---

FILING SHEET FOR EASTERN CAPE JUDGMENT

ECJ no :

PARTIES:

TERENCE NEWBURY PRICE PLAINTIFF

AND

MUTUAL & FEDERAL INSURANCE CO. LTD DEFENDANT

REFERENCE NUMBERS -

• Registrar: CASE NO: 2052 / 2003

• Magistrate:

• Supreme Court of Appeal/Constitutional Court:

DATE HEARD:

DATE DELIVERED: 6/09/05

JUDGE(S): SANGONI J

LEGAL REPRESENTATIVES -

Appearances

:

• for the State/Applicant(s)/Appellant(s): Adv PJ de Bruyn SC & Adv B J

Pienaar

• for the accused/respondent(s): Adv S H Cole

Instructing attorneys:

• Applicant(s)/Appellant(s): Friedman Scheckter

• Respondent(s): Joubert Galpin Searle

CASE INFORMATION -

• Nature of proceedings : Civil Action

• Topic:

• Keywords:

IN THE HIGH COURT OF SOUTH AFRICA

(SOUTH EASTERN CAPE LOCAL DIVISION)

CASE NO: 2052/2003

In the matter between:

TERENCE NEWBURY PRICE PLAINTIFF

and

MUTUAL & FEDERAL INSURANCE CO. LTD DEFENDANT

JUDGMENT

SANGONI J:

1. In this matter the plaintiff who is practising law as an advocate has instituted an action against the defendant claiming compensation for damages to his motor vehicle, arising from a collision with another motor vehicle on 25 April 2003.

2. The defendant is a registered insurance company. At the time of the collision there was a valid policy of insurance issued by the defendant in favour of the plaintiff providing, inter alia, a cover for damages incurred as a consequence of a motor vehicle accident. This cover is subject to certain terms and conditions.

3. One such condition as contained in the special exclusions clause, reads:

“We will not be liable for loss, damage, injury or liability caused, sustained or incurred;

3.1 …

3.2 while the vehicle is being driven or used

3.2.1 …

3.2.2 …

3.2.3 by you while under the influence of intoxicating liquor or drugs or while the concentration of alcohol in your blood exceeds the statutory limit;”

4. Defendant’s case is that at the time of the collision the plaintiff was driving his motor vehicle whilst under the influence of intoxicating liquor or drugs, alternatively, whilst the concentration of alcohol in his blood exceeded the statutory limit. It is accepted by the parties that once that is proved the defendant will not be liable for damages. They agreed that the questions relative to merits and quantum be separated and I then so directed.

5. They further agree that the only issue to be decided is whether the plaintiff did or did not violate the condition set out in paragraph 3 above.

6. The collision that gave rise to this action, between the plaintiff’s vehicle and a motor vehicle driven at the time by one Mr Botha, occurred at a robot-controlled intersection. The question of who was to blame for the collision is by agreement of the parties not to be considered for purposes of the issue to be decided.

7. At the scene of the accident both the plaintiff and Mr Botha were breathalysed, one after the other, by means of the same instrument, a Lion Alcohol Meter Model S-D2 bearing number 69/255 (Lion S-D2 for short). There is no dispute that the results for Mr Botha were a breath alcohol reading of 0,47 milligrams per litre (“mg/1”) and 0,44 mg/1 as regards the plaintiff. For completeness I should mention that the plaintiff maintains that such readings were not shown to him and it is on this basis that he does not dispute them.

8. Blood samples were later taken from both drivers for purposes of analysis by means of Gas Chromatography, so as to establish blood alcohol concentration (BAC). The sample of blood extracted from Mr Botha was so analysed and it is not in dispute that the Chief Forensic Analyst in the employ of the State found that the concentration in his blood sample was 0.11 grams per 100 millilitres. Plaintiff’s blood sample could not be analysed as the Chief Forensic Analyst found, as stated in his affidavit, that the blood sample taken leaked out of the glass sample holder and dried.

9. The defendant bears the onus, on a balance of probabilities, to prove that the plaintiff drove his vehicle whilst under the influence of intoxicating liquor or drugs, alternatively, whilst the concentration of alcohol in his blood exceeded the statutory limit.

10. It is trite law that a driver is under the influence of liquor when “the skill and judgment normally required of a driver in the manipulation of a vehicle were diminished or impaired as a result of the consumption of the intoxicating liquor.”

(Motor Law by W.E. Cooper Volume One Page 554 and authorities referred to therein)

11. It is section 65 (2) of the National Road Traffic Act 93 of 1996, that prescribes the limit in blood alcohol content. It reads:

“(2) No person shall on a public road –

(a) drive a vehicle; or

(b) occupy the driver’s seat of a motor vehicle the engine of which is running, while the concentration of alcohol in any specimen of blood taken from any part of his or her body is not less than 0.05 gram per 100 millilitres, or in the case of a professional driver referred to in section 32, not less than 0,02 gram per 100 millilitres.”

12. Defendant’s case is that the concentration of alcohol in the breath of the plaintiff was 0,44 mg/l when measured by the breathalyser and that therefore, a conversion to blood alcohol concentration would be in excess of the statutory blood alcohol concentration of 0,05 gram per 100 millilitres. As will be dealt with later, the basis for the conversion, according to experts’ opinion, is the scientific assumption that the concentration in the breath is directly proportional to the concentration in the blood. In terms of what is referred to as Henry’s law, at constant temperature the concentration of gas dissolved in a liquid is proportional to its concentration in the air directly above the liquid. Applied to the concentration of alcohol in the breath and in the blood this means the concentration in the expired alveolar air is taken as directly proportional to the concentration in the blood.¹

13. The test of alcohol in a sample of breath is known in our law. Section 65 (5) of the National Road Traffic Act 93 of 1996 provides:

“(5) No person shall on a public road –

(a) drive a vehicle; or

(b) occupy the driver’s seat of a motor vehicle the engine of which is running, while the concentration of alcohol in any specimen of breath exhaled by such person is not less than 0,24 milligrams per 1 000 millilitres, or in the case of a professional driver referred to in section 32, not less than 0,10 milligrams per 1 000 millilitres.”

14. In terms of subsection (7) a prescribed equipment is required to measure the concentration of alcohol in the breath for purposes of subsection (5) referred to above. Regulation 332 to the National Road Traffic Act provides that such equipment shall be evidential breath testing equipment complying with the requirements of the South African Bureau of Standards in terms of SABS 1793. The equipment comprises the following:

(a) Drager Alcotest 7110 MKIII part number 8314647 (Germany);

(b) Drager Alcotest 7110 MKIV part number 35307791 Australia); and

(c) Intoxilyzer 1-400 (South African version – Intoxilizer 5000 P-SA).

Needless to say that the Lion S-D2 which is the instrument used in this case is not one of those.

15. Before dealing with the scientific evidence by experts, regarding the breath tests, the proportionality and correlation between breath and alcohol tests and related issues, I should first deal with the non-scientific evidence especially the evidence relating to the happenings after the collision on the day in question.

16. In this regard defendant called three witnesses, all traffic officers who attended at the scene after the collision.

17. The first to come to the scene was Traffic Officer Ludicke at 19h09. According to her, when she arrived at the scene she found the plaintiff. She knew him before that day. She then elicited information from him as to how the accident occurred. Mr Botha, the driver of the other vehicle came up to the scene later. He also supplied information relative to the accident.

18. She testified that the plaintiff requested her to conduct a breathalyser test on Mr Botha and her response was that she would test both drivers. It is standard, she said, that if they test one they should test the other. She observed that Mr Botha was under the influence of liquor but the plaintiff, according to her, was normal. Both drivers were rude to each other, one blaming the other for the accident. Mr Botha reported to her that he earlier had to leave the scene as the plaintiff attacked him with a fist. While conversing with the plaintiff, the plaintiff told her that he had had a few drinks but that he had hit Mr Botha with a fist out of shock.

19. She testified further that she alco-tested Mr Botha by means of a Lion Alcohol Meter Model S-D2 breathalyzer instrument which had been calibrated on 14 February 2003. As a matter of policy, such machines were calibrated every three to six months. The person who calibrated the machine according to her, was Senior Traffic Officer Gallant. From her point of view, it was working properly. The result of the test was shown to the other traffic officers as well as to the drivers by her.

20. She took the measurements on the scene, drawing up the sketch plan and key and also making a report. The time of the accident, as reflected in the report, was 18h50. According to her she got that information from the plaintiff. Mr Botha confirmed that the accident occurred before 19h00.

21. The witness then asked Senior Traffic Officer Camagu to test the plaintiff by means of the same machine. A different mouthpiece was used. She did not witness that test as she subsequently left the scene to take Mr Botha to hospital for purposes of taking a blood sample from him. That was done in her presence; the blood sample was sealed in a container in the prescribed manner.

22. According to her she followed the current procedures when breath testing Mr Botha and everything was done correctly when a sample of his blood was taken. She estimated the lapse of time at about four minutes between the time Mr Botha was breathalysed and the time the plaintiff was breathalysed.

23. To a question under cross-examination as to whether she had any scientific knowledge as to how the apparatus used works and how the results were achieved, she said the training she received dealt with how to use the machine. She was not in possession of any written instructions that should be followed. She was trained in 1999 at the Traffic College at Korsten. She insisted that she told the plaintiff that he would also be breathalysed not as an attempt to appease Mr Botha but because the policy stipulated that if you test one driver the other should be tested as well.

24. Traffic Officer Simakuhle also gave evidence. According to him he got to the scene while the witness Ludicke was already there. He observed that the plaintiff was well dressed and when he went closer to him he perceived some smell of liquor. At a later stage while travelling with the plaintiff from the police station to the hospital she noticed that the plaintiff’s eyes appeared ‘reddish’. Traffic Officer Ludicke instructed him to keep an eye on the two drivers, which he did. He witnessed the breathalysing on both Mr Botha and the plaintiff. Both of them were shown the results, according to him.

25. He testified that the plaintiff was asked to blow once on the machine, not three or four times as was suggested. When it became known that the reading was beyond the limit he explained to the plaintiff what his rights were. He testified that Traffic Officer Camagu asked the plaintiff whether in the past 24 hours that had gone by the plaintiff did drink and the answer was in the affirmative.

26. It was put to the witness by Mr De Bruyn SC, representing the plaintiff, that one Mr Griebenow a friend of the plaintiff was on the scene when the breath tests were done. The witness answered that he only observed him at the police station.

27. Simakuhle stated that he is the officer who arrested plaintiff at the scene, not for driving whilst under the influence of liquor but for having exceeded the necessary limit as shown by the breath test.

28. It was put to the witness that he gave evidence as to how a breathalyser test should be normally done not how the tests were done. This seems to acknowledge that if they happened the way the witness related, they were done correctly.

29. Senior Traffic Officer Camagu confirmed that he conducted the breathalyser test on the plaintiff. He substantially corroborated the evidence of Simakuhle and Ludicke as to how these tests were conducted. He confirmed that the plaintiff did say he consumed liquor in the 24 hours that had gone by.

30. According to him the traffic officers, including himself, received training on how to use the breathalyser machines on regular basis. He denied that there were two attorneys present when the test on the plaintiff was conducted. He also denied that he called on the plaintiff to blow on the machine four times but stated it was only once.

31. The plaintiff did not testify. Mrs Price and attorney Griebenow however did so on his behalf.

32. The main thrust of their evidence was that they attended at the scene and that there was nothing they observed indicating that the plaintiff was under the influence of intoxicating liquor, no liquor smell on his breath and no red or reddish eyes as alleged by Traffic Officer Simakuhle. Mrs Price went further to say in her testimony that at about 4h30 p.m. on that day the plaintiff was at home and there was nothing pointing at the plaintiff as being under the influence of liquor. In fact, she said she requested the plaintiff to take their son Dominique to another area to fetch his girlfriend. In his testimony Mr Griebenow stated that he was present when the defendant was being breathalysed, which was denied by the traffic officers. The plaintiff was called upon to blow on the machine four times in terms of his testimony.

33. The critical point in their evidence is whether there were any clinical signs indicating that the plaintiff was under the influence of liquor. They were not able to point out any irregularities in the testing process. Mr Griebenow testified that he was told some days after the test, that it was irregular for the plaintiff to have been called upon to blow four times. No further evidence came up to substantiate how irregular that was. The expert witness, Dr Williams, stated that blowing four times was not necessarily irregular. I should mention that in the letter of demand the plaintiff addressed to the defendant he refers to three times. From the reading of the letter I agree with Mr Cole’s interpretation that the plaintiff refers to three times.

34. Traffic Officer Simakuhle was the only one, out of the witnesses who had been to the scene, who perceived a smell of liquor on the breath of the plaintiff and who observed that his eyes were reddish. His evidence was thus not corroborated to that effect. It would be risky therefore to rely on his evidence in this regard.

35. Much as Mr Griebenow testified that he got to know and has been close to the plaintiff for the past twenty years, as well as the fact that Mrs Price knows the plaintiff very well, which I accept, none of them was able or did say that the plaintiff had not consumed liquor in the course of the afternoon of that day apart from the 500 ml beer that is admitted he took before 14h00. They were not in a position to deny that the plaintiff did say to the traffic officers he had had a few drinks or he had taken some drink.

36. In light of the foregoing I accept that the instrument was correctly operated according to protocol. The challenge to the manner the instrument was operated was not far- reaching. I also proceed on the basis that there were no clinical signs indicating that the plaintiff was under the influence of intoxicating liquor at the time he was at the scene or at any time thereafter.

Accuracy and Reliability of the Machine

37. The issue here is whether the Lion Alcohol Meter, Model S-D2 is accurate and reliable. That is an enquiry in the context of the machine as part of the generic class and whether or not the specific instrument was in good working order at the time of the breath tests on both drivers. As regards the first aspect the defendant relies on the evidence of two expert witnesses, Mr Evans² and Doctor Williams³ who is from the United Kingdom, for the second aspect largely on the testimony of Mr Gallant.

38. Before looking at what was done in an attempt to calibrate the measuring instrument in question I should state that I take guidance from the dictum of Van Dijkhorst J in S v Van der Sandt 1997 (2) SACR 116 (W), where it was stated:

“…the State has to prove that the measuring instrument gives the correct measurement. This entails that its operation be explained, that it is proved to be trustworthy in its operation and that its result is proved to be correct. This includes proof that it is properly calibrated to official measurements.”

39. Examining the evidence of the Senior Traffic Officer Gallant which deals with the calibration of the machine, I refer to his testimony with regard to what he did on 14 February 2003. His account goes like this:

“Now from the calibration process, how is it being done? …Now if a traffic officer is bringing his machine in, we have got a cylinder within the radio room then I would take a little pipe that you can use with your mouth, then I would connect it to the cylinder itself, then I will turn the cylinder to open and I also take this breathalyser and I also attach it to this pipe, little small pipe, then there is also a button on the cylinder, then I will press the button, and then the breathalyser must be one having a red light, then if the green light comes on then I will press a second button on the breathalyser and then it will give me a reading.

Now what reading do you expect and what do you do with the reading? …I am expecting to find a reading of 0,24.

If the reading is higher than that figure what do you do? … Now on the side of breathalyser there is a little ‘skroefie’ there, then I will take a screwdriver then I will turn the ‘skroefie’ to turn the reading to what it is suppose to be which is 0,24.

So can you amend that figure to go up and down? … Yes M’Lord.

Now if you get the correct figure on the breathalyser which is 0,24 do you do something to establish that figure is correct? … That is correct so M’Lord. I will then measure the breathalyser on the cylinder as I explained previously and I will follow the same method.” (My underlining)

40. To calibrate as defined in the Concise Oxford Dictionary, means to “correlate the readings of (an instrument) with those of a standard.” Many definitions on the Web are in agreement with this definition. One such goes like this:

“Calibration is correcting a measuring instrument by measuring values where true values are known.”⁴

It is common cause that the reason for proper, regular instrument calibration is to prevent concentration readings that are inaccurate and to ensure, therefore, the accuracy of the instrument. For proper calibration it is essential to follow the manufacturer’s guidelines and even as regards the frequency. One of the rules is that a certified calibration gas be used before the expiration date. The instrument reading is compared to the actual quantity of the gas present in the cylinder. It should be noted that no such guidelines were brought to the attention of the Court.

41. It is surprising that the witness was not able to say whether the cylinder he connected to the little pipe, in terms of his evidence, was a gas or an alcohol cylinder. It would be fair to conclude that the content of the cylinder would influence the result. He did not know what was blown into the machine or the cylinder. He had no knowledge, technical or otherwise as to how the cylinder or the machine works. If he gets the reading of 0,24 he would check its correctness by following the same process of connecting the machine to the cylinder by means of a pipe. That would mean, in my view, repeating the process of purported calibration to verify its correctness. His task aimed at achieving the correct reading was simply to turn the reading on the instrument up and down to the reading 0,24. Once done that would tell him that the instrument was properly calibrated. That raises the question as to why do it again when the figure 0,24 appears on the machine already. How many times should it be done in ascertaining its correctness?

42. In the Rule 36(9) (b) notice filed in respect of Mr Gallant he is described as a Senior Traffic Officer employed by the Traffic Department of the Mandela Metropolitan Municipality. During April 2003 he was a shift leader working from the radio control room at the traffic department. He was then responsible, amongst other things, for the calibration of the breathalyser units,. He said he was taught by a previous shift leader as to how to do the calibration. As a shift leader himself, it would be expected the information given is what would be passed on by him to whomever he were to teach on calibration. No details of the lessons he got were given. The information sounds simple and not necessarily scientific a process, it does not appear to involve any meaningful intellectual activity. The witness’s field of expertise remains unclear.

43. I understand Mr De Bruyn’s submission to be that there is no reliable evidence to prove that the instrument in question was properly, effectively and adequately calibrated. Mr Cole argued on behalf of the defendant that this approach amounted to litigation by ambush as it was not put to Mr Gallant that he did the calibration wrongly. I am of the view that the questions put to Messrs Gallant and Evans under cross-examination, though perhaps not overtly, sufficiently indicate that the correctness and effectiveness of the calibration process and therefore the accuracy of the machine were put in issue.

Mr Evans did not corroborate the calibration procedures set out by Mr Gallant. He testified that the recommendation of the manufacturer is that the instrument should be calibrated over a period of six months. He confirmed that the relevant instrument was tested by his technician for accuracy in June 2004. He also tested it. That was a normal check up, “simply to check the calibration of the machine”, but not to do anything to the machine. Apart from the fact that that was sixteen months after the February 2003 calibration, it does sound like the calibration or check up in June 2004 was aimed at the current proceedings. The outcome of such calibration was not disclosed except to say there was nothing wrong with the machine. Later, to say, the machine read 6% below the calibration gas value which is determined at so much distance from the sea level, the value being .29 mg per litre in Pretoria. He said nothing about Mr Gallant’s technical know how, the use of a cylinder containing either gas or alcohol or some other substance and the procedures followed by him at calibration.

44. I was referred to the case of S v Strydom 1978(4) SA 768 (ECD). This is a criminal case and obviously the standard of proof is proof beyond a reasonable doubt. It was held that the evidence of the expert witness in that case could not be accepted as proof that the analysis of the appellant’s blood by the instrument used had been accurate. Though the appellant had not given any evidence in his defence he had however in his plea statement stated that the analysis of the specimen had not given a reliable result. One of the points taken on his behalf was that there was paucity of evidence towards establishing the reliability of the analysis performed.

45. Stewart J, after examining the evidence of Van Riet, who had analysed the blood specimen, concluded that “Van Riet did not pretend to understand the process which the blood underwent in the machine …” and further that “it is apparent from his evidence that the working of the machine itself was not something which fell within his field of competence.”

46. On page 75 A-C Stewart J said the following:

“In the present case the machine itself is used to test its own accuracy. What is missing is proof that, if the machine gives a particular reading when tested with standard alcohol, it is reliable reading when analysing blood. The proof would be by way of expert technical evidence to that effect by a person able to describe the process in the machine and to vouch for its reliability, of by testing the machine against another unrelated method of analysis. The absence of such evidence, more particularly when there is reliable evidence that the result of the analysis of the appellant’s blood is out of keeping with the findings of the district surgeon on clinical examination and with the police constables observation, raises a doubt as to the reliability of the analysis.” (My underlining)

47. Mindful of the fact that the standard of proof is different in a civil case, I am, however, of the view that the concerns expressed in the Strydom case apply equally to the current case. The apparent lack of technical knowledge on the part of Mr Gallant casts doubt as to the correctness of the calibration, more particularly because there is evidence placed before the Court relating to the calibration records kept, showing the extent of variations between the figures this machine and other similar machines reflect at the start of the calibration process and the figures they have been calibrated to. The range of such variations surprised Mr Evans. Out of fifteen cases referred to the variable range is from 92% over to about 65% under the figure of 0,24 to which the machines were calibrated to. On 14 February 2003 the machine in question registered 33% under before calibration. No explanation was given for such figures. Mr Evans stated that they made no sense to him in view of the fact that they regarded the machines to be accurate. When asked as to whether there was something seriously wrong, he responded as follows:

“I don’t say that there is something seriously wrong with the machine, I am saying that there is something here which I just don’t understand, I am not seeing them using the machine, or anything like that.”

The written records of calibration kept by the municipality in respect of each instrument, help show, inter alia, which instruments are prone to erratic readings. The records referred to paint a very bleak picture.

48. As alluded to earlier the defendant relies on the evidence of two expert witnesses, Mr Evans and Dr Williams, in support of its contention that the relevant Lion S-D2 as one of a kind, is accurate and reliable. In my view what matters is whether the specific machine used was in good working order at the time and less important is its brand status. It is trite that with regard to evidential breath testers there are standards or specifications to be met that are laid down or prescribed within the relevant jurisdictions. Even with the EBTs one is dealing with a brand in that some brands are prescribed like the Drager Alcotest in the South African situation. Such specifications seek to eliminate the risk of a false result with regard to breath alcohol value. With such specifications and stringent requirements one in effect moves from the generality of a brand to the specificity of an individual instrument with a view to eliminating the risk of false or incorrect results.

49. As background information I refer to the article of A.W.Jones⁵ and K. A. Jonsson issued in the year 1993 following the 12^th International Conference on Alcohol and other related issues. Its title is DETERMINATION OF ETHANOL IN BREATH AND ESTIMATION OF BLOOD ALCOHOL CONCENTRATION WITH ALCOMETERS S-D2.

50. In their opening paragraph the authors had this to say:

“The Lion Alcolmeter S-D2 is a multipurpose hand-held device designed originally for use as a road side screening test. However, its easy-to-read digital display and highly reproducible breath sampling system, has created many applications for this instrument in alcohol research, industrial chemistry, and clinical and emergency medicine. Although the Lion Alcolmeter S-D2 has been available for many years, hardly any studies have dealt with its precision and its accuracy.”

51. The Lion S-D2 instrument was once manufactured in South Africa by a close corporation called Alcosafe CC. The witness Evans is a member of that corporation but now only involved in distribution. The parent company is Lion Laboratories Limited based in Wales in the United Kingdom. The nature of the relationship between the close corporation and this company has not been spelled out clearly. Dr Williams however, described Alcosafe CC as Lion’s exclusive distributor in South Africa. CMI is Lion Laboratories Limited sister company based in the United States of America. Both companies are wholly owned subsidiaries of an American company known as MPD. Lion Laboratories Limited uses Alcometer as its brand name whereas CMI’s brand name is Intoxilyser, hence the Alcolmeter S-D2 is the same instrument as in Intoxilyser S-D2, according to Dr Williams. This specific instrument that was used in conducting the breath tests on the plaintiff as well as on Mr Botha was manufactured by Alcosafe CC.

52. Both Mr Erasmus and Dr Williams described the Lion S-D2 instrument as a stable machine that is widely known for its accuracy. They have both agreed that this instrument is used in most countries as a screening device and that the various stringent specifications required of evidential testers do not necessarily apply in every case where the S-D2 machine is used. In South Africa, for instance, there are no such specifications applicable to a lion S-D2. In jurisdictions where the S-D2 machine is used as an EBT, for example the US Department of Transport, and North Dakota, examples given by Dr Williams, there are some standard requirements in place.

53. By way of an example, I was referred to Exhibit “Q” which represents a document⁶ issued by the office of the Attorney General in North Dakota. It deals with operating instructions for an intoxilyser S-D2 and other screening devices. In the article the S-D2 instrument is recognised as a screening device with the following rules laid down:

“The screening test or tests must be performed by an enforcement officer certified as a chemical test operator by the state toxicologist. The results of such screening test must be used only for determining whether or not a further test shall be given …”

54. The certified chemical test operator in terms of the article, refers to a person certified by the state toxicologist as qualified to perform analysis for alcohol in a person’s blood, breath, saliva or urine. It is also stated in the article that the instrument should be checked once a month and be calibrated as needed. Such stipulations lend support to Prof Saayman’s viewpoint that the Lion S-D2 instrument is not designed to provide definitive values as in South Africa, particularly, there are no procedural guidelines and technical requirements to safeguard that the machine is properly maintained, operated to ensure a correct result.

55. Dr Williams testified that the general view nowadays amongst the scientists, is that breath instruments whether they are so-called screening or so-called evidential are all of a similar accuracy and there is therefore no need for nomenclature distinctions. He however, stated that in most countries, including the United Kingdom, the S-D2 instrument is used as a screening device. It is clear however, that the distinction between the two categories is maintained. In a situation where the S-D2 is used as an EBT there are certain requirements specified including, in some cases, the minimum qualification or training of the operator of the instrument.

56. When asked under cross-examination why there was a need for other instruments if S-D2 was such a wonderful machine, his response was that the main reason is that the other machines, with their facilities with regard to print-outs with date and data, are geared for challenges, litigation and disputes that might arise to reinforce evidence in contested matters. This suggests that the Lion S-D2 instrument may not be an appropriate machine for such purposes.

57. What comes out clearly from the experts is that the Lion S-D2, as a screening device, is a high quality performer, recognised by scientists of note to be scientifically reliable, subject to the principle that, as a machine, it is fallible and its high quality performance is contingent on its proper maintenance and use by trained and knowledgeable operators. I can only accept the Lion S-D2 instrument as a high quality performer in that context. What is pertinently relevant, in my view, is whether the individual instrument used in this case has been shown to be capable of yielding the correct results.

Conversion of Breath Alcohol to Blood Alcohol

58. In the article – Knight’s Forensic Pathology published in 1996, authored by Pekka Saukto & Benard Knight, the following appears:

“Breath is now used by many jurisdictions to measure alcohol intake, either as a screening test before blood is taken for analysis, or as an evidential method instead of blood or urine. There is still some controversy over the scientific accuracy of this method, but usually the results are so high that errors are immaterial – or in marginal results, more accurate blood testing is indicated.”

59. It is not in dispute that there is a significant volume of literature advocating the indirect determination of the blood alcohol concentration (BAC) via the measurement of alcohol in an appropriate sample of breath. Once breath alcohol concentration (BRAC) has been established the idea is to convert it into (BAC) by multiplying the breath alcohol reading by a factor called the partition ratio⁷. It is not in dispute that the conversion of breath alcohol concentration to blood alcohol concentration raises a number of contentious issues.

60. From the reading of scientific publications as well as the evidence of the expert witnesses in this case, Dr Wagner, Dr Williams, both on behalf of the defendant, as well as Dr Saayman, on behalf of the plaintiff, it becomes clear that the partition ratio varies⁸ from person to person and also varies within a given individual at different times, depending on a number of factors, some physiological and others not. Such factors may be, for example, the effect of body temperature of the individual, cellular composition of blood, etc. According to some academic or scientific publications it is not unknown that a ratio with regard to an individual person can be as low as 600:1. Dr Williams sees it possible in a situation where one would take a large dose of alcohol very quickly on an empty stomach and then a sample is taken from the venous blood and breath. He argues that it is not a genuine partition ratio but it is something artefactual. It is common therefore to establish population based averages and the commonly used is 2100:1 blood- breath ratio. It is common cause that as the ratios vary the variance can produce erroneous test results. The view of Dr Williams, is that the applicable ratio is 2300:1 which translates to a multiplication factor of 0,23.

61. To illustrate the extent of the controversy on this subject of partition ratio I was referred to the book edited by James C Garriott⁹. An extract from the book reads:

“Statues have erroneously assumed that the same partition ratio applies to everyone. Scientific studies have shown ACTUAL partition ratios to be as low as 1,100:1 and higher than 2,700:1. A true average for all persons may more accurately be stated at 2,300:1 and 2,350:1. All breath testing devices used in America assume that the test subject’s breath will be exactly 2,100 times greater than the concentration of alcohol in the person’s blood. Statutes also erroneously assume that alcohol is absorbed and eliminated at identical rates by every person, when in vio testing thoroughly refutes this assumption.”

Further on it reads:

“Thus any, “fixed” ratio that the law mandates is nothing more than a legal fiction to help facilitate the use of breath testing devices.”

62. The lesser the ratio the more favourable it is to the person tested, in other words the lesser the BAC will be. Dr Wagner’s contention is that it is equitable to use the 2100:1 ratio in the current case as it is the average that is being used in South Africa. He refers to the Knights Forensic Pathology article referred to which gives the normal ratio to be between 2100 and 2400:1. According to him the current statutory limits of 0,05 g/100 ml and 0,24 mg per 1000 millilitres are a product of a calculation based on this ratio. Indeed mathematically it works up to that but who knows if it is a direct product of that calculation. Applying the ratio to the reading of 0,44 mg/l, according to him, would amount to the blood alcohol reading of 0,09 g/100 ml which obviously is well above the statutory limit of alcohol in the blood.

63. The contention on behalf of the plaintiff is that the average ratio should not be applied to an individual person as there can always be individual variations. In terms of the evidence the average ratios are established by collating ratios, some high and others low, from individual persons.

64. Mr Cole’s submission is that the partition ratio applicable to the population at large is within the range of 2100:1 to 2400:1 and the court should then accept on a balance of probabilities that a ratio of 2100:1 applies to the general population and that average ratio should then be applied to the plaintiff, and that would result in BAC of 0,09 g/100 ml. He submits further that the plaintiff bears the evidential burden to satisfy the court that he is an exception to the general rule.

65. It appears to me that whatever average partition ratio is accepted to be, that will not address the issue as to why such ratio should be applied to an individual case in a situation where there is no per se prescript to that effect. Even in the National Road Traffic Act the average ratio is not prescribed. It may well be that the ratio 2100:1 was used in establishing the statutory limits. It appears to me logical and equitable that the individual person’s ratio be established as a basis for the conversion of the alcohol content in his/her breath in order to determine the level of alcohol in his/her blood.

66. A consideration would have to be given to whether there is a burden of rebuttal resting with the plaintiff to show that the average partition ratio should not apply in his circumstances. I now turn to consider whether the evidential burden lies upon the plaintiff.

67. Firstly, the manner of establishing the average ratio, as is clearly coming out of the evidence adduced and the relevant scientific publications, is due to collation of ratios that may be less or more, if not far less or far more, than the ratio that is finally established. It is not proven scientifically that an average partition ratio, say 2100:1, applies to each individual or that it is the minimum ratio there can be. One wonders therefore what purpose the average ratio serves in a situation where there is no ratio prescribed by law, where a duty would be created for one to rebut the applicability thereof to his/her specific case. The average ratio is not established on the basis that each tested person’s ratio is round about 2100:1, for instance. In my view that is a significant factor to consider when deciding whose responsibility it is to establish and prove the personal ratio of an individual in a given situation.

68. Secondly, the question that arises is what the plaintiff would have been expected to do if he had the evidential burden. Would he have been expected to have determined his ratio on the day of the accident or at any other time before October 2004 when expert notices were filed? It appears that it was only then that the issue relating to ratio was formulated. Given the fact that it has been conceded that factors that affect the partition ratio vary from individual to individual and may, at different times, vary within the same individual, plaintiff, it appears, would not be expected to have done anything as from October 2004 towards establishing its ratio as at the time of the accident. In Kruger vs Stoltz & Another 2003 JDR 0800 (T) Patel J found that the onus to prove that the defendant drove the insured vehicle whilst under the influence of liquor or whilst his blood alcohol content exceeded the prescribed legal limit had not been discharged. The effect was that the exemption clause contained in the contract of insurance could not be invoked. One of the reasons for the decision was that it was conceded by an expert witness on behalf of the litigant that bore the onus that “the only way in which one could have obtained the defendant’s conversion ratio as at the time of the accident was to have procured both his blood-alcohol and breath- alcohol levels and then to make the calculations.” Amongst other things, this highlights that if the intimation with regard to partition ratios came to the notice of the plaintiff in October 2004, it would be meaningless to attempt to establish his ratio as at the date of the accident. It also indicates that Patel J accepted that it was the responsibility of the litigant who had the onus to prove the driver’s ratio.

69. In the pleadings the issue regarding partition ratios has not been raised. The plaintiff’s case is not that his partition ratio is different from whatever ratio that is considered average. Plaintiff’s case is that his own conversion ratio, which may be different from the average ratio, is the one that should be applied. His partition ratio is unknown and the issue is not that it “does not conform to that of the public at large” as Mr Cole so suggested. The defendant does not allege and it appears was not in a position to positively allege, that the average partition ratio is the one that is capable of producing the correct results in the conversion exercise involving the plaintiff. I have thus come to the conclusion that the issue of an average conversion ration, even if established, does not cast an evidential burden on the plaintiff to come up with factors that would show that the average ratio does not apply to him. Placing the evidential burden upon the plaintiff would be tantamount to applying what in American jurisprudence is known as the ‘rule of convenience’ which states that “when the exonerating fact was peculiarly within the defendant’s control, the burden was on the defendant to present [such] evidence”. That is a rule that is not generally applied.¹⁰

70. The defendant also relies on the fact that Mr Botha’s breathalyser reading was 0,45 mg per litre and in the blood sample taken about 1½ hours after the breath test the blood alcohol concentration was established at 0,11 gm per 100 ml. The submission therefore is that it is a plausible inference that plaintiff’s blood, considering the breath alcohol content of 0,44 gm per 100 ml, would have registered a reading above the statutory limit of 0,05 gm per 100 ml if the blood sample had been analysed.

71. To use the words of Doctor Wagner, the accuracy of the Lion S-D2 machine used “was verified unequivocally by the blood sample of Botha which was analysed by the accepted Headspace Gas Chromatography method.” He acknowledged that the blood samples from Mr Botha and the plaintiff were taken almost at the same time which is about 1½ hours from the time the breath tests were conducted. He goes on to say the analysis of Mr Botha’s blood sample showed that his BAC was well over the statutory limit and therefore on a balance of probabilities it has to be accepted that a BRAC reading of 0,44 indicated that the blood level exceeded the legal limit as regards the plaintiff.

72. Doctor Williams agrees with Dr Wagner, but speaking of Mr Botha’s breath and blood alcohol levels, he expresses some kind of reservation in the following sense:

“His actual blood specimen taken about 90 minutes later was found to contain 0.110%, whereas one might expect it to have decreased to a lower level than this in the intervening period. However, as we do not know when PRICE consumed his last drink before driving that evening, or indeed what he had also had to eat, it may well be that his blood alcohol level was still rising or was at least static on a plateau during the intervening period [between breath and blood].”

73. When Dr Williams was questioned under cross-examination as to whether the equation would be proper if, for instance, Mr Botha had partaken of a large amount of alcohol just prior to the accident, Dr Williams respondent as follows:

“I fully agree with you there, but the point is ,47 equates to ,114 and the blood was actually ,110, so in a period of ninety minutes there has only been an ,004 loss, which means if anything he was either tested on the plateau or the ratio has been skewed because he was still absorbing.”

74. The extract referred to above from Garriott’s book warns of the error commonly made by assuming that alcohol is absorbed and eliminated at identical rates by every person. It appears that the common view by experts is that once alcohol drink is consumed some of it gets absorbed directly through the walls of the stomach and through to the small intestines. This is normally a quick process but there may be a delay caused by strong solutions of alcohol so that the alcohol does not quickly get into the bloodstream. The question of an empty or full stomach also impacts on the pace taken by alcohol in its journey to the bloodstream. The metabolic breakdown that takes place in the liver results in the elimination of alcohol in the blood. The rate of such elimination is dependant on numerous factors like the quantity of food in the stomach before the consumption, and this may differ from person to person. A W Jones and L Andersson in the article titled - Comparison of Ethanol Concentrations in Venous Blood and End-Expired Breath during a controlled drinking study, had this to say:

“The closeness of agreement between blood- and breath-alcohol concentration profiles during absorption, distribution and elimination stages of ethanol metabolism deserve further investigation.”

In the same article they further state that:

“In conclusion, the results of this controlled study show that both blood- and breath-alcohol concentration-time data can be used to make forward or backward estimations during the post-absorptive phase of ethanol kinetics. Translation of the BrAC into BAC before these projections are made is not recommended and can give misleading information because the BAC/BrAC factor is a moving target.”

75. Professor Saayman maintains that the corroboration sought by the test results of the blood alcohol analysis of Mr Botha’s sample raises some doubts when one takes into account the long time lapse between the taking of the breath samples and the blood samples. He maintains that there would certainly be changes in the BAC due to interim absorption or elimination or metabolism. His opinion is that the elimination of consumed alcohol takes place on an average rate of 0,01 to 0,02 gm per cent per hour subject to individual variations which may be quite large. For the given period therefore close correlation between BRAC and BAC for Mr Botha is unexpected. He ends up by saying in his report:

“precautionary warning or reservations regarding calculations pertaining to metabolism and/or the effect of alcohol in specific individuals, must be incorporated in any expert report and duly recognised. This is probably the biggest limiting factor in terms of expressed expert opinion in legal proceedings pertaining to specific cases.”

76. In light of the aforegoing it appears it is inappropriate to make an inference from what is perceived as correlation between the breath alcohol reading and blood alcohol concentration of Mr Botha as support for the view that the plaintiff’s blood alcohol concentration had exceeded the statutory limit at the time he was driving.

77. I therefore come to the following conclusions.

78. There is no evidence to justify that the plaintiff drove whilst under the influence of intoxicating liquor or drugs. As indicated above, it is only the witness Simakuhle who referred to a liquor smell on the breath of the plaintiff. It is only him who observed that the eyes of the plaintiff were reddish. This he did after he had been with the plaintiff at the scene for some time without noticing anything in the plaintiff’s eyes. It was only when he was travelling with him from the police station to hospital when he made that observation. It appears that even that witness was not convinced that the plaintiff was under the influence of liquor as he testified that he did not arrest him for that but for having exceeded the statutory limit whilst driving. The fact that the plaintiff admitted to having consumed a few drinks or having taken a drink does not take the matter any further. It is no proof that his skill and judgment were diminished or impaired as a result of the consumption of such drinks. I am not persuaded that there is any merit in the contention that the attack on Mr Botha by the plaintiff resulted from alcohol in his system. This is a perception not supported by the facts, more especially that the attack was allegedly blamed on shock. I am not aware of any evidence supporting what is referred to as ‘reduced peripheral vision’ and memory deficits on the part of the plaintiff.

79. Earlier in the judgment I referred to the concerns raised in the case of S v Strydom which, in my view, cast doubt on the correctness of the calibration carried out by Mr Gallant. He admitted that he knows nothing more about calibration as well as the instrument itself than connecting a small pipe to a cylinder on the instrument and then turning on the figure that appears until it registered 0,24. It follows that if the figure is already 0,24 when he starts the process, there is nothing more to do unless it is for verification purposes. The verification process itself is unsatisfactory as the instrument can perpetuate its faults by exhibiting the figure 0,24 again and again. This means there is no independent way of checking the correctness of the instrument.

80. The defendant relies only on the evidence of Mr Gallant in this regard. I agree with Mr De Bruyn that after listening to the evidence of Mr Gallant one is left uncertain what kind of cylinder that was used, what was inside the cylinder and whether the calibration was done according to specifications, if there are. This criticism emanates largely from what was revealed by the witness under cross-examination. I am of the view that it is not fair therefore to talk of litigation by ambush.

81. It has been conceded that a partition ratio that would bring out the correct BAC of the plaintiff is the one that applies to him as a person. The argument on behalf of the defendant is that where there is no case made out by the plaintiff as to the factors you meet to his personal situation that might show that his personal partition ratio is different from the average one the latter ratio should then apply. I have dealt with the issue of onus and evidential burden above. It is my view that without having his personal ratio established and applied the defendant has not discharged the onus it bears.

82. For the sake of clarity, I make no decision on the question of which of the specific population based average ratios is applicable as it is not necessary for the purpose of this judgment. Suffice to say it does appear that 2100:1 ratio is commonly used albeit largely based on statutory mandate.

83. Taking the totality of evidence into account I come to the conclusion that in view of the calibration shortcomings referred to above the defendant has not discharged its onus to prove that the breathalyser reading of 0,44 mg per litre reflected the correct level of alcohol in the breath of the plaintiff. I also come to the conclusion that it would fall within the parameters of the defendant’s onus to establish the plaintiff’s individual partition ratio and if needs be, to apply it in the conversion exercise from BRAC to BAC. Such onus has also not been discharged in this regard.

84. As regards the issue of costs Mr Cole’s submission is that if the plaintiff is successful an award of costs should stand over pending the determination of quantum as it may well fall within the jurisdiction of the magistrate’s court. The counter argument by Mr De Bruyn is that the proceedings have now reached the end of the merits stage and the principle that costs follow the result should now apply.

85. The court has a discretion towards costs at this stage. The issues agreed upon by the parties at pre-trial stage are factors relevant in the exercise of the discretion and also in deciding whether the costs should be awarded on a High Court scale.

86. In the current case the parties agreed, inter alia, that merits and the quantum be separated and that “this matter cannot be resolved by mitigation/arbitration, or in another jurisdiction”. The case was tried in the High Court as a consequence of that agreement. It is fair therefore that the costs be on a High Court scale irrespective of the quantum subsequently proved or agreed upon. (Brauns v Shoprite Checkers Pty Ltd 2004 (6) SA 211 (E) p221C-E).

87. It is not disputed that this matter is highly technical, complex and very important to both parties so as to justify employment of two counsel by the plaintiff. There is however no merit in the submission by Mr de Bruyn that costs should be on an attorney and client scale. Nothing points out as contended on behalf of the plaintiff, to the defendant having attempted to show that the plaintiff “had orchestrated inter alia the loss of the police docket and the loss of his blood sample, etc.”

88. The defendant failed to discharge the onus to prove that the plaintiff drove the relevant motor vehicle whilst under the influence of intoxicating liquor or drugs or whilst the concentration of alcohol in his blood exceeded the statutory limit.

In the result I make the following order:

1. That the plaintiff’s claim is not excluded by the special exclusions section in plaintiff’s insurance policy with the defendant, referred to in paragraph 3 of this judgment;

2. That defendant should pay to the plaintiff such damages, as the consequence of the accident on 25 April 2003, the plaintiff is able to prove or are agreed upon;

3. The defendant is ordered to pay the plaintiff’s costs in respect of the merits on a High Court scale, such costs to include the costs occasioned by the employment of two counsel and, such costs to include qualifying expenses of Professor Saayman.

C T SANGONI

JUDGE OF THE HIGH COURT Date:

Counsel for the Plaintiff : Adv P J de Bruyn SC

Adv B J Pienaar

PORT ELIZABETH

Attorneys for the Plaintiff : Friedman Scheckter

PORT ELIZABETH

Counsel for the Defendant : Adv S H Cole

GRAHAMSTOWN

Attorneys for the Defendant : Joubert Galpin Searle

PORT ELIZABETH

T N Price vs Mutual and Federal Ins Co Case no 2052/2003

---