Evidence in Criminal Proceedings: Hearsay and related topics [1997] EWLC 245(7) (19 June 1997)

BAILII is celebrating 24 years of free online access to the law! Would you consider making a contribution?
No donation is too small. If every visitor before 31 December gives just £1, it will have a significant impact on BAILII's ability to continue providing free access to the law.
Thank you very much for your support!

7.1 We now consider how the rule against hearsay should be formulated. We focus on the distinction between assertions and direct evidence, (1) and a variety of cases in which this distinction has given rise to difficulty. (2) We explain the formulation that we proposed in the consultation paper for the purpose of avoiding these difficulties, (3) and how we have modified it. (4) Finally we consider the special problem of statements generated by machines. (5)

ASSERTIONS AND DIRECT EVIDENCE

7.2 There are two basic ways of proving a fact in issue. First, it may be proved by proving some other fact which renders it more likely to be true: the other fact is directly probative of the fact to be proved. Second, it may be proved by means of a persons assertion that it is true. The hearsay rule applies only to the latter form of proof.

7.3 An assertion can consist of words or conduct or both. But, merely because a persons words or conduct are relied upon as evidence of a fact, it does not follow that they are an assertion of that fact. For example, a persons words may betray guilty knowledge without necessarily amounting to a confession of that persons guilt.

7.4 Often it will be clear whether a persons words or conduct are adduced as proof of a fact on the basis that they are directly probative of it in other words, if it were not true then that person would probably not have spoken those words or acted in that way or on the basis that they amount to an assertion of it. But it is sometimes debatable which of these is the case.

Borderline cases

"Implied assertions"

7.5 In Wright v Doe d Tatham, (6) for example, the issue was whether letters written to a man in which the writers appeared to assume the sanity of the recipient could be evidence of his sanity. Parke B held that the letters were hearsay because they were not directly probative of the fact to be proved, but only an assertion of it. He explained his decision with a now notorious illustration of a sea-captain who boards a ship, from which fact a court might be tempted to infer that the ship was sea-worthy. Parke B said the hearsay rule would apply to such conduct, and evidence of it would be inadmissible. (7)

7.6 Parke Bs approach was approved by a bare majority (8) of the House of Lords in Kearley. (9) The issue was whether, on a charge of possessing drugs with intent to supply, a prosecutor could rely on evidence by the police that they had been to the home of the defendant when he was not there, and had there received telephone and personal calls from people (who were not called as witnesses) asking about drugs that the defendant had for sale. It was held that the hearsay rule applies where it is sought to draw an inference of a fact from words or conduct which are intended to be assertive of some other fact, or are not intended to be assertive at all. As evidence of the fact that the defendant dealt in drugs, the callers words were therefore hearsay; and, being unable to find any applicable exception to the rule, the majority of the House held them inadmissible. (10)

7.7 Evidence of the kind that was excluded in Wright and Kearley is usually referred to as an implied assertion. This is a somewhat unfortunate expression, for two reasons. First, it begs the question of whether the words or conduct in question are an assertion of the fact that they are adduced to prove. It is at least arguable that they are not assertive at all, but directly probative in which case it would follow that they should not be caught by the hearsay rule.

7.8 Second, the word "implied" is here used in an unusual sense. Normally it refers to a statement which is not expressly spoken or written but is intended to be understood from what is said or done. But where there is an assertion of the fact to be proved, it is immaterial whether that assertion is express or (in the ordinary sense) implied. An assertion of a fact is no less of an assertion because it is implicit in an express assertion of a different fact, or because it takes the form of non-verbal conduct such as a gesture. An assertion can therefore be implied (in the ordinary sense) without being what is described in the context of hearsay as an "implied assertion". (11)

7.9 Some respondents argued that "implied assertions" should fall within the hearsay rule, on the ground that, if an assertion would be inadmissible hearsay if made expressly, it should not make any difference that that assertion is implied. As one respondent put it, "Both are hearsay, plain and simple". We agree that all assertions should be caught by the rule if they are adduced as evidence of the fact asserted, irrespective of whether they are express or implied. But this is not the issue, and it is only the use of the expression "implied assertions" that suggests it is. The question is whether, in a case such as Wright or Kearley, there is an assertion at all.

