Indian Economic & Social

History Review http://ier.sagepub.com

Caste as Maratha: Social categories, colonial policy and identity in early

twentieth-century Maharashtra
Prachi Deshpande
Indian Economic Social History Review 2004; 41; 7
DOI: 10.1177/001946460404100102

The online version of this article can be found at:

http://ier.sagepub.com/cgi/content/abstract/41/1/7

Published by:

http://www.sagepublications.com

Additional services and information for Indian Economic & Social History Review can be found at:

Email Alerts: http://ier.sagepub.com/cgi/alerts

Subscriptions: http://ier.sagepub.com/subscriptions

Reprints: http://www.sagepub.com/journalsReprints.nav

Permissions: http://www.sagepub.com/journalsPermissions.nav

Downloaded from http://ier.sagepub.com at CALIFORNIA DIGITAL LIBRARY on September 19, 2007

© 2004 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
Caste as Maratha: Social categories,
colonial policy and identity in early
twentieth-century Maharashtra
Prachi Deshpande
Colorado State University

In the light of recent scholarship emphasising the historicity of caste, this article tracks the
transformation of the category ’Maratha’ from its precolonial register as a military ethos to
that ofa caste in the early twentieth century. Surveying the category’s genealogy in non-Brahman
literature and colonial ethnographic writings and policy, it argues that this caste-based register
of ’Maratha’ was shaped through a complex, interactive process by both colonial and Indian
discourses. In doing so, the article attempts to historicise ’Maratha’ and emphasises the impor-
tance of locating the modern history of caste and its encounter with colonialism in regional/
local contexts.

One of the striking features of the colonial encounter in western India and the
transformation in vocabularies of community and political identity was a change
in the understanding and usage of the category ’Maratha’. The term recalls a
precoi_onial warrior heritage, embodied most strikingly in the figure of the eques-
trian--and now ubiquitous-Shiva~i, and continues to signal, in popular parlance
as well as scholarly literature, the historical polity that resisted Mughal expansion
into the Deccan in the seventeenth and eighteenth centuries. Its more dominant
usage in modem Maharashtra, especially in the twentieth century, however, has
been as the marker not of an entire polity, but of a specific social group: the Marathas
(often referred to as ’Maratha-Kunbis’ as well), who are today the politically domi-
nant, upper-caste group in the state. This dominance of the Maratha caste in
postcolonial Maharashtra and its expression in terms of land control, political alli-
ances and rural networks of power has been well documented by political scien-
tists.’ Although detailed historical studies of the non-Brahman movement have
See, for instance. Anthony T. Carter, Elite Politics in Rural India: Political Stratification and
1
Political Alliances in Western Maharashtra. London, New York, 1974; V. M. Sirsikar, The Rural Elite
in a Developing Society: A Study in Political Sociology, New Delhi, 1970; Jayant Lele, Elite Pluralism
and Class Rule: Political Development in Maarashtra, India, Toronto, Buffalo, 1981; Mary C. Carras.
The Dynamics of Indian Political Factions: A Study of District Councils in the State of Maharashtra.

Downloaded from http://ier.sagepub.com at CALIFORNIA DIGITAL LIBRARY on September 19, 2007

© 2004 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
 8/

pointed to multiple imaginings of what and whom ’Maratha’ represented over the
colonial period, scholarship on the postcolonial period has often tended to project
the category’s current avatar unproblematically into the colonial and precolonial
past. The historiographical implications of the transformation of a broad, histori-
cal category to a narrow, specific caste group have attracted less attention.
The changing meanings of ’Maratha’, however, may be seen as an example of
the historicity of social categories, especially caste categories, in modem South
Asia. Recent interventions in the study of caste, despite various ideological differ-
ences of emphasis, have highlighted and documented its historicity and the impact
of the colonial encounter in producing the practice and politics of caste identities
as we know them today.’ In particular, scholars have shown an increasing interest
in exploring the influence of colonial enumeration and classification practices from
the later nineteenth century and the colonial representative framework (which re-
lied heavily on such practices) on caste politics and identity. Perhaps the most
valuable outcome of this shift from the ’immutability’ of caste to its ‘modernity’ is
the rejection of an over-generalised, uniform approach to caste based on its nor-
mative aspects and the acknowledgement of messy contradictions and geographi-
cal variations in the development and practice of caste identities.
In light of these recent interventions, this article tracks the transformation of
the category ’Maratha’ from its dominant precolonial register as a historical, mili-
tary ethos to the bounded marker of a caste group. The principal focus here is on
-the discursive contestations that marked the content and meaning of this category
in the early twentieth century and its growing importance in structures of colonial
policy. The central argument the article makes is that the caste-based register of
’Maratha’ that came to dominate by the late colonial period was shaped through a
complex, interactive process both by colonial policies of classification and repre-
sentation, as well as Maharashtrian attempts to engage with new vocabularies of

identity. It not only surveys the many changes the category underwent in the dis-
course of the non-Brahman movement, but also tracks the category’s genealogy

through a series of colonial ethnographic writings and official policy, thus point-
ing to the interpenetration of both discourses. In doing so, the article attempts to
historicise the category Maratha and emphasises more broadly the importance of
locating the modem history of caste and its encounter with colonialism in regional/
local contexts.

Cambridge Eng., 1972; Livi Rodrigues, Rural Political Protest in Western India, Delhi, New York,
1998. Rodrigues’ work, in particular, focuses on the late nineteenth and early twentieth centuries, but
treats the ’Maratha-Kunbi’ category as practically fixed.
2
The most important works in this regard are Bernard Cohn, An Anthropologist among the Historians
and Other Essays, Delhi,1987; Susan Bayly, Caste, Society and Politics in Indiafrom the Eighteenth
Century to the Modern Age, New York, 1999; and Nicholas Dirks, Castes of Mind: Colonialism and the
Making of Modern India, Princeton, 2001.

Downloaded from http://ier.sagepub.com at CALIFORNIA DIGITAL LIBRARY on September 19, 2007

© 2004 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
 9

’Maratha’ in Precolonial and Early Colonial Contexts

Details of the origins of the term ’Maratha’ are still relatively unknown, but it has
been argued that these lie in the long period of Musfim rule in the Deccan between
the fourteenth and seventeenth centuries, especially in the states of Ahmednagar
and Bijapur. ’Marathas’ were initially Marathi-speaking units in the armies of these
states and gradually came to identify, by Shivaji’s time in the seventeenth century,
the many local lineages and elites who had found avenues for social mobility
through civil and military employment in these states. Many of these chieftains,
who claimed Rajput ancestry and descent from a set of elite 96 Kshatriya families,
called themselves ’Marathas’, but the vast numbers of Kunbi cultivators of west-
ern India who served under them also belonged to ’Maratha’ armies.3

Early British commentators such as James Grant Duff or John Malcolm used
the term ’Mahratta’ to encompass the entire polity that held sway over western
and central India in the eighteenth century.’ Thomas Broughton’s entertaining, if
somewhat acerbic, 1813 L~tters from a Mahratta Camp described at length the
’Mahratta legend of fear’ and repeatedly referred to the Marathas as a ’race’ and
’tribe’ full of rather regrettable military practices and values:
they were, according
to him, ’deceitful, treacherous, narrow-minded, rapacious [sic] and notorious li-
ars’.5 He understood them as being Hindus and found their riotous participation in
Mohurrum ceremonies ’curious’, but at another point also described Baboo Khan,
a Muslim, as ’a Mahratta chief of some rank and consideration’.6 Broughton did

recognise a general hierarchy among the Marathas, describing ’two grand classes’
of Brahmans and ’all the inferior castes of the Hindoos, but composed chiefly of
Aheers or shepherds, and Koormees or tillers of the earth .... The various castes of
the second class are freer from religious prejudice, as to eating, than say any other
Hindoos’.’
Richard Jenkins, Resident at the court ofl~lagpur and author ofthe 1827 Report
on the Territories of the Raja of Nagpore, however, was more aware that ‘Maratha’
itself might be flexibly applied:

The term Mahratta, though applied by the other tribes to the inhabitants of
Maharashtra in general, seems among the Mahrattas themselves to be limited
to a few distinct classes only. The Jhari and Mahratta Kunbis are considered the
genuine Mahrattas by all the other classes: besides these the term is more par-
ticularly applied to the numerous tribes and families from whom the most

3 Stewart Gordon, The Marathas, 1600-1818. Cambridge/New York, 1993, pp. 182-208.
J.P. Guha, ed., James Grant Duff, History of the Mahrattas, Two Volume Edition, New Delhi.
4
1971[1826]; Sir John Malcolm, A Memoir of Central India, London, 1823.
5
Thomas Duer Broughton, Letters Written in a Mahratta Camp During the Year 1809, Descriptive
of the Characters, Manners, Domestic Habits, and Religious Ceremonies, of the Mahrattas, London.
1977[1813], p.72.
, p. 50.
Ibid.
6
., p. 71. Emphasis in original.
Ibid
7

Downloaded from http://ier.sagepub.com at CALIFORNIA DIGITAL LIBRARY on September 19, 2007

© 2004 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
 10/

celebrated Mahratta leaders have sprung. The number of these families is ...

ninety-six.’
Jenkins’ comment aside, we do not have substantial contemporary evidence to
indicate just how central this social category was, the different spheres in which it
might have been most strongly invoked, or the degree to which it corresponded
with jati divisions in the precolonial period. We do have some evidence that it was
relatively flexible and open to appropriation by humbler, but enterprising families
through military service, marital alliance opportunities and negotiations with chiefs
and rulers.’ The most celebrated example of the Maratha claim to Kshatriya status
was, of course, Shivaji himself, whose Vedic coronation in 1674 took place in the
face of loeal Brahman protests about his uncertain jati origins. Sumit Guha’s richly
detailed and thorough discussion of the opportunities for upward mobility (in-
cluding, in some cases, Rajput status) afforded to groups such as the Kolis and
Mavlis through military service and engagement with successive regimes in West-
ern India attests to the fact that this was a widespread phenomenon.’° Recently,

