INDRA SA\VHNEY AND ORS. ETC. ETC. v. UNION OF INDIA AND ORS. ETC. ETC.

room for any amount of abuse. According to the learned counsel, the political executive, owing to the degeneration of the electoral process, B normally acts out of political and electoral compulsions, for which reason it may not act fairly and indepenently. If, on the other hand, the provision I is to be made by the legislative wing of the State, it will not only provide an opportunity for debate and discussion in the Legislature where several shades of opinion are represented but a balanced a~d unbiased decision c free from the allurements of electoral gains is more likely to emerge from such a deliberating body. Sri Venugopal cites the example of Tamil Nadu where, according to him, before every general election a few communities are added to the list of backward classes, only with a view to winning them over to the ruling party. We are not concerned with the aspect of what is ideal or desirable but with what is the proper meaning to be ascribed to the expression 'provision' in Article 16(4) having regard to the context. The use of the e.xpression 'provision' in clause ( 4) of Article 16 appears to us to be not without design. According to the definition of 'State' in Article 12, it includes not mere'.y the government and Parliament of India and Government and Legislature of each of the States but all local authorities and other authorities within the territory of India or under the control of the Government of India which means that such a measure of reservation can be provided not only in the matter of services under the Central and State Governn1ents but also in the services of local and other authorilt.s referred to in Article 12. The expression 'Local Authority' is defined in F Section 3(31) of the General Clauses Act. It takes in all municipalites, Panchayats and other similar bodies. The expression 'other authorities' has received extensive attention from the court. It includes all statutory authorities and other agencies and instrumentalities of the State Govern- mcntiCcntral Governmenl. Now, \vould it be reasonable, possible or prac· ticablc to say that the Parliament or the Legislature of the State should G provide for reservation of posts/appointments in the services of all such ho<lics beside~ pr9vi<ling for in respect of services under the CentraVStatc Cfovcrnmcnt? Thi'..; aspect would become clearer if we notice the definition of "Law" in Article 13(3)(a). It reads:

H "13(3) Int hi~ article, unless the context otherwise requires,-

INDRA SJ\ WHNEY v. U.0.1. 609

(a) 1'Law'' includes any ()rdinance, order, bye-law, rule, A regulation, notification, custom or usage having in the territory of India the force of law; ....... .''

The words ''order", ~'bye-law", "rule11 and "regulation" in this definition are significant. Reading the definition of "State" in Article 12 and of "Law" in Article 13(3)(a), it becomes clear that a measure of the nature con- B templated by Article 16(4) can be provided not only by the Parlia- ment/Legislature but also by the executive in respect of Central/State services and by the local bodies and "other authorities" contemplated by Article 12, in respect of their respective services. Some of the local bodies and and some of the statutory corporations like Universities may have thier c own legislative wings. In such a situation, it would be unreasonable and inappropriate to insist that reservation in all these services should be provided by Parliament/Legislature. The situation and circumstances of each of these bodies may vary. The rule regarding reservation has to be framed to suit the particular situations. All this cannot reasonably be. done by Parliament/Legislature . • Even textually speaking, the contention cannot be accepted. The very use of the word 11 provision 11 in Article 16(4) is significant. Whereas clauses (3) and (5) of Article 16 - and clauses (2) to (6) of Article 19 - use the , word 11 Law", Article 16(4) uses the word "provision!!. Regulation of service conditions by orders and Rules made by the Executive was a well known feature at the time of the framing of the Constitution. Probably for this reason, a deliberate departure has been made in the case of clause (4). Accordingly, we hold, agreeing with Balaji, that the 11 provision11 con- templated by Article 16( 4) can also be made by the executive wing of the F Union or of the State, as the case may be, as has been done in the present case. Bajaji has been followed recently in Comptroller and Auditor General of India v. Mohan Lal Mehrotra, A.I.R.1991 S.C.2288. With respect to the argument of abuse of power by the political executive, we may say that there is adequate safeguard against misuse by the political executive of the power under Article 16(4) in the provision itself. Any determination of G backwardness is not a subjective exercise nor a matter of subjective satis- faction. As held herein - as also by earlier judgments ' the exercise is an objective one. Certain objective social and other criteria has to be satisfied before any group or class of citizens could be treated as backward. If the executive includes, for collateral reasons, groups or classes not satisfying H

610 SUPREME COURT REPORTS [1992] SUPP. 2 S.C.R.

A the relevant criteria, it would be a clear case of fraud on power.

Question l(b) : U1iether an executive order making a 'provision' under Article 16(4) is enforceable forthwith?

56Copy pinpoint56. A question is raised whether an executive order made in terms B of Article 16(4) is effective and enforceable by itself or whecher it is 1 necessary that the said 'provision is enacted into a law made by the 11

appropriate Legislature under Article 309 or is incorporated into and issued as a Rule by the President/Governor under the proviso to Article 309 for it to become enforceable? Mr. Ram Jethmalani submits that Article C 16( 4) is merely declaratory in nature, that it is an enabling provision and that it is not a source of power by itself. He submits that unless made into a law by the appropriate Legislature or issued as a rule in terms of the proviso to Article 309, the "provision 11 so made by the Executive does not become enforceable. At the same time, he submits that the impugned Memorandums must be deemed to be and must be treated as Rules made D and issued under the proviso to Article 309 of the Constitution. We fmd it difficult to agree with Sri Jethmalani. Once we hold that a provision under Article 16( 4) can be made by the executive, it must necessarily follow that ' such a provision is effective the mement it is made. A Constitution Bench of this Court in B.5. Yadav (1981 S.C. 561), (Y.V. Chandrachud, C.J., E speaking for the Bench) has observed:

"Article 235 does not confer upon the High Court the power to make rules relating to conditions of service of judicial officers attached to district courts and the courts subordinate thereto. Whenever it was intended to confer F on any ·authority the power to make any special provisions or rules, including rules relating to conditions of service, the Constitution has stated so in express terms. See, for example Articles 15(4), 16(4), 77(3), 87(2), 118, 145(1), 146(1) and (2), 148(5), 166(3), 176(2), 187(3), 208, 225, 227(2) and (3), 229(1) and (2), 234, 237 and 283 (1) and G (2)."

Be that as it may, there is yet another reason, why we cannot agree that the impugned Memorandums are not effective and enforceable the moment they are issued. It is well settled by the decisions of this court that the appropriate government is empowered to prescribe the conditions of

INDRA SAWHNEY v. U.0.1. 611

service of its employees by an executive order in the absence of the rules made under the proviso to Article 309. It is further held by this court that even where Rules under the proviso to Article 309 are made, the goverment .can issue orders/instructions with respect to matters upon which the. Rules are silent. [see Sant Ram Sharma v. State of Rajasthan [1968] 1S.C.R.111]. This view has been reiterated in a recent decision of this court in Comp- troller and Auditor General v. Mohan/al Mehrotra [1990] 2 S.C.C. 1 wherein it is held:

"The High Court is not right in stating that there cannot be an administrative order directing reservation for Scheduled Castes and Scheduled Tribes as it would alter the statutory c rules in force. The rules do not provide for any reservation. In fact it is silent on the subject of reservation. The Govern- ment could direct the reservation by executive orders. The administrative orders cannot be issued in contravention of the statutory rules but it could be issued to supplement the statutory rules [See the observations in Santram Sharma v. D State of Rajasthan, [1968] I SCR 111 AIR 1967 SC 1910]. In fact similar circulars were issued by the Railway Board introducing reservations for Scheduled Castes and Scheduled Tribes in the Railway Services both for selection and non-selection categories of posts. They were issued to E implement the policy of the Central Government and they have been upheld by this Court in Akhil Bhartiya Soshit Karamchari Sangh (Railways) v. Union of India, [1991] 1 sec 246 = AIR 1981 SC 298". F It would, therefore, follow that until a law is made or rules are issued under Article 309 with respect to reservation in favour of backward classes, it would always be open to the Executive (Government) to provide for reservation of appointments/posts in favour of Backward Classes by an executive order. We cannot also agree with Sri Jethmalani that the im- pugned Memorandums should be treated as Rules made under the proviso G to Article 309. There is nothing in them suggesting even distantly that they were issued under the proviso to Article 309. They were never intended to be so, nor is that the stand of the Union Government before us. They are executive orders issued under Article 73 of the Constitution read with clause (4) of Article 16. The mere omission of a recital "i'l. the name and H

612 SUPREME COURT REPORTS [1992] SUPP. 2 S.C.R.

A by order of the President of India" does not affect the validity or enfor- ceability of the orders, as held by this court repeatedly.

Question 2(a). Whether clause (4) of Article 16 is an exception to clause (1)?