Negative assertions

7.10 Closely connected to the problem of "implied assertions" is that of negative assertions. For the purposes of the hearsay rule it is obviously immaterial whether the fact to be proved is positive or negative, provided that an assertion of the fact is adduced as evidence of its truth. The difficulty arises where it is debatable whether the evidence of a negative fact is an assertion of it, or a fact suggesting in some other way that it is true.

7.11 Suppose, for example, that the fact to be proved is the fact that a particular event did not occur. The fact-finders may be invited to reason that, if it had occurred, its occurrence would have been recorded; and that, since its occurrence was not recorded, it did not occur. But is the non-recording of the event an assertion that it did not occur, or is it directly probative? In Shone (12) the evidence of a stock clerk and a sales manager that workers would have made entries on record cards if certain items had been lawfully disposed of, that there were no such entries, and that those items must therefore have been stolen, was held not to be hearsay but direct evidence of that fact. It seems that, if an inference is drawn from what a document says, the document is hearsay; but if an inference is drawn from what it does not say (or from the fact that no document exists), that is direct evidence.

7.12 Similarly, the hearsay rule may not apply where the court is invited to infer a negative fact from the fact that certain words were not spoken. Thus in Harry (13) the accuseds counsel sought to ask police witnesses about seven telephone calls made to the premises which Harry had occupied with the co-accused, P. None of the callers had asked for the appellant, and most had asked for P. The jury were to be invited to infer that it was P, not Harry, who was dealing in drugs. The fact that the callers had asked for P was held inadmissible, either as evidence against P (14) or to exculpate Harry; but the fact that they had not asked for Harry was admissible. What the callers said was hearsay; what they did not say was direct evidence.

Identification evidence

7.13 Yet another difficult case is that in which it is sought to adduce an utterance or writing as evidence of identification. The person identified may be the very person who is alleged to have spoken or written the words relied upon, or some other person. In either case it may be doubtful whether the words in question are an assertion of that persons identity, or are directly probative of it.

7.14 In Rice (15)the Crown adduced a used airline ticket to Manchester, bearing the names "Rice and Moore", in support of the evidence of a co-defendant named Hoather that he had flown to Manchester with Rice at about the time of the flight to which the ticket related, and that Rice had booked their ticket. (16) The Court of Appeal doubted that the ticket could be admissible as evidence that the booking had been made by a person called Rice: for that purpose it was hearsay. But it was held to be permissible for the jury to infer that the ticket had been used by someone called Rice, because of

the balance of probability recognised by common sense and common knowledge that an air ticket which has been used on a flight and which has a name upon it has more likely than not been used by a man of that name (17)

7.15 The distinction between these two uses of the ticket seems artificial: it was no more likely that the ticket had been used by someone called Rice than that it had been issued to someone of that name. (18) In our view it would have been better to treat the ticket as direct evidence, and admissible, on the latter issue as well as the former.

7.16 Rice was considered, and a similar conclusion reached, in Lydon. (19) A gun, allegedly used in a robbery, had been found by the side of a road which would have been used by the getaway car. Nearby were found two rolled-up pieces of paper bearing the words "Sean rules". The appellants first name was Sean. It was held that this was not hearsay but direct evidence.

The inference that the jury could draw from the words written on the piece of paper is that the paper had been in the possession of someone who wished to write "Sean rules", and that person would presumably either be named Sean himself or at least be associated with such a person, and thus it creates an inferential link with the appellant. (20)

It would have made no difference to the evidences admissibility, but only to its weight, if the pieces of paper had borne the appellants name in full. They were not so much an assertion by the writer (that his name was Sean) as something that a person not named Sean (and not associated with such a person) would be unlikely to write.

Our provisional proposal: intention to assert

7.17 These borderline cases are not precisely analogous to one another. But they appear to have at least one feature in common, and one which is absent in the case of statements to which the hearsay rule clearly applies. This common feature is the fact that, while the words or conduct relied upon may in fact cause others to draw certain inferences, it is not the intention of the person whose words or conduct are in question that they should have that effect. The crucial question, we believe, is not whether any particular kind of assertion should be excluded from the ambit of the hearsay rule, but whether words or conduct which are not intended to assert a fact should be treated, for the purposes of the rule, as amounting to an assertion of that fact at all.

7.18 If it is known that a person spoke or acted in such a way as to cause someone else to infer the truth of a particular proposition, two inferences may be drawn: first that that person at that time believed that proposition to be true, and second that that belief was correct. Neither inference is inevitable: the person may have been seeking to mislead, or may have been mistaken. The hearsay rule recognises that if both these risks are present then, in the absence of an opportunity to cross-examine the person in question, there is good reason to exclude evidence of his or her words or conduct.

7.19 If, however, the risk of deliberate fabrication can be discounted, the possibility of a mistake is not necessarily sufficient reason to exclude evidence of the words or conduct. An example of this is the principle of res gestae, which, as preserved by our draft Bill, (21) permits evidence of a statement which "was made by a person so emotionally overpowered by an event that the possibility of concoction or distortion can be disregarded".

7.20 Where there is a substantial risk that an out-of-court assertion may have been deliberately fabricated, therefore, we think it right that the assertion should fall within the hearsay rule whether it is express or implied. It follows that the rule should extend to any conduct which is intended to give the impression that a particular fact is true, and is adduced as evidence of that fact. But where that risk is not present in other words, where the person from whose conduct a fact is to be inferred can safely be assumed to have believed that fact to be true we do not think a court should be precluded from inferring that fact merely because that person may have been mistaken in believing it. And if that person did not intend anyone to infer it, it follows that that person cannot have been seeking to mislead anyone about it.

7.21 We therefore take the view that a persons words or conduct should not be regarded as asserting a fact, and therefore should not be caught by the hearsay rule if adduced as evidence of that fact, unless that person intends to assert that fact.

7.22 In the consultation paper we noted that several other jurisdictions have excluded unintentional assertions from the ambit of the hearsay rule, (22) or have proposed doing so; (23) and we made a provisional proposal to this effect. Our proposed formulation of the hearsay rule was as follows:

7.23 On consultation, the thrust of this proposal met with much support. Some respondents argued that it should make no difference whether an out-of-court assertion is express or implied: if it is repeated in court, and the person who made it does not give evidence, the other party faces difficulties in challenging the reliability of the assertion and the credibility of its maker. But this argument appears to be directed at the admission of implied assertions in the ordinary sense, and not at evidence which is not intended to be assertive at all. We still believe that the borderline cases discussed above ought not, in general, to be caught by the hearsay rule.

Modification of the provisional proposal

7.24 However, many respondents, while agreeing that "implied assertions" should be taken outside the hearsay rule, had misgivings about our proposal to do this by formulating that rule in terms what the putative declarant intended to assert. We share these misgivings, and in the consultation paper drew attention to two practical difficulties with this approach.

First, admissibility then comes to depend on the chance of how an invidividual has expressed himself or herself, whether in a question or a direct statement. Secondly, cogent evidence would still be excluded. For example, a caller may say "Can I have my usual stuff?", which would be admissible, as containing no factual assertion, but the words of a caller who says "The stuff you sold me last week was bad" will be inadmissible. Yet there is no obvious reason why the second statement is any less reliable as evidence than the first, if the court is not interested in the quality of the drugs supplied. (25)

We now consider whether our provisional proposal can be modified in such a way as to meet these objections.

What is an intention to assert?

7.25 As more than one respondent pointed out, the idea of an "intention to assert" is ambiguous. The crucial question, we have argued, is whether the person whose words or conduct are in question intended to convey the impression that the fact which it is now sought to infer from those words or that conduct was true. Only if that person did not intend to convey that impression can it safely be assumed that he or she was not deliberately seeking to mislead. It follows that what is crucial is not the way in which that person happened to express himself or herself, but the impression that his or her words or conduct were intended to convey.

7.26 Thus evidence that a caller said "Can I have my usual stuff?" would be admissible to prove that the accused did habitually supply unlawful drugs; but this is not because the callers words do not amount to an assertion. It is because the callers intention is not to give anyone the impression that the person addressed is a drug-dealer, but simply to request drugs. The caller intends the words to be heard only by a person whom the caller believes to be a drug-dealer: obviously the caller has no wish to convince that person, or anyone else, that that person is a drug-dealer.

7.27 From this point of view it makes no difference that the caller says "The stuff you sold me last week was bad". The inference to be drawn from these words is essentially the same as in the case of the words "Can I have my usual stuff?" namely that the person whom the caller intends to address is in the habit of supplying drugs to the caller. On its face, admittedly, this is an express assertion that the person addressed sold drugs to the caller last week; and, since the caller obviously intends to say exactly what he or she does say, in one sense the caller intends to assert that fact. But it is not the callers intention to cause the person addressed to infer that that fact is true, since he or she already knows it. In that sense there is no intention to assert.

7.28 The point may be further illustrated by reference to Teper. (26) The defendant was charged with arson of his own shop. A woman had been heard to shout to a passing motorist "Your place burning and you going away from the fire". If the womans intention were to draw the attention of bystanders to the fact that Teper was leaving the scene, her words would be hearsay, since she might have been trying to mislead the bystanders. If, however, she was intending only to indicate to the motorist that she knew he was Teper, she could not be seeking to mislead anyone about who he was. If he was Teper, he knew he was; and if he was not, she could not hope to convince him that he was. She might still be asserting that he was Teper, but she would not be intending to persuade anyone of this.

7.29 We recognise that it may sometimes be difficult to ascertain what impression, if any, the words or conduct in question were intended to convey, and legal argument may result. (27) But a party seeking to adduce evidence must always show, to the appropriate standard of proof, that the facts are such as to render the evidence admissible; and it will usually be possible to infer from the circumstances the intentions of the person in question.

7.30 It may be difficult in some cases for the prosecution to prove beyond reasonable doubt that that person did not intend to convey to an observer the fact that it is now sought to prove; but, in view of the risk of fabricated evidence being admitted without the opportunity for cross-examination, we think it is right that there should be this safeguard.

7.31 In the case of defence evidence the point will have to be proved only on the balance of probabilities; but, if it appears more likely that the relevant intention did not exist than that it did, we think it right that the defence should be permitted to adduce the evidence. The possibility that it may have been fabricated can be taken into account in assessing the weight that it should be given.

7.32 This reasoning suggests that the crucial question should be, not whether the maker of the statement appears to have intended to assert the fact which the statement is adduced to prove, but whether he or she appears to have intended to cause another person to believe that fact.

7.33 However, we think it would be going too far to say that a statement should never fall within the hearsay rule unless it appears to have been intended to cause another to believe the fact stated. We have argued that a statement should be regarded as hearsay if it seems possible that it may have been deliberately fabricated; and there may be cases where a statement is deliberately fabricated although it is not intended that another person should believe it to be true. This may be so if it is intended that another person, while not necessarily believing the fact stated, should act on the basis that it is true.

7.34 Suppose, for example, that As job involves reimbursing his colleagues for their travelling expenses. It is sought to prove that his colleague B travelled to Glasgow on a particular date, by adducing her claim form in which she stated that she had done so. We believe that that evidence should fall within the hearsay rule, because the claim might be fabricated. But if it were necessary, in order to bring a statement within the hearsay rule, to show that it was made with the intention of causing a person to believe it, it might be argued that Bs claim is made with no such intention. It may be that, when a claim is submitted to A, all he is required to do is to check that it complies with the rules laid down for such claims; and if it does, he automatically pays it. If B tells him she has been to Glasgow on business, he will pay for that journey. He will not consider whether he believes that B has been to Glasgow, and it is probably of no concern to B whether he believes it or not. (28) But she does intend that he should act on the basis that it is true; and we believe that this should be sufficient to bring her statement within the hearsay rule.

7.35 Similarly, we believe that a statement should fall within the hearsay rule in any case where it is made with the intention that some action should be taken on the basis that the fact stated is true even if the taking of that action involves no human intervention, but only the operation of a machine. Suppose, for example, that the processing of Bs expenses claim is carried out not by A but by a computer system, which has been programmed to print a cheque for the appropriate amount if the information provided by the claimant appears to meet the specified criteria. Clearly the risk of fabrication is just as great as if the information were given to a human, and it would be arbitrary to apply different rules to the two cases.

Intention and purpose

7.36 We have so far been referring to the intention with which the statement in question is made. But the word "intention" is ambiguous. In some contexts it refers only to the purpose with which a person acts the objective that that person hopes to achieve by acting as he or she does. In others, it includes not only purpose but also what is sometimes called "oblique" intention. In this wider sense, a person "intends" not only the consequences that he or she wishes to bring about, but also those that he or she knows to be an inevitable side-effect of the consequences that he or she desires. In our recent reports we have used the word in the latter sense, reserving the word "purpose" for the former. (29)

7.37 The point is perhaps unlikely to be of great practical importance, but we have considered whether the applicability of the hearsay rule to a particular statement should depend on the intention (in our wider sense) or only on the purpose with which the statement is made. Should a persons words or conduct count as a hearsay statement of a fact if that person does not positively desire that another should thereby be caused to believe that fact (or that another should be caused to act, or a machine to operate, on the basis that it is true), but knows that this will inevitably occur?

7.38 Our reason for focusing on the "intention" of the putative declarant is that, if that person is not seeking to convey a particular impression, it follows that he or she cannot be seeking to convey a misleading impression: the possibility of deliberate fabrication is thus ruled out. (30) But this argument seems equally applicable where, although he or she knows that a particular inference will inevitably be drawn, that is not his or her purpose. Moreover, if we were to include within the hearsay rule the case where he or she knows that a particular inference will be drawn, it is hard to see any rational basis for excluding the case where he or she knows that it may be drawn. We believe that the most defensible place to draw the line is between those consequences that it is the putative declarants purpose to bring about, and those that it is not.

Multiple purposes

7.39 On the other hand we see no reason to confine the hearsay rule to statements which are made solely, or even primarily, for one of the specified purposes. Where a person has more than one purpose for what he or she says or does, we believe it should be sufficient that at least one of those purposes falls within the categories we have identified.

Our recommended formulation

(1) that (subject to the exceptions we recommend) in criminal proceedings a statement not made in oral evidence in the proceedings should not be admissible as evidence of any matter stated, and

(2) that a matter should be regarded as stated in a statement if (and only if) the purpose, or one of the purposes, of the person making the statement appears to the court to have been

7.41 In the borderline cases we described above, we believe that the difficulty of applying the law would be reduced by our recommended formulation of the hearsay rule. In a case such as Kearley or Harry, (32) for example, we think a court would normally have little difficulty in concluding that it was not the callers purpose to cause anyone to believe that anyone was (or was not) selling drugs. Similarly, where a person has failed to record an event, (33) it will often be clear that that failure was not intended to give the impression that the event had not occurred. Either the event did not occur, or that person did not realise that it had: in either case, it will not have been his or her purpose to cause anyone to believe that it had not happened, because it will not have crossed his or her mind that anyone might think it had. And therefore the record will be direct evidence that the event did not occur.

STATEMENTS PRODUCED BY MACHINES

7.42 There is an additional difficulty where it is sought to adduce a statement generated by a machine. Usually the statement will be included in a document produced by the machine, such as a computer printout; (34) but it could equally be a reading on a gauge, or even the mechanical equivalent of an oral statement.

7.43 The present law draws a distinction according to whether the statement consists of, or is based upon, only what the machine itself has observed; or whether it incorporates, or is based upon, information supplied by a human being.

Statements not based on human input

7.44 The hearsay rule does not apply to tapes, films or photographs which record a disputed incident actually taking place, (35) or to documents produced by machines which automatically record an event or circumstance (such as the making of a telephone call from a particular number, (36) or the level of alcohol in a persons breath). (37) In such a case the court is not being asked to accept the truth of an assertion made by any person. The evidence is not hearsay but real evidence.

7.45 Our draft Bill preserves this rule by confining the word "statement" to a representation made by a person. (38) The conclusions printed out (or "spoken") by a machine are not a statement for the purposes of the Bill, and therefore the hearsay rule does not apply to them.

Statements based on human input

7.46 By contrast, the present law does sometimes exclude evidence of a statement generated by a machine, where the statement is based on information fed into the machine by a human being. In such a case, it seems, the statement by the machine is admissible only if the facts on which it is based are themselves proved.

7.47 In Wood, (39) for example, it was sought to prove that certain metal found in the appellants possession was of the same type as a stolen consignment, by adducing evidence of figures produced by a computer which had analysed the results of X-rays and other tests carried out by chemists. It was held that this was not hearsay because the chemists had given oral evidence of the results of the tests. In the absence of admissible evidence of those results, the computers analysis of the results would not have been admissible either. In R v Coventry Justices, ex p Bullard, (40) on the other hand, a computer printout stating that a person was in arrears with his poll tax was held to be inadmissible hearsay because it must have been based on information "implanted" into the computer by a human, which had not been properly proved.

