ASHOK KUMAR YADAV AND ORS. ETC. ETC. v. STATE OF HARYANA AND ORS. ETC. ETC.

ASHOK KUMAR v. H~RYANA (Bhagwati, J,) 693

figures relied upon by tbe Division Bench may create a suspicion in one's mind that some element of arbitrariness might have entered A the assessment in the viva voce examination. But suspicion cannot take the place of proof and we cannot strike down the selections made on the ground that the evaluation of the merits of the candi- dates in the viva voce examination might be arbitrary. It is necessary to point out that the Court cannot sit in judgment over 8 the marks awarded by interviewing bodies unless it is pro,.ed or obvious that the marking is plainly and indubitably arbitrary or affected by oblique motives. It is only if the assessment is patently arbitrary or the risk of arbitrariness is so high that a reasonable person would regard arbitrariness as inevitable, that the assessment of marks at the viva voce test may be regarded as suffering from the C vice of arbitrariness. Moreover, apart from the only three candi- dates, namely. Trilok Nath Sharma, Sbakuntala Rani and Balbir Singh one of whom belonged to the general category and was related to Sbri Raghubar Dayal Gaur and the other two were candidates for the seats reserv(d for Schedukd Castes and Mre related to Shri R.C. J) Marya, there was no other candidate in whom the Chairman or any member of the Haryana Public Service Commission was interested, so that there could be any motive for manipulation of the marks at the viva voce examination. There were of course general allegations of casteism made against the Chairman and the members of the Haryana Public Service Commission, but these allegations were not substantiated by producing any reliable material before the Cou1t. The Chairman and members of the Haryana Public Service Commission in fact belonged to different castes and it was not as if any particular caste was predominant amongst the Chairman and members of the Haryana Public Se1vice Commission so as even to f' remotely justify an inference that the marks might have been mani· pulated to favour the candidates of that caste. We do not think that the Division Bench was right in striking down the selections made by the Haryana Public Service Commission on the ground that they were vitiated by arbitrariness or by reasonable likelihood of biu. G That takes us to the next ground of challenge which found acceptance with the Division Bench. This ground of challenge was streneously urged on behalf of the petitioners and it was sought to be supported by reference to the decision of this Court in Ajoy Hasia H v. Khalid Mujjubudin.( 11 The contention of the petitioners under this \IJ tJ~ilJ 2 S.C.R. 7 9

694 SUPREME COURT REPORTS (!985J SUPPL. s.c.R.

ground of challenge was tJ:lat in comparison to the marks allocated A to the written examination, the proportion of the marks allorated to the viva voce test was excessively high ar,d that introduced an irredeemable element of arbitrariness in the selection process , 0 as to offend Articles 14 and 16 of the Constitution. It is necwary in order to appreciate this contention and to adjudicate upon its validity 8 to consider the relative weight attached by the relevant rules to the written examination and viva voce test. We have already referred to the Punjab Civil Service (Executive Branch) Rules 1930 as applicable in the Stale of Haryana. Rule 9 of these rules prescribes that a competitve examination shall be held in accordance with the Regulations set out in Appendix 1 for the purpose of selection by c competition of candidates to the Haryana Civil Service (Executive Branch) and other allied services and under Regulations I and 5 every ex-service officer has to appear in a written examination in 5 compulsory subject' carrying in the aggregate 400 marks and a viva voce test carrying 200 marks and likewise, every candidate belong·ng to the general category has to appear in a written examination in 8 subjects carrying in the aggregate 700 marks and for him also there is a viva voce test carrying 200 marks. The argument of the petitioners was that in ca'e of ex-service officers the marks allocated for the viva voce test were 200 as against 400 allocated for the written examination so that the marks allocated for the viva voce test came to 33.3% of the total number of marks taken into account for the purpose of making selection. So also in the case of candidates belonging to the general category, the marks allocated for the viva voce test were 200 as against 700 allocated for the written exami- nation with the result that the marks allocated for the viva voce test came to 22.2% of the total number of marks for the compet1t1ve e>amination. This percentage of 33.3% in the case of ex-service officers and 22.2% in the case of other candidates was, according to the Division Bench, unduly high and rendered the selection of the candidates arbitrary. The correctness of this view has been challenged before us on behalf of the respondents. G This Court speaking through Chinnappa Reddy, J. pointed out in Liladhar v. Stare of Rajasthan( 1) that the object of any process of selection for entry into public service is to secure the best and the most suitable person for the job, avoiding patronage and favouritism. H Selection based on merit, tested impartially and objectively, is the

(11 11982) l S.C.R. 329.

