Riley v Seip [2008] ACTSC 72

3. The appellant was present at the accident scene. The collision had been reported to police at 11.37 pm and police attended at 11.47 pm. The appellant was subjected to a roadside screening test. It proved positive.

4. In consequence, the appellant was taken to Tuggeranong Police Station and required to provide a sample of breath for analysis.

5. The sample was purportedly analysed by a Drager Alcotest 7110 instrument. A result of 0.067g/100ml was recorded. According to a certificate pursuant to s 41(1)(B) of the Road Transport (Alcohol and Drugs) Act 1977 (ACT) (RT (A & D) Act) the sample of breath was provided at 00:30 hours on 20 August 2004 and the analysis was completed at 00:31 hours on that day.

6. The respondent, Constable Rab Seip, noted that during the period of time he had the appellant under observation the appellant's "face was flushed, he had watery eyes, his speech was very confused, and that he made a 'wheezing' noise when breathing". Another officer had observed a "slight odour of intoxicating liquor" on the appellant.

7. The appellant had been involved in a single vehicle collision. There was no evidence that these symptoms led to any conclusion as to the appellant's state of intoxication. Nor, as his Honour commented, was there any adverse inference capable of being drawn from the fact that the accident had occurred. The appellant acknowledged falling asleep at the wheel.

8. The appellant pleaded not guilty and gave notice, through his lawyer, that he intended to challenge the accuracy of the analysis.

9. The prosecutor, Ms Caffery, called an expert witness on that issue. The other facts were, by consent, proved by the tender of witness statements.

10. Mr John Farrar was the qualified expert called by the prosecutor. He explained the operation of the Drager breath analysing instrument. He also inspected and interpreted the records of the calibration on 19 June 2004 of the instrument in question. They indicated to him that the machine was then reading slightly low.

11. The machine contained two measuring systems, one infrared, the other electrochemical. The intent of those two systems was that if one of them "drifted" too far out of synchronisation with the other in respect of a sample of breath, then a malfunction would be signalled and the analysis aborted. He could see no evidence on the records of the test performed on the appellant and the subsequent calibration test to indicate that the result of the test should have been any less than 0.067g/100ml.

12. He was referred to a technique for calculating blood alcohol levels based on the consumption of alcohol over time in a person of a certain weight and sex referred to as the Widmark calculation. He noted that, as a biological assumption had to be made, the results of such a calculation could only be an approximation. Even so, of course the result of that calculation may be so far different from the breath analysis as to be inconsistent with it. Also, as Mr Farrar pointed out, the Widmark blood alcohol calculation is highly dependant, not only on the physiology of the individual which is reasonably objectively ascertainable, but, more significantly, on the reported consumption of intoxicating liquor, both as to quantity and timing.

13. Mr Farrar agreed, however, that, if the recorded result of the breath analysis was correct, the apparent assumptions as to the pattern of consumption made by the appellant's expert produced a conclusion inconsistent with that result. The two results could be reconciled only if the quantity of alcohol consumed over the relevant period had been significantly underestimated by the appellant.

14. Whilst Mr Farrar agreed with Mr Gill, for the appellant, that the Drager instrument needed to be recalibrated every six months, the specified anticipated drift was, he said, about 0.003 and its inherent tolerance was also 0.003. He was taken to the mechanical assumption inherent in the instrument, that the ratio of alcohol in the breath to alcohol in blood, expressed as a proportion of weight to volume, was 2100:1. He also agreed that that ratio was regarded as less than the "usual ratio" for the population as a whole. That was 2300:1 so that, usually, the result returned would be a modest understatement of slightly less than 10%. Conversely, of course, if the individual had a ratio of less than 2100:1, the result returned by the machine would be similarly over-stated, though, mathematically, the disparity is greater for a lesser ratio than a greater one of the same numerical order, for example, +200:1, produces a lesser numerical understatement than, say, -200:1 produces as an overstatement.

15. It also appeared that if the infrared analysis differed from the electrochemical analysis by more than 10% then the machine would report an error and abort the analysis. For the result of 0.067 to be an overstatement of an actual result below .05 the overstatement would have to have been approximately 30%. Again, that is merely a mathematical extrapolation.

16. If the subject had a breath content that included a non-alcohol contaminant, diethyl ether was offered as an example, it would affect the electro-chemical test but not the infrared test and a disparity error would be generated ("Error 31").

17. Mr Farrar did concede, however, that if in the field there was an undetected disparity between the two sensors drifting in the same direction a disparity of plus 20% between them might go undetected, though it would be "an extreme scenario".

18. Whilst he had never observed a machine drift into error to that extent and then drift back to conformity between calibrations, Mr Farrar fairly conceded that one could not know if that occurred but considered that it would be "quite rare".

19. Constable Michael Templeman gave evidence of the servicing of the instrument. It was calibrated on 19 June 2004. Again, on 5 October 2004, it was sent in for service, for consistently, though intermittently, displaying "Error 31". It was then reading low. That is, it had drifted into greater than 10% disparity between the two sensor systems.

20. There was some discussion before his Honour as to whether apparent signs of intoxication were relevant. It is true that apparent effects say little about the level of blood alcohol, particularly as between .067 and less than .05. However, a gross disparity between the signs and symptoms and a claim to have drunk minimal alcohol would tend to support the accuracy of an analysis displaying a relatively high reading. I consider his Honour was correct to have regard to that evidence, though as his Honour conceded, the weight to be accorded to it would not have been great. The circumstances of an accident, if produced by the aberrant driving of the accused person, would equally be relevant though, in this case, the weight of that circumstance could only have been slight.

21. The primary dispute centred around the evidence of alcohol consumption by the appellant and the consequences of that evidence, if not rejected, upon the asserted analytical result.

22. Professor Starmer, a consultant pharmacologist, gave a report to defence lawyers dated 15 March 2005. His expertise was both relevant and unquestioned. His evidence concerning the Drager Alcotest 7110 instrument was based on the assumption that the appellant had consumed:

... two [2] bottles [375ml] of Hahn Light beer with dinner between 8.00 pm and 9.30 pm and one [1] bottle [375ml] of Hahn Light beer at 10 pm.

23. He assumed that the appellant was a 50 year old man, 5'6" tall and of stocky/overweight build.

24. Given those assumptions, Professor Starmer was of the opinion that the appellant, had his consumption been as he suggested, should have had a blood alcohol level at the time of the actual analysis of not more than .002g/100ml, that is, virtually nil.

25. In contrast, if the breath test result was accurate, the appellant would have had to have consumed, over the same time frame, approximately nine [9.1] bottles of Hahn Light or five bottles of Hahn Premium [5.1], each of a 375 ml capacity to have produced that result.

26. I note that Constable Heazlewood, who accompanied the respondent and the appellant back to the police station, also observed that there was a clear smell of intoxicating liquor emanating from the appellant.

27. That observation was similar to that recorded by Constables Fulton and Seip. The appellant's wife and son were also present at the accident scene. Neither of them gave evidence before his Honour.

28. The appellant did give evidence before his Honour. He said he had attended, at between 7.30 pm and 8.00 pm on 19 August 2004, a work production meeting. It was held at Ruby's Restaurant, Dickson, ACT.

29. He deposed that, with a mixed Chinese meal, he had consumed Hahn Light beer only. He did that following a drink-driving conviction in 2001 which had motivated him to take care with alcohol consumption. He recalled consuming three Hahn Light beers in 'stubby' bottles. That was what he told his solicitor Mr Saeedi a few days later. He was confident no more than that had been consumed by him before the accident.

30. There were about 11 co-workers at the dinner. The bill was paid by the general manager. No evidence was called from any of those persons nor was a copy of the restaurant account tendered.

31. A copy of Mr Saeedi's file note was tendered. It recorded the version of events at the dinner and personal details given in evidence by the appellant which were later relayed to Professor Starmer as assumptions to be made to found his opinions.

32. Character evidence was also called. Mr John Smith, who had known the appellant for over 10 years and worked with him, attested to his honesty. Mr Bruce Selleck had worked with the appellant for more than five years and also attested to his reputation for honesty.

33. In his submissions to his Honour, Mr Gill relied upon Perkins v Pohla-Murray (1983) 51 ACTR 3 (per Kelly J); Harrington v Zaal (1992) 106 FLR 175 (per Higgins J); Looper v Forbes (1992) 112 ACTR 29; 111 FLR 449 (per Miles CJ) and Green v Tongs (2005) ACTSC 7 (per Higgins CJ).

34. The underlying principle is not in dispute. That was acknowledged as early as 1970 in Saxe v Kellett [1970] VicRp 79; [1970] VR 600, 602 per Anderson J. Each of the above decisions acknowledged the same principle.

35. That is, that the prosecution must prove that the accused person had driven a motor vehicle and during a specified period had present in his blood the prescribed concentration of alcohol. The prescribed concentration for present purposes was between .05 and .08grams/100 millilitres of blood.

36. The certificate evidencing the administration of a breath test, as provided for by the RT (A & D) Act, provides prima facie evidence of the presence of the concentration so recorded. It also evidences that the machine producing the result was operating and was operated correctly. There is an evidentiary assumption that the machine will return an accurate result.

37. That evidentiary assumption may be displaced by other evidence which casts reasonable doubt upon it.

38. In Perkins v Pohla-Murray (supra) that doubt was founded on evidence, accepted by the learned Magistrate, as at least a reasonable hypothesis, that the appellant had consumed an insufficient quantity of intoxicating liquor to have produced a prescribed concentration. On appeal Kelly J found that it had been open to the learned Magistrate so to conclude.

39. In Harrington v Zaal (supra) the doubt arose from the result of a blood test undertaken by the defendant upon release from custody about one and a half hours after the breath analysis. There was some evidence from a person who had been with the defendant earlier in the day as to his alcohol consumption. That evidence was not disputed. There was also expert evidence that the result, as displayed by the instrument, was inconsistent both with the consumption attested to by the defendant and his corroborating witness, and with the blood test result. Thus, I found, on appeal, that there was, on those facts, a reasonable doubt as to the accuracy of the breath analysis result.

40. In Looper v Forbes (supra), Miles CJ considered an appeal from a finding by a Magistrate that the appellant was guilty of an offence against s 19 of this Act (then entitled Motor Traffic (Alcohol and Drugs) Act). The result, as shown by the instrument, was 0.200g/100ml of blood. The appellant was observed to be "moderately" affected by intoxicating liquor. He had admitted to consuming a large quantity of intoxicating liquor the previous evening. Dr Frank Slater gave expert evidence which seems to have been directed to raising a doubt, not as to the result of the breath test, but rather, as to the consumption history. There was, in his Honour's view, no error in the Magistrate accepting the breath test result as accurate. Nor was he obliged to accept the testimony of the appellant and his mother as to the quantity of drink he had consumed which, if accurate, would have challenged the accuracy of the test result.

41. That is not to assume that, once an adverse result is obtained by breath analysis, the subject of it bears any legal onus of proof to displace it. However, it does cast an evidential burden upon that subject to cast reasonable doubt upon the result so obtained. That evidential burden had not been discharged (see s 58(7) Criminal Code 2002 (ACT) for the meaning of that expression).

42. Green v Tongs (supra) was an entirely different case. The learned Magistrate in that case had evidence of alcohol consumption from the respondent. Unless that evidence was rejected as untrue, the result returned by the instrument was inconsistent with it. The evidence he gave concerning his alcohol consumption was consistent with the minimal signs of intoxication apparent to observing police. The only issue was whether his Worship (as then he was) was able, without expert evidence, to conclude that the result was inconsistent with the history of consumption deposed to by the respondent. I held that it was open to his Worship to have accepted the respondent's unchallenged evidence as to consumption and, as a matter of judicial notice (see s 144 Evidence Act 1998 (Cth)), to have concluded that the result was markedly inconsistent with the result given by the instrument so that he could entertain, as he did, a reasonable doubt as to its accuracy.

43. It is essentially a factual question. Mr Gill did refer his Honour to the evidence of the appellant's alcohol consumption which, if accepted, would, on the expert evidence of Professor Starmer, be inconsistent with the breath analysis result. He also referred to evidence of an error being displayed some time after the test in question indicating disparity between the analysing systems within the instrument. That raised, he submitted, a reasonable possibility that the two systems had drifted in the same direction so as to produce an exaggerated result undetected by the electro-chemical cell. It was further submitted that the appellant might have had an extraordinarily low breath to blood ratio or that, in any event, the appellant had made an honest and reasonable mistake as to the extent of his alcohol consumption.

44. Neither of the latter two submissions were urged upon me on this appeal nor is there any evidence upon which either could be supported.

45. The prosecutor, Ms Caffery, accepted the general principles referred to above but pointed out that there was no indication at the time of the breath analysis that the machine was faulty. The systems contained in it were designed to preclude undetected error.

46. Further, there was evidence of observations by police that some signs of intoxication were present. Though slight, they were not inconsistent with the analysis by the instrument.

47. She further submitted that, whilst there was evidence from the appellant of relatively modest consumption of light beer, it was not inconsistent with the appellant having lost count of his alcohol consumption or of some other error in respect of it.

48. On 16 November 2007, his Honour delivered a written decision. His statement of principle was, I consider, unexceptionable. His Honour said at [4]:

Doubt as to the proper operation of the breath analysing instrument may arise in a number of ways. There may be direct evidence of some defect in the instrument, or there may be evidence from which it may be inferred that the instrument was not functioning properly.

49. His Honour did not consider that the "Error 31" message subsequently intermittently appearing cast doubt on the proper operation of the instrument on the occasions when it did not appear.

50. More significant, of course, was the evidence of the appellant as to his drinking pattern in the light of the expert evidence concerning it. As his Honour correctly noted, the issue is as to the weight to be accorded to such evidence. It is not for the defendant to adduce such evidence. A defendant bears no onus of proof or of persuasion, but the breath test result, particularly in the context of evidence positively supporting the scientific accuracy of the instrument itself, may well suggest that such evidence must be inaccurate or misleading whether by error or otherwise. It is the more difficult to reject such contrary evidence, the more it is corroborated. The appellant's evidence was not supported by any such corroborative evidence.

51. His Honour accepted that the police observations were also as consistent with the evidence of the appellant as with the test result. That result was, however, inconsistent with a blood alcohol level of 0.01g/100ml or thereabouts. That left two possible findings. One was that the evidence of the appellant could not be rejected with certainty, in which case, unlikely though it may be, the breath analysing instrument must have failed. The other was that the evidence of the appellant was not reliable, due to either deliberate falsehood or faulty recollection, and could, therefore, be rejected with certainty as lacking accuracy.

52. His Honour was satisfied that the latter alternative was the case and found the offence proved.

53. This was not a case in which the competing hypotheses were equally well supported, as was the case in Harrington v Zaal. Nor was there any corroboration of the appellant's drinking pattern, as was the case in Perkins v Pohla-Murray. Whether a reasonable doubt is raised is a question of fact, which is within the power of the decision-maker to determine. It may be that another decision-maker might have been positively persuaded by the appellant's evidence, even if uncorroborated, sufficiently to entertain a reasonable doubt.

54. His Honour appropriately gave weight to the appellant's good character. He did not find a deliberate deception on the part of the appellant. He found that, for whatever reason, though he was otherwise convinced of it by the time he consulted Mr Saeedi, the appellant had consumed more alcoholic beverages than he recalled. It was unnecessary for his Honour to consider whether the appellant would have had a defence had he reasonably believed that he had not consumed enough alcoholic beverages to risk exceeding 0.05g/100ml. The offence under s 19 (RT (A & D) Act) was created before the enactment of the Criminal Code 2002 (ACT) (the Code) thus there is no specific enactment classifying it as an offence to which s 36 of the Code is, or is not available as a defence (honest and reasonable mistake of fact). That is not to say that such a belief would not be relevant to penalty, even if not as to criminal responsibility.

55. It may have strengthened the appellant's case to have produced a copy of the restaurant account. On the other hand, it may have revealed he had made a mistake in his recollection. So also if fellow diners had been interviewed soon after the event, it might have revealed whether light beer was in fact served or, whether, in error standard beer was served.

56. In the absence of such evidence his Honour was, in my opinion, not in error in finding that the appellant's evidence had not been accurate. Nor could an honest and reasonable belief, even if available as a defence, have been supported by an erroneous belief as to the extent of alcohol consumption. I leave aside for another case the innocent ingestion of alcohol introduced by the act of a third party. It follows that the appeal must be dismissed.

57. The orders made by the learned Magistrate are confirmed save that the disqualification and time for payment will date from today.

I certify that the preceding fifty-seven (57) numbered paragraphs are a true copy of the Reasons for Judgment herein of his Honour, Chief Justice Higgins.

Solicitor for the respondent: Director of Public Prosecutions for the ACT