R v. Taiters; ex parte Attorney-General of Queensland [1996] QCA 232 (16 July 1996)

The circumstances in which these questions have arisen from the trial of the accused Taiters may be briefly stated.

On 24 December 1993 a fight occurred in the street between Taiters and a man called Cooper. Cooper suffered certain injuries and died in consequence on 3 January 1994. In the course of the fight Taiters had struck Cooper causing him to fall heavily and strike his head on the cement footpath. Cooper was taken to hospital but did not initially remain an inmate and was allowed to leave. Certain symptoms persisted and he was readmitted when it was discovered that he had a fractured skull. Contusions and swelling of the brain became apparent and notwithstanding treatment that was administered, Cooper's death followed. The present matter falls to be dealt with on the basis that Taiters caused the death. We are not concerned with the physical mechanisms that played a part following the initial injury caused.

Taiters faced trial on a manslaughter charge. The Crown led its evidence and after hearing argument the trial Judge ruled that he would direct the jury that the accused should be acquitted on the basis that it was not open to the jury to be satisfied beyond reasonable doubt that the accused was guilty of the offence charged. That course being indicated by the Judge, the Crown prosecutor sought the return of the indictment and entered a nolle prosequi. We are not concerned with the correctness of the action of the learned Judge in concluding that in the exercise of his function as trial judge the matter could and should be taken from the consideration of the jury. No question of the kind considered by the High Court in Doney v. R [1990] HCA 51; (1990) 171 C.L.R. 207 presently arises.

In the argument put to us on the reference it was suggested that trial Judges continue to experience difficulty in giving correct and clear directions to juries when cases arise involving the application of s.23 of the Code. We are particularly concerned with the following part of that section:

"Subject to the express provisions of this Code relating to negligent acts and omissions, a person is not criminally responsible for an act or omission which occurs independently of the exercise of [the person's] will, or for an event which occurs by accident."

Leaving on one side the reference there made to "omission" it is seen that attention is directed both to acts which occur independently of the exercise of the will and, separately, to events which occur by accident. A number of earlier debates upon the operation of this section have been settled by the decision of the High Court in Kaporonovski v. The Queen [1973] HCA 35; (1975) 133 C.L.R. 209 and in particular the judgment delivered in that case by Gibbs J agreed in by Stephen J. The reasons there delivered by Gibbs J. have acquired additional weight as a result of the pronouncement of the majority of the High Court in Van Den Bemd [1994] HCA 56; (1994) 179 CLR 137. There the High Court refused special leave to appeal from a judgment of this Court. In refusing special leave the majority appeared content to accept the correctness of the interpretation placed upon s.23 by this Court in Van Den Bemd [1995] 1 Qd.R. 401 supported as it was by comments made by members of the High Court in Kaporonovski, particularly Gibbs J.

It is clear, then, that certain statements of principle of this Court in Van Den Bemd will be acted upon by trial Judges when, in the future, they sum up to juries. What was said in Van Den Bemd is of considerable importance. It should now be taken that in the construction of s.23 the reference to "act" is to "some physical action apart from its consequences" and the reference to "event" in the context of occurring by accident is a reference to "the consequences of the act". Even if, as has been said, there can on occasion be some difficulty, in an exceptional case, in distinguishing the border line between act and event so viewed, this theoretical distinction is clear. Taking an example from Kaporonovski itself, the thrusting of the glass by the accused was the act and the injury to the victim's eye which constituted the grievous bodily harm was the event. A number of occurrences can as a result of the operation of one or more chains of causation follow upon the doing of an act. However, s.23 is concerned to excuse from criminal liability so the relevant event for the purpose of the section should be taken to be the one which, apart from the operation of the section, would constitute some factual element of an offence which might be charged. In cases when grievous bodily harm is charged the state of bodily harm will be the relevant event and when unlawful killing is charged, the death will be the relevant event. A study of the judgments in Kaporonovski and Van Den Bemd is persuasive that this view should be accepted as correct notwithstanding the observations to be found in certain earlier cases and in R v. Hodgetts and Jackson [1990] 1 Qd.R. 456.

The principle now established in respect of the second part of the rule appearing in the quoted words of s.23 may be taken as stated by Gibbs J at 231:

"It must now be regarded as settled that an event occurs by accident within the meaning of the rule if it was a consequence which was not in fact intended or foreseen by the accused and would not reasonably have been foreseen by an ordinary person".

In terms of what the Crown must prove to exclude the operation of the defence of accident under s.23 it has been noted that double negatives can sometimes appear in the words in which trial judges sum up and it has been suggested that they may be troublesome for juries. Such phrasing has been subject to adverse comment, see e.g. the observations of Stanley J in R v. Knutsen [1963] Qd.R.157 at 178 and the observations of this Court in R v. Whiting [1995] 2 Qd.R.199 at 201. In the latter case the Court said:

"It is desirable in directing a jury to avoid where possible the use of a double negative, which is a form of speech favoured by lawyers rather than by ordinary members of the public."

"It would no doubt be legitimate for a Judge in summing up to reformulate the critical question in affirmative terms without doing injustice to the accused."

However, it was not found necessary in Whiting to suggest an appropriate positive form of direction such as would be consistent with the words of the statute as they now stand authoritatively interpreted. Reference was made in argument to a further case, R v. Ellem (1994) 75 A.Crim.R.370, as illustrating problems which continue to be experienced when trial judges direct on s.23.

It is possible that one observation made by this Court in Van Den Bemd may subsequently have been misconstrued and so become the source of some of the difficulties. At page 405 it was said:

"The test thus appears to be one of foreseeability of the happening of the consequence as a matter of probability or 'likelihood'".

Reference was immediately thereafter made to R v. Knutsen (supra) and R v. Tralka [1965] Qd.R.225. Later, however, on the same page, this Court said:

"In the present context that means that the relevant question was whether the jury were satisfied beyond reasonable doubt that Bankier's death was not such an unlikely consequence of the punches delivered by the accused that it could not have been foreseen by an ordinary person in the position of the accused".

This last statement is in obvious conformity with the words of the test propounded by Gibbs J. (It should be noted that the original uncorrected reasons of the Court of Appeal read "Bankier's death was such an unlikely consequence ..." which is the form in which the quotation appears at page 138 of the report of proceedings in the High Court in volume 179, CLR. The sense of the passage requires the "not" omitted from the quoted phrase to be read in.)

One question is whether the first statement quoted from page 405 of Van Den Bemd was intended to introduce some departure from or modification of the test or was, on the contrary, meant to be no more than a neutral statement of it. It is to be noted that the Code itself contains no definition of "accident". In turning to provide an exact meaning of the s.23 phrase, Gibbs J. did not directly and in positive form state what the Crown has to prove. Since, as has been indicated, the onus lies on the Crown, it will be called upon to exclude the excuse raised by s.23 when the evidence in particular proceedings requires the issue to be dealt with. Any judge choosing to cast the Gibbs J. test in positive form will not be obliged to adhere to a particular formula, but this being said he will, of course, be obliged to put the substance of the matter correctly. It seems that the time has come for this Court to attempt to state the formula in an acceptable positive form.

It is important to continue to bear in mind that under the Code we are concerned with something that is there described by the term "accident". It is also clear that within the Code concept there is an objective element, something beyond what might have been subjectively contemplated by the perpetrator of the act in question. If it is to be an accident in the Code sense it must be not intended or foreseen by the one acting. This is one essential requirement. What is involved in the concept of subjectively "foreseen" can be put to one side for the moment while attention is given to the further objective element.

To be accidental in the relevant sense there is the additional requirement that the event must be one which "would not reasonably have been foreseen by an ordinary person". Looking again at the first quoted statement by this Court on the nature of the test at 405 in Van Den Bemd the question is whether it was there intended to expound at all upon the degree of likelihood or extent of probability involved. The alternative view is that the Court did not mean to make any precise statement upon necessary degrees of likelihood and probability and was intending to do no more than draw general attention to the fact that matters of likelihood and probability are involved in the notion of "foreseeable". Immediately after making the statement referred to at 405 the Court mentioned the two cases of Knutsen and Tralka. When examined, it is seen that the judgments in those cases make no precise rulings on the degree of likelihood or extent of foreseeability which will exclude the finding of accident when the objective aspect of the test comes to be applied. Thus the judgment of Philp J. in Knutsen talks of an event which "might well" occur (see at 168) while that of Stanley J. (see at 174) refers to matters that are foreseeable, natural and probable as relevant in the application of the test. A factor which may be contributing to difficulty is that there is ambiguity rather than precision in the use of the words "likely or probable" in the present context. It may be of some assistance at this point in the discussion to note that in dealing with the differently phrased provisions of the Tasmanian Code, Kitto J. in Vallance v. The Queen [1961] HCA 42; (1961) 108 C.L.R. 56 offered certain helpful commentary upon the extent of prevision which will exclude a conclusion that an event has, within the meaning of the Tasmanian Code section occurred "by chance" (see at 65). Of course, the solution for the problems arising under the Queensland Code provision must be independently investigated.

The meaning that the Courts now give to "accident" in the Code has evolved out of past attempts to make various parts of the Code work harmoniously together and in this process some attention has been given, for example, to s.289 which deals with negligent acts causing personal injury. There was also an obvious need to give reasonably distinct meanings to "acts" such as might occur independently of the will and the contrasted accidental "events" both of which concepts are referred to in s.23. It was probable that some need was felt to make any meaning attributed to "accident" one which was reasonably conformable with accepted notions of criminal responsibility.

When the meaning of a word in the Code is doubtful it has been declared permissible to look beyond the bare context of the Code itself to find the correct meaning: see per Stanley J. in Knutsen (supra) at 171 and Gibbs J. in at Stuart v. The Queen [1974] HCA 54; (1974) 134 C.L.R. 426 at 437.

With this history and driven by influences of the kind referred to, we now have the comparative certainty of meaning of s.23 which is established in Kaporonovski and Van Den Bemd. It is important not to retreat into the realm of uncertainty which preceded these authoritative pronouncements. The fact that the Code version of "accident" must satisfy an objective requirement as well as a subjective one may have the effect that the meaning is somewhat different from that accepted in other more general contexts. However, Gibbs J. in effect indicated in R v. Tralka (supra) at 233 that it should not have been supposed that the sense of the word as it appears in s.23 would be wholly subjective.

If the suggestion made in Whiting (supra) is acted upon and an acceptable formula in positive form is sought, the authoritative statements found in Kaporonovski and Van Den Bemd must first be regarded. As has been mentioned double negatives tend to intrude when some conversion of the test in terms of what the Crown must prove is undertaken so that the summing up appears in some form like this - `must show that it is not such an event that it could not reasonably have been foreseen, etc.'.

In undertaking the necessary analysis, it has to be said first that an event cannot qualify as an accident within the meaning of s.23 simply because a reasonable person, although regarding the consequence as being a likely outcome, would have thought it more probable that it would not happen than that it would. To conclude differently on this point would, it is suggested, involve a rejection of the concept as expounded in Kaporonovski. The discussion may be carried further and instances at either end of the spectrum looked at. If the outcome of some action is regarded as certain or even just more probable than not, it cannot legitimately be called accidental. Even if there is a substantial likelihood although something less than a preponderance of probability that a particular outcome will occur and the risk of the outcome is voluntarily accepted by the one acting, it should not, if it results, be called accidental. On the other hand, something which a reasonable man might think of as no more than a remote possibility which does not call to be taken into account and guarded against can, when it happens, be fairly described as accidental.

The references which have been made in the cases to "reasonably" and "ordinary person" in the context under discussion, give an emphasis to the fact that the relevant test calls for a practical approach and is not concerned with theoretical remote possibilities. It directs inquiry to what would be present in the mind of an ordinary person acting in the circumstances with the usual limited time for assessing probabilities, this being a factor which is applicable to a great deal of human activity. However, it should not be accepted that some real risk of an outcome which an ordinary person in the circumstances would have been conscious of, can be disregarded by the doer of an action, yet still, if it eventuates, be called accidental within the meaning of the section. In the subjective part of the expression being considered under s.23 ("an event which occurs by accident"), i.e. when it is necessary to consider "foreseen" by the accused, the same degrees of likelihood will be regarded as those discussed in connection with the objective test.

By way of summary and looking at the matter from the point of view of the prosecution, it can be said that if the circumstances of the case call for the s.23 defence of accident, i.e. that based on the words "an event which occurs by accident", to be excluded, the applicable onus will be sufficiently stated if the jury is told that:

"The Crown is obliged to establish that the accused intended that the event in question should occur or foresaw it as a possible outcome, or that an ordinary person in the position of the accused would reasonably have foreseen the event as a possible outcome. "

This casts the matter in an acceptable positive form. If this direction is given it will be desirable for the trial judge to add that in considering the possibility of an outcome the jury should exclude possibilities that are no more than remote and speculative.

Although the questions posed for the Court's answer do not direct attention to all of the critical details which have been discussed, they should now be responded to: