---

THE HIGH COURT OF SOUTH AFRICA

(WESTERN CAPE HIGH COURT)

Appeal Case No: A135/2013

Trial Case No: 1221/2007

In the matter between: Not reportable

OMAR FARIED PANDIE	APPELLANT
And
RETHAAN ISAACS	RESPONDENT

Coram: BLIGNAULT, SALDANHA & ROGERS JJ

Heard: 26 JULY 2013

Delivered: 4 SEPTEMBER 2013

______________________________________________________________

JUDGMENT

______________________________________________________________

ROGERS J:

Introduction

1. The appellant and respondent were the defendant and plaintiff in the court below and I shall refer to them as such. The plaintiff sued the defendant, who is a gynaecologist and obstetrician, for damages for performing a sterilisation procedure on her on 4 November 2004 to which she had allegedly not consented. The sterilisation procedure in question is known as a bilateral tubal ligation. The trial judge upheld the claim and awarded her damages of R410 172,35 comprising R4 000 for past medical expenses, R200 000 as general damages, R198 172,35 for future medical expenses and R8 000 for loss of future earnings.

2. With the leave of the trial judge the defendant appeals to a full bench. In his appeal the defendant attacks the judge’s findings both on liability and quantum.

Factual overview

3. The plaintiff was 32 at the time the sterilisation procedure was performed. The sterilisation was carried out as an adjunct to a caesarean section during which the plaintiff’s fourth child was born. The operation was performed at the Chris Barnard Memorial Hospital (‘the CBMH’ or ‘the hospital’). There is no complaint by the plaintiff regarding the performing of the caesarean section. Her case is that she did not consent to the sterilisation.

4. The plaintiff’s pleaded case was understood by the parties to advance the following causes of action in the alternative, namely that the performing of the sterilisation without the plaintiff’s consent, informed or at all, constituted an assault; that it was a wrongful and negligent delictual act; that it was a breach of the plaintiff’s contract with the defendant; and that it was a breach of the Sterilisation Act 44 of 1998 and of certain guidelines published by the Medical and Dental Professions Board of the Health Professions Council of South Africa (‘the HPCSA’). The two main assertions in the pleadings which formed the foundation of these causes of action were [a] that in the consultations which preceded the performance of the operation on 4 November 2004, including the consultation on 3 November 2004, the plaintiff in oral discussion with the defendant made clear that she did not want to undergo sterilisation; and [b] that the defendant had been negligent in not personally checking, before doing the operation, whether the plaintiff had signed the written consent to sterilisation required by the Sterilisation Act.

5. There was a sharp factual dispute between the plaintiff on the one hand and the defendant and a nursing sister Desiree du Plessis (‘Du Plessis’) on the other as to what happened on the day of the operation (4 November 2004) and in the plaintiff’s prior consultations with the defendant (particularly on 3 November 2004). I shall identify these areas of dispute in my factual overview.

6. The plaintiff fell pregnant with her fourth child in February or March 2004. The father was Anwar Prinsloo (‘Prinsloo’) to whom she had been married for about ten years. She already had three children by that time, born in 1991, 1996 and 1998. The second and third children were delivered by caesarean section. The plaintiff testified that Prinsloo was an abusive husband and that the marriage had irretrievably broken down. She also testified, however, that the pregnancy was planned as a last attempt to repair the marriage. On the other hand, the psychologist who testified on her behalf, Mr Graeme Lewis (‘Lewis’), said that she told him that she and Prinsloo were intending to get divorced but then discovered that she was pregnant; that in terms of their Muslim faith the divorce had to be deferred until 100 days after the birth of the child; and that she saw the pregnancy as a last opportunity to restore the marriage. (In the event the plaintiff and Prinsloo did get divorced after the birth of the fourth child.)

7. The plaintiff first consulted the defendant (on the referral of her GP) on 9 March 2004 for reasons unrelated to pregnancy. When he saw her again on 7 April 2004 he ascertained that she was pregnant. Further antenatal consultations took place on 5 May 2004, 2 June 2004, 5 July 2004, 2 August 2004, 31 August 2004, 29 September 2004, 21 October 2004 and 3 November 2004 (the day before the operation). During these consultations it was agreed that the plaintiff’s baby would be delivered by caesarean section.

