The Journal of Peasant Studies, Vol.27, No. 1, October 1999, pp.35-60
The Bahujan Samaj Party in Uttar Pradesh IAN DUNCAN
Ian Duncan, School of African and
Asian Studies, University of Sussex, Brighton BN I 9QN. The
INTRODUCTION
What is distinctively new about the BSP is that it has achieved an electoral success,
particularly in the state of Uttar Pradesh (UP), never experienced before by a political
party seeking to represent predominantly 36 THE JOURNAL OF PEASANT STUDIES the interests of ex-Untouchables. The party has established itself as the preferred electoral choice of many of the poorest and most marginalised sections of the north Indian population and specifically it has gained the support of peasants and agricultural labourers in large numbers. It has been seen as 'the massive electoral participation of the untouchable "chamars" for the first time' [Srinivasulu, 1994: 159] and as a party specifically oriented towards the Scheduled Castes that has generated a solidity and stability among its supporters 'which constitutes a fundamental change from the past' [Brass, 1997: 2419]. In the process of mobilising this following the party has contributed to a radical realignment of social forces and created an autonomous political identity among the Dalits that has cut away the support they were previously willing to give to other parties. The decline in the influence of the Congress especially in UP, and the subsequent reverses for the party nationally, can in large part be attributed to the loss of much of its Dalit support base to the BSP in the 1990s. The unprecedented success of the party can also be measured by the fact that it has been able, albeit in alliance with or depending upon the support of other parties, to take power in governments in UP on three occasions during the 1990s and in the process bring to power the first Dalit Chief Minister of the state, Ms Mayawati. Although these administrations have proved to be extremely unstable and short-lived they mark a dramatic break from the pattern of coalitions of social forces that have ruled the state previously [Duncan, 1997b; Prasad, 1995]. Although the BSP has undoubtedly managed to generate the enthusiastic support and loyalty of a wide range of the rural poor of UP it does not seem to have done so as a consequence of any major campaigns or particular emphasis on issues affecting landless labourers or marginal landholders as such. Rather its appeal and campaigning appear to be far more caste based than class oriented and it has been primarily concerned with issues of social oppression and exclusion than with those relating to economic exploitation. One comment on its early activities remarked that 'the mobilisation of the rural poor is seldom done on the issues emanating from capitalist agriculture' [Singh, 1993: 109] and other observers see the party as another expression of the unprincipled factionalism characteristic of UP politics and not going much beyond 'relatively narrow objectives such as caste-based reservations of public sector employment' [Dreze and Gazdar, 1998: 104]. The BSP in this respect can be seen as a part of the growing politicisation of caste in India in recent times that has resulted in a changed focus of claims and demands on the part of mobilised groups with more attention being given to social status and political power than to economic advancement [Kothari, 1994: 1590]. Although there is a substantial congruence between class and caste and some observers [Lieten: 1994] have DALITS AND POLITICS IN RURAL NORTH INDIA 37 interpreted the resurgence of caste mobilisation as a surrogate form of class politics and others see caste as inextricably bound up with relations of production and surplus extraction [Omvedt, 1994a: especially 21-58], the absence of any class based element in the strategy of the party places severe limitations on the further expansion of the electoral appeal of the BSP. Similarly the absence of any class based policy framework calls into question the ability of the BSP in government to begin to remedy the profound economic inequalities and poor performance of the state in issues of public policy. Whilst the solid core of support for the BSP certainly rests on its base among the ex-untouchables, the party also aspires and claims to be a political expression and social movement of the Bahujan, the majority, rather than just the vehicle of the Dalits. Included among this potentially wider collectivity which the BSP seeks to mobilise are the so-called Other Backward Classes (OBCs), these are the low but not ex-untouchable caste groups, the Scheduled Tribes, and the Muslim population. The extent to which it has succeeded in this respect will be a central concern of this article.
DALITS IN THE RURAL ECONOMY
OF UP
The SC cultivators are also more likely to be operating much smaller holdings than the non-SC and many of them returned in the census as landholding cultivators are likely to be working as labourers as well. Comprehensive contemporary data on comparative landholdings for the two categories are not available for the various districts of the state but figures collected in a 20 per cent sample survey during the 1961 Census give us some indication of the degree of discrepancy in landownership, which if anything has probably increased since then. As can be seen from Table 2 nearly 60 per cent of SC cultivating households operated holdings of less than 2.5 acres compared to below 35 per cent for the non-SC population. It should also be pointed out, in relation to these figures, that the approach to the classification of landholders, conducted on the basis of cultivating households rather than individual operational holdings, resulted in a significant underestimation of the numbers cultivating the smallest holdings [Brass, 1980: 405, 424]. The 1961 method of enumerating cultivating households reported 39 per cent under 2.5 acres [Government of India, 1966: 16-53] but the 1971 Agricultural Census found 67 per cent of operational holdings under the roughly equivalent size of one hectare DALITS AND POLITICS IN RURAL NORTH INDIA 39 [Government of Uttar Pradesh, 1978: 53-54], a discrepancy unlikely to have been produced by land transfers in the space of ten years.
The brief account above of the position of the Dalits shows the extent to which the
various SC groups are typically to be found predominantly in the ranks of agricultural
labour or small landholders. As well as occupying an extremely precarious position in
the rural economy of the state, Dalit groups, despite undeniable improvements in their
social position in recent decades, are still firmly rooted to the bottom rung of the caste
hierarchy. In electoral terms the Dalits suffer from the further disadvantage that they
rarely form the numerically dominant caste in any part of UP and are typically evenly
spread and not found in the geographical concentrations that give other caste groups
electoral advantage in particular constituencies. Jatavs for example are usually found
as the second most numerous caste group in villages right across the state
[Schwartzberg, 1965; Schwartzberg, 1968].
Any party seeking to mobilise Dalits as an electoral force faces formidable difficulties
given their position in the social and economic structure. These difficulties are
compounded when it also aspires to go further and recruit support from a wider
constituency of groups seen to be the subject of similar social oppression and
economic exploitation. In the past Ambedkarite Dalit parties, if at all successful, have
found themselves the subject of tempting offers of co-option from ruling parties at the
state and national level [0ommen, 1984: 54], to which many have succumbed.
It is a common complaint that Dalit parties, and individual Dalit politicians, are all too
likely to become chamcha, puppets or stooges, of dominant interests once they
begin to have any degree of electoral success. In response to this there has emerged a
strong insistence on autonomy and organisational independence in contemporary
Ambedkarite social and political movements which often appear to behave in an
openly hostile 40 THE JOURNAL OF PEASANT STUDIES manner even to sympathetic potential allies [Kothari, 1994: 1593; Duncan, 1997b:986]. Dalit political parties are also constantly vulnerable to the accusation that they are 'casteist' and aware that if they are too successful in consolidating a particular caste group they may appear to be exclusive and uninterested in the welfare of others. The dilemma was stated clearly by one of the BSP leadership in 1991 when he addressed an Ambedkar anniversary rally: 'It is possible to consolidate the Harijan vote, but that will not suit my purpose. If the Harijans vote overwhelmingly for the BSP it might alienate other components of the Bahujan like Muslims and Backwards. I am trying to unite all these social groups.' (40.4) The remainder of this article looks at the extent to which the BSP has succeeded in this aim in the context of its performance in UP in the elections it has contested since 1989. ORIGINS AND DEVELOPMENT OF THE BSPThe origins of the BSP can be traced to the formation in the late 1970s of the Backward and Minority Central Government Employees Federation (BAMCEF). This organisation had started in 1971 as a co-ordinating committee of Dalits based in a government scientific research institute in Pune and the key figure in its leadership was Kanshi Ram. Initially conceived as a loosely structured body putting Dalit government employees in touch with each other, by the late 1970s the organisation had expanded and become more formally structured with a central office in Delhi, regular publications and a programme of activities and campaigns which emphasised the importance of self-activity by Dalits and allied groups [Joshi, 1987; Pai, 1993: 63-7; Mendelsohn and Vicziany, 1998: 218-26]. Kanshi Ram represented a tendency within BAMCEF that advocated giving priority to electoral participation and for this purpose he formed an allied organisation popularly known as DS-4 (Dalit Shoshit Samaj Sangharsh Samiti) in 1981 which contested elections in Haryana in 1982. The launch of the BSP followed in 1984 and it seems that these organisational transformations were accompanied by considerable internal party strife [Joshi, 1987:92; Mendelsohn and Vicziany, 1998: 222]. In the Legislative Assembly (UP state government) elections held in March 1985 the BSP contested the majority of the seats and although its overall performance was not impressive (it polled something like 2.5 per cent of the vote) the results showed that an organisation with the ability to mount a serious electoral challenge had been forged. (40.5) It was also the first evidence that the BSP had successfully been transformed from a body primarily based upon urban government employees to one which could sustain a political and electoral organisation in the hinterland of the Indian DALITS AND POLITICS IN RURAL NORTH INDIA 41 countryside. In December 1985 the BSP performance in the Bijnor Lok Sabha by-election provided the first evidence that the party also had the capacity to influence the outcome of, and possibly win, Parliamentary and Assembly contests. In a high profile campaign with two very well-known Dalit candidates (Meira Kumar and Ram Vilas Paswan) standing for the Congress and the main opposition party, Ms Mayawati, by now the major BSP figure in UP, took nearly one-fifth of the vote and her intervention contributed to a reduction of the Congress majority from 100,000 to just 5,000. (41.6) Mayawati comes from a Jatav/Chamar family which moved to Delhi after living in western UP. The family was involved in Ambedkarite politics but the extent of this activity is not clear. She gained degrees from Meerut University and later a further law degree from Delhi University. Mayawati first met Kanshi Ram when she was a student in 1977 and after working as a teacher she gave up her job in 1984 to become a full-time activist for the BSP. From an early stage her political ability and commitment impressed just about everyone with whom she came into contact. As the former Governor of UP was later to remark: 'Her political antennae are indeed very finely tuned. She has a political sense and uncanny intuitive reactions' [Bhandari, 1998: 172]. Mayawati strengthened her claim to the leadership of Dalit politics in UP in March 1987 when she contested another Lok Sabha by-election in Hardwar and came second to Congress and took nearly one-third of the total vote. In a series of Assembly by-elections held at the same time the BSP on average polled 25 per cent of the vote. Kanshi Ram also entered the election fray in UP when he stood in the highly publicised by-election in Allahabad in June 1988, where most attention focused on the contest between former defence minister V.P. Singh, recently expelled from the Congress, and the Congress candidate. Kanshi Ram polled 72,000 votes in a constituency where the BSP nominee had received only 1,700 votes in the previous election. Whilst these election results provided valuable publicity for the BSP and a great fillip for Dalit activists the party still needed to show it could perform consistently in general elections when its resources were stretched across the countryside in a state-wide campaign. Any assessment of the subsequent overall performance of the BSP during the late 1980s and 1990s is made difficult by the complicated pattern of alliances entered into by the party in Legislative Assembly elections; with the Samajwadi Party in 1993 and with Congress in 1996. The pattern is simpler for elections to the Lok Sabha where the party stood in its own right with just a few minor seat adjustments. As can be seen from Tables 3 and 4 the BSP has enjoyed steadily rising electoral popularity throughout the 1990s but seems to have reached a plateau of support at around 20 per cent of the vote and still 42 THE JOURNAL OF PEASANT STUDIES experiences considerable problems in converting votes into a proportional share of seats. This plateau was probably reached, in terms of general popular support rather than actual votes, around the 1993 election when the small number of candidates standing, as a consequence of the electoral alliance with the Samajwadi Party, reduced its share of the overall vote. The evidence for this general level of support seems to be confirmed by results in the pattern of results for by-elections and for elections to rural local bodies held in May 1994 and April/May 1995 and for elections to urban councils in November 1995 (Panchayati Raj Update, 24 Dec. 1995)
THE BSP AS GOVERNMENT AND
CAMPAIGNER DALITS AND POLITICS IN RURAL NORTH INDIA 43 election campaign it formed an alliance with the Samajwadi Party of Mulayam Singh Yadav and agreed to stand in less than half the seats it had contested in 1991 (see Table 4). It was a tactical move that paid dividends as the alliance came to power as a coalition government. But co-operation was short-lived as acrimony and mutual suspicion increased and eventually in June 1995 Mayawati withdrew from the coalition [Duncan, 1997b]. She then formed a government of the BSP alone, supported from outside by the Bharatiya Janata Party (BJP), that lasted just over four months. In the next elections to the State Assembly held in 1996 the BSP sprang another tactical surprise by forming an alliance with the Congress. Even more remarkable was the agreement of Congress to act as the junior partner by fielding only 126 candidates leaving 296 seats for the BSP. No party could form a government until March 1997 when the BSP once again surprised everyone by forming a coalition government with the BJP, with the added innovation that the office of Chief Minister was to be rotated every six months. Mayawati became Chief Minister for the second time, completed six months in office, handed over to the BJP in September but then withdrew from the government just one month later. In just four years the BSP had been in either electoral alliances or government coalitions with every other major political party in the state. Whilst this record may seem contradictory and opportunistic it is in fact entirely consistent with the view held by Kanshi Ram and Mayawati that all other considerations must be subordinated to the paramount concern that Dalits should take every opportunity to exercise political power. For Kanshi Ram 'expenditure of effort on any object other than the capture of government is seen to be superfluous' [Mendelsohn and Vicziany, 1998: 223] and when in government the BSP relentlessly pursued projects to promote the Dalit identity and presence in public life. To this end in its three periods of government it has initiated such programmes as the installation of thousands of statues of Ambedkar in towns and villages across the state and the creation of a massive commemorative Ambedkar Park in the state capital Lucknow. This project, still uncompleted and the subject of judicial enquiries, was estimated to have cost over f 14 million and Mayawati as Chief Minister ordered that work was to continue round the clock in an effort to complete the park [Bhandari, 1998: 168]. The Mayawati regimes also saw massive transfers of civil servants and police personnel, a common practice in UP, but on a scale that led the then Governor of UP to conclude that there 'was no doubt that officers of the Scheduled Caste had been favoured' [Bhandari, 1998: 170; also Mendelsohn and Vicziany, 1998: 228-9]. There were however other more tangible material benefits as a consequence of the BSP periods of government. Whilst some aspects of the 44 THE JOURNAL OF PEASANT STUDIES government's policy of transfer of officials may have been questionable, in the eyes of many Dalits Mayawati had done them a service by replacing officials who were perceived to have been corrupt [Pai and Singh, 1997: 1358] and she 'dealt particularly severely with officials judged to have failed to protect the most vulnerable people in a particular District' [Mendelsohn and Vicziany, 1998: 229]. The Governor of UP later wrote that on her frequent trips to inspect District administrations it was rare that a visit ended without an officer being sacked or transferred [Bhandari, 1998: 169]. The Mayawati governments also expanded the special welfare and development programme of the 'Ambedkar Villages'. These villages were chosen on the basis of having a high proportion of Dalit inhabitants and received special infrastructural development funds that were usually spent in the Dalit quarters of the villages on public amenities and housing [Pai, 1997: 2314]. Over 25,000 Ambedkar villages in UP were designated and participation in the scheme was often cited by Dalits as a reason for supporting the BSP [Brass, 1997: 2415]. Sometimes land that had been allotted to Dalits many years before, but never actually transferred, was finally handed over during the Mayawati regime [Pai and Singh, 1997: 1358]. Overall there can be no doubt that during the periods of Mayawati's tenure as Chief Minister she pushed forward programmes and projects that were both symbolically and practically relevant to the Dalit population of the state and by so doing contributed further to the construction of a wider and more robust and self-confident Dalit identity across the state. The efforts of the BSP in government were complemented by the mobilising work of a dedicated cadre of activists at the grass roots. The local level activities of the BSP need to be located within the broader context of the general social movements of which the political and electoral campaigning of the BSP were just one part. In particular, if we see social movements to be 'sustained interactions between aggrieved social actors and allies, and opponents and public authorities' [Tarrow, 1994: 33] then India has witnessed a spectacular growth in the social movements of Dalits in recent decades. At present what has become known as 'Ambedkarisation' -- 'the tremendous growth in the consciousness among Dalits about the life and ideas of Ambedkar' [Pai, 1997: 2314] - has been the subject of some studies [Zelliot, 1992; Omvedt, 1994b; Kothari, 1994] but nothing has been produced on the scale of the equivalent work done on rural social movements in, say, Latin America [Foweraker, 1995; Veltmeyer, 1997]. Indeed with very few exceptions [Routledge, 1996] the new social movements approach has received very little attention in