ASHOK KUMAR v. HARYANA (Bhagwati, J.) 695

~ssential foundation of any useful and efficient public service. So -0pen competitive e.amination has come to be accepted almost univer- A sally as the gateway to public services. But the question is how should the competitive examination be devi~ed1 The competitive examination may be based exclusively on written examination or it may be based exclusively on oral interview or it may be a mixture of both. It is entirely for the Government to decide what kind of competitive examination would be appropriate in a given case. To quote the words of Chinnappa Reddy, J. "In the very nature of things it would not be within the province or even the competence of the court and the Court would not venture into such exclusive thickets to discover ways out, when the matters are more appro- priately left" to the wirdom of the experts. It is not for the Court to lay down whether interview test should be held at all or how many marks should be allowed for the interview test. Of course the marks must be minimal so as to avoid charges of arbitrariness, but not necessarily always. There may posts and appointments where the only proper method of selection may be by a viva voc-e test. Even in the case of admission to higher degree courses, it may sometimes be necessary to allow a fairly high percentage of marks for the viva voce test. That is why rigid rules cannot be laid down in these matters and not · y courts. The expert bodies are generally the best judges. The Government aided by experts in the field may appropriately decide to have a written examination followed by a viva voce test. E

It is now admitted on all bands that while a written examination as,esses the candidate's kvo" ltdge and intellectual ability, d viva voce test seeks to assess a candidate's overall intellectual and personal qualities. While a written examination bas certain distinct advant· F ages over the viva voce test, there are yet no written tests which can evaluate a candidate's initialive, alertness, resourcefulnes..-, depen.. dableness, cooperativeness, capacit) for clear and logical presentation, effectiveness in discussion. eff<ctiveness in rr.eeting and dealing \lith others, adaptability, judgment, ability to make decision, ability to lead, intellectual and moral integrity. Some of these qualities can be G evaluated, perhaps with some degree of error, by a viva voce test, much depending on the constitution of the interview Board.

Glenn Stahl has pointed out in his book on Public Personnel Administration that the viva voce test does suffer from ceitain B disadvantages such as the difficulty of developing a valid and reliable oral test, the difficulty of securing a reviewable record of an oral test

696 SUPRFME COURT REPORTS '1985] SUPPL. s.c.R.

and public suspicion of the oral test as a channel for the exertion of political influence and. as pointed out by this Court in Ajay Hasia's case (supra), also of other corrupt, nepotistic or extraneous conside- rations, but despite these acknowledged disadvantages, the viva voce test has been used increasingly in the public personnel testing and bas become an important instrument whenever tests of personnel attributes are considered essential. Glenn Stahl proceeds to add that "no satisfactory written tests have yet been devised for measuring such personnel characteristics as initiative, ingenuinity and ability to elicit cooperation, many of which are of prime importance. When properly employed, the oral test today deserves a place in the battery used by the technical examiner." There can therefore be no doubt c that the viva voce test performs a very useful function in assessing personnel characteristics and traits and in fact, tests the man him- •• self and is therefore regarded as an important tool along with the written examination. Now if both written examination and viva voce test are accepted as essential features of proper selection in a given case, the question may arise as to the weight to be attached respectively to them. "In-the case of admission to a college for instance". as observed by Chinnappa Reddy, J. in Liladhar's case, "where the.candidate's personality is yet to develop and it is too early to identify the personal qualities for which greater importance may have to be attached in later lief, greater weight bas perforce to be given to performance in the written exam;nation" and the impor- tance to be attached to the viva voce test in such a case would therefore necessarily he minimal. lt was for this reason that in Ajav Hoisa's case this Court took the view that the allocation of as high a percentage of marks as 33.3°'. to the viva voce test was "beyond all reasonable proportion and rendered the selection of the candidates arbitrary". But, as pointed out by Chinnappa Reddy, J., "in the case of services to which recruitment has necessarily to be made from persons of mature personality, interview test may be the only way subject to basic and essential academic and profeS&ional G requirements being satisfied". There may also be services "to which recruitment is made from younger candidates whose personalities are on the thresh hold of development and who show sings of great pormise" and in case of such services where sound selection must combine academic ability with personality promise. some weight has to be given to the viva voce test. There cannot be any bard and fast II rule regarding the precise weight to be given to the viva voce test as against the written examination. It must vary from service to ;ervice

AHOSK KUMAR v. HARYANA (Bhagwati. J.) 697

according to the requirement of the service, the minimum qu!ilifica- tion prescribed, the age group from which the selection is to be A made, the body to which the task of holding the viva voce test is proposed to be entrusted and a host of other factors. It is essentially a matter for determination by experts. The Court does not possess the necessary equipment and it would not be right for the Court to pronounce upon it, unless to use the words of Chinnappa Reddy, J. 8 in Li!adhar's case "exaggerated weight has been given with proven or obvious oblique motives."

We may now, in the background of this discussion, proceed to consider whether the allocation of as high a percentage of marks as c 33.3 per cent in case of ex-service officers and 22.2 per cent in case of other candidates, for the viva voce test renders the selection process arbitrary. So far as ex-service officers are concerned, there can be no doubt that the percentage of marks allocated for the viva voce test in their case is unduly high and it does suffer from the vice of arbitrariness. It has been pointed out by the Division Bench in a D fairly elaborate discussion that so far as the present selections in the category of ex-service officers are concerned, the spread of marks in the viva voce test was inordinately high compared to the spread of marks in the written examination. The minimum marks required to be obtained in the written examination for eligibility for the viva voce test are 180 and as against these minimum 180 marks, the highest marks obtained in the written examination in the category of ex-service officers were 270, the spread of marks in the written examination thus being only 90 marks which works out to a ratio of 22 .2 per cent. But when we turn to the marks obtained in the viva voce test, we find that in case of ex-service officers the lowest marks obtained were 20 while the highest marks secured were 171 and the spread of marks in the viva voce test was thus as wide as 151 in a total of 200 marks, which worked out to an inordinately high percentage of 76. The spread of marks in the viva voce test being enormously large compared to the spread of marks in the written examination, the viva voce test tended to become a deter- G mining factor in the selection process, because even if a candidate secured the highest marks in the written examination, he could be easily knocked out of the race by awarding him the lowest marks in the viva voce test and correspondingly, a candidate who obtained the lowest marks in the written examination could be raised to the top most position in the merit list by an inordinately high mar king in the viva voce test. It is therefore obvious that the allocation of

698 SUPREME COURT REPORTS (1985] SUPPL. S.C.R.

such a high percentage of marks as 33.3 per cent opens the door wide for arbitrariness, and in order to diminish, if not eliminate the risk of arbitrariness, this percentage need to be reduced. But while considering what percentage of marks may legitimately be allocated for the viva voce test without incuring the reproach of arbitrariness it must be remembered that ex-service officers would ordinarily be 8 middle aged persons of mature personality and it would be hard on them at that age to go through a long written examination involving 8 subjects and hence it would not be unfair to require them to go through a shorter written examination in only 5 sub1ects and submit to a viva voce test carrying a higher percentage of marks than that might be prescribed in case of younger candidates. The personalties c of these ex-service officers being fully mature and developed, it would not be difficult to arrive at a fair assessment of their merits on the basis of searching and incisive viva voce test and therefore in their case, the viva voce test may be accorded relatively greater weight. But in any event the marks allocate for the viva voce test cannot be as high as 33.3 per cent.

The position is no different when we examine the question in regard to the percentage of marks allocated for the viva voce test in case of persons belonging to the general category. The percent· age in the case of these candidates is Jess than that in the case of ex-service officers, but even so it is quite high at the figure of 22.2. Here also it has been pointed out by the Division Bench by giving facts and figures as to how in the case of present selections from the general category the spread of marks in the viva voce test was inordinately high compared to the spread of marks in the written examination so that a candicate receiving low marks in the written examination could be pulled up to a high position in the merit list by inordinately high marking in the viva voce test. The viva voce test in the general category, too, would consequently tend to become a determining factor in the process of selection, tilting the scales in favour of one candidate or the other according to the marks awarded to him in the viva voce test. This is amply horned out by the observations of the Kothari Committee in the Report made by it in regard to the selections to the Indian Adminis· trative Service and other allied services. The competitive examina· tion in the Indian Administrative Service and other allied services also consists of a written examination followed by a viva voce test. Earlier in 1948 the percentage of marks allocated for the viva vocc

