Evaluating Reserved Seats as a Measure of Substantive Representation for Scheduled Castes and Scheduled Tribes through the Analysis of Parliamentary Questions in the Lok Sabha (1999–2016)

Introduction

Long before Anne Phillip highlighted the importance of the ‘presence’ of social identities in legislatures, India underwent heated debates over whose presence in the legislature would provide substantive representation to disadvantaged minorities like Muslims, Christians, Sikhs, untouchables, tribal groups, Anglo-Indians and Europeans. Special measures of representation ranging from separate electorate, weightages to reserved seats were devised by the colonial rulers to address the question of representation for these social and cultural groups, though, with a limited franchise. As an outcome of a bitter contestation between Ambedkar and Gandhi, reserved seats with double vote were evolved as a compromised formula for future elections to represent untouchables. The provision of reserved seats for SCs in legislative bodies has been celebrated as one of the most empowering measures for them to overcome historical disadvantage. The increased presence of SC members in the parliament has been credited to the policy of reserved seats; still, not much studies have been done to apprise their substantivenesses as a method of representation for the target groups. A number of studies suggest the presence of scheduled caste/scheduled tribe (SC/ST) representatives in decision-making bodies made the voices of their community better heard and influenced policy decisions in favour of their parent community (Galanter, 1979; McMillan, 2005; Pande, 2003). This argument has been countered by other studies such as Jensenius (2012, 2015) and Vundru (2013, 2018). Jensenius (2012) used the National election study survey data 2004 and analysed the reserved seats for SCs and STs and found that there is no sense of feeling in the reserved constituencies between SCs and STs and non-SCs voters. In addition, she suggests that background of SC and ST voters who live in reserved constituencies is not effective in SC constituencies (Jensenius, 2012). Similarly, Jensenius (2015) in her second paper has covered about 3000 state assembly constituencies in the 15 largest Indian states. The paper compares the development indicators in reserved and unreserved constituencies. She used literacy rate and the percentage of non-workers among SCs and non-SCs as development indicators. She finds that development would not affect the SC constituencies and non-SC constituencies and the result is insignificant. She finds that there is no correlation between development and reserved constituencies (Jensenius, 2015). Vundru (2018) argued that reserved seats failed to provide substantive representation to SC. On the other hand, Irma Clots-Figueras (2011) studied the effect of female political representation in the state legislature on the issues of public goods and expenditures. Her findings suggest that female legislators in reserved seats from SC/SC background in local governance invest more in health and early education and favourite ‘women-friendly law’. Conversely, her findings also suggest that unreserved seats for women have no positive effect on women-friendly legislation, and they oppose land reforms, preferably investing in higher education, and reduce social expenditure in their daily life. (Clots-Figueras, 2011, p. 664). Chin and Prakash (2011) found no significant impact of reserved seats in reducing poverty among SCs but reported a better positive correlation between reserved seats and reduction of poverty among STs. This study attempts to empirically know the ‘representativeness’ of SC/ST members of legislative bodies as far as substantive-ness of their representation is concerned by comparatively analysing parliamentary questions (PQs) raised by them and non-SC MPs on the issues concerning SCs. The first section of the paper traces the historical trajectory of reserved seats as a special measure of representation for untouchables and contestations surrounding it. Taking clues from section I, the context of the study is framed along with certain questions to consider substantiveness of reserved seats as a mode of representation. The next section describes the methodology and data sources for the paper and presents detailed data models. The last section is a conclusion with a discussion on obtained data.

I Contestations over Political Representation of Untouchables in Colonial India

The advent of Dr Ambedkar in the political field of colonial India marked the assertion of untouchables (Depressed Classes, Dalits or Harijans) as a political community. He bargained hard with the colonial rulers on the one hand and Gandhi and the Congress on the other to accept separate electorate for untouchables as a formula to overcome the historical injustice suffered by them. In 1919, Ambedkar presented his demand of separate electorate for the first time in front of the Southborough Committee, like that which was already available to Muslims through the Morley-Minto reforms, 1909. While the committee rejected the separate electorate for untouchables, it came up with the option of nomination of the member of the untouchable community to the provincial legislature in the Government of India Act 1919 (Vundru, 2013, pp. 2–3). Although Congress and Gandhi were sympathetic to the cause of untouchables, their approach was drastically different. While Ambedkar advocated a language of empowering rights, Gandhi relied on social reforms for the emancipation of untouchables. Similarly, the Congress expressed sympathy to untouchables but found it impossible to concede to separate electorate for them. In 1927, Ambedkar sought separate electorate for Depressed Classes in front of the Simon Commission along with other civil rights for them. Whereas, as a reaction to Simon Commission, Congress organized all India party conference and produced the Nehru Report 1928 which rejected all claims of separate electorate including that of Muslims, Sikhs and Depressed Classes. The report argued that separate electorate was ‘bad for the growth of a national spirit’ and particularly bad for minorities themselves and ‘make the majority wholly independent of the minority’ and ‘under separate electorate, therefore, chances are that minority will always have to face hostile majority’ (Nehru Report 1928). This report was strongly opposed by Muslim community led by Mohammad Ali Jinnah and the Depressed Classes led by Dr Ambedkar, both of them believed that the report had not taken sufficient care of rights of their respective communities and denied their rightful claims of representation.

Gandhi never accepted elevation of untouchables as a politically active community and resisted demand of separate electorate as it would be divisive for the Hindu community (Tejani, 2007; Vundru, 2018, p. 37) and would perpetuate the practice of untouchability (Vundru, 2018, p. 35). The divide between Gandhi and Ambedkar became more evident in post-Simon failure scenario when Prime Minister Ramsey MacDonald invited Indian leaders for Round Table Conferences in London. In the absence of Congress, the First Round Table Conference (1930–1931) was represented by 89 members from different Indian social groups, communities and princely states and it was largely inconclusive. While Sikh, Muslim and Christian leaders advocated separate electorate for their respective communities, Ambedkar successfully argued for representation of Depressed Classes through the separate electorate. Ambedkar argued in the conference that despite Depressed Classes being perceived as Hindus and inalienable constituent of Hindu community, they constitute a separate group (Tejani, 2007, p. 214; Vundru, 2018) because of the kind of treatment they received in the community. The Second Round Table was inaugurated in September 1931, significantly different from the first, because of Gandhi’s presence who participated as a sole representative of the Congress in the conference as a result of Gandhi-Irwin Pact. This conference was full of heated discussions as Gandhi was not ready to concede any community- or group-specific demand of separate representation except for the Muslims and Sikhs in the background of Lucknow Pact of 1916. This position was resisted by the leaders representing Depressed Classes, Christians and Anglo-Indians. Thus, an informal session was organized to settle communal claims. But Gandhi never convinced to recognize untouchables as a separate political community within Hinduism (Tejani, 2007, pp. 215–216; Vundru, 2018, pp. 21–34). Both Gandhi and Ambedkar questioned each other’s authority to be a spokesman of Depresses Classes. Gandhi even claimed to receive correspondences from the Depressed Classes and Christians discarding special representation (Tejani, 2007, pp. 215–216; Vundru, 2018, pp. 30–31).

The Second Round Table Conference did not reach any agreed solution, and thus, representatives of minorities came up with an informal pact about their rights in the future constitution including that of political representation for Depressed Classes. PM Macdonald accepted the pact and accordingly came up with the famous ‘communal award’ on 17 August 1932, recognizing separate electorates for the Muslims, Christians, Sikhs, Europeans, Anglo-Indians and Depressed Classes. The award, moreover, conferred weightages to certain communities and designed the representative system mutually separating a number of socio-economic identities on the lines of community, race and religion, and of economic interests and other cultural differences. The Depressed Classes were given ‘double vote’ that they were given separate electorate in their majority areas and also allowed to vote in general seats to avoid the perpetuation of their segregation (Tejani, 2007, p. 225; Vundru, 2018, p. 36). Ambedkar was a bit dissatisfied because of the scaling down of Depressed Classes representation. It was the Congress and Gandhi who found the communal award a threat to national unity in general and separate electorate for Depressed Classes as divisive for Hindu community in particular. Gandhi, who was imprisoned for initiating a second round of civil disobedience, wrote from Yeravda prison that

