[1956] EACA 33

## APPELLATE CRIMINAL

## Before Sir Kenneth O'Connor, C. J., De Lestang and Edmonds, J. J.

| THE QUEEN, Appellant | | | | |----------------------|--|--|--|

$\mathcal{A} = \mathcal{A}_1$

$v$ .

## RAOJIBHAI GIRDHARBHAI PATEL AND ANOTHER, Respondents. Criminal Appeal No. 2 of 1956

**CASE STATED**

Criminal Procedure Code, section 367—Indian Evidence Act—Whether wire recorder and wire recording admissible in evidence.

This was a case stated by a resident magistrate under section 367 of the Criminal Procedure Code, upon the application of the Attorney-General.

The first respondent was charged before a resident magistrate in Nairobi on three counts of forgery of jury summonses, and both respondents were charged on a fourth count with conspiracy to defeat the course of justice contrary to section 112 $(a)$ of the Penal Code. The magistrate acquitted the respondents on all the charges; but, at the instance of the Attorney-General, stated a case for the determination by the Supreme Court of certain points of law arising in connexion with the charge of conspiracy against both respondents.

The case is reported only on the question of admissibility in evidence of a wire recorder, wire recording and transcripts.

In the Court below the prosecution tendered a wire recorder by means of which a magistrate had recorded the numbers of a panel of jurors chosen by ballot by him for a forthcoming Supreme Court trial. It was alleged by the prosecution that a list containing other numbers had, in pursuance of an unlawful conspiracy, been substituted by the respondents for the numbers selected by the magistrate. The wire recorder, a recording made by it, and transcripts of that recording were tendered by the prosecution as evidence of the actual numbers which had been chosen, but the lower Court held that they were not admissible in evidence.

Held (14-2-56).—Inter alia that the Magistrate was not correct in deciding that the wire recorder and recording and transcription were not admissible in evidence.

Cases cited: R. v. Santokh Singh, Cr. Case 2323/1954; R. v. Kapurchand, Cr. App. 187 of 1954; R. v. Jono Kenyatta, Cr. App. 276 of 1953; Nicholas v. Penny, (1950) 2<br>K. B. 466; Emperor v. Ashootosh, I. L. R., 4 Cal. 483; Buxton v. Cumming, (1937) 71 S. J. 232.

Authorities cited: Sarkar on Indian Evidence Act, 9th edn., p. 24, 520, 521; Monir on Indian Evidence Act, 3rd edn., p. 15; Phipson on Evidence, 9th edn., 493; Maxwell on Statutes, 10th edn., p. 79.

Conroy, Solicitor-General, with him D. P. R. O'Beirne, for the Crown.

**B.** O'Donovan for the respondents.

The opinion of the Court (O'Connor, C. J., de Lestang and Edmonds, J. J.) was delivered by O'Connor, C. J.

JUDGMENT (Relevant portions only).—Question 6 asks whether the magistrate was correct in deciding that the wire recorder and recording tendered by Mr. Sampson and the transcriptions of the recording made by him and John Henry Baker (P. W. 63) were not admissible in evidence.

The matter arose in this way. As already mentioned, Mr. Sampson had been forewarned by the police of the plot to "rig" the jury and, when he drew the additional 50 numbers on 11th October, he had, concealed upon his person, a wire recorder connected to a small microphone in the shape of a wrist watch. He stated that he made a record of the numbers which he drew, by means of the wire recorder: that he handed the recorder to A. S. P. Lawrence later that day. In Court he identified the wire recorder and said that when he listened to the playing of the wire recorder he had heard his own voice and had heard what his voice was saying: he had made a transcript of what he heard his voice saying.

The recorder had been kept in police custody. Senior Superintendent Baker gave evidence that he received it on 12-10-55. The machine then had a spool inside it. This was played back by him and he made a transcript of what it said. He produced the transcript, which he said was accurate. The recorder was still kept in police custody. On 17th October he made another transcript which he also produced. The transcript which Mr. Sampson had made and the wire recorder and the transcripts made by Senior Superintendent Baker were tendered in evidence, but were objected to by the defence.