8. It is common cause that the question of sterilisation arose at the consultation on Wednesday 3 November 2004, a consultation also attended by Prinsloo. (Prinsloo was not called as a witness.) The plaintiff testified that the defendant had raised this question with her during two earlier consultations; that on both occasions she had said she did not want to be sterilised; and that she repeated this somewhat irritably when the defendant raised it again on 3 November 2004. The defendant, by contrast, said that this was the first time he raised the matter with her; that he asked her whether she wanted more children, to which she replied in the negative; that he explained the medical reasons for considering sterilisation following what would be her third caesarean section; that he told her what a tubal ligation would entail; and that she then said she wished to be sterilised.

9. Towards the end of the consultation, and after the discussion of sterilisation, the defendant wrote out in the presence of the plaintiff and Prinsloo a letter for her to hand to the hospital on admission the next day. He gave this to her in a sealed envelope. She did not see its contents. The letter informed the hospital that the plaintiff was being admitted for an elective caesarean section and tubal ligation (the letter used the abbreviations ‘C/S’ and ‘T/L’).

10. The medical evidence established that it was sound practice for a gynaecologist to draw to the attention of a patient who was to undergo a third caesarean section the increased risks associated with further pregnancies and to recommend a sterilisation. These risks were not, however, of a kind that would make it proper for the gynaecologist to press the case for sterilisation if the patient was reluctant or wanted further children.

11. The hospital records reflect that the plaintiff was admitted to the hospital at 10h56 on Thursday 4 November 2004 (this was probably on the 8^th floor). She was accompanied by her husband. She was taken by an assistant nurse, Ms Z Samsodien (‘Samsodien’), to the maternity ward on the 11^th floor. The plaintiff gave Samsodien the sealed letter from the defendant. What happened thereafter is a matter of some controversy. What is common cause is that Samsodien prepared the consent form for the plaintiff’s signature on the basis that the plaintiff was consenting to a caesarean section and a sterilisation by way of bilateral tubal ligation – one can assume that Samsodien did so on the basis of the defendant’s letter. It is also common cause that in final form the references on the consent form to sterilisation and bilateral tubal ligation were crossed out and initialled.

12. The plaintiff’s evidence was that when Samsodien presented her with the consent form referring to sterilisation she refused to sign it until those parts of the form were deleted. The defendant’s case, based on the evidence of Du Plessis, was that the plaintiff signed the form as completed by Samsodien; that the plaintiff subsequently told Samsodien she no longer wanted the sterilisation; that Samsodien consulted Du Plessis about this development; and that the two of them returned to the plaintiff, at which point the deletions were made and initialled by the plaintiff. Du Plessis testified that she told the plaintiff to inform the defendant when she saw him in the labour ward (which was on the 12^th floor next to the theatre) that she had decided not to have the sterilisation. Du Plessis also testified that she told Samsodien to bring this fact to the attention of the nursing staff in the labour ward. All of this was denied by the plaintiff. (Although the defendant’s counsel put to the plaintiff in cross-examination what Samsodien would supposedly say, Samsodien was not called as a witness.)

13. One of the hospital records, called a Nursing Progress Report, records against the time 11h30 that the plaintiff was admitted to the ward (ie the maternity ward) for a caesarean section and further: ‘Patient and husband verbalise they don’t want Tubal Ligation done.’ This note appears to have been made and signed by Samsodien.

14. There is another hospital record, called a Peri-Operative Patient Record, which, in a section headed ‘Pre-Operative Assessment and Preparation for Theatre’, records that informed consent was obtained but there is the handwritten qualification: ‘No T.L. please’ (ie no tubal ligation). This section of the Patient Record was signed by Samsodien. Samsodien at 12h00 also signed the next section of the Patient Record headed ‘Baseline Observations’.

15. According to the Nursing Progress Report, the plaintiff was transferred from the maternity ward on the 11^th floor to the labour ward on the 12^th floor at 12h30. It appears that she was taken there by Samsodien. The maternity operating theatre was also on the 12^th floor. Nurse J Solomons (‘Solomons’) worked in the labour ward. The section of the Patient Record headed ‘Final Pre-Operative Check at Theatre Reception/Handover’ bears her signature against the times 12h30 and 12h55. This section is on the same page as the two sections signed by Samsodien.