when the minorities claim was presented I said that I should resist with my life the grant of separate electorate to the Depressed Classes…. I am not against their representation in legislature…. But I hold that separate electorate is harmful to them and for Hinduism. (cited in Tejani, 2007, pp. 226–227)

Gandhi opined that the separate electorate would destroy Hinduism. According to him, ‘Untouchable hooligans’ together with ‘Muslim hooligans’ would make ‘common constituent cause’ and would kill Hindus. He went on ‘fast unto death’ for withdrawal of separate electorate for Depressed Classes and to save the unity of Hinduism. In response, Macdonald entrusted caste Hindus and Depressed Classes leaders to arrive at some alternative position to adjust the provisions of the award. Majority of people were behind Gandhi including a section from Depressed Classes. But Ambedkar and the like-minded considered Gandhi’s fast as ‘political stunt’, ‘coercive’ and ‘manipulative’ which was morally inconsistent (Tejani, 2007, pp. 228–231). Finally, Ambedkar surrendered to the countrywide pressure and, thus, gave up the demands of separate electorates and signed the famous Poona Pact in 1932 which provisioned reserved seats in joint electorate characterized with panel system. As per the Pact, primary elections were to be held through the separate electorate, consisting of voters of Depressed Classes electing a panel of four candidates from the community. The secondary election was to be, then, held through joint electorate electing one candidate out of the panel of four (Poona Pact 1932). During the course of the Poona Pact and afterward, Ambedkar showed dissatisfaction over the adoption of this system. During the discussion on Poona Pact, he wanted a panel of four to a panel of two candidates. After signing the Pact, Ambedkar again met Gandhi on 23 April 1933 to suggest replacement of the Panel system and two-stage elections with single election along with the percentage system. According to his new proposal, voters of Depressed Classes and general voters should vote together for Depressed Class candidates in separate ballot boxes in a single election. The Depressed Class candidate who would have secured less than 25 per cent of the votes from his community was to be eliminated. The votes of general voters, then, would be counted for the candidates securing more than 25 per cent votes of classes. And in case none of the candidates secured 25 per cent votes, top two candidates were to be selected for the counting of general votes. Gandhi refused this proposal (Vundru, 2018, pp. 65–67).

The Poona Pact extinguished the chances of Depressed Classes breaking away from Hindu community that had settled Hindu community as permanent numerical majority. It was the spectre of the disappearance of the identifiable majority that had brought Hindu nationalists, liberals and Gandhians together against separate electorates for the Depressed Classes (Tejani, 2007, p. 228). The agreement reached in the Poona Pact was kept intact in the Government of India Act 1935 (which also introduced the term ‘Scheduled Caste’ to describe the castes scheduled to be beneficiaries of special representation in legislative bodies) and later in the Constitution (Tejani, 2007, p. 232).