Philip Constable has also shown how Mahar soldiers participated in the precolonial
military labour market through this open-ended, inclusive Maratha category signi-
fying military naukari.&dquo;
Some eighteenth-century sources, however, suggest that this register of ’genu-
ine’ or ’most celebrated Mahrattas’, with its attendant Kshatriya and Rajput ances-
try claims might well have been part of the broader military Maratha ethos itself.
Numerous Marathi bakhars (chronicles) narrating important battles and family
sagas were composed in this period; elsewhere I have argued that these texts not
only commemorate important Maratha battles and warriors, but also articulate a code
of honour and military-cultural values specific to Maratha warriors, often through
an admixture of defiance and admiration for Rajput fighting skills and valour.&dquo;

8
Richard Jenkins, Report on the Territories of the Raja of Nagpore, Nagpur, 1 923, p. 19. I owe this
reference to Sumit Guha.
9
Rosalind O’Hanlon, Caste, Conflict, and Ideology: Mahatma Jotirao Phule and Low Caste Protest
in Nineteenth-Century Western India, Cambridge/New York, 1985, pp. 15-23. An oft-quoted Marathi
proverb captures this feature of upward mobil ity in rural society perfectly: kunbi mazla, maratha zhala,
meaning, ’When a Kunbi prospers. he becomes a Maratha’. The anthropologist Iravati Karve’s
characterisation of the Maratha-Kunbi caste groups as a ’cluster’ of castes attempts to indicate precisely
this mobility between groups that had different ritual and social practices, but displayed nevertheless a
certain unity of customs, ties and linkages. See Iravati Karve, Hindu Society: An Interpretation, Pune,
1968.
10 Sumit Guha, Environment and Ethnicity in India, 1200-1991, New York, 1999, pp. 85-107.
11
Constable argues that this inclusiveness was gradually eroded in favour of an increasing Kshatriya
exclusivity over the eighteenth century, especially under the Brahmanical policies of the Peshwas, but
unfortunately fails to discuss contemporary sources that demonstrate such a turn towards Kshatriya
jati
exclusivity. Philip Constable, ’The Marginalization of a Dalit Martial Race in Late Nineteenth and
Early Twentieth Century Western India’, Journal of Asian Studies, Vol. 60(2), 2001, pp. 439-78.
12
Prachi Deshpande, ’Narratives of Pride: History and Regional Identity in Maharashtra, c187-
1960’. unpublished Ph.D Dissertation, Tufts University, 2002, Ch. 2.

Downloaded from http://ier.sagepub.com at CALIFORNIA DIGITAL LIBRARY on September 19, 2007

© 2004 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
 11

’Rajput’ and ’Maratha’ in these narratives certainly appear as elite categories, but
not as specific jati groups; instead, they are attributional terms embodying specific

military values, with the former frequently serving as a category for emulation. As
several works on the military labour market in medieval and early modem India
have argued, Rajput/Kshatriya connections looked good on military resumds, some-
thing the Marathas were not unaware of;&dquo; the repeated references in the bakhar
narratives to correct behaviour, prestige, valour and Kshatriya dharma also under-
scored the fighting qualities and a code of honour for the top brass among the
Maratha chiefs to celebrate and emulate, but not specifically a Maratha Kshatriya
jati. For example, in the famous Bhausahebanchi bakhar, one of the most riveting
late eighteenth-century accounts of events leading up to the Maratha defeat at
Panipat in 1761, the chieftain Jayappa Shinde describes a skirmish between his
forces and those of the Rajput chief Bijesing:

These are Marwadi Rajputs, incredibly valorous; their bodies dance around
even if they are beheaded Ihey also have a lot of firepower. Our people are
...

faint of heart to begin with, with steel weapons, tied to a tree they will uproot it
to try and flee .... [The] courage [of the Marwari Hara Rajputs] was not sur-
prising. But the Marathas did put up a brave show .... many Marathas were
killed, but even so, they must be feel that blessed were the Rajput mothers that
bore such sons.&dquo;

Of course, this military, flexible register of ’Maratha’ sits uneasily with (he
increased Brahmanisation of the Maratha state under the Peshwas in the eigh-
teenth century. As is well known, the Peshwas vigorously sought to enforce jati
boundaries and rules of jati discipline, especially relating to intermarriage,
interdining and, most importantly, varna status claims.&dquo; Condemning the Peshwas’
relegation of all non-Brahman groups, from high-ranking officials to ordinary
people, to a low Shudra ritual status was also a prominent feature of non-Brahman
polemic in the colonial period. Kshatriya claims by Maratha families after the
Peshwa debacle, thus, are often seen as proof of this policy during Peshwa rule
itself, but we really know very little about how this Brahmanisation impacted the
’Maratha’ category itself. It is remarkable that secondary works on the subject
point to an overwhelming number of cases involving numerous Brahman jatis and
The most widely cited work is D.H.A. Kolff,
13 Naukar. Rajput and Sepoy: The Ethnohistory of the
Military Labour Market in Hindustan, 1450-1850, Cambridgeshire, 1990, but see also Stewart Gordon.
’Zones of Military Entrepreneurship in India, 1500-1700’, in Stewart Gordon, Marathas, Marauders
and State Formation in Eighteenth Century India, New York, 1994, pp. 182-207; Sumit Guha,
Environment and Ethnicity in India, pp. 83-107, and Philip Constable, ’The Marginalization ofa Dalit
Race’.
14
M.S. Kanade, ed., Bhausahebanchi Bakhar, Pune 1975, pp. 84-91. Also, see an excellent English
translation of this bakhar in Ian Raeside, Marathi Historical Papers and Chronicles: The Decade of
Panipat (1751-61), Bombay 1984, pp. 1-101.
15 The most oft-cited work in this regard is Hiroshi Fukazawa, ’State and Caste System (Jati) in the
Eighteenth Century Maratha Kingdom’, Hilotsubashi Journal of Economics. Vol. 9(1), 1968, pp. 32-44.

Downloaded from http://ier.sagepub.com at CALIFORNIA DIGITAL LIBRARY on September 19, 2007

© 2004 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
12

ritual rules interaction and hierarchy, and an urgency to

governing their mutual
prevent groups such as the Prabhus and the Daivadnya Sonars (a goldsmith caste)
from claiming higher varna status (Kayastha and Brahman respectively) and Vedic
ritual. ‘6 The one Maratha/Kshatriya case that appears is the successful insistence
of Jijabai, queen ofKolhapur from the Shankaracharya of Karveer, on Vedic death
rites for her husband Sambhaji, much to the Shankaracharya’s irritation.&dquo;
Examining the Peshwas’ jati policy in detail is beyond the scope of this article,
but it is worth noting the ways in which further research could qualify the rather
general picture of this Brahmanisation we currently have, and reveal a more subtle
understanding of the ’Maratha’ label under the Peshwas. As Susan Bayly has ar-
gued, Brahmanisation across various successor states in the eighteenth century
did not just mean a greater number of Brahmans in the administration, but also an
urge and ability among various scribal and commercial groups to exploit opportu-
nities in competing states through varna claims to Brahman, Kayastha or Vaishya
status.18 Viewed in the context of the Peshwas’ penchant for their own Chitpavan
jati-fellows in political appointments as well as commercial linkages, it is worth
exploring in greater detail whether such trends encouraged the regime to apply jati
laws as much towards monitoring Brahman groups and preventing claims to simi-
lar qualities and qualifications by scribal competitors such as the Prabhus, as to-
wards keeping lower castes and those considered untouchable in their place in the
varna hierarchy. Moreover, it is crucial to investigate further the success of these

attempts in actually enforcing jati discipline and vama claims, and also the degree
to which these were able to control both influential Maratha chiefs as well as
ordinary Maratha or Kunbi soliders and peasants. It is arguable that while such
rules greatly increased the monitoring and exploitation of ordinary lower-caste
folk, those considered untouchable, as well as the Brahman groups (especially
Brahman women, as Uma Chakravarti has shown),’9 the military context and
opportunities across the Maratha dominions in western and central India, espe-
cially outside the Peshwa’s direct control, made the enforcing of Shudra status for
Marathas, particularly the more influential ones, more difficult, thus keeping it
open to appropriation and inclusion. A fascinating observation by Grant Duff in
the opening pages of his work hints at how Peshwa concerns over jati might have
had to do much more with emphasising Brahman exclusivity within the broader
’Maratha’ military fold than demarcating a Maratha Shudra jati:

... the Mahratta is applicable in some degree to all [the inhabitants of

name

Maharashtra], when spoken of in contradistinction to men of other countries,

16
See Fukazawa, ’State and Caste System’, p. 42; P.E. Gavali, Society and Social Disabilities
under the Peshwas, Delhi. 1988, pp. 105-13; Sudha Desai, Social Life in Maharashtra under the
Peshivas, Bombay, 1980, pp. 30-42.
17
P.E. Gavali, Society and Social Disabilities. p. 110.
18
Bayly, Caste, Society and Politics in India, p. 70.
19
Uma Chakravarti, ’Gender, Caste and the State in Eighteenth Century Maharashtra’, in Uma
Chakravarti, Rewriting History: The Life and Times of Pandita Ramabai. New Delhi, 1998, pp. 3-42.

Downloaded from http://ier.sagepub.com at CALIFORNIA DIGITAL LIBRARY on September 19, 2007

© 2004 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
 13

but amongst themselves a Mahratta Bramin will carefully distinguish himself

from a Mahratta. That term, though extended to the Koonbees, or cultivators,
is, in strictness, confined to the military families of the country, many of whom
claim a doubtful, but not improbable descent from the Rajpoots.20

To be a ’Maratha’ in the precolonial was not to be part of an

period, then,
enumerable and bounded jati; depending the context, the category could sig-
on

nify certain military values, the fierce armies that struck terror in people’s hearts,
a political force in the subcontinent, or the elite of a broad military-cultural group.

Broughton’s account of the Maratha chiefs’ celebrations and activities in the camps,
Mohurrum and Holi, also suggests that expressing Hindu religious and caste dif-
ference in daily life was not central to being a Maratha, especially in a military
environment.
In 1818, the British installed Pratapsinh, a descendant of Shivaji, as the nomi-
nal ruler at Satara to offset the recently deposed Peshwa and Brahman power in
Pune. Company rule drastically reduced the military avenues for social mobility
within western Indian society, and brought the Peshwai’s attempts at policing jati
discipline to an end. Both developments were to have profound consequences on
the composition and understanding of the category ’Maratha’. The first couple of
decades witnessed many such claims from various groups to higher varna status,
both Kshatriya and Brahman, with rearguard action from Brahmans in Pune.1’
The most famous ofthese, of course, was Pratapsinh’s successful use of the changed
power configurations to claim Kshatriya and Vedic ritual status for his family, the
Bhosales, and those of other Maratha chiefs in 1830, following a decade-long
conflict with Brahman opinion in Pune.
As Rosalind O’Hanlon has argued, the public debate that finally secured
Kshatriya status for Pratapsinh, however, brought to the fore and legitimised as
acceptable ritual and dining practices that were rather loosely defined and widely
practised in rural society (two of these mentioned are meat-eating and eating out
of a common plate); these criteria thus enabled not just influential landed chiefs
but also many modest Kunbi families to put forward Kshatriya claims, despite
Pratapsinh’s attempts to limit them to a small, elite circle.22 From the mid-
nineteenth century, contemporary Marathi observers commented on the increased
tendency among upwardly-mobile Kunbi groups, some newly urbanised, but also
those benefiting from the recent commercialisation of agriculture, to take up the
sacred thread and the appellation of ’Maratha’ .23 As we shall see below, colonial
officials also began recording these ongoing changes from the 1870s onwards. It

20
Duff, History of the Mahrattas, p. 8.
21
See N.K. Wagle, ’A Dispute between the Pancal Devajna Sonars and the Brahmanas of Pune
Regarding Social Rank and Ritual Privileges: A Case of the British Administration of Jati Laws in
Maharashtra, 1822-1825’, in N.K. Wagle, ed., Images of Maharashtra: A Regional Profile of India.
London. 1980, pp. 129-59.
22
O’Hanlon. Caste. Conflict and Ideology, pp. 24-49’.
23
Ibid., pp. 41-45.

Downloaded from http://ier.sagepub.com at CALIFORNIA DIGITAL LIBRARY on September 19, 2007

© 2004 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
14 /

was, thus, in this assertion of higher jati

status and ritual claims that the early
colonial period witnessed attempts at social mobility; combined with the decline
of military opportunities, these activities were significant in shifting the dominant
martial register of’Maratha’ to that of a more bounded and exclusive community
over the nineteenth century.

Early Non-Brahman Protest

The rise of low-caste protest against Brahman dominance in the later nineteenth
century gave these activities a sharp political twist. The overwhelming dominance
of Brahman groups in the new colonial order in Maharashtra and the
preponderance of Brahmans in the nationalist middle classes have been well docu-
mented.24 Another striking feature of the colonial encounter in this region was the
strident presence in political and social discourses of narratives from the past,
from the period that came to be known as ’Maratha history’. Themes and symbols
from this past served as a prime cultural resource for different social groups to not
only express both identity and difference, but to also imagine a modern,
Maharashtrian regional identity.25 B.G. Tilak’s well-known invocation of Shivaji
as a nationalist hero in the 1890s was one among many such uses of this history
made by Brahman nationalists; in this narrative, the Maratha conflict with the
Mughals and others was a patriotic one where all Marathi-speaking social groups
worked together as Marathas. Despite disagreements within the broad Brahman
nationalist position, these historical invocations were shot through with the idea of
a natural caste hierarchy that placed Brahmans at the helm: the Brahman Peshwa
served as the perfect example for the natural social leadership of Brahmans in
Maharashtrian society. 21
The low-caste critique spearheaded by activists like Jotirao Phule focused, of
course, on the overwhelming presence of Brahmans in every walk of life and the
re-inscription of Brahman social and ritual power under the new colonial order
through privileged access to western education and employment in the colonial
government. Writers and activists from various non-Brahman groups, however,
also invoked the Maratha past in their protests against Brahman dominance, laying
bare the tacit assumptions of Brahman leadership in many nationalist narratives.

24
Gordon Johnson, ’Chitpavan Brahmans and Politics in Western India’, in E.R. Leach and S.N.
Mukherjee. eds, Elites in South Asia, Cambridge, 1970; Gail Omvedt, Cultural Revolt in a Colonial
Society: The Non Brahman Movement in Western India, 1873 to 1930, Bombay, 1976, pp. 76-78;
Veena Naregal, Language Politics, Elites, and the Public Sphere. New Delhi, 2001; Uma Chakravarti.
Rewriting History, pp. 43-106.
25
Deshpande, Narratives of Pride’.
26This claim of Brahman indispensability to the Maratha project was demonstrated most clearly in
the elevation of the Brahman poet and contemporary of Shivaji, Samarth Ramdas, as the chief moral
force and spiritual guide behind the establishment ofMaratha freedom in the writings of many Brahman
nationalists. Pamphlets, novels and poetry on Ramdas as Shivaji’s guru tlooded the printed sphere in
the early twentieth century. and were certainly a response to non-Brahman critiques of Brahman narratives
on the Maratha past. Ibid., Ch. 4.

Downloaded from http://ier.sagepub.com at CALIFORNIA DIGITAL LIBRARY on September 19, 2007

© 2004 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
 15

They sought to root the political position of the ’non-Brahman’ in regional history
and culture, and put forward their own versions of Maharashtra’s history and tra-
ditions. Indoing so they made the Maratha past a prime site for the articulation of
caste conflict and identity. Shivaji’s own conflict with local Brahmans in the sev-
enteenth century over his right to a Vedic coronation gave this protest a potent
symbolic resource. In particular, the Peshwai’s attempts at enforcing jati differ-
ence came to neatly represent the worst of Brahman dominance. As O’Hanlon has
shown, the attempts to give the category ’Maratha’ a new meaning were central to
these processes. Non-Brahman activists disagreed sharply among themselves over
the content and meaning of the category, but were successful in constructing it as
an explicitly political expression of non-Brahman protest and a social category
that specifically excluded Brahmans. James Grant Duff, for instance, came under
criticism for giving his monumental historical work the misleading title History of
the Mahrattas, when it contained information about many groups like Brahmans
who were not really Marathas.2’
One of the earliest of such attempts to exclude Brahmans from the ’Marathas’
category was by Narayan Meghaji Lokhande.’-8 In an article titled Are the Brahmans
Marathas?’ in the Din Bandhu of 17 January 1886, Lokhande criticised the Gover-
nor of Bombay, Lord Harris, for using ’Maratha’ to denote all Marathi-speakers.’-9

Maratha means those of the Kshatriya varna. In this Kshatriya varna, there are
ninety six families, and many sub-families within these. The people who were
bom into these families are the true Marathas (Kshatriyas). Those who hold
surnames from among these families can become Marathas; other people can
never do so If, in this country of Maharashtra, the Brahmans can become
....

Marathas, then even the Muslims and other people could call themselves Marathas.
There is not to be found amongst those who call themselves Brahmans the
similarity in manners and customs, deities and religion, and in families and
lineage, which there is amongst all the Maratha people .... We can never ever
allow the Brahmans to take the liberty of calling themselves Marathas.30

As is apparent from the above quotation, he acknowledged the elite nature of the
Maratha category with its ninety-six families, but allowed for its extension to in-
clude families of other castes who had the same surnames and could, thereby,
’become’ Marathas. The fact that many such surnames were common across rural
caste groups made this a significant extension. He attempted, like Phule, to yoke

27
O’Hanlon, Caste, Conflict and Ideology, pp. 290-302.
28
Lokhande (1848-1907) was a contemporary of Phule’s, and like him, belonged to the Mali
jati.
He was active in the Satyashodhak Samaj not only as the editor of the Din Bandhu newspaper, but also
as one ofthe earliest organisers of labour in the Bombay cotton mills. See Manohar Kadam. Bharatiya

Kamgar Chalvaliche Janak Narayan Meghaji Lokhande (The Father of India’s Labour Struggle Narayan
Meghaji Lokhande). Pune, 1995.
29
Din Bandhu, 17 Jan. 1886. Report on Native Newspapers, week ending 23 Jan. 1886.
30
Din Bandhu, 21 Oct. 1904, quoted in O’Hanlon. Caste, Conflict and Ideology, p. 291.

Downloaded from http://ier.sagepub.com at CALIFORNIA DIGITAL LIBRARY on September 19, 2007

© 2004 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
16/

the ongoing Sanskritising tendencies within rural society to a radical edge, but it is
important to note that he did not clearly specify that all of rural society could
belong to his Maratha community. Equally importantly, he made it clear that the
Muslims had no place in it. Despite its affirmation of a core Maratha elite of ninety-
six families, however, Lokhande’s understanding of the category Maratha remained
one of the most radical within non-Brahman ideology. In I887, he formed the
Maratha Aikyecchu Sabha (Society for Maratha Unity) to ensure that the demands
of education made for Marathas was suitably broad based.&dquo; Adroitly, he avoided
mentioning specific caste groups and focused instead on wresting the legacy of
Shivaji from Brahmans.
Other non-Brahmans were more explicit than Lokhande. For the Deccan Maratha
Education Association (DMEA), one of the many non-Brahman organisations in
the late nineteenth century, Maratha meant the cluster of elite Marathas and hum-
bler Kunbi families linked through kinship; it excluded other agricultural castes
like the Malis, who were very active in non-Brahman politics. The DMEA sought
to claim the historical heritage of the Maratha struggle for this cluster, arguing that

the Maratha and Kunbi population form the muscle and bone of native society.
Their helplessness and ignorance is a national disgrace This condition of ....

is
things by no means an inevitable evil. At one time, not very distant in the
past, they numbered among them some of the renowned leaders of the
Maharashtrian armies, and many filled its ranks. In fact, these classes were the
mainstay of the Maratha power in its palmy [sic] days.&dquo;
Another organisation catled ’the Society for the Maratha caste for putting for-
ward the Dharma of Kshatriyas and for the raising of funds for that Dharma’ invoked
a military past to claim the category for an even more limited group of families:

The name Maratha has realty only ever been given to those who were Kshatriyas.
All other people were happy to accept the name of their trade as their caste-
name ; but the name of Maratha has come to be given permanently to all those
who have kept their mastery of their own land and who take pride in putting
their lives at stake to protect it. Our habit of using Maratha for our caste name
is really a matter of great joy: it means that our very name proclaims that we are
the people of this land of our birth.&dquo;

O’Hanlon has argued, quite rightly, that it was precisely to avoid such San-
skritising tendencies and the resultant cleavages between various low-caste groups
that the most creative and farsighted of non-Brahman thinkers, Jotirao Phule,

"
Ibid., p. 294.
32
Ibid.. p. 292. This statement was made by a speaker at a meeting of the society, which was
founded in August 1882 in Bombay by Tukaramji Haraji Patil Salunkhe.
33
Ibid., p. 299.