16. The hospital records show that the plaintiff was taken into the theatre at 13h10. It is common cause that the defendant did not see the plaintiff in the labour ward before she was taken to theatre. His evidence was that he arrived at the labour ward shortly after 13h10 in order to talk to her but was told by Solomons that the plaintiff had already been taken to theatre. He said his reaction to this news was: ‘Ms Prinsloo, the patient for the caesarean section and sterilisation?’, to which Solomons allegedly replied in the affirmative. (Solomons was not called as a witness.)

17. The medical personnel in theatre were the defendant, his assistant Dr Kriel; an anaesthetist Dr Whitehead, and a paediatrician Dr Zieff. The senior theatre nurse was Sister Kim Venter (‘Venter’). There was at least one other nurse in attendance (her name was not established in evidence – I shall refer to her as the ‘junior theatre nurse’). Dr Whitehead administered a spinal anaesthetic which permitted the plaintiff to remain conscious and alert during the operation. The area of her body on which the defendant operated was shielded from her sight by a screen behind which the defendant worked. Prinsloo was in the theatre with her. The defendant began the caesarean section at 13h25, and at 13h29 the baby was delivered. After closing the uterus the defendant performed the bilateral tubal ligation which took only a couple of minutes.

18. The events in theatre are a matter of dispute. Apart from the parties themselves, the defendant called Dr Whitehead as a witness. Venter was deceased by the time of the trial. The other medical and nursing personnel who were in attendance were not called by either side.

19. The plaintiff’s version in summary was that prior to the defendant’s arrival at theatre one of the nursing sisters looked at her folder and said that she noticed the plaintiff was not having the sterilisation. The plaintiff did not know the nurses’ names but described this particular nurse as ‘darker skinned’ than the other nurse (who was fair with red/brown hair). The plaintiff’s expert, Dr Rosemann, in setting out the plaintiff’s version in one of his reports, refers to the former nurse as an ‘Asian nursing sister’. It seems likely that this was the junior theatre nurse. The plaintiff testified that Dr Whitehead also looked at her folder and remarked that he noticed she was not having the sterilisation. She described the defendant’s demeanour on his arrival as aloof and distant. He did not have any conversation of relevance with her nor did she hear him say anything to any other personnel about sterilisation. Upon completion of the surgical procedures, one of the nurses held up and showed her a bottle containing two pieces of floating tissue (the plaintiff testified that this was the fairer-skinned nurse, and it appears probable that this was Venter). Dr Whitehead then whispered in her ear, ‘But you didn’t ask for a sterilisation’, at which she began to cry, eventually falling asleep. When she awoke she was back in the maternity ward.

20. The defendant’s version was quite different. He testified that on his arrival in theatre he was met by Venter. He greeted everyone, including the plaintiff, and then went to the scrub room. As he began to operate he told the plaintiff he was proceeding with the caesarean section. Upon delivering her child he congratulated her on the birth of a son. After closing up her uterus he turned to Venter and asked: ‘Are we proceeding with this sterilisation?’. She said yes and handed him the necessary instruments. After performing the tubal ligation he went to the change room. Venter came there and said that the plaintiff had not been meant to have a sterilisation. He was shocked. He went to the plaintiff, who was in the recovery room, and told her he was very sorry and would need to find out where the change occurred. He testified that she was not asleep though she may have been dozing off. He then proceeded to the 9^th floor where he had another operation scheduled.

21. Dr Whitehead testified that in the ordinary course he would have seen a patient such as the plaintiff in the labour ward 15 to 30 minutes before the operation though he had no independent recollection of her case. He said it would make no difference to his anaesthetic procedures whether the caesarean section was with or without a tubal ligation. He could not remember whether he was told that the plaintiff’s caesarean section was with or without a tubal ligation. He denied that he checked her folder in theatre and spoke to her as she alleged. He was quite positive that there had been no conversation raising any controversy – it was the sort of thing he would remember. He also denied that on completion of the operation he whispered to her as the plaintiff claimed.

22. The plaintiff remained in the maternity ward until she was discharged at noon on 7 November 2004. Much time was spent during the trial on what happened in the postnatal phase. The adequacy of the defendant’s postnatal care was not an issue in the case and these events are irrelevant except to the extent that they may bear on credibility. It is common cause that the defendant did not raise the sterilisation with the plaintiff during the postnatal phase – on his version, this was because he was unable to ascertain the facts from the nursing staff prior to the plaintiff’s discharge and because she did not thereafter attend the consultation scheduled for two weeks after the operation or respond to his secretary’s telephone calls. It is also common cause that the plaintiff did not raise the sterilisation with the defendant, though he attended on her in the ward on several occasions prior to discharge. She said that she was too angry to do so.