II The Context of the Study

Ambedkar, the staunchest advocate of untouchables’ ‘cause, wanted an electoral system that could send true representatives of untouchables to legislatures without isolating them from the majority community. He wished to ensure the crucial role of untouchables in the elections of general candidates (Vundru, 2018, p. 101) so that they had to be responsive to untouchable interests and wellbeing. Contrary to that, reserved seats, as adopted in the constitution and altered later, made the SC candidates dependent upon general voters to win an election, which forced the SC candidates to be responsive to the interests of general voters than to the SC voters whose role was no more decisive since the abolition of the multi-member constituencies. Both Lelah Dushkin (1972) and Marc Galanter (1979) argued that doublemember constituencies were abolished due to the fear of defeat of the general candidates by reserved candidates. Presenting a nuanced analysis of political representation of SCs, Galanter (1979) pointed out the structural constraints imposed by the arrangement of the reserved seats on SC representatives. Galanter argued that ‘to promote integration and acceptance, the reservations are designed in a way which filters and muffles the representation of the interests of Scheduled Castes and Tribes in so far as they diverge from the interests of other groups’ (Galanter, 1979, p. 442). Moreover, a single-member reserved constituency, argued Gopal Guru (1986), was used by the Congress as a dumping ground for the areas where Congress was not popular. Further, among 79 reserved constituencies, only on 39 seats SC population constituted or exceeded 20 per cent of the total population, and that is why Dalit parties like Bahujan Samajwadi Party (BSP) and Republic Party of India found it difficult to win a reserved seat (Vundru, 2018, p. 155). In fact, the impact of SC/ST reservations on overall electoral politics is found to be peripheral as parties with meagre support base among SCs/STs are able to secure a win from the reserved constituency (McMillan, 2005). Adding an explanation to this argument, Guru has pointed out that the SC candidates who are supported, financed and nominated by strong political parties are able to win the election. Once elected, these winners are inclined to project their parties as SC friendly (Guru, 1986) and their role then becomes quite limited in advocating the interest of their communities (McMillan, 2005). The accountability of reserved candidates to their respective communities has not been sufficiently ensured; not much incentive is there for them for being responsive to the interests of their respective communities (ibid.). In this situation, political reservation of SCs has become merely symbolic. In his analysis of issues in SC representation, Vundru (2013, 2018) found that the SC voters were dissatisfied with their representative as they worked on the commands of their political parties under the fear of losing their tickets for next elections. SC voters were found to believe that their representatives were ashamed to raise the issues faced by their community. Strikingly, Vundru (2018) found that SC voters were not satisfied with the present electoral system and wanted a change in form of assurance that Dalit representatives would be elected only by Dalit voters (2018, p. 157). Whereas 76.8 per cent SC respondents believed that their member of legislative assembly did not raise Dalit issues in the legislative assembly, 57.6 per cent believed that their MPs had not ever raised their issues in the parliament (2013, p. 24). Therefore, special measures like reserved constituencies, proportional representation and nominations which are considered necessary to make SC voices heard in the institutions of representation have produced ambiguous results.

Duflo (2005) analysed whether political reservation has any bearing on the provision of public goods in India in favour of the target group. She has found that reservation significantly increases the access of disadvantaged groups to political decision making and brings to power a group of relatively inexperienced and less-educated politicians. Chattophadhya and Duflo (2004) have analysed the reservation of women and schedule castes in Panchayati Raj in the state of Rajasthan and West Bengal. They find that women members coming from reserved seats tend to invest more in the water sector and SCs members from reservation background invest more in public goods for their community. Duflo, Fischer, and Chattophadhya (2005) provided evidence to show that a SC pradhan provides more goods to SC villages. By contrast, Bardhan, Mukherjee, and Torrado (2010) studied the impact of political reservation on SCs/STs and women in local government elected members of Gram Pradhan in the state of West Bengal. The result shows that Women Gram Pradhan in local government has a negative effect on ST/SC in terms of material benefits and SCs/STs Gram Pradhan slightly better than women but without significant negative effect. On the other hand, Dunning and Nilekani (2013), who studied the effect of quotas in the state of Karnataka, found that quotas have very little impact on public goods and redistribution. Chin and Prakash (2011) studied redistributive effects of political reservation for SCs and STs and confirmed a positive correlation between political reservation and poverty reduction which was found to be more significant for STs than SCs. Pande (2003) saw a positive relationship between increased SC representation and increase in job share of SCs but refrained from strongly concluding that ‘political reservation is a welfare-enhancing policy’ (p. 1147).

Whereas western literature on reserved seats examined weather quotas affect substantive representation. Hodžić and Mraović (2015) examined the effect of the reserved seats for minorities in Bosnia and Herzegovina and fonud a positive effect between reserved seats representative and substantive representation. By contrast, Jusić and Stojanović (2015), who studied reserved seats for national minorities in Bosnia and Herzegovina, did not find a positive effect. Furthermore, Banducci, Donovan, and Karp (2004) found a positive relationship between reserved seats representative and substantive representation. On the other hand, Lublin and Wright (2013) analysed that reserved seats and lower thresholds increase in minority representation but could not find correlation between reserved seats and substantive representation. With his significant contribution in the literature of politician representation, Kroeber (2017) indicated a positive relationship between the reserved seats representative and substantive representation. Kroeber concluded that political reservation is the only tool for ethnic minorities, which limits the electorate to minority voters only so that representatives could be perfectly accountable to the target group. Under this condition, minority representatives can achieve the terms of substantive representation (Kroeber, 2017, pp. 2011–2012).

Mixed conclusions from above-cited studies again prepare a compelling ground for fresh empirical research to find out whether political reservation for SCs in the form of reserved constituencies is acting as a proficient measure to provide substantive reservation to them. Taking clues from previous studies, certain questions are framed to evaluate substantiveness of reserved seats. Thus, this paper seeks to know whether the doubts raised by Ambedkar on reserved seats as an efficient and effective measure of representation of SCs are relevant after their working of more than seven decades. Therefore, it is pertinent to ask whether reserved seat MPs are actually able to provide substantive representation by raising questions concerning their community irrespective of their party affiliation or their placement in ruling or opposition party.

Hypothesis

The available literature on reserved seats suggests that reserved seats are proficient to promote interests and well-being of disadvantaged minorities. On this basis, we assume that in India, reserved seat MPs are more likely to engage with substantive representation to their concerned community than other MPs. We expect that reserved seat MPs would be more likely than other MPs to ask questions of a ‘substantive nature’. Moreover, we assume that irrespective of the party affiliations, reserved seat MPs tend to provide substantive representation to their community members. Likewise, reserved seat MPs of opposition parties are more likely to raise substantive issues than the MPs who are in power sitting in the government to sustain accountability of SC representatives towards their community.

III Data and Research Methods

Our observations consist of all PQs tabled in 1999–2016 (17 years and four general elections) in the Lok Sabha (lower of house of the parliament). In hypotheses, (a) we outlined our expectations that reserved seat MPs would be more likely than other MPs to ask PQs of ‘substantive nature’ in the Lok Sabha. To test for these hypotheses, we analysed the PQs asked by reserved seat MPs in general and other MPs in particular during 1999–2016. The PQ data set includes starred and un-starred PQs of MPs of reserved seats and non-reserved seats in Lok Sabha between the years 1999 and 2016. Data for this study were collected through keyword searches on parliament website and library. The PQs are an important tool for measuring an individual legislator’s role orientation and the functions of parliaments (Martin, 2011). It is also provided to identify the MP’s behaviour and legislative activities in the parliament. This paper has conducted content analysis and identified the themes of PQs related to the SCs and STs and compiled into one group, call it substantive issue questions. The primary purpose of this paper is not simply to present a summary of PQs asked by MPs but to use these data to analyse empirical and hypothetical assumptions (Martin, 2011). In total, 258,617 PQs had tabled during 1999–2016 (see Table 1). For the purpose of this paper, only those PQs are selected that were either tabled by the MPs of reserved seats and non-reserved seat MPs dealing with pertinent SC issues. Our data sets consist of 1708 MPs from reserved or unreserved seats during 1999–2016. Out of 1708, 425 MPs belong to the schedule caste and schedule tribes elected from reserved constituencies. Of this number, 39 were female and 386 were male MPs (see Table 1).

OPEN IN VIEWER

Table 1.

