainly were connections with what, many years ago, I
had described as a 'constructive swadeshi' trend during the movement against the Partition of Bengal. This valorized autonomous
self-help efforts in indigenous enterprise, education and village
organization over the politics of both 'Moderate1 and 'Extremist'
varieties, and was embodied most notably in the writings and
activities of Rabindranath Tagore and Satish Mukherji (editor of
Dawn, founder of the Dawn Society, and key figure in the national
education movement of the Swadeshi years).4s
The appeal of the state/society disjunction at this specific historical conjuncture was clearly related to intelligentsia disillusionment
both with 'improvement* under colonial hegemony and initiative,
and the possibilities of oppositional politics of what had widely
come to be termed the 'mendicant' kind a dual loss of faith that
was not always accompanied by enthusiasm about the new politics
of the extremist or terrorist varieties. Certain structural features of
colonial rule also provided a basis for the conceptualization of samaj
as autonomous from rashtra, and identifiable in the main with religious community rather than territorial nationhood. Census classification and enumeration helped to consolidate community boundaries
defined in terms of religion and caste, while colonial justice made personal and family laws into distinct religious domains within which
textualized norms of high-caste or ashraf social behaviour were
sought to be universalized in unprecedented ways.46 In census and
* Charles Taylor, The Politics of Recognition', in Amy Gutman (ed.),
Multiadtxralism (Princeton, 1994).
Sumit Sarkar, Smulabi Movement in Bengal, 1903-1908 (New Delhi, 1973),
Chapter II, and passim.
Warren Hastings laid down in 1772 that 'inheritance, marriage, caste and
other religious usages or institutions' were to be administered in different ways
to Hindus and Muslims, according to the 'Shaster' and Islamic jurisprudence
respectively: pandits and ulema were therefore appointed to aid British Indian
courts in a system which continued till 1864. In form, this was a continuation
of Mughal practice which had been marked by a similar duality in Diwani, as
distinct from Nizamat, adaiats, where Muslim criminal law prevailed. 'But by
far the greater part of litigation was never brought before Muslim officials, but
was settled by recourse to traditional methods of resolving disputes, which
differed according to the caste, the status in society, and the locality of the parties.'
J.D.M. Derrett, Religion, law, and the State in India (London, 1968), pp. 229,233.
For a specific instance of the legal homogenization brought about under British
23
law alike, 'colonial knowledge', it needs to be added, was not just a

Western superimposition: such an interpretation gravely underestimates the extent and significance of inputs from relatively privileged
Indian groups with autonomous interests and inclinations.47
The focus upon samaj as counterposed to politics could acquire
alternative stresses, logically distinct even though quite often intermingled within the same activity or text. One strand was clearly
populist, and manifested itself through appeals to the urban bhadralok to re-establish links with rural life. Concrete efforts in that
direction included village reconstruction initiatives, attempts to
promote elementary education and cottage crafts (as distinct from
Swadeshi textile mills or an alternative, Calcutta-cased 'national'
university), and the gathering-together of invaluable collections of
folk literature, songs and fairy tales.48 Yet samaj was simultaneously
all too often conceptualized in Hindu, high-caste gentry, and paternalist terms, and these were the nuances that the term itself tended
to carry over from earlier usages.4*
Both tendencies can be seen at work in extensions of constructive
swadeshi moods into historical retrospects. An initial consequence
of the valorization of samaj as actual or potential site of autonomy
rule, in this case actually restricting the rights of tower-caste women even while
implementing an undoubtedly progressive legislation, see Lucy Carroll, 'Law,
Custom and Statutory Social Reform: The Hindu Widows' Remarriage Act of
1856', in J. Krishnamurti (ed.), Women in Cobnut India (Delhi, 1989).
47
The Bengal Census Report of 1901 provides a concrete example of such
an input in a crucial area. E A Gait interpreted Census Commissioner H . H .
Risle/s instruction to classify castes in each region "by social precedence as
recognized by native public opinion' to mean that 'the decision must rest with
enlightened public opinion and n o t with public opinion generally.' Bur 'enlightened opinion would inevitably mean the views of highly educated Hindus,
i.e. overwhelmingly of upper-caste men. N o t surprisingly, the vast majority of
claims for higher status by subordinate castes were summarily rejected by the
Bengal Census Report Census (India), 1901, Volume 1.1, p . 538; Census (Bengal),
1901, vol. vi, pt I, pp. 354, 378-84.
48
Two striking examples would be Dakshinaranjan Mitra Majumcfar's Thahtmarjbuti (Grandma's Tales, published in 1907), and Dineshchandra Sen's discovery of the Mymensingh folk ballads- For an account of the many dimensions of
constructive swadeshi, see Swadeshi Movement in Bengal, Chapters 2 , 3 , 4 , and 10.
49
The Bengali word samaj referred to collectivities or gatherings of particular
castes (in the sense ofjatis, and more particularly, sub-divisions of the latter)
or religious sects (e.g. Brahmo 'Samaj') before its late-nineteenth-early-twentiethcentury extension to signify also more wide-ranging notions of society or
community.
24

was a fairly remarkable and precocious interest in social and cultural

history: very different, really, from the supposed British colonial
prototype, for it developed precisely around the time when the turn
towards professionalized accuracy on the Ranke model was making
such themes disreputable in Western academic scholarship. The two
main areas of original research were ancient Indian culture and
religion and, in Bengal particularly in the wake of the Swadeshi
upsurge, regional and local histories as well as extremely important
surveys of the development of Bengali language and literature. Here,
pioneering use was made of vernacular literary texts, oral folk
traditions, artistic works, and a very wide range of cultural artefacts.
To take one remarkable instance: Dineshchandra Sen's Bnhat Banga,
published in 1935 as the fruit of two decades of labour, began with
a declaration that 'the social, artistic and religious evolution of a
civilization does have some relationship with political history, but
the connection is not necessarily always close or vital/ A later chapter
argued that wanderings among the common people, and studying
their patterns of life, crafts and traditions, could often reveal more
about true history than poring over inscriptions or written texts. It
went on to illustrate this proposition through inferences teased out
from peasant ways of learning measurements and predicting the
weather, folksongs, and the material culture of boatmaking, houseconstruction, the weaving of quilts, and the preparation of sweets.
For Sen, it appears at times, the true repositories of Bengal's culture
have been plebeian, low-caste people bound up with everyday material
production, not the Brahman bearers of high Sanskrit learning. The
weakening of that high-culture-bearing strata under Pathan rule is
presented, most uncharacteristically, as a boon which opened the
way for the development of the Bengali ianguage.50
If Sen's populism, probably inspired by one kind of reading of
Vaishnava traditions, represented an effort at a kind of peoples'
history, there was also the slightly later and much more carefully
crafted initiative of Niharranjan Ray {Bangalir Itibas, Calcutta, 1949).
This attempted a veritable total history of pre-thirteenth-century
Bengal, with sections on ecology, economic conditions, land relations, caste and class structures, statecraft, religion, culture, and
everyday life. Ray, too, had spent years wandering through Bengal's
countryside, but as a Left-nationalist activist, and his work reveals
signs of Marxian influence.
so Dineshchandra Sen, Brihat Banga, 2v (Calcutta, 1935; repr., 1993), pp. v,
895-946.
25
It must be added immediately, however, that Niharranjan Ray

and even Dineshchandra Sen were hardly typical of the bulk of
writings built around the rashtra/samaj dichotomy. A local history
like Jogendranath Gupta's Bikrampurer Itibas (Calcutta, 1909) would
be a much more representative example. Gupta enumerated in great
and loving detail the past and present achievements of the Bengali
Hindu bhadralok in what was one of its classic heartlands. No one
would guess from reading his book that more than half the population of the region he was writing about were Muslims, or that, among
the Hindus, Namashudras considerably outnumbered the Brahmans,
Baidyus and Kayasthns combined. The bhadralok history of Bikranv
pur was emphatically not about people like them.51 As Niharranjan
Ray pointed out in the Introduction to his Bangalir Itibas, 'samaj'
had generally been understood 'in a very narrow manner*, excluding
the plebeian strata, even by those who had grasped what Ray
reiterated as the key feature of pre-colonial Indian life: its centring
around 'samaj1, not 'rasjjtra'.52
Partha Chatterjee's Retailed account of writings about history in
colonial Bengal does make fleeting reference to a trend in early
nationalist historiography which 'denied the centrality of the state
in the life of the nation'. The general framework he has adopted
leads him to locate 'the principal difficulty with this view, which
has many affinities with the later politics of Gandhism', in 'its
inherent vulnerability to the overwhelming sway of the modern
state.'53 Other kinds of vulnerabilities, of tie son implicit in the
silences of Gupta's history of Bikrampur, appear more obvious and
vital to me, at least so far as Bengal historiography is concerned.
The dearest evidence for them comes horn more general or programmatic statements about history conceived in terms of the state/society
disjunction, made during the Swadeshi years by Rabindranath Tagore
and Satish Mukherji.
Rabindranath wrote often about history between 1901 and 1912,
51
For some details o f Gupta's book, as well as census data about the religious
and caste composition of the Bikrampur region, see m y 'Kalki-avatar of
Bikrampur: A Village Scandal in Early Twentieth Century Bengal', in Ranajit
Guha (ed.), Subaltern Studies VT (Delhi, 1989).
52
Niharranjan Ray, Bangalir Itibas: Adiparba, vol. I (Calcutta, 1949, 1980),
PP- 2, 5.
53
The reference* extends over two paragraphs i n a forty-page analysis of T h e
Nation a n d Its Pasts', and 'Histories and Nations', in The Nation and Its
Fragments, p p . 76-115.
26
and these essays provide rich indications of a mind grappling with

a rashtra/samaj framework, and then in important ways going
beyond it through an auto-critique of some of his own earlier
assumptions: a complicated and contradictory process that attained
abiding literary form through Gora (serialized between 1907 and
1909). Bbaratvanher Itihas (1902), probably the best known of
Rabindranath's historical essays, used an interesting language that
pressed the politics/culture divide towards a Muslim/Hindu dichotomy which is never explicitly avowed in the text Thus the narrative
of wars and invasions, proclaimed by Tagore at the beginning of
this article to be no more than a bad dream not genuine, valuable,
history is immediately defined by him to have extended 'from
Mahmud's invasions to the imperial boasts of Lord Curzon*. 'In
the darkness caused by the storm and thunder of Mughal and
Pathan, our ancient temples had to cover their heads, while the
marbled, ornate tombs of the mistresses of Sultans soared to kiss
the stars.' The general denigration of statecraft, it seems, does not
extend to the wars and conquests of a Samudragupta, while the
exaltation of culture quickly slips into a firmly Hindu mould.54
In Nababanha and Brahman, two other 1902 essays published a
few months before Bharatvanbcr Itihas, Rabindranath spelled out
what was then his notion of ideal Hindu samaj, in terms explicitly,
even aggressively Brahmanical and patriarchal. (A combination that
seems almost inseparable: we may recall the conflation of over-mighty
Shudras and disorderly wives in the dystopia of Kaliyuga.) Inequality
is inevitable in all human societies, he argued, but India has given
appropriate respect to 'low and high, women and men'.55 He counterposed the entire society of gentlefolk (bbadrasampraday) who should
be given dwija (twice-born) status, to those considered 'Shudras1, in
ancient India as well as today 'Santals, Bhils, Kols, bands of
sweepers' for in a proper samaj 'neck and shoulders must not be
lowered to the level of the ground.* *We want to become dwijas, not
fcringhas't whereas today there was the danger of all Brahmans
degenerating into 'a vast society of tired clerks worn out by excessive
work.*56 And, still on the theme of the necessary inequality of
Bharatvanber Itihas (Bhadra 1309/1902), Rabindra Racbanabali, vol. IV
(Calcutta, 1940, 1975), pp. 377, 379.
ss Nababanha (Baisakh 1309/1902), ibid., p. 373.
56 Brahman (Asar 1309/1902), ibid., pp. 395, 401-2. I find this counterpositioning of proper hierarchy to the miseries of contemporary clerical life,
where 'the Brahman has to work with lowered head in the office of the sahib'
27
humankind, Rabindranath in his Nababanha essay attacked the

'modern wife' who, in imitation of the West, feels ashamed of
'serving' her husband and children. In the Indian tradition, however,
'sweeping the floor, bringing water, preparing food, eating after
everyone else . . . considering even husbands without any exceptional qualities akin to gods' have been rightly taken to be the hallmarks
of the 'grihalakshmi', the embodiment of true feminine grace and
beauty.57
Rabindranath, as is well known, soon moved away from most of
these positions. The Hindu-Muslim riots in East Bengal in 1906-7,
and the failure of the Swadeshi movement to enthuse the bulk of
the peasantry, set him thinking about the problematic features of
the samaj he had briefly idealized. Tagore, after about 1907, developed a powerful and consistent anti-communal critique, and by
1909 was condemning the samaj based on hierarchized caste difference as a 'gigantic system of cold-blooded repression'.58 The
repudiation of gender inequality was, perhaps less sharp or consistent,
but still a short story like StnerPatra (1914) stands in utter and total
contrast to the passage I have just quoted.
For a more consistent elaboration of the implications of a Hindu
communitarian ideology grounded in hierarchy we need to turn to
Satish Mukherji and his Dawn. Mukherji's The Question of Caste1
(Dawn, August 1903) proclaimed axiomatically 'that in all ages and
by virtue of a law of nature, there shall be inequalities and distinctions between man and man.' He reminded those who objected to
caste as being hereditary that property, too, descended 'from rather
to son*. Admittedly, an element of flexibility enabling some promotions or demotions on the basis of merit was advisable to allow
'proper placement and chance of transfer': but this necessarily
presupposed 'a group who can make the needed choices'. The
Brahmans, Mukherji concluded, have the best qualifications and
traditional expertise for this job of guardianship. Thus a mildly
reformist criticism of caste was neatly co-opted into a defence of
(p. 393) extremely significant, and will be discussing its significance in Chapters
6 and 8. Late-nineteenth-century modulations o f the Kaliyuga myth, we shall
see, are important primarily as an entry point into representations o f this
clerical world.
57 Ibid., p. 374.
58
Letter to Myron Phelps^4 January 1909, reprinted mModern Review, August
1910. For more details about the change in Rabindranath's views after 1907,
see my Swadeshi Movement in Bengal, pp. 62, 82-91.
28
Brahmanical order. Only through caste hierarchy, Dawn editorials

and articles repeatedly proclaimed, could 'progress' be reconciled
with 'order' or 'stability*. The editorial for the November 1900 issue,
for instance, urged the need to understand the specific 'laws1 of
India's 'social evolution'. It admitted the need to overcome 'the
present inertia of Indian society in many matters in respect of which
its hands are free, e.g., social, educational, religious and industrial',
but emphasized that this should be done only through pursuing 'a
course that is consistent with stability*}9 The language reminds us
that Satish Mukherji had some connections with a Positivist group
in late-nineteenth-century Bengal which had inflected the doctrines
of Comte in a highly conservative, Brahmanical direction. As so
often, caste and gender hierarchy were seen as interdependent, and
in March 1903 Dawn gave great prominence to the views of the
'eminent Hindu Positivist, the late Jogendrachandra Ghosh, Zemindar', that India's progress 'must be securely based on continuance
of the traditional family system.'60 The Bengal Positivists, I will argue
in a subsequent chapter, provide an illuminating case study of the
ways in which specific aspects of colonial structures and discourses
(legal recognition of a sphere of community-based personal law, and
fragments of Comtean theory) were used as resources to reaffirm
Brahmanical hegemony.61
Dawn regularly counterposed Brahmanical and patriarchal order
against the incessant competition and 'gospel of enjoyment' of the
West. It related the craze for increasing 'consumption per head' to
the 'undue importance attached to the doctrine of rights*, as manifested,
in its opinion, notably in the advance of democracy in Victorian
England.62 Its very first issue emphasized the need for India 'to steer
clear of the Labour Problem of Christendom'.63 The anti-capitalistic
note is quite striking, but so is the precise angle of attack as revealed
5 Italics in original. 'Indian Social Evolution and Reform' (Damt, November
1900). See also 'Principles of Social Order: The Statical Aspect' (Editorial, Doom,
March 1900) which reiterated that 'all progress is built on Order and is
delusive, and even mischievous, when it is not built on order . . . *.
60
For a more detailed discussion of this Positivist<um-Brahmanical inflection of the rashtra/samaj, see Chapter 9. The pioneering, and still the most
detailed study of Comte's Bengali disciples is Geraldine Forbes, Positivism in
Bengal (Calcutta, 1975).
Chapter 9.
*2 'Social Movements Round a Centre' (anon., Dawn, August 1897).
63
The Situation in India: A Problem and an Illustration* (anon., Dawn,
March 1897).
2?
by the kind of (totally inegalitarian) samaj or community being

posited against it. And here, it has to be added, the anticipations in
Dawn of some very contemporary trends become really startling. An
article entitled Western Ideal of Nationalism (unsigned, but probably
by Satish Mukherji) in Dawn, June 1911, contrasted the "Western'
notion of 'political nationalism', focussed upon the 'development of
men's activities as members of a state', with the Hindu ideal of community based on regulated, hierarchized difference. The unity sought
by Hindu society was not something 'homogeneous', but based on
dbarmashastras that laid down differentiated 'standards of righteous
conduct adapted to various and varying . . . classes and divisions of
people'. Unity came also from 'allegiance to the framers of these
Laws, who form a distinct spiritual order. . . .' Western 'political
nationalism', in contrast, sought 'a homogeneous political existence'
through 'a suppression of all diversity*. Mukherji traced its origins
back to 'France during the Revolutionary epoch of the eighteenth
century*, and in particular to 'the French Encyclopaedists . . . who
in the name of equality and fraternity had preached a jehad against
all that men and nations held sacred.' Such ideas, the article implied,
had been imposed on India by Bentham and Mill, with a minor
modification that substituted utility for the 'goddess of Reason'. Large
parts of this essay, one is tempted to comment, could walk into a
contemporary anti-'Orientalist' collection with a minimum of editorial updating: they counterpose, in remarkably clear language, an
ideal of cultural difference premised on internal hierarchy against
notions of universal rights which are felt to be homogenizing.
But of course the argument was anything but purely indigenous
or traditional: even a cursory glance through the issues of Dawn
indicates a striking degree of derivation from that other, more
insidious kind of Orientalism that patronizes and praises, instead
of denouncing, an equally essentialized Orient. A Cambridge don,
Oscar Browning, was quoted with great approval for his statement
that his Indian visit had taught him 'to tolerate purdah, and to have
an admiration for caste',64 while many of the strongest assertions of
patriarchal values come from the pens of Annie Besant and Sister
Nivedita.65
Dawn, July 1903.
6
* Thus Annie Besant argued in an article entitled The Education of Hindu
Youth' (Daawjune 1897) that passing the matriculation examination was useless
for Indian girts, who should be trained rather in 'devotion and piety'. Any
imitation of the West with respect to the education of women could 'break up
30
Assertions of values that otherwise would have been considered

socially retrogressive are quite often explained or even justified today
as valid responses to the all-pervasive colonial authority and interference. It therefore becomes important to note that the Dawn variety
of'constructive swadeshi' was not conspicuously anti-colonial, so far
as politics was concerned. Indeed, the politics/society disjunction in
this case permitted at times a rather remarkable degree of loyalism.
In February 1898 Dawn justified Satish Mukherji's Bhagavat Chatuspathi (a 'Hindu Boarding Religious Institution' to train students
"under a system of Hindu discipline') in part on the grounds that
'in this way alone could we live happily amongst our rulers, and
setting an example of lofty character repay them tenfold the debt
which we owe them for the era of uninterrupted peace and tranquillity which India had not enjoyed for many and many a day until
she came by the dispensation of an All-wise Providence under British
overlordship.* That was 1898: thirteen years later, after the storms
of Swadeshi had come and gone, Satish Mukherji was hailing the
'transcendental importance* of the visit of the King-Emperor, and
arranging to present George V with a full set of the copies of his
Doom.66 The important point that seems to emerge is that the
refurbishing, or invention, of ideologies of Brahmanical hegemony
and patriarchy under colonial rule did not necessarily flow from
anti-colonial impulses atone: more internal compulsions and power
relations also deserve attention.
I have been emphasizing a precocious, if in many ways problem-ridden, thrust towards social history. A qualification that needs
to be made at this point is that such tendencies had developed
the family system, drive the women out in the world to earn their living, make
them competitors with men
' A n earlier essay by Besant on Hindu women
(1894), which Dawn reprinted in October 1901, had extolled the charms of
chaste widowhood. It admitted that Hindu ideals of womanhood could have
no place in the West, but pleaded, in a classic statement of one kind of
Orientalism: 'Leave the Hindu woman untouched by Western thought, and do
not destroy a type just because it is unique . . . We have women enough, who
are brilliantly intellectual and competent: let us leave unmarried the one type
which is the incarnation of spiritual beauty.' Nivedita, too, was full of admiration for the 'nun-like qualities* of the Hindu widow, and felt that 'there are
few great relationships in human life like that between a Hindu man and his
mother.' (Dawn, May 1903.)
** "The Imperial Visit1 (Dawn, December 1911). For the presentation of Dawn
to George V, see Haridas and Uma Mukherji, Origins of the National Education
Movement (Calcutta, 1957), p. 249.
31
primarily outside the world of the professional Indian historian,

which in any case took quite some time to constitute itself.
University departments began systematic research in history rather
late, as they had been primarily examining-cum-teaching bodies till
the early twentieth century, and an all-India organization of the
historical profession (the Indian History Congress) was floated only
in the 1930s. As had happened in the West a generation or two
earlier, the turn to the Ranke model of academic precision and
strictly archive-based history placed a heavy premium on politicalmilitary-administrative narratives. The tone was set by the major
British Indian surveys: Vincent Smith's Oxford History of India (1919)
and the Cambridge History series published during the inter-war
years, while among Indian historians Jadunath Sarkar emerged as
the most respected and influential scholar through his predominantly political works on the Mughal empire and the Maratha
kingdoms.67 A state-oriented Indian history had thus come to
dominate academia now, largely displacing social-cultural interests68
without however fundamentally modifying many of the underlying premises of that other kind of work. Thus the more abundant
and precise data that had now become available about the ancient
(or 'Hindu') period led to dynastic histories marked often by an
uncritical preference for alleged periods of 'imperial unity', particularly the Guptas, Asoka Maurya remaining a bit suspect because
of his Buddhist affiliations. Imperial unity, however, ceased to be
such a plus point if the rulers happened to be Muslims, for, as in
nineteenth-century historical novels, the wars of sections of Rajputs,
Marathas, and Sikhs with centralizing Muslim rulers were generally
given the status of national struggles. Another revealing discrepancy
consisted in a variation across time of the degree of attention
professional historians were prepared to give to social-cultural
matters. Ancient Indian civilization and culture still attracted a lot
of attention in syllabi and research alike, quite often in highly
apologetic, even revivalist forms. Similar themes were much less
studied or taught for the 'Muslim' period, except by a few firmly
67
Jadunath Sarkar, however, had the imagination, flexibility and grace to

