Page 1

1

ITEM NO.1A               COURT NO.12               SECTION XVI

              S U P R E M E  C O U R T  O F  I N D I A                        RECORD OF PROCEEDINGS

CIVIL APPEAL NO. 2051 OF 2015 (Arising out of Petition(s) for Special Leave to Appeal (C)   No(s). 27582/2014) SUJASHA MUKHERJI                           Petitioner(s)                                 VERSUS THE HON'BLE HIGH COURT OF CALCUTTA,  THR. REGISTRAR & ORS. Respondent(s) Date :19/02/2015 This Appeal was called on for Judgment Today. For Petitioner(s)                      Mr. Partha Sil,Adv.                    For Respondent(s)                      Mr. G. S. Chatterjee,Adv.                

Hon'ble Mr. Justice Vikramajit Sen pronounced the  Reportable  Judgment  of  the  Bench  comprising  His  Lordship and Hon'ble Mr. Justice C. Nagappan.

Leave granted.

  The Appeal is allowed and the Impugned Judgment is  set aside.

 (NEELAM GULATI)      COURT MASTER

  ( SAROJ SAINI)           COURT MASTER  

 ( Signed Reportable Judgment is placed on the file)

Page 2

2

REPORTABLE

IN THE SUPREME COURT OF INDIA

CIVIL APPELLATE JURISDICTION

CIVIL APPEAL NO.   2051      OF 2015 [Arising out of S.L.P.(C) No.27582 of 2014]

SUJASHA MUKHERJI         ….APPELLANT

Versus

THE HON’BLE HIGH COURT OF  CALCUTTA THROUGH  REGISTRAR  & ORS                                                                   .....RESPONDENTS    

J U D G M E N T

VIKRAMAJIT SEN, J.

1 Leave granted.

Page 3

3

2 Succinctly stated the significant and singular facts of the case are  

that the Writ Petitioner/Appellant was placed in the second position of the  

Written  Test  for  recruitment  to  the  cadre  of  the  District  Judge  (Entry  

Level) through Direct Recruitment from the Bar-2012.   Had her marks  

not been moderated from 55 per cent to 37 per cent in Paper No.II her  

aggregate marks would have been 307 which is higher than the candidate  

at Serial No.1 by 6.5 marks; in other words, she was the topper in the  

Written  Test  comprising  5  papers.   After  moderation  was  carried  out,  

[which it  appears  was conducted only in respect  of  Paper No. II],  she  

stood disqualified from further consideration, i.e. appearing for the final  

stage  of  selection,  viz.,  the  Interview/viva  voce  for  the  reason  that  

obtainment of minimum marks of 40 per cent in each paper was the pre-

requisite for being called for the Interview.  These facts have struck us as  

extremely significant for the reason that a candidate who stood First in the  

Written Examination (in five papers) has not been found suitable for even  

being called for the final step in recruitment, i.e. the Interview.  It has been  

asserted by the Writ Petitioner/Appellant that she is a position-holder in  

the Calcutta University; she quite obviously also possesses extraordinarily  

high academic and scholastic merit.  It has been vehemently contended  

before us, as also before the learned Single Judge and the learned Division  

Bench of the Calcutta High Court, that the moderation exercise has been  

undertaken even though it had not been notified or clarified at any stage

Page 4

4

that the examination would be subject to this scrutiny.   Whilst the 2006  

Guidelines were placed before and were duly approved by the Full Court,  

it appears that the 2012 Guidelines had not been placed before the Full  

Court but were followed by the three Judge Committee.

3 Learned  Senior  Counsel  for  the  High  Court  of  Calcutta  has  

strenuously  submitted  that  moderation  has  been  carried  out  strictly  in  

conformity with the decision of this Court in Sanjay Singh v. U.P. Public   

Service  Commission,  Allahabad (2007)  3 SCC 720.   It  deserves  to  be  

immediately underscored that the Rules for that examination envisaged a  

moderation exercise whereas this feature is absent so far as the subject  

examination is concerned.    We must immediately express the view that  

this argument has no merit since Moderation is merely a method to ensure  

that the marking or valuation is free from even unintended discrimination  

or inequality.