Number of Parliamentary Questions Tabled by MPs in the Lok Sabha, Their Religious and Social Background, Type of Constituency and the Party Status

Variables	Frequency	Percentage (%)
Religious background
Hindu	246,963	95.49
Muslim	11,654	4.51
Gender
Male	240,440	92.97
Female	18,177	7.03
Social background
UR	184,731	71.43
OBC	15,351	5.94
SC	39,873	15.42
ST	18,662	7.22
Constituency type
Reserved constituency	58,535	22.63
Unreserved constituency	200,082	77.37
Party parliamentary status
Government	125,983	48.71
Opposition	132,634	51.29

Source: Parliamentary Questions Data Set, 1999–2016.

Note: UR – Unreserved.

As already mentioned, the unit of analysis is PQs asked during 1999–2016, through which we measure the dependent variable, that is, substantive questions/issues concerning. Because our dependent variable is a categorical variable, we apply logistic regression model analysis. To distinguish between reserved seat constituency and non-reserved seat member of the parliament, we have included and divided the reserved and non-reserved constituency, as well as dummy variable code ‘1’ if the MPs belongs to non-reserved constituency MPs and ‘0’ for reserved seat MPs. We also coded dummy variable ‘1’ in case of the opposition (party) and ‘0’ for government (or party) in power. To determine the effect of substantive issues, we use dummy variable: reserved seat MPs, other MPs. In addition, we include caste, female, party status; opposition and government; and type of constituency as explanatory variables. Opposition and government parties tend to have different attitudes about the substantive issues, and constituency party might overlap as constituency party more or less likely to be willing to act in the interest of minority (Dunning & Nilekani, 2013; Jensenius, 2012; Kroeber, 2017; Zuber, 2015).

In Model 1 (Table 2), for this study, only one dependent variable was used, that is, whether a MP was elected from a reserved constituency or from unreserved constituency. The findings confirm that SC MPs elected from reserved constituencies are more likely to raise issues related to their community they represent as compared to other MPs. The odds ratio is significant and greater than 1. After including other control variables, it has been found in all the models that major political parties are more likely to raise the issue of deprived communities as compared to the left parties. Except the Congress party, major political parties are more likely to raise the issue of Dalit community in the parliament. The odds in favour of the major political parties (Bharatiya Janata Party [BJP], BSP, other independent MPs and other smaller parties) in raising the issues of the deprived community are greater than one. Moreover, across gender, the result shows that the odds of raising questions related to development community by female MPs is significantly 1.26 times higher than male MPs, irrespective of which community these female MPs come from. Similarly, after incorporating other variables such as MPs in opposition and government, we found that in the context of opposition, MPs in government asking PQs related to substantive issues of SC/ST communities is less significant, that is 1.08 times as compared to opposition MPs, while there is no significant difference in the odds of asking questions related to substantive issues of SC/ST communities between MPs elected on reserved seats in opposition or in government.

OPEN IN VIEWER

Table 2.

Parliamentary Questions, Reserved Seats and Unreserved Seats MP in Lok Sabha 1999–2016

Variables	Model 1	Model 2	Model 3	Model 4	Model 5
MPs in Lok Sabha
Dalit interests	1.915***	1.826***	1.757***	1.882***	1.681***
	(0.0432)	(0.0419)	(0.0606)	(0.0579)	(0.0467)
Party
Congress		0.996	1.019	1.019	0.987
		(0.0522)	(0.0537)	(0.0537)	(0.0522)
BJP		1.220***	1.234***	1.234***	1.178***
		(0.0621)	(0.0628)	(0.0628)	(0.0605)
BSP		1.693***	1.662***	1.662***	1.714***
		(0.128)	(0.126)	(0.126)	(0.130)
State		0.978	0.971	0.971	0.969
		(0.0509)	(0.0506)	(0.0506)	(0.0505)
Independent		2.271***	2.150***	2.150***	2.138***
		(0.199)	(0.191)	(0.191)	(0.189)
Smaller		1.201***	1.195***	1.195***	1.202***
		(0.0758)	(0.0755)	(0.0755)	(0.0765)
Govt./Opp.
Govt. in power			0.925***	0.925***	0.910***
			(0.0250)	(0.0250)	(0.0204)
Gender
Female					1.261***
					(0.0310)
Religion
Muslims					0.937
					(0.0550)
Social groups
ST					1.273***
					(0.0495)
SC					–
OBC					1.033
					(0.0505)
Interaction
R.S. MP × Opp.			1.072
			(0.0494)
R.S. MP × Govt.				0.933
				(0.0431)
Constant	0.0300***	0.0277***	0.0286***	0.0286***	0.0365***
	(0.000399)	(0.00133)	(0.00140)	(0.00140)	(0.00222)
Observations	258,617	258,617	258,617	258,617	258,617

Source: Parliamentary Questions Data Set, 1999–2016.

