all towns are one, all men our kin.
|Home||Whats New||Trans State Nation||One World||Unfolding Consciousness||Comments||Search|
Tamils - a Nation without a State
The Changing Politics of Tamil Nadu in the 1990s
John Harriss and Andrew Wyatt
In May 2004 the Dravida Munnetra Kazhagam (DMK) led Democratic People�s Alliance won all 40 seats in Tamilnadu and Pondicherry in the Lok Sabha elections. The DMK itself won 15 seats, the Congress ten, the Pattali Makkal Katchi (PMK) six, the Marumalarchi Dravida Munnetra Kazhagam (MDMK) four, the Communist Party of India and the CPI(M) two each and the Indian Union Muslim League one. The Frontline report on the election results commented that �A tidal wave of people�s anger routed the AIADMK [All India Anna Dravida Munnetra Kazhagam] and its ally, the Bharatiya Janata Party�.
The result is a striking indication of the change that has taken place in the politics of Tamilnadu over the seventeen years since the death of M G Ramachandran [MGR] (who dominated Tamil politics like no other political leader before or since), and since the publication of David Washbrook�s discussion of the politics of the state in the Frankel and Rao collection of studies (Washbrook 1989). It is not that the DMK should have defeated the AIADMK � which had won a stunning victory in the last state assembly elections only three years before, in 2001 � that is significant, for the two major Dravidian parties have fairly regularly alternated in office for many years.
Tamilnadu has had its share of electoral �waves�, and the 2004 election verdict may fairly be seen as yet another of these (there being evidence of widespread dissatisfaction with the performance of the AIADMK government in the state). What is significant, however, and what is emphatically different from the past � at least up till 1998 - is that the DMK should have shared its success with so many other parties, and that M.Karunanidhi, the long-standing leader of the DMK, should have commented to reporters that �I will not say that the DMK�s strength is behind the victory. It is the alliance�s strength�.
This was a remarkably candid admission of the fact that alliance or coalition politics are by now well established in Tamilnadu. State politics are no longer dominated so absolutely by the two major Dravidian parties. Between them, they do still command the major share of the popular vote (each about 30 per cent), but whereas in the past smaller parties could exercise very little influence, they can do so now.
Tamilnadu now has a multi-party system, that includes the BJP and a number of significant caste-based parties (notably the PMK). These are all new features in the politics of the state. Our purpose in this paper is briefly to describe what has happened, to probe into what underlies the changes that have taken place in the party system, and finally to reflect upon their implications.
The history of Tamil politics may be seen now in terms of three phases of political mobilisation. Congress electoral mobilisation gave way to Dravidian populist mobilisation in the 1960s. This has been extraordinarily successful, and it has made the politics of Tamilnadu quite exceptional. But for whatever reasons � different writers have reached different conclusions on this matter � the last decade has seen the decline of Dravidianism. The most recent, third phase of political mobilisation in the state may be described as �fragmentary mobilisation�. What do these trends connote?
From a �Two and a Half Party� System to Multi-party Politics
One of the cardinal facts about the politics of Tamilnadu � that marks it out from almost every other major state - is that the Congress lost power in the state in 1967, to the DMK, and has never regained it subsequently. The DMK�s return to power in the state elections in 1971 was assisted by the split in the Congress; and thereafter, following the split in the DMK itself, and the formation by MGR in 1972 of the AIADMK, the two Dravidian parties have slugged it out with each other in the Assembly, in the towns and villages of Tamilnadu, and in successive elections, with the Congress only ever in a supporting role. It was a �two and a half party system�, in which the Congress, as the �half� party, largely conceded influence in state politics whilst securing for itself a substantial share of the seats in Tamilnadu in Lok Sabha elections, in alliance with the DMK (1980) or more usually the AIADMK (1984, 1989, 1991.
Throughout the time whilst MGR was alive, from the elections of 1977, such was the hold of the puratchi talaivar (�revolutionary leader�) or vadyar (�teacher� � the more commonly used nickname for him) on the imagination of ordinary Tamilians that the AIADMK retained office in the state over three successive elections (1977, 1980 and 1984).
The party was then split over the succession to MGR after his death in 1987, between his wife Janaki, and his sometime co-star and mistress Jayalalitha. In these circumstances the DMK was able to return to power in 1989 with a thumping majority, only to lose it, disastrously (the party was reduced to holding only two seats in the Assembly), in elections in 1991, following the assassination of Rajiv Gandhi.
The pattern was only broken following the formation of the breakaway Tamil Maanila Congress (TMC) by G K Moopanar in 1996, in protest � he said - against the decision of the Congress High Command to ally with the AIADMK which he considered to be corrupt (though Moopanar�s decision probably stemmed from the frustrations that he and other Congress leaders felt at the minor role they were able to play in state politics).
Up till this time (for a period of twenty years) the Congress had been the balancing party in the state party system, and smaller parties had little chance of making any significant impact. Thus the PMK (formed in 1989 by S.Ramadoss) and MDMK (formed by V.Gopalsamy - �Vaiko� - in 1993 breaking away from the DMK against what Vaiko argued was the undue influence within it of members of the Karunandhi family and their betrayal of the ideals of Dravidianism) contested the 1996 elections, but the larger parties felt confident enough to ignore them. Their campaigning strategies and coalitions had no place for these smaller parties. The DMK and the TMC together won sweeping victories in both Lok Sabha and state Assembly elections in 1996 (the AIADMK was shattered in the latter on the scale that the DMK had endured five years before).
By 1998, however, thanks to the split in the Congress in Tamilnadu, and to the decline in the strength and influence of the party nationally, it ceased to hold the balance between the Dravidian parties. A number of new parties were drawn into the competing coalitions, and were able to exercise influence as they had not done before.
In the Lok Sabha elections of 1998 the TMC, still in alliance with the DMK, suffered a humiliating defeat, while in the same elections the success of the AIADMK owed a good deal to an alliance with the Vanniyar party of S Ramadoss, the PMK, that had come to command a good deal of influence in the northern part of the state where the Vanniyars are numerous. We may date the emergence of a multi-party political system in Tamilnadu to those elections.
One set of factors underlying the change in the party system of Tamilnadu is what may be termed the federal incentive structure. The fundamental element of this incentive structure is the attractiveness of controlling the state government - state governments having extensive powers that politicians are anxious to seize (Manor 1995: 48-51). The redrawing of the state boundaries along linguistic lines from the 1950s gave the federal system greater stability and reduced the significance of language as a divisive issue (Adeney 2003:191); while the reorganisation of the states also created some units in which caste-based competition became a more obvious way of mobilizing support (Chhibber 1999: 56-57).
India�s size and diversity makes organising a bid to take national power extremely difficult. This, along with the range of powers available to the state assembly, makes regionally based parties attractive to aspiring political entrepreneurs. The primary focus of DMK and AIADMK leaders between the late 1960s and the late 1980s, therefore, was capturing and holding on to power in Tamil Nadu and an accommodation with the Congress Party supported this objective. The dominance of Congress in New Delhi combined with the strong central powers granted by the Constitution convinced Dravidian politicians that the centre-state relationship was potentially coercive.
The threat of central intervention, indeed, encouraged the DMK to follow a moderate course once in power (Kohli 1991:161, 162 & 165). The main objective of the Congress leadership in New Delhi, meanwhile, was to maximize the number of MPs returned to the Lok Sabha from Tamil Nadu. This stifled new political initiatives as the political market in the state was closed: the Dravidian parties controlled the state government and Congress dominated the national representation of the state. Electoral alliances that included Congress and one of the Dravidian parties meant smaller parties were more or less irrelevant at election time.
While a good deal of Tamil politics continues to be regionally oriented - the ongoing dispute with Karnataka over water sharing is one marker of this - the interaction between Tamil Nadu and the centre has changed in significant ways since 1989. The 1989 Lok Sabha election was a watershed, the defeat of the Congress reflecting the decline of the Congress Party organisation that had been more or less evident for some time (Hewitt: 1989). In retrospect the result can be seen as having marked the beginning of serious electoral decline for Congress, the party not having won a majority at the national level in any ensuing election. The federal incentive structure has changed in consequence. Cabinet posts at the centre became available to either the DMK or the AIADMK if they wished to support a national coalition government; while from 1998 onwards it became clear that Congress was unable to deliver the votes in the state that would secure the margin of victory for its alliance partner. Other small parties could perform this function, however, and they could also get MPs elected and cabinet posts for themselves at the centre.
Leaders of smaller parties were no longer doomed to a life of political obscurity � for even if they could not rule at the state level they could increase their representation in the national parliament and the state assembly. The leaders of national parties now had to take them seriously and they won grudging respect from the previously imperious Dravidian parties.
The arrival of multi-party politics in Tamilnadu was very clearly demonstrated in the 2001 elections to the state Assembly. Of these elections the Economic and Political Weekly�s correspondent wrote:
At the time he was writing the alignments of significant political parties in Tamilnadu had been thrown up in the air following the decision of the PMK to quit the National Democratic Alliance, in which it had stood alongside the DMK since the Lok Sabha elections of 1999, and to ally itself with the AIADMK. Shortly afterwards further confusion was added when another participant in the NDA, the MDMK, parted company with the DMK, even whilst remaining within the alliance at the national level.
Finally, the two alliances that confronted each other in the polls in 2001 included strange bedfellows. The DMK, the inheritor of the (supposedly) rationalist, secular and socially radical tradition of the Dravidian Movement was allied (as it had been since 1999) with the Hindu nationalist BJP, and with several newly formed caste-based parties.
Hardly less strange was that Moopanar�s TMC was in alliance with the AIADMK, of which he had previously been so strident a critic, and with the rump Congress party.
Almost as strange was the fact that both the communist parties should have fallen over themselves in their eagerness to support the so-called �Secular Front� headed by the ADMK, and that they should have been so ready to welcome into the Secular Front an explicitly caste-based party such as the PMK.
Two Dalit parties, Dr K.Krishnasamy�s Puthiya Tamizhagam, strong amongst Pallars in the southern districts of the state, and R.Thirumavalavan�s Dalit Panthers, ranged against the Vanniyars in the northern districts, were drawn, meanwhile, into the NDA-linked alliance given the participation of the PMK in the Secular Front, and the support for the AIADMK amongst the locally dominant Thevars in the south.
Never before had Tamilnadu witnessed such complicated alliance politics.
In the event Jayalalitha had got the alliance arithmetic right, and Karunanidhi had got it wrong, and in spite of polling evidence which showed a fairly high degree of satisfaction with the DMK government that had been elected in 1996, it was ejected from office. Frontline reported that
Yogendra Yadav reported findings from CSDS surveys that showed that had the two major Dravidian parties gone it alone they would each have got more votes, and �More important �that the DMK commands a greater vote-share than the AIADMK�. There was no �wave effect� in Tamilnadu in the 2001 elections, therefore; there was no anti-incumbency vote (though both have occurred on other occasions). The outcome was explained by the alliance arithmetic.
In the Lok Sabha elections of 2004, however, the DMK won a famous victory, not so much in its own right but � like the AIADMK in 2001 � because it got the coalition arithmetic right, as Karunandhi conceded in the remarks we quoted earlier. Now the PMK, the MDMK, the recently re-united Congress (following Moopanar�s death), and the communist parties stood alongside the DMK, and against the AIADMK-BJP combine (the Dalit parties forming, on this occasion, a third front called the People�s Alliance).
Patterns of Political Mobilisation Under the Ascendancy of the Dravidian Parties
Here we want to explain the pattern of politics in Tamilnadu that has been broken down during the 1990s. There is disagreement amongst scholars about the implications of the shifting ideologies of the Dravidian Movement, but substantial agreement, on the other hand, about its extraordinary success in terms of political mobilisation. Though the ideology of Dravidianism originally focused on ethnicity � the identity of the �Dravidian� Shudra against the �Aryan�, �North Indian� Brahmin � it rapidly became much more of a populist discourse, of a plebeian stamp, emphasising the notion of the common (Tamil) man (Barnett 1976: chapter 7). The genius of Annadurai and of others was in their ability to create and to communicate a mytho-history, celebrating the great achievements of the Tamilians, that had meaning for ordinary people in a way that the �scientific�, developmental project of the Nehruvian state did not.
The emerging Tamil political elite was extraordinarily adept in building precisely that �common thinker we-ness �and a single political language� that, according to Kaviraj, the elite of the Nehruvian state neglected (Kaviraj 1991). No matter what its policy achievements � and they were considerable � or its success in maintaining support amongst the �big men� of the Tamil country, Congress gradually lost out through the 1950s and 1960s to the world of meaning created by the DMK, as well as to its increasing organisational strength. The Dravidian parties were adept, Subramanian argues, at building �inclusive sub-cultures� amongst the intermediate and the lower social strata, �within which were linked caste and religious groups that might otherwise have come into conflict with each other.
The DMK built such sub-cultures in the northern plans and the Kaveri valley from the 1950s to the early 1970s, and the AIADMK in the rest of Tamilnadu (the western and southern plains, and the Tamirapani valley) in the 1970s and 1980s�(Subramanian 2002: 126). These were social groups that had been largely ignored by the Congress, which built support �on thin foundations � large-scale industrialisation, pan-Indian nationalism and the aggregation of already organised interests�; and also by the communists who were identified primarily with the propertyless (Subramanian 1999: 47ff).
The cultural nationalist agenda of the Dravidian parties, and its moral claims for social justice for the common people (to be achieved by modest redistribution, or �sharing� through welfare programmes rather than by changing the distribution of assets), was immensely successful in the context of a society which, as Washbrook argued �The possibilities of class politics �have been reduced by the extent to which capitalism has promoted the almost infinite �petit bourgeoisification� of interests� (1989: 220).
Writing as we do at the time of the US Presidential Election of 2004, in which polls have shown that the issue that concerned the greatest proportion of the electorate was �moral values� we are tempted to suggest certain similarities between political mobilization around cultural issues in Tamilnadu under the Dravidian parties and that encouraged by the contemporary Republican party in the US. Of this Slavoj Zizek has written: �although the ruling class disagrees with the populist moral agenda [which creates �an opposition between honest, hard-working Christian Americans on the one hand, and decadent latte-drinking liberals who drive foreign cars, mock patriotism and advocate abortion and homosexuality on the other�] it tolerates it as a means to keep the lower classes in check. The �moral war� allows the lower classes to articulate their fury without disturbing dominant economic interests.
Culture war, in other words, is class war by other means: the one is a displacement of the other�. The kind of �culture war� that went on in Tamilnadu, driven by the Dravidian parties, has also allowed �the lower classes to articulate their fury without disturbing dominant economic interests�. The economically dominant class fractions in Tamilnadu have thus been able very largely to stand apart from politics.
During the heyday of the Dravidian parties, political allegiances in Tamilnadu were not strictly defined by social divisions, or in other words cleavages were not fully exploited by the DMK and the AIADMK. The two main parties devised strategies to appeal to certain social groups but there was no sense in which these cleavages were stable or had primordial roots. The DMK and the AIADMK were unable to claim exclusive loyalty among the social groups they targeted. Though each party identified itself with a broad social category the populist styles of the DMK and the AIADMK meant that they were also catch-all parties. Their supporters could be found on either side of key social divides.
Language, caste, gender and class are obvious ways in which political opinion could be divided in the state, but the dominance of the Dravidian parties ensured that no single line of division became significant - at least until the changes of the recent past, with the emergence of a multi-party system. With regard to language, the distinction between Tamil and non-Tamil speakers was of course politicized very effectively by the DMK in the 1960s (Barnett: 1976). Tamil speakers form a massive majority in the state � 85% of the population claim the language as their mother tongue (Swamy 1996b: 192). But this very asymmetry makes it difficult to translate language into a permanent division around which partisan loyalty could be built. It has been impossible for one party to take possession of the language issue over a long period of time. The DMK could not hold a monopoly of Tamil speaking opinion and after 1972 the AIADMK confirmed that the language issue had ceased to be the sole property of one party.
Caste in Tamilnadu is a more obvious source of social division that might be exploited by political parties and it has had an impact on politics, no doubt� but without providing a stable cleavage structure around which a party system could be built. Entrepreneurial political leaders have at different times opened up and closed down the issue of caste. One reason why it has been difficult to use caste as the basis of a stable cleavage structure is that the state has no numerous dominant castes such as characterise some other major states (like Okkaligas and Lingayats in Karnataka); and the basic units of the caste system, the individual jatis, are never more than a significant minority in a constituency.
This means that even if all the members of the largest caste group vote together for a single party, which is too strong an assumption to hold true, they are likely to be defeated by another party that has assembled a cross-caste coalition. Caste-based parties are not ineffective, however, in circumstances where a party system is fragmented. Contemporary Uttar Pradesh offers a variation on this theme. In some constituencies in the state the winning party might only need to win a little over 20 per cent of the vote to secure victory. Parties based around a single caste group are unlikely to reach the winning threshold.
However some of the larger caste groups have been used to provide the core support for parties that present modified appeals to caste sentiment. The Samajwadi Party appeals to a coalition of Muslim and backward caste voters. The Bahujan Samaj Party gets strong support from Chamar voters but has worked hard to extend its appeal to other groups. Many of these fall under the �Scheduled Caste� label but candidate selection has been used to broaden the coalition of voters. . Caste-based parties can succeed in a fragmented party system where electoral alliances are routinely formed to contest elections � as has been the case in Tamilnadu, as we have seen, since 1998.
Another reason why caste has not emerged as the defining cleavage in the Tamilnadu party system is that there are at least four ways in which caste might be used to constitute a cleavage, and one form of caste-based mobilization can potentially be countered by another interpretation of the caste issue.
The first dividing line that was used to organise party politics was the Brahmin-non-Brahmin divide, but the non-Brahmin category proved too amorphous to become the basis of an enduring cleavage.
The political success of certain �forward� caste groups in the Justice Party was resented by other �backward� caste groups � and this represented a second caste-based division along which the party system might have been organized. Members of the backward castes successfully pressed the government of the Madras Presidency to attend to the needs of the �backward classes� in the 1940s. The DMK exploited this backward caste resentment but it was destined to become more than just a backward caste party (Barnett 1976: 115).
A third axis that has intermittently operated as a cleavage is that of the divisions between individual caste groups. The large Vanniyar caste group, which comprises approximately 12 per cent of the population of the current state of Tamil Nadu, has been particularly amenable to this type of mobilization - Vanniyar parties were briefly successful in the 1950s and the PMK has proved remarkably durable since 1989.
The dividing lines between individual backward castes and between the �backward� castes and the scheduled castes create divisions that are salient in the everyday experience of the majority of the population, and this makes broader categories, such as that of the backward castes, difficult to invoke as the basis of political action.
The fourth division that has been exploited periodically has been the dividing line between those inside and those outside (the Dalits) the caste system. This cleavage was used again in the 1990s as the DPI and the PT claimed to represent the interests of the Dalits in the state.
As with the issue of language, non-Brahminism and links to the backward castes formed elements of Dravidian mobilization against Congress but the DMK did not convert this division into a political cleavage that could be sustained over a long period of time.