Downloaded from http://ier.sagepub.com at CALIFORNIA DIGITAL LIBRARY on September 19, 2007

© 2004 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
 17

consciously avoided the category ’Maratha’ in his imagination of a rural, non-

Brahman solidarity. Phule used ’Shudra’ as well as ’Kshatriya’ (derived from ksh-
etra or land to denote aboriginal inhabitants of the land) to evoke this solidarity
and a generalised sense of pride and bravery.34 It is important to note, however,
that Phule’s ingenious interpretation of ’Kshatriya’, too, remained iconoclastic,
even within the non-Brahman movement. As we shall see, successive polemicists

preferred and advanced more conventional vama connotations of the term, which
increasingly covered only a small, elite section of non-Brahmans.
’Maratha’ in the Early Twentieth Century
In this period, the non-Brahman movement took the contestation of historical nar-
ratives from the relatively sedate sphere of newspapers into the streets and the
public arenas of the Ganpati and Shivaji festivals. Dressed up like Maratha sol-
diers in the Chhatrapati mela, non-Brahman youth penned ballads and songs that
claimed ’Maratha’ as a source of non-Brahman pride and heaped scorn on the
Peshwas by holding them responsible for losing Maratha sovereignty to the Brit-
ish. Through strong, colourful language, these songs depicted Brahman attempts
to be a part of the Maratha past as illegitimate:

Awaken 0 Marathas, this is a time of freedom, awaken to your glory!

Shiva-ba, who protected our faith is called Shudra by the beggar priests
And yet we stand silent with our heads bowed! ...

Having fanned these flames the priesthood watches the fun

Here is a traitor and you feel nothing?
The beggar priests robbed you of freedom ...
And brought the glory of Satara to dust
Shivaji is our source of joy and spirit
Come and prove to the world your grit
Remember Shiva-ba and embrace your courage
Sing for your freedom, Har Har Mahadev3s
At its peak in the early twentieth century, the non-Brahman movement came
under the patronage of Chhatrapati Shahu, ruler of the princely state of Kolhapur
and descendant of Shivaji, and underwent significant changes from the earlier
days of Phule and Lokhande in both discourse and social participation. Shahu, who
began supporting the cause after a bitter clash in the 1890s with Tilak and conser-
vative Brahman opinion over the issue of Vedic rituals for his family, provided

34
Ibid., pp. 140-63.
35
Keshavrao Jedhe, Chhatrapati Mela Sangraha (Songs ofthe Chhatrapati Mela), Pune, 1928, p. 19.
For a good, general survey of non-Brahman politics and personalities in the early twentieth century, see
Rosalind O’Hanlon, ’Acts of Appropriation: Non-Brahman Radicals and the Congress in Early Twentieth
Century Maharashtra’, in Mike Shepperdson and Colin Simmons, eds, The Indian National Congress
and the Political Economy of India, 1885-1985, Brookfield. 1988, pp. 102-46.

Downloaded from http://ier.sagepub.com at CALIFORNIA DIGITAL LIBRARY on September 19, 2007

© 2004 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
18

much-needed financial as well as symbolic support. In the later years of his reign,
he also campaigned actively for the removal of untouchability, opening hostels
and schools, and providing jobs for non-Brahmans.36
Shahu’s position on caste was ambiguous: he championed an array of non-
Brahman causes which had their defiance of Brahman authority as the common
denominator. His campaign against untouchability and support for Ambedkar, for
instance, was matched with a strident insistence on Vedic Kshatriya rights for the
elite Marathas, including the establishment of a Kshatriya priesthood for Marathas
to do away with Brahmans altogether in ritual life. Despite generous financial
support to the Satyashodhak Samaj, he refused to become a member, choosing the
less radical and Vedas-friendly Arya Samaj instead. ,

Several scholars have noted that in his personal attitudes and approach to caste
divisions, Shahu became increasingly radical with time and often annoyed some
of his close elite Maratha associates.&dquo; Shahu’s leadership did serve to bring diverse
discontents against Brahman authority under one cause, but another broad conse-
quence of his championing of the Vedokta cause and other policies was that many
newly-urbanised and respectable Kunbi families were attracted to the non-Brahman
movement.38 Gail Omvedt is right in arguing that Shahu himself, especially in his
later years, sought a gradualist, liberal middle ground between the conservative
and radical extremes of the non-Brahman movement, but the overall effect of his
policies was to ensure a predominance of elite Marathas, or well-off Kunbis seek-
ing Maratha status, within non-Brahman politics. Benefiting from the cash crop
boom in the early twentieth century, such upwardly mobile agricultural groups
organised under the labei ’Maratha’ in a spate of Maratha caste conferences eagerly

36
For an early account of Kolhapur politics and Shahu’s interaction with the British government in
enforcing policies favourable to non-Brahmans, see Ian Copland, ’The Maharaja of Kolhapur and the
Non-Brahmin Movement, 1902-10’. Modern Asian Studies, Vol. 7(2), 1973, pp. 209-55, but a more
comprehensive study in Engl ish is Gail Omvedt, Cultural Revolt in a Colonial Society, pp. 124-36. An
excellent study in Marathi is Y. D. Phadke, Shahu Chhatrapati ani Lokamanya (Shahu Chhatrapati and
the Lokamanya), Pune, 1986.
37
An official reported in1919, for instance, that his tea party with Mahars in Nashik had ’a disturbing
effect in the neighbourhood and may have caused a rupture among the Marathas of the Maharajah’s
own caste, as the well-to-do Maratha of the rural area does not move easily with the times and cannot be

expected to receive the Mahars with open arms at such short notice’. Maharashtra State Archives
(hereafter MSA) Mumbai, Home (Special) Files, No. 363, 1919, ’Brahmans vs. Non-Brahmans’.
38
In 1894, Brahmans occupied a total of 104 jobs in Shahu’s government, while non-Brahmans
occupied18. No untouchable occupied any government post at this time. By 1922, Brahmans occupied
69 of these jobs, non-Brahmans 168, and one untouchable had been appointed. Similarly, the number
of Brahman students in the Kolhapur schools in 1894 was 2,522, with 8,088 non-Brahman students and
264 untouchables. By 1922, however, while the number of Brahman students had increased only to
2,722, the number of non-Brahman students had shot up to 21,027, and that of untouchable students to
2,162. Although these figures do not give us a caste breakdown of Marathas and others in the non-
Brahman category, it is worth postulating that the proportion of Marathas was higher, because of the
many special educational privileges like hostels, scholarships, etc., that they got, both as part of Shahu’s
promotion and the numerous Maratha caste organisations. Y.D. Phadke, VisavyaShatakatila Maharashtra
(Maharashtra in the Twentieth Century), 5 Vols., Vol.1, Pune, 1989, p. 216.

Downloaded from http://ier.sagepub.com at CALIFORNIA DIGITAL LIBRARY on September 19, 2007

© 2004 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
 19

patronised by Kolhapur and other Maratha princely states like Gwalior and Baroda.
In Vidarbha in particular, the non-Brahman leadership was characterised by the
participation of large landed Deshmukhs, who were economically powerful, but
not generally accepted as being part of the older elite Maratha families entrenched
in the Deccan; many of these strongly supported the elite Kshatriya classification
of Marathas and lent support to such conferences. Several Patil Conferences, bring-
ing together village headmen who were usually Marathas of some standing, were
also held under the non-Brahman umbrella in the i920s.~
Non-Brahmans also entered the formal political arenas of legal councils, coop-
erative credit societies, local boards and municipa! councils in the early twentieth
century. Given the severely restricted property and education franchise for elec-
tions to any government body in both Bombay and the Central Provinces at this
time (roughly only a meager 9 and 8 per cent of the population in the provinces
respectively),40 both candidates and electors were drawn overwhelmingly from
the richer peasantry.41 As Omvedt has shown in considerable detail, this increas-
ingly elite dimension to the caste conferences in this period as well as in rural
institutional power structures, therefore, heralded the dominance of well off,
upwardly-mobile Maratha-Kunbis-increasingly organising only as Marathas and
claiming Kshatriya vama status-in the non-Brahman movement.a2
This changing face of the movement, in its approach to ’Maratha’ but also in
composition, served to blunt the radical edge Phule, Lokhande and Satyashodhak
ideology had given non-Brahman protest, and resulted in an increased ambiva-
lence towards questions of untouchability and lower-caste unity. Contestations
between radical and conservative activists over the definition and appropriation of
‘Maratha’, however, continued. Activists from other caste groups, such as the radical
Pune-based writer and editor Mukundrao Patil, continued to take the Satyashodhak
line.43 He bitterly criticised what he saw as an obsession with Kshatriya status,
warning that it would one day bring to dust all the good work done by Phule’s
Satyashodhak Samaj.44 In his writings, he emphasised an inclusive ‘Maratha’ cat-
egory. At the other, conservative, end was the Amravati-based prolific writer K.B.

39 MSA Mumbai, Home (Special) Files, No. 363, ’Brahmans vs. Non-Brahmans’, B.V. Jadhav to
A.L. Robertson, 16 May 1915.
40
Oriental and India Office Collections, (hereafter OIOC), British Library, London, Files of the
Indian Franchise Committee, Q/IFC/71 (Bombay) and Q/IFC/30 (CP-Berar).
41
A.R. Bhosale, History of the Freedom Movement in Satara District, 1885-1947, unpublished
Ph.D. Dissertation. Shivaji University, 1978, pp. 195-96.
42
For a detailed description ofcaste conferences and the politics of local boards and representative
politics in this period as part of non-Brahman organisation, see Omvedt, Cultural Revolt in a Colonial
Society. pp. 171-206.
43
Mukundrao Patil (1886-1953), a Mali, was the son of one of Phule’s associates Krishnarao
Bhalekar, but was adopted by a relative Ganpatrao Patil and grew up in relative comfort in Ahmednagar.
He not only authored popular non-Brahman polemical tracts such as Hindu ani Brahman (Hindus and
the Brahman), and Kulkarni Lilamrut (The Kulkami Saga), but also diligently edited the longest running
non-Brahman newspaper Din Mitra from 1910-46.
" Y. D. Phadke. Keshavarao Jedhe, Pune, 1982, p. 24.