Note: Reference Category: MPs in Lok Sabha: Non-R.S. MP; Party: left; Govt./Opp.: Opp.; Gender: male; Religion: Hindu; Social Groups: General; Standard error in parentheses; Significance level: ***p < 0.01, **p < 0.05, *p < 0.1.

Results: We present the results of regression analysis below.

Interestingly, the study could not find any significant difference among the Muslim and non-Muslim MPs in raising the issue of deprived communities in the parliament. Similarly, the coefficient of the MPs belonging to the Other Backward Class (OBC) groups is also insignificant, while as the coefficient of the MPs belonging to the ST category is greater than one which means that ST MPs are more likely to raise the issue of deprived sections in the parliament as compared to the MPs belonging to the general or OBC groups. Our results actually confirm what Zuber (2015) rightly observes in his six formulas, where he draws the line between reserved seat representative and substantive representation. In addition, he argues that party affiliations affect that line when the MP comes from an independent party or ethnic party with ethnic votes. Our results show that it is more likely for ethnic party (BSP) or independent party MPs to ask PQ on the substantive issue related to the SC/ST community as compared to other parties.

Conclusion and Discussion

Our data suggest SC MPs are closely connected to the interests of their community, confirming that reserved seat MPs are able to provide substantive representation to Schedule Caste community. Therefore, reserved seats remain pertinent special measure of political representation for disadvantaged sections of society against the findings of Vundru. However, this inference needs a closer analysis as it is noteworthy that non-reserved seat MPs (mostly OBCs and general category) are not eager to raise issues concerning SCs.

On the other hand, the fact ensures that the present shape and design of the reserved seats impose constraints on SC representatives. An SC MP has to prove himself as representative of all the electors and not just that of SCs of his constituency. He has to be comparatively more worried about the party he comes from than the SC electors whose vote actually do not play a decisive role in the first past the post system in securing an electoral victory on reserved seat. For example, a SC representative securing 30 per cent of the total votes in a reserved constituency, which is more than the number of votes secured by other contenders, would be declared elected. There is no obligation on him to secure any percentage of SC votes or whether SC community living in his constituency be satisfied with his performance. In a situation, reserved constituency has SC population between 20 and 30 per cent, he could successfully win the elections without appealing to SC population or appealing to other caste populations. It is worth recalling at this point that Ambedkar wanted mandatory 25 per cent SC votes for SC candidates to be open for general voters (if the candidate would have secured 25 per cent SC votes, then votes given by general voters to that candidate would have been counted for the final result of the election).

Along with earlier studies, this paper suggests that the performance of SC MPs in parliament is contingent upon their membership in ethnic, national, centralist, leftist or rightist party. Although SC MPs across political parties provide substantive representation by asking questions concerning their community, their political parties do not show similar concern for SCs. Ethnic parties like BSP along with independent candidates and smaller political parties are found to be keen in raising SC issues in the parliament, but major parties such as Congress, BJP and other left parties are showing similar concern for SC. While this paper acknowledges the significance of descriptive representation for SCs in form of reserved constituency, it suggests a revisit to Ambedkar’s criterion of mandatory 25 per cent SC votes for winning election on SC reserved seat. Introduction of this criterion would make major political parties accountable to SC voters of reserved constituency and strengthen reserved constituency as a measure of descriptive representation. They could not treat reserved seat as ‘dumping ground’ for places where they lack popular support.