23. The plaintiff testified that she did not know exactly what procedure the defendant had performed on her (apart from the caesarean section). She eventually consulted a new gynaecologist, Dr Elmarie Basson, in September 2005 to confirm that she had indeed been sterilised. She testified that it was Dr Basson who explained that her fallopian tubes had been cut on both sides. Dr Basson informed her that a reversal of the bilateral tubal ligation could be attempted (by way of a procedure known as reanastomosis).

24. During May 2007 the plaintiff’s then attorneys wrote to the defendant concerning the matter (their letter is not in the record). He replied in a letter dated 4 June 2007. Summons was issued in September 2007. The case was heard during October and November 2010. The plaintiff testified and called as experts Dr GWE Rosemann (a gynaecologist) and Mr G Lewis (a psychologist) (‘Lewis’). The defendant testified and called as an expert Dr JOT van Helsdingen (a gynaecologist). The defendant also called the nursing sister, Du Plessis and Dr Whitehead. Judgment was delivered on 16 May 2012.

25. To complete the factual overview, I should mention that the plaintiff got divorced from Prinsloo at some stage after the operation. The date is unclear – she testified that the divorce was about a year and a half after the operation (which would be in about May 2006) though she told Lewis that she got divorced at the end of 2005. She then formed a relationship with one Ricardo Davids (‘Davids’). When this relationship started and ended is not possible to ascertain with any confidence from the record. The plaintiff testified that she started seeing Davids about a year and a half after the divorce (which would be sometime between May and December 2007) but she told Lewis variously that the relationship started in early 2006; or that it started six to seven months after her fourth child’s birth (which would be in June/July 2005). She testified that the relationship lasted two and a half years, though she said to Lewis that the relationship spanned ‘almost two years’. According to Lewis, the plaintiff told him that Davids wanted to marry her and was excited about the possibility of a reversal of the sterilisation. However, the reversal procedure scared her and she was not sure she could go through with it. David’s sister also told her that Davids deserved a women who did not already have children and who could still bear children. The plaintiff thus decided to break off the relationship.

The judgment a quo

26. After summarising the evidence of the various witnesses, the trial judge made certain credibility findings. The plaintiff impressed him as an honest and credible witness. The defendant, by contrast, was said to have been ‘very economic with the truth’. The judge considered that the defendant ‘cracked’ under cross-examination, particularly concerning two apparently different versions of his file notes and concerning supposed contradictions between his oral testimony on the one hand and his letter of fortune 2007 on the other. The trial judge labelled him an ‘outright liar’. Du Plessis also did not impress the trial judge as an honest and credible witness. Dr Whitehead, so the judge concluded, had attempted to distance himself from anything to do with the matter. Regarding the two gynaecological experts, Dr Rosemann was found to be a ‘very impressive witness’ and the judge had no doubt as to his honesty and credibility. Dr van Helsdingen, by contrast, was according to the trial judge ‘not at all honest’ on the important question as to whether it was the duty of a surgeon to check the patient’s written consent form prior to commencing an operation. Lewis impressed the trial judge as honest and credible.

27. The trial judge did not set out his specific factual findings on the material areas of dispute. In the light of his credibility findings, though, it may safely be assumed that he adjudicated the case on the basis that the plaintiff made it clear to the defendant in the consultation of 3 November 2004 that she did not want a sterilisation; that this was a repeat of what she had said on two earlier occasions; that on admission to the hospital on 4 November 2004 she refused to sign the consent form until the references to tubal ligation were deleted; that Du Plessis said nothing to her about telling the defendant that she had changed her mind; and that the nurse and Dr Whitehead spoke to her in theatre as she claimed (and that Dr Whitehead’s contrary evidence was thus to be rejected). Whether the trial judge rejected the defendant’s evidence regarding what Solomons said to him in the labour ward and regarding his query to Venter in theatre and her reply is not clear from the judgment. The plaintiff in the nature of things had no personal knowledge as to what passed between the defendant and Solomon. And as to the alleged query by the defendant to Venter in theatre, the plaintiff conceded that it was possible that there was some conversation between the defendant and Venter which she did not hear.