hail Niharranjan Ray's book as a landmark in what h e declared would be
increasingly recognized as the 'highest' kind o f history: social history. See his
preface to the first edition o f Bangalir hibas, p. x.
68
In Bengali-language works, too, the new political focus was exemplified
in Ramaprasad Chanda's Gaur Rajmala (Rajshahi, 1912) and Rakhaldas Bandopadhyay's Banglar Itiftas (Calcutta, 1914).
32

anti-communal nationalist historians who tried to highlight 'syncretic' Bhakti-Sufi movements and foregrounded Akbar against
Aurangzeb.69
One curious feature of history-wiring about the 'modern* or
colonial period prior to the 1950s needs some emphasis. Nationalist
historiography developed on sites some distance from what, on
logical grounds, should have been its proper location: the rich and
growing traditions of contemporary anti-colonial movements. There
was virtually no professional research on such themes (or on 1857)
till some years after independence, and the history of colonial India
consequently remained very much a narrative of viceroys, Afghan
or Burmese wars, and administrative and 'constitutional' reforms.
Home Department files or private papers for recent years were
largely inaccessible, most academic historians worked in government-controlled or financed institutions, and with the rise of mass
nationalism (as well as revolutionary terrorism and Left formations)
the factor of censorship (and, more often, self-censorship) had
probably become much more important than in the more placid
late nineteenth century. Even the critique of British Indian economic
policies worked out by Moderate Congress intellectuals like Naoroji
or R.C. Dutt seldom entered standard history textbooks: certainly
I cannot recall such themes in my college courses, a decade after
1947. The social-historical impulse also tended to wither away for
the colonial period, with the major exception of the middle class
studying its own cultural origins in an increasingly self-adulatory
manner through the renaissance myth. Caste and religion in colonial
times were probably felt to be divisive themes, from the perspectives
of countrywide unity and anti-British struggle.
Through silences and stresses alike, the bulk of late-colonial
Indian professional historiography came to have a tilt that was
strongly Hindu, as well as North Indian. The alternatives that
sometimes emerged within that same milieu were based on simple
inversions of mainstream assumptions, and hence did not mark
any qualitative break. Thus there were occasional writings which
glorified Muslim rulers, eras of pan-Islamic grandeur, and powerful
Southern or regional dynasties, in equally uncritical ways. What
remained fairly ubiquitous were views from the top: whether Northern, Southern, or regional, Brahmanical/high caste or asbraf. By far
69 An obvious example is Tarachand, Influence of Islam on Indian Culture
(1922; reprinted Allahabad, 1963).
33
the most influential model, of course, was that of a fundamentally

harmonious 'Indian' civilization and culture, all too often implicitly
identified with 'Hindu' traditions of alleged catholicity, with an
underlying Brahmanical or high-caste slant. In a simultaneous move,
the sting was sought to be removed from inconvenient questions
of gender oppression by postulating a Vedic or ancient golden age
of learned and respected women subsequently shattered by foreign,
usually Muslim, intrusions.70
There were some signs of an inversion of a more fundamental
kind. In Maharashtra and Tamilnadu, in the wake of powerful
lower-caste movements, alternative versions of history were constructed which stood the theory of the assimilative spread of 'Aryan*
civilization on its head, and projected a counter-myth of NorthernBrahmamcal foreign conquest and tyranny over the indigenous
'bahujan samaj' of intermediate and low castes. The Shivaji projected
in Jyotirao Phule's ballad about him in 1869 was primarily a
Kunbi-Maratha folk hero distinguished by concern for peasants,
while Gbutamgiri (1873) dismissed 'fictions* about his 'freeing the
motherland from MUccbas and protecting Brahmans and cows' as
'false religious patriotism'.71 In the 1920s, the Chamars of Punjab
would use the recent discovery of Harappan civilization to develop
a similar anti-Brahmanical Aryan rhetoric.72 Even in Bengal, much
less known for its caste politics, the thrust towards bhadralokdominated social and regional histories of the Swadeshi and postSwadejhi years actually coincided with a quite independent stream
70
Traces of this myth can be seen already in a paper on the conditions of

Indian women presented to the Derozian Society for Acquisition of General
Knowledge (Maheshchandra Deb, A Sketch of the Condition of Hindu Women,
reprinted in Gautam Chattopadhyay). It attained the status of a historical
commonplace through Altekar's The Position of Women in Hindu Civilisation
(1938). Two excellent recent critiques are Uma Chakrabarti, 'Whatever Happened
to the Vedic Dasi? Orientalism, Nationalism, and a Script for the Past', in
Kumkum Sangari and Sudesh Vaid (eds), Recasting Women: Essays in Colonial
History (New Delhi, 1989), and Kumkum Roy, '"Where Women are Worshipped,
there the Gods Rejoice": The Mirage of the Ancestress of the Hindu Woman',
in Tanika Sarkar and Urvashi Butalia (eds), Women and the Hindu Right (New
Delhi, 1995).
7
1 Rosalind O'Hanlon, Caste, Conflict and Ideology (Cambridge, 1985), Chapter
10; Collected Works ofMahatma Jyotirao Phule, Volume I (Slavery) (Bombay, 1991),
p. 26.
72
M a r k Juergensmeyer, Religion as Social Vision: The Movement against Untouchability in Twentieth-Century Punjab (California, 1982).
34
of tracts. These expressed lower-caste grievances and aspirations, and

constituted for them, sometimes, imagined pasts built out of selective
appropriations from elite myths and histories. Thus the Rajbansi
claim to Kshatriya status could be buttressed through the Brahmanical myth of the destruction of Kshatriyas by Parashuram,73 while the
metrical biography of the founder of the Matua sect which had laid
the foundations of the Namashudra movement in Faridpur laid
claim to the anti-caste heritage of the Buddha (and Kabir) a generation before Ambedkar.74 Alternative historiographies like these have
been generally, and symptomatically, ignored by academic scholarship. Today they seem on the point of becoming a formidable force,
as an opportunist BJP-BSP (Hindu upper<aste and trader with Dalit,
lower caste) alliance in Uttar Pradesh breaks down partly through
Dalit insistence on celebrating, precisely in the state where Ayodhya
is located, an anti-Brahman leader of far-off Tamilnadu who had
publicly burned pictures of Ram on Madras beach in 1956.75 Implicit
here is a very different way of imagining, not only the subcontinent's
pasts, but perspectives of national unity or integration.
Ill
I have been emphasizing the differences within late-colonial Indian
historical thinking, in particular a contrast between social-historical
impulses mainly generated outside the formal historical profession,
and statecraft-oriented narratives written from within its confines.
Certain features common to both appear equally significant, however,
when looked at from today's perspectives, and in terms of their
conditions of production and dissemination. These demarcate the
late-colonial situation quite sharply from the many historical worlds
of today, and consequently offer a vantage point for a brief review
of contemporary opportunities and predicaments.
There were two notable absences. The Asiatic Society and the
73
The ancestors of the Rajbansis, it was claimed, were Kshatriyas who had
taken refuge in the wilds of North Bengal to escape the wrath of Parashuram,
and had subsequently forgotten their high-caste origin and customs. Harakishore
Adhikari, Rajbansi Kulapradip (Calcutta, 1908).
w Tarakchandra Sarlcar, Sri Sri Harililamrita (P.O. Olpur, Faridpur, 1916).
For a more detailed discussion of such alternative constructions, see Chapter 9.
75
For details about Periyar*$ burning of images of Ram, as well as his other
violent attacks on the Ramayana, see Paula Richman, 'E.V. Ramaswami's
Reading of the Ramayana', in Paula Richman (ed.), Many Ramayanas: The
Diversity of a Narrative Tradition in South Asia (Delhi, 1994).
35
Anthropological Survey apart, official funding for pure research,

detached from pedagogy, hardly existed, and there was very little of
today's accelerating globalization which has made trips abroad for
degrees, research or seminars an important part of the more prestigious kinds of academic life. Opportunities for any kind of higher
education were more restricted and therefore even more dass-cumcaste defined than today, given the far fewer universities and colleges.
Within this smaller educated community, however, the hierarchical
divisions between research/teaching, university departments/undergraduate colleges/schools, metropolitan/provincial universities seem
to have been somewhat less sharp. Repositories of books, manuscripts, art objects and cultural artefacts were often built up by
autodidacts, gentlemen with access to local resources and antiquarian
interests but little formal academic training: a zamindar, lawyer or
schoolteacher could sometimes contribute as much or more as a
university professor. For Bengal, one thinks immediately of the
Bangiya Sahitya Parishad, many local libraries, and the Varendra
Research Society, the latter located in a small North Bengal district
town (Rajshahi) yet enjoying at one time an academic prestige which
it would be difficult for any non-metropolitan centre to emulate
today. Another example of this relative absence of internal hierarchization within a smaller educated elite is provided by the career
of Sir Jadunath Sarkar (1870-1958). A Rajshahi zamindar's son,'
Jadunath's formal degrees were in English, and till retirement he
combined research with the teaching of History, together sometimes
with English and Bengali, mainly to undergraduate students (at
Ripon, Metropolitan and Presidency Colleges in Calcutta, followed
by Patna and Cuttack, and then briefly at the Benaras Hindu
University). Jadunath became internationally renowned but never
went abroad.76
Late-colonial histories, then, were generally written by teachers
for students or general readers. Very many of the topmost professional scholars also produced textbooks, and most of them published
original works both in English and in indigenous languages. There
was therefore much less of a gap than is evident now between the
best and the worst or even average histories. But it would be
dangerous to romanticize: inadequate funding for full-time research,
confinement within national or regional parameters in the absence
of opportunities for wider contacts, the restrictive aspects of a
Biographical data taken from S.P. Sen (ed.), Dictionary of National Biography,
vol. IV (Calcutta, 1974).
37
nationalist paradigm shot through with unstated class and high-caste

assumptions (quite often sliding into communalist attitudes), all
exerted a price. The 'best' scholarship of those times, with rare
exceptions, appears unacceptably limited, parochial and unselfquestioning today.
Post-independence historiographical developments, in contrast,
have been marked by a dialectic which simultaneously enhanced
standards vastly at elite levels, while paying far too little attention
to histories being taught to the majority of college and school
students as well as diffused through other means among the general
public. Advanced historical research has come to have as its intended
audience one's academic peer-group, research students of the best
universities, and, increasingly, international conferences. Meanwhile
the now very seriously dated historiography of a past generation has
kept on getting reproduced and disseminated, in diluted and crude
forms, at other, inferiorized and neglected levels. Thus has come to
be constituted a 'common-sense' using that term in the most
negative of Gramsci's several different formulations77 open to
appropriation and orchestration by organizations such as the Sangh
Parivar.
There has certainly been a qualitative transformation in the work
of the leading practitioners since the 1950s, bound up with very
significant shifts in basic approaches and choice of research questions. In ancient and medieval Indian historiography, where the
changes have been most obvious, work from the late 1950s has
focussed on themes like 'social formations',78 debates about the
existence and nature of Indian feudalism, or the possibilities of
capitalistic developments in pre-colonial times. Inscriptions and land

grants have been probed, no longer primarily for information about
kings, dynasties or conquests, but for the inferences that could be
teased out from them on broader socio-economic relationships and
questions of state formation. Impressive detailed studies of medieval
agrarian, artisanal or commercial structures have similarly taken the
place of old-fashioned dynastic and military histories.
Ranke, then, has been displaced in considerable measure by Marx.
The specific direction of this change owes much to the overall
conjuncture of the 1950s and 1960s, marked by a strong and apparently growing Left presence in Indian political and intellectual life.
Mere enhancement in research opportunities obviously cannot explain it> nor was it just a question of wider international contacts.
It was not mainstream British or American historiography, not even
writings on South Asian themes, but a journal like Past and Present,
the 'transition debate', and the work of historians like Hill, Hobsbawm and Thompson often sought to be marginalized by academic
establishments in the West that appeared most stimulating to
Indian scholars exploring new ways of looking at history.
The new history had been iconoclastic in the 1960s; today, in
leading universities as well as in the Indian History Congress (though
hardly elsewhere), it has been functioning as a kind of establishment
for almost a generation. This provokes, nowadays, a certain legitimate
impatience about the occasionally simplified and restrictive nature
of its applications of Marxism, and attempts at more wholesale
repudiation are not unlikely, given the context of the collapse of
socialist regimes and the sharp Right turn in recent world and Indian
politics. It is important, therefore, to retain a sense of the sheadistance that separates the post-Kosambi or post-Irfan Habib historical world from what had preceded it, even while developing the
qualities of self-criticism vital for any living tradition of radical
historiography.
An aspect of this 'shift in the paradigm',79 one which has not
been much emphasized but is particularly relevant for my present
argument, is the rupture with a conventional nationalist historiography which, when transposed into ancient or medieval times, all
too often had become indistinguishable from communalism. To cite
only a few obvious instances: the casualties of the transformation
36
77 Gramsci, who always made clear that for him there could never be 'just
one c o m m o n sense', quite often emphasized its 'fragmentary, incoherent, and
inconsequential' aspects, as a kind of bricollage of residues from the high
cultures and 'prejudices from all past phases of history . . . ' But it could also
include elements evolving from below, as it were, from shared experiences in
labour and in social relations, with an embryonic oppositional potential.
Assertions of lower-caste identities through imagined histories have obviously
n o intrinsic superiority in sheer academic terms over dominant-caste constructions, but they do seem to include elements of common-sense of the latter
kind. Antonio Gramsci, Selections from Prison Notebooks, ed. Quentin Hoare and
Geoffrey N . Smith (New York, 1971), pp. 323-7, 419-25. See also the helpful
comments of E.P. Thompson, Customs in Common (London, 1991, 1993),
pp. 10-11.
7
8 The major breakthrough, of course, was D.D. Kosambi*s^ Introduction
to the Study of Indian History (Bombay, 1956).
79
Romila Thapar has used this phrase to describe the impact of Kosambi:
The Contribution of D.D. Kosambi to Indology', in Romila Thapar, Interpreting
Earfy India (Delhi, 1992).
40
at times, it is true, by Left studies of class struggles of workers and

peasants. But once again socio-cultural dimensions tended to get
marginalized, as simplified Leninist frameworks gave priority to a
combination of economic pressures and 'external* organization, and
tended often to get lost in debates on the correctness or otherwise
of party strategies.
The would-be social historian of modern India had perforce often
to turn for guidance to social anthropologists. But that too has been
a domain full of problems, where conservative attitudes have often
blended with structural-functionalist premises to produce an abundance of bland, tension-denuded categories. Thus all kinds of caste
mobility, including radical protest, have been grouped together under
one label, 'Sanskritization', indicative primarily of the most assimilative kind of change. A parallel instance would be the' jajmani system',
where elements of mutuality in relations between the Brahman ritual
expert and the householder have been extended to the rather different
transactions of peasants and artisans, even landholders and agricultural labourers.
That the standard anti-colonial nationalist model, at least in its
more totalizing, un media ted versions, can be constrictive is indicated
by the fact that much interesting and stimulating recent work has
been taking place outside, or in a tangential relationship to, its
boundaries. Take for example the notable growth areas of economic
history, and studies, often over the long-term, of the grossly neglected
South. The framework of abrupt and cataclysmic late-eighteenth
and early-nineteenth-century economic decline that still works fairly
well for some regions of early colonial penetration (most notably,
Bengal) does appear somewhat less helpful, for other areas. (A work
like Chris Bayly's Rulers, Townsmen and Bazaars*1 however controversial at times, cannot be dismissed as simply neo-colonialist) The
rapidly growing genre of environmental or ecological history, to
take a second example, often has to work with scales of time that
need not coincide with conventional periodizations. Parts of the
South, again, seem to have had patterns and rhythms of their own:
looking outwards commercially towards and across the Indian
Ocean rather than inland, and marked in the inter-War years more
by lower-caste assertions (and in some regions by Left movements)
than mainstream Gandhian nationalism or Congress and League
politics.
u C A . Bayly, Rnlm, Townsmen and Bazaars: North Indian Society in the Age of
British Expansion, 177&-1870 (Cambridge, 1983; rpt. Delhi, 1993).
1HE MANY WORLDS OF INDIAN HISTORY
41
There developed, in the 1970s and early 1980s, a conjuncture

during which conventional nationalist and Left-nationalist premises
seemed on the point of more direct challenge, even decisive ovwthrow. As in the 1950s, a changed overall context was crucial. Its
constituent elements included the (slightly delayed) academic fell-out
from worldwide moods of radical optimism characteristic of the
1960s and early 1970s, and the rise within India of a variety of
extreme-Left tendencies which combined disillusionment with organized Marxist parties with hopes of an impending peasant revolution.
As the prospects of radical change withered, there was a proliferation
of volunteer groups engaged in constructive work at the grassroots,
while women's movements with self-consciously feminist perspectives
emerged as a novel and permanent element in the Indian scene.
-Meanwhile, lower-caste protests were gathering strength, forcing, after
the Mandal flare-up in the mid-nineties, considerable rethinking
among Left activists who had for long underestimated its autonomous appeal. History-writing was modified in this changed conjuncture in two more or less parallel but largely unconnected ways:
the sudden popularity of 'histories from below* (early Subaltern
Studies, of course, but also quite a lot of work outside and sometimes
preceding it),83 and a quantum leap, virtually from scratch, in
women's studies, increasingly informed by feminist approaches.
The first wave of Indian feminist scholarship original, powerful,
but nowadays largely neglected or forgotten questioned the triumphal narrative of unilinear advance in the 'status of women' through
male-initiated nineteenth-century social reform, followed by women's
participation in Gandhian, revolutionary-terrorist or left-led movements. More nuanced and ambiguous patterns were suggested, emphasizing the contradictions of reform and die ways in which nationalism could have displaced the women's question and recuperated
patriarchal ideologies and structures, even while opening up public
spaces for women, and there were interesting efforts to relate shifting
gender relations to detailed studies of socio-economic processes.84
83
I attempted a survey-cum-analysis of the early moves towards histories

from below in Popular Movements and Middle-Class Leadership in Late Colonial
India (Deuskar Lecture, given at the Centre for Studies in Social Sciences*
Calcutta, 1982; published Calcutta, 1983); see particularly p. 74, fh. 3. Like my
Modern India 1885-047 (Delhi, 1983), I had drafted this lecture before reading
Subaltern Studies I (Delhi, 1982).
84
1 owe this assessment of early feminist historiography to Tanifea Sackac
"Women's Histories and Feminist Writings in India: A Review and A Caution*
43
Meanwhile, there was a spate of research publications on tribal,

peasant and labour movements, as well as a few pioneering, sympathetic studies of lower-caste initiatives in large part independent
of, or even hostile to, mainstream nationalism. The generalizations
that emerged from some of this work were not dissimilar at times
to those being worked out independently by historians of women.
An inverse relationship was suggested between moments of popular,
specifically peasant, autonomy, and Gandhian nationalism in its
more organized forms and phases. Subaltern Studies, in particular,
began with a programmatic statement simultaneously critiquing the
elitism of both colonialist and nationalist historiographies.85 The
habit of looking at history solely from the top downwards, in terms
of leaders mobilizing the masses through ideals, charisma, or
manipulation, it was cogently argued, has often coincided with
economistic assumptions: both had combined, even in Left historiography, to obstruct efforts at studying the consciousness and
culture of subaltern groups.
those embodied in that alleged quintessence of post-Enlightenment

rationality, the bureaucratic nation-state). The adoption of the single
criterion of subordination or otherwise to modern Western powerknowledge is not too distant, surely, from the familiar digits of
cultural nationalism. And the related tendency to valorize all
assertions of indigenous community values is likely to inhibit
sympathetic explorations of a vast range of initiatives by or on
behalf of subordinated groups: women, lower-castes, Left-led peasant
and workers movements, all of which have selectively appropriated
elements from Western ideologies. Later chapters will provide instances of how such inhibitory pressures are already at work, even
in the writings of scholars with undeniably radical values.87 Histories
from below have ceased to be in vogue, being displaced by a focus
on colonial or elite discourses, and feminist studies of the nineteenth
century often dwell obsessively on the limitations of West-inspired
reform initiatives.
42
I have argued later in this volume that the possibilities that had
opened up a decade or so back, the chances of a social-historical
breakthrough, have today become restricted once again, and that
this, too, has happened in contexts both worldwide and specific
to India.86 For the moment I will merely suggest a connection
between such closures and a paradoxical kind of nationalist recuperation associated with critiques of colonial discourse, particularly in
the dominant strand within today's Subaltern Studies. Paradoxical,
both in terms of the starting point of that project, and because
the critique of official, state-centred nationalism has not been given
up. But a two-fold displacement has occurred: from colonial domination to Western cultural conquest; and from subaltern, usually
peasant, consciousness (often marked by the centrality of religion,
but not detached from questions of class, exploitation, and power)
to affirmations of community consciousness in effect defined by
religion and abstracted from indigenous power relations (other than
(Plenary Session Address, Seventh Berkshire Conference, Chapel Hill, North
Carolina, June 1996), forthcoming. A fine example of this earlier work is J.
Krishnamurti (ed.). Women in Colonial India: Essays on Survival, Work and the
State (Delhi, 1989), consisting of essays published several years earlier in the
Indian Social and Economic History Review.
85
Ranajit Guha, 'On Some Aspects of the Historiography of Colonial India',
in Guha (ed.), Subaltern Studies I (Delhi, 1982).
** See Chapter 3 below.
A clarification may be needed at this point I am far from

suggesting any rupture with basically anti-colonial parameters in
writing the history of colonial India. Colonial exploitation and
domination of course constituted the central set of relationships
during these centuries. In so far as its cultural manifestations have
been highly productive of undifferentiatcd illusions of progress,
modernity or reason, the critique of colonial discourse does have its
uses though more perhaps in the West, and in English Literature
circles, than in the specifically historical world of South Asia, where
many of its findings sound rather familiar. There remains a need to
recognize nuances and mediations, variations in the extent of colonial
cultural or other domination across times, regions, social spaces, and
the possibility of earlier tensions (around caste and gender, notably)
being reproduced in ways no doubt conditioned by the colonial
presence, but not uniquely determined by it. The traditional, orthodox-Marxist way of looking at the colonial world in terms of a
series of class-determined oppositions to an alliance between imperialism and a subordinated feudalism rightly appears problemridden, stilted, and reductionist today. But it did provide some space
87
A major example would be certain significant silences in Partha
Chatterjee's comprehensive recent work, The Nation and Its Fragments: Colonial
and Post-Colonial Histories (Delhi, 1994). Chatterjee's subsequent article,
'Secularism and Toleration', Economic and Political Weekly, XXK, 28, 9 July 1994,
is the clearest embodiment so far of the slide from subaltern through peasant
to religious community. For a more detailed discussion, see Chapter 3.
44
45
for distinguishing between varieties of nationalism in terms of their

social perspectives and composition. A glance back at the pages of a
journal like Dawn suggests that we are today in some danger of
unwittingly reproducing the assumptions and values of a particularly
narrow and elitist cultural nationalism. What is required, perhaps, is
some equivalent of the 'doubled' (or better, multiple) vision socialistfeminist historians have been struggling to attain. Gender in capitalist
societies cannot be understood in total separation from class: it has
repeatedly proved disastrous to collapse the one into the other.88
Much more is at stake here than merely academic historiography.
I have argued elsewhere that the shift towards criteria of indigenous
'authenticity' and 'community' can constitute, however unwittingly,
certain dangerous common discursive spaces, for already some of
the more sophisticated ideologues of Hindutva have started using
similar categories and arguments.89 My glance back at non-statist,
samaj-oriented patriotic histories should raise some doubts also
about the strategy, often advocated nowadays in the spate of writings
about communalism (provoked by recent developments), of postulating traditional catholicities against the homogenizations being
projected by Hindutva or Muslim fundamentalist groups: 'authentic
community-consciousness', so to say, against 'communalisms' ultimately attributable to colonial discourses.90 A journal like Dawn,
for instance, carefully kept away from the communal numbers game
that had begun in Bengal soon after the decline of the Swadeshi
movement,91 argued that homogenization was contrary to the true
spirit of Hinduism and an offshoot of Western cultural domination,
and yet quite aggressively asserted high-caste, patriarchal values.
Such assertions were actually more blatant at times in catholic,

non-communal texts that were not primarily engaged in efforts to
build Hindu unity against the Christian or Muslim Other. Projects
for such unity, in partial contrast, on occasion seemed to demand
assimilative caste reform: U.N. Mukherji, once again, provides an
excellent Bengal example.92 The emergence, already around the
Swadeshi years and on a vastly enhanced scale today, of alternative
lower-caste histories and conceptions of solidarities makes the
reiteration of indigenous, undifferentiated community-values highly
problematic. Their concordance with contemporary feminist values
would be as difficult.
Much of the appeal of late Subaltern Studies, as well as some of
the criticisms it has evoked, flow from its apparent affinities with
aspects of postmodernism or more precisely perhaps with what
has come to be called postcoloniality. These include the centrality
of anti-Enlightenment rhetoric, the oscillation between 'community'
and 'fragment', and an occasional toying with moods of epistemological uncertainty. Postmodernisms in recent years have swung
sharply between what can amount to affirmations of identity-politics
(as counterposed to the allegedly homogenizing politics of universal
and equal rights going back to the Enlightenment), and celebrations
of hybridization, of identities disintegrating as globalization intensifies.93 I have argued in another chapter that I find the one pole as
unacceptable as the other, particularly in post-Ramjanmabhumi
India, but also that the parallels with postmodernism, whether drawn
in admiration or as critique, I believe to be based in large part on
misrecognitions.94 Rhetoric against other people's metaphysical totalities apart, there has been very little, really, in late Subaltern Studies
of reflexive, self-doubting moods and methodological disquiet raised
by the, problematizations of language in recent years. This to me is
a matter of some regret, for such reflexivity can have considerable
value. It has helped to make an increasing number of historians far
more self-aware and questioning about the representations they use
J o a n Kelly, 'The Doubled vision of Feminist Theory 1 , in Judith Newton,

Mary Ryan a n d J u d i t h Walkowitz (eds), Sex and Class in Women's History
(London, History W o r k s h o p Series, 1983).
89
Sumit Sarkar, T h e Anti-Secularist Critique of Hindutva: Problem of A
Shared Discursive Space', in Germinal- Journal of Department of Germanic and
Romance Studio, Delhi University, vol. I, 1994. See also Chapter 3.
90
That, roughly, seems to be Gyanendra Pander's argument in The Construction of Communalism in Colonial North India (Delhi, 1990).
See the pioneering discussion of this theme in P.K. Datta, ' "Dying
Hindus": Production of Hindu Communal Common-Sense in Early 20th
Century Bengal', Economic and Political Weekly, 20 June 1993. Dawn seems to
have carefully side-stepped the question of the alleged decline of Hindu numbers
in Bengal as compared to Muslims, raised by U.N. Mukherji in 1909: see the
unsigned essay *Wno are Hindus and Who are Not', in Dawn, January, February,
June 1911.
92
See Chapter 9 for some elaboration of this a r g u m e n t

For t w o recent critiques o f this 'tendency . . . t o waver constantly between
the opposing polarities of cultural difTerentialism and cultural hybridity1, see
Aijaz Ahmad, T h e Politics of Literary Post-Coloniality', from which I have
taken the quoted passage (Race and Class, 36, iii, January-March 1995), and
Terry Eagleton, 'Where do Post-Modernists Come From?', Monthly Review, 47,
iii, July-August 1995.
See Chapter 3.
93
46
as their 'sources', the categories they employ, and the rhetoric implicit
in their writings, and much fine work is being done nowadays
dancing, as it were, at the edges of relativism.
Yet it is difficult to deny that a complete surrender to relativistic
positions tends to become self-contradictory, being far easier to apply
to other people's positions than to one's own. Paradoxically, it can
also become a kind of soft option: if all statements are really on the
same level, what matters is presentation, display, command over
up-to-date style, not the toil of hard research or genuine auto-critique.
Charles Taylor has argued recently that extreme subjectivism ignores
the fundamentally dialogic nature of human life, language and
knowledge, its development, always, through interaction and exchange.95 A dialogical imagination, further, need not necessarily
abstract from power relations, though that has happened at times
in some readings of Mikhail Bakhtin. What it necessarily emphasizes
are the non-monologic, social, conditions of production of consciousness. The effort to develop a social history of historical awareness acquires, then, an added significance. It can point towards ways
of recognition of the reflexive turn that do not have to succumb to
complete subjectivism.
more India-rooted, post-1950s Left-nationalist historiography beyond

'higher' academic circles. The spread-effects of History Congress
sessions, the possibly more effective state-level conferences conducted
through regional languages, sporadic translation efforts, and occasional refresher courses, remain fairly limited, and the possibilities
of democratic dialogue often get further restricted, even within these
limits, by the prevalence of hierarchized structures and attitudes.
And it is surely symptomatic that the high degree of interest in
Western Marxist and radical historiographies has never been extended to include efforts to learn from 'history workshop* experiments. In Britain, Germany and some other Western countries, these
have sought to go beyond the academic guild through extra-mural
adult and workers' education initiatives. They have encouraged
workers and other ordinary folk to write or speak about their
experiences and memories, and tried to form groups of local "barefoot
historians'.96 In India, however, with the important and honourable
exception of gender studies, which has offered considerable opportunities at times for fruitful interaction between activists and academics, research and teaching tend to remain highly hierarchized even
among Left intellectuals.
I have argued implicitly throughout this essay that an exploration

of the social conditions of production of history cannot afford to
remain a merely intellectual project. It needs to become part of wider
and far more difficult efforts to change these conditions. The paradox
of postcolonial front-ranking historiography has been that the affirmation of socially radical values and approaches (unimaginable for
old masters like Jadunath Sarkar or R.C. Majumdar, for instance)
has been accompanied by more, rather than less, elitism in structures
of historical production and dissemination. Late Subaltern Studies, as
the first Indian historiographical trend to achieve an international
prestige largely prior to, and in excess of, its reputation within India,
is peculiarly open to a critique in terms of its 'politics of location*.
But of course elitism operates within academic structures inside the
country too, and at many different levels. Residence, or even language
writing in Hindi or other indigenous languages rather than in
English will not automatically eliminate hierarchies.
The marginalization of the JNU historians' manifesto was a
reminder that there has been relatively little sustained or effective
attempt to spread the methods, findings, and values of even the
The contrasting experiences of two efforts at preparing school

textbooks can serve in conclusion as indicators of problems and
possibilities. In the schools where they have been in use, the National
Council of Education, Research and Training (NCERT) textbooks
commissioned in the mid-1970s from front-ranking (and mostly
Delhi-based) historians have certainly helped to eliminate the blatant
communal bias at the level of prescribed texts (through not necessarily from actual teaching),97 and outdated histories have been
displaced to some extent by the findings and approaches of postindependence research. But their impact has been reduced by over-
95 Charles Taylor, The Politics of Recognition1.
96
47
It is seldom remembered that E.P. Thompson's Making of the English

Working Class originated as lectures to adult education classes for Yorkshire
workers. T h e History Workshop Journal contains abundant information about
extra-guild initiatives, emerging from British New Left and socialist-feminist
movements. For the less known b u t important West German developments in
the 1980s, see Alf Ludtke's Introduction to Ludtke (ed.), The History of Everyday
Life (Frankfurt, 1989; Eng. trans., Princeton, 1995), and Geoff Eley, 'Labour
History, Social History, Alltagsgescichte: Experience, Culture and the Politics
of the Everyday A New Direction for German Social History?', Journal of
Modern History, 61, June 1989.
97
RSS-run schools in Delhi often use N C E R T textbooks, n o doubt interpreting them in their own way.
49
48
burdened syllabi, bureaucratic management, and a concentration on

providing 'correct* factual information and interpretation rather
than imaginative pedagogical presentations. The texts were written
by university scholars with little possibility of contact with secondary
education: inputs through discussions with schoolteachers, difficult
to organize for such a centralized, Delhi-based project, seem to have
been minimal.
A decade or so later, the Eklavya volunteer group was able to
work out much more interesting and innovative history texts and
teaching methods through sustained grassroots work in the not
particularly propitious atmosphere of Hoshangabad's small town
and village schools in Madhya Pradesh. There were consultations
with metropolitan historians (the initiators of the history textbooks
project were themselves JNU graduates), but also repeated rounds of
discussions with local schoolteachers. Eklavya history texts contain
less factual detail than the NCERT books: combined with constant
attention to teaching methods, they do seem geared towards much
more classroom discussion and creative assimilation.98 Eklavya recently organized a teaching-cum-research seminar of Madhya Pradesh
college teachers, and has had plans for collecting historical material
through its far-flung and socially diverse local contacts. Money of
course remains a very major constraint, for Eklavya, unlike most
NGOs today, has so far kept away from all international funding
agencies.
Such experiences are a reminder that what is needed is not just
more effective channels of communication through which high
academic wisdom can be disseminated downwards, but efforts to
democratize also the production of historical knowledge, to work
towards a new kind of historical culture. There is a need to pioneer
ways of developing interaction among researchers, teachers, and
activists drawn from, or working among, diverse social strata. On
98
The ancient Indian history textbook, to take a specific example, includes

a number of stories, some taken from the Jatakas and other texts, others
invented. A story set in a hunting-foodgathering community is followed by
questions as to what its members would do if hunters fail to find game. Initial
responses, I was told, often suggest going to the market, or, more commonly,
borrowing from the mahajan. Further classroom discussions can then highlight
what is and is not possible in a particular historical situation, thus introducing
basic notions about the logic of a social formation far more imaginatively and
effectively than any formal definition. Incidentally, I recall being amazed by
the level of animated discussion in an Eklavya village class admittedly in
one of their best schools.
a long-term scale, collaborative research works and textbooks could

emerge, enriched by multiple social and pedagogical experiences,
and based on a mutual reformulation of perspectives.
I know this will sound hopelessly Utopian, and particularly so
because any suggestion for moving beyond the professional guild
tends to get equated with some form of 'going to the people*, which
is then dismissed as unrealistic for the vast majority of academics.
What I am suggesting as beginnings are far more modest things.
There seems no reason, for instance, why participants at the many
seminars that are constantly being held in cities like Delhi should
not include at least some schoolteachers. The hierarchical divisions
between scholars at research institutes, university teachers, and those
working in undergraduate colleges are visibly deepening: surely
something should be done to reduce these barriers. One way could
be informal discussion groups inevitably middle class, perhaps,
but still including people other than academics: for history, surely,
is a subject in which intelligent interest does not demand great
professional knowledge. I can recall some groups like these, one of
them consisting of trade union activists, as well as dedicated efforts
to bring out a historical journal in the vernacular, from the Calcutta
of my youth, and no doubt there have been many such instances
elsewhere. Small beginnings: but surely we can agree that the many
worlds of Indian history must not be allowed to fly totally apart, as
the social base of producers and intended audiences of front-ranking
South Asian scholarship narrows, even while reaching out towards
global horizons.

Sumit Sarkar

Uploaded by

Copyright:

Available Formats

Sumit Sarkar

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Sumit Sarkar

Uploaded by

Copyright:

Available Formats

Writing social history / Sumit

Sarkar; New Delhi: Oxford

ntrospection about their own location in society has not been

WRITING SOCIAL HISTORY

ways, both these sets of developments have in effect projected views

THE MANY WORLDS OF INDIAN HISTORY

production and dissemination of historical awareness in today's

WHITING SOCIAL HISTORY

THE MANY WORLDS OF INDIAN HISTORY

'derivative discourses'. A total rupture then has to be presumed

WRITING SOCIAL HISTORY

THE MANY WORLDS OF INDIAN HISTORY

suitable material for a colonial discourse approach, I hope to

Ranajit Guha, An Indian Historiography of India: A Nineteenth-Century Agenda

(Chronicle of Kings, Serampur, 1808). Its author was an orthodox

WRITING SOCIAL HISTORY

historians for his unusually critical treatment of sources and striving

THE MANY WORLDS OF INDIAN HISTORY

WRITING SOCIAL HISTORY

Orthodox Brahmanical denunciations of Kalavada often associated

THE MANY WORLDS OF INDIAN HISTORY

The relevant points, in considering the importance of these varied

Pre-colonial histories, Ranajit Guha has argued, consequently tended to

WRTTING SOCIAL HISTORY

THE MANY WORLDS OF INDIAN HISTORY

not encourage, on the whole, any total view of history as diachronic

provoked a search for sustenance in past glories, real or imagined,

Notions of time, now assumed to be linear and abstract (i.e. no

The place of history in the new educational system thus needs

WRITING SOCIAL HISTORY

THE MANY WORLDS OF INDIAN HISTORY

so often made that the culture imposed through Western education

Historical Anecdota took its place next to Aesop's Fabks as reading

lessons 'illustrative of justice, fidelity, probity and humanity*. To

The myth of Kahyuga vanished quickly from formal Indian

A similar continuity is noticeable in^he focus, i n 'renaissance' and 'Saidian'

WRITING SOCIAL HISTORY

THE MANY WORLDS OF INDIAN HISTORY

Indian history provides at first sight a particularly telling instance

This collective self, however, was for Bankim almost invariably

Indira Chowdhuiy Sengupta, 'Colonialism and Cultural Identity: T h e

WRITING SOCIAL HISTORY

of the kind of inquiry that has remained foreclosed,32 both in the

THE MANY WORLDS OF INDIAN HISTORY

between the easy acceptance of some key aspects of Anglo-Indian

WRITING SOCIAL HISTORY

THE MANY WORLDS OF INDIAN HISTORY

framework of ancient Hindu glory/medieval Muslim tyranny and

in large part because 'like other modern ideologies, it allows for a

Chatterjee, Nation, pp. 91, 110.

WRITING SOCIAL HISTORY

seminal recent analysis by Charles Taylor. The specific contexts

THE MANY WORLDS OF INDIAN HISTORY

law alike, 'colonial knowledge', it needs to be added, was not just a

WRITING SOCIAL HISTORY

was a fairly remarkable and precocious interest in social and cultural

It must be added immediately, however, that Niharranjan Ray

WRITING SOCIAL HISTORY

and these essays provide rich indications of a mind grappling with

THE MANY WORLDS OF INDIAN HISTORY