relation to India. At the heart of the BSP organisation, and the wider movement of Ambedkarisation, at the grass roots level is a cadre of dedicated local activists who have been characterised as young, politically conscious, better DALITS AND POLITICS IN RURAL NORTH INDIA 45 educated and upwardly mobile [Pai and Singh, 1997: 1359]. Kanchan Chandra on the basis of her observations across seven states during the 1996 election campaign states that the BSP workers at the local level consist mainly of Scheduled Caste and other lower caste activists 'in whom anger against the upper castes and the idealism of social transformation is deeply rooted' [Chandra and Parmar, 1997: 216]. The activist cadre has assumed a crucial role in the leadership of village level politics and act as the intellectuals of their caste group, questioning all aspects of the social order and taking a decisive role in mobilising their castes especially in political matters [Pai and Singh, 1997: 1359]. Party workers for the BSP have proved to be inventive and innovative in their approach to social and political mobilisation. Joshi [1987: 91] reports that BAMCEF had established 'Awakening Squads' who used song, informal drama and poster displays to spread the message of the organisation across the country. V.B. Singh [1992: 250] describes the campaigning of the party in Azamgarh and says that as early as 1989 BSP activists had adopted novel and effective methods, including forming theatre groups aimed at dramatising the position of the Dalits and exposing the oppressive role of the upper castes. The BSP continued the tradition of earlier Ambedkarite organisations by encouraging its followers to use a whole number of rhythmic couplets and chants on public processions and demonstrations. Many of these were deeply offensive to the upper castes and, although they provided huge amusement and enjoyment to the Dalit participants, were hardly likely to assist in the construction of a wider political alliance. (45.8) Processions through public areas were often initiated to coincide with the installation of new Ambedkar statues and the designation of Ambedkar villages and became an important means by which the BSP symbolically asserted the new role of Dalits in the public arena. ELECTORAL ECOLOGY OF THE BSP The electoral performance of the BSP can be analysed from a variety of perspectives and there exist a number of very useful studies based both on survey and aggregate data. By far the most impressive information based on survey data comes from the work conducted by the Centre for the Study of Developing Societies (CSDS) in Delhi. The results of their National Election Survey carried out across the country during the 1996 Lok Sabha election was supplemented by a further study conducted to coincide with Legislative Assembly elections in UP later in the same year. (45.9) From the CSDS Lok Sabha survey results we get strong evidence that the BSP had succeeded in constructing a very solid electoral base among the poor and the Dalits of UP. Their data show almost 40 per cent of the very poor in the 46 THE JOURNAL OF PEASANT STUDIES sample reporting that they voted for the BSP, almost 20 per cent above the proportion of the very poor supporting any other party, and the BSP vote showed a greater skewness towards the poor than any other major party. The survey also found around 65 per cent of the Scheduled Caste population supporting the party with even higher support (74 per cent) among the Jatav/Chamar caste group. (46.10) The CSDS data also reveal significant support for the BSP among some OBC groups. On the other hand the survey suggests that only around 2 per cent of the non-SC population supported the BSP. The figures from the Legislative Assembly election survey show a very similar pattern. These findings are supported by other more locally based surveys and village studies. For example Brass [1997: 2415-16] reports overwhelming support for the BSP in the 1996 Lok Sabha elections on the part of Jatavs/Chamars, but less enthusiastic support from the Pasis in central UP. The study of four villages in western UP by Pai and Singh [1997] completed before and after the 1996 Lok Sabha elections reveals almost total support for the BSP from the Jatavs. My own impression during visits to districts of western UP and Ruhelkand in December 1997 and March 1998, both before and after the 1998 general election, was entirely the same. Indeed it is hard to disagree with Brass that talking to villagers in UP these days about their voting behaviour tends to produce predictable statements, particularly in relation to the electoral preferences of particular caste groups, and no great surprises [Brass, 1997: 2408].
Consistency and Continuity of Support DALITS AND POLITICS IN RURAL NORTH INDIA 47 elections between 1989 and 1996. Table 5 shows the correlations of the share of the total vote gained by the BSP in those constituencies where the party stood candidates in Lok Sabha elections between 1989 and 1998. Table 6 is the equivalent matrix for Legislative Assembly elections between 1989 and 1996. Table 7 considers the BSP overall vote by district in Legislative Assembly elections and includes all constituencies, irrespective of whether the BSP stood candidates or not.
48 THE JOURNAL OF PEASANT STUDIES
The general pattern revealed by these data support the view that the BSP did construct a solid electoral support base from the late 1980s onwards. (48.12) The figures from the Lok Sabha elections, where the BSP had no electoral allies, provide the strongest evidence for this and the figures for state level elections need to. be appreciated in the context of the varied pattern of electoral alliances. In particular the fact that in 1996 the BSP chose to contest in alliance with the Congress and that it increased the number of its candidates from 164 in 1993 to 296 in 1996 needs to be taken into account. Although we consider here only those seats where it stood in all elections, it has to be remembered that in 1993 and 1996 it was adding two quite different support bases to its own core group of voters. It should also be noted that its solid support does not necessarily mean it can retain seats and
DALITS AND POLITICS IN RURAL NORTH INDIA 49 like every other political party in UP over the last ten years the BSP is subject to a high turnover in individual constituencies. In Lok Sabha elections since 1989 it has managed only once to retain a seat won in the previous general election (Akbarpur in 1998); the turnover in LA seats is shown in Table 6, but again this needs to be interpreted in the context of the shifting pattern of election alliances.