Downloaded from http://ier.sagepub.com at CALIFORNIA DIGITAL LIBRARY on September 19, 2007

© 2004 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
20

Deshmukh, who was only concerned with claiming ritual Kshatriya status for the
Marathas and uninterested in its application to the wider non-Brahman commu-
nity. Deshmukh’s immensely popular books on the history of the Marathas (A
History of the Maratha Kshatriyas, A New Sacred Thread for the Kshatriyas and
The Kshatriyas and Vaishyas Face Off with the Brahmans) focused on the upper-
caste status, genealogies and surname lists of ‘legitimate’ Kshatriya families and
analyses of Puranic texts for proof of this Kshatriya heritage. The rhetoric in such
narratives was often indistinguishable from those of Brahman conservatives and
Hindu nationalists in their defence of Vedic traditions, and statements of upper-
caste difference against lower-caste groups and untouchables.45
These conflicts and contradictions in non-Brahman discourse and the ambigu-
ities underlying Maratha were often papered over in the eagerness, especially on
the part of young radicals in the 1920s such as Keshavrao Jedhe and the firebrand,
short-lived Dinkarrao Javalkar, to put up a united non-Brahman front against con-
servative Brahman ideologues.46 Jedhe was a fervent supporter of temple entry
struggles for those considered untouchable, especially of Ambedkar’s famous 1927
Satyagraha at the village water tank at Mahad. At the same time, he also partici-
pated in the Maratha claim for Kshatriya status led by Chhatrapati Shahu and the
latter’s Kshatriya priesthood. Like many other young activists, he participated in
all these activities under the broad ambit of ’non-Brahmanism’, ignoring the im-
plications of such contradictions for the inequalities within non-Brahman society
itself. In doing so, he ended up ratifying the growing idea that ’Maratha’ was not
an all-inclusive marker for all of rural society, but the term for the peasant upper-

caste elite now dominant in the non-Brahman movement. Despite Jedhe’s support
for Ambedkar, therefore, local Marathas strongly protested against the Mahad
Satyagraha. They held a meeting to consider the readmission of their caste fellows,

45
Biographical information about Deshmukh has been difficult to trace, but we do know he was a
wealthy landed deshmukh in Amravati and financed the publication and advertisement of his books on
his own steam. In a 1921 tract, he argued that Brahmans were actually Muslim immigrants from Egypt
and the progeny of prostitutes and servants and had subsequently tampered with Vedic texts to place
themselves in a position of power. He also added that comparing the Marathas to Shudra castes such as
Mahars, Mangs, Gonds and Dhors was atmaghaataki, or a blow to the Kshatriya Marathas’ very soul.
K.B. Deshmukh, Kshatriya Vaishyancha Brahmananshi Samana (the Kshatriyas and Vaishyas Face
Off with the Brahmans), Amravati, 1921, p. 28.
46
Jedhe (1896-1959), in fact, represents this growing Maratha/Maratha-Kunbi presence in the
movement. His family, urbanised and upwardly mobile in Pune, claimed descent from the Jedhe deshmukh
contemporaries of Shivaji, and provided considerable support and leadership to the movement in Pune.
Jedhe went on to be an important Congressman in the 1940s and 1950s, briefly joining the Peasants and
Workers Party in the 1950s. Y.D. Phadke’s political biography provides an excellent overview of his
politics and ideological approach. Phadke, Keshavrao Jedhe. Javalkar (1898-1932) was alsoa Maratha-
Kunbi. but with humbler origins. His fierce anti-Brahman polemic was featured in controversial tracts
such as Deshache Dushman (Enemies of the Country). Following a trip to England in 1929-30, he
became dissatisfied with non-Brahman caste rhetoric and became increasingly Communist in his
orientation, advocating peasant issues before his untimely death in 1932. His writings can be found in
Y.D. Phadke, ed., Dinkarrao Javalkar Samagra Vangmaya (The Collected Writings of Dinkarrao
Javalkar), Pune1984.

Downloaded from http://ier.sagepub.com at CALIFORNIA DIGITAL LIBRARY on September 19, 2007

© 2004 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
 21

who were believed to have been ’polluted’ by participating in the Satyagraha.

Remarkably, Jedhe is reported as being present at this meeting as weil.4’
Non-Brahman newspapers of the early twentieth century, such as Shripatrao
Shinde’s Vijayi Maratha and Bhagwantrao Palekar’s Jagru~i, also provide clues to
the growing preponderance of an elite Maratha peasant group in non-Brahman
discourse.48 Recurrent advertisements and announcements in these newspapers
suggest that it was primarily well-off and educated Maratha and Maratha-Kunbi
readers who patronised them. Matrimonial advertisements, for instance, were in-
variably from ’Maratha famil[ies] of high birth, seeking contacts with similar fami-
lies’ .4<) It is rarely that we see families from other lower-caste groups adver-tising
in these papers, or a readiness among Maratha families to make contact with them.
Also prominent are advertisements for a wide range of religious and pamphlet
literature pertaining to Kshatriya ancestry-surname lists that one could consult
to confirm such ancestry, guides to appropriate Vedic rituals, special Kshatriya
almanacs for discerning Maratha families, and so on. One such advertisement for
an almanac, published by the Shree Shivaji Kshatriya Vedic School, announced
that it was attractive not only because all the major non-Brahman leaders had
certified it, but also because it had beautiful pictures of Shivaji and Shahu on the
cover.&dquo; Histories of elite Maratha families were also prominent.
Colonial Sociology and Representation
Let us now consider the development and impact of colonial sociological prac-
tices the transformation of the category Maratha,’. From the mid-nineteenth
on

century, the development of the colonial state into a full-fledged bureaucratic ap-
paratus heralded the construction of a much more varied and detailed body of
official information about native society, history and culture through massive
projects such as the census, gazetteers and land surveys. These were comprehen-
sive compilations of information on regional social groups, customs and rituals,
religious beliefs, and theories of their origins and history. This body of informa-
tion was important not only because of its detailed treatment of caste and ritual
practice, but also because it laid claim to a much higher standard of scientific
accuracy and finality.

47 MSA Mumbai, Home (Spl) 355 (64) II, ’Mahad Satyagraha’, CID Report, 31 Jan. 1928.
48
Shripatrao Shinde, also a Maratha-Kunbi, was the son of a merchant from Kolhapur. A prot&eacute;g&eacute; of
Shahu, he combined his work as a Satyashodhak activist with a job as a colonial policeman, but quit the
latter job to move to Pune in1918and edit the newspaper Viyaji Maratha as a non-Brahman alternative
to the hugely popular Kesari, mouthpiece of nationalist brahmans such as Tilak. Bhagwantrao Palekar
(1982-1973) also came from a Maratha-Kunbi family which was not well-off, but had Satyashodhak
connections; they moved in search of financial opportunities from Nasik to Baroda, where Palekar had
some schooling. He worked in several newspaper offices and printing presses in Bombay and Baroda
before starting the non-Brahman newspaper Jagruti, which he edited from 1917-49. His editorials and
autobiography are available in Sadanand More, ed., Jagrutikar Palekar (Marathi). Pune, 1996.
49See Jagruti. 15 Feb. 1919; Vijayi Maratha, 6 Jan. 1930 for examples.
Vijayi Maratha,7 Apr. 1930. See also the advertisement ’Excellent Histories of the Kshatriyas’
50
for a series of books by K.B. Deshmukh of Amravati in Maratha Navjeevan, 7 Apr. 1936, p. 9.

Downloaded from http://ier.sagepub.com at CALIFORNIA DIGITAL LIBRARY on September 19, 2007

© 2004 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
22

Caste became, in this bureaucratic project, a much-valued index of patterns of

behaviour, an idea that was fuelled not only by the colonial state’s desire to predict
and anticipate its subject popuiation’s actions, but also by the overwhelming influ-
ence of the theories of race and ethnology.&dquo; Bureaucratic information of all kinds,
therefore, provided break-ups of statistics by caste with some information about
the caste itself. For example, Dr H.V. Carter wrote enthloodastically in a report on
leprosy in the Bombay presidency in the early 1870s:
The subject of caste is full of instruction to the antiquary and the ethnologist: it
is a mine as yet unworked, but which holds information sufficient, by analysis
of details, to explain many curious anomalies in the opinions and condition of
the existing native races, if not to throw light on their proclivities. 52

Early colonial writings had had much to say about the military proclivities of the
Marathas and the category remained a favourite subject of the colonial sociologi-
cal pronouncements that multiplied in the late nineteenth century. For instance,
the second edition of the Imperial Gazetteer, published in 1885, stated:

The Marathas have a distinct national individuality. They are an active, ener-
getic race liable to religious enthusiasm and full of military ardour ... the chief
caste or tribe among them is the agricultural Kunbi ... Shivaji himself belonged
to this fighting class of the Kunbi peasantry Altogether the Marathas ac-
0

....

knowledge upwards of two hundred castes, including thirty-four septs [sic] of

Brahmans.&dquo;

Although the Imperial Gazetteer retained the earlier idea of the Marathas as a
polity with common religious and martial attributes, the provincial gazetteers pub-
lished throughout the 1880s told a different story. Concerned as they were with
recording in painstaking detail the practice of custom and ritual, particularly those
relating to social status and marriage, these gazetteers focused a great deal of at-
tention on the idea of ’Maratha’ as a marker of social/jati status. All of them also
recorded the flexibility of the distinction between Marathas and Kunbis. The
Kolhapur volume recorded that
the martial classes among the Marathi-speaking middle classes called them-
selves Marathas. Some families have perhaps an unusually large strain ofNorth-
ern or Rajput blood, but as a class Marathas cannot be distinguished from

Marathi-speaking Deccan Kunbis, with whom all eat and the poorer intermarry.&dquo;
51
Rashmi Pant, ’The Cognitive Status of Caste in Colonial Ethnography: A Review ofthe Literature
on the Northwest Provinces and Awadh’, Indian Economic and Social History Review, Vol. 24(2),
1987, p. 148.
52
Quoted in ’Castes of the Bombay Presidency’, Indian Antiquary, Vol. 2, 1873, p. 154.
53
W.W. Hunter, ed.. The Imperial Gazetteer of India, 2 ed, 9 Vol s., Vol. 3, London, 1885, pp. 50-51.
54
The Gazetteer of Bombay Presidency, Kolhapur District. Vol. 24, Bombay. 1886, p. 65.

