Sarjoo Prosad, C.J.
1. A Division Bench of this Court has referred the following points for decision by the Full Bench :
'Whether the authorities of this Court in (i) Dhokalsingh v. State, ILR (1953) 3 Raj 762 and (ii) Ranjita v. State, in Appeal No. 152 of 1957, D/- 14-2-1958 (Raj), lay down the correct law on the following matters :
(1) Whether it is necessary that entries shouldbe made in the various police records of the precautions that were to be taken for keeping the accused person ba-parda while under Police custody;
(2) Whether it should be specified in the warrant of commitment of the accused, when he is sent to the judicial custody that he is to be kept ba-parda till the identification parade takes place, and what precautions should the jail authorities take for keeping the accused ba-parda.
(3) Whether necessarily-entries should be made in the jail records for keeping the accused ba-parda while he is in the judicial lock up.' In making the reference, the Bench was of the opinion that the above questions were of vital importance and arose for consideration in a large number of cases; and since from time to time the observations made in those cases had led to misunderstanding on the part of the subordinate Courts as if they contained a rigid and inflexible rule of law and not a rule for the appreciation of evidence, the Bench thought that the points raised required examination by a larger Bench. It is significant that one of the Hon'ble Judges, who is a party to this reference, was himself a party to the decision in the later case of Ranjita v. State, in Appeal No. 152 of 1957.
2. For the purpose of deciding the points under reference, it is but necessary to quote the relevant observation made in Dhokalsingh's case ILR (1953) 3 Raj 762, (supra), which has been quoted also in the order of reference itself which has presented a source of doubt and difficulty. The passage runs as follows :--
'It is incumbent oh the prosecution in every case, where an accused person is not known from before, to put him up for identification. If therefore, it is desirable that an identification should take place it is the duty of the police in every such case, when an accused person is arrested, to warn him to keep his face hidden as he is to be put up for identification. It is also the duty of the police to take steps themselves to see that such an accused is conveyed Ba-parda (under proper cover) from the place of his arrest to the thana or wherever else he is taken and from there to the lockup or the jail. It is also necessary for the police to see that, while such an accused is kept at the thana, there are proper arrangements so that no one is able to see him. Further, if the accused is kept in a lock-up or a jail from which he has to be taken out for purposes of easing himself or for his bath or for any other purpose, it is again the duty of the police to see that proper arrangements are made to keep him Ba-parda when he is so taken out, till the identification is over. These precautions become all the more necessary in a case like the present where the appellants started saying from the moment they were put up for identification that they had been shown to the witnesses. It is also necessary that proper entries in the various police records should be made of the precautions that have been taken to keep the accused persons Ba-parda, and evidence of police constables or other police officers who having made those entries and who kept the accused Ba-parda is produced in court. This evidence, in our opinion, should be produced in every case, and it is absolutely essential to produce it where the police knows that the accused is contending that he was shown to the witnesses before the identification.'
A perusal of the passage indicates that it is based upon a strong suspicion of the conduct of the police and that of the identifying witnesses. But is it reasonable or fair to start with any such assumption? No one doubts the position that where the accused are put up for identification by witnesses to whom they are strangers, every precaution should be taken against the witnesses having an opportunity of seeing the accused prior to the stage of the test identification parade. If an accused has been shown to the witnesses or the witnesses have had the opportunity of seeing him prior to the test identification, the identification becomes meaningless and loses all its significance. In that event, the witnesses would easily pick out the accused from a number of other persons in the test identification parade. Therefore, precautions are necessary against any such opportunity being given to the witnesses, either at the instance of the witnesses themselves or at the instance of the police.
The Court should as a rule of caution and prudence insist on proper identification proceedingsduring the course of investigation. But the point is, whether there is any justification in law or prudence for compelling the police necessarily and in all such cases, to follow the elaborate procedure indicated in the passage quoted above. Where the identifying witnesses are present in the locality and the record shows that there was a possibility of their seeing the accused in the lock-up or out-side, it should be necessary to demand a satisfactory explanation from the prosecution that in truth the witnesses had not seen the accused and that precautions had been taken against any such possibility. Otherwise, what justification is there for insisting that the police should always keep the accused Ba-parda even when they are taken out of the lock-up for attending to calls of nature. An accused, who is innocent, might as well resent any such indecent restraint on his movements. Even from the practical point of view the directions contained in the above passage, if I may say so with, the utmost respect, do not carry much significance. There would not be much sense in keeping the accused Ba-parda if a witness were really so revengeful as to be out to identify him or to see his face at any cost, or the police officer were dishonest enough to show the accused to the witnesses. The mere entry in the police diary or in the warrant of custody that he had been or should be kept Ba-Parda will be no safeguard, for those entries can be made even without complying with the actual requirement. A dishonest police officer can always make such a routine entry and defeat the very purpose of the directions.
I also fail to understand why the mere failure of an identifying witness to state in his deposition that he had not seen the accused prior to the stage of identification, should discredit his testimony. To throw out the sworn and valuable testimony of the witness on that ground alone, is as bad as accepting his testimony on the bare statement that he had not seen the accused prior to the identification parade. I, therefore, do not see what safety or guarantee is provided merely by making an entry in the police diary or in the jail warrant that the accused was kept Ba-Parda, when in fact it is not observed in practice. A dishonest police officer might even destroy the prosecution case by omitting to make such an entry if courts rigidly insist upon, the observance of this formality, for then on a mere omission to make such entries, the evidence of the eye-witnesses will have to be discarded.
Why should we assume that the witnesses, who are strangers to the accused, will be necessarily out to identify them at any cost; and why should we assume that the police officers, who in most cases, are equally strangers to the accused, would be so dishonest as to show them to the witnesses and even persuade them to identify the accused in the manner suggested by the police officers? In my opinion, these assumptions are clearly unjustified. I am inclined to agree with the observations of James. J. in Ashrafi v. The State, AIR 1961 All 153, where the learned Judge said :
'In dealing with such questions it is often ignored that the accused is a total stranger to thewitnesses and that save for exceptional cases he is a stranger to the police too, hence neither the witnesses nor the police have any motive for incriminating him falsely.'
In our opinion, rulings of this nature as illustrated by the passage quoted from Dhokalsingh's case, ILR (1953) 3 Raj 762, unfortunately accentuate the difficulties of honest investigating officers and truthful witnesses, unless they are kept within their proper perspective.
3. At this stage it is important to remember the purpose which test identification actually serves in the investigation of a crime or in the trial of the criminal. The evidence of personal identification in Court has been the subject-matter of numerous conflicting decisions which peculiarly attracts the warning given by the Supreme Court of the United States in Aldridge v. United States, (1930) 283 US 308, and which has been adopted with approval in other authoritative pronouncements in India :
'Courts ought not to increase the difficulties by magnifying theoretical possibilities. It is their province to deal with matters, actual or material, to promote order and not to hinder it by excessive theorising or by magnifying what in practice is not really important.'
We wish that this wholesome advice should be always kept in view when Courts are dealing with the evidence on record. It is well established that the result of an identification parade conducted at the stage of investigation is not a piece of substantive evidence in a case and cannot be the basis of conviction. The substantive evidence against the accused, is that which is given in Court by the identifying witnesses in the witness-box; yet test identification parades have their own importance in establishing the identity of the culprit. They provide a very good piece of corroborative evidence and greatly enhance the credibility of the evidence of identification given in Court. Identification parades are, therefore, held not for the purpose of giving defence Advocates materials to work upon, but in order to satisfy investigating officers of the bona fides of the prosecution witnesses and that they are on the right track.
The whole idea of such a parade is to see if the witness, who claims to have seen the accused at the time of the occurrence, can identify him from amongst a number of other persons without aid from any other source. If he can, then it becomes more or less certain that the evidence of identification as deposed to by the witness in Court is true and genuine. As a rule of prudence before a Court can accept the evidence of identification as sufficient to establish the identity of an unknown accused, the Court necessarily looks to find some good corroborative evidence in support of that evidence. That corroborative evidence may come from various circumstantial factors; one of them being the evidence of test identification where the witness picks out the accused from amongst a number of persons with whom he is mixed up. A precaution of the kind, which is insisted upon, to support the evidence of identification given in Court, is generally necessary, because of the frailty of human memory. It is possible that the witness who saw a person at the time of the commission of the crime may have had only a fleeting impression of the miscreant, which impression is likely to fade off as time passes.
There may be therefore the possibility of a bona fide mistake or error when after a long lapse of time the witness comes to identify the accused for the first time in Court. The impression may be strengthened when it is fresh in the mind of the witness if an opportunity is given to him to identify the stranger culprit soon after the commission of the crime under circumstances where there is no danger of the culprit being picked out by the identifying witness on account of any extraneous factor or influence. It cannot be denied that the mere fact that a person is in the dock as an accused is often likely to influence the mind of a witness and make him think that the person in the dock is the person whom he had seen committing the crime and thus reduce or nullify the evidentiary value of the identification evidence given in Court.
4. In a criminal offence, identification serves a two-fold object. Firstly it is meant to satisfy the investigating authorities before sending up a case for trial to Court that the person arrested, but not previously known to the witness, was one of those who committed the crime, or the property concerned was the subject of such crime; secondly, it is to satisfy the Court that the accused was a real offender or the article was concerned with the crime which was under trial. Identification proceedings, therefore, are as much in the interest of the prosecution as in the interest of the accused. When a witness points to a stranger in Court that he was the offender or claims an article and affirms that it was his property which had been stolen, there is no guarantee necessarily of the truth of his assertion and corroborative factors should be looked into to see if the evidence of the witness is reliable.
Consequently, in order to have some assurance of the truth, a test identification is held. The witness at an earlier stage is confronted with the alleged offender not standing alone but mixed up With a number of innocent persons of almost similar built, age-group and feature, or the suspected article is mixed up with a number of other articles which resemble it, that is to say, it is to give credence to the evidence of a witness who has not known the accused from before or who has not seen the article subsequent to the commission of the offence that a test identification is held. With all this precaution, however, the Court should never lose sight of the fact that the main evidence before it is the deposition, in Court.
It would be wrong to start with a suspicion of the evidence given on oath and insist that unless the witness adhered to some kind of a formula in stating that he had not actually seen the accused, who was a stranger to him, after the occurrence and before the test identification parade, his evidence must be rejected. Even in the absence of test identification the evidence may be such as to convince the Court of its truth and veracity. Corroboration of the evidence may be derived from various circumstantial factors: for instance, the sufficiency of the light, the distance from which the witness watched, the period for which he could see the accused and the part which the accused played in the occurrence. All these circumstances may enable the witness to have a clear and vivid impression of the features of the accused who committed the crime.
There may be also cases where the witnesses themselves received injuries at the hands of certain participants in the crime whose features they had sufficient opportunity to note. There is no reason why in the presence of these circumstantial details which corroborate the evidence of personal identification, the Court should not act on the convincing evidence given before it, merely because there was no test identification parade held earlier. It would be, therefore, wrong to insist upon the value of evidence at test identification as an invariable corroborative factor like a mathematical formula in each case. What the Court is concerned to judge is the substantive evidence which is placed before it. These rules are mere rules of caution and prudence and must not be subjected to excessive theorising or converted into mathematical formula so as to completely fetter the judgment of the Court in weighing the evidence before it.
5. Phipson in his Law of Evidence observes:
'In criminal cases it is improper to identify the accused only when in the. dock; the police should place him, beforehand with others, and ask the witness to pick him out.''
Similarly, in Davies v. The King, 57 Com WLR 170 (181), appears the dictum that it was
'indisputable that a witness, if shown to the person to be identified singly and as the person whom the police have reason to suspect, will be much more likely, however fair and careful he may be, to assent to the view that the man he is shown corresponds to his recollection.''
In Craig v. The King, 49 Com WLR 429 (450), the learned Judges quoted the observations of Lord Guthrie:
'It would not be safe to convict the prisoner merely on the evidence of personal impression of his identity with the man seen fleeing from the house, on the part of strangers to him, without reference to any marked personality or personal peculiarities, and without corroboration derived from other kinds of evidence. My proposition involves a distinction between the identification, by personal impression, of a strange person, and the identification, by personal impression, of a familiar person.'