57Copy pinpoint57. In Ba/aji it was held - "there is no doubt that Article 15(4) has to be read as a proviso or an exception to Articles 15(1) and 29(2)". It was observed that Article 15(4) was inserted by the First Amendment in the light of the decision in Champakam, with a view to remove the defect pointed out by this court namely, the absence of a provision in Article 15 C corresponding to Clause (4) of Article 16. Following Balaji it was held by another Constitution Bench (by majority) in Devadasan - "further this court has already held that clause (4) of Article 16 is by way of a proviso or an exception to Clause (1)''. Subbarao, J., however, opined in his dissenting opinion that Article 16(4) is not an exception to Article 16(1) but that it is only an emphatic way of stating the principle inherent in the main provision itself. Be that as it may, since the decision in Devadasan, it was assumed by this court that Article 16(4) is an exception to Article 16(1). This view, however, received a severe set-back from the majority decision in State of Kera/a & Ors. v. N.M. Thomas, [1976] 1 S.C.R. 906. Though the minority (H.R. Khanna and A.C. Gupta, JJ.) stuck to the view that Article 16(4) is an exception, the mejority (Ray, CJ., Mathew, Krishna Iyer and Fazal Ali, JJ.) held that Article 16(4) is not an exception to Article 16(1) but that it was merely an emphatic way of stating a principle implicit in Article 16(1). (Beg. J. took a slightly different view which it is not necessary to mention here). The said four learned Judges - whose views have been referred to in para 41 - held that Article 16(1) being a facet of the doctrine of equality enshrined in Article 14 permits reasonable classification just as Article 14 does. In our respectful opinion, the view taken by the majority in Thomas is the correct one. We too believe that Article 16(1) does permit reasonable classification for ensuring attainment of the equality of opportunity assured by it. For assuring equality of opportunity, it may well be necessary in certain situations to treat unequally situated persons unequally. Not doing so,· would perpetuate and accentuate inequality. Article 16(4) is an instance of such classification, put in to place the matter beyond controversy. The "backward class of citizens" are classified as a separate category deserving a special treatment in the nature of reservation of appointments/posts in the services of the State. Accordingly, we hold that clasue (4) of Article 16

INDRA SAWHNEY "- U.O.l. 613

is not exception to clause (1) of Article 16. It is an instance of classification implicit in and permitted by clause (1). The speech of Dr. Ambedkar during the debate on draft Article 10(3) [corresponding to Article 16(4)] in the Constituent Assembly - referred to in para 28 - shows that a substantial number of members of the Constituent Assembly insisted upon a "provision (being) made for the entry of certain communities which have so far been outside the administration'', and that draft clause (3) was put B ·_in recognition and acceptance of the said demand. It is a provision which must be read along with and in harmony with clause (1). Indeed, even without clause (4), it would have been permissible for the State to have evolved such a classification and made a provision for reservation of appointments/posts in their favoµr. Clause (4) merely puts the mailer beyond any doubt in specific terms.

Regarding the vie\v expressed in Ba/aji and Devadasan, it must be remembered that at that time it was not yet recognised by this court that Article 16(1) being a facet of Article 14 does implicitly permit classifica- tion. Once this feature was recognised the theory of clause (4) being an exception to clause (l) became untenable. It had to be accepted that clause (4) is an instance of classification inherent in clause (1). Now, just as Article 16(1) is a facet or an elaboration of the principle underlying Article 14, clause (2) of Article 16 is also an elaboration of a facet of clause (l). If clause (4) is an exception to clause (I) then it is equally an exception to clause (2). Question then arises, in \Vhat respect is clause (4) an Cx:ception to clause (2), if 'class' docs not means 'caste'. Neither clause (1) nor clause (2) speak of class. Does the contention mean that clause (1) does nut pern1il classification and therefore clause (4) is an exception to it. Thus, from any point of vic\v, the contention of the petitioners has no merit. F Question 2(b): Whether Article 16(4) is exhaustive of the concept of reseroations in favour of backlvard classes?

58Copy pinpoint58. The question than arises whether clause (4) of Article 16 is exhaustive of the topic of reservations in favour of backward classes. Before G we answer this question it is \Vell to examine the meaning and content of the expression "reservation". lls 1neaning has to be ascertained having regard to the Context in \Vhich it OCCUTS. rhe relevant words are "any provision for the reservation of appointments or posts." The question is \Vhethcr the said words contemplate only one form of provision namely H

614 SUPREME COURT REPORTS (!992J SUPP. 2 S.C.R.

A reservation simplicitor, or do they take in other forms of special provisions like preferences, concessions and exemptions. In our opinion, reservation ·is the highest form of special provision, while preference, concession and exemption are lesser forms. The Constitutional scheme and context of Article 16(4) induces us to take the view that larger concept of reservations takes within its sweep all supplemental and ancilliary provisions as also B lesser types of special provisions like exemptions, concessions and relaxa- tions, consistent no doubt with the requirement of maintenance of efficien- cy of administration - the admonition of Article 335. The several concessions, exemptions and other measures issued by the Railway Ad- ministration and noticed in Karamchari Sangh are instances of supplemen- c tay, incidental and ancilliary pruvisions made with a view to make the main provision of reservation effective i.e., to ensure that the members of the reserved class fully avail of the provision for reservation in their favour. The other type of measure is the one in Thomas. There was no provision for reservation in favour of Scheduled Castes/Scheduled Tribes in the matter of promotion to the category of Upper Division Clerks. Certain tests were required to be passed before a Lower Division Clerk could be promoted as Upper Division Clerk. A large number of Lower Division Clerks belonging to S.C./S.T. were not able to pass those tests, with the result they were stagnating in the. category of L.D.Cs. Rule 13AA was accordingly made empowering the government to grant exemption to mem- bers of S.C./S.T. from passing those tests and the Government did exempt them, not absolutely, but only for a limited period. This provision for exemption was a lesser form of special treatment than reservation. There is no reason why such a special provision should not be held to be included within the larger concept of reservation. It is in this context that the words F 11 any provision for the reservation of appointments and posts'1 assume significance. The word "any" and the associated words must be given their due meaning. They are not a mere surplusage. It is true that in Thomas it was assumed by the majority that clause (4) permits only one form of provision namely reservation of appointments/posts and that if any conces- G sions or exemptions are to be extended to backward classes it can be done only under clause (1) of Article 16. In fact the argument of the writ petitioners (who succeeded before the Kerala High Court) was that the only type of provision that the State can make in favour of the backward classes is reservation of appointments/posts provided by clause (4) and that H the said clause does not contemplate or permit granting of any exemptions

INDRA SAWHNEY v. U.O.l. 615

or concessions to the backward classes. This argument was accepted by A Kerala High Court. This Court, however, by a majority (Ray,C.J., Mathew, Krishna Iyer and Fazal Ali, JJ.) reversed the view taken by Kerala High Court, holding that such exemptions/concessions can be extended under clause (1) of Article 16. Beg, J. who joined the majority in exemption provided by impugned notification was indeed a kind of reservation and was warranted by and relatable to clause (4) of Article 16 itself. This was because - according to the learned Judge - clause (4) was exhaustive of the provisions that can be made in favour of the backward classes in the matter of employment. We are inclined to agree with the view taken by Beg, J. for the reasons given hereinabove. In our opinion, therefore, where the State finds it necessary - for the purpose of giving full effect to the provision of reservation to provide certain exemptions, concessions or preferences to members of backward classes, it can extend the same under clause (4) itself. In other words, all supplemental and ancilliary provisions to ensure full availment of provisions for reservation can be provided as part of concept of reservation itself. Similarly, in a given situation, the State may think that in the case of a particular backward class it is not necessary to provide reservation of appointments/posts and that it would be sufficient if a certain preference or a concession is provided in their favour. This can be done under clause (4) itself. In this sense, clause (4) of Article 16 is exhaustive of the special provisions that can be made in favour of 11 the backward class of citizens". Backward Classes having been classified by the E Constitution itself as a class deserving special treatment and the Constitu- tion having itself specified the nature of special treatment, it should be presumed that no further classification or special treatment is permissible in their favour apart from or outside of clause (4) of Article 16. F Question 2(c): Whether Article 16(4) is exhaustive of the very concept of reservations?

59Copy pinpoint59. The aspect next to be considered is whether clause (4) is exhaus- tive of the very concept of reservations? In other words, the question is whether any reservations can be provided outside clause (4) i.e., und~r clause (1) of Article 16. There are two views on this aspect. On a fuller consideration of the matter, we are of the opinion that clause (4) is not, and cannot be held to be, exhaustive of the concept of reservations; it is exhaustive of reservations in favour of backward classes alone. Merely H

616 SUPREME C:OlJiff REPORTS (l992j SUPP. 2 S.C.R.

A bacausc, one form of classification is stated as a specific clause, it docs not follow that the very concept and power of classification implicit in clause (1) is exhausted thereby. To say so would not be correct in principle. But, at the same time, one thing is clear. It is in very exceptional situations, - and not for all and sundry reasons - that any further reservations, of whatever kind, should be provided under clause (1). In such cases, the B State has to satisfy, if called upon, that making such a provision was necessary (in public interest) to redress a specific situation. The very presence of clause (4) should act as a damper upon the propensity to create further classes deserving special treatment. The reason for saying so is very simply. If reservations are made both under clause (4) as well as c under clause (1), the vacancies available for free competition as well as reserved categories would be correspondingly wittled down and that is not a reasonable thing to do.

Whether clause (I) of Anicle 16 does not permit any reservations?

60Copy pinpoint60. For the reasons given in the preceding paragraphs we must reject the argument that clause (l) of Article 16 permits only extending of preferen- ces, concessions and exemptions, but does Dot permit reservation of appoint- ments/posts. As pointed out in para (54) the argument that no reservations can be made under Article 16(1) is really inspired by the opinion of Powell, J. E in Bakke. But in the very same paragraph we had pointed out that it is not the unanimous opinion of the Court. In principle, we see no basis for acceding to the said contention. What kind of special provision should be made in favour of a particular class is a matter for the State to decide, having regard to the facts and circumstances of a given situation - subject, of course, to the observations in the preceding paragraph. F PART - W (QUESTIONS 3, 4 AND 5)

Question 3(a): Meaning of the expression "Backward Class of citizens" in Anic/e 16(4).