28. The trial judge upheld all the pleaded causes of action (assault; delictual negligence; breach of statutory duty; and breach of contract), essentially on the basis that the defendant should himself have checked the written consent form before proceeding with the operation. This finding was in accordance with Dr Rosemann’s opinion in oral evidence.

29. In regard to damages, the amount of R4 000 for past medical expenses was apparently agreed.

30. As to general damages, the trial judge considered that R200 000 was fair and reasonable. He made reference in this regard to loss of amenities, particularly the plaintiff’s child-bearing capacity; disfigurement (on appeal, disfigurement was said to relate to the scars from a future reanastomosis); the plaintiff’s emotional suffering; and pain and suffering in general.

31. The sum awarded in respect of future medical expenses, namely R198 172,35, was arrived at by applying a 10% contingency deduction to a gross amount of R220 191,50. Although not spelt out in the judgment, we were told the gross amount was calculated thus:

• the cost of the reversal procedure (reanastomosis) – R35 000;

• five cycles of in vitro fertilisation (‘IVF’) treatment at R35 000 per cycle – R175 000;

• 17 sessions of psychotherapy at R599,50 per session – R10 191,50.

32. Finally, an amount of R8 000 was awarded in respect of future loss of earnings, based on the fact that reanastomosis and five cycles of IVF would require the plaintiff to be off work for about a month.

The legal framework

33. Physical interference with another’s body is delictually wrongful unless there is a ground of justification (Van der Walt & Midgley Principles of Delict 3^rd Ed para 78 and case there cited). One such ground is consent, encapsulated in the maxim volenti non fit iniuria. In the context of physical interference in the form of medical treatment, the consent required is informed consent as explained in Castell v De Greef 1994 (4) SA 408 (C).at 426F-H. If a doctor’s treatment is wrongful because there is no informed consent, he will be liable to the patient in delict if his wrongful conduct was perpetrated with fault (dolus or culpa) and caused the harm of which the patient complains.

34. The plaintiff in this case pleaded, among other things, that the defendant’s conduct was an assault, alternatively a wrongful and negligent act. The use of the word ‘assault’ with reference to the administration of medical treatment without consent was questioned in Broude v McIntosh 1998 (3) SA 60 (A) at 67H-68F (and see also Van der Walt & Midgley op cit para 78). In Louwrens v Oldwage [2004] 1 All SA 532 (C) para 99 the doubt expressed by the Appellate Division was said to have been obiter, and Castell (a full bench decision) was said to support the classification of such conduct as an assault. Louwrens was overruled by the Supreme Court of Appeal without reference to this aspect – 2006 (2) SA 161 (SCA). Castell does not, as I read it, address this particular issue. In my respectful view, the use of the word ‘assault’ is unnecessary in this setting and is apt to confuse. In criminal law, where one has various specific crimes with their own distinctive elements, assault is an offence which requires among other things that the accused should have acted with dolus, which encompasses knowledge of the wrongfulness of his conduct. If, for example, the accused mistakenly believes he is acting in legitimate self-defence, he will not be guilty of assault. Delictual liability, by contrast, is determined with reference to overarching general principles of wrongfulness, causation, fault and damage. Unlike English law, we do not have a collection of separate torts. In English law, for example, a doctor who performs an operation without consent may be sued for the tort of battery or the tort of negligence or both (see Jones Medical Negligence (2003) paras 2.018 – 2.019 and paras 6.003 – 6.162 where there is detailed discussion of the differing rules relating to consent and causation for the two torts). In our law, by contrast, the wrongful invasion of another’s body causing harm gives rise to delictual liability if the perpetrator acted with dolus or culpa. If a defendant has acted with dolus (encompassing knowledge of the wrongfulness of his conduct) his conduct might coincidently amount to a criminal assault but it adds nothing to a delictual analysis to use this label. But the word ‘assault’ is also misleading where it is used in a medical setting to denote merely that the doctor administered treatment without informed consent, because such conduct, even though it is wrongful, does not without more give rise to liability and would rarely be perpetrated with dolus – it might not even be negligent.

35. Mr Bhoopchand for the plaintiff did not argue before us that the defendant’s performance of the sterilisation was an assault in the sense of treatment administered in the knowledge that the plaintiff had not consented to it. As will appear from my later analysis of the facts, Mr Bhoopchand’s concession was clearly right. I thus do not think the trial judge should have typified the defendant’s conduct in this way.