Downloaded from http://ier.sagepub.com at CALIFORNIA DIGITAL LIBRARY on September 19, 2007

© 2004 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
 23

Other volumes for the districts of Poona, Satara, Ratnagiri, Berar and Nagpur
also recorded the eagerness with which Kunbi families were taking to the sacred
thread, and the connection that some families claimed with Rajput and Kshatriya
ancestry.
As a result of field surveys and research, the analysis that the gazetteers summed
up about the Marathas was shaped by the changing patterns of ritual and status
claims that were taking place in the late nineteenth century. In these writings,
Maratha emerged not as a term to be stretched to the entire Marathi-speaking popu-
lation, but the marker of a specific caste group. When the new edition of the Impe-
rial Gazetteer was published in 1908, it abandoned its earlier ’national’ descrip-
tion of the Marathas and wrote that ’of the total population of the Deccan districts,
thirty percent are Marathas, between whom intermarriage is permissible’.55
This tension between the historical antecedents of the Marathas as a polity or a
tribe and their contemporary avatar as a caste group remained a central feature of
colonial sociological writings well into the twentieth century. This debate was
influenced not only by the need to identify large numerical majorities in different
regions for administrative reasons, but also by the larger ethnological debate on
the Aryan racial presence in the subcontinent and its expression in different caste
groups. H.H. Risley, the leading proponent of the racial view of caste origins, was
convinced, on the basis of anthropometric measurements of people from the Deccan,
that the Marathas were of Scythian origin. To him,

The physical type of the people of this region accords...well with this theory
[of Scythian origin] while the arguments derived from language and religion do
not seem to conflict with it... on this view the wide-ranging forays of the Marathas,
their guerilla methods of warfare, their unscrupulous dealings with friend and
foe, their genius for intrigue and their consequent failure to build up an enduring
dominion, might well be regarded as inheriied from their Scythian ancestors.5’

In commenting on the ’character’ of the Marathas, however, Risley was not refer-
ring to the Marathas as a caste; he was referring to the Marathi-speaking popula-
tion as a whole.
R.E. Enthoven, the Superintendent of Ethnography for the Bombay presidency
and the author of Tribes and Castes of Bflmbay, disagreed strongly with Risley’ss
conclusion, on the grounds that it clashed strongly with contemporary evidence of
social hierarchy prevailing in the Deccan. The fact that a Chitpavan Brahman and
a member of the untouchable Mahar caste could have the same cephalic index
measurements was to Enthoven ’at least disconcerting’ and he added that ’the
Mahar would not be expected in such strange company’.5’ To Enthoven, the

55
The Imperial Gazetteer of India. 3 ed, Calcutta. 1908, p. 302.
56
H.H. Risley, The Imperial Gazetteer of India. Calcutta 1885, quoted in R.V. Russell, The Tribes
and Castes of the Central Provinces of India, London. 1916, p. 201.
57
R.E. Enthoven, The Tribes and Castes of Bombay, Bombay, 1920, p. xv.

Downloaded from http://ier.sagepub.com at CALIFORNIA DIGITAL LIBRARY on September 19, 2007

© 2004 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
 24/

Marathas were possibly the descendants of an aboriginal tribe, which might then
have mingled with some influences from the north. He wrote at length about the
village guardian deities or devaks that were important to the Maratha caste rituals,
concluding that these totems were evidence of a ’pre-Aryan element in the
Marathas’. Kunbi itself, he argued, was an occupational term, and derived from
the Sanskrit word for husbandman, kutumbika. As such, then, Marathas and Kunbis
could not exactly be tenned a tribe, but a group with mixed origins that developed
into a caste. In explaining this development, Enthoven turned to history.

There is probably no substantial difference in origin between the landholding

and warrior section, i.e. Marathas Proper, the cultivators, i.e. Maratha Kunbis
and the numerous local occupational castes....the rise of the Maratha power in
the 17the century induced the fighting classes...to claim for themselves Kshatriya
rank and to discourage widow remarriage. It is chiefly on this ground that they
claim to be superior to the Kunbis. But by descent the Maratha appears to be
one with the Maratha Kunbis and certain other occupational castes in the
Deccan.&dquo;

R.V. Russell, who supervised the compilation of the Tribes and Castes of the
Central Provinces, also wrote at length about the Marathas’ origins. According to
him, the present incarnations of caste groups in Indian society were the result of its
organisation into village communities. This was definitely the case with the Kunbis,
who as a tribe might have settled in fresh areas and slowly developed into a caste. 59
As for the Marathas, they were a caste ’of purely military origin constituted from
the various castes of Maharashtra who adopted military service [under Shivaji],
although some of the leading families may have had Rajputs for their ancestors.&dquo;’
Various census reports published from the 1890s to the 1930s also grappled
with the question of the Marathas’ origins. The Central Provinces Census Report
for I891, expressing a great deal of frustration at the lack of any uniformity in
answers to questions about caste groupings, finally decided to get rid of local
divisions and classed the Marathas as a whole as a ’military tribe’ under the sec-
tion of Dominant Agricultural and Military Castes.61 The Bombay Census of 1901]
referred to the Marathas as the ’descendants of Shivaji’s warriors’, a tribe who were
then ’split asunder by virtue of social inequalitites’.6z It classed them as a caste of
a ’rational’ type, on the basis of their having a lower degree of racial purity than
tribal castes; their racial admixture was of a comparatively recent origin and the
basis for their unity political and not racial.&dquo; Census reports for both these provinces
followed the occupational classification of caste laid out by leading ethnographers

Ibid., pp. xviii.

58
59
Russell, Tribes and Castes of the Central Provinces, p. 43.
60
Ibid., p. 201.
61
Census of India, Central Provinces, Vol. 12. Calcutta, 1891, p. 165.
62
Census of India, Bombay Presidency, Vol. 8, Calcutta, 190 1, p. 184.
63
Ibid, p. 183.

Downloaded from http://ier.sagepub.com at CALIFORNIA DIGITAL LIBRARY on September 19, 2007

© 2004 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
 25

such as John Nesfield

j
and William Crooke as opposed to the racial formula of
Herbert Risley. Racial origins, however, continued to haunt census officials and
their approach to caste, as suggested by a comment in the Central Provinces Cen-
sus Report of 1931:

Caste is so mixed up with race that for ethnological purposes a continuous

record is essential. To treat in a single class the Maratha Brahman and the Maria
of Chanda, the Rajput and the Saonr of Saugor ... or the Bairagi and Chamar of
Chhatisgarh would be openly to flout science. 64
The confusion among colonial ethnographers about the exact nature of ‘Maratha’
was clearly linked to the ongoing processes of status claims within rural society
since the mid-nineteenth century. The claims of many Maratha families to Kshatriya
and Rajput status brought to the fore the issue of so-called Aryan origins, which
ethnographers like Enthoven and Russell, arguing from the perspective of village
communities, totem deities and the like, were not willing to endorse. The appear-
ance of pamphlets claiming to provide proof of these Rajput and Kshatriya origins

through Puranic origin myths, surname lists and so on in the early twentieth cen-
tury-K.B. Deshmukh’s works discussed above are an excellent example-forced
ethnographers to take note of these claims and their possible veracity. Enthoven
gave considerable space to these narratives in his work, trying in vain to analyse
them scientifically and logically. Despite his insistence on the non-Aryan origins
of the caste group, he finally concluded that at present 54 Maratha families (whose
names he provided) could logically be said to have legitimate Kshatriya and Rajput

ancestry
Colonial sociological materials, thus, differed widely over the meaning of
’Maratha’, with extensive debate about its classification and nature as a tribe rooted
in history or a caste group characterised by common practice. In both cases, it is
necessary to reiterate, broader intellectual influences of race and ethnology were
important in shaping colonial lines of enquiry, but the debates and practices offi-
cials encountered on the ground, as it were, within Maharashtrian society, consid-
erably muddied the end result of their investigations. On the one hand, the consis-
tent invocation of Shivaji and his history in writings about the Marathas, particu-
larly at the height of the non-Brahman movement, made it impossible to ignore a
sense of tribal--or racial--unity that fit in neatly with the late nineteenth-century

understanding of political entities in racialised terms. On the other hand, changing

ritual practices within Maratha-Kunbi groups, the foregrounding of Brahman ex-
ploitation in non-Brahman discourse and the communitarian debates over who
really belonged in the category Maratha made it imperative to take note of it in
64
Census of India, Central Provinces, Vol. 12, Delhi, 1931, p. 352.
65
Enthoven uses extensively a Marathi pamphlet Kshairavanshsagar (Ocean of Kshatriya Families)
’published by a Maratha of high birth’ called Mr Patankar. Enthoven, Tribes and Castes of Bombay,
Vol. III, p. 23.I have been unable to locate this tract itself, but it appears to be quite similar to the kind
of literature being produced in the 1920s and 1930s by conservative Marathas like K.B. Deshmukh.

