Electoral Politics in the Time of Change: India's Third Electoral System, 1989-99 Author(S): Yogendra Yadav Source: Economic and Political Weekly, Vol

Home , Lok Shakti, Yadav (surname)

Electoral Politics in the Time of Change: India's Third Electoral System, 1989-99 Author(s): Yogendra Yadav Source: Economic and Political Weekly, Vol. 34, No. 34/35 (Aug. 21 - Sep. 3, 1999), pp. 2393- 2399 Published by: Economic and Political Weekly Stable URL: http://www.jstor.org/stable/4408334 . Accessed: 09/01/2014 20:07

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp

. JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].

Economic and Political Weekly is collaborating with JSTOR to digitize, preserve and extend access to Economic and Political Weekly.

http://www.jstor.org

This content downloaded from 128.59.62.83 on Thu, 9 Jan 2014 20:07:04 PM All use subject to JSTOR Terms and Conditions Electoral Politics in the Time of Change India's Third Electoral System, 1989-99 Yogendra Yadav

The popular reading of the last decade of electoral politics is of a rapid decline and impending collapse of Indian democracy. This essay attempts to contest this all too familiar view from above without yielding to the temptation of building its mirror-immage. The changing outcome of the electoral game in this decade is seen here as the reflection of a fundamental transformation in the terrain of politics which in turn is anchored in the process of social change. The voters choose differently in this decade, for, the structure of choice inherent in elections has undergone a change. Yet the choice often turns out to be far from the act of sovereignty that the fiction of liberal democracy makes it out to be. AN attemptto understandthe politics of Interpretingmodem politics is a demand- gone a change.The size, the composition electoral choice in contemporaryIndia ing activity in the best of times. But the andthe self-definition of-those who choose bringsus face to face with a tensionthat cognitive demandsplaced on the inter- havechanged. More and more citizen from lies at the heartof India's currentdemo- preter of contemporaryIndian politics the lower rungs of society participatein cratic transition,if not at the heart of appearexcessive even by that standard. this ritualof choosing, they come not as democratic theory itself. The current For one thing, the received conceptual individualbut as groups,and they bring decaderepresents the full unfoldingof the frames of various hues were never de- withthem their own tastes and worldviews. contradictionbetween the logic of politi- signedto graspthe specificityof the path They havemuch more to choosefrom and cal equalityand that of social inequality, of democracy in a poor, non-western they exercise their right to reject very something that Ambedkarhad warned society. Besides, there was no way the frequenctly,as the site of choicehas moved againstin the Consititutionassembly it- feeble intellectualattention given to the closer home. Yet the choice turnsout to self. The dynamicsof political equality taskof collectingrelevant information and be farfrom the act of sovereigntythat the triggeredoff by theinstitution of universal makingsense of it in Indiacould have met fiction of liberaldemocracy makes it out adultfranchise and the self-reproductive the scale andthe paceof eventsin the last to be. The voterscan choose froma given processes of the structureof socio-eco- decade. set, but they cannotdetermine which set nomic inequalityinherited from the past The most commonreading of the last to choose from.They can elect someone have both crossed a certain threshold decade of electoral politics, in its aca- and then throw himself out, but there is withoutbeing able to tamethe othercom- demic and popularversions, is a storyof preciouslittle that they can do to ensure pletely. the rapiddecline and the impendingcol- thatthe representativesactually represent As theparticipatory upsurge leads to the lapse of Indiandemocracy. It finds signs them once in power. downward of the of election to spread democracy, widespread fatigueleading THIRDELECTORAL SYSTEM political processes begin to disturbthe populardisinterest in and cynicism to- inheritedritual social hierarchy and trouble wardspolitics. The last decadestands out The period since 1989 is best charac- theestablished elite with hitherto unknown for suddenoutburst of some of the mala- terisedas a new electoralsystem, the third anxieties.Since elections are associated dies inherentin our system:the endemic one since the inaugurationof democratic withmuch of this upheaval,the spectacle multiplicationin the numberof political elections in 1952. Normallythe expres- of election has come to occupy an aura parties and the fractionalisationof the sion 'electoralsystem' is usedin the limited of 'samudramanthan', the grand ritual politicalspace; the rise of regionalparties sense of the formalrules thatspecify the churningof the social sea by the small and caste-communitybased partiesthat procedure(first-past-the-post or the pro- gods of democracywith the help of the threatento unleashfissiparous tendencies portionalsystem, for example)followed modem instrumentof a vote. But just as anda clashof primordialloyalties; end of in elections. In that sense therehas been this churningseems poised to yield the ideology-basedpolitics and the declineof no change in our electoral system. But proverbial'amrita', it comes up against politicalmorality; and, of course,exces- politicalhistorians have also used the same therocky bed of economicinequality. The sive politicalcorruption, non-governance, expressionto capturea certainconstella- mayaof collectiveaction and the invisible disorderand instability.To sum up: our tion or patternof the outcome and the 'rakshasa'calledIdeology conspire against politics is one big mess. determinants of electoral choice that all the smallgods. The churningdoes not The principleambition of thisoverview characterisea certainperiod of history.l stop, the spectaclegoes on. Yet at some of the electoralpolitics of the last decade In this substantivepolitical sense, we can level it turnsinto a harmlessgame, subtly is to contestthis all too familiarview from talk of a new electoralsystem when we delinkedfrom some of the gravestthreats above withoutyielding to the temptation detecta destabilisationof andits replace- it posedto thelords who inheritedthe sea. of buildingits mirror-image.The chang- ment by a new patternof electoralout- It is hardlysurprising that the enactment ing outcomeof the electoralgame in this comes as well as its determinants. of this storythrough a series of elections, decadeis seen here as the reflectionof a In this substantivesense the first four all in the course of 10 years or so, has fundamentaltransformation in the terrain generalelections from 1952 to 1967 fall stumpedanalysts of Indianpolitics, often of politicswhich in turnis anchoredin the underthe first election system. The one deceiving them into a surface reading processof socialchange. The voters choose partydominance of the Congressmeant of the entire process as no more than a differentlyin thisdecade, for the structure that the elections in this periodwere not horse-race of political entreprenures. of choice inherentin electionshas under- seriouslycompetitive in thisperiod marked