61Copy pinpoint61. What does the expression "Backward Class of Citizens" in Article 16( 4) signify and how should they be identified? This has been the single- most difficult question tormenting this nation. The expression is not defined in the Constitution. What does it mean then? The arguments before us mainly revolved round this question. Several shades of opinion have been presented to u.s ranging from one extreme to the other. Indeed,

INDRA SAWHNEY v. U.0.1. 617

it may be difficult to set out in full the reasoning presented before us orally and in several written propositions submitted by various counsel. We can mention only the subslance of and the broad features emerging from those submissions. At one end of the spectrum stands Sri N.A.Palkhiwala (sup- ... ported by several other counsel) whose submissions may briefly be sum- marised in the following words: a secular, unified and caste-less society is a basic feature of the Constitution. Caste is a prohibited ground of distinc- B tion under the Constitution. It ought be erased atogether from the Indian Society. It can never be the basis for determining backward classes referred to in Article 16( 4). The Report of the Manda! Commission, which is the basis of the impugned Memorandums, has treated the expression "back- ward classes" as synonymous with backward castes and has proceed to c identify backward classes solely and exclusively on the basis of caste, ignoring all other considerations including poverty. It has indeed invented castes for Non-Hindus where none exists. The Report has divided the nation into two sections, backward and forward, placing 52% of the population in the former section. Acceptance of Report would spell dis- D aster to the unity and integrity of the nation. If half of the posts are reserved for backward classes, it would seriously jeopardise the efficiency of the administration, educational system, and all other sevices resulting in back- wardness of the entire nation. Merit will disapear by deifying backward- ness. Article 16(4) is broader than Article 15(4). The expression "backward class of citizens" in Article 16(4) is not limited to "socially and educationally E backward classes" in Article 15(4). The impugned Memorandums, based on the said report must necessarily fall to the ground along with the Rcporl. Jn fact the main thrust of Sri Palkhiwala's argument has been against the Manda) Commission Report. F

62Copy pinpoint62. Sri K.K.Yenugopal appearing for the petitioner in Writ Petition No.930 of 1990 adopted a slightly different approach while reiterating that the expression "backward classes of citizens" in Article 16(4) cannot be constured as backward castes. According to him, backwardness may be social and educational and may also be economic. The authority appointed to identify backward classes must first settle the criteria or the indicators G for determining backward classes and then it must apply the said criteria to each and every group in the country. In the course of such identification, it may well happen that certain castes answer and satisfy the criteria of backwardness and may as a whole qualify for being termed as a backward class. But it is not permissible to start with castes to determine whether a H

618 SUPREME COURT REPORTS [1992] SUPP. 2 S.C.R.

A caste is. a backward class. He relied upon the provision in clause (2) of Article 38 and Article 46 to say that the objective is to minimize the inequalities in income not only among individuals but also among groups of persons and to help the weaker sections of the society. The economic criterion is an important one and must be applied in determining backward classes and also for excluding those sections or identified groups who may B for the sake of convenience be referred to as the 'creamy layer'. Since castes do not exist among Muslims, Christians and Sikhs, caste can never be the basis of identification. The learned counsel too pointed out the alleged basic errors in the approach adopted by and conclusions arrived at by the Manda! Commission. c Smt. Shyamala Pappu also took the stand that caste can never be the basis for identification. According to her, survey to identify backward classes should be from individual to individual; it cannot be caste-wise. To the same effect are the submissions of Sri P.P. Rao appearing for the D Supreme Court Bar Association. According to him, the only basis for identifying backward classes should be occupation-cum-means as was done in the State of Karnataka at a particular stage which aspect is dealt with and approved by this court in Chitra/ekha & Ors. v. State of Mysore, (1964] 6 S.C.R.368. A secular socialist society, he submitted, can never counterance identification of backward classes on the basis of caste which E would only perpetuate and accentuate caste differences and generate antagonism and antipathy between castes.

63Copy pinpoint63. At the other end of the spectrum stands Sri Ram Jethmalani, counsel appearing for the State of Bihar supported by several other coun- F sel. According to him, backward castes in Article 16(4) meant and means only the members of Shudra casts which is located between the three upper castes (Brahmins, Kshatriyas and Vaishyas) and the out-castes (Panchamas) referred to as Scheduled Castes. According to him, Article 16(4) was conceived only for these "middle castes" i.e., castes categorised as shudras in the caste system and for none else. These backward castes G have suffered centuries of discrimination and disadvantage, leading to their backwardness. The expression "backward classes 11 does not refer to any current characterstic of a backward caste save and except paucity or inadequacies of representation in the apparatus of the Government. Pover- ty is not a necessary criterion of backwardness; in is in fact irrelevant. The H provision for reservation is really a programme of historical compensation.

INDRA SAWHNEY v. U.O.L 619

It is neither a measure of economic reform nor a poverty alleviation programme. The learned counsel further submitted that it is for the State to determine who are the backward classes; it is not a matter for the court. The decision of the Government is not judicially reviewable. Even if reviewable, the scope of judicial review is extremdy limited - to the only question whether the exercise of power is a fraud on the Constitution. The learned counsel referred to certain American decisions to show that even in that country several programmes of affirmative action and compensatory discrimination have been evolved and upheld by courts.

64Copy pinpoint64. Dr. Rajiv Dhawan, learned counsel appearing for Srinarayana Dharama Paripalana Yogam (an association of Ezhavas in Kerala) sub- C mitted that Article 16(4) and 15(4) occupy different fields and serve different purposes. Whereas Article· 15(4) contemplates positive action programmes, Article 16(4) enables the State to undertake schemes of positive discrimination. For this reason, the class of intended beneficiaries under both the clauses is different. The social and educational backward- ness which is the basis of identifying bakcwardness under Article 15(4) is only partly true in the case of 'backward class of citizens' in Article 16(4). The expression nany backward class Of citizens" occurring in Article 16(4} must be understood in the light of the purpose of the said clause namely, empowerment of those groups and classes which have been kept out of the administration - classes which have suffered historic disabilites arising from discrimination or disadvantage or both and who must now be provided entry into the administrative apparatus. In the light of the fact that the Scheduled Castes and Scheduled Tribes were also intended to be beneficiaries of Article 16( 4) there is no reasor: why caste cannot be an exclusive criteria for determining beneficiaries under Article 16(4). Coun- F sel emphasised the fact that Article 16(4) speaks of group protection and not individual protection.

Sri R.K.Garg appearing for the Communist Party of India, an Inter- venor, submitted that caste plus poverty plus location plus residence should be the basis of identification and not mere caste. According to the learned G counsel, a national 1,;oncensus is essential to introduce reservations for 'orther backward classes' under Article 16(4) and that efforts must be made to achieve such a concensus.

65Copy pinpoint65. Sri Siva Subramanium appearing for the State of Tamil Nadu H

620 SUPREME COURT REPORTS [1992] SUPP. 2 S.C.R.

A ·supported the Mandal Commission Report in its entirety. According to him, backward classes must be identified only on the basis of caste and that no economic criteria should be adopted for the said purp~se. He submitted that economic criteria may be employed as one of the indicators for identification of backward classes but once a backward class is identified as such, there is no question of excluding any one from that class on the - B basis of income or means or on any other economic criterion. He referred to the history of reservations in the province of Madras prior to inde- pendence and now it has been working there successfully and peacefully over the last several decades.

c Sri P.S.Poti appearing for the State of Kerala ~upported the iden- tificaion of backward classes solely and exclusively on the basis of caste. He submitted that the caste system is scientifically organised and practiced in Kerala and, therefore, furnishes a perfectly scientific basis for identifica- tion of backward classes. He submitted that besides the vice of un- touchability, another greater vice of 'unapproachability' was also being practiced in that State.

Sri Ram Awadesh Singh, M.P., President of Lok Dal and President of All India Federation of Backward Classes, Scheduled Castes, Scheduled Tribes and Religious minorites submitted that caste should be the sole criteria for determining backwardness. He referred to c..:enturies of injustice meted out by upper castes to shudras and panchamas and submitled that these castes must now be given a share in the governance of the country which alone will assure their dignity besides instilling in them a sense of confidence and a spirit of competition.

66Copy pinpoint66. Sri K.Parasaran, learned counsel appearing for the Union of India urged the following submissions:

(!)The reservation provided for by clause (4) of Article 16 is not in favour of backward citizens, but in favour of backward class of citizens. What is to be identified is backward class of citizens and not citizens who can be classified as backward. The homogeneous groups based on religion, race, caste, place of birth etc. can form a class of citizens and if that class is backward there can be a reservation in favour of that class of citizens.

(2) Caste is a relevant consideration. It can even be the dominant consideration. Indeed, most of the lists prepared by the States are prepared

INDRA SAWHNEY v. U.0.1. 621

with reference to and on the basis of castes. They have been upheld by this court.

(3) Article 16(2) prohibits discrimination only on any or all of the grounds mentioned therein. A provision for protective discrimination on any of the said grounds coupled with other relevant grounds would not fall within the prohibition of clause (2). In other words, if reservation is made in favour of backward class of citizens the bar contained in clause (2) is not attracted, even if the backward classes are identified with reference to castes. The reason is that the reservation is not being made in favour of castes simplicitor but on the ground that they are backward castes/classes which are not adequately represented in the services of the State. c (4) The criteria of backwardness evolved by Manda! Commission !s perfectly proper and unobjectionable. It has made an extensive investiga- tion and has prepared a list of backward classes. Even if there are instances of under-inclusion or over-inclusion, such errors do not vitiate the entire D exercise. Moreover, whether a particular caste or class is backward or not and whether it is adequately represented in the services of the State or not are questions of fact and are within the domain of the executive decision.