ASHOK KUMAR v. HARYANA (Bhagwati, J.) 699

test was 22 and it was marginally brought down to 21.60 in 1951 and then again in 1964, it was scaled down to 17.11. The Kothari A Committee in its Report made in 1976 pleaded for further rednction of the percentage of marks allocated for the viva voce test and strongly recommended that the viva voce test should carry only 300 out of a total of 3l 00 marks. The Kothari Committee pointed out that even where the percentage of marks allocated for the viva vocc B test was 17.11, nearly I /4th of the candidates selected owed their success to the marks obtained by them at the viva voce test. This proportion was regarded by the Kothari Committee as "somewhat on the high side". It is significant to note that consequent upon the Kothari Committee Report, the percentage of marks allocated for the viva voce test in the competitive examination for the Indian C Administrative Service and other allied services was brought down still further to 12.2. The result is that since the last few years, even for selection of candidates in the Indian Administrative Service and other allied services where the personality of the candidate and his personnel characteristics and traits are extremely relevant for the D purpose of selection, the marks allocated for the viva voce test constitute only 12.2 per cent of the total marks. Now if it was found in the case of selections to the Indian Administrative Service and other allied services that the allocation of even 17.11 per cent marks for the viva voce test was on the higher side and it was responsible for nearly I/4th of the selected candidates securing a E place in the select list owing to the marks obtained by them at the viva voce test, the allocation of 22.2 per cent marks for the viva voce test would certainly be likely to create a wider scope for arbitrari- ness. When the Kothari Committee admitredly an Expert Com- mittee, constituted for the purpose of examining recruitment policy F and selection methods for the Indian Administrative Service and other allied services took the view that the allocation of 17.1 per cent m:irks for the viva voce test was on the higher side and required to be reduced, it would be legitimate to hold that in case of selections to the Haryana Civil Services (Executive Branch) and other allied services, which are services of similar nature in the G State, the allocation of 22.2 per cent marks for the viva voce test was unreasonable, We must therefore regard the allocation of 22.2 per cent of the total marks for the viva Voce test as infecting the selection process with the vice of arbitrariness. H But the question which then arises for consideration is as to what is the effect of allocation of such a high percentage of marks

700 SUPREME COURT REPORTS [1985j SUPPL. s.c.R.

for the viva voce test, both in case of ex-service officers and in case A of other candidates, on the selections made by the Haryana Public Service Commission. Though we have taken the view that the percentage of marks allocated for the viva voce test in both these cases is excessive, we do not think we would be justified in the exercise of our discretion in setting aside the selections made by the 8 Haryana Public Service Commission after the lapse of almost two years. The candidates selected by the Haryana Public Service Commission have already been appointed to various posts and have been working on these posts since the last about two years. More· over the Punjab Civil Service (Executive Branch) Rules 1930 under which 33.3 per cent marks in case of ex-service officers and 22.2 per c cent marks in case of other candidates, have been allocated for the viva voce test have been in force for almost 50 years and everyone has acted on the basis rules. If selections made in accordance with the prescription contained in these rules are now to be set aside, it will upset a large number of appointments already made on the D basis of such selections and the integrity and efficiency of the entire administrative machinery would be seriously jeopardised. We do not therefore propose to set aside the selections made by the Haryana Public Service Commission though they have been made on the basis of an unduly high percentage of marks allocated for the viva voce test. E Now if the allocation of such a high percentage of marks as 33.3 in case of ex-service officers and 22.2 in case of other candidates, for the viva voce test is excessive, as held by us, what sbuld be the proper percentage of marks to be allocated for the viva voce test in F both these cases. So far as candidates in the general category are concerned we think that it would be prudent and safo to follow the percentage adopted by the Union Public Service Commission in case of selections to the Indian Administrative Service 8nd other allied services. The percentage of marks allocated for the viva voce test by the Union Public Service Commission in case of selections to the G Indian Administrative Services and other allied service is 12.2. and that has been found to be fair and just, as striking a propor balance between the written examination and the viva voce test. We would therefore direct that hereafter in case of selections to be made to the the Haryana Civil Services (Executive Branch) and other allied services, where the competitive examination consists of a written examination followed by a viva voce test, the marks allocated for

MHOK KUMAR 1'. HAl!.YANA (Bhagwali, J.) 701

the viva voce test shall not exceed 12.2 per cent of the total marks taken into account for the purpose of selection. We would suggest that this percentage should also be adopted by the Public Service Commissions is other States, because it is desirable that there should be uniformity in the selection process throughout the country and the practice followed by the Union Public Service Commission should be taken as a guide for the State Public Service Commissions to adopt and follow. The percentage of marks allocated for the viva voce test case of ex-service officers may, for reasons we have already discussed, be somewhat higher than the percentage for the candidates belonging to the general category. We would therefore direct that in case of ex-service officers, having regard to the fact that they would ordinarily be middle aged persons with personalities c fully developed the percentage of marks allocated for the viva voce test may be 25. Whatever selections are made by the Haryana Public Servire Commission in the future shall be on the basis that the marks allocated for the viva voce test shall not exceed 12.2 per cent in case of candidates belonging to the general category and 25 per cent in case of ex-service officers. D

Before we part with this judgment we would like to point out that the Public Service Commission occupies a pivotal place of importance in the State and the integrity and efficiency of its ad- ministrative apparatus depends considerably on the quality of the selections made by the Public Service Commission. It is absolutely essential that the best and finest talent should be drawn in the administration and administrative services must be composed of men who are honest, upright and independent and who are not swayed by the political winds blowing in the country. The selection of candidates for the administrative services must therefore be made strictly on merits, keeping in view various factors which go to make up a .strong, efficient and people oriented administrator. This can be achieved only 1f the .Chairman and members of the Public Service Commission are eminent men possessing a high degree of calibre competence and integrity, who would inspire confidence in the pub]"~ G mind about the objectivity and impartiality of the selections to ~e made by them. We would therefore like to strongly impress upon every State Government to take care to see that its Public Service Com- mission i.s mann~d by co~petent, honest and independent persons of outstandmg ab1hty and high reputation who command the confidence of the people and who would not allow themselves to be deflected by

702 SUPREME COURT REPORTS [1985] SUPPL, S.C.R.

any extraneous considerations from discharging their duty of making selections strictly on merits. Whilst making these observations we would like to make it clear that we do no for a moment wish to suggest that the Chairman and members of the Haryana Public Service Commission in the present case were lacking in calibre, competence or integrity. B We would also like to poiut out that in some of the States, and the State of Haryana is one of them, the practice followed is to invite a retired Judge of the High Court as an expert when selections for recruitment to the Judicial Service of the State are being made - and the advice given by such retired High Court Judge who partici- c pates in the viva voce test as an expert is sometimes ignored by the Chairman and members of the Public Service Commission. This practice is in our opinion undesirable and does not commend itself to us. When selections for the Judicial Service of the State are being made, it is necessary to exercise the utmost care to see that compe- D tent and able persons possessing a higb degree ofrectitude and integrity are selected, because if we do not have good, competent and honest judges, the democratic polity of the State itself will be in serious peril. It is therefore essential that when selections to the Judicial Service are being made, a sitting Judge of the High Court to be nominated by the Chief Justice of the State should be invited to participate in the interview as an expert and since such sitting Judge comes as an expert who, by reason of the fact that be is a sitting High Court Judge, knows the quality and character of the candidates appearing for the interview, the advice given by him should ordinarily be accepted, unless there are strong and cogent reasons for not accepting such advice and such •trong and cogent reasons must be recorded in writing by the Chairman and members of the Public Service Commission. We are giving this direction to the Public Service Commission in every State because we are anxious that the finest talent should be recruited in the Judicial Service and that can be secured only by having a real expert whose advice constitutes a determinative factor in the selection process.

We accordingly allow the appeals, set aside the judgment of the Punjab and Haryana High Court and reject the challenge to the validity of the selections made by the Haryana Public Service H Commission to the Haryana Civil Services (Executive Branch) and other allied services. But ;n view of the fact that an unduly large number of candidates were called for interview and the marks

ASHOK KUMAR V. HARYANA (Bhagwati, J.) 703

allocated in the viva voce test were excessively h.igh, it is possible that some of the candidates who might have otherwise come in the select list were left out of it, perhaps unjustifiably. We would there· fore direct that all the candidates who secured a minimum of 45 per cent marks in the written examination but who could not find entry in the select list, should be given one more opportunity of appearing in the competitive examination which would now have to be held in accordance with the principles laid down in this Judgment and this opportunity should be given to them, even though they may have passed the maximum age prescribed by the rules for recruitment to the Haryana Civil Services (Executive Branch) and other allied services. We would direct that in the circumstances of the case the fair order of costs would be that each party should bear and pay his c own costs throughout.

M.L.A. Appeals allowed