INDIA - Dalits and the Right to Food: Discrimination and Exclusion in Food-related Government Programs
by Joel Lee*
Researcher, Indian Institute of Dalit Studies
Sukhadeo Thorat*
Professor, Jawaharlal Nehru University, New Delhi
Director Indian Institute of Dalit Studies, New Delhi
Working Paper
Introduction
I. The ProblemAs a customary system of social and economic governance based on graded hierarchy, caste in Indian society lives in tension with the legal system of the Indian state, which is based on ideals of liberty and equality. Dalits, i.e. the Scheduled Castes or ?untouchables? who constitute almost one-fifth of the Indian population, suffer most acutely from the social and economic violence of the caste system, which prescribes their position as uniquely at the bottom of the graded hierarchy. Dalits in Indian society negotiate social and economic transactions in many spheres of life from this inherited position at the bottom, while in other spheres ?untouchability? excludes them from transactions with dominant caste society altogether.
Recognizing that caste discrimination and exclusion, particularly against Dalits, continues to thrive, and acknowledging that such discrimination and exclusion contradicts the spirit and letter of India?s Constitution, the central government has enacted various anti-discriminatory measures and legislation over the years, intended to redress these problems. The ?Anti-Untouchability Act? of 1955, the ?Protection of Civil Rights Act? of 1955/1976, and the ?Scheduled Caste/Scheduled Tribe Prevention of Atrocities Act? (SC/ST POA Act) of 1989 all acknowledge existing forms of discrimination and exclusion, and introduce new measures to eradicate them. A system of reservations in government employment, entrance in government educational institutions, etc., has also been enacted with the intent of redressing conventional caste discrimination and exclusion in these state-controlled sectors. In response to the growing Right to Food movement of recent years, the Supreme Court, following the Constitution, has required the central government, when introducing new food-related schemes in response to drought, unemployment, starvation, and so forth, to include specific provisions to prevent caste discrimination and exclusion in the implementation of these programs.
Conventional caste forces, however, are often able to subvert, evade, hijack or manipulate the progressive legal framework of the state, nullifying the transformative potential of its anti-discriminatory measures. Even in government schemes designed specifically for the welfare of communities traditionally discriminated against, members of those communities often find themselves excluded both from participation/partnership in the implementation of the scheme, as well as from the material or other benefits of the scheme. This is notably the case in Right to Food-related government programs such as the Midday Meal Scheme and the Targeted Public Distribution System.
II. Objectives
The Indian Institute of Dalit Studies (IIDS), therefore, undertook a study to look at caste discrimination and exclusion against Dalits specifically in the implementation of these Right to Food-related government welfare schemes. While violent atrocities and other egregious violations of the human rights of Dalits have received some (warranted) attention in recent years, the area of discrimination in government programs remains a relatively neglected area of study.
The purpose of the IIDS survey was to obtain an accurate, current, ground-level view of how, where, and to what degree caste discrimination and exclusion operate in the above-mentioned government programs as they are implemented in villages across India. Specifically, this was examined in terms of:
1. Access ? Can Dalits access the benefits of the government schemes?
2. Participatory Empowerment/Ownership ? To what degree are Dalits able to participate meaningfully in the implementation of the schemes, from ?having a say? to having ownership stakes in the schemes? material capital?
3. Treatment ? Are Dalits treated fairly and on equal terms with dominant caste communities in the implementation of the schemes?
.
.
.
III. Findings
A. Access:
1. Existence of a functioning MMS
This was measured jointly by the responses to the first question on the IIDS survey, ?Is there a Midday Meal Scheme in your village?? the fourth question, ?Is the midday meal actually served every day? If not, then how many days per month is the midday meal actually provided?? and the fifth question, ?Are the midday meals hot, cooked meals, or uncooked meals??
Results are positive. Out of all 306 villages surveyed, only five villages (three in Andhra Pradesh, two in Tamil Nadu) reported that there had been a MMS, but it had been closed. Of those, one village (TN) reports that the cause of closure was that dominant castes opposed the scheme because it would benefit SC/ST children; two villages (one AP, one TN) report that there was no initiative or leadership for the scheme; one village (AP) reports a problem with funding, and the fifth village (AP) gives no reason for the closure. The other 301 villages reported having a current MMS in which cooked food was prepared every school day.
2. Location of MMS
Location of the MMS has two components: setting and locality. For setting, the survey asks, ?Where is the midday meal held?? in the sense of, ?in which sort of physical space is it held?? with options of school, temple, public building, or other ? please specify. In 93% of respondents? villages, the midday meal is held in the school building itself, while in another 3% the meal is held in another public building. Only in two cases, both in Tamil Nadu, is the government midday meal held in a temple, raising immediate questions of exclusion for Dalit children, who are generally forbidden entry into temples, as well as for other non-Hindu children. See Table 2.