7.48 We believe that this distinction is well-founded and should clearly be preserved. In a case such as ex p Bullard it would be absurd to admit the printout without requiring proof of the input on which it was based. The question is, on what basis should such evidence be excluded? One view is that it is hearsay, because it is tantamount to a statement made by the person who fed the data into the machine. (41) An alternative view is that the statement by the machine, properly understood, is conditional on the accuracy of the data on which it is based; and that, if those data are not proved to have been accurate, the statement therefore has no probative value at all. The question of hearsay does not arise, because the statement is simply irrelevant.

7.49 We believe that the latter view is closer to the truth, and that it is therefore unnecessary to complicate our hearsay rule by extending it to statements made by machines on the basis of human input. On the other hand we do not think it would be safe to assume that everyone will share this view. We must anticipate the argument that, if such statements are inadmissible at present, that is because they are hearsay; that, under our recommendations, they would no longer be hearsay, because our formulation of the rule would apply only to representations made by people; and that they would therefore cease to be inadmissible.

7.50 We have therefore concluded that a separate provision is necessary, independent of the hearsay rule. We recommend that, where a representation of any fact is made otherwise than by a person, but depends for its accuracy on information supplied by a person, it should not be admissible as evidence of the fact unless it is proved that the information was accurate. (42) (Recommendation 3)

(7) Cross and Tapper points out, at p 591, that the illustration was unnecessary to Parke Bs conclusion, and that "It remains to be seen whether there is any authority to support it".

(8) Lords Bridge of Harwich, Ackner and Oliver of Aylmerton. Lords Griffiths and Browne-Wilkinson dissented.

(10) The majority of the judges in Kearley held the evidence in question to be inadmissible as being irrelevant in any event: per Lord Ackner at pp 253E254A, Lord Oliver of Aylmerton at p 271, and Lord Bridge of Harwich at p 243CG.

(11) Conversely, it may sometimes be arguable that even an express assertion of a fact should not be treated as being truly assertive in view of the purpose with which it is made. See para 7.27 below.

(14) See Kearley [1992] 2 AC 228 (paras 4.19 4.21 above), in which Harry was approved by Lords Ackner and Oliver of Aylmerton.

(16) Moore was the name of another co-defendant. It was suggested that Rice had booked the ticket with the intention that Moore should accompany him, and that Hoather had taken Moores place; but the jury were directed, rightly in the view of the Court of Appeal, that the ticket was not evidence against Moore.

(22) In Scotland such assertions have never been seen as falling within the hearsay rule. The Federal Rules of Evidence exclude them (rule 801(a)), as does the Australian Evidence Act 1995, s 59.

(23) See cl 27(2)(b) of the Evidence Code prepared by the Law Reform Commission of Canada, and NZLC Preliminary Paper No 15, Evidence Law: Hearsay (1991) pp 3233.

(27) This problem was anticipated by the Crown Prosecution Service, the Society of Public Teachers of Law, Mr Justice Curtis, Alan Suckling QC, Professor John Jackson, Peter Mirfield and David Ormerod.

(28) There is a similar difficulty in the law of deception: where a person assumes that everything is as it should be, and the defendant dishonestly takes advantage of that assumption, is the defendant obtaining by deception? See A T H Smith, "The Idea of Criminal Deception" [1982] Crim LR 721.

(29) See Legislating the Criminal Code: Offences Against the Person and General Principles (1993) Law Com No 218, paras 7.1 7.14, and cl 1(a) of the Criminal Law Bill annexed to that report.

(34) In that case, s 69 of PACE lays down further requirements which must be satisfied before the statement can be admitted, whether it is hearsay or not: see Part XIII below.

(35) Eg Dodson (1984) 79 Cr App R 220, in which the two accused were photographed by security cameras during their attempted robbery of a building society.

(37) Castle v Cross [1984] 1 WLR 1372. In that case, the disputed statement from the machine was not a blood-alcohol reading, but a statement that the defendant had failed to provide a sample of his breath large enough for it to analyse. See also Owens v Chesters (1985) 149 JP 295.

The Law Commission