Economic and Political Weekly August 21-28, 1999 2393

This content downloaded from 128.59.62.83 on Thu, 9 Jan 2014 20:07:04 PM All use subject to JSTOR Terms and Conditions by a low level of electoralparticipation. ideologies was adaptedto suit popular politics:Mandal, Mandir and Market. The The choice was betweenthe omnipresent taste.2 almost simultaneousand sudden occur- Congressand its regionally fragmented On the face of it, the 1989 electoral rance of these three events - the imple- opposition;often the opposition came from verdict appearedno different from the mentationof the MandalCommission's within the Congress.Electoral loyalties earlierwave electionsof the secondelec- recommendationsfor OBC reservations, werefixed at the nationallevel unless the toral system.In many ways thatelection the BJP's rathyatrathat catapultedthe constituencylevel preferences dictated indeedbelonged to the earlierperiod. The Babarimasjid dispute into national promi- short-termdeviations from it. The voter rise of V P Singh had galvanised the nence and the forex crisis leadingto the of course did not vote as an individual, opposition to the Congress. The anti- implementationof the first phaseof IMF butrather as a memberof politicised'jati'. Congresswave in northIndia followed the sponsoredpackage of 'liberalisation'- Nextto thecandidates' party, theirjati had same logic of oppositionunity (captured createdan extraordinaryopportunity for some effect on the voting behaviour.In so wellby DavidButler and Prannoy Roy' s reworkingthe establishedpolitical align- social terms the castes that enjoyed the index of oppostion unity) as the 1977 ments.All the threeoffered the possiblity benefits of early entry into modem edu- wave. The Congressand the opposition of creatinga new cleavagethat cut across cation or early politicisationthrough the tried theirold social alliances.Yet there theestablished cleavage structure and thus nationalmovement or bothdominated the weresigns of thenew orderyet to be born. engaging in a new kind of political list of elected representatives.The 1967 Underthe umbrellaof JanataDal, state- mobilisation.Eventually, not all the three electionhad already signalled a transition, wide political formations had already cleavages could be activatedin politics, for the monopolyof the Congressand the begun to excrcise a significant role in at least not in the same degree. But the savarnajatis was challengedfor the first nationalpolitics. Devi Laland Biju Patnaik simultaneityof this change did resultin timein northIndia. The process had started were alreadymajor political players. The a transitionof the electoral system and much earlierin the south. electionresult also carriedsigns of shift- allowedseveral latent forces to surfacein Although Indira Gandhi's unprec- ing socialbasis of politics:Muslims in UP electoral politics. edented electoral victory in the 1971 anddalits and OBCs in Biharhad already The 1991verdict finally inaugurated the election was initiallyseen as the restora- begunto upsetthe given social equations new system. The earlierlogic of regime tion of the Congress dominance,in ret- forpolitical parties. In thatsense the 1989 alterationand thatof Congressvictory in rospectthat election looks like a begin- electionlies at the cuspof the secondand the context of dividedopposition clearly ning of the second electoralsystem. The the thirdelectoral system. indicateda massive victoryfor the Con- apparentcontinuity of the Congresswas The decisive stimulusfor changecame gress in 1991, a repetitionof the 1980 deceptive;the Congress that Indira Gandhi between the 1989 and the 1991 in what wave.But it didnot happen.The Congress led to powerin 1971 was a new partythat was christenedas the threeMs of Indian improvedonly marginally,not enoughto hadto negotiatea new terrainof electoral TABLE1:VOTE SHARE OF MAJOR FORMATIONS INLOK SABHA 1989-1998 politics.The move towardsthe new sys- ELECTIONS, tem was triggeredoff by the first demo- 1989 1991 1996 1998 cratic upsurge in the late 1960s. The a new en- Congress 39.5 36.6 28.8 25.9 upsurgebrought great many Congress allies 0.3 2.1 0.9 0.3 trantsfrom the 'middle' castes orthe OBCs BJP 11.5 20.0 20.3 25.5 into the game of electoral politics and BJP allies - 0.8 4.6 10.6 turnedit trulycompetitive. Congress was Left Front 10.2 9.7 9.1 7.8 no longer the single dominantparty but Janata family 18.7 15.1 11.4 11.2 BSP 2.1 1.6 3.6 4.7 throughoutthe 1970s and 1980s it con- Independents 5.3 3.9 6.3 2.4 tinuedto the naturalparty of governance, Other 23.9 10.2 15.0 11.6 the pole aroundwhich electoralcompe- tition was organised.The success or the Notes: Congess allies: 1989 = UDF partners. 1991, 1996 = UDF partners,AIADMK. 1998 = UDF BJP allies: 1991 = Shiv Sena. 1996 = Shiv Sena. 1998 = failureof the attemptsby the 'opposition' partners. Samata, HVP, Samata, to a unitedfront the Con- HVP, Shiv Sena, Akali Dal (Badal), Lok Shakti, AIADMK, MDMK, PMK, TDP, BJD, put up against TrinamulCongress, LoktantrikCongress. Left Front = CPIM, CPI, FBL, RSP in 1989, 1991, gress made a decisive difference to the 1996 and 1998. Janatafamily: 1989 =JD, JP. 1991 = JD, JP(S). 1996 = JD, SP. 1998 = JD, electoraloutcome. Elections turnedinto SP, RJD. plebiscites where the effective unit of Source: CSDS Data Unit. choice was the entire political nation, TABLE 2: NUMBER OF PARTIES: TOTAL AND EFFECTIVEAT NATIONAL, STATE AND CONSTITUENCY sometimessplit along the north-south lines. LEVEL,1952-1998 A typicalelectoral verdict in this period took the formof a nation-wideor some- Election year 52 57 62 67 71 77 80 84 89 91 96 98 times state-widewave for or againstthe Total no of parties in Congress. The local specificities of a Lok Sabha 21 13 21 19 25 19 18 22 25 25 29 40 constituencysimply did not matter.These Effective No of parties electoral waves flattenedthe terrainof [National] 4.1 3.5 4.2 4.7 4.4 3.4 4.2 3.9 4.7 5.1 6.9 6.9 electoral little room Effective no of parties competition,leaving [sverage of all states] 3.1 3.4 3.4 3.6 3.0 3.5 3.2 3.0 2.5 3.0 2.7 3.0 for local variations.In social termsit was Effective no of parties [average a period of cross-sectionalmobilisation of all constituencies] 3.3 3.0 2.8 2.9 2.4 2.1 2.7 2.4 2.5 2.9 3.0 2.7 via state-widejati alliances like KHAM in or AJGAR in UttarPradesh. Note: All the calculation of the effective number of political parties in this table is based on the Gujarat vote share of the parties in various Lok Sabha elections and uses the Taageperaand Shugart Ideologically,it was an era of populism, (1989) formula. as the borrowedframework of western Source: Heath 1999: 66-71.

2394 Economic and Political Weekly August 21-28, 1999

This content downloaded from 128.59.62.83 on Thu, 9 Jan 2014 20:07:04 PM All use subject to JSTOR Terms and Conditions have a clear majority.But for the addi- theCongress is no longerthe natural party some adivasis.The BJPand its allies were tionalvotes broughtin by RajivGandhi's of governance.In fact thereis good evi- the highestvote getternot only amongthe assassination,it would have remainedat dence(see Heathand Yadav in this issue) upper class Hindus but also among the the same level as in 1989. The BJP's that the Congress vote in the 1990s is OBCs as a bloc. In political-ideological success in UP, Gujaratand Karnataka definedby its opponent.Congress picks terms,the partyhas expandedto win the defied all the established patterns of up residualvotes afterits opponentshave confidence of various allies who have political sociology and political geogra- targetedand mobiliseda certainsection. little patiencewith its Hindutvaideology. phy. The state partiesdid not look like a The votefor the Left Front in WestBengal In this respect the formationof the Na- flash in the pan: they were here to stay. and Keralahad long ceased to be merely tional DemocraticAlliance, the alliance Clearlywe werein a new politicalera. The an anti-Congressvote; it is more mean- with DMK and PMKin TamilNadu and near majority that Congress mustered ingful to see the Congressvote in West with splinterJD in Bihar and Karnataka allowed it to postponethe arrivalof the Bengal as anti-LeftFront vote. In Delhi, on the eve of 1999 elections symbolises by now familiarconsequenes for the new Rajasthanand Gujaratthe Congresshas the extentto whichthe BJPhas been able system by anotherfive years. The full been replacedby the BJP as the natural to eraseits statusof a politicaluntouchable implicationswere to unfold in the 1996 party of governance.In Bihar, UP and in the aftermathof the demolitionof the andthe 1998Lok Sabhaelections. But the Tamil Nadu any talk of Congress/anti- Babarimasjid. outlinesof the new system were clear in Congressvote is boundto invite ridicule The thirdspace - the alternativeoutside the series of assembly elections held today.Unlike the BJP, the Congressis yet the Congressand the BJP fold - has seen between 1993 and 1995.3 to accept theimparatives of coalition majorups and downs in this period.Yet Electoraloutcomes in the subsequent politics and alliance building. The way some developmentsare hereto stay. The elections have establishedbeyond doubt threeof its strongunits (Maharashtra, West rise of the BSP as a recognisednational that we now live in a differentera. But Bengaland Tamil Nadu) broke away from party deserves more notice than it has thereis no need to recountall the changes the parentparty and does not augurwell received.Afterall, it is the firsttruly new here. The rise and fall in the fortuneof for the futureof the party.The Congress party- whose genealogycannot be traced politicalparties and leaders in this period is there to stay but the Congressera in backto the partiesat the time of indepen- is too obvious to be missed out even by politics is behind us. dence- thathas managedto crossthe high a casualspectator. What is requiredhow- The rise of the BJP to power and a thresholdof viabilityunder our first-past- ever is an accountof what has changed, consistentrise in its vote sharewith every the-postsystem. Most othernew entrants notby way of eventsand personalities but electionsince 1984has been accompanied have managedto cross the thresholdonly withreference to the processesand struc- by a three-dimensionalexpansion in the at the state level. In this context the rise turesthat bind the day-to-dayroutine of 1990s (Olliver Heath in this issue). In of state partieshas changedthe natureof the electoralrace. For an obsession with geographicterms it has expandedmuch political alternatives. Parties like the the rapidly swinging fortunes in the beyond its northIndian, Hindi heartland SamajvadiParty, Samata Party, Rashtriya politicalgames can turn our attention away core to includeGujarat and Maharashtra Janata Dal, Nationalist Congress Party from the silent yet fundamentalchanges in its core areas.More importantlyit has and arguablyMamata's Trinamool Con- in the groundrules of this game.Also, we developed substantial presence in gress or the TMC in Tamil Nadu are not need to ask whetherthese fundamental Karnataka,Andhra Pradesh, Orissa and cast in the same mould as the classical changes have resulted in a meaningful Biharand a footholdin West Bengal and regionalparties like the Akalis or the DMK realisationofthe promise ofelectoral demo- TamilNadu. In socialterms it is no longer or the TDP. Their political presence is cracy, the promiseof self-governance. an urban bania-brahminparty. It has state specific but their political vision is a formidablerural base, ex- not. Each of these would out of their POSTCONGRESS POLITY developed go tending well into the lower OBCs and way to claimthat they are a nationalparty. The most obviousand easily identified characteristicof the thirdelectoral TABLE3: AGGREGATEVOLATILITY BETWEEN LOK SABHAELECTIONS, 1977-1998 system (Per cent) is the changein the choice set available to the voter. Table 1 summarises the Election years 1977-1980 1980-1984 1984-1989 1989-1991 1991-1996 1996-1998 changes in the vote share for the major Aggregate volatility 23.8 21.5 16.1 11.8 12.9 14.7 political formationssince 1989. A con- tinuousdecline in the vote sharefor the Notes: Aggregate volatility is calculated by measuring the--'swing' between a party's vote share all the in election standsout as between two elections. The value reported in the table is the sum of the swing for Congress every parties divided by two. Where new parties have emerged, or split, the entire vote share of the definingfeature of this distributionof the new partyis taken as the swing. In this calculation Only partiesthat gained either at least vote shares. Clearly the changes are so one seat or at least 1% of the national vote are included in the swing calculations. basic that we arejustified in talkingof a Independentshave been excluded. Only parties that gained either at least one seat or at least new partysystem. In the firsttwo systems I per cent of the national vote are included in the swing calculations. Independents have underthe erathe voter exercised been excluded. Congress Source: CSDS Data Unit. only one choice, whetherto vote for or againstthe Congress. That is no longerthe TABLE4: TOTALSEATS RETENTION BETWEEN LOK SABHA ELECIONS, 1977-98 case, andnot just becausethere are many 1980 1984 1989 1991 1996 1998 non-Congressalternatives. Congress is no longer the pole against which every Seat 154 338 200 301 264 263 politicalformation is defined.In thissense Per cent 28 62 37 55 49 48 we are in a now. post-Congresspolity Notes: The table reports the number and percentage of the seats retained by all the parties in any Evenin thosestates where there is a direct Lok Sabha election race betweenthe Congressand its rival, Source: CSDS Data Unit,

Economic and Political Weekly August 21-28, 1999 2395

This content downloaded from 128.59.62.83 on Thu, 9 Jan 2014 20:07:04 PM All use subject to JSTOR Terms and Conditions In some sense the Left parties belong to Mahasangha in Maharashtra, to name a composition of Lok Sabha has not changed the same category. They are the only few. The eventual outcome has either been very much, an extraordinaryproportion of formation not to have undergone a basic a complete failure (even in those areas seats have changed hands between two change in its geographical or regional where these movements and their leaders elections. In the last two elections more profile in the last decade, though have lost have for a long time enjoyed mass sup- than half of the seats have changed hands, some votes, are more confined to their port) or a gradual absorption after initial a very high figure by any standard.Table 5 regional pockets-andare a little more lower success. Clearly the present system has confirms this impression by examining class- and caste-based than before. The little room for micro political action to the 'hard core' evidence of survey panel. third space has emerged in this decade as translate into effective political alterna- It shows that all the parties, including the spring of political alternatives. While tive. The failure of any third force to those in ascendancy lose anything beween its failure to consolidate into an all-India develop downward linkages with 35 to 50 percent of their voters across two third formation has greatly reduced its say grassroots movements and upward link- elections, another unusually high figure. in national politics, this failure has also ages with each other in a national frame- A third way to examine the volatility is kept open a diversity of political choices work largely accounts for an the absence to look at how the incumbentgovernements within the mainstream of national politics. of genuine and effective political alterna- fared in the various assembly elections tive in the Third Electoral since 1989. Table 6 shows that the failure POLITICALFRAGMENTATION? System. Does the voter use the vote more effec- rate of the incumbent governements at the If the electoral choices available to the tively? If electoral volatility is any mea- hustings is as high as 77 per cent. The voter are to be judged by the number of sure of the effective use of vote in mass figure touches 82 per cent if we look political parties that seriously contest the elections, a simple look at the data may exclude the small states and Uts where elections and can put up viable candidates, suggest that the voters are less decisive competitive politics is sometimes yet to there is some expansion of the choices now than they were before. The era of take off. In other words, barring a few available in the thirdelectoral system. The wave elections witnessed massive swings exceptions, all the state governments have numberof political partiesthat have gained between two elections. The 1990s have faced defeat at the polls in this period. And entry to the Lok Sabha has gone up dra- not witnessed anything like that. Table 3 those who did not were governments that matically in the last Lok Sabha. This does quantifies this impression and shows that did perform in one respect or the other. not indicate their real strength. There is if we sum up all the swings for various Clearly, whatever the choice set avail- little evidence to supportthe popularbelief parties, there has been a decline in it in able to the voters, the election outcome that this figure reflects a fragmentation of the 1990s. The last three elections have of the last decade indicates that they use the party system or a mad proliferation of not seen anything of the kind of swing it quite effectively and often in a very parties. Yet if we calculate the number of witnessed in the previous three elections. discerning manner. parties by their effective share of votes or The figures for the last two elections would The voters have also been effective in seats, we find a marginal increase. But that have been lower than the table depits, if bringing about a noticeable change in the is a function of artificial aggregation at the the artificial effect of the formation of new composition of the political elite. Politics national level. If we look at the number parties or party splits was taken out of the of presence or representational democ- of effective parties at the state level or at calculation. racy is the most immediate form of asser- the constituency level, we find basically But as in many other cases, the aggre- tion by the newly enfranchised commu- a bipolar competition. Duverger's law that gate figures conceal more than they re- nities. To those already enfranchised, this postulates the emergence of two-party veal. As Tables 4, 5 and 6 show quite is the time to go for the higher forms of system in a first-past-the-post electoral conclusively, the elections of the 1990s political representation. To be sure this procedure does work in India too, except have seen a very high degree of volatility. change is very uneven and holds mainly that it works at the state level. Instead of Table 4 shows that although the overall for numerically large and electorally producing a simple bipolarity at the na- TABLE5: VOTERETENTION BYPARTIES/ALLIANCES BETWEEN LOK SABHA ELECTIONS, 1991-1998 tional level, it has resulted into 'multiple bipolarities' (Sridharan nd). It does in- INC+ BJP+ NF/UF LF BSP dicate an in choice, but not of expansion 1991-96 53 65 53 68 56 the kind suggested by the first impression 1996-98 53 63 38 61 41 of the national figure of the number of political parties. Note: Table entries refer to the percentage of those who voted for a party in the previous elections who continued to vote in the next one. The data includes The data Notwithstanding the new entrants, our only panel respondents. on 1991 is based on recall in 1996 pre-poll survey. system continues to maintain high barriers Source: NES 96, NES 98 for newcomers, especially for those that operate at the level below that of a state. TABLE6: INCUMBENTPARTY'S PERFORMANCE INASSEMBLY ELECTIONS. 1989-1999 In practice all the peoples movements that Total Number Information Incumbent Won Incumbent Defeated stand for attempts to build political alter- States Available on native operate at a local level and find it No No No Per Cent No Per Cent impossible to cross the minimum thresh- Major states 35 33 6 18 27 82 old of viability. The last decade has seen Minor states/UTs 28 24 7 29 17 71 various experiments by the erstwhile non- All states 63 57 13 23 44 77 party political formations to register their in the elctoral arena: Note: Major states include all the states with 10 or more Lok Sabha seats. The rest, including Uts presence Chhattisgarh with elected assemblies are treated here as minor states/UT. Those cases where it was Mukti Morcha, KamatakaRajya Rayyata difficult to determine who was the incumbent or who really won the election have been Sangha,UTJAS in northBengal, Samata excluded as No Information. Sangathanin Orissaand Bihar,Bahujan Source: CSDS Data Unit

2396 Economic and Political Weekly August 21-28, 1999

This content downloaded from 128.59.62.83 on Thu, 9 Jan 2014 20:07:04 PM All use subject to JSTOR Terms and Conditions decisive communities. As such the OBCs encies groupedby theirdegree of urban- rawnarratives of socialjustice articulated arethe biggest gainerin the race for political ity. We know thatin the first two decade by a KanshiRam or a LalooPrasad Yadav parties to accommodate them in their of electoral democracy,urban areas re- achievedwhat Lohia's sophisticatedphi- organisationaland legislative wings. There corded higher turnout. Now the rural losophyof historyfailed to do threedecades is no appreciable increase in the presence constituencieshave overtakenthe urban ago, namely,to makeit respectableto talk of women in the legislatures, nor have the constituenciesin electoralturnout. aboutcaste in the public-politicaldomain. Muslims gained much from it. Yet times As shudraparticipation has increasedin The emergence of 'social justice' as a of political turbulence, close electoral the last ten years, the socially and the rubricto talk aboutcaste equity,political contests and frequent elections bring economicallyprivileged sections of soci- representaionof castes and communities numerous gains to marginal groups. A ety have recordeddecreasing levels of and issues of communitarianself-respect large number of lower OBC castes have politicalparticipation. An urban,educated, and identityis a distinct achievementof had their first MLA or the first MP in this uppercaste citizen is far more likely to this period. decade. It is in this decade that India had be non-participantand cynical about Yet theparticipatory upsurge has not led two prime ministers from the south and matters political than his counterpart to anythinglike an effectivecontrol of the the first dalit woman as the chief minister. amongthe rural,uneducated, lower caste. lower orderson the issues or the agenda Even a look at the DEMOCRATIC UPSURGE cursory participation TABLE 8: ODDS-RATIO FOR VOTING: 1971 AND profilein otherdemocracies is enoughto 1996 AND1998 If there is so much turbulence at the level demonstratethat this trend is very un- 1971 1996 1998 of electoral outcomes, one of the funda- usual. The textbookrule about political mental reason for it is that the participa- participationis thatthe higheryou are in Community tory base of electoral democracy is ex- socialhierarchy, the greaterthe chanceof Hindu Upper 1.11 0.90 0.97 panding in the 1990s. Once again, aggre- your participatingin politically activity Hindu OBC 0.82 1.07 0.94 can be here. The India is SC 1.04 1.22 1.21 gate figures deceptive includingvoting. Contemporary ST 0.65 0.91 0.95 upsurge does not always show up in any perhapsthe only exception to this rule Muslim 1.59 0.92 1.12 dramatic jump in the overall turnout fig- among functioningdemocracies in the Sikh 1.53 0.86 1.60 ures and that is why analysts have tended worldtoday. The participatoryupsurge of Christian 2.29 1.13 1.05 not to notice it. Table 7 presents these the shudrasis the definingcharacteristic Economic status overall turnout figures. But now there is of the ThirdElectoral System. The con- Very Poor 0.89 1.24 0.92 sufficient evidence to conclude that the tinuousinflux of from Poor 0.98 1.13 1.05 peopleincreasingly Middle 1.14 0.94 1.18 electoral arena in this has the lower ordersof in the arena political period society Upper middle 1.06 0.89 0.96 witnessed greater participation and more of democraticcontestation provides the Upper 1.38 0.75 0.75 intense politicisation than before.4 This setting,the stimuliand the limits to how Education tendency is particularlystrong among the the election system unfolds. Illiterate 0.82 1.03 0.91 shudras, used here as a Up to middle 1.49 1.01 1.33 generic category CREOLISATIONOFDEMOCRACY for all the social includ- College 1.29 1.05 0.91 marginal groups Graduate 1.07 0.70 0.66 ing women.5 The dominant peasant pro- The influx of lower orders into the field prietorOBCs were alreadyfairly politicised of democraticcontestation has left its Notes: The variable for constructing the in the second system, but now there is an impressions on the vocabulary of this economic status of the respondentswas extension of this trend to the lower OBCs for the new entrants derived from their type of occupation contestation, brought and amount of land and in 1971 as well. Table 8 the odds ratio withthem their beliefs as well. Forthe first owned, presents and 1996, family's monthly income and for voting for different caste, community timethe neat arrangement of theborrowed in 1998 the type accommodation. or class groups.The odds that a dalit will high ideologicalspectrum was disturbed Source:National Election Study [NES] 1971 vote are much higher today than that of by homespunideological fragments. The and NES 1996 and NES 1998 an upper caste. This has been accompanied by a significant rise in their sense of TABLE7: TURNOUTINLOK SABHA AND MAJOR ASSEMBLY ELECTIONS, 1952-1998 efficacy and their involvement in more active forms of This Lok Sabha elections Major State Assembly Elections political participation. Year Turnout (Per Cent) Year No of States Turnout (Per Cent) is perhaps due to the fact that it was only in this decade that a sizeable chunk of dalit 1952 45.7 1952 22 46 electors were able to actually exercise 1957 47.7 1957 13 48 their for the first 1962 55.4 1960-62 15 58 voting right time, 1967 61.3 1967 20 61 especailly in northIndia. The huge turnout 1971 55.3 1971-72 21 60 gap between the adivasi and the non- 1977 60.4 1977-78 24 59 adivasi voter has been bridged in the last 1980 57.2 1979-80 16 54 three elections, though the adivasi citizen 1985 64.1 1984-85 18 58 is yet to catch up in the active forms of 1989 61.9 1989-90 18 60 Table 9 looks 1991 55.9 1991 - - political participation. 1996 57.9 1993-96 25 67 closely at the women's turnout to show 1998 62.1 1998 4 63 that the proportion of women among the voters has gone up after stagnating for Note: A 'major' round of assembly elections is defined here as one which involved, within a year two decades. Their or two, elections to at least 2000 assembly constituencies. The figures for the last column nearly proportion do not conform to this definition. Calculations include for state among the politically active citizen has provisional figures assembly elections held in 1995, the report for which has still not been released by the Election also registereda major leap. Table 10 Commission. looks at the turnoutfor differentconstitu- Source: CSDS Data Unit.

Economic and Political Weekly August 21-28, 1999 2397

This content downloaded from 128.59.62.83 on Thu, 9 Jan 2014 20:07:04 PM All use subject to JSTOR Terms and Conditions of elections.The very rawnarratives that vara like jati groupingsas the primary test betweentwo catch-allparties, UP and ensuredthe presenceof social justice as social bloc for politicalmobilisation. The Biharshow evidence of extremejati/vama a theme also allowed for the marginal- emergenceof uppercaste or upper/lower polarisation.If religiouscleavage accounts isationand containmentof this concern. OBCor dalitas moreor less homogenous for voting differentials in Kerala and The homespunideological fragmentsof categoryin some northIndian states is a Punjab,it is class in Delhi and Bengal. thisperiod were based on verysmall visions pointerin thisdirection. The same process If the Congressis a partyof the down- thatfailed to develop linkageswith other can be seen at workin the emergenceof troddenwhen it faces theBJP in thenorth- largerissues. A supportfor caste based numericallylarge castes by combining west, it represents privileges where it reservations,for instance, did not translate morethan one traditionaljatis: Kshatriyas confrontsthe Left Front.The BJP com- into any concrete stance on the Babari in Gujarat,Maratha-Kunabi in Maha- bines its uppercaste votes with different masjid dispute or on liberalisationand rashtra,Ahirs in Bihar, Rajbanshisin social groupsin differentstates. The erst- globalisation.This failure was most marked Bengalor Bhandarisin Goa arebut a few while JanataDal representedthree radi- in therealm of theeconomic policy which exampleof the shiftingdefinition of the cally differentsocial groupsin the three was effectivelywithdrawn from the menu primarysocial unititself. To be sure,this states where it mattered:cross-sectional availableto the voter. developmentcannot be dated back only supportwith accent on the uppercastes This was made possible in large mea- to the last ten years.Many of these com- in Orissa,dominant peasant proprietors in sure by the bifurcationof the electorate munitieshad startedevolving much ear- Karataka anddalit-lower OBC in Bihar. into two circles of intelligibility.6While lier. Nor does this new process work to Such a differentiationof the social cleav- the new entrantsto democraticpolitics the exclusionof the earliermodel of jati ages is itself a new featureof the third struggledto express their local concerns alliances.Yet it capturessomething of the electoralsystem. While it allows various and interestsin the alien vocabularyof qualitativechange in this respect. marginalcommunities to havea say in the liberaldemocracy, the elite marchedfur- Beyondthis level of generality,there is statepolitics, such a largevariety has not ther away for its rendezvouswith the so very little thatis commonto all the states allowedthe buildingof a largercoalition calledglobal village. This gave riseto two in termsof thepatterns of socialcleavages of marginalisedcommunities that might radicallydifferent languages of politics activatedby politics. Not only are the give them a role in the nationalpolitics. correspondingto the 'Bhasha'/English specificcombinations unique to eachstate Anothercommon belief in this respect divide. The former,a languageof demo- dueto its socialdemography, but thenature is that whateverthe exact natureof the cratic rights and social justice, was de- of cleavageitself varies from state to state. social cleavages, there is a tendencyto- ployed