Downloaded from http://ier.sagepub.com at CALIFORNIA DIGITAL LIBRARY on September 19, 2007

© 2004 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
26/

caste terms, in addition to the

classificatory value that caste itself had gained in the
process of ’knowing’ Indian society.
Maharashtrian writers, both Brahman and non-Brahman, also creatively retash-
ioned colonial points of view in putting forward their own social and political
claims.66 The pioneering nationalist historian V.K. Rajwade’s writings are a good
example of Brahman statements of social leadership couched in scientific, histori-
cist language.67 Rajwade was clearly influenced, like many others of his genera-
tion, by the ongoing debates over caste and race in colonial ethnography and used
the broad framework of Aryan settlement from the north into the Deccan to dis-
cuss the ancient social history of Maharashtra.6$ His application of the theory,

however, posited the Brahmans as the only true preservers of the Aryan tradition.
The Marathas appeared as the m ixed products of contacts between inferior Kshatriya
peoples who had tired of Buddhist ideas in the Gangetic plain and settled further
south, and the aboriginal Naga peoples of the Deccan. These people were, he
wrote, ’totally dependent on the priesthood’, incapable of government and were
conquered successively by various Kshatriya peoples of the north: the Chalukyas,
Rashtrakutas, Yadavas, etc. Some of these Kshatriya influences permeated into a
few elite families, among them Shivaji’s Bhosale lineage, but the bulk of the Maratha
people remained quite uncultured. ’With no proper deities, no definite religion, no
alphabet or sense of history, these people were responsible for the downfall of
kings and the godly Brahmans. ’6~ Grant Duff had concluded that the lack of any
architectural achievements by the Marathas was testimony to their cultural weak-
ness as a nation. Rajwade, otherwise bitterly critical of Duff, used precisely this

argument to criticise the Marathas, but as a caste. Underlying this criticism, of

course, was his own agenda of refuting non-Brahman claims to being the genuine
shapers and inheritors of Maharashtra’s history and culture.
In refuting arguments such as those of Rajwade, non-Brahman writings also
used sections of culonial discourses in putting forward their own interpretations of
Maratha history or the Maratha caste group. Jotirao Phule’s skillful use of the

66
cf. Sumit Sarkar, Writing Social History, New Delhi, 1997. Sarkar argues persuasively that it is
important to recognise that this skillful borrowing of colonial discourses reflects not simply the agency
of the colonised, but also elite attempts to perpetuate the very real inequalities that existed within
Indian society along the lines of caste and gender.
67
V.K. Rajwade ( 1863-1926) was the pioneer of a modern, national ist historiography of the Marathas.
Between 1898 and 1920 he collected and published, at great personal expense, 20 volumes oforiginal
sources from the Maratha administrative record Marathyanchya Itihasachi Sadhane (Sources of Maratha

History). He wrote extensively on history and language, but it was his pronouncements on Brahman
social leadership and superiority in Maratha history that became the most controversial. Rajwade’s
collected works are now available in Itihasacarya V. K. Rajavade Samagra Sahitya (The Collected
Works of V.K. Rajwade), 13 Vols, 1995-98.
68
See in particular the articles ’The skin-colour of Aryans in India’, ’Our Puranas and the new
discoveries in Assyria’. ’Then who are the Brahmans?’ and ’The Inclusion of non-Hindus in Hindu
society’ included in L.S. Joshi. ed., Rajwade Lekhasangraha, (Selected Writings of Rajwade). 4 cd.
New Delhi. 1992.
p. 140.
Ibid,
69

Downloaded from http://ier.sagepub.com at CALIFORNIA DIGITAL LIBRARY on September 19, 2007

© 2004 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
 27

Aryan race debates to delegitimise Brahman leadership over Hindu society by

depicting them as outsiders oppressing aboriginal peoples is well known. Other
polemicists such as K.S. ’Prabodhankar’ Thackeray, also a severe critic of James
Grant Duff, nevertheless selectively borrowed from his History of the Mahrattas
tohighlight the picture of Brahmans as crafty and untrustworthy.’° Wasudeo Lingojii
Birze, librarian to the prince of Baroda and author of two popular books called
Who are the Marathas? and The Life of Kshatriyas leaned heavily on Grant Duff,
James Tod and Harry Acworth, the compiler of Marathi ballads, to bolster his
claim of an ancient, upper-caste Kshatriya genealogy for the Marathas and to contest
Brahman arguments that no Kshatriyas remained in the Kaliyuga or the present time.&dquo;
Non-Brahman campaigners encouraged prosperous peasants to drop the Kunbi
tag during local elections as well, as the case of the prosperous Tirole Kunbis-
turned-Marathas of Khandesh demonstrates.&dquo; There were also concerted efforts
to influence the production of colonial discourse on the Marathas through cam-
paigning before census operations. Many Maratha organisations within the non-
Brahman movement attempted, through conferences and articles in various news-
papers just before the census surveys in 1921 and 1931, to persuade the broader
Kunbi population to return itself not as Kunbi but only as Maratha:

All literate Maratha people know that in the census times many illiterate villag-
ers call their caste ’kulvadi’ or ’kunbi’ rather than ’Maratha’. Except for Leva
Kunbis in Khandesh, all those who call themselves ’Kunbis’ or ’kulvadis’ in
Maharashtra, Konkan, Berar etc. are of ‘Maratha’ caste. Only out of ignorance
do people not call themselves ’Marathas’. Educated Marathas should clearly
inform any ignorant Maratha The days of the rule of wealth have gone and
....

the day of the rule of numbers has come: we hope our educated Maratha society
will remember this.73

A couple of the census officers also remarked on the urgency of these attempts,
which, it appears, did pay off. The numbers of Kunbis all across the Marathi-
speaking areas dropped significantly in the 1921 and 1931 Censuses, with notable
increases in the number of people who now called themselves simply ’Maratha’ .74

70
K.S. Thackeray, Gramanyacha Sadyanta Itihasa, Arthat, Nokarashahiche Banda (a
Comprehensive History of Rebellion), Mumbai, 1919, pp. 32-33. Playwright, journalist and father of
the current Shiv Sena chief, Thackeray is better known as ’Prabodhankar’ in Maharashtra because of
the popular non-Brahman weekly Prabodhan he edited for many years. Although not a Satyashodhak,
he spearheaded the non-Brahman critique of Brahman historiography and is famous for his sarcastic
and biting language. His collected works are now available in Prabodhanakar Thackeray Samagra
Vangmaya (Collected Works of Prabodhankar Thackeray), Mumbai, 1998.
71
Wasudeo Lingoji Birze, Who are the Marathas?, Bombay, 1896 and Kshatriya Ani Tyanche
Astitva (the Life of Kshatriyas), 2 cd, Bombay, 1912.
72
Omvedt, Cultural Revolt in a Colonial Society, p. 133.
73
This was an appeal by the Kshatrajagadguru, the Maratha priestly head instituted by Shahu,
published in the Satyavadi newspaper in February 1931. Quoted in ibid., p. 134.
74
Census of India, Central Provinces, Calcutta, 1921, p. 147; see also Census of India, Central
Provinces, Calcutta, 1931, p. 368.

Downloaded from http://ier.sagepub.com at CALIFORNIA DIGITAL LIBRARY on September 19, 2007

© 2004 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
28/

Two important policies of the colonial state served to further consolidate the
attempts of the elite non-Brahman sections to claim ’Maratha’ as the marker of an
upper-caste Kshatriya status. The first was the creation of political categories of
representation through the Montague-Chelmsford Reforms put in place in 1921.
Non-Brahman leaders demanded separate electorates for a series of non-Brahman
castes, but instead, the reforms granted seven reserved seats for a single category
termed ’Marathas and Allied Castes’. Although O’Hanlon has pointed out that
this umbrella category served as an official recognition of the common non-Brah-
man cause and the terminology Marathe ani Itar (Marathas and Others) that activ-
ists themselves applied, it is important to note that this official category also high-
lighted, simultaneously, the very real and perceived differences that existed be-
tween non-Brahman castes and the growing dominance of Marathas and Maratha-
Kunbis within the non-Brahman fold.75 Official comments on this dominance in
progress reports on the reforms were matched by protests from representatives of
the ‘Allied castes’ throughout the 1930s against Maratha control of these piece-
meal constitutional and electoral benefits allotted by the colonial government.’6
The second important colonial policy was the delineation of’Maratha’ for pur-
poses of recruitment into the army. The Maratha military past was clearly impor-
tant in the identification of the Marathas as a ’martial race’, and the reiteration of
the martial overtones of this history in colonial sociological materials about the
Marathas as a caste no doubt played a crucial role in earmarking them as fit for
recruitment. Constable also points out that the Eden Commission’s emphasis on
localised recruitment, combined with the move towards greater social
homogenisation of companies and battalions, increased the Bombay Army reli-
ance on local Marathas from the 1880s, but was matched by the growing dissatis-
faction with the fighting qualities of these recruits and doubts about their true
martial attributes.&dquo; It is perhaps for this reason that the 1908 recruitment hand-
book issued by the army insisted on restricting recruitment only to ’genuine
Marathas’ and painstakingly detailed the ’pure’ groups, the ones ’attempting to
pass’, and the means to detect any such ’deceit’ .78 The book also provided a list of
96 family names as a reliable guide to determine whether the person was a genuine
Maratha. It highlighted that, keeping their high-caste Hindu status in mind, they
would also have the benefits of cooks and water-carriers available to them.

A recruit on enlistment is asked his caste by the recruiter and again by the

Recruiting Staff Officer. In the Konkan, a recruit when asked if he is a Maratha

will say ’yes’. He should then be asked whether he is a ’Rao’ or a ’Naik’. If he

75
Rosalind O’Hanlon, ’Acts of Appropriation’, pp. 111-12.
76 OIOC. British Library, London. Files of the Indian Delimitation Committee, Q/IDC/7, Report of
the Delimitation Committee in Bombay and Sind, 1935 ; N.C. Surve. Deposition to the Delimitation
Committee, 9 Sep. 1935.
77
Constable, ’The Marginalization of aDalit Race’, p. 455.
78
R.M. Betham, Marathas and Dekhani Musalmans:Compiled under the Orders of the Government
of India, Calcutta. 1908, p. 49.