Instead, as Subramanian argues, the DMK used a number of overlapping categories, including caste and language, to create a sense of a Tamil community distinct from the elite that dominated state politics in the 1950s.
The DMK �sought the valorisation of plebeian norms, rather than the policing of ethnic boundaries� (Subramanian 2002: 128). Dravidian populism thus displaced overt appeals to caste as a method of mobilization and the Dravidian parties succeeded in cultivating partisan support that crossed social boundaries and transcended particular caste identities. The very public incantation of particular caste identities that marks contemporary state politics was redundant while Tamil cultural nationalism was in the ascendent.
With regard to class, the populist strategies devised by the two Dravidian parties were predicated � as we have argued - on ideas of difference but they did not generate categories that could be easily converted into conventional modes of cleavage-based mobilization. The populist strategies of the two parties have been analysed in terms of their attempts to target different socio-economic groups: the DMK is generally thought to have linked its fortunes to intermediate and moderately well off groups, and the AIADMK to have made a much more explicit attempt to target the poor with its paternalist approach.
Opinion poll data gathered in 1991 by the Department of Statistics at Madras Christian College can be used to explore the extent to which support for the two parties was socially differentiated (Swamy 1998: 128-129). These data did indeed reflect a rural bias towards the AIADMK, and could broadly be construed as providing evidence of a proto-class-based cleavage.
What is especially striking, however, about these data is that they show that support for both parties was socially mixed. The AIADMK had the decisive edge among women and landless labourers but it also had a lead over the DMK among most categories of wealthier voters (Swamy 1998: 129). This suggests that lingering partisan loyalty among a minority of poorer voters sustained the DMK through its lean years during the rule of MGR and the AIADMK.
In his study of Madurai in the late 1980s Kohli concluded that �(n)either caste nor class can provide a ready basis for aggregation of political interests in contemporary Madurai� (1991: 179). Trade unions in the city, though largely class rather than caste-based, were divided by loyalties to different parties (Kohli 1991: 177-178). The lack of substantive class-based mobilization is one consequence of successful populist mobilisation. Swamy describes the party competition favoured by the Dravidian parties �as arising not from parties grounded in identifiable social groups or ideologies, but through the competition between broad rhetorical strategies� (Swamy1996a, cited in Harriss 2000: 336). The affective links based on identities cultivated by the DMK and the AIADMK cut across class lines and are powerful motivators of political action.
Finally, let us consider gender. Swamy has outlined in some detail how the populist strategy employed by the AIADMK under the leadership of MGR crosscut other cleavages by opening up a gender gap between the parties. Opinion polls and data gathered from gendered segregated polling booths in 1989 demonstrated the AIADMK lead among women voters (Swamy1996b: 202, 205, 207-208).
Policies designed to appeal to women included the temporary prohibition of alcohol. The Chief Minister�s Nutritious Noon Meal Scheme was a high profile initiative that established the party�s welfarist reputation. Initially targeted at poor school children the scheme was extended to include other indigent groups including widows. Jayalalitha continued to develop this aspect of the AIADMK�s constituency when she assumed leadership of the party in 1989 (Widlund 2000); and the DMK has not been able to close the gender gap, as the CSDS poll conducted before the 2001 assembly elections showed. The AIADMK was found to enjoy a nine percentage point lead over the DMK among female respondents who expressed a preference (Frontline June 8, 2002).
Following her return to power in 2001 Jayalalitha took care to back a number of policies to sustain the AIADMK tradition of taking care of women. Though the substantive contribution of these policies might be called into question the gendered differences between the DMK and the AIADMK as �parties in the electorate� are well documented.
Historically, therefore, identity or cleavage based politics have been transcended by the populist mobilisations of the Dravidian parties. But this is changing quite rapidly and identity based politics are now becoming much more important in Tamilnadu. The forms of identity politics are not always new.