Regional Patterns of Support The failure of the BSP to establish a stronger electoral presence in the western districts of the state needs to be looked at more closely. It is particularly puzzling as this region is where movements representing the Ambedkarite tradition have found strong support in the past among the substantial Jatav population. Indeed in the 1960s it was the region of UP where the Republican Party of India had its best successes. Part of the answer is to be found in the way in which Congress and the Janata Dal, despite their general decline in the 1990s, managed to maintain some influence in this part of the state. Other explanations have drawn attention to the poor Organisation of the BSP in the west and Mayawati's role in it and to the continuing influence of the veteran Jatav campaigner and Congress 50 THE JOURNAL OF PEASANT STUDIES activist B.P. Maurya. His long association with the Congress, after starting his political life in the Ambedkarite Republican Party, was given a higher profile when he was made one of the Vice-Presidents of the All India Congress Committee [Mendelsohn and Vicziany, 1998: 225]. (50.16) Whilst these factors may have had some influence there also seem to be some deeper Structural features which could explain the differences in the performance of the BSP in the east and west. The contrasting experiences of scheduled caste agricultural labourers in the east and west of the state has been closely studied by Jens Lerche [1994, 1995, 1998] who challenges the often cited orthodoxy that the economic development experienced in the west, 'dynamic-capitalist', has resulted in a better position for agricultural workers there than those in the eastern 'semi feudal' system. He stresses the need to locate local level struggles within the framework of broader relationships and particularly within the regional balance of power. In particular he points out that the pattern of land reform and caste relations has created a tension between former landlords and smaller landowners in the east and consequently a more fluid context in which agricultural labourers find room for manoeuvre and bargaining. In the west the dominance of established landholders was cemented in place by the pattern of agrarian reforms and subsequent agricultural development. In this region agricultural labourers find limited scope for economic bargaining or political activity within the 'village republics' dominated by the pervasive well organised landholding interests. (50.17) In fact one commentator has gone as far as arguing that agricultural labourers in reality hardly exist in much of the western region at all and those recorded by the census have no stable or permanent relationship with the landed peasant proprietors for whom they work only intermittently [Gupta, 1992: 165].
Lerche provides important insights into the different political economies of rural areas
in the east and west and useful insights into possible reasons for the differential
performance of the BSP. It is also undoubtedly the case that the Scheduled Castes are
locked out of the landholding community in many western areas. Apart from the hill
areas of UP the lowest proportion of scheduled caste landowners are to be found in
some of the Jat dominated districts of the west.
(50.18) There is also evidence that in spite of
impressive economic advances in the west the region remains socially backward and
fares no better than the east in a whole range of available indicators of well being and
social advancement and indeed may well lag behind in terms of child mortality and
gender equality [Dreze and Gazdar, 1998]. The general impression of a
dynamic west and a stagnant east also needs to be appreciated in the light of some
evidence suggesting that the east has in fact experienced considerable advances in
some aspects of agricultural development [Alagh, 1988: 35; Sharma and
Poleman, 1994: 42; Duncan,
Social and Economic Basis of Rural Support
On the basis of the evidence discussed so far it seems that there is a range of issues
concerning the electoral performance of the BSP in rural areas worth investigating by
means of aggregate ecological analysis. First
Second, there is the question of support for the BSP among the SC in general and
among particular caste groups. As survey and village level research indicates, there has
been particularly strong support for the BSP among Jatavs/Chamars, the largest
proportion of this caste group. The other non-Jatav/Chamar SC, as well as the SC in
general are considered here. For the general SC population figures the 1991 census
data are used, for the proportions of individual SC groups only 1961 data are
available, but this is reasonably reliable.
(i) Agricultural labour: As Table 9 shows support for the BSP in rural areas
before 1996 does seem to be associated with the proportion of agricultural labourers
in the workforce, but the association is not apparent in the election of that year. The
grounds for seeing this as an important indicator of BSP support need to be tempered
by the fact that correlations at the regional level do not show consistent patterns. The
regional correlations are not presented here but only in Ruhelkhand, Bundelkhand and
the Lower Doab are consistent patterns of strong association apparent. However when
the SC and non-SC components of the agricultural labour force are considered
separately a quite different pattern emerges with strong and significant correlations of
the BSP vote with SC agricultural labourers but no such relationship with the numbers
of non-SC agricultural labourers. This general pattern also holds across the various
regions of the state when they are considered separately. On the basis of this evidence
there seem to be good grounds to suggest that whatever support there was for the
BSP among agricultural labourers this was coming disproportionately from those
belonging to the SC.
(ii) Small landholders: The correlation matrix for the BSP vote in rural areas and the proportion of smaller landowners in different categories is given in Table 10. There is no need to discuss these results extensively as they point to the same sort of conclusion as those reached by Pai 119931 in her study of the 1989 election. It appears from the evidence of aggregate analysis that there is no discernible association between the level of support for the BSP and the pattern of distribution of landholdings among smaller landowners. Neither do the results for individual regions demonstrate any particular pattern and certainly no evidence for support for the BSP among the landed poor. Older data giving the distribution of landholdings by size for the SC and non-SC separately have also been analysed and they too produce no evidence that the BSP has established any sort of distinctive support base among either the SC or non-SC small landholders. Adjusting the data to allow for the different average sizes of holdings in different parts of the 5tate similarly produces no significant patterns or trends.