These observations merely lead to the conclusion that the sworn testimony of witnesses in Court as to the identity of strangers, speaking generally, require corroboration which may be in the form of an earlier identification proceeding. These rules of caution, however, do not tie down the discretion of the Court to this extent that even if it is satisfied that the evidence of a particular witness is such that it can safely rely on it without the precaution of an earlier identification proceeding, the Court is powerless to base a conviction on that evidence. There are exceptional cases where theCourt on account of the intrinsic value of the evidence on record and its adequacy would be prepared to rely safely on the evidence in Court. The matter appears to have been clarified and explained in a subsequent decision of this Court in Bhurgiri v. The State, ILR (1954) 4 Raj 476, where the learned Chief Justice, who was a party to the earlier decision, has himself discussed the importance and limitations, of a test identification parade on the lines stated by us earlier.
It is obvious from these discussions that the value of sworn testimony of witnesses in Court should not be lightly discarded on the mere assumption that there was no test identification parade held earlier or that the test identification was not in proper form, provided that the evidence On record on the point of identification is otherwise adequate and convincing. In our opinion, therefore, it would be going too far to suggest that all that precaution of Ba Parda with corresponding routine entries in the police diary or in the jail warrant are absolutely essential. All that is necessary is to see whether there is anything on the record to indicate that the accused had been either deliberately shown to the witnesses prior to the test identification parade or that the witnesses had an opportunity of seeing the accused and being pointed out their identity before the test identification.
If there is any such indication on the record and the investigating authorities have taken no precaution against it, then alone no value should be attached to the test identification otherwise there can be no justification for insisting that the identifying witnesses should unnecessarily go on repeating in their evidence that they had not seen the accused before the test identification parade and if they failed to do so all the value of their evidence in Court should be lost merely on that account.
As we said, if we insist on any such formula, there is nothing to prevent the witnesses, if they are dishonest enough to do so, to go on repeating this formula in their examination in chief. It is always open to the defence to bring out any such facts in cross-examination of the witnesses which may lead to the inference that the witnesses had already an opportunity of seeing the accused and, therefore, no reliance should be placed upon their test identification. Normally, it is expected that the police, before they place the accused for identification at a test identification parade, will take every precaution to see that the witnesses have not already seen the accused, unless it appears that they had some personal motive of reward or revenge or undue professional enthusiasm in having a person identified as an accused at any cost by even winning over the identifying witnesses to act according to their desire.
These are undoubtedly far-fetched assumptions, though in some cases the evidence might justify an inference of this kind. It is also erroneous, in my opinion, to attach too much weight to the assertion of the accused, unsupported by any cogent material, that he or they had been shown to the witnesses prior to the test identification parade. Experience shows that almost invariablythis defence is taken in order to destroy the value of the test identification and if the Courts were to act merely on the ipse dixit of the accused and thereby throw an unnecessary burden on the prosecution to make the identifying witnesses repeat in each case that they had not seen the accused earlier, they would be, in my opinion, adopting a meaningless formula which is neither warranted by logic or law.
6. In conclusion, therefore, we are of opinion that the propositions laid down in ILR (1953) 3, Raj 762, cannot possibly be regarded as a rule of law. If I may say so with the utmost respect, they have been laid down far too broadly to merit acceptance even as practical propositions and can only lead to the accentuation of the difficulties of honest investigating officers and truthful witnesses. To the extent, therefore, that these propositions are in conflict with the views expressed by us earlier, they cannot be regarded as useful guide to Courts in the appreciation of the evidence on record. We would, therefore, answer the questions under reference in the following manner :
(1) It is not necessary that entry should be made in the various police records of the precautions that were taken for keeping the accused persons Ba Parda while under police custody.
(2) It is also not necessary to specify in the warrant of commitment of the accused when he is, sent to judicial custody that he is to be kept Ba Parda till the identification parade takes place, nor is it necessary to specify the precautions that the jail authorities are to take for keeping the accused Ba Parda.
(3) It is also not necessary that entries should be made in the jail records for keeping the accused Ba Parda while he is in the judicial lock up.
7. It is for the police authorities to specifyadministratively what precautions they would liketo take in order to avoid the accused being seenby the identifying witnesses prior to the test identification so that the value of their identificationmay not be lost; but it is unnecessary for theCourt to lay down hide-bound rules for the conduct of the police in a matter of this nature. Muchwill depend upon the circumstances of each casein evaluating the evidence of identification. Tolay down any hard and fast rule would be to unduly curtail the judicial discretion of the Courtwhich after all, is the best judge of the evidenceplaced before it.