The second aspect of location, locality, is raised in the next survey question, ?Where is this space located?? with options of SC colony, dominant caste colony, and other ? please identify. In all three states, the villages in which the MMS is held in the Dalit colony are in a minority. After that point, however, the interstate variation is considerable. At 46%, Andhra Pradesh has a significantly higher proportion ? more than double ? of MMSs held in Dalit colonies than either Tamil Nadu (19%) or Rajasthan (12%).
One weakness of the survey is that while the option ?dominant caste village? was intended to imply the area inhabited by all non-SC/ST castes (i.e. whether ?upper caste? or BCs/OBCs; who, while often segregated amongst themselves, still share a common village from which the Dalit colony typically remains separate and outside) and the option ?other place? was intended to imply some truly neutral place outside of caste colonies (e.g. on the roadside apart from the village), these intended meanings did not effectively translate during the conducting of the survey. Some researchers and respondents followed the intended meanings, while others understood ?dominant caste village? to imply only ?upper caste? neighborhoods, and therefore marked MMSs held in ?backward caste? neighborhoods as ?other place?. Since the distinction between neutral, un-casted space and space owned and occupied by dominant castes (of whatever grade, since, typically, segregation/exclusion of Dalits is common to all non-SC/ST castes) was thus blurred, conclusions can only be drawn from the figures for MMSs held in Dalit colonies.
The survey does not directly ask the distance of the place where the MMS is held from the Dalit colony. However, when discussing the location of the MMS with research teams, a number of respondents brought attention to distance as a factor that makes their children?s access to the MMS problematic. In three villages, all in Tamil Nadu, respondents said that Dalit children simply cannot participate in the MMS because the school is too far from their homes.
3. Subjective comments on access
In Kamalaputhur Village in Thiruvannamalai District in Tamil Nadu, respondents related that a Dalit girl in second standard was prohibited from eating and chased out of her government school (located in the dominant caste locality) by the dominant caste community when she tried to partake of the MMS . In five other villages (three in Andhra Pradesh and two in Tamil Nadu), respondents stated that some Dalit children are barred from participation in the MMS on account of either caste discrimination generally or ?untouchability? specifically. While these six villages constitute only 2% of the 306 villages surveyed, the living practice of outright exclusion has profound implications for the Right to Food, and Dalits? access to that Right.
More common in the data than the outright exclusion of Dalit children is ?inclusion with negative distinction?, i.e. caste discrimination, in the MMS. While this too impinges on access, it will be discussed under treatment.
B. Participatory empowerment/ownership:
1. Percentage of Dalit cooks
As with location of the MMS, so also are there sharp divergences between states in the percentages of MMS cooks who are Dalit. In Rajasthan, only 8% of villages surveyed had Dalit cooks for the MMS, another 4% had ST cooks, and the remaining 88% had dominant caste cooks. In Tamil Nadu, 31% of villages surveyed had Dalit cooks, another 4% had both Dalit and dominant caste cooks, and 65% had dominant caste cooks (no ST cooks were found in any of the villages surveyed in TN). In striking contrast, 49% of villages surveyed in Andhra Pradesh had Dalit MMS cooks, 1% ST cooks, 3% mixed cooks (some from each community), and 47% dominant caste cooks. This gives a three state average of 29% Dalit cooks, 67% dominant caste cooks, and 2% each ST cooks and mixed cooks.
2. Percentage of Dalit organizers
The ?organizer? or ?in-charge? of the MMS means the person ultimately responsible for making the midday meal take place on a day-to-day basis. Most often the organizer is a schoolteacher or school headmaster, but in some places the PDS dealer, sarpanch or other community member is given responsibility for the MMS.
The survey data for Dalit organizers follows a similar pattern to that for Dalit cooks, except that there are even fewer Dalits employed as organizers (a decision-making position of authority) than there are as cooks. In Rajasthan, for instance, in not a single of the villages surveyed was there a Dalit in-charge for the MMS; 86% of respondent villages had dominant caste organizers, while 14% had ST organizers. Tamil Nadu follows with 73% dominant caste organizers and 27% Dalit organizers. Again, Andhra Pradesh has the highest degree of Dalit participation/ownership of the MMS, with 45% Dalit organizers, 51% dominant caste organizers, and 2% each ST organizers and organizers of various castes. See tables 11-16.