54 THE JOURNAL OF PEASANT STUDIES (iii) Caste: The evidence for a strong association between proportion of the SC population and support for the BSP is apparent in the correlation matrix presented in Table 11. However, what is also clear is that these data are heavily influenced by the large proportion of the Jatavs in the SC population. Table 11 also shows that if the Jatav and non-Jatav population are considered separately then support for the BSP among the Scheduled Castes appears to be confined to the Jatavs. Again regional figures are not given in detail but all correlations of Jatavs/Chamar rural population with the BSP vote are high, positive and, with the exception of the west and the lower Doab, significant (at the five per cent level).The regional pattern for support among the non-Jatavs shows broadly a negative correlation with exception of the lower Doab which shows strong support for the BSP among non-Jatav SC. There appears to be very strong evidence that the BSP has certainly built a strong following among the Dalits of UP but that this is restricted to the Jatav caste group with little evidence of general support from other sections of the Scheduled Castes, although it is possible that individual non-Jatav Scheduled castes did support the party.
CONCLUSION DALITS AND POLITICS IN RURAL NORTH INDIA 55 that it has managed to consolidate the Dalits of UP to the extent it would wish and the evidence presented here suggests that like other Ambedkarite movements before, particularly the Republican Party, it relies disproportionately on the support of a single caste group. In the present political conjuncture of Uttar Pradesh, and India more generally, the targeting of particular caste groups for political mobilisation remains an attractive prospect for any party or movement. It is often the case that by emphasising the particular aspects of a caste that it has in common with others and by creating ambiguity and blurring the edges of what separates one caste from another, more generalised identities and political collectivities can be constructed through political action initially based on caste. The recent political history of UP has seen at least one such example, the succession of parties led by Charan Singh, of the construction of a state wide peasant and farmer identity which both utilised and transcended caste identities and at the same time created a collectivity of rural interests of substantially different economic positions [Duncan, 1997a]. The important issue is not so much the political expression of caste identity but what caste can become and how it is perceived as a consequence of its mobilisation in the political arena through the purposeful activities of parties and movements. On the other hand social movements and political parties can become so obliged to emphasise the particularistic characteristics of a caste group, in order to establish themselves or to survive in the world of electoral realities, and the BSP for all its radical and subversive rhetoric is totally committed to institutional politics, that they remain ensnared and trapped within the confines of narrow caste politics. For the moment the BSP is following the second of these alternatives.
NOTES 1. The term 'Dalit' is used by many members of ex-Untouchable caste groups to refer to THEmselves in preference to Harijan or Scheduled Caste or a variety of other terms and for that reason I use it here. It is, however, open to some ambiguity as it is often used to refer to a wider section of the population than those administratively designated as Scheduled Castes. I am using the term here to refer to the ex-Untouchables caste groups and those officially recognised as Scheduled Castes. [Back] 2. For background to Dalit movements see Mendelsohn and Vicziany [1998], Gore [1993], Omvedt [1994a], Mahar [1972]. [Back] 3. For a discussion of these issues at the all-India level see Nadkarni [1997]. [Back] 4. India Today, 15 May 1991 [Back] 5. As the BSP was not recognised as an officially registered party in the 1985 election, its candidates stood as Independents. Estimates vary about its number of candidates and share of the vote. Joshi [1987: 92] reports it stood 335 candidates and won 2.4 per cent of the vote, Pai [1993: 66] claims it won 4 per cent. Vote share throughout the article is taken as the percentage of the total valid vote. All following election data are taken from the relevant 56 THE JOURNAL OF PEASANT STUDIES Election Commission publications or from the website of the Election Commission www.eci.gov.in. [Back] 6. Mendelsohn and Vicziany [1998: 224] are mistaken in their view that Congress won the seat easily. A majority of 5,000 in a constituency with an electorate of 750,000 is a closely election. In 1989 Mayawati won the seat. [Back] 7. For accounts of politics in UP during this time see Hasan [1998], Graff [1998], Duncan [1997b], Wright [1995]. [Back] 8. The Arnbedkarite movement has a long tradition of inventive and often provocative chanting of couplets on demonstrations and processions. In the 1960s supporters of the Republican Party, then the main vehicle of Ambedkarite politics, were sometimes brought to trial after election campaigns charged with seeking to win votes on the basis of caste or religion -- an offence under Indian Election Law - by chanting slogans with a heavy caste content. Some of these cases reached the Allahabad High Court [Government of Uttar Pradesh, 1962-65: 12.5.62, 26,5.62, 15.2.64, 29.2.64, 30.5.64, 12.12.64, 23.1.65.1 and are therefore well documented. One such slogan was Thakur, Brahman aur Lalal Kardo inka munha kola (Thakurs, Brahmans and Banias / make their faces black). The exhortation is to disgrace the high castes (by blackening their faces), and at the same time it is a statement of the Jatav accusation that they are labelled black by the supposedly fairer-skinned higher castes. This is emphasised by the rhyming Lalalkala where kala means black and Lala as well as being a vernacular expression for Banias (usually traders) also means red. Another well known chant at the time Jatav Muslim bhai bhailHindu kaum kehan se aye (Jatavs and Muslims are brothers/Where do these Hindus come from?) reflected the Republican alliance with Muslims at the time and the belief of some Jatavs that they were the original inhabitants of India conquered by invading high castes. The BSP supporters continue this tradition and take up similar themes. Particularly popular with activists at the moment is Tilak, taraju aur talwarllsko maro jute char. The call once again is to disgrace the high caste Brahmans, Banias and Thakiirs by beating them with shoes, but with the added