For example, the Vanniyar mobilization of the 1980s and the formation of the PMK had a precedent in the form of the Commonweal Party and the Tamilnad Toilers of the 1950s (Rudolph and Rudolph 1967: 56-58). The political significance of cleavages has not been constant over time. Elite political actors are sometimes able to exploit social divisions and re-align the cleavages that shape a party system.
It is important, however, not to overstate the efficacy of such elite agency and to recognise, for instance, that the recent electoral history of the Tamilnadu is littered with examples of failed political entrepreneurs who have attempted to exploit the obvious lines of social division in the state and build cleavage-based parties. But the structural circumstances of the 1990s � that we described above - have enabled political entrepreneurship, and in favourable circumstances some party leaders have been able to achieve a great deal, and in the process they have changed the party system.
The Decline of the Dravidian Parties, and the Rise of Caste Politics and of Hindu Nationalism
Change in the party system at the national level brought about changes in the incentive structure in state politics, as we noted, and new opportunities in the party system in Tamilnadu have been successfully exploited by those, notably S Ramadoss of the PMK and Vaiko of the MDMK, whom we may describe as effective political entrepreneurs.
Another aspect of the changes that have taken place in Tamil politics, part cause perhaps, and part effect, has been the decline of the two Dravidian parties and of the party political system, dominated by these parties, that we have just described. Yogendra Yadav noted of the 2001 elections that they had �brought into focus a process that has been going on in the state for well over a decade [i.e for a period that extends back beyond the watershed in national politics of 1989]. The Dravidian parties are slowly losing their capacity for cross-sectional mobilisation. They can no longer meet the various sub-regional and sectional aspirations that have found political articulation in the form of small parties�.
Narendra Subramanian amplifies the point, noting that the combined vote share of parties that have emerged since 1989 rose to 25 per cent in 2001, and arguing that: �These new political forces emerged in response to the diminished social presence of the Dravidian parties [our emphasis, AW and JH]�These parties ceased to be associated with distinctive political visions, and their links with activists and supporters as well as their contributions to civic life weakened� (2002: 138). One marker of this has been the decline in political participation in Tamilnadu, shown up in declining turnouts at elections.
What accounts for the diminished social presence of the Dravidian parties? Part of the explanation, no doubt, has to do with the changes in the incentive structure for political leaders that we described above, and the changed context in which Tamil politics now operates.
But it is hard not to think that the Dravidian parties have played an important part themselves in the weakening of their position in state politics.
One set of factors has to do with the political economy of the state. Washbrook was surely prescient when he wrote that
In short, the Government of India, in the context of the economic reforms of the 1990s, has indeed ceased to be willing or able to subvent the DMKs� populist policies, resources for redistribution (�sharing�) have dried up, and the stability of the state�s political system has, effectively, disintegrated.
The AIADMK governments led by Jayalalitha, which have continued to pursue what Subramanian describes as �paternalist populism� (as distinct from the �affirmative populism� of the DMK), have been those most seriously affected by new resource constraints.
The present difficulties of the Jayalalitha government, and the �tidal wave of anger� towards it reflected in the Lok Sabha elections in 2004, are the outcomes of its inability to meet the expectations of different groups of actual and potential supporters. As the Frontline correspondent noted: �
A long list of factors worked against the AIADMK-BJP alliance. �[including] anaemic implementation of mid-day meal scheme for children [pretty much the spearhead of paternalist populism under MGR in the 1980s]; the stopping of the supply of eggs to children under the scheme; the termination of service of about 10,000 road workers�the denial of bus passes to a section of students�..Government employees and teachers turned against the party after it ruthlessly put down a strike by them. The government reduced the retirement benefits of government employees and teachers. Transport workers lost several thousands of rupees when their monetary benefits were compounded. The government shut down many single-teacher schools �The Jayalalitha government antagonised every section of society�. But such are the budget constraints that the government labours under that it had very little choice on many of these matters.
Secondly, though Subramanian has been critical of the arguments of M S S Pandian and S Ananthi about the decline of the social radicalism of the earlier Dravidian Movement, and those of Geetha and Rajadurai about what they consider to have been its �ideological regression� (Geetha and Rajadurai 1996), he too argues that the Dravidian parties �ceased to be associated with distinctive political visions�.
It is neither the adoption nor the subsequent abandonment of Tamil nationalism, or the declining emphasis on the issues of language and state autonomy that matter so much here, but rather the loss of the social vision of Dravidianism.
M S S Pandian has consistently argued that for all its twists and turns and political opportunism the Dravidian Movement was successful in infusing a sense of self-respect amongst Non-Brahmins in the state. But he argues that uneven developments across caste groups created serious tensions.
�The materially more advanced sections of the non-Brahmins, such as the Chettiar elite, who during their economic ascendancy endorsed and funded the movement, found it no longer to be of any great relevance� (1994: 221), while on the other hand Dalits and groups like the Vanniyars have fallen away because their aspirations have not been met by the Dravidian parties. �The most important and obvious reason for this tragic political shift is the growing power and arrogance of the Backward Class elites at the local level, which has often translated itself into Adi-Dravida [Dalit] violence� (1994: 221) � which has then often been ignored by the leaders of the Dravidian parties.
Their failure historically to deal with the issues surrounding untouchability is the most important marker of the loss of the radical social vision expressed, though erratically, by E.V. Ramasami Naicker (�Periar�) � the inspirational leader of the Dravidian Movement in the 1930s and 1940s. It is unsurprising, in the circumstances, that political movements should have arisen amongst Dalits in Tamilnadu, or that the two parties, the Dalit Panthers of India and Puthiya Tamizhagam, should have been formed � though, of course, it is also important to note that these two parties are organised around different Dalit caste groups (Paraiyars and Pallars, respectively).
Latterly Karunanidhi, as Chief Minister, apparently prevaricated in his response to the police attack on Dalits in Tirunelveli in July 1999, presumably because he wanted to try to win the support of the local dominant Thevars over from the AIADMK � exactly as Jayalalitha had done also, in similar circumstances in 1995. The MDMK leader Vaiko is himself a representative of the elite Backward Classes, coming from a big landlord household. Pandian notes finally that �with their recently acquired economic strength [the elite sections of the Nadars, Kallars, Gounders and others] are finding it possible to express a pan-Indian desire which is partly reflected in their drift to the Hindu right� (1994: 223).
This �drift to the Hindu right� is the other significant aspect of the decline of Dravidianism. Subramanian is emphatic, however, that �Hindu nationalism was impeded for long in Tamilnadu not by the ideology of the early Dravidian Movement, but by the DMK�s construction of cohesive partisan subcultures incorporating networks linking various caste and religious groups� (2002: 131).
Consequently the increasing disengagement of the DMK from its links with different support groups in society and the declining cohesion of its subcultures since the 1980s, has left room for the growth of Hindu nationalism.
But this is not the only part of the story, for as Subramanian also argues, the AIADMK has always been more open to Sanskritic culture and to upper caste mores (reflected in the fact, noted earlier, that it has usually won stronger support than has the DMK amongst upper caste people), and it has not promoted cooperation across religious boundaries. The AIADMK government of 1991-96 provided considerable state support for the Hindu religion, which was then continued by the DMK in office.
Only shortly before the 2001 elections Karunanidhi was reported in The Hindu as having boasted at a conference of village temple priests about the increased number of temple dedications under his DMK government since 1996. Of course, Hindu religious observance, involvement in temple construction and participation in public religious performances such as those at Vinayaka Chaturthi which were initially encouraged in Tamilnadu by the Hindu Munnani (Fuller 2001), do not in themselves equate with support for Hindu nationalism.
But there is evidence that at different social levels and amongst different social groups the new religiosity builds up a sub-stratum of ideas that disposes people towards Hindu nationalist ideas (Harriss 2002). The BJP still has little electoral purchase in Tamilnadu, but there is no doubting that there is widespread sympathy for Hindu nationalist ideas in different sections of Tamil society.