67Copy pinpoint67. In paragraphs 33 to 42, we have noticed how this court has been grappling with the problem over the years. In Venkataraman's case, a E Seven-Judge Bench of this court" noticed the list of backward cl~sses mentioned in Schedule III to the Madras Provincial and Subordinate Service Rules, 1942, as also the fact that backward classes were enumerated on the basis of caste/race. It found no objection thereto though in Cham- pakam, rendered by the same Bench and on the same day it found such a F classification bad under Article 15 on the ground that Article 15 did not contain. a clause corresponding to clause ( 4) of Article 16. In Venkataraman's case this court observed that in respect of the vacancies reserved for backward classes of Hindus, the petitioner (a Brahmin) cannot have any claim inasmuch as "those reserved posts (were reserved) not on the ground of religion, race, caste etc. but because of the necessity for G malcing a provision for reservation of such post in favour of a backward class of citizens." The writ petition was allowed on the ground that the allocation of vacancies to and among communities other than Harijans and backward classes of Hindus cannot be sustained in view of clauses (1) and (2) of Article 16. H

622 SUPREME COURT REPORTS (1992] SUPP. 2 S.C.R.

68Copy pinpoint68. Though Balaji was not a case arising under Article 16(4), what it said about Article 15( 4) came to be accepted as equally good and valid for the purpose of Article 16( 4). The formulations enunciated with respect to Article 15(4) were, without question, applied and adopted in cases arising under Article 16(4). It is, therefore, necessary to notice precisely the formulations in Balaji relevant in this behalf. Gajendragadkar,J. speaking B for the Constitution Bench found, on an examination of the Nagangowda Committee Report, "that the Committee virtually equated the class with the castes." The learned Judge then examined the scheme of Article 15, the meaning of the expression 'class', the importance of caste in the Hindu social structure and observed, while dealing with social backwardness: c "Therefore, in dealing with the question as to whether any class of citizens is socially backward or not, i: may not be irrelevant to consider the caste of the said group of citizens .........Though the caste of the group of citizens may be relevant, its importance should not be exaggerated. If D the classification of backward classes of citizens was based solely on the caste of the citizen, it may not always be logical and may perhaps contain the vice of perpetuating the caste themselves. 11

E The learned Judge further proceeded to hold:

"Besides, if the caste of the group of citizens was made the sole basis for determining the social backwardness of the said group, the test would inevitably break down in relation to many sections of Indian society which do not recognise castes in the c0nventional sense known to Hindu society. How is one going to decide whether Muslims, Christians or Jains or even Lingayats are socially backward or not? The test of castes would be inapplicable to those groups, but that would hardly justify the exclusion of these groups in to to from the operation of Art.15(4). It is not uolikely that in some States some Muslims or Christians or Jains forming groups may be socially backward. That is why we think that though castes in relation to Hindus may be a relevant factor to consider in determining the social back- H wardness of groups or class of citizens, it cannot be made

INDRA SAWHNEY v. U.0.1. 623

the sole or the dominant test in that behalf. Social back- A wardness is in the ultimate analysis the result of poverty to a very large extent... .. .!! is true that social backwardness which results form poverty is likely to be aggravated by considerations of caste to which the poor citizens may belong, but that only shows the .relevance of both caste and poverty in determining the backwardness of citizens. 11 B

The learned Judge stressed the part played by the occupation, con- ventional beliefs and place of habitation in determining the social back- wardness. Inasmuch as the identification of backward classes of Nagangowda Committee was based almost solely on the basis of caste, it was held to be bad.

'The criticism of the Respondents' counsel against the Judgment runs thus: While it recognises the relevance and significance of the caste and the integral connection between caste, poverty and social backwardness, it yet refuses to accept caste as the sole basis of identifying socially backward classes, partly for the reason that castes do not exist among non-Hindus. The Judgment does not examine whether caste can or cannot form the starting- point of process of identification of socially backward classes. Nor does it consider the aspect - how does the non-existence of castes among non-Hindus (assuming that the said premise is factually true) makes it irrelevant in the case of Hindus, who constitute the bulk of the country's population. There is no rule of law that a test of basis adopted must be uniformly applicable to the entire population in the country as such.

Before proceeding further it may be noticed that Ba/aji was dealing with Article 15(4), which clause contains the qualifying words "socially and educationally" preceding the expression "backward classes". Accordingly, it was held that the backwardness contemplated by Article 15(4) is both social and educational. Though, clause (4) of Article 16 did not contain any such qualifying words, yet they came to be read into it. ln!anaki Prasad Parimoo, Palekar, J. speaking for a Constitution Bench, took it as "well-set- G tied that the expression 'backward classes' in Article 16(4) means the same thing as the expression 'any socially and educationally backward class of citizens' in Article 15(4)". More of this later.

69Copy pinpoint69. In Minor P.Rajendran, the caste vis-a'vis class debate took a sharp turn. The ratio in this case marks a definite and clear shift in emphasis. H

624 SUPREME COURT REPORTS f1992J SUPP. 2 S.C.R.

A (We have dealt with it at some length in para 36). Suffice it to mention here that in this decision, it was held that "a caste is also a class of citizens and if the caste as a whole is socially and educationally backward reserva- tion can be made in favour of such a caste on the ground that it is a socially and educationally backward class of citizens within the meaning of Article 15(4) ....... .It is true that in the present case the list of socially and educa- B tionally backward classes has been specified by caste. But that does not necessarily mean that caste was a sole consideration and that persons belonging to these castes are also not a class of socially and educationally backward citizens." This principle was reiterated in Peeriakampan. Bairam and Trilokinath-I/. We have referred to these decisions at some length in paras 38 and 39. In Peeriakamppan, Hegde,J. concluded, "a caste has always been recognised as a class."

70Copy pinpoint70. This issue was gone into jn some detail in Vasant Kumar, where all the five Judges constituting the Constitution Bench expressed different opinions. Chandrachud,CJ. did not express himself on this aspect but other four learned Judges did. Desai, J. recognised that "in the early stages of the functioning of the Constitution, it was accepted without dissent or dialogue that caste furnishes a working criterion for identifying socially and educationally backward class of citizens for the purpose of Art.15(4)." He also recognised that "there has been some vacillation on the part of the judiciary on the question whether the caste should be the basis for recog- nising the backwardness." After examining the significance of caste in the Indian social structure, the learned Judge observed:

F "Social hierarchy and economic position exhibit an indis- putable mutuality. The lower the caste, the poorer its member. The poorer the members of a caste, the lower the caste. Caste and economic situation, reflecting each other as they do are the Deus ex-Machina of the social status occupied and the economic power weilded by an individual G or class in rural society. Social status and economic power are so woven and fused into the caste system in Indian rural society that one may without hesitation, say that if poverty be the cause, caste is the prilnary index of social backward- ness, so that social backwardness is often readily identifi- H able with reference to a person's caste. 11

INDRA SAWHNEY v. U.0.1. 625

The learned Judge also recognised that caste system has even A penetrated other religions to whom the practice of caste should be anathema. He observed:

"So sadly and oppressively deep-rooted is caste in our country that it has cut across even the barriers of religion. The caste system has penetrated other religious and dis- B sentient Hindu sects to whom the practice of caste should be anethma and today we find that practitioners of other religious faiths and Hindu dissentients are some times as rigid adherents to the system of caste as the conservative Hindus. We find Christians Harijans, Christian Madars, Christian Reddys, Christian Kamm as, Mujbi Sikhs, etc. etc. c InAndhra Pradesh there is a community known as Pinjars or Dudekulas (known in the North as 'Rui Pinjane Wala'): (professional cottonbeaters) who are really Muslims but are treated in rural society, for all practical purposes, as a Hindu caste. Several other instances may be given." D

Having thus noticed the pernicious effects of the caste system, the learned Judge opined that the only remedy in such a situation is to devise a method for determining social and educational backward classes without reference to caste. He stressed the significance of economic criterion and E of poverty and concluded that a time has ·come when the economic criterion alone should be the basis ior identifying the backward classes. Such an identification has the merit of advancing the secular character of the nation and will tend towards nuilifying caste influence, said the learned Judge. F

71Copy pinpoint71. Chinnappa Reddy,J. aealt with the question at quite some length. The learned Judge quoted Mh Weber, according to whom the three dimensions of social inequality· are class, status and power - and stressed the importance of poverty in this matter. Learned Judge opined that caste system is closely entwined with economic power. In the words of the G learned Judge:

nsocial status and economic power a:re so woven and fused into the caste system in Indian rural society that one may without hesitation, say that if poverty be the cause, caste is the primary index of social backwardness, so that social H

626 SUPREME COURT REPORTS (1992] SUPP. 2 S.C.R.

A backwardness is often readily identifiable with reference to a person's caste.'1

The learned Judge too recognised the percolation of caste system into other religions and concluded his opinion in the following words:

B ''Poverty, caste, occupation and habitation are the principal factors which contribute to brand a class as socially back- ward .........But mere poverty it seems is not enough to invite the Constitutional branding, because of the vast majority of the people of our country are poverty-struck but some among them are socially and educationally forward and c others backward ........ True, a few members of those caste or social groups may have progressed far enough and forged ahead so as to compare favourably with the leading forward classes economically, socially and educationally. In such cases, perhaps an upper income ceiling would secure the benefit of reservation to such of those members of the class who· really deserve it ...... Class poverty, not individual poverty, is therefore the primary test... .. ~ Once the relevant conditions are taken into consideration and the backwardness of a class of people is determined, it will not be for the court to interfere in the matter. But, lest there be any misunderstanding, judicial review will not stand excluded."

72Copy pinpoint72. A.P.Sen,.J. dealt with this question in a short opinion. According to him: F ".... .The predominant aP.d only factor for making special provisions under Art.15(4) or for reservation of posts and appointments under Art.16{4) should be poverty, and caste or a sub-caste or a group should be used only for purposes of identification of persons comparable' to Scheduled Cas- G tes or Scheduled Tribes, till such members of backward classes attain a state of enlightenment and there is eradica- tion of poverty amongst them and they become equal partners in a new social order in our national life. 11

73Copy pinpoint73. "E.S.Venkataramiah,J. too dealt with this aspect at some length.

INDRA SAWHNEY v. U.0.1. 627

After examining the origins of the caste and the ugly practices associated with it, the learned Judge opined:

An examination c: the question in the background of the 11

Indian social conditions shows that the expression 'back- ward classes' used in the Constitution referred only to those who were born in particular castes, or who belonged to particular races or tribes or religious mii1orities which were 11 backward.

The learned Judge then referred to the debates in the Constituent Assembly on draft Article 10 and other allied articles, including the speech of Dr. Ambedkar and observed thus: c "The whole tenor of discussion in the Constituent Assembly pointed to making reservation for a minority of the popula, tion including Scheduled Castes and Scheduled Tribes which were socially backward. During the discussion, the D Constitution (First Amendment) Bill by which Article 15(4) was introduced, Dr. Ambedkar referred to Article 16(4) and said that backward classes are 'nothing else but a collection of certain castes. This statement leads to a reasonable inference that this was the meaning which the E Constituent Assembly assigned to classes' at any rate so far as Hindus were concerned.11

The learned Judge also supported the imposition of a means test as was done by the Kerala Government in KS.Jayasree & Anr. v. State of Kera/a & Anr., [1977] 1 S.C.R.194. F The aboVe opinions emphasise the integral connection between caste, occupation, poverty and social backwardness. They recognise that in the Indian context, lower castes are and ought to be treated as backward classes. Rajendran and Vasant Kumar (opinions of Chinnappa Reddy and G Venkataramiah, JJ.) constitute important milestones on the road to recog- nition of relevance and significance of caste in the context of Article 16(4) and Article 15(4).

74Copy pinpoint74. At this stage, it would be fruitful to examine, how he words "caste" and class 11 were understood in pre-Constitution India. We shall first refer H 11

628 SUPREME COURT REPORTS [1992] SUPP. 2 S.C.R.

A to various Rules in force in several parts of India, where these expressions were used and notice how were these expressions defined and understood. In the Madras Provincial and Subordinate Service Rules, 1942, framed by the Governor of Madras under Section 241(2)(b) read with 255 and 275 of the Government of India Act. 1935, the expression "backward classes" was defined in clause 3(A) of Rule 2. (The porvinces of Madras at that time B covered not only the present State of Tamil Nadu but also a major portion of the present State of Andhra Pradesh and parts of present States of Kerala and Karnataka.) The definition read as follows:

"3(A)."Backward classes" means the communities men- c tioned in Schedule III of this part."

Schedule III bore the heading "backward classes". It was a collection of castes and tribes under the sub-heading '1race, tribe or caste. 11 The backward classes in the Schedule not only included the backward castes and tribes in Hindu religion but also certain sections of Muslims in the D nature of castes. For example, item (23) in Schedule lII referred to 'Dudekula' who, as is well known, is a socially disadvantaged section of Muslims - in effect, a caste - pursuing the occupation of ginning and deaning of cotton and preparing pillows and mattresses. In this connection, reference may be had to Chapter III - 'History of the Backward Classes Movement in Tamil Nadu' - of the Report of the Tamil Nadu Second E Backward Classes Commission (1985), which inter alia refers to formation of 'The Madras Provincial Backward Classes League, an association rep- resenting the various backward Hindu communities' in 1934 and its demand for separate representation for them in services.

The former State of Mysore was one of the earliest States, where F certain provisions were made in favour of Backward Classes. The opinion of E.S.Venkataramiah, J. in Vasant Kumar, (at pages 442-443) traces briefly the history of reservations in the State of Mysore from 1918-21 upto the re-organisation of State. The learned Judge points out how the expression 'backward classes' and 'backward communities' were used interchangeably. G All the castes/communities' except Brahmins in the State were notified as backward communities/castes. As far back as 1921, preferential reeruit- men~ was provided in favour of "backward communities", in Government services.

In Bombay province, the Government of Bombay, Finance Depart- H men! Resolution No.2610 dated 5.2.1925 defined "Backward Classes" as all

INDRA SAWHNEY v. U.0.1. 629

except Brahmins, Prabhus, Marwaris, Parsis, Banyas and Christians. Cer- A tain reservations in Government service were provided for these classes. In 1930, the State Committee noticed the over-lapping meanings attached to the expressions "depressed classes" and 11'ackward classes11 and recom- mended that "Depressed Classes" should be used in the sense of untouch- ables, a usage which "will coincide with existing common practice." They proposed that the wider group should be called "Backward Classes", which B should be subdivided into Depressed Classes (i.e., untouchables); Aboriginals and Hill Tribes; Other Backward Classes {including wandering tribes). They opined that the groups then currently called Backward Clas- ses should be renamed "intermediate classes". In addition to 36 Depressed classes (approximate 1921 population 1.475 millions) and 24 Aboriginal c and Hill Tribes (approximate 1921 population 1.323 millions), they listed 95 Other Backward Classes {approximate 1921population1.041 millions)".

75Copy pinpoint75. In the former princely State of Travancore, the expression used was "Communities", as would be evident from the Proceedings of the Government of His Highness the Maharaja of Travancore, contained in D Order R. Dis. N.893/general dated Trivandrum, 25th June, 1935. It refers to earlier orders on the subject as well. What is significant is that the expression "communities" was used as taking in Muslims and certain sec- tions of Christians as well; it was not understood as confined to castes in Hindu social system alone. The operative portion of the order reads as follows: E

".....Accordingly, Government have decided that all com- munities whose population is approximately 2 per cent of the total population of the State or about one lakh, be recognised as separate communities for the purpose of F recruitment to the public service. The only exception from the above rule will be the Brahmin community who, though forming only 1.8 per cent of the total population, will be dealt with as a separate community. On the above basis the classification of communities will be as follows:- G A. HINDU

1. Brahmin.

2. Nayar.

3. Other Caste Hindu.

4. Kummula. H

630 SUPREME COURT REPORTS (1992] SUPP. 2 S.C.R.

A 5. Nudar.

6. Ezlmva.

7. Cheramar (Pulaya)

8. Other Hindu.

B. MUSLIM. B C. CHRISTIAN.

1. Jacobite.

2. Marthomite.

3. Syriac Catholic. c 4. Latin Catholic.

5. South India United Church.

6. Other Christian."

In the then United Provinces, the term "Backward Classes" was understood as covering both the untouchable classes as well other "Hindu D Backward" classes. Marc Galanter says:

"The United Provinces Hindu Backward Classes League {founded in 1929) submitted a memorandum which sug- gested that the term "Depressed" carried a connotation "of untouchability, in the sense of causing pollution by touch as in the case of Madras and Bombay" and that many communities were ·reluctant to identify themselves as depressed. The League suggested the term "'Hindu' Backward"' as a more suitable nomenclature. The list of 115 castes submitted included all candidates from the untouchable category as well as a stratum above. "All of the listed communities belong to non-Dwijas or degenerate or Sudra classes of the Hindus." They were described as low socially, educationally and economically and were said to number over 60% of the population." G The expression !!depressed and other backward classes" occurs in the Objectives Resolution of the Constituent Assembly moved by Jawaharlal Nehru on December 13, 1946.

76Copy pinpoint76. We may also refer to a speech delivered by Dr. Ambedkar on H May 9, 1916 at the Columbia university of New York, U.S.A. on the subject

INDRA SAWHNEY v. U.0.1. 631

"castes in India: their mechanism, genesis and development" (the speech was published in Indian Antiquary-May 1917-Vol.XLI), which shows that as early as 1916, 11 class" and 11 caste" were used inter-changeably. In the course of the speech, he said:

"....society is always composed of classes. It may be an exaggeration to assert the theory of class-conflict, but the existence of definite classes in a society is a fact. Their basis illay differ. They may be economic or intellectual or social, but an individual in a society is always a member of a class. This is a universal fact and early Hindu society could not have been an exception to this rule, and, as a matter of fact, we know it was not. If we bear this generalization in mind, c our study of the genesis of caste would be very much facilitated, for we have only to determine what was the class that first made itself into a caste, for class and caste, so to say, are next door neighbours, and it is only a span that separates the two. A Caste is an Enclosed Class." D

A little later he staied:

"We shall be well advised to recall at the outset that the Hindu society, in common with other societies, was c0m- posed of classes and the earliest known are the (1) Brah- E mins or the priestly class; (2) the Kshatriya, or the military class; (3) the Vaishya, or the merchant class and (4) the Shudra or the artisan and menial class. Particular attention has to be paid to the. fact that this was essentially a class system, in which individuals, when qualified, could change their class, and therefore classes did change their person- nel. At some time in the history of the Hindus, the priestly class socially detached itself from the rest of the body of people and through a closed-door policy became a caste by itself. The other classes being subject to the law of social division of labour underwent differentiation, some into large, other~ into very minute groups."

77Copy pinpoint77. In Encyclopaedia Britannica Vol.16, the following statement oc- curs under the heading "Slavery, Serfdom and Forced labour" under the sub-heading "servitude in Ancient India and. China~ - 11 castes in India." 11 H

632 SUPREME COURT REPORTS [~92] SUPP. 2 S.C.R.

A "More abundant than slavery were serfdom. Within the rigid classification of social classes in ancient India, the Sudra caste was obliged to serve the Ksatriya, or warrior caste, the Brahmins, or priests, and the Vaisyas, or farmers, cattle raisers and merchants. There is an unbreakable barrier, however, separating these castes from the inferior B Sudra caste, the descendants of .the primitive indigenous people who lived in serfdom.

In those times it was not a person's economic wealth that gave him his social rank but rather his social and racial c level; and thus one of the Mano's laws says" Although able, a Sudra must not acquire excess riches, since when a Sudra acquires a fortune, he vexes the Brahmans with his in- solence." The barrier separating the servile castes took on extreme cruelty in some laws:

D The legal condition of the Sudra left him only death as a means of improving his condition."

In Legal Thesaurus (Regular Edition) the following meanings are 1 given to the word 'class": E Assortment, bracket, branch, brand, breed, caste, category, classification, classes, denomination, designation, divi- sion ..... ; gradation, grade, group, grouping hierarchy..... . sect, social rank, social status ..... ".

F The following meanings are given to the word "caste" in Webster's English Dictionary:

"(1) a race, stock, or breed of men or animals (2): one of the hereditary classes into which the society of India is divided in accordance with a system fundamental to Hin- G duism, reaching back into distant antiquity, and dictating to every orthodox Hindu the rules and restrictions of all social intercourse and of which each has a name of its own and special customs that restrict that occupation of its members and their intercourse with the members of the other classes (3)(a): a division or class of society comprised

INDRA SAWHNEY v. U.0.1. 633

of persons within a separate and exclusive order based variously upon differences of wealth, inherited rank or privilege, profession, occupalion ...... (b) the position con- ferred by caste standing. (4) a system of social statification more rigid than a class and characterized by hereditary status, endogamy and social barriers rigidly sanctioned by custom law or' religion." B

All the above material does go to show that in pre-Independence India, the expressions 'class' and 'caste' were used interchangeably and that caste was understood as an enclosed class. c

78Copy pinpoint78. We may now turn to Constituent Assembly debates with a view to ascertain the original intent underlying the use of words "backward class of citizens". At the outset we must clarify that we are not taking these debates or even the speeches of Dr. Amhedkar as conclusive on the meaning of the expression 11backward classes." We are referring to these D debates as furnishing the context in which and the objective to achieve which this phrase was put in clause (4). We are aware that what is said during these debates is not conclusive or binding upon the court because several members may have expressed several views, all of which may not be reflected in the provision finally enacted. The speech of Dr. Ambedkar on this aspect, however, stands on a different footing. He was not only the E Chairman of the Drafting Committee which inserted the expression "back- ward" in draft Article 10(3) [it was not there in the original° draft Article 10(3)], he was virtually piloting the draft Article. In his speech, he explains the reason behind draft clause (3) as also the reason for which the Drafting Committee added the expre~sion 11 backward" in the clause. In this situation, F we fail to understand how can anyone ignore his speech while trying to ascertain the meaning of the said expression. That the debates in Con- stituent Assembly can be relied upon as an aid to interpretation of a constitutional provision is borne out by a series of decisions of this Court. See Madhu Limaye, A.LR. 1969 S.C. 1014 at 1018; Go/aknath v. State of Punjab, A.LR. 1967 S.C.1643 at 1657 (Subba Rao, CJ.); opinion of Sikri, G CJ., in Dhillon v. Union of India, [1972] 2 S.C.R.33 and the several opinions in Keshavananda Bharati, [1973] 4 S.C.C.225 where the relevance of these debates is pointed out, emphasising at the same time, the extent to which and the purpose for which they can Le referred to). Since the expression "backward" or "backward class of citizens" is not defined in the Act, H

634 SUPREME COURT REPORTS [1992] SUPP. 2 S.C.R.

A reference to such debates is permissible to ascertain, at any rate, the context, background and objective behind them. Particularly, where the Court wants to ascertain the 'original intent' such reference may be un- avoidable.

79Copy pinpoint79. According to Dr. Ambedkar (his speech is referred in para 28 B and need not be reproduced here), the Drafting Committee was of the opinion that such a qualifying expression was necessary to indicate that the classes of citizens for whom reservations were to be made are those Tlcommunities which have not had so far representation in the State. 11 It was also of the opinion that without such a qualifying expression (like C 'backward') the "exemption made in favour of reservation will ultimately eat up the rule altogether". This was also the opinion of Sri K.M.Munshi, who too was a member of the Drafting Committee. In his speech (referred to in para 27) he explains why the said qualifying expression "backward" was inserted by the Drafting Committee in draft Article 10(3). His speech, in so far as it is relevant on this aspect, has been quoted in extenso in para D 28 and need not be repeated here.

In our opinion too, the words 1'class of citizens - not adequately represented in the services under the State" would have been a vague and uncertain description. By adding the word "backward" and by the speeches of Dr. Ambedkar and Sri K.M.Munshi, it was made clear that the "class of E citizens ........ not adequately represented in the services under the State" meant only those classes of citizens who were not so represented on account of their social backwardness.

Reference can also be made in this context to the speech of Dr.Am- F bedkar in the Parliament at the time the First Amendment to the Constitu- tion was being enacted: It must. be remembered that the Parliament which enacted the First Amendment was the very same Constituent Assembly which framed the Constitution and Dr. Ambedkar as the Minister of Law was piloting the Bill. He said that backward classes "are nothing else but G a collection of certain castes". (the relevant portion of his speech is referred to in papa 32) and that it was for those backward classes that Article 15(4) was being enacted.

80Copy pinpoint80. Pausing here, we may be permitted to make a few observations. The speeches of Dr. Ambedkar may have to be understood in the context of the H then obtaining ground realities viz.. (a) Hindus constituted 84% of the total

INDRA SAWHNEY v. U.0.1. 635

population of India. And among Hindus, caste discrimination was unfor- A tunately an unpleasant reality; (b) caste system had percolated even the Non-Hindu religions - no doubt to varying extents. Particularly among Chris- tians in Southern India, who were converts from Hinduism, it was being practised with as much rabidity as it was among Hiridus. (This aspect has been stressed by the Manda! Commission (Chapter 12paras11to16) and has also been judicially recognised. (See, for inst:ince, the opinions of Desai and B Chinnappa Reddy, JJ. in Vasant Kumar). Encyclopaedia Britannica-Il- Micropaedia refers to existence of castes among Muslims and Christians at pages 618 and 619. Amo!lg Muslims, it is pointed out, a distinction is made between 'Ashrats' (supposed to be descendants scondants of Arab im- migrants) and non-Ashrafs (native converts). Both are divided into sub- c groups. Particularly, the non-Ashrafs, who are converts from Hinduism, it is . pointed out, practice caste system (including endogamy)" in. a manner close to that of their Hindu counter-parts." All this could not have been unknown

- to Dr_. Ambedkar, the keen social scientist that he was.

( c) It is significant to notice that throughout his speech in the Constituent Assembly, Dr. Amb~dkar was using the word 11 communities 11 (and not 'castes') which expression includes not only the castes among the D

Hindus but several other groups. For example, Muslims as a whole were treated as a backward community in the princely State of Travancore besides several sections/denominations among the Christians. The word E "community" is clearly wider than 11 caste 11 - and "backward communities" meant not only the castes - wherever they may be found - but also other groups. classes and sections among the populace.

81Copy pinpoint81. Indeed, there are very good reasons why the Constitution could not have used the expression "castes" or "caste" in Article 16(4) and why the word "class" was the natural choice in the context. The Constitution was meant for the entire country and for all time to come. Non-Hindu religions like Islam, Christianity and Sikh did not recognise caste as such though, as pointed out hereinabove, castes did exist even among these religions to a varying degree. Further, a Constitution is supposed to be a permanent document expected to last several centuries. It must surely have been envisaged that in future many classes may spring-up answering the test of backwardness, requiring the protection of Article 16(4). It, therefore, follows that from the use of the word "class" .in Article 16(4), it cannot be concluded either that "class" is antithetical to '1casten or that a caste cannot H

636 SUPREME COURT REPORTS [199i] SUPP. 2 S.C.R.

A be a class or that a caste as such can never be taken as a backward class of citizens. The word "class" in Article 16(4), in our opinion, is used in the sense of social class - and not in the sense it is understood in Marxist Jargon. -- In Rajendran, Trilokinath-II, Bairam and Peerikarupan, this reality B was recognised and given effect to, notwithstanding the fact that they had to respect and operate within the rather qualified formulation of Ba/aji.

For the sake of completeness, we may refer to a few passages from Vasant Kumar to show what does the concept of 'caste' signify? DA. Deasi, J. defines and describes "caste" in the following terms: c What then is a caste? Though caste has been discussed by scholars and jurists, no precise definition of the expression has emerged. A caste is a horizontal segmental division of society spread over a district or a region or the whole State D and also sometimes outside it. Homo Hierarchicus is ex- pected to be the central and substantive element of the caste/system which differentiate it from other social sys- tems. The concept of purity and impurity conceptualises the caste system .......There are four essential features of the caste system which maintained its homo hierarchicus char- E acter: (1) hierarchy (2) commensality: (3) restrictions on marriage; and (4) hereditary occupation. Most of the caste are endogamous groups. Intermarriage between two groups is impermissible. But 'Pratilom' marriages are not wholly known." F Venkataramiah,J. also defined "caste" in practically the same terms. He said:

11 · A caste is an association of families which practice the custom of endogamy i.e. which permits marriages amongst G the members belonging to such families only. Caste rules prohibit its members from marrying outside their caste ..... . A caste is based on various factors. Sometimes it may be a class, a race or a racial uDil. A caste has nothing to do with wealth. The caste of a person is governed by his birth, in a H family. Certain ideas of ceremonial purity are peculiar to

INDRA SAWHNEY v. U.0.1. 637

each caste ....... Even the choice of occupation of members A of castes was predetermined in many cases, and the mem- bers of particular caste were prohibited from· engaging themselve:; in other types of callings, profession or occupa- tions. Certain occupations were considered to be degrad- ing or impure. 11 B

82Copy pinpoint82. The above material makes it amply clear that a caste is nothing but a social class - a socially homogeneous class. It is also an occupational grouping, with this difference that its membership is hereditary. One is born into it. Its membership is involuntary. Even if one ceases to follow that occupation, still he remains and continues a member of that group. To C repeat, it is a socially and occupationally homogenous class. Endogamy is its main charateristic. Its social status and standing depends upon the nature of the occupation followed by it. Lowlier the occupation, lowlier the social standing of the class in the graded hierarchy. lit rural India, occupa- tion-caste nexus is true even today. A few members may have gone to cities or even abroad but when they return - they do, barring a few exceptions - they go into the same fold again. It doesn't matter if he has earned money. He may not follow that particular occupation. Still, the label remains. His identity is not changed. For the purposes of marriage, death and all other social functions, it is his social class - the caste - that is relevant._ It is a matter of common knowledge that an overwhelming majority of doctors, engineers and other highly qualified people who go abroad fdr higher studies or employment, return to India and marry a girl from their own caste. Even those who are settled abroad come to India in search of brides and bridegrooms for their sons and daughters from among their own caste or community. As .observed by Dr. Ambedkar, a caste.is an enclosed class and. it was mainly these classes the Constituent Assembly had in mind - though not exclusively - while enacting Article 16(4). Urbanisation has to some.extent broken this caste- occupation nexus but not wholly. If one.sees around himself, even in towns and cities, a barber by caste continues to do the same job - may be, in a shop (hair dressing saloon). A washermaD. ordinarily carries on the same job though he may have a laundry of his own. G May be some others too carry on the profession of barber or washerman but that does not detract from the fact that in the case of an ove~-whelming majority, the caste-occup3ti~n nexus subsists. In a rutal c9ntext, of course, a member of barber caste .carrying on the occupation of a washerman or vice versa would indeed be a rarity - it is simply not done; There, one is H

638 SUPREME COURT REPORTS [1992] SUPP. 2 S.C.R.

A supposed to follow his caste occupation, ordained for him by his birth. There may be exceptions here and there, but we are concerned with generality of the scene and not with exceptions or aberrations. Lowly occupation results not only in low social position but also in poverty; it generates poverty. 'Caste- occupation-poverty' cycle is thus an ever present reality. In rural India, it is strikingly apparent; in urban centers, there may B be some dilution. But since rural India and rural population is still the overwhelmingly predominant fact of life in india, the reality remains. All the decisions since Ba/aji speak of this 'caste-occupation-poverty' nexus. The language and emphasis may very but the theme remains the same. This is the stark reality notwithstanding all our protestations and abhorrence c and all attempts at weeding out this phenomenon. We are not saying it ought to be encouraged. It should not be. It must be eradicated. That is the ideal - the goal. But any programme towards betterment of these sections-classes of society and any programme designed to eraC:icate this evil must recognise this ground reality and attune its programme accord- D ingly. Merely burying our heads in the sand - Ostrich-like - wouldn't help. One cannot fight his enemy without recognizing him. The U.S.Supreme Court has said repeatedly, if race be the basis of discrimination - past and present - race must also form the basis of redressal programmes though in our constitutional scheme, it is not necessary to go that far. Without a doubt, an extensive restructuring of socio-economic system is the answer. E That is indeed the goal, as would be evident from the preamble and Part IV (Directive Principles). But we are concerned here with a limited aspect of equality emphasised in Article 16(4) - equality of opportunity in public employment and a special provision in favour of backward class of citizens to enable them to achieve it. F (b). Identification of "backward class of citizens".

83Copy pinpoint83. Now, we may turn to the identification of "backward class of citizensn. How do you go about it? Where do you begir.? Is the method to very from State to State, region to region and from rural to urban? What G do you do in the case of religions where caste system is not prevailing? What about other classes, groups and communities which do not wear the label of caste? Are the people living adjacent to cease-fire line (in Jammu and Kashmir) or hilly or inaccessible regions to be surveyed and identified as backward classes for the purpose of Article 16(4)? And so on and so forth are the many questions asked of us. We shall answer them. But our

INDRA SAWHNEY v. U.0.l. 639

· answers will necessarily deal with generalities of the situation and not with problems or issues of a peripheral nature which are peculiar to a particular State, district or region. Each and every situation cannot be visualised and answered. That must be left to the appropriate authorities appointed to identify. We can lay down only general guidelines.

At the outset, we may state that for the purpose of this discussion, we keep aside the Scheduled Tribes and Scheduled Castes (since they are admittedly included within the backward classes), except to remark that backward classes contemplated by Article 16( 4) do comprise some castes - for it cannot be denied that Scheduled Castes include quite a few castes. c Coming back to the question of identification, the fact remains that one has to begin somewhere - with some group, class or section. There is no set or recognised method. There is no law or other statutory instrument prescribing the methodology. The ultimate ide.a is to survey the entire populace. If so, one can well begin with castes, which represent explicit identifiable social classes/groupings, more particularly when Article 16( 4) seeks to ameliorate social backwardness. What is unconstitutional with it, more so when caste, occupation, poverty and social backwardness are so closely inter-twined in our society? [Individual survey is out of question, since Article 16(4) speaks of class protection and not individual protec- tion]. This does not mean that one can wind up the process of identification with the castes. Besides castes (whether found among Hindus or others) there may be other communities, groups, classes and denominations which may qualify as backward class of citizens. For example, in a particular State, Muslim community as a whole may be found socially backward. (As a matter of fact, they are so treated in the State of Karnataka as well as in the State of Kerala by their respective State Governments). Similarly, certain sections and denominations among Christians in Kerala who were included among backward communities notified in the former princely State of Travancore as far back as in 1935 may also be snrveyed and so on and so forth. Any authority entrusted with the task of identifying backward classes may well start with the castes. It can take caste 'A', apply the criteria of backwardness evolved by it to that caste and determine whether it qualifies as a backward class or not. If it does qualify, what emerges is a backward class, for the purposes of clause (4) of Article 16. The concept of 'caste' in this behalf is not confined to castes among Hindus. It extends to castes, wherever they obtain as a fact, irrespective of religious sanction H

640 SUPREME COURT REPORTS [1992] SUPP. 2 S.C.R.

A for such practice. Having exhausted the castes or simultaneously with it, the authority may take up for consideration other occupational groups, communities and classes. For example, it may take up the Muslim com- munity (After excluding those sections, castes and groups, if any, who have already been considered) and find out whether it can be characterised as a backward class in that State or region, as the case may be. The approach B may differ from State to Stale since the conditions in each Slate may differ. Nay, even within a State, conditions may differ from region to region. Similarly, Christians may also be considered. If in a given place, like Kerala, there are several denominations, sections or divisions, each of these groups may separately be considered. In this manner, all the classes among the c populace will be covered and that is the central idea. The effort should be to consider all the available groups, sections and classes of society in whichever order one proceeds. Since caste represents an eAisting, identifi- able, social group spread over an overwhelming majority of the country's population, we say one may well begin with castes, if one so chooses, and then go to other groups, sections and classes. We may say, at this stage, that we broadly commend the approach and methodology adopted by Justice 0.Chinnappa Reddy Commission in this respect.

We do not mean to suggest - we may reiterate - that the procedure indicated hereinabove is the only procedure or method/approach to be adopted. Indeed, there is no such thing as a standard or model proce- dure/approach.- It is for the authority (appointed to identify) to adopt such approach and procedure as it thinks appropriate, and so long as the approach adopted by it is fair and adequate, the court has no say in the matter. The only object of the discussion in the preceding para is to emphasise that if a Commission/Authority begins its process of identifica- F tion with castes (among Hindus) and occupational groupings among others, it cannot by that reason alone be said to be constitutionally or legally bad. We must also say that there is iJ.o rule of law that a test to be applied for identifying backward classes should be only one and/or uniform. In a vast conntry like India, it is simply not practicable. If the real object is to discover and locate backwardness, and if such backwardness is found in a caste, it can be treated as backward; if it is found in any other group, section or class, they too can be treated as backward.

83A. The only basis for saying that caste should be excluded from consideration altogether while identifying the Backward Class of Citizens

INDRA SAWHNEY v. U.0.1. 641

for the purpose of Article 16( 4) is clause (2) of Article 16. Thi' argument, however, overlooks and ignores the true purport of clause (2). lt prohibits discrimination on any or all of the grounds mentioned therein. The sig- nificance of the word "any"· cannot be minimised.** Reservation is· n~t being made under clause (4) in favour of a 'caste' but a 'backward class'. Once a caste satisfies the criteria of backwardness, it becomes a backward class for the purposes of Article 16(4). Even that is not enough. It must be further found that that backward class is not adequately represented in the services of the State. In such a situation, the bar of clause (2) of Article 16 has no application whatsoever. Similarly, the argument based upon secular nature of the Constitution is too vague to be accepted. It has been repeatedly held by the U.S. Supreme Court in School desegregation cases c that if race be the basis of discrimination, race can equally form the basis of redressal. In any event, in the present context, it is not necessary to go to that extent. It is sufficient to say that the classification is not on the basis of the caste but on the ground that that caste is found to be a backward class not adequately represented in the services of the State. Born Heathen, D by baptism, it becomes a Christian - to use a similie. Baptism here means passing the test of backwardness.

84Copy pinpoint84. Another contention urged is that only that group or section of people, who are suffering the lingering effects of past discrimination, can alone be designated as a backward class and not others. This argument, inspired by certain American decisions, cannot be accepted for more than one reason. Firstly, when the caste discrimination is still prevalent, more particularly in rural India (which comprises the bulk of the total popula- tion), the theory of lingering effects has no relevance. Where the dis- crimination has ended, does that aspect become relevant and not when the discrimination itself is continuing. Secondly, as we have noticed hereinabove, the said theory has practically been given up by the U.S. Supreme Court in Metro Broadcasting. In this case, it is held sufficient for introducing and implementing a race-conscious programme that such .. In Air India v. Nargesh Mina, this court held: "What Articles 15(1) and 16(2) prohibit is that discrimination should not be made only and only on the ground or sex. These G

Articles of the Constitution do not prohibit the State from making discrimination on the ground of sex coupled with other considerations. On this point, the matter is no longer res intergra but is covered by several authorities of this court." Reference is then made to Ywuf Abdul Aziz v. State of Bombay, (1954} S.C.R. 930 and Miss C.B. Mu:hamma v. Union of India, (1979) 4 S.C.C. 260. H

642 SUPREME COURT REPORTS [1992] SUPP. 2 S.C.R.

A programme serves important State objectives. In other words, according to this test, it is no longer necessary to prove that such programme is designed to compensate victims of past societal or governmental discrimination. Thirdly, the basic premise of the theory of lingering effects is not accepted by all the learned Jndges of U.S. Supreme Court. If one sees the opinion of Douglas, J. in Defunis and of Marshall, J. in Bakke and Fullilove. It would become evident. They also say that discriminatory practices against blacks and other minorities have not come to an end but are still persisting. In this country too, none can deny - in the face of the material collected by the various Commissions including Manda! Commission - that dis- crimination persists even today in India The representation of the socially backward classes in the Government apparatus is quite inadequate and that conversely the upper classes have a disproportionately large representation therein. This is the lingering effect, if one wants to see it.

Whether the backwardness in Anicle 16(4) should be both social and D educational7

85Copy pinpoint85. The other aspect to be considered is whether the backwardness contemplated in Article 16(4) is social backwardness or educational back- wardness or whether it is both social and educational backwardness. Since the decision in Balaji, it has been assumed that the backward class of citizens contemplated by Article 16(4) is the same as the socially and educationally backward classes, Scheduled Castes and Scheduled Tribes mentioned in Article 15(4). Though Article 15(4) came into existence later in 1951 and Article 16(4) does not contain the qualifying words 'socially and educationally' preceding tl:te words "backward class of citizens' the same meaning came to be attached to them. Indeed, it was stated in Janaki Prasad Parimoo (Palekar,J. speaking for the Constitution Bench) that:

'Article 15(4) speaks about socially and educationally -backward classes of citizens." However, it is now settled that the expression "backward class of citizens" in Article G 16(4) means the same thing as the expression 'any socially and educationally backward class of citizens" in Article 15(4). In order to qualify for being called a 'backward class citizens' he must be a member of a socially and education- ally backward class. It is social and educational backward- H ness of a class which is material for the purposes of both

INDRA SAWHNEY v. U.0.1. 643

Article 15(4) and 16(4)." A It is true that no decision earlier to it specifically said so, yet such an impression gained currency and it is that impression which finds expression in the above observation. In our respectful opinion, however, the said as" sumption has no basis. Clause (4) of Article 16 does not contain the qualifying words 'socially and educationally' as does clause (4) of Article 15. It may be remembered that Article 340 (which has remained unamended) does employ the expression 'socially and educationally backward classes' and yet that expression does not find place in Article 16(4). The rea5on is obvious: "backward class of citizens" in Article 16{4) takes in Scheduled Tribes, Scheduled Castes and all other backward classes of citizens including the socially and educationally backward classes. Thus, certain classes which may not qualify for Article 15(4) may qualify for Article 16(4). They may not qualify for Article 15(4) but they may qualify as backward class of citizens for the purposes of Article 16(4). It is equally relevant to notice that Article 340 does not expressly refer to services or to reservations in services under the D State, though it may be that the Commission appointed thereunder may recommend reservation in appointments/posts in the services of the State as one of the steps for removing the difficulties under which SEBCs are la,bour- ing and for improving their conditions. Thus, S.E.B.Cs, referred to in Article 340 is only one of the categories for whom Article 16(4) was enacted; Article 16(4) applies to a much larger class than the one contemplaied by Article 340. E It would, thus, be not correct to say that barkward class of citizens' in Article 16(4) are the same as the socially and educationally backward classes in Article 15(4). Saying so would mean and imply reading a limitation into a beneficial provision like Article 16(4). Moreover, when speaking of reserva- tion in appointments/posts in the State services - which may mean, at any level whatsoever - insisting upon educational backwardness may not be quite appropriate.

Further, if one keeps in mind the context in which Article 16(4) was enacted it would be clear that the accent was upon social backwardness. It goes without saying that in Indian context, social backwardness leads to educational backwardness and both of them together lead to proverty - which in turn breeds and perpetuates the social and educational backward- ness. They feed upon each other constituting a vicious circle. It is a well known fact that till independence the administrative apparatus was manned almost exclusively by members of the 'upper' castes. The Shudras, the H

644 SUPREME COURT REPORTS [1992] SUPP. 2 S.C.R.

A ·Scheduled Castes and the Scheduled Tribes and other similar backward social" groups among Muslims and Christians bad practically no entry into the administrative apparatus. It was this imbalance which was sought to be redressed by providing for reservations in favour of such backward classes. In this sense Dr. Rajiv Dbawan may be right when he says that the object of Article 16(4) was "empowerment" of the backward classes. The idea was to enable them to share the state power. We are, accordingly, of the opinion that the backwardness contemplated by Article 16(4) is mainly social backwardness. It ,would not be correct to say that the backwardness under Article 16(4) should oe both social and educational. The Scheduled Tribes and the Scheduled Castes are without a doubt backward for the purposes of the clause; no one bas suggested that they should satisfy the test of social and educational backwardness. It is necessary to state at this stage that the Manda! Commission appointed under Article 340 was con- cerned only with the socially and .educationally backward classes con- templated by the said Article. Even so, it is evident that social backwardness bas been given precedence over others by the Manda! Commission - 12 out of 22 total points. Social backward:aess - it may be reiterated - leads to educational and economic backwardness. No objection can be, nor is taken, to the validity and relevancy of the criteria adopted by the Manda! Commission. For a proper a appreciation of the criteria adopted by the Manda! Commission and the difficulties in the way of evolving the criteria of backwardness, one must read closely Chapters lII and XI of Volume I along with Appendixes 12 and 21 in Volume II. Appendix XII is the Report of the Research Planning Team of the Sociologists while Appendix 21 is the 'Final List of Tables' adopted in the course of socio-educational survey. In particular, one may read paras 11.18 F to 11.22 in Chapter XI, which are quoted hereunder for ready reference:

"11.18. Technical Committee constituted a Sub-Committee of Experts (Appendix-20, Volume II) to help the Commis- sion prepare 'Indicators of Backwardness' for analysing data contained in computerised tables. After a series of meetings and a lot of testing of proposed indicators against the tabulated data, the number of tables actually required for the Commission's work was reduced to 31 (Appendix- 21 Volume II). The formulation and refinement of in- dicators involved testing and validation checks at every stage.

INDRA SAWHNEY v. U.0.l. 645

11.19. In this connection, it may be useful to point out that in social sciences no mathematical formulae or precise bench-marks are available for determining various social traits. A survey of the above type has to read warily on unfamiliar ground and evolve its own norms and bench- marks. This exercise was full of hidden pitfalls and two simple examples are given below to illustrate this point. B

11.20. In Ba/aji's case the Supreme Court held that if a particular community is to be treated as educationally backward, the divergence between its educational level and that of the State average should not be marginal but sub- c stantial. The Court considered 50% divergence to be satis- factory. Now, 80% of the population of Bihar (1971 Census) is illiterate. To beat this percentage figure by a margin of 50% will mean that 120% members of a caste/class should be illiterates. In fact it will be seen that in this case even 25% divergence will stretch us to the maximum saturation point of 100%.

11.21. In the Indian situation where vast majority of the people are illiterate, poor or backward, one has to be very careful in setting deviations from the nornts as, in our conditions, norms themselves are very low. For example, Per Capita Consumer Expenditure for 1977-78 at .current prices was Rs.991 per annum. For the same period, the poverty line for urban areas was at Rs.900 per annum and for rural areas at Rs.780. It will be seen that this poverty line is quite close to the Per C~pita Consumer Expenditure of an average Indian. Now following the dictum of Ba/aji case, if 50% deviation from this average Per Capital Con- sumer Expenditure was to be accepted to identify 'economically backward' classes, their income level will have to be 50% below the Per Capital Consumer Expen- diture i.e. less than Rs.495.5 per year. This figure is so much below the poverty line both in urban and rural areas that most of the people may die of starvation before they qualify for such a distinction. H

646 SUPREME COURT REPORTS [1992] SUPP. 2 S.C.R.

A 11.22. In view of the above, 'Indicators for Backwardness' were tested against various cut-off points. For doing so, about a dozen castes well-known for their social and educa- tional backwardness were selected from amongst the castes covered by our survey in a particular State. These were treated as 'Control' and validation checks were carried out B by testing them against 'Indicators' at various cut-off points. For instance, one of the 'Indicators' for social backward- ness is the rate of student dropouts in the age group 5-15 years as compared to the State average. As a result of the above tests, it was seen that in educationally backward c castes this rate is at least 25 per cent above the State average. Further, it was also noticed that this deviation of 25% from the State average in the case of most of the 'Indicators' gave satisfactory results. In view of this, wherever an 'Indicator' was based on deviation from the State average, it was fixed at 25%, because a deviation of D 50% was seen to give wholly unsatisfactory results and, at times, to create anomalous situations."

It is after these paragraphs that the Report sets out the indicators (criteria) evolved by it, set out in Paras 11.23 and 11.24 of the Report. E The S.E.B.Cs. referred to by the impugned Memorandums are un- doubtedly 'backward class of citizens' within the meaning of Article 16(4).

(d) 'Means' test and 'creamy layer':

86Copy pinpoint86. 'Means test' in this discussion signifies imposition of an income limit, for the purpose of excluding persons (from the backward class) whose income is above the said limit. This submission is very often referred to as "the creamy layer" argument. Petitioners submit that some members of the desig- nated backward classes are highly advanced socially as well as economically and educationally. It is submitted that they constitute the forward section of that particular backward class - as forward as any other forward class mem- ber - and that they are lapping up all the benefits of reservations meant for that class, without allowing the benefits to reach the truly backward members of that class. These persons are by no means backward and with them a class cannot be treated as backward. It is pointed out that since Jayasree, almost every decision has accepted the validity of this submission.

INDRA SAWHNEY v. U.0.1. 647

On the other hand, the learned counsel for the State of Bihar, Tamil A Nadu, Kerala and other counsel for respondents strongly oppose any such distinction. It is submitted that once a class is identified as a backward class after applying the relevant criteria including the economic one, it is not permissible to apply the economic criteria once again and sub-divide a backward class into two sub-categories. Counsel for the State of Tamil Nadu submitted further that at one stage (in July 1979) the State of Tamil B Nadu did indeed prescribe such an income limit but had to delete it in view of the practical difficulties encountered and also in view of the repre- sentation received. In this behalf, the learned counsel invited our attention to Chapter 7-H (pages 60 to 62) of the Ambashankar Commission (Tamil Nadu Second Backward Classes Commission) Report. According to the c respondents the argument of 'creamy layer' is but a mere ruse, a trick, to deprive the backward classes of the benefit of reservations. It is submitted that no member of backward class has come forward with this plea and that it ill becomes the members of forward classes to raise this point. Strong reliance is placed upon the observations of Chinnappa Reddy,J. in Vasant D Kumar, to the following effect:

11 ••• 0ne must, however, enter a caveat to the criticism that the benefits of reservation are often snatched away by the top creamy layer of backward class or caste. That a few of the seats and posts reserved for backward classes are E. snatched away by the more fortunate among them is not to say that reservation is not necessary. This is bound to happen in a competitive society such as ours. Are not the unreserved seats and posts snatched away, in the same way, by the top creamy layers amongst them on the same prin- F ciple of merit on which the non reserved seats are taken away by the top layers of society. How can it be bad if reserved seats and posts are snatched away by the creamy layer of backward classes, if such snatching away of un~ reserved posts by the top creamy layer of society itself is not bad? 11 G

In our opinion, it is not a question of permissibility or desirability of such test but one of proper and more appropriate identification of a class - a backward class. The very concept of a class denotes a number of persons having certain common traits which distinguish them from the others. In a H

648 SUPREME COURT REPORTS jl992} SUPP. 2 S.C.R.