irony that leather working and making shoes is the traditional caste occupation of the Jatav/Chamars. See also Mendelsohn and Vicziany [1998: 223]. [Back] 9. Thanks to CSDS for invitation to attend the workshop on the 1996 NES held December 1997 in Delhi and to everyone attending for their papers, comments, advice and hospitality. My thanks as well to Richard Harris for accommodation and transport arrangements in both December and March 1998 and to Satish Jacob for giving me the opportunity to visit Ruhelkand. [Back] 10. For an account of the methodology of the 1996 CSDS survey see Chandra and Parmar [ 1997: 220-21]. [Back] 11. Brass [1997: 2415, 2419] finds positive and highly significant correlations between the vote share of the BSP in Lok Sabha elections 1989-96. His figures for the correlation coefficients are: 1989-91 .5477 (N = 42); 1989-96 .3000 (N=61); 1991-96 .5100 (N=48). In the following Tables the values presented refer to the simple (Pearson product moment) correlation coefficient which measures the direction and strength of the linear association between two variables. A value of +I would suggest perfect positive association and a value of -1 perfect negative association. The statistical significance (p) of values depends on the size of the sample (N). In all tables if the significance (p) is at the five per cent level or better (that is such results are likely to occur by chance only five times or less in one hundred) the value is marked by *. [Back] 12. The other major northern state where the BSP has built a solid base of support is Madhya Pradesh. Here its electoral performance during the 1990s has shown a similar pattern of consistency though its support is more regionally fragmented than in UP. It is particularly strong in the northern areas of the state bordering UP where the Jatav/Chamar caste group are found in large numbers. In elections for the Lok Sabha it won one seat in 1991 which it retained in 1996 in addition to winning one more. Both these seats however were lost in 1998. Its share of the vote in Lok Sabha elections has also risen steadily (1991 - 3.5 per cent, 1996 - 8.2 per cent, 1999 - 8.7 per cent) Its performance in individual Lok Sabha constituencies where it stood candidates in all three elections shows a highly significant degree of correlation (1991-96 .8145; 1991-98 .8196; 1996-98 .9414; N=18 in all cases). DALITS AND POLITICS IN RURAL NORTH INDIA 57 In elections to the Madhya Pradesh Legislative Assembly held in 1990 and 1993 the BSP gained 3.5 per cent and 7.1 per cent of the overall vote. Two BSP candidates were elected 1990 and both retained their seats in 1993 when the party won an additional eight seats. The correlation of vote share in 176 constituencies were the party stood in both elections is .7544. At the time of writing Legislative Assembly elections are pending in Madhya Pradesh. [Back] 13. These districts are defined on the basis of the average District wide BSP vote in them being at or above the state-wide average vote share (or below it) in all four elections since 1989. Districts boundaries are constantly being re-organised but although there are now 83 districts in UP in most cases data can be re-aggregated relatively simply. As this article forms part of a wider study of UP politics since 1984, Districts boundaries used here are those in place before extensive reorganisation. Allowance is made for all District re-organisations in the data presented. A particular problem occurs with the conventional regional boundaries which in some cases have been breached by the creation of new Districts. Finally it has always seemed to me that Allahabad, though certainly part of the lower Doab region physically, has far more in common, socially and economically, with the eastern Districts and is therefore included with them. [Back] 14. These districts are: Faizabad, Basti (including Sidarthnagar), Gorakhpur (including Maharajganj), Azamgarh (including Mau), Jaunpur, Allahabad, Banda, Hamirpur, Jalaun. [Back] 15. These districts are: Bareilly, Shahjahanpur, Lucknow, Rae Bareli, Gonda, Farrukhabad, Mathura, Bulandshahr, Meerut (including Ghaziabad), Muzaffarnagar, Dehra Dun. [Back] 16. B.P. Maurya certainly claims that his influence in the area is still strong and is proud of the fact that he thinks he has kept the BSP at bay. He defected from the Congress before the 1998 general election to join the BJP and maintains that this resulted in many Jatavs voting for the BJP for the first time in that election (Interview, B.P. Maurya, 20 March 1998). [Back] 17. The village republic system of the western areas was epitomised by the powerful influence exercised by the Bharatiya Kisan Union in recent years, see Bentall and Corbridge [1996], Corbridge [1997], Gupta [1988], Gupta [1992], Hasan [1989], Singh [1992]. [Back] 18. Over the state as a whole the SC form 18.4 per cent of the landholding cultivators, but in Muzaffarnagar only 4.7 per cent, Meerut 4.3 per cent, Ghaziabad 5.7 per cent and Saharanpur 9.2 per cent [Government of India, 1994]. [Back] 19. 1 have not discassed here the efforts of the BSP to recruit Muslim support which it ccriainly has seen as part of the Bahujan collectivity. The party at one time appeared to be in a strong position among the Muslim population and repeating the success of the Republican Party in the 1960s in recruiting substantial Muslim support and giving official party positions and candidatures to Muslims. This support and involvement by Muslims is now dissipated, partly as a consequence of their experiences within the party and finally as a result of the electoral alliance of the BSP with the Congress in 1996. The Congress Party is seen by UP Muslims as holding a major responsibility for the events at Ayodhya in December 1992 [Duncan, 1997b: 987-92]. [Back]
REFERENCES
Back to the top. | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| American Political