4 Learned Single Judge was of the opinion that the ratio of  Sanjay  

Singh had not been comprehensively followed, in that neither was a Head  

Examiner appointed, nor was a meeting held for the purpose of discussing  

the question paper  and the possible/model  answer thereto.  The learned  

Single Judge, therefore, found in favour of the Writ Petitioner/Appellant.  

The learned Single Judge had also noted that of the three examiners the  

junior-most  judge have  been  appointed  as  the  moderator.   It  was  also

Page 5

5

emphasised by the learned Single Judge that instead of moderation, in fact  

a re-assessment of the answer book of the Writ Petitioner/Appellant of  

Paper  No.II  has  been  carried  out.   Noting  that  if  the  Appellant  had  

received  three  marks  more  in  Paper  II  even  after  moderation  (i.e.  a  

deduction of 15 marks instead of 18 marks) she would have qualified to  

participate in the viva voce/Interview, the learned Single Judge held that  

the Appellant was unjustifiably excluded from the zone of consideration  

and was, therefore,   entitled to the relief as claimed in the petition.  The  

direction that was issued was to award the Appellant 55 marks in Paper II  

(i.e. without any moderation whatsoever) and to recast her position in the  

merit list accordingly; and further that the Selection Board should take her  

interview  within  the  least  possible  time  and  if  the  Appellant  did  not  

qualify after the Interview the candidate who would be otherwise entitled  

as per her/his merit panel should be appointed.

5 In the Impugned Judgment the endeavour of the learned Division  

Bench was palpably to decide the matter within the confines of  Sanjay  

Singh.  However, they have pointedly clarified the Judgment is not to be  

read and interpreted like a statute, which is the ratio of judgment of this  

Court.  Quite palpably, this clarification was necessitated by the fact that  

the  High  Court  had  not  ordained  a  selection  procedure  which  was  

completely in sync with Sanjay Singh.  In contrast to the Single Judge the

Page 6

6

learned Division Bench has played down and discounted the fact that a  

Head Examiner had not been appointed and that the Committee had not  

thought it essential to hold a meeting to discuss the questionnaire as well  

as agree on the acceptable/suitable/model answers thereto.  It is also not  

controverted that the candidates had not been notified that any or all of the  

papers  may  be  subjected  to  moderation  which  also  is  a  distinguished  

feature to the examinations process in Sanjay Singh.

6 We note that the senior-most Judge of the three Judge/Examiners is  

the  author  of  the  Impugned  Judgment.   It  requires  to  be  immediately  

stated that this is alarmingly irregular and tantamounts to being a Judge in  

one’s own cause.  It was, therefore, imperative for the learned Judge to  

recuse himself from the adjudication; and this facet would ordinarily be  

sufficient  to  set  aside  the  Impugned  Judgment.   However,  keeping  in  

perspective  the  gravity  and urgency  of  the  matters  in  issue  before  us,  

rather than remand the dispute to the High Court for a fresh determination  

by a Division Bench comprising learned Judges who are not connected in  

any manner  to  the  subject  selection,  we  think it  proper  to  proceed  to  

decide the dispute on its merits.   

7 The ratio of  Sanjay Singh, which is the fulcrum of the discussion  

of every aspect of this case, is discernable from the following extract:

Page 7

7

“23. When a large number of candidates appear for an  examination,  it  is  necessary  to  have  uniformity  and  consistency  in  valuation  of  the  answer-scripts.  Where  the  number of candidates taking the examination are limited and  only  one  examiner  (preferably  the  paper-setter  himself)  evaluates the answer-scripts, it is to be assumed that there  will  be  uniformity  in  the  valuation.  But  where  a  large  number of  candidates take the examination,  it  will  not be  possible to get all the answer-scripts evaluated by the same  examiner. It, therefore, becomes necessary to distribute the  answer-scripts among several examiners for valuation with  the paper-setter (or other senior person) acting as the Head  Examiner.  When  more  than  one  examiners  evaluate  the  answer-scripts relating to a subject,  the subjectivity of the  respective examiner will creep into the marks awarded by  him to the answer-scripts allotted to him for valuation. Each  examiner will apply his own yardstick to assess the answer- scripts.  Inevitably  therefore,  even  when  experienced  examiners receive equal batches of answer-scripts, there is  difference in average marks and the range of marks awarded,  thereby  affecting  the  merit  of  individual  candidates.  This  apart,  there  is  “hawk-dove”  effect.  Some  examiners  are  liberal  in valuation and tend to award more marks.  Some  examiners are strict and tend to give less marks. Some may  be moderate and balanced in awarding marks. Even among  those who are liberal or those who are strict, there may be  variance in the degree of strictness or liberality. This means  that  if  the  same  answer-script  is  given  to  different  examiners,  there is all  likelihood of different marks being  assigned. If a very well-written answer-script goes to a strict  examiner  and  a  mediocre  answer-script  goes  to  a  liberal  examiner, the mediocre answer-script may be awarded more  marks than the excellent answer-script. In other words, there  is “reduced valuation” by a strict examiner and “enhanced  valuation”  by  a  liberal  examiner.  This  is  known  as  “examiner  variability”  or  “hawk-dove  effect”.  Therefore,  there is a need to evolve a procedure to ensure uniformity  inter  se  the  examiners  so  that  the  effect  of  “examiner  subjectivity”  or  “examiner  variability”  is  minimised.  The  procedure  adopted  to  reduce  examiner  subjectivity  or  variability is known as moderation. The classic method of  moderation is as follows:

Page 8

8

(i) The paper-setter of the subject normally acts as the  Head  Examiner  for  the  subject.  He  is  selected  from  amongst  senior  academicians/scholars/senior  civil  servants/judges. Where the case is of a large number of  candidates,  more  than  one  examiner  is  appointed  and  each of  them is  allotted around 300 answer-scripts  for  valuation.

(ii) To achieve uniformity in valuation,  where more  than one examiner  is  involved,  a  meeting of  the Head  Examiner with all  the examiners is held soon after the  examination. They discuss thoroughly the question paper,  the possible answers and the weightage to be given to  various  aspects  of  the  answers.  They  also  carry  out  a  sample  valuation  in  the  light  of  their  discussions.  The  sample valuation of scripts by each of them is reviewed  by the Head Examiner and variations in assigning marks  are further discussed. After such discussions, a consensus  is  arrived at  in regard to the norms of valuation to be  adopted.  On  that  basis,  the  examiners  are  required  to  complete  the  valuation  of  answer-scripts.  But  this  by  itself, does not bring about uniformity of assessment inter  se  the  examiners.  In  spite  of  the  norms agreed,  many  examiners tend to deviate from the expected or  agreed  norms, as their caution is overtaken by their propensity  for  strictness  or  liberality  or  erraticism or  carelessness  during the course of valuation. Therefore, certain further  corrective steps become necessary.

(iii)  After  the  valuation  is  completed  by  the  examiners,  the  Head  Examiner  conducts  a  random  sample survey of the corrected answer-scripts to verify  whether the norms evolved in the meetings of examiner  have  actually  been  followed  by  the  examiners.  The  process of random sampling usually consists of scrutiny  of some top level answer-scripts and some answer books  selected  at  random from the  batches  of  answer-scripts  valued by each examiner. The top level answer books of  each examiner are revalued by the Head Examiner who  carries out such corrections or alterations in the award of  marks as he, in his judgment, considers best, to achieve  uniformity.  (For  this  purpose,  if  necessary  certain  statistics like distribution of candidates in various marks  ranges, the average percentage of marks, the highest and

Page 9

9

lowest  award  of  marks,  etc.  may  also  be  prepared  in  respect of the valuation of each examiner.)  

(iv)  After  ascertaining  or  assessing  the  standards  adopted  by  each  examiner,  the  Head  Examiner  may  confirm the award of  marks without any change if  the  examiner  has  followed  the  agreed  norms,  or  suggests  upward  or  downward  moderation,  the  quantum  of  moderation varying according to the degree of liberality  or strictness in marking. In regard to the top level answer  books  revalued  by  the  Head  Examiner,  his  award  of  marks is accepted as final. As regards the other answer  books below the top level, to achieve maximum measure  of  uniformity  inter  se  the  examiners,  the  awards  are  moderated  as  per  the  recommendations  made  by  the  Head Examiner.

(v) If in the opinion of the Head Examiner there has  been  erratic  or  careless  marking  by  any  examiner,  for  which it is not feasible to have any standard moderation,  the answer-scripts valued by such examiner are revalued  either by the Head Examiner or any other examiner who  is found to have followed the agreed norms.

(vi) Where the number of candidates is very large and  the examiners are numerous, it may be difficult for one  Head Examiner to assess the work of all the examiners.  In  such  a  situation,  one  more  level  of  examiners  is  introduced. For every ten or twenty examiners, there will  be a Head Examiner who checks the random samples as  above.  The  work  of  the  Head  Examiners,  in  turn,  is  checked by a Chief Examiner to ensure proper results.

The  above  procedure  of  “moderation”  would  bring  in  considerable uniformity and consistency. It should be noted  that  absolute  uniformity  or  consistency  in  valuation  is  impossible to achieve where there are several examiners and  the effort is only to achieve maximum uniformity”.

8 It  appears  to  us  to  be  uncontrovertibly  comprehensible  that  the  

cornerstone, nay keystone, of the method of moderation enunciated by this  

Court in Sanjay Singh postulates the existence of a Head Examiner who

Page 10

10

is usually the paper-setter also.  In this case the Respondents have neither  

asseverated nor  established that  a  Head Examiner had been appointed.  

Learned Senior Counsel had adumbrated that a multi person Committee of  

the High Court was entrusted with this fundamental duty and that in all  

probabilities this Committee selected the questionnaire from the multitude  

of questions suggested for the subject.   The fundamental predication of  

the paragraph 23(i) and (ii) of  Sanjay Singh, therefore, does not exist.  

Even if that were to be overlooked, no meeting was convened in which the  

examiners were present so as to discuss the substance of the questions and  

reach  a  consensus  as  regards  suitable/model  answers  thereto.  Quite  

evidently, the keystone on which the structure of Sanjay Singh had been  

painstakingly  constructed,  has  been  removed  with  the  result  that  the  

edifice has crumbled down.   It is not logical for the basic features to be  

ignored and thereafter to follow other elements for that will become an  

incorrect  extrapolation.    Furthermore,  we find no justification  for  the  

junior-most  Judge/examiner  to  have  been given the  formidable  task of  

moderation.  In the course of argument, we had requested learned Senior  

Counsel for the High Court to provide us with the curriculum vitae of the  

learned Judges in order to appreciate this decision; it could have been that  

he possessed an academic background or previous experience with regard  

to the conduct of Examination which made him the most suitable amongst  

the three Examiners to perform the task of moderation, but we could not

Page 11

11

find any additional criteria to support his candidature as the Moderator.  

Furthermore, we find that the marks awarded by the 1st Examiner have  

been left unchanged, except of cosmetic alterations.   In fact, there mostly  

appears to be a variation of nil or 0.5 marks after the moderation whereas  

large  scale  changes  have  been  effected  so  far  as  the  2nd Examiner  is  

concerned.   Assessment of answer books by an examiner is intrinsically a  

subjective exercise making it  a  rarity for two of them award the same  

percentage or marks out of 100.    It is, therefore, surprising to note that  

the Moderator’s subjectivity is almost identical to that of the 1st Examiner,  

but drastically different to the 2nd Examiner.   This has persuaded us to  

conclude that what has transpired in actuality is a fresh assessment and not  

a moderation of marks already awarded by an examiner.  This is not the  

purpose or objective behind moderation.  As has been clearly spelt out in  

Sanjay  Singh,  where  there  are  numerous  examiners  it  is  but  to  be  

expected that one may be more liberal when compared to another, who  

may even be strict, giving birth to the ‘hawk-dove’ effect, which has so  

perspicuously and graphically been explained in  Sanjay Singh.     The  

avowed  purpose  behind  moderation  is  to  “to  achieve  uniformity”,  to  

eradicate as far as possible the ‘hawk-dove’ effect.  If mistaking in the  

valuation of Answer-books are found to be rampant in the opinion of the  

Head Examiner, a fresh evaluation would have to be undertaken,  since  

moderation by definition cannot remove widespread mistakes.  It appears

Page 12

12

to us that sub-para (iii) of Sanjay Singh has been misconstrued and hence  

misapplied in the Impugned Judgment whilst it has been correctly applied  

by the learned Single Judge.    

9 Revaluation as envisaged in the paragraph 23 of Sanjay Singh has  

to be undertaken by the Head Examiner/Paper Setter who, as has already  

been noted, is non-existent in the present case.   The effort would be to  

eradicate the ‘hawk-dove’ syndrome, and this is achieved by computing  

the ‘mean’ and, thereafter, to add or deduct, across the board, in all the  

Answer–sheets.  It cannot be disputed that this is not what has transpired  

in the present case since quite apparently moderation has been carried out  

in respect of the assessment/marking of the 2nd Examiner and that too in  

Paper No. II.   So far as most of the candidates whose answer scripts had  

been reassessed afresh, the reduction averages 10 marks which, therefore,  

constitutes the mean.  Therefore, the deduction of as many as 18 marks so  

far  as  the  Appellant  is  concerned  is  not  logical  or  justified  as  a  

consequence of moderation.   We also think that a moderator should give a  

long  and  serious  thought  to  the  correctness  of  his  assessment  on  the  

realization he finds that the top-most candidate stands disqualified by the  

purported exercise of  moderation.    As we have already noted above,  

instead of deducting 18 marks if even 15 marks had been deducted, the  

Appellant who has scored the highest marks before moderation and the

Page 13

13

second highest  marks  even after  moderation,  would have  qualified for  

being called to the Interview/viva voce.  A grave injustice has been caused  

to the Appellant.  The learned Division Bench should have been alive to  

this injustice since it had before it the judicial determination of the learned  

Single Judge.  We shall abjure from making any further observation.

10 On the first hearing of this matter, we had been informed by Mr.  

Gopal Subramanium, learned Senior Counsel that one vacancy has been  

preserved during the pendency of the Writ Petition as well as the Appeal,  

which position we had ordered should continue.

11 The Appeal  is  allowed and the Impugned Judgment  is set  aside.  

Since  an  effort  of  moderation  has  been  carried  out  we  hold  that  a  

deduction of 10 marks being the appropriate mean arrived at be deducted  

from the initial marks of 55 obtained by the Appellant in Paper II.  Despite  

this deduction the Appellant will remain at second position in the merit  

list.  In fact, the position would be identical even if the Appellant were to  

be  awarded the minimum marks  of  40  per  cent,  i.e.,  by  deducting 15  

marks  from  the  original  marks  obtained  by  her.   We  direct  the  

Respondent-High Court to interview the Appellant within one week from  

its  receiving knowledge of  this  Judgment.   The result  thereof  must  be

Page 14

14

declared  within  one  week thereafter.   The  Hon’ble  Chief  Justice  shall  

ensure that the Interview Committee does not comprise any of the three  

Examiners.  If the result remains unfavourable to the Appellant the post  

shall remain unfilled for a period of 30 days therefrom.  

12 There will be no orders as to costs.

......………………….…J.      [VIKRAMAJIT SEN]

                                                                                 ......………………….…J.

    [C. NAGAPPAN] New Delhi; February 19, 2015.