36. Apart from the case based on delictual negligence, the plaintiff pleaded a concurrent action in contract. This is permissible in the medical context (LAWSA 2^nd Ed Vol 8 para 53; cf Lillicrap, Wassenaar and Partners v Pilkington Brothers (SA) (Pty) Ltd 1985 (1) SA 475 (A) at 499A-500A and 501H-502G, inter alia explaining certain dicta of Innes CJ in Van Wyk v Lewis 1924 AD 438 which was a medical case). However, and given the centrality of consent to the delictual and contractual analysis, Mr Bhoopchand correctly accepted that if the delictual claim failed the contractual one would fare no better. If, on the other hand, the delictual claim was correctly upheld by the trial judge, a concurrent finding in contract would not yield better results for the plaintiff than the delictual one. On the contrary, in a claim based on breach of contract the plaintiff would be confined to claiming pecuniary loss – no general damages for pain and suffering could be recovered (see Administrator, Natal v Edouard [1990] ZASCA 60; 1990 (3) SA 581 (A) at 595D-597H).

37. The plaintiff referred in her pleadings to the Sterilisation Act and the HPCSA guidelines. The guidelines do not have the status of law and are merely part of the evidential material to be weighed in determining the standards reasonably to be observed by doctors.

38. Section 2(2) of the Sterilisation Act stipulates that a person capable of consenting to be sterilised may not be sterilised ‘without his or her consent’. The word ‘consent’ is defined in s 1 as meaning ‘consent contemplated in section 4’. As at November 2004 s 4 read thus: ¹

‘4. Consent. – For the purposes of this act, “consent” means consent given freely and voluntarily without any inducement and may only be given if the person giving it has –

1. been given a clear explanation and adequate description of the –

1. proposed plan of the procedure; and

2. consequences, risks and the reversible or irreversible nature of the sterilisation procedure;

2. been given advice that the consent may be withdrawn at any time before the treatment; and

3. signed the prescribed consent form.’

39. Regulation 2(1) of the regulations prescribed under s 8 of the Act provides that a person who is capable of consenting and who requests that a sterilisation be performed on him or her ‘shall complete Part A and Part B of Form 1 and submit such form to the head of the health facility, together with the standard consent form’. The phrase ‘standard consent form’ is defined in the regulations as meaning ‘a form provided by the health facility for recording consent to the administration of a local or general anaesthetic, and an accompanying surgical procedure’.

40. Part A of Form 1 provides for the insertion of the following particulars: name, ID number, age, race and number of children. Part B is a signed declaration requesting the sterilisation.

41. Although the regulations contemplate two documents – a signed request for sterilisation in prescribed form and the hospital’s standard consent form – no reference was made in the present case to the presence or absence of a signed request. In so far as the Act is concerned, the case was fought on the basis that what was relevant was the hospital’s standard consent form and the manner in which it was amended at the plaintiff’s insistence.

42. The precise significance of the Sterilisation Act in the context of a civil claim for damages is a question not without difficulty. The plaintiff appears to have relied on the Act as conferring a distinct right of action. Mr Albertus SC for the defendant submitted that no civil remedy was conferred by the Act (though contravention is a criminal offence – see s 9). In Steenkamp NO v Provincial Tender Board, Eastern Cape 2006 (3) SA 151 (SCA) Harms JA said (paras 20-22) that a statutory and common law duty may overlap. Where the legal duty involved is imposed by statute the focal question is one of statutory interpretation, namely whether the statute confers a right of action or provides a basis for inferring that a legal duty exists at common law. He said that our case law is not clear when it comes to drawing the boundary between liability due to breach of statutory duty and common law liability. He expressed the view that if the statute on a proper interpretation provides a remedy in damages a common law duty cannot arise; and that if the statute does not provide for a damages remedy, to infer a common law remedy would be contrary to the statutory scheme.

43. If a statute on a proper interpretation confers a civil remedy for damages the further question may arise as to whether the liability is strict or requires proof of fault (LAWSA 2^nd Ed Vol 8 para 74).