The magistrate heard legal argument as to the admissibility in evidence of the wire recorder, the transcript made by Mr. Sampson and the two transcripts made by Senior Superintendent Baker, and he heard the evidence of a Mr. Cripps (P. W. 65) who spoke as an expert on wire recorders. Mr. Cripps testified *inter alia* that on the morning he gave evidence, that is on 5th December, 1955, he and A. S. P. Lawrence had made a test of the wire recorder and its microphone and found them to be in good working order. When played back, he recognized his own voice and that of A. S. P. Lawrence and it was a true recording of what they had said. He played back another part of the same spool of wire and heard someone shouting out some numbers. He was satisfied that the spool was in good working order. This witness then explained the working of wire recorders, and what happ.ens to the wire when they are operated. His explanation may be summarized as under :-The diaphragm of the microphone is vibrated by the speaker's voice and a small electric current is generated which is amplified and passes ro the recorder head which magnetizes the wire at variable amounts of magnestism along it : this changes the position of electrons in the wire, but makes no visible marks on the wire. When played back, the impulses put on to the wire generate a small electric current whioh is amplified and, when connected to the diaphragms of headphones or the diaphragm of a loudspeaker, makes it "possible to hear an exact reproduction of what was said into the microphone". No change in the wire is visible to the naked eye or (apparently) to the eye assisted by a lens or a microscope: there is nothing on the wire perceptible by the visual sense.

The ground of objection raised by the defence to the admission in evidence of the wire recorder and transcriptions was that they were not "evidence" within section 3 of the Indian Evidence Act which defines evidence as used in the Act (unless a contrary intention appears from the context) as (l) statements made by witnesses, i.e. oral evidence and (2) documents produced for the inspection of the Court. "Document" is defined as meaning "any matter expressed or described upon any substance by means of letters, figures or marks, or by more than one of t,hose means, intended to be used or which may be used, for the purpose of recording that matter". Jt was argued that the wire in the recorder was clearly not a statement of a witness and neither was it a document, as it had no letters, figures or marks expressed or described upon it. Three cases were cited in support of the proposition that the recording was not admissible: *-Reg. v. Santokh Singh* (Criminal Case 2323/54); *Kapurchand v. Reg.* (Criminal Appeal 187 of 1955); and *Joma Kenyatta v. Reg.* (Criminal Appeal 276 of 1953). In *Santokh Singh's case* the magistrate decided not to admit evidence of a wire recording of a conversation, as there was no evidence of recognition of the voices. In *Kapurchand's case* a tape recording was not permitted by the magistrate · to be played in court, but a transcription was admitted. On appeal to this Court, it was argued that the tape recording was a "document" and should have been permitted to be played. It was argued that the matter recorded on it was "marks" within the definition of "document" in section 3 of the Indian Evidence Act. This Court rejected that argument and said (at page 8 of the judgment"): -

"We do not decide that in no case could a tape recording be permitted to be played in Court, if a proper foundation were to be laid for this to be

done or that such a recording could never be 'evidence' or a matter which, though not 'evidence' within the definition, a Court might legitimately take into consideration *(see Monir,* 3rd edition, page 15). What we do decide is that the tape was not shown to be 'evidence' in this case and that. having regard to the fact that the transcription and translation were admitted, no prejudice has been occasioned to the appellant, whether the learned magistrate was right in refusing to have the tape played in Court or not."