3. Subjective comments on participatory empowerment: denial of right to work
Survey questions framed in terms of subjective experience are equally if not more important to understanding the nature of exclusion and caste discrimination, than questions framed in terms of quantifiable variables. The pitfalls of subjective questions can be largely overcome by a process of verification and specification through particularizing follow-up questions; and the IIDS survey self-consciously attempts to do exactly that.
In response to the general question, ?Is there caste discrimination in the MMS in your village?? 52% of respondents from Rajasthan, 24% from Andhra Pradesh and 36% from Tamil Nadu (giving a three-state national average of 37%) report that there is indeed a problem of caste discrimination in the Midday Meal Scheme in their village.
In response to the following, particularizing question, ?If so, then how?? the data furnishes 79 valid responses from the three states. Of these, 42 of respondent villages, constituting the largest proportion, report that opposition to Dalit cooks is either the primary problem or one of the problems in their MMS.
?Opposition to Dalit cooks? is actually a blanket term describing several different patterns of specific acts of caste discrimination and exclusion observed in the IIDS study. The patterns can be grouped into five, taking place at different points during the process of MMS institution and continuance. First, when local administrators are putting the MMS into place, dominant caste community members intervene to block the hiring of Dalit cooks, favoring dominant caste cooks instead. Where a Dalit cook has been hired, dominant caste parents then begin sending their children to school with lunches packed at home, or require their children to come home for lunch, in any case forbidding their children to eat food prepared by the Dalit cook. In the third stage, dominant caste parents or community members pressure the local administration to dismiss the Dalit cook, on any pretext, and hire a dominant caste cook instead. Where this is ineffective, or sometimes without the intervening step, the dominant caste parents campaign to shut down the MMS in the village school altogether. Finally, some dominant caste parents react to the hiring and keeping of a Dalit cook by withdrawing their children from the school, and sometimes admitting them in a different school where the cook is not Dalit.
Some examples may help illustrate the above patterns. The first is adequately exemplified by Komara Village in West Godavari District, Andhra Pradesh. There, dominant caste women organized in the state government?s ?Dwacra? (Development of Women and Children in Rural Areas) scheme successfully mobilized community and administrative support to bar a Dalit women?s Dwacra group from obtaining employment as cooks in the village MMS.
Bhunabhay Village in Ajmer District, Rajasthan, illustrates two of the trends identified above. In Bhunabhay, when the MMS began in July, 2002, Sunita Bhil, a Scheduled Tribe widow, was hired to prepare the midday meal of ghughri. Dominant caste parents, considering Sunita polluted on account of her caste, ordered their children not to eat the midday meal at their school, effectively launching a proxy hunger strike through their children. Alongside this, the dominant caste parents met with and pressured the dominant caste headmaster of the school to dismiss Sunita Bhil from employment. Ultimately, the headmaster consented, expelled Sunita Bhil, and hired a dominant caste woman to cook in her place.
Typifying the fourth trend, the MMS of a government school in Ranga Reddy District, Andhra Pradesh, lasted exactly ten days before the dominant caste community, incensed that Dalit cooks had been hired, shut down the school .
Behind all of these trends of dominant caste behavior is the classic Hindu understanding of purity and pollution, according to which food prepared by a Dalit ? that is, an ?untouchable? ? is considered ?polluted? by virtue of its contact with the intrinsically polluted Dalit. On another level, dominant caste opposition to Dalit cooks also represents a power struggle over livelihood rights. In the manner of social boycotts, concerted dominant caste opposition to Dalit cooks functions to break Dalit economic aspirations, i.e. Dalit entry into new livelihood domains such as government employment at the village level. The rural dominant caste establishment, which traditionally enjoys the economic dependence of the Dalit community, perceives Dalit entries into new economic spheres as threatening, and therefore responds with a backlash. This is most evident in Andhra Pradesh, where a greater proportion of Dalits have secured employment as MMS cooks.
C. Treatment:
1. Seating arrangement/eating arrangement
Referring again to Tables 18 and 19, the second most commonly reported manifestation of caste discrimination in the MMS is segregated seating arrangements. The psychological scarring and other detrimental effects of segregation have been well documented worldwide, and do not require rehearsal here. As has been famously observed, ?separate is never equal?. This observation is entirely applicable to segregation in the Midday Meal Scheme, which violates Indian law as well as international law to which India is signatory, and which has obvious negative effects on the actualization of the Right to Food.
Several variations on the theme of segregation surface in the IIDS survey data. 31% of the villages that specify the form of caste discrimination in their MMS identify separate seating in their schools. In these instances, Dalit children are required to sit apart from the dominant caste children; sometimes simply apart within the same space, other times outside of the school building while the dominant caste children sit inside, or on a lower level than their dominant caste peers.
9.2% of villages report that the Dalit children and dominant caste children are required to eat separate meals altogether. This is most often the case where there are two MMS cooks for the same school, one Dalit and one dominant caste. The practice of separate meals usually implies segregated drinking water arrangements as well.
Interestingly, segregated seating is not always an institution from the beginning of the MMS. Paradigmatic of a trend most visible in the southern states, in Enathi Village in Sivagangai District of Tamil Nadu, in 2001, the dominant caste community instituted segregated seating in the MMS in a primary school where Dalit and dominant caste children previously had been sitting and eating together. In Enathi, following a dispute between a dominant caste woman and a Dalit woman over the latter?s right to draw water from a public well, the dominant caste woman?s community attacked the Dalit colony, causing the Dalit woman and her husband to be hospitalized. When the Dalit community approached the police and administration for justice, Enathi?s dominant caste community organized a rigorous social boycott of the Dalits, physically enforced by barricading the Dalit colony; and it was in this context, as part of the boycott, that segregation was launched in the previously shared MMS.
Similar incidents reported in the IIDS survey suggest a trend in which a caste conflict unrelated to the MMS flares up in a village, often as a result of Dalit assertion of rights, and as part of the effort to reestablish hegemony, the dominant caste community inaugurates new forms of segregation (in the MMS, for instance) and asserts new practices of untouchability. In one village in Tamil Nadu, respondents to the IIDS survey state that the dominant caste government schoolteacher ?solved? the caste tensions in their village by introducing segregated seating. While it is common in popular discourse to describe phenomena of caste discrimination as ?remaining?, ?still continuing?, and ?lingering?, such language does not accurately characterize the ground reality. Discourse aimed at eradicating caste discrimination must take into account the dynamism of caste phenomena, erroneously portrayed by the dismissive language of inevitable social progress.
2. Preferential treatment
Respondents in another eight villages (Table 18b) report more subtle forms of discrimination. In these villages, dominant caste teachers practice caste favoritism in serving the MMS, treating the dominant caste children preferentially and reserving the smaller or less desirable portions for Dalit children.
D. Issues in the pre-MMS dry grain distribution: the case of Bihar and Uttar Pradesh:
Two of the states most in need of a functioning Midday Meal Scheme to help improve their dismal education and nutrition records, Bihar and Uttar Pradesh have yet to comply with the Supreme Court?s six-month deadline (nearly two years old at the time of writing) in implementing the MMS, and continue instead with the distribution of dry grains that was intended only as a precursor to the MMS.
In Bihar, 99% of researchers in the IIDS study left the MMS section of the survey blank after answering the introductory question, ?Is there a Midday Meal Scheme in your village?? in the negative. In Uttar Pradesh, on the other hand, a majority of researchers completed the data in the MMS section by applying the questions to the existing program of dry grain distribution. The following brief discussion, then, will be based on this UP data and informal observations from a number of survey researchers in Bihar.
A. Access
1. Existence of a Functioning MMS
Setting aside the free, shared, cooked, midday meal required by law to be served at all government schools, is the system of free dry grain distribution to all government school children up and running in Bihar and UP?
Because some researchers treated the dry grain distribution system as equivalent to the MMS for the purposes of the survey and others did not, the survey data for ?Is there a Midday Meal Scheme in your village?? cannot be used. In UP, however, nearly every respondent village that elected to address the MMS questions pointed out two basic, congenital problems with the functioning of the dry grain distribution. First, while the distribution is intended to be 3 kilograms of wheat or rice (the grain of choice varies by location) per child per month, the full 3kgs are rarely provided; instead, the PDS distributor, sarpanch, teacher, or combination thereof, distribute 2kgs, 2.5kgs, or some other amount less than 3kgs, per child, and misappropriate the rest. Second, the distribution rarely takes place on a monthly basis; rather, distribution takes places every two, three, or four months, or in some cases less often still, without regularity or assurance. Some researchers in Bihar and UP reported a complete absence of the dry grain distribution system, or a system that had once functioned but had become defunct. Still, the majority of villages appear to have the dry grain distribution, but with abundant corruption and no regularity.
2. Location of the MMS
In 57% of valid responses in Uttar Pradesh, the actual event of grain distribution takes place in the school itself, as it should. In an alarming trend, however, another 37% of respondent villages report that their children have to go to the home or shop of the PDS dealer to receive their 3kg of grain.
Whether in school or in the PDS dealer?s home, in 52 out of 61 valid responses in Uttar Pradesh, the event of grain distribution to school children takes place in the dominant caste village. In six villages, the distribution takes place in a Dalit colony, and in three villages, in some other place. See Table 21. Even when not outright prohibited from entering, Dalit children are still in a situation of disadvantage and heightened vulnerability in dominant caste localities. The extreme bias of location evidenced by this data is thus a considerable impediment to Dalit children?s free and equal access to monthly government grain, and, by extension, the Right to Food generally.
B. Participatory empowerment/ownership
The issue of cooks does not arise in the dry grain distribution system. The proportion of Dalit organizers, however, to the total number of organizers, will provide a clear and measurable indicator of Dalit participatory empowerment and ownership of the pre-MMS dry grain distribution system.
In 42 out of 45 respondent villages in Uttar Pradesh, that is, in 94% of cases, the dry grain distribution organizer is dominant caste. Two villages reported a Dalit organizer, and one village, a ST organizer.
It is also noteworthy that in Uttar Pradesh, the distribution organizers are overwhelmingly male (32 out of 35 valid responses, or 91%). This contrasts significantly with the gender configuration of MMS organizers in the states where the cooked MMS is functioning. See Table 23. When considering the degree of reported corruption and discrimination in the functioning of the pre-MMS system in UP, the 91% male, 94% dominant caste composition of the grain distribution in-charges may prove significant.
C. Treatment
Discriminatory treatment in the pre-MMS dry grain distribution system is widely reported in the UP surveys in which respondents applied MMS questions to the grain distribution, as well as by survey researchers in Bihar. One commonly reported phenomenon is that the previously mentioned stinting of grain by the distributor (2 or 2? kgs instead of 3) is reserved for Dalit children, while dominant caste children receive the full quantity. Elsewhere, respondents explained that the entire quantity of grain allotted to Dalit children is simply consumed by the government intermediaries and Dalit children and parents are informed that the grain supply has not come, or was insufficient. In other places, if Dalit children miss one or more days of school in any given month, the teacher refuses to give those children their allotment of grain, whereas this rule is not applied to dominant caste children.
In Sonadi Village in Ghazipur District, UP, respondents report that the dominant caste teacher arbitrarily withholds the monthly rice allotment from some SC children, while giving it to other SC children and all of the dominant caste children. When the Dalit community approached the PDS distributor to lodge a complaint, the distributor responded that the grain is ?not for your children anyway?.
IV. Concluding Summary of MMS data
IIDS survey data clearly identifies the strong points and shortcomings of the Midday Meal Scheme as it is being conducted in India today, in terms of exclusion and caste discrimination as barriers to Dalit attainment of the Right to Food. A review of this data will highlight the primary issues Dalits face in the MMS, and will bring forth interstate variations that indicate possible solutions.
Access to the Midday Meal Scheme is first and foremost contingent on the implementation of the scheme by state governments. On this point, Uttar Pradesh and Bihar, where one third of India?s Dalits live , deny Dalit and other poor children access to their legislated entitlements from the very beginning, by simply refusing to implement the shared, cooked, Midday Meal Scheme. In the distribution of dry grains to government school children that continues to substitute for the MMS in Bihar and UP, regularized corruption and caste-based discriminatory distribution are widely reported, and in some cases outright exclusion of Dalit children from distribution is reported. Uttar Pradesh and Bihar, then, are currently blocking, rather than facilitating, Dalit children?s access to food through the MMS.
The governments of Rajasthan, Andhra Pradesh and Tamil Nadu, on the other hand, have achieved the initial step towards access, by implementing the MMS. Out of 306 villages surveyed in these three states, 301 villages, or 98.4%, have a functioning MMS in the government school in their village. A functioning MMS, however, does not always assure access. In a small number of respondent villages in Andhra Pradesh and Tamil Nadu, Dalit children are completely barred from functioning MMSs by dominant caste communities.
A second critical factor affecting Dalits? access to the MMS is the setting and location of the program. 93% of respondent villages in Rajasthan, Andhra Pradesh and Tamil Nadu hold the MMS in the school building itself, as is appropriate. Requiring immediate relocation, however, are two villages in Tamil Nadu in which the MMSs are currently held in temples, spaces from which Dalits are excluded.
In Uttar Pradesh, on the other hand (data is not available for Bihar), in only 57% of respondent villages is the pre-MMS distribution of dry grain to government schoolchildren conducted in the school building itself, while in another 37% of villages it is conducted in an ?other place?, unacceptably often the home or shop of the PDS dealer.