Science Review Vol. 93, No 4 (December 1999), p. 999 Quest for Power: Oppositional Movements and Post-Congress Politics in Uttar Pradesh. By Zoya Hasan. New Delhi: Oxford University Press, 1998. 292 p. $24.95. Leela Fernandes, Rutgers University Contemporary politics in India poses a unique case for the study of economic and political transitions in comparative contexts. In contrast to cases in regions such as Eastern Europe and Latin America, which have been characterizedby both the economic transition from state planning to liberalized economies and the political transition of democratization, India's nascent economic reforms have been initiated within the political context of a parliamentary democracy. In the Indian context, political transition appears to reside in growing democratic instability with the successive fall of fragile central governments and the rise of exclusivist Hindu nationalist political forces. Zoya Hasan's Quest for Power is a compelling analysis of contemporary Indian politics and suggests that such instability represents a structural political transition as India shifts away from a national system dominated by the Congress Party in the postindependence period. The appearance of political disorder, as the book demonstrates, conceals a deeper process of grassroots political democratization within the context of a formal democracy. The central argument of Quest for Power is that India's political transition must be understood as a structural shift in the relationship between state and civil society. Hasan argues that the failure of the state to change the material and social conditions of marginalized sectors of society and the exclusion of these groups from the Congress Party's centralized power structure led to the independent political mobilization of new social constituencies. According to Hasan, this mobilization has placed historically marginalized groups at the center of the political system at both the local and national level and has led to the decline of the Congress Party. Drawing on archival research, government reports, qualitative interviews, and secondary sources, the book examines contemporary politics in the northern state of Uttar Pradesh, which has played a critical role in shaping both national political discourses and electoral politics. Hasan presents a comprehensive analysis of the declining political fortunes of the Congress Party in this state and the rise of three critical social movements that have effectively challenged it: a farmers movement, mobilization of "backward castes," and the Hindutva (Hindu nationalist) movement. This focus moves away from the traditional assumption that these social groups are merely passive votebanks manipulated by political leaders. Hasan demonstrates that the decline of the Congress Party stemmed in large part from its inability to accommodate newsocial groups demanding a voice in both the local and national political system. The unwillingness of the local Congress organization to broaden its upper-caste base to include the socially mobile backward castes (p. 38) and the Congress Party's "pragmatic communalism" (p. 235) provided the political space for both the rise of regional parties and the emergence of the Bharatiya Janata Party (BJP) as a powerful national force. Hasan analyzes the social factors and local political processes that have allowed new regional political parties in Uttar Pradesh, such as the Samajwadi Party (SP) and the Bahujan Samaj Party (BSP), to displace the Congress and claim to represent the interests of social groups, such as backward castes and Muslim constituencies. The significance of the book is underlined by recent political events. Both the BSP and the SP played a critical role in the recent fall of the BJP coalition government. In 1999, the BSP's last-minute, unanticipated vote against the BJP led to the fall, and the SP's refusal to support the Congress Party prevented the return of a Congress-led government. Hasan examines the socioeconomic and historical processes that have led to this regionalization of Indian politics, which enables local parties to affect substantially the direction of national electoral politics. The emphasis on the significance of social mobilization and the effects of expanded political participation lend an important and timely perspective to current debates on India's "crisis of govern ability" (see Atul Kohli, ed., Democracy and Discontent: India's Growing Crisis of Governability, 1994). Hasan's argument recasts this debate by suggesting that what Book Reviews: COMPARATIVE POLITICS December 1999, p.1000 appears as a crisis of democratic instability is in fact a case of growing democratic political participation. India's crisis of governability, ironically, reflects a process that is firmly linked to the electoral politics of parliamentary democracy. A particular strength of Quest for Power is its refusal toromanticize the political activism of the peasant and caste based social groups. Hasan examines the complex social and economic hierarchies that led to the marginalization and subsequent mobilization of these groups, but she also points to the limits of such mobilization. For instance, she argues that it has restructured the political world but has not had a similar effect on class-based structural inequality. On the contrary, it primarily benefits backward caste groups that have already acquired economic clout (p. 163). Hasan also argues that rural mobilization of farmers in Uttar Pradesh was "class driven" (p. 97), as the leadership was oriented toward the rich and middle peasantry (for a contrasting perspective, see Ashustosh Varshney, Democracy, Development and the Countryside, 1995). The book's analysis of shifting social relations and its departure from purely statist approaches hold potential insights for understanding the ways in which grassroots social movements affect national political transitions in comparative contexts. Yet, because the book does not have a generalized theoretical framework or discussion, it will be of interest mainly to scholars of contemporary Indian politics. In addition, the emphasis on political processes within civil society at points results in an overestimatation of the "retreat of the state" and the corresponding decline of the Nehruvian project (p. 3) in the past two decades. The current halting pace of economic reforms, for instance, demonstrates the need for a qualification of this potential retreat. More important, the state still remains a major employer. In fact, instances of political mobilization examined in the book often involve caste-based claims for reservations in government employment (p. 152). Such demands tend to represent an extension or a consequence of rather than a clear decline in the Nehruvian model, which transformed the state into the primary source of secure jobs in postcolonial India. On the whole, Quest for Power provides an important and convincing explanation of the decline in the Congress- dominated political system. A short review cannot adequately capture the nuanced empirical analysis of the central trends now shaping Indian politics. This rich analysis makes the book worthwhile reading for anyone interested in gaining a deeper understanding of the increasingly complicated permutations of national politics in contemporary India. [END] Back to the top. XX Ideas and Writings on South Asia
Webbed by Philip McEldowney |