In *Jomo Kenyatta's case* (which was decided before *Kapurchand's case* but which was not drawn to the attention of the Court in that case) a tape recording of a speech at a meeting had been sought to be put in evidence by the defence. The magistrate held that this was not a "document". On appeal this Court held *inter alia* that the tape recording was not a "document" within the definition of section 3 of the Indian Evidence Act. The Court also held that a sound-track on a film was not a "document" as, though it had marks, these were not capable of conveying information to the visual sense. For the same reason, the Court expressed the opinion that a gramophone ,record, though it had ma.tier on it expressed by marks, was not a "document" within the definition. No authority was cited for the decision as to the sound-track or the gramophone record or for the proposition that "marks" must be capable of conveying information to the visual sense. These questions do not arise in the instant case. We have found difficulty with the passage in the judgment in the Kenyatta appeal in which the learned Judges who heard that appeal, in holding that a witness could not be called to prove that t,he sound-track and recording were the same as the words used by him at the meeting, said "the only admissible evidence as to the sound-track and recording would have been the evidence of someone who was in a position to say on oath 'I was present at the Kiambu meeting and it is within my personal knowledge that this particular sound-track and this particular tape recording are respectively a sound recording and a tape recording of that speech'." It is not necessary in order to prove the signature of X, to call a witness (other than X) who saw the signature made: neither is it necessary, in order to prove that a photograph is a photograph of X, to call the person who took the photograph. The evidence of a witness who is familiar with the signature or the features of X can be admitted to prove respectively x·s signature or that the photograph appears to be a photograph of him. Similarly, the evidence of **X** himself can be admitted to prove his signature or his photograph. With the greatest respect .to the learned Judges who heard the Kenyatta appeal. we see no reason why the evidence of a witness whose words have been recorded should not similarly be admissible to say that he has played back the record and that he then recognized his voice and the substance of what he said on the occasion in question. There was evidence in the insta!1t case that it is difficult fur a person to recognize his own voice when recorded. But that is a matter of the weight of the evidence not of its admissibility. Howeve.r, the point dealt with in ohe abovequoted passage from Kenyatta's appeal does not strictly arise in the instant case, as the recorder and spool were identified as those which Mr. Sampson used.

The learned magistrate held a "trial within a trial" to decide whether the recorder, recording and transcripts were admissible. Mr. Sampson testified *inter alia* that when the recorder was played back to him on 1-11-55 he heard his own voice reading out numbers consecutively: that he recognized the ·numbers and thought from certain numbers which he recognized that the recording was that made by him while drawing the numbers on 11th October; and that was the only occasion when he had spoken into any recorder such a list of numbers. He said that he had heard the recording played again on 4-11-55 and had then recognized i,t as the same recording as he had heard on 1-11-55: and he recognized his voice and the contents as being the same as he had heard on 1-11-55. He then (on 4-11-55) made a note of what he said. **He** also recognized on tihe

recording another voice, which appeared to be the voice of the first respondent. Mr. Sampson gave, in his evidence in chief, ten numbers, of which he said that had a recollection independently of the recording. In cross-examination, however, he said that he had recited the numbers as a result of his recollection having been refreshed by the recording.

As already mentioned, Mr. Cripps gave evidence as to the mechanism of wire recorders and as to tests made on the recorder in question.

The learned magistrate rejected the recorder, the wire and the transcriptions on the grounds that "There is no evidence before this Court that this recorder is an accurate machine". Clearly that finding was erroneous. There was some evidence before the magistrate of the accuracy of the recorder. It had been in police custody from 11th October to 5th December, 1955, when Mr. Cripps tested it and he said that it and the identical spool on another part of which Mr. Sampson had made his recording were in good working order on 5th December and that the mechanism of wire recorders made it possible to "hear an exact reproduction of what was said into the microphone". There was also Mr. Sampson's evidence that he recognized his voice and the voice of the first respondent and some of the numbers. Clearly this was some evidence from which an inference of the accuracy of the machine and its recording on 11th October might have been drawn. It is correct, however, that there was no evidence that the machine was tested for accuracy on or about 11th October and no direct evidence that it was then accurate. This may have been what the magistrate meant.

Evidence of the accuracy of the recording was not necessary as a matter of law to render the recording admissible. As Lord Goddard, L. C. J., said in Nicholas v. Penny, (1950) 2 K. B. 466, "If evidence is given that a mechanical device such as a watch or a speedometer—and I can see no difference in principle between a watch and a speedometer—recorded a particular time or speed, which it is the purpose of that instrument to record, that can by itself be prima facie evidence on which a Court can act, of that time or speed. It might be that in any particular case a Court would refuse to act on such evidence..." Nicholas v. Penny was a case of a traffic policeman following a speeding car and testing its speed by the speedometer on his own car. It was held, upon a Case Stated, that the justices could convict on the policeman's testimony of what his speedometer recorded without evidence of the accuracy of the speedometer. We think that the principle enunciated by Lord Goddard, L. C. J., is applicable to the case of a tape recorder. It is the purpose of such an instrument to record sounds conveyed into its microphone. What emerges from a playback is prima facie probative of what went into the microphone and the evidence of a witness as to what it says is receivable; but, in any particular case, a Court might refuse to act on the recording or on such evidence without proof of the accuracy of the machine. While, in the present case, the magistrate was wrong in saying that there was no evidence of accuracy, there was no evidence that the recorder had been tested and found to be accurate on or about 11th October, which was the crucial time. And it is reasonably plain that the magistrate, rightly or wrongly, did not feel convinced that the machine was accurate. This was a question which went to weight and not admissibility, but it is a factor which we cannot ignore in deciding whether to order a rehearing.

Was the recorder itself admissible in evidence? We think that clearly it was. Mr. Sampson said "I used a recorder and this is it". The Crown produced evidence of its condition. It was open to the Court to require its production for inspection under the second proviso to section 60 and section 165 of the Indian Evidence Act. Sirkar in his commentary on the proviso to section 60 says (9th edition at page 521): "The definition of evidence in section 3 does not cover everything that the Court has before it in arriving at a decision. There are other

media of proof, e.g. material objects produced. . . " The fallacy in the argument for the defence was that it assumed that a Court can only take into consideration "evidence" as defined in section 3 of the Indian Evidence Act. But "evidence" is used in the Act in a very restricted sense and by no means comprehends all the matters which a Court may take into consideration. The learned authors of Woodroffe and Amir Ali's commentary on the Law of Evidence (9th edition at pages 113 and 114) point out that the definition in section 3 of the Act does not include all matters which the Court may take into consideration, e.g. real evidence such as the property stolen, weapons, etc. These are relevant facts to be proved by "evidence", i.e. by oral testimony of those who know of them: section 3 is an interpretation clause and the legislature only explained by it what it intended to denote whenever the word evidence is used in the Act: Emperor v. Ashootosh 4 Cal. 483, 492 (F. B.). Sarkar (9th edition at page 24) writes:-

"The definition of the word "evidence" does not cover everything that the court has before it in arriving at a decision. Besides depositions taken in court and documents which come under the head of evidence, the Judge may have other matters on which to found his conclusions: (1) Material objects may be produced in court and so by section 60 it is provided that if oral evidence refers to the existence or condition of a material thing, other than a document, the court may require the production of it, and on the same principle, section 218 Cr. P. Code directs the transmission by the committing magistrate to the sessions Court of weapons and other things referred to by the witnesses."

Sarkar lists a number of other matters which a Court may take into consideration which are not included in the definition of "evidence" as that expression is used in the Act, and says (at page 25): "In addition to the depositions of witnesses and documents all these materials are matters which according to the last paragraph but two of section 3, the Court has to consider in arriving at a conclusion." Monir in his commentary on the definition of "evidence" in section 3 of the Indian Evidence Act (3rd edition at page 15) writes: -

"Therefore, matters other than the statements of witnesses and documents" produced for the inspection of the Court, e.g. statements made by parties when expressed otherwise than as witnesses, demeanour of witnesses, the results of local investigation or inspection, and material objects other than documents, such as weapons, tools, stolen property are not 'evidence' according to the definition given in the Act. They are, however, matters which the Court may legitimately take into consideration ... 'evidence' as defined by the Act is not the only medium of proof."

We have no doubt that the tape recorder and its spool were admissible as real evidence for inspection. What other use could be made of them and whether the spool (wire) could properly have been played back in Court is a more difficult question.

The learned Solicitor-General has argued that the magistrate should have listened to a playback and should have himself compared the sounds thus produced with Mr. Sampson's voice, and, if he came to the conclusion that this was a recording made by Mr. Sampson, should have admitted its contents in evidence. We think, however, that it is in principle undesirable that the Court should (onstitute itself an expert to decide whether a recorded voice is identical with a voice of a witness heard in Court, and that (on the analogy of proof of handwriting) a witness familiar with the voice of the person who made the record and who has listened to the playback should be called to prove the identity. Evidence of identity from the person who himself made the record and afterwards recognized the playback would be admissible; but, owing to the fact that some persons cannot recognize their own recorded voices, the evidence of another person would carry more weight. Subject to identity being properly

established, if the question in issue is what was said at a particular time in the. course of, e.g. an interview, or a speech or declaration, and evidence is given that a wire recording of it was made simultaneously, we do not see any reason in principle why the recording should not be admitted in evidence and be permitted to be played in Court. It is not a "document" and it is not "evidence" within the restricted meaning of that term used in the Indian Evidence Act; but, as we have already said, many matte.rs may be taken into consideration by a Court working under the Indian Evidence Act which are not "evidence" in the sense defined in section 3. The recording is evidentiary in the sense that it is a matter of fact the tendency of which is to produce in the mind a persuasion of the existence of some other matter of fact. *(See* Bentham's and Best's definitions of "evidence" quoted in *Sarkar,* 9th edition, at page 21.)

*Buxton v. Cumming,* (1937) 71 **S. J.** 232, was a case in which Buxton, a bookmaker, claimed from Cumming £1,000 under an agreement in writing whereby Cumming had agreed to pay that sum. The £1,000 claimed was in respect of gaming and wagering t·ransactions which the defendant had entered into with the plaintiff. The matter had been referred to the National Turf Protection Society. The secretary of that society gave evidence of an interview which he had had with the defendant when he had endeavoured to get the defendant to agree to pay in order to avoid being reported to Tattersalls. This witness stated that immediately after the interview he had dictated into dictaphone his recollection of what took place. Swift, J ., then raised the question whether a dicta phone record had ever been accepted in evidence by the Courts, and upon counsel replying that he did not Uhink so, said that he saw no reason why such a record as the one which the witness said he had made should not be put in evidence. Presumably this must have meant that the record could be playe.d. It does not a.ppear from the report in the *Solicitors' Journal* that the record was actually played; but in *Phipson* on Evidence (9th edition at page 493) the learned author says: "In *Buxton v. Cumming* 71 **S. J.** 232, a dicta phone record of a witness's recollection of an interview, made by him immediately after the interview, was allowed to be put in". *Sarkar* (9th edition at page 520) cites *Buxton v. Cumming* as authority for the proposition that "the dictaphone !fecord of an interview may be a.ccept.ed in evidence." In *Kapurchand's case (supra)* this Court mentioned that a gramophone record had been permitted to be played in a Magistrate's Court in England (1954 Criminal Law Review, page 4). This Court went on to say that it was unaware of any case of a ta.pe recording having been permitted to be· played in a Court in England and t1hat, in any event, caution would have to be exercised in following English precedents in this matter, as there was not in England any statutory definition of "evidence" or of "document" which requires matter to be exp,ressed or described by letters, figures or marks. This Court was then unaware of the *Buxton v. Cumming case.* On going into the matter further than it was necessary to pursue it in *Kapurchand's case,* we are now more firmly convinced than we were in *Karpurchand's case* that the restrictive effect of the definition of "evidence" and "document" in section 3 of the Indian Evidence Act it not important, since it seems to be well settled vhat a Court may take into consideration matters which do not fall within the definition of "evidence" in the Act. Accordingly, we now have less hesitation in following the English decisions. It is clear from Sarkar's reference to *Buxton v. Cumming,* mentioned above, that that learned author considered it to be an authority applicable to India. We do not think that it makes any difference whether tlhe d·ictaphone in *Buxton's case* was of a type where marks are impressed upon a disc or cylindeir or which operates, like a tape recorder or wire recorde.r, by invisible rearrangement of magnetized electrons.

The authorities are extremely meagre, but we a-re of opinion that a tape recording may be played back in a Court in Kenya if a proper foundation has been laid. That founda,tion would include evidence of a particular recording having been made on the tape recorder and of the nature of the matter recorded, and evidence (which should not be merely the opinion of the Judge, jury or magistra,te) of the identity of the voice or voices of the person or persons who had made the recording with the voice or voices sounding from the record as played back.

There is one respect in which *Buxton v. Cumming* differs fundamentally from the instant case. In *Buxton's case* the dicta.phone record, being a record of dictated notes of the witness's recollection of an interview, could, presumably only be played to refresh the witness's memory of the interview. We notice that *Buxton's case* is cited in Phipson's treatise on Evidence in the section which deals with refreshing memory. We do not think that a witness could refresh his memory *while under examination in a Court in Kenya* by having a tape recording played to him. The reason is that section 159 of the Indian Evidence Act which deals with refreshing memory in Court refers expressly to a "writing" and we do not think that a tape recording is a "writing".

As stated, the instant case differs from *Buxton v. Cumming* in that what was recorded was not someone's recollection of an interview dictated afterwards (which could only be used for refreshing memory) but the actual interview itself. We think that this was probative *per se* (which a note of an interview made after it is over is not) and that, accordingly, this could have been given in Court otherwise than for the purpose of refreSlhing memory. If we are right, and the recording of the interview could have been played in Court, then we think that the transcriptions previously made could also be admitted when properly proved, for the purpose of testing the accuracy and consistency of the machine.

The answer to question 6 is, therefore, No.

We have come to this conclusion without g1vmg effect to the principle referred to by the Solicitor-General that the language of a statute may be extended to new things which could not have been contemplated by the legislature when it was passed: tha.t this occurs when the statute deals with a genus and the thing which afterwards comes into existence is a species of it. *(Maxwell* Statutes, l 0th edition, page 79). This principle is well settled. The learned Solicitor-General went on to argue that a tape ,recording was in the genus "document" or "writing". If so, and if this principle is applicable to extend the meaning of "document" and "writing" in the Indian Evidence Act to include tape recordings, then *cadit quaestio:* they must be admissible in evidence; but we have doubts whethe,r the genus "document" or "writing" includes tape recordings of t•he spoken word.

Notwithstanding that we a.re of opinion that, in the instan,t case, the recording and transcriptions could have been admitted, we shall not direct a rehearing with the recording admitted because we think it would be undesirable at this stage and unnecessary and because, in any event, the magistrate (in the absence of direct evidence of the accuracy of the machine on or about 11th October) does not seem to have attached weight to the recording.

We have indicated that Mr. Sampson could not have been a1lowed to refresh his memory while under examination in Court by having the recording played back. But there was nothing to prevent him ;refreshing his memory of the numbers from the recording before he came to Court, provided that be disclosed (as he did) that this had been done. His evidence based on recollection so n,freshed was admissible. The weight to be attached to it might depend upon whether his ,recollecti9n was partly :independent or entirely dependent on the recording, and upon whether the magist-rate was satisfied as to the accuracy of the recording. *(Nicholas v. Penny (supra.})*