Counting for equality

The recent Supreme Court ruling on reservations recognises the diversity within SC/ST. However, targeted action needs caste census data

Caste-based reservation has long been one of India’s most divisive issues, yet recent trends increasingly favour extending reservations to backward groups.

In State of Punjab vs Davinder Singh, the Supreme Court of India held that the sub-classification of Scheduled Castes/Tribes (SCs/STs) to provide intra-category reservations for more backward groups is constitutional. While sub-classification is a step forward, it remains a complex issue, dividing even the proponents of positive discrimination.

Earlier, in E V Chinnaiah vs State of Andhra Pradesh, the court held that once castes are identified by the President under Articles 341 and 342, they form a conglomerate to be deemed as SCs/STs.

The court reasoned that by creating the deeming fiction, the constitution suggests that all identified castes constitute a homogenous class due to their shared experience of untouchability.

However, in Davinder Singh, the court disagreed, asserting that Articles 341 and 342 only deal with caste identification and do not imply homogeneity among SCs.

Further, by espousing homogeneity in Chinnaiah, the court overlooked societal realities. Scholar Christophe Jaffrelot distinguishes between traditional and graded inequality: In traditional inequality, dominant groups confront each other as unified forces, whereas in graded inequality, even the dominated groups are internally divided and stratified.

B R Ambedkar explained that in a system of graded inequality, every class enjoys some privileges, except those at the very bottom of the social hierarchy.

Graded inequality persists in several Indian states, including Punjab and Maharashtra. In Punjab, stratification exists among Balmikis, Mazhabis, and Ravidasi Sikhs, the last being relatively more advanced. Similarly, in Maharashtra, Chambhars, Mahars, and Mangs practice untouchability against one another.

Article 341(2) states that only Parliament can modify the presidential list by including, excluding, or altering groups. The dissenting opinion in Davinder Singh applied a literal interpretation, concluding that sub-classification amounts to inclusion or exclusion of groups and hence modifying the list, which is impermissible.

However, this is a mischaracterisation. Article 341(1) concerns only the identification of SCs, and sub-classification does not alter the list. Instead, it reallocates reservation benefits among the SCs/STs based on social advancement without including or excluding any group. This modifies the reservation policy, not the presidential list — a distinction overlooked in both opinions.

The court in Chinnaiah overlooked that Articles 15(4) and 16(4), read with Article 14, grant the state broad powers to provide reservations for backward classes. Article 15(4) permits the state to make ‘any special provisions’ for socially and backward classes, SCs, and STs, with ‘any’ being broad enough to include sub-classification. Similarly, Article 16(4) allows the state to provide reservations for ‘backward classes’ who are ‘not adequately represented’ in the state services, which can include a caste or sub-group within SCs/STs if they meet these criteria. 

Davinder Singh correctly applies the principle of reasonable classification under Article 14, interposed with Articles 15(4) and 16(4), as upheld in State of Kerala vs N M Thomas, to sub-classification.

This classification must meet two criteria: an intelligible differentia between classes and a rational nexus with the legislation’s objective. Given the graded inequality within SCs/STs, there is a clear distinction between more backward groups and relatively advanced ones. Providing reservations to these groups has a rational nexus with the goal of achieving substantive equality.

Davinder Singh falls short on two points. Firstly, it doesn’t clarify whether the power to subclassify belongs to state legislatures/governments or to the broader ‘State’ defined in Article 12, which includes the central government, Parliament, and local authorities.

If state governments have this power, a detailed explanation is necessary. Ambedkar feared state governments might revel in electoral politics in identifying SCs, which is why Parliament was given the power to alter the presidential list. Similar concerns arise with sub-classification, where state governments could factor in political considerations. This ambiguity around ‘State’ could lead to litigation.

Secondly, the court applied the creamy layer principle to SCs/STs, which traditionally has been applied to OBCs, despite it not being argued by the parties and sub-classification being distinct from creamy layer.

Sub-classification operates at the group level, reallocating reservation benefits to most backward groups, while the creamy layer operates at the individual level by excluding those individuals who have achieved a certain economic status. Justice Gavai acknowledged that creamy layer cannot be applied to SCs/STs in the same way as to OBCs but failed to clarify what ‘creamy layer’ means here.

Apart from its minor shortcomings, the court’s verdict is a welcome step towards securing social justice for the most marginalised. According to the court, sub-classification policies have to be backed up by quantifiable data. Given this, a nation-wide or state-specific caste census is imperative for the effective implementation of this policy.

(The writers are students at NALSAR University of Law, Hyderabad)