If the physical setting of the MMS is important, the locality in which that space is situated is equally if not more significant. Rajasthan and Tamil Nadu have very low percentages of villages in which the MMS is held in a Dalit locality (12% in Rajasthan and 19% in Tamil Nadu), whereas villages in more than double that percentage in both states hold the MMS in dominant caste localities. In notable contrast, 46% of respondent villages in Andhra Pradesh hold the MMS in a Dalit locality, which goes a long way toward assuring Dalit access, and should help erode dominant caste prejudices against entering Dalit localities.
In Uttar Pradesh, in 85% of respondent villages, the distribution of dry grain to government school children takes place in dominant caste localities, while in less than 10% of villages is the distribution conducted in Dalit localities. In UP, Rajasthan and Tamil Nadu, then, the vast majority of Dalit children must enter an area of heightened vulnerability, tension and threat, in order to avail themselves of the midday meal or its dry grain equivalent. Access for Dalit children is thus conditional, and hostage to the fluctuating state of caste relations in the village or region. Incidents like those at Enathi and Kamalaputhur villages in Tamil Nadu demonstrate how Dalit children?s access to the MMS, already tenuous because it is held in the dominant caste locality, is then cut off when dominant castes feel the need to reassert their hegemony.
In measuring Dalits? participatory empowerment in and ownership of the MMS, the IIDS survey data unearths interesting patterns both in terms of national trends and interstate variations. In hiring practices, Rajasthan is consistently the least likely to employ Dalits, with 8% of respondent villages having a Dalit cook, and not a single respondent village having a Dalit MMS organizer. Tamil Nadu hires proportionally more Dalits, while still keeping them firmly in the minority, with 31% of respondent villages having Dalit cooks, and 27% having Dalit organizers. Andhra Pradesh leads the three states in indicators of Dalit empowerment and ownership of the MMS, with 49% and 45% of respondent villages having Dalits as cooks and organizers, respectively. See again Tables 10 and 16.
One argument against hiring Dalit cooks is that where the society is not prepared to accept a shared meal cooked by a Dalit, it will ?create tension?, schools will be paralyzed and (dominant caste) children?s attendance will drop, thus defeating the purpose of the MMS. This argument has been made with reference to Rajasthan, formulated something like, ?In a socially conservative environment like Rajasthan?s, where people are not ready for Dalit cooks, hiring them now will cause more harm than good?. Significantly, however, opposition to Dalit cooks is the most frequently reported problem not only in Rajasthan but in Andhra Pradesh as well (and is a close second to segregated seating in Tamil Nadu). In other words, Andhra Pradesh?s success in hiring a significant proportion of Dalit cooks is not due to lack of opposition; but is rather a matter of political will (generated by sustained pressure from people?s movements). That Andhra Pradesh?s relatively progressive hiring practices have not been accompanied by a corresponding crisis of dropping school attendance or paralysis of the school system suggests that the above argument against hiring Dalit cooks, speculative in nature anyway, is in fact spurious.
Measurable indicators point to an extremely low level of Dalit participatory empowerment and ownership of the pre-MMS dry grain distribution system in Uttar Pradesh. In 94% of respondent villages in UP, the distribution organizer is dominant caste; SC and ST organizers are found in a combined 6% of respondent villages.
In terms of treatment of Dalits in the MMS, 27 respondent villages in Rajasthan, Andhra Pradesh and Tamil Nadu reported segregated seating in the MMS, and eight villages reported segregated meals altogether. In another eight villages, respondents reported that Dalit children are served food inferior to or in lesser amounts than their dominant caste classmates.
A three-state national average of 37% of respondent villages reports that caste discrimination does, in fact, afflict the Midday Meal Scheme in their village. Opposition to Dalit cooks it the single most common problem (48.3%), followed by segregated seating (31%), segregated meals (9.2%), and unfavorable treatment in food allotment (9.2%).
Taking a closer look at the 37% national average, considerable interstate variation comes to light. In Rajasthan, 52% of respondent villages report caste discrimination in the MMS; in Tamil Nadu, 36%; in Andhra Pradesh, 24%. It can hardly escape notice that this configuration is a nearly exact inverse of the interstate variation for percentage of villages with Dalit cooks, percentage of villages with Dalit organizers, and percentage of villages in which the MMS is held in a Dalit locality. That is to say, Andhra Pradesh, which has the highest percentage of Dalit cooks, Dalit organizers, and Midday Meal Schemes held in Dalit localities, simultaneously has the lowest percentage of reported caste discrimination in the MMS. Rajasthan, which has the lowest percentage of Dalit cooks and organizers and Midday Meal Schemes held in Dalit colonies, simultaneously has the highest rate of reported caste discrimination; and Tamil Nadu stands about midway between Rajasthan and Andhra Pradesh in each of these variables.
Do these matching patterns in the data indicate causality?
Quantitative and qualitative evidence from the field suggests that the above variables share at least an influential, if not directly causal, relationship with the degree of reported discrimination in each state. Considering each variable alone, for instance, the rates of reported discrimination are consistently lower when Dalit organizers are in charge of the MMS, when Dalit cooks are cooking the midday meal, and when the MMS is held in a Dalit colony, than when dominant caste organizers are in charge of the MMS, dominant caste cooks are cooking the midday meal, and when the MMS is held in dominant caste localities. Subjective comments from researchers and respondents also affirm that these trends are interlinked.
A look at Andhra Pradesh further supports this understanding of the data. What sets Andhra Pradesh apart? Does the lower incidence of reported caste discrimination in the MMS in Andhra simply reflect a more caste-free, egalitarian society than neighboring Tamil Nadu or distant Rajasthan? A quick glance at any of the literature on the subject, for instance the annual reports of the National Commission for Scheduled Castes and Scheduled Tribes, or the annual reports of human rights organizations such as Sakshi-Human Rights Watch AP, will disabuse the reader of any such notion. In fact, to take one commonly used indicator, rates of reported crimes committed against Dalits are higher in Andhra Pradesh than in Tamil Nadu, Bihar, Karnataka or Orissa, while lower than in Rajasthan or Uttar Pradesh . Andhra Pradesh?s relatively lower levels of reported discrimination in the MMS in IIDS survey data, then, cannot be linked to lower levels of casteism in the society generally.
Instead, as the patterns in data suggest, the higher percentages of Dalit cooks and organizers, and higher percentage of villages in which the MMS is held in Dalit localities, appear to be responsible for Andhra?s relatively low incidence of reported caste discrimination in the MMS. But how is it that Andhra Pradesh has come to have these higher levels of Dalit participatory empowerment, and Midday Meal Schemes held in Dalit colonies? One primary reason is that the Andhra Pradesh government conducts the MMS through local women?s organizations known as DWACRA (Development of Women And Children in Rural Areas) groups. As an alternative to implementing the scheme only through the usual channels of entrenched government machinery, known for corruption, casteism and unaccountability, having a joint set-up between the government and local social organizations, appears to have an invigorating effect on all actors involved. Given the opportunity to take up leadership roles and local level government employment, mothers of government school children take an increased interest in and engagement with the school and the MMS, and begin to demand access and extract accountability from government machinery.
Sustained mass action by mobilized people?s movement in Andhra Pradesh should be credited with creating the political atmosphere in which the state government has been forced to engage and cooperate with local non-governmental organizations in implementing its schemes. While DWACRA groups are government sponsored, they are clearly influenced by the models provided by social movements. Just as people?s participation has a proven record of decreasing corruption by government officials, so likewise it seems that people?s participation, particularly Dalit people?s participation, is beginning to bring down levels of exclusion and caste discrimination in government schemes.
All of that being said, it would be misinformed to suggest that the DWACRA group model is alone sufficient to eradicate the problems of the MMS or that Andhra Pradesh has ?arrived? in terms of enabling Dalits? Right to Food through the MMS. There is still a 24% rate of reported caste discrimination in Midday Meal Schemes in Andhra Pradesh, meaning that Dalit children in one in four schools face segregated seating, opposition to their community?s cooks, segregated meals altogether, or other forms of discriminatory treatment. IIDS survey data furnishes several instances of dominant caste women?s DWACRA groups practicing exclusion or discrimination in the MMS, for instance by rallying the dominant caste community to bar Dalit women?s DWACRA groups from employment as MMS cooks. Three of the cases of brazen exclusion of Dalit children from the MMS also remain for Andhra Pradesh?s government to eradicate before it can make any meaningful claims about the success of its policies.
A study of Andhra Pradesh?s Midday Meal Schemes provides not an ideal model, but a work in progress, with mixed success, from which some directional ideas can be taken. What policies/approaches seem to be working there, that can be applied in other states? First, increase the proportion of schools and MMS centers in Dalit colonies. Second, promote Dalit participatory empowerment and ownership of the MMS through hiring and promoting larger proportions of Dalit cooks and Dalit organizers. This can be catalyzed partly by implementing the MMS through or with the collaboration of people?s movements and local organizations such as Dalit women?s self-help groups.
The IIDS survey data, from Uttar Pradesh?s dry grain distribution system to Rajasthan, Tamil Nadu and Andhra Pradesh?s cooked MMS, sketches two vivid pictures of possible Midday Meal Schemes, at different ends of a broad spectrum. On one end is a government welfare scheme, in which the entrenched, dominant caste intermediaries of the government machinery are entrusted with giving monthly handouts to poor children. In practical terms, this scheme is conducted strictly on the terms of the dominant caste intermediaries, in their locality, with preferential treatment for their children, routine embezzlement, corruption, unaccountability, and anything from apathy to contemptuous hostility toward the children and parents of the Dalit community. On the other end of the spectrum is a government program jointly operated and monitored by the government and local, empowered community groups with Dalit leadership and/or representation, in which Dalit children and dominant caste children daily share a hot, cooked, nutritious meal, as often in a Dalit locality as in a dominant caste locality, and as often with a Dalit cook as with a non-Dalit cook. In practical terms, the only exclusion in this picture is the elective self-exclusion of the more conservative elements in dominant caste society, who withdraw their children from the program initially, but eventually, when their pressure tactics on a government committed to actualizing its Constitution and laws fail, re-enroll their children and slowly adjust to a truly democratic society. Dalit access to and fair treatment in the MMS in this picture is assured through Dalit decision-making empowerment and real ownership stakes in the program, which as part of the fabric of a vigilant civil society keeps the government accountable.
The Public Distribution System
I. Context
The Indian Government?s Targeted Public Distribution System (TPDS, or, often, simply PDS) is reputed to be the largest system of controlled food distribution in the world. In it, government stocks of essential food commodities, notably rice, wheat, sugar and oil, are distributed through the Food Corporation of India to needy areas, where people of Below Poverty Line (BPL) status can purchase the goods at subsidized, below-market prices fixed by the government. At the local level, stocks are provided through government-recognized ?Fair Price Shops?, or PDS shops, run by the local, government-recognized PDS dealer.
Unlike the Midday Meal Scheme, the TPDS is functioning throughout India. In the IIDS study, all 531 villages surveyed in Rajasthan, Uttar Pradesh, Bihar, Andhra Pradesh and Tamil Nadu, addressed the PDS in the survey.
II. Approach
As in the MMS section of the study, so also in the PDS section, exclusion and caste discrimination are identified in terms of access, participatory empowerment/ ownership, and treatment. The following measurable indicators are used to evaluate access. First, the existence and number of PDS shops in respondent villages ? are there PDS shops in your village, and how many? Second, the location of the PDS shops ? are PDS shops located in dominant caste localities, Dalit colonies, or other places?
The measurable indicator for participatory empowerment/ownership is the proportion of Dalit PDS dealers to total PDS dealers ? do Dalits have always to receive goods from dominant caste PDS shops, or are there Dalit PDS dealers in the community as well?
Measurable indicators for treatment in the PDS include subjective responses to questions regarding, first, discrimination in quantity ? do Dalits receive less than the legislated amount of goods for the price? Second, discrimination in price ? do PDS dealers charge Dalits more than dominant caste members for the same products? Three, caste-based unfavorable treatment by the PDS dealer ? do dominant caste PDS dealers favor their own community in distribution? And four, the practice of untouchability by PDS dealers ? do PDS dealers employ untouchability practices in the physical act of distribution or sale?
III. Findings
A. Access
1. Existence and number of PDS shops
To the credit of the Public Distribution System, PDS shops are largely up and running. As a national average, 87% of respondent villages in the IIDS study report having at least one functioning PDS shop in their village: 73% of respondents have exactly one PDS shop in their village, while 14% have more than one PDS shop per village. 13% of respondent villages, however, have no PDS shop in their village, and must travel outside to avail themselves of their legislated entitlements of subsidized goods.
Most of the villages without PDS shops are in Uttar Pradesh and Bihar. Taking a look at interstate variation, Uttar Pradesh shows itself the most recalcitrant in assuring PDS accessibility, with 39% of respondent villages lacking PDS shops, and only 7% having more than one shop. Bihar follows with 16% of villages lacking a PDS shop, and only 10% having more than one. Rajasthan and Tamil Nadu score about equally; 100% of Rajasthan?s admittedly small sample size of respondent villages report having exactly one PDS shop per village, while Tamil Nadu has a marginal 7% and 9% of villages lacking PDS shops and having more than one PDS shops, respectively. Access appears most assured in Andhra Pradesh, where 44% of respondent villages have more than one shop, 53% have exactly one shop, and only 3% have no shop. See Table 24.
While the IIDS survey did not directly ask about the distance between Dalit dwellings and the nearest PDS shop, 29 respondent villages made a point to remark on distance as a factor that impinges upon their access to the PDS. Of these, 26 villages, or 5% of all surveyed, commented that they had to travel ?far? or ?more than 2 kilometers? to reach the nearest PDS shop.
World Prout Assembly: Dalits and the Right to Food: Discrimination and Exclusion in Food-related Government Programs