to win elections while the latter, If Andhraand Rajasthan represent a con- wardsgreater political polarisation along a languageof macro-economicand bu- reaucraticmanagement, guided the fram- TABLE9: WOMEN'STURNOUT AND WOMEN AS PROPORTION OFTOTAL VOTERS, ing of policies. While one sectionhad the LOKSABHA ELECTIONS, 1957-98 consolationof winningthe elections, the Year Women Men (Per Cent Women (Per Cent Turnout Odds Ratio othercould continue to rule.Just when the (Per Cent Turnout) Turnout) of Total Voters) Index lower orderhad some access to political the most economic 1957 38.8 55.7 38.3 0.81 0.50 power. significant 1962 46.6 62.1 39.8 0.84 0.53 decisionswere removed from the political 1967 49.0 61.0 42.0 0.88 0.62 agenda. 1971 49.2 61.2 42.4 0.89 0.61 If one goes by the popularaccounts, the 1977 54.9 65.4 43.6 0.91 0.65 rise of casteismand its gripover electoral 1980 51.2 62.2 43.1 0.90 0.64 politics is the distinctiveattribute of the 1984 59.3 68.4 44.5 0.93 0.67 1990s. But such an accountsuffers from 1989 57.5 66.4 44.1 0.93 0.68 1991 50.5 60.7 42.9 0.88 0.66 serious flaws. For one thing, caste has 1996 53.4 62.1 44.0 0.92 0.70 been operatingin electoralpolitics for as 1998 61.0 65.9 46.9 0.98 0.81 long as we know. Candidates'jati was a crucialfactor in electoralsuccess even in Notes: (1) Column 4 shows the percentage of women as a proportionof the total turnout. (2) The turnoutindex controls for the uneven size of the male and female electorate by expressing 1950s. Besides, the relationshipbetween the proprtionof women in turnoutrelative to the proportionof women in the electorate. For caste and politics is not just one way example, in 1998 women accounted for 46.9 per cent of the total voters and 47.7 per cent traffic.Politics has affected castes as much of the total electorate. We can therefore express the proportionin turnoutas a fraction of the as caste affects politics. What is distinc- proportionof the electorate. Thus, 46.9 / 47.7 = 0.98. If women had accounted for 47.7 per tive about the currentphase is not the cent of the total voters then their proportionin turnout would have been equivilent to their mix of these two or the vicious proportionamongst the electorate and the turnoutindex would have equalled 1. 3) The odds 'deadly' ratios measure the ratio between the odds of women voting and men voting. grip of caste over politics but ratherthe Source: CSDS Data Unit. mannerin which politics has come to TABLE 10: TURNOUT BY DEGREE OF URBANITY IN LOK SABHA 1977-1998 shape caste identities . ELECTIONS, In the first electoralsystem. the effec- Locality (Per Cent Urban) 1977 1980 1984 1989 1991 1996 1998 tive social bloc was a jati in its local In the Second re- Rural (up to 25 per cent) 60.3 56.7 64.1 62.6 56.6 58.7 63.0 setting. System, jati PredominantRural (26-50 per cent) 60.4 57.8 64.8 62.7 56.8 59.0 62.9 mainedthe primarybloc but the game of Urban (51 per cent and above) 62.1 57.9 62.9 60.0 50.4 53.4 56.7 manufacturingelectoral majorities led to Notes: Table entries are for turnout (in for the three of Lok Sabha thebuilding of state-widealliance ofjatis. average per cent) groups An extensionof the samesearch for stable constituencies according to the estimated proportionof urbanelectorate. The estimates have been developed by comparing the 1991 Census figures for urban population with the electoralmajorities has led to the replace- electorate size in 1991 lok Sabha elections. mentof jati by state-widejati clustersor Source: CSDS Data Unit.

2398 Economic and Political Weekly August 21-28, 1999

This content downloaded from 128.59.62.83 on Thu, 9 Jan 2014 20:07:04 PM All use subject to JSTOR Terms and Conditions thosecleavages. Once againthe available a contextfor massification is prepared.Far notoriouslyweak. People often use elec- evidencedoes not supportthis belief.8 It fromsetting the stage for further fragmen- tions effectivelyto choose theirrepresen- is truethat the rise of a partylike the BJP tationor effectivedecentralization within tative and the governmentbut rarelycan thathas skewed social base did increase these states,the rise of the statesindicates theyuse electionsto choosepolicies about the overall level of ethnic/social a high degree of homogenisationwithin issues that mattermost to them. polarisationin the 1990s. In a state each of these states.Whatever the early impli- Notes like UP where the BJP faces two other cationsfor nationalintegration, it is fairly skewed parties in the BSP and the SP, clear that its implicationsfor Indiande- 1 In the context of the US political history, see there is an unusually high level of mocracyare positive. The contextof state Paul Kleppner, The Third Electoral System, of votes caste/commu- choice closer to 1853-1892: Parties, Voters and Political polarisation along politics brings political Cultures Hill: of North lines for the last decade.But that is the of the (Chapel University nity every-daypolitical experience Carolina Press, 1979). The case for taking not trueof the rest of the country.Infact people and gives them an opportunityto seriously the US political history of the late the 1998 elections revealeda decline in evaluatethe governancethat affects them 19th century for understandingcontemporary the overall level of polarisation.And, most. That still leaves out a nuanced India has been argued persuasively by Robin whateverthe resultof the 1999 election, constituencyspecific assessmentby the Archer 'American Communalism and Indian the is to be even lower but the micro-levelanti- Secularism:Religion and Politics in India and polarisation likely voters, arguably the Economic and Political this time. at workin the 90s takescare West', Weekly, incumbency XXXIV (15), April 10, 1999. Thefinal feature, namely the emergence of that gap. 2 I have narratedthe story of the last fifty years of state as the effective unit of electoral Inthe limited sense described above, the at greaterlength in my contributionon 'Politics' choice, does not follow fromany of those thirdelectoral system represents a relative to the special issue, India Briefing: A mentioned above; infact there is some expansionof democraticchoice for the TransformativeFifty Year (Armonk,NY: M E case for thatthe above mentioed votersas well as their But if we Sharpe, 1999) arguing efficacy. 3 For an overview of thatround and a first sketch featuresare a resultof the last one. Elec- raisethese at a morefundamen- questions of the outlines of the new system see my 'Re- toralchoice in the Lok Sabhaelections in tal level, we confrontthe limits of what configurationin IndianPolitics: State Assembly the 1950s and 1960s was hardly influ- may have been achieved. Does the in- Elections, 1993-95', Economic and Political encedby theboundaries of thestate. Under crease in the numberof politicaloptions Weekly, XXXI (2-3), January 13-20, 1996. theSecond system, the statewere a distant represent a substantive expansion of 4 I have triedto look at a wide varietyof evidence considerationin the nation-wide choice?Or do we have moreof the in my 'Second Democratic Upsurge: Patterns secondary same, of in Electoral Politics stateslike Keralaand or more of less? This is not to Bahujan Participation plebiscites.Barring perhaps in the 1990s' in the volume edited by Francine West Bengalthat had emergedas excep- reiteratethe universalscepticism about Frankel (Delhi: Oxford University Press, in tions, it was the north-southdistinction the natureof choice in liberaldemocra- press). The conclusions offered in this and the thatmade any differenceto the electoral cies. In recenttimes the politicalspace in following para are taken from that paper. outcomes ratherthan state boundaries. India has shrunkmarkedly because the 5 I follow here the original meaning of the term the 1970s and 1980s mainstream the 'shudra politics' by Lohia. Throughout people parties,mainly Congress 6 This is borrwoed from in state and like otherin expression Sudipta voted the assembly elections as the BJP, are more each Kaviraj'sanalysis of the nationalistdiscourse. if theywere electing a primeminister. The all the crucialpolicy matters.The entire Several SubalternStudies historians have made 1990s have seen a radical shift in this politicalspectrum of mainstreampolitical a similar point about the disjunctionbetween respect.There is an unmistakablefore- optionsoffers a verysmall range of policy elite ideologies and mass beliefs. Partha groundingof her state in the political options. This problem becomes com- Chatterjee's remarks on the parallel between horizonof an citizen. Political if we addthe of exer- the peasant insurgencies in colonial India and ordinary pounded difficulty electoral waves in post-colonial times point loyalties,opinions and even social iden- cising whateverpolicy options may be in the direction of a sub-field of reseach that tities are now chosen at the level of the available.Here againthe referenceis not is waiting to be developed. state.Now peoplevote in the parliamen- merelyto thegap between the promise and 7 D L Sheth's paper 'Secularisation of Caste' taryelections as if theyare choosing a state the reality of collective action in any is to my mind the most comprehensive and government.State specific electoralver- situation.In contemporaryIndia the chain nuanced recent statement on the vexed therise state-wide andthe thatlinks needsto theirfelt desire relationship between caste and politics. dicts, of parties peoples' 8 conclusions based on the on- of state-wide clus- to their articulateddemand to its Preliminary joint emergence jati/vama aggre- going research on the changing pattern of tersas theeffective categories of electoral gation and finally to its translationinto community and vote from 1962 to 1998 by mobiltsationin the 1990s areall manifes- public policy is impossibly long and me and Anthony Heath. tations of this structuralattribute. Therise of statepolitics has been viewed with considerablesuspicion by the En- For the Attention of Subscribers and as the of glish press beginning political Outside India fragmentationif not balkanisationof In- Subscription Agencies dia. Such a fails to note thatthis reading It has come to our notice that a number of to the EPW from outside developmentis a functionof anaggregative large subscriptions ratherthan at work the country together with the subscription payments sent to supposed subscription agents disintegrativeprocess in India have not been forwarded to us. in ourpolity. It is betterinterpreted as the first on the Indian towardsthe We wish to point out to subscribersand subscriptionagencies outside India that all foreign step path with the must be forwarded to us and not creationof a mass society throughthe subscriptions,together appropriateremittances, mechanismof In a to unauthorisedthird parties in India. competitivepolitics. We take no whatsoever in of not with us. continentalsize polity like ours, it is responsibility respect subscriptions registered preciselyby articulatingrather than sup- MANAGER pressingthe distinctivenessof statesthat

Economic and Political Weekly August 21-28, 1999 2399

This content downloaded from 128.59.62.83 on Thu, 9 Jan 2014 20:07:04 PM All use subject to JSTOR Terms and Conditions