Downloaded from http://ier.sagepub.com at CALIFORNIA DIGITAL LIBRARY on September 19, 2007

© 2004 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
 /29

says he is neither or tries to explain what he is, it may be assumed he is not a

pure Maratha and he should be rejected. In the Dekhan, a man may say he is a
Kunbi, a Maratha or both; he can be accepted. Those who claim to be only
Marathas and not Kunbis are probably of better birth. If a man tries to explain
he is neither by some other class of Maratha, he is not a Maratha at all and
should be rejected.’9

Maratha recruitment was stepped up after World War I, after a series of

commendable performances by the Maratha Light Infantry coincided with non-
Brahman efforts to step up the recruitment of young non-Brahman men into gov-
ernment jobs, political office and the army. The favourite grounds for recruitment
were the strongholds of elite Maratha power, Satara and Kolhapur. The handbook’s s

guidelines testified to the elasticity of the category but attempted to fix it as a

narrow upper caste; actual recruitment practice once again, therefore, served the
interests of elite Marathas and upwardly-mobile Maratha-Kunbis. Throughout the
early twentieth century, various Maratha associations and non-Brahman newspa-
pers enthusiastically promoted the attractions of the army, but according to
Constable, Maratha military officers also showed an increased intolerance towards
lower-caste and Dalit recruits in their ranks.&dquo; The All India Maratha Conference’ss
statement, published in the Jagruti issue of I March 1919, was directed as much at
the Maratha community as to the colonial government:

The Conference of course expresses its loyal support [to the government] and
expresses delight about the victory in the recent war. In battlefields across the
world, brave Maratha heroes displayed their dazzling Kshatriya qualities in
this war. Along with this report, therefore, we put forward a request that the
Marathas be allowed to play a greater role in the army and display their military
qualities, and that some educational institutions, especially a college be estab-
lished to allow them to further develop these qualities.$’

Constable has argued that the colonial martial race ideology and its insistence
on a high-caste Maratha pedigree for recruitment was ‘less a hegemonic colonial
strategy invented by the British colonial establishment for British social and mili-
tary control of India, than colonial &dquo;accommodation&dquo; for strategic purposes of
higher-caste Maratha claims to social exclusivity as kshatriyas’ ,82 Indeed, we do
witness here the impact of non-Brahman debates over ’Maratha’ on colonial
ethnography and policy in general, but we also see the need for colonial discourse
to develop a fixed formula that would transcend these bewildering claims. More-
over, the colonial state’s search for, and ability to fix, such formulas regarding

79
Ibid., p. 96.
80
Constable, ’The Marginalization of aDalit Race’, p. 459.
81
The All India Jagruti
Maratha Conference’, I Mar. 1919.
,
82
Constable. ’The Marginalization of a Dalit Race’, p. 459.

Downloaded from http://ier.sagepub.com at CALIFORNIA DIGITAL LIBRARY on September 19, 2007

© 2004 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
30

caste in structures of recruitment and representation were instrumental in deciding

which claims were consolidated and which ones lost out. In identifying the Maratha
as a high-caste Kshatriya category different from other rural caste groups, the army
recruitment policy as well as the reserved seats category ’Marathas and others’
ended up consolidating the elite and exclusivist strand within non-Brahman dis-
course. By the 1930s, the radical, inclusive interpretations that had made ’Maratha’
a new vision of community and a site for articulating a more popular alternative to
Brahman nationalism had all but dissipated. In the formal political arena as well as
in the public sphere, the idea that the Marathas were upper-caste Hindus, a Kshatriya
caste group with a glorious military tradition, became dominant. Those who ben-
efited the most from the crystallisation of this view in the formal political appara-
tus were rural elites best placed to successfully establish this pedigree.
In the 1930s, younger non-Brahmans, increasingly unhappy with the non-
Brahman party’s loyalism and influenced by socialist ideas, established links with
Gandhian Congressmen and entered nationalist politics.g3 Jedhe’s efforts in
particular led Congress membership to soar from 45,915 members in 1936 to
1,56,894 the next year.84 Rural participation was certainly higher during the Civil
Disobedience and Quit India movements than in the non-cooperation struggles of
the 1920s; official daily and weekly reports from across Maharashtra suggest that
this move of non-Brahmans towards nationalism was certainly crucial to the
Congress’ success in rural mobilisation.85 The Congress tapped into the rural net-
works built by the non-Brahman movement in the 1920s: following Jedhe’s move,
to cite just one example, the party was able to capture an unprecedented 10 out of
1 I district local boards in Bombay in 1935.86 _

Y.B. Chavan records in his memoirs that the Congress’ entry into these institu-
tions through the nomination and election of non-Brahman Congressmen was
crucial to the party’s rural base; importantly, it paved the way for rural Maratha
leaders to gradually displace older, Brahman Congressmen.81 The non-Brahman

83
O’Hanlon, ’Acts of Appropriation’, pp. 126-46.
84
MSA Mumbai, Home (Spl), 922 (2), Extract from the Bombay Presidency Weekly Letter No. 47,
27 Nov. 1937, p. 31.
85
MSA Mumbai, Home (Spl), 1 43-E-I, ’Non-Brahman Party’, CID Weekly Report, Satara District,
31 May 1930; 800 (74) 21, Weekly Reports, Ratnagiri District (1932-34) 28 Jan. 1932; 800 (106)-D
(4) Weekly Reports, Ahmednagar District, 26 Feb. 1937; 800 (74) (11) Confidential Report No. 7.5/
1932, Dhule District; MSA Nagpur, CP Political & Military Files, Case # 12, Akola District, 1941.
86
Phadke, Keshavrao Jedhe, p. 164. The Gandhian Congress was also able, remarkably, to win the
elections to the Pune municipality at this time, and storm the stronghold of Tilakite conservative
Brahmans.
87
Y.B. Chavan, Krishnakath: Atmacharitra (on the Banks of the Krishna. an Autobiography), Vol.
1. Pune, 1984, pp. 172, 176, 229. Chavan, of humble Maratha-Kunbi origin, joined rural politics in
Satara district as a young recruit in the later 1920s and rose to prominence in the 1942 Satara struggle.
He led the Bombay Congress in 1956 and became the first chief minister of the bil ingual Bombay state
in 1956 and also of United Maharashtra in 1960. His diplomatic skills were instrumental in negotiating
a successful settlement with the Central Government over the demands for the new Maharashtra state,
and he went on to serve as Minister for the Home. Defence and Finance portfolios in the 1960s.

Downloaded from http://ier.sagepub.com at CALIFORNIA DIGITAL LIBRARY on September 19, 2007

© 2004 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
 /31

movement, despite all its caste and economic differentiation, had always shown a
greater inclination towards voicing rural issues and grievances than the urban,
Brahman Congress; its pro-peasant rhetoric in this mass phase went a long way in
giving the Congress party itself a rural face, and the Marathas an edge in the new
nationalist politics. It was this new nationalist rural Maratha leadership and its
networks built through non-Brahman activism that spearheaded the parallel gov-
ernment of Satara during the 1942 Quit India movement. The movement thrived
on underground networks and overwhelming popular support from the local peas-

antry, with one of the largest concentration of Marathas.g8 It also propel led younger
Maratha leaders such as Nana Patil and Chavan himself, with their experience of
grassroots political activity, into prominence in the Congress Party in the 1940s;
this experience and contact with rural networks was crucial to the Maratha pre-
dominance in electoral politics following Independence.

Conclusions

In her analysis of the use of ’Maratha’ by non-Brahman activists in the 1890s,

Rosalind O’Hanlon has rightly remarked that this ambiguity over the category’s
content and meaning was a source of both strength and weakness. While it al-
lowed non-Brahmans to exclusively claim the regional history and heritage of
pride and struggle of the precolonial past and delegitimise Brahman claims to
social leadership, its elite inflections made it difficult for all non-Brahman groups
to lay equal claim to it.89 Activists in the twentieth century such as Mukundrao
Patil struggled to keep alive this radical edge to the category by using it to refer to
all non-Brahman caste groups in their rhetoric, but with increasing difficulty. The
Kshatriya claims within non-Brahman discourse together with the growing domi-
nance of Maratha-Kunbis within the movement and in rural networks of power
ensured that the radical possibilities the category had been invested with during
the late nineteenth century lost out to exclusivist, upper-caste claims. Finally, the
imperatives of colonial ethnography and its crystallisation in policies of enumera-
tion and representation consolidated this emerging elite ’Maratha’ dominance over
rural society.
I have attempted in this article to track one example of the multiple ways in
which social categories were formulated and debated in the colonial period. This
overview of the discourses and policies surrounding the definition of the category
’Maratha’ emphasises the importance of keeping regional histories and contexts
in play when plotting the modern history of caste and caste identity. Recent schol-
arship on caste has been invaluable in historicising caste, but has resulted in a
rather polarised debate over the degree of blame to be attached to colonialism for
the central role that caste occupies in Indian social and political life today. An

88
A.B. Shinde. The Parallel Government of Satara: A Phase of the Quit India Movement. New
Delhi, 1990; Gail Omvedt, ’The Satara Prati Sarkar (the Parallel Government of Satara)’, in Gyanendra
Pandey, ed., The Indian Nation in 1942, Calcutta. 1988.
89
O’Hanlon, Caste, Conflict and Ideology, pp. 275-77.

Downloaded from http://ier.sagepub.com at CALIFORNIA DIGITAL LIBRARY on September 19, 2007

© 2004 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
32

analysis of colonial discourse and official policy regarding ’Maratha’ indicates

that colonial sociology was not homogeneous, and that colonial attempts to under-
stand the ’Maratha’ category in all its sociological, political and historical impli-
cations had a much more complex relationship with the ongoing debates within
Maharashtrian society, with both spheres influencing and significantly borrowing
from each other. To both colonial observers as well as Maharashtrian writers, the
specific precolonial history of Maharashtra, be it Shivaji’s military adventures or
the Peshwai’s rigid Brahmanical strictures, played a crucial role in determining
how the Marathas were to be understood, organised and represented: this under-
scores the importance of viewing these debates and policies in their appropriate

regional setting.
Recognising this two-way borrowing between colonial and Indian discourses
is not to reduce the colonial state’s tremendous power in setting the terms of the
debate or in influencing the larger political and social environment within which
Indian writers and activists themselves functioned. Indeed, as we have seen, colo-
nial policies regarding electoral categories and military recruitment played a cru-
cial role in consolidating changes taking place within non-Brahmam discourse.
Instead, it is to highlight the complex interactive process through which caste
categories were constructed, the selective and skillful use of colonial discourses
by different groups of Indians to advance different social and political claims, the
many ways in which these claims themselves influenced colonial categories of
representation, and the need to consider regional particularities in locating this
agency.

Downloaded from http://ier.sagepub.com at CALIFORNIA DIGITAL LIBRARY on September 19, 2007

© 2004 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.