44. In my view it is unnecessary in the present case to search for a right to claim civil damages in the Sterilisation Act. The interference with another person’s body has always been prima facie wrongful at common law, and this is as true for an act of sterilisation as any other physical interference. If the wrongful act causes harm and is performed with dolus or culpa there is a common law remedy. The question is whether the common law ground of justification in the form of consent is – in relation to sterilisations – now qualified by the Act, so that what a defendant must prove to negative wrongfulness is consent in accordance with the Act (in which event informed oral consent would not suffice). To hold that the Act regulates for all purposes, including civil law, the manner in which justification by consent is to occur in relation to sterilisations is not to say that the Act confers a civil remedy for damages – that remedy is already supplied by the common law.

45. In my opinion, the Act should indeed be interpreted as regulating the manner of consent to sterilisations for all purposes, including as a justification ground in delict. The Act is directed at protecting the individual interests of patients in relation to their reproductive capacity and respecting their individual right of choice. It strikes me as anomalous that a prima facie wrongful act may be legitimised by obtaining a consent which does not comply with the Act and where the performance of the sterilisation is (subject to questions of mens rea) a criminal offence. It may be argued that it would be unsatisfactory to hold a doctor liable where he has obtained informed oral consent but has failed to procure a written consent in accordance with the Act. The answer to that objection is that at common law the doctor is not liable merely because the act of sterilisation is wrongful; it is also necessary to show that he acted with fault and that the element of causation is satisfied. Where a patient has given informed oral consent but complains ex post facto that there was no written consent, the test for causation would entitle one hypothetically to substitute a lawful course of conduct in which the doctor sought the written consent before operating; if one knows that the patient gave informed oral consent, one could probably conclude that he or she would have signed the consent form if requested to do so, so that the doctor’s wrongful conduct was not the cause of harm.

Factual analysis

46. In my view, an application of the law to the facts of this case leads to the conclusion that written consent as required by the Act was not present, so that the performance of the sterilisation, which was prima facie wrongful at common law, was not successfully justified by the defendant; but that the defendant performed the sterilisation in the bona fide belief that the plaintiff had consented, and that in so doing he acted in accordance with prevailing practice in his profession and was not negligent. He should thus not have been found liable. The evidence indicates that the hospital staff were negligent in dealing with the obtaining of the plaintiff’s written consent. Whether the defendant might have been held vicariously liable for their negligence was not, however, a matter which was pleaded and it would not be permissible or fair to decide the case on the basis of vicarious liability at this late stage of proceedings.

47. These conclusions require, firstly, a consideration of the evidence as to what occurred at the consultation on 3 November 2004. If the plaintiff, as she testified, rejected the sterilisation, it is obvious that what the defendant did the next day must either have been knowingly wrongful or at least grossly negligent. I do not think Mr Albertus SC for the defendant suggested otherwise. The position is, however, different (as will appear) if, as the defendant testified, the plaintiff gave oral consent on 3 November 2004 (though Mr Bhoopchand did not concede that this would be decisive). I should add, at this point, that although the pleadings made reference to the need for consent to be informed, the case was not contested on the basis that if the defendant’s version was true the plaintiff’s consent on 3 November 2004 was nevertheless not an informed one. On the defendant’s version he gave a detailed explanation of the risks underlying his advice in favour of sterilisation and of the nature of the sterilisation procedure. If anything was missing from the alleged advice, it was not shown to have been significant or to have been something which would have affected the plaintiff’s response on that day.

48. The version of the plaintiff on the one hand and the defendant and Du Plessis on the other were in direct conflict. The correct approach to resolving the factual disputes arising from their irreconcilable versions is the one described by Nienaber JA in Stellenbosch Farmers’ Winery Group Ltd & Another v Martell et Cie & Others 2003 (1) SA 11 (SCA) para 5:

’To come to a conclusion on the disputed issues a court must make findings on (a) the credibility of the various factual witnesses; (b) their reliability; and (c) the probabilities. As to (a), the court’s finding on the credibility of a particular witness will depend on its impression about the veracity of the witness. That in turn will depend on a variety of subsidiary factors, not necessarily in order of importance, such as (i) the witness’ candour and demeanour in the witness-box, (ii) his bias, latent and blatant, (iii) internal contradictions in his evidence, (iv) external contradictions with what was pleaded or put on his behalf, or with established fact or with his own extracurial statements or actions, (v) the probability or improbability of particular aspects of his version, (vi) the calibre and cogency of his performance compared to that of other witnesses testifying about the same incident or events. As to (b), a witness’ reliability will depend, apart from the factors mentioned under (a)(ii), (iv) and (v) above, on (i) the opportunities he had to experience or observe the event in question and (ii) the quality, integrity and independence of his recall thereof. As to (c), this necessitates an analysis and evaluation of the probability or improbability of each party’s version on each of the disputed issues. In the light of its assessment of (a), (b) and (c) the court will then, as a final step, determine whether the party burdened with the onus of proof has succeeded in discharging it. The hard case, which will doubtless be the rare one, occurs when a court’s credibility findings compel it in one direction and its evaluation of the general probabilities in another. The more convincing the former, the less convincing will be the latter. But when all factors are equipoised probabilities prevail.’

49. In my respectful view the trial judge failed properly to evaluate the conflicting evidence in accordance with this approach. He mentioned supposedly unsatisfactory features of the defendant’s evidence without addressing their real significance or the defendant’s explanations; he did not refer to some unsatisfactory features of the plaintiff’s evidence; and most importantly, he did not make reference to the inherent probabilities. His credibility findings did not, as far as I can discern, rest on his assessment of the demeanour of the witnesses but were based rather on his assessment of the objective quality of their evidence. In the circumstances, and despite the well-known reluctance of appellate courts to upset the credibility findings of a trial judge, I consider that this is a case where this court should revisit the factual findings (see Santam Bpk v Biddulph 2004 (5) SA 586 (SCA) para 5).

50. I have summarised the competing versions as to what happened at the consultation on 3 November 2004 on the subject of sterilisation. On both versions Prinsloo was present, and on the plaintiff’s version he participated in the decision – she testified that when the subject arose Prinsloo said that he did not want her to be sterilised but that it was her body to decide. The plaintiff did not call Prinsloo as a witness. It is true that by the time of the trial they were divorced and that he had apparently been an abusive husband, but the plaintiff’s testimony concerning his conduct during the consultation and at the hospital does not suggest that he was hostile or uncaring towards her in relation to the birth of their child. She did not testify that their relationship subsequent to the divorce was such that she could not call him to corroborate her version.

51. There are various inherent probabilities which point against the plaintiff’s version that she refused consent at the consultation on 3 November 2004 (of these, the third and fourth mentioned below are the most significant):

[a] Firstly, it is common cause that there was no financial or other incentive for the defendant to press the plaintiff to undergo a sterilisation. His fee for a tubal ligation was not more than R250. It was not put to him in cross-examination that he was a paternalistic surgeon who was intent on performing the procedure because ‘he knew best’. If on a prior occasion the defendant had mentioned sterilisation and the plaintiff had said no, it seems unlikely that the defendant would have raised the matter again. Yet on the plaintiff’s version he did so on three occasions, her answer being the same each time. It seems more probable that the matter was (as the defendant testified) raised only once, and that this was on 3 November 2004.

[b] Second, the plaintiff was 32 years old in November 2004 (she turned 33 in December 2004) and was about to give birth to her fourth child. While apparently being reasonably comfortable financially, she was not affluent. There were certain risks associated with further pregnancy. While a woman may obviously choose to have further children at a later age, it is not improbable that a woman in her position would have agreed to a sterilisation.

[c] Third, immediately after the discussion about sterilisation the defendant wrote out a referral letter in the presence of the plaintiff and Prinsloo which stated that she was to be admitted the next day for a caesarean section and tubal ligation. The plaintiff’s counsel did not put to the defendant nor argue to us that the defendant included the reference to tubal ligation deliberately, knowing it to be contrary to the plaintiff’s wishes. On the plaintiff’s version, her rejection of sterilisation was clearly conveyed to the defendant without scope for misunderstanding [5/449-450]. Mr Bhoopchand submitted that the defendant could have added ‘+ T/L’ mistakenly, ie absent-mindedly. He pointed to the fact that the defendant in the same letter mistakenly described the plaintiff’s child-bearing history as ‘G ₃P₄’ (medical shorthand for a woman who has been pregnant three times and given birth to viable babies twice) whereas she was in fact a G ₄P₃. However, there is a qualitative difference between the two ‘errors’: in the one case the defendant intended to record the plaintiff’s child-bearing history but made an error with the digits; in the other case, one would have to find that he inserted additional words which should not have been there at all. This is, I suppose, possible but it strikes me as most unlikely.

[d] