Police v Smith, Mark Joseph

The first preliminary issue - is Dr Moses' evidence admissible (Evidence Act, s 79 & s 135)?

1. The prosecutor at the outset of these proceedings indicated that he objected to Dr Moses' evidence on the basis that it was opinion evidence which did not fall within the exception in s 79 of the Evidence Act; or alternatively that it should be excluded under s 135. By agreement, Dr Moses' evidence was taken in its entirety, along with all the other evidence in the case, and the question of admissibility was deferred until the end of the evidence. Such a course was both pragmatic and entirely appropriate in the context of the issues in this case. It is of course now necessary to resolve the question of admissibility before moving to any substantive determination.
2. The particular value of Dr Moses' evidence to the accused is said to be this: Dr Moses says the drug zolpidem, which is the active ingredient of stilnox, can on rare occasions induce a state of "parasomnia", in which people perform such complex actions as sleep-walking, sleep-cooking, sleep-eating, sleep-smoking, and, most relevantly, sleep-driving. Dr Moses describes it in this way (at p 13): The parasomnias are a a group of complex semi-purposeful physical or verbal behaviours that occur during the sleep period, of which sleepwalking is one.... Sleepwalking results from incomplete or partial arousal from slow-wave sleep. ... During sleepwalking the person appears to be awake, as their eyes are open and they can perform complex tasks and engage in conversation, albeit often in slurred or bizarre speech. If obtained, a sleepwalker's EEG (electrical brainwave activity) pattern usually shows they are asleep. Although many people perform long complex tasks such as driving and cooking whilst sleepwalking, it is characteristic for sleepwalkers to have no memory or, at best, 'glimpses' of their sleep-related behaviour.
3. Dr Moses goes on to express the opinion that in Mr Smith's case, the accused was "very likely" in such a zolpidem-induced state of parasomnia at the time he drove, and crashed, his vehicle.
4. There is no dispute that Dr Moses is a pharmacologist of considerable experience and renown. So far as s 79 is concerned, the prosecution contend that Dr Moses' opinion relies largely on "untested, anecdotal, self-reported evidence, from people who have had adverse reactions to stilnox and people who have been charged by police with criminal offences after these alleged adverse reactions." Specifically, as I understand the submissions, the objections to the evidence are twofold: 1. Dr Moses does not have "specialised knowledge" regarding zolpidem and parasomnias, because such specialised knowledge does not exist. Put another way, the state of the current scientific literature is not sufficient to establish either that "sleep-driving", as a phenomenon, can exist, or that zolpidem can cause it. 2. If such "specialised knowledge" does exist, then Dr Moses' opinion does not sufficiently demonstrate the process of reasoning by which she expresses the opinion that the accused in this case was probably in a state of zolpidem-induced parasomnia.
5. Dealing with the first issue, I note this. Although the prosecution advanced some skepticism in relation to the existence of zolpidem-induced parasomnia, as I understand the evidence they did not advance a direct forensic challenge to it as a phenomenon. Dr Judith Perl gave evidence on behalf of the prosecution. Her expertise was not called into question. Importantly though, Dr Perl herself did not deny that the phenomenon may exist; her evidence was limited, as I have described, to advancing skepticism, both with respect to the phenomenon generally, and with respect to its application in this case specifically.
6. Because of the way the evidence fell, and because this is a criminal prosecution (with the associated criminal standard of proof), it does not fall to me to try and make a determination as to whether the phenomenon is real or not. The ultimate question for me is (subject to the evidence I find to be admissible), whether the prosecution have proved beyond a reasonable doubt, either that it is not a real phenomenon, or if there is a reasonable possibility it is, that it was not affecting the accused at the time he was driving. Because of that, it is not necessary to review the minutae of Dr Moses' assertion that the phenomenon is real. I am satisfied that Dr Moses' evidence demonstrates sufficient scientific support for it to render the evidence admissible under s 79, and Dr Perl does not appear to me to put a contrary position. In addition to the detailed references to the evidence base contained in Dr Moses' report, I draw some comfort from the fact that the Therapeutic Goods Administration, this country's peak regulatory body, has, since 21 February 2008, imposed a box warning on zolpidem-containing medications which includes this: Zolpidem may be associated with potentially dangerous complex sleep-related behaviours which may include sleep walking, sleep driving and other bizarre behaviours.
7. So far as the second part of the objection is concerned, namely that Dr Moses' opinion does not sufficiently identify the process of reasoning through which she ultimately comes to her opinion, I disagree. The facts of this case, and the reasoning by which she applied them to her specialised knowledge, are quite clearly set out, particularly at sections 13 and 14 of her report, beginning at page 21.
8. Accordingly, I am of the view that the evidence of Dr Moses falls within the terms of s 79 of the Evidence Act.
9. The objection under s 135 has not been forcefully put. So far as it is based on s 135(a), that the evidence would be unfairly prejudicial to the prosecution, the objection is without merit. Unfair prejudice is the risk that the tribunal of fact will use the evidence in some illogical or irrational way, whether by affording it more weight than it deserves, or by some other means. It would be a rare case indeed in which such a risk arose where the tribunal of fact is a professional arbiter, such as a Magistrate or a Judge sitting alone. In my view, I am quite capable of assessing the rational force of the evidence, whilst taking into account the forensic criticisms made of it. The same can be said of the objection under s 135(b).
10. Accordingly, I admit the evidence.

The second preliminary issue - can evidence of zolpidem intoxication be taken into account on the question of voluntariness (s 428G, Crimes Act)?

1. Section 428G of the Crimes Act is in these terms: 428G Intoxication and the actus reus of an offence (1) In determining whether a person has committed an offence, evidence of self-induced intoxication cannot be taken into account in determining whether the relevant conduct was voluntary. (2) However, a person is not criminally responsible for an offence if the relevant conduct resulted from intoxication that was not self-induced.
2. In this case, Dr Moses asserts that the accused's involuntariness was brought about by ingestion of zolpidem. It is clear, though, that he also ingested alcohol. Dr Moses says that the consumption of alcohol can add to the effects of zolpidem, and, importantly, may increase the risk of a zolpidem-induced parasomnic episode. It is common ground that those who take zolpidem should not also drink alcohol, and that both doctors and pharmacists are under a continuing obligation to ensure that their patients are warned not to do so. The accused, though, asserts that he has never been given such a warning.
3. The Crown case, then, is that if the accused was in a state of parasomnia, it was brought about by a combination of alcohol and zolpidem, each of which he freely took. In those circumstances, if the alcohol triggered the zolpidem-induced parasomnia, then his condition is properly described as "self-induced intoxication", which, according to s 428G(1), cannot be taken into account on the question of voluntariness.
4. Self-induced intoxication is a defined term. The definition, contained in s 428A, is (relevantly) as follows: Self-induced intoxication means any intoxication except intoxication that: (a) ... (b) ... (c) results from the administration of a drug for which a prescription is required in accordance with the prescription of a medical practitioner, a registered nurse whose registration is endorsed under the Health Practitioner Regulation National Law as being qualified to practise as a nurse practitioner, a registered midwife whose registration is endorsed under the Health Practitioner Regulation National Law as being qualified to practise as a midwife practitioner, or dentist, or of a drug for which no prescription is required administered for the purpose, and in accordance with the dosage level recommended, in the manufacturer's instructions.
5. That definition of self-induced intoxication is, to my mind, drafted in exhaustive terms. In this case, the accused has given sworn evidence that the drug was prescribed for him by a doctor; that it was taken for a purpose for which it had been made, namely as a sleeping medication; and that he took the amount he was instructed to take. Not surprisingly, there is no evidence to the contrary. Applying the terms of the (somewhat cumbersomely drafted) definition, then, Mr Smith's evidence supports an assertion that any intoxication by zolpidem:

• resulted from the administration of a drug
• for which a prescription is required
• in accordance with the prescription of a medical practitioner
• administered for the purpose
• and in accordance with the dosage level recommended
• in the manufacturer's instructions.

1. Put in those terms, taking the definition of self-induced intoxication on its face, any intoxication by zolpidem was not self-induced, and hence must be taken into account on the question of voluntariness.
2. In Hadba v R (2004) 146 A Crim R 291, the ACT Supreme Court considered a very similar provision to the definition of self-induced intoxication in s 428A. In considering a similar question as to whether the appellant's intoxication was relevantly "caused" by alcohol or by some other prescription medication he was then taking, the Court said this (at [19]): The provisions of ss338 and 339 are not concerned with the voluntary consumption of alcohol or drugs per se, but rather with self-induced intoxication and it is the causation of the relevant state of intoxication that is the decisive issue. At least in the civil law it is now clear that issues of causation must be determined not by reference to the "but for" test but by the application of the "common sense" test described by the High Court in March v E & MH Stramare P/L [1991] HCA 12. In our opinion any question of whether intoxication came about as a consequence of one or more of the factors specified in s338(2) should be approached in a similar manner.
3. I confess that I respectfully hold some reservations about the correctness of their Honours' approach in Hadba, given that the term "self-induced intoxication" is, to my mind, exhaustively defined in s 428A. However, I need not venture further along those lines other than to express my reservation, since the nature of the evidence in this case is such that, even on the test in Hadba, the accused's intoxication in this case falls outside the definition of being self-induced. It is very clear from Dr Moses' evidence that the underlying pharmacological issue is that zolpidem is capable of inducing a state of parasomnia. The circumstances in which it will do so are not directly predictable. Alcohol is one risk factor, but there are others. As I understand her evidence, although drinking alcohol whilst taking zolpidem may in some individuals trigger a parasomnic episode, it will certainly not always do so, even within the same individual. Similarly, some individuals may have a zolpidem-induced parasomnic episode without taking alcohol. The parasomnic state, then, is fundamentally brought about, or caused, by the zolpidem, not anything else. Where, as here, the zolpidem was taken in accordance with a prescription, and in accordance with the other elements of the definition of non-self induced intoxication in s 428A, any behavioural sequellae must be characterised as non-self induced.
4. Accordingly, I find that the evidence relating to intoxication by zolpidem is to be taken into account on the question of voluntariness.

The third preliminary issue - has the issue of automatism been enlivened?

1. The prosecution submits that the accused has not discharged his evidentiary burden of raising the issue of automatism; that the evidence in the case has not sufficiently "enlivened" the issue. In particular, the prosecution submit that although the evidence in the defence case has raised a "bare" possibility that the accused was in a state of automatism, it has not raised a "reasonable possibility".
2. With respect to the prosecution, it seems to me that this submission may be attended by error. Although it was not specifically stated, this submission resonates strongly with the reasoning of the High Court in CTM v R (2008) 236 CLR 440, but the analogy is not valid. In CTM, the High Court determined that the defence of honest and reasonable mistake of fact applies to offences under s 66C of the Crimes Act, of sexual intercourse with a child under 16. The High Court confirmed that an honest and reasonable mistake of fact as to the age of the complainant is an available means of exculpation, and that, once raised, it is for the prosecution to disprove beyond reasonable doubt.
3. However, the majority in CTM found that the issue of honest and reasonable mistake had not been sufficiently "enlivened" in that case, and that the accused had not discharged his evidentiary burden. Specifically, the majority found that although there was evidence raising the bare issue of mistake, there was no evidence capable of raising either the honesty or the reasonableness of that mistake.
4. The prosecution argue at page 4 of their submissions that although there may be evidence to point to a "possibility" that the accused was in a state of automatism, there is no evidence as to its reasonableness - they submit that "the defendant has not raised to any satisfactory evidentiary standard that it is a 'reasonable possibility'". As I have noted, whilst that submission does have some resonance with the language of the High Court in CTM, there is a fundamental difference between CTM and the present case. Although neither honest and reasonable mistake nor automatism are a defence in the true legal sense, they nevertheless have different components - something akin to elements - that the prosecution must disprove. An honest and reasonable mistake of fact has three such components - that it is honest, that it is reasonable, and that the mistake is one of fact. Where there is some evidence capable of "enlivening" each of those three matters, then the onus rests on the prosecution to disprove them beyond reasonable doubt. Reasonableness, however, is not an independent component of automatism, and the only involvement of the concept of "reasonableness" is that the prosecution must disprove automatism beyond "reasonable" doubt.
5. In this way, it appears that the prosecution may have somewhat conflated the issue of exactly what the content of the evidentiary burden should be, perhaps (either advertently or inadvertently) because of the similar language in CTM. Whatever be the case, I am firmly of the view that the evidentiary onus on the accused goes no further than to raise the issue of automatism in a genuine way upon the evidence, and that having done so, the onus is on the prosecution to prove beyond reasonable doubt that it is not so. Attempts at parsing the meaning of the criminal onus and burden of proof are fraught with difficulty, as courts at all levels have long recognised: see for example Green v R (1971) 126 CLR 28.
6. Ultimately, then, on this second preliminary issue, I am satisfied that the accused has discharged the evidentiary burden of putting automatism in issue. That being so, it remains to be determined whether the prosecution have proved, beyond reasonable doubt, that the accused was not in such a state at the time of driving.
7. Before turning to answer that question, though, and the last remaining preliminary issue, there is one other observation I should make regarding the onus of proof, which is this. The prosecution submissions focus on many ways in which the accused could have supported his claim, but failed to do so. Those matters are collected at page 11 of the written submissions, and include such matters as the fact that the accused has not produced any prescription, and that he did not call any doctor or pharmacist to give evidence on his behalf. There was cross-examination about some of those matters, but not others. In approaching those submissions, I am mindful of the need to ensure that I do not reverse the onus of proof. A number of the matters referred to in the submissions, including the hospital records, and, once Dr Moses' report was furnished, evidence from treating doctors and dispensing pharmacists, could have been obtained by the prosecution. As a consequence of the cross-examination, some of those matters bear upon the accused's credibility, but beyond that, I remind myself that the burden of proof is on, and remains on, the prosecution.

The fourth preliminary issue - is the evidence directed towards a finding of sane or insane automatism?

1. I mention this issue for completeness. It has not been strongly litigated by either party, but since in my view it arises on the evidence, it is appropriate to make brief comment upon it.
2. It is clearly the case that not everyone who takes stilnox will experience parasomnia. Nor will everyone who takes stilnox and alcohol together have such an experience. Indeed, there is material in Dr Moses' report which is suggestive of an underlying predisposition in the accused to have such episodes. At page 9 she says this: It should be noted that throughout the ten years of taking zolpidem, Mr Smith experienced at least three episodes of parasomnia that he knows about. These episodes are relevant to this case in that they represent Mr Smith's idiosyncratic predisposition/vulnerability to the sleep-walking side effects of zolpidem, since only a minority of people are affected in this way.
3. The distinction between sane and insane automatism is explained in paragraphs [65]-[67] of Woodbridge v R [2010] NSWCCA 185, which was helpfully extracted in the prosecution's written submissions. The Court said this: 65 In The Queen v Radford (1985) 42 SASR 266 at 276 King CJ explained the difference between sane automatism and insane automatism as follows: The significant distinction is between the reaction of an unsound mind to its own delusions or to external stimuli on the one hand and the reaction of a sound mind to external stimuli, including stress producing factors, on the other hand. 66 This statement was said by Hunt J in R v Youssef (1990) 50 A Crim R 1 at 5 to be the clearest statement of the distinction. King CJ's discussion of the distinction was approved in The Queen v Falconer (1990) 171 CLR 30 at 41-42, 48-49, 60, 76 and 85. 67 In Falconer Gaudron J said (at 85): In general terms, a recurring state which involves some abnormality will indicate a mind that is diseased or infirm, but the fundamental distinction is necessarily between those mental states which, although resulting in abnormal behaviour, are or may be experienced by normal persons (as, for example and relevant to the issue of involuntariness, a state of mind resulting from a blow to the head) and those which are never experienced by or encountered in normal persons. ... And in Radford, King C.J. (75) distinguished between "an underlying pathological infirmity of the mind, ... which can be properly termed mental illness" and "the reaction of a healthy mind to extraordinary external stimuli". 68 As Deane and Dawson JJ make clear in Falconer in the passage the trial judge set out at para [14] of his judgment, there may be cases in which the evidence allows alternative contentions, that an accused's acts were involuntary either by reason of mental disease or natural mental infirmity on the one hand, or by reason of the operation of a defence upon a normal mind.
4. Given what is said to be an underlying predisposition on the part of the accused to zolpidem-induced parasomnia, the question naturally arises as to whether there is an "underlying pathological infirmity of the mind", such that any automatism is more appropriately characterised as insane, rather than sane, automatism. However, as I have already noted, this question has not been significantly explored, and nor is it strongly contended by the prosecution that this is a case which, on the balance of probabilities, is of insane rather than sane automatism. In an appropriate case it may be so, but here there is insufficient material to establish insanity.

Have the prosecution proved the guilt of the accused?

1. Having dealt with the preliminary issues, I turn, finally, to the substantive question. Have the prosecution proved beyond reasonable doubt that the accused's driving was voluntary? Put another way, have the prosecution removed the reasonable possibility that the accused was driving in a parasomnic state? The answer to these questions turns ultimately on a number of factual matters.
2. As I have already noted, Dr Moses claims that in certain cases, stilnox may cause parasomnia, and people can and do undertake complex tasks within that state. If they do, they are not exercising free and voluntary choice, since they are fundamentally asleep. A person in such a state can properly be regarded as an automaton as that term is used in the authorities.
3. Dr Perl, on behalf of the prosecution, has expressed some well-founded and legitimate skepticism as to whether stilnox-induced parasomnia is a real phenomenon, but in my view she stops short of asserting that it does not exist.
4. Taking those two opinions together, I am of the view that it is at least a reasonable possibility that the phenomenon exists. I would venture no higher than that, but nor do I need to for present purposes.
5. Dr Moses expresses the opinion, based on matters told to her by the accused, that there is a high probability the accused was indeed in such a parasomnic state, brought about by taking stilnox. There is no doubt that he took stilnox, it was in his blood. But the key facts underpinning her opinion, most of which come from the accused, include these: 1. He had no memory of driving, and claims that he remembers nothing from when he went to bed until he woke up the next morning in hospital. 2. He claims that he had walked home from the hotel and had gone to bed and to sleep - he therefore had absolutely no reason to drive. 3. Immediately before the collision, the accused did not swerve, brake or take any evasive action. 4. Witnesses who saw him at the scene of the collision described him as slurring his words, incoherent and not speaking in complete sentences. 5. He had on at least three previous occasions exhibited activity consistent with stilnox-induced parasomnia, indicating that he had an idiosyncratic predisposition to the phenomenon. 6. As well as consuming stilnox and alcohol, the accused took a central nervous system (CNS) depressant, epilim, which is believed to increase the risk of parasomnia.
6. These facts underpin Dr moses' opinion and are essential to it. Except in respect of matters 3 and 4, they are founded on the truthfulness of the accused's self-reporting. The prosecution submit that I should reject his evidence about these critical facts, and if I do, then the whole basis of Dr Moses' opinion falls away. I agree that that would be the consequence of completely rejecting the accused's evidence.
7. Speaking frankly, like the prosecution I am deeply skeptical about the accused's evidence. I certainly would not accept it on the balance of probabilities. Besides the inherent implausibility of it, I note the following:

• There is something of a convenient coincidence in the fact that the only documented instance of the accused potentially being in a parasomnic state coincides with him being charged with a criminal offence.
• There is an implicit unlikelihood in the notion that he could perform such a complex action not merely of driving, but of driving for perhaps more than 2kms before crashing.
• It is at least highly surprising that even in a parasomnic state, despite having already gone to bed, he had the presence of mind to take his wallet with him.
• Dr Moses said that those in a parasomnic state are capable of performing complex actions that are familiar and reflexive, but the accused's car was out of registration and had been for some time. Assuming he was acting lawfully, it is unlikely that he had driven at all in quite some time.
• At the time of the collision, the accused was driving towards his home rather than away from it - plainly, he had gone to an undisclosed end-point in his journey, then turned around and started to return home.
• The driving occurred about 4 hours after taking the stilnox. Dr Perl gave evidence about the half-life of stilnox and expressed the opinion, which does not appear controversial, that most of the effects of the zolpidem would have been concentrated much earlier in time.

1. There are other unsatisfactory aspects of the accused's evidence, including the following: 1. The accused denies any knowledge of any risks or warnings associated with the use of stilnox, despite a number of high profile media campaigns over a number of years, and despite there being stringent legal obligations on those who prescribed and dispensed the stilnox to him over 10 years. It is difficult to accept, in particular, that so many health professionals could have been so derelict in their duties for so long as to not provide him with those warnings. 2. The accused claims he had 3 previous incidents suggestive of parasomnia which he never followed up on, including one which apparently resulted in a significant injury to him. It is difficult to accept that although he is obviously intelligent, yet he made no further inquiries, and took the matters no further. Even when he was told by a doctor that it might have been the stilnox, he claims that he asked no further questions and made no further investigations. 3. It is difficult to accept that Dr Pickles, one of his treating physicians, had not taken the matter any further after the report to her of an episode of possible stilnox-related parasomnia, and, despite the extant warnings, had not taken him off it or made further investigation at all. 4. It is deeply implausible that despite allegedly having suffered harm as a result of the lack of proper medical oversight, including during the collision that has led to these charges, he continues to go to the same medical practitioners even to this day. 5. It is difficult to accept that even when stilnox was freshly prescribed to him by his GP in 2014, after a long period during which he did not use that drug, still he was never given any warnings about it, and no investigations were made or questions asked about the potential for adverse reactions. 6. The evidence the accused gave about his involvement with Camp Quality, in particular his explanation for why the nature of that evidence changed after being served with the statement of Peta Kennedy, was less than satisfactory. 7. His explanation for why his car was parked on the street in the way that it was, despite being so far out of registration, was also less than satisfactory. 8. It is noteworthy that there is nothing in the medical notes that were tendered which indicated a loss of memory, even though such a thing would plainly be medically significant.
2. However, having noted those matters, on the other hand I note: 1. At 0.110, his level of intoxication was certainly not so high that one would ordinarily expect a person to have a significant amount of amnesia; nor is it so high as to make likely a single-car, low speed collision with a stationary object on a relatively straight piece of road, such as the accused had. 2. There is, as the defence have pointed out, no evidence to contradict the accused's sworn claim that he had no memory of the driving. There is for example nothing in the medical notes that have been tendered. As I have noted elsewhere, it was open to the prosecution to obtain all medical records relating to the accused's hospitalisation, as well as to his transport in the ambulance, but that was not done. 3. Given that the accused could and did walk to a licensed premises earlier in the night, it seems largely inexplicable that he would then be voluntarily out in his car at 3.00 in the morning, after having gone to bed. Although there are some suggestions in the brief that he was in possession of a receipt from the Dee Why RSL club, that receipt was not put into evidence, and there is absolutely no evidence before me to suggest that he was at that club at any time relevant to the issues before the Court.

---

DISCLAIMER - Every effort has been made to comply with suppression orders or statutory provisions prohibiting publication that may apply to this judgment or decision. The onus remains on any person using material in the judgment or decision to ensure that the intended use of that material does not breach any such order or provision. Further enquiries may be directed to the Registry of the Court or Tribunal in which it was generated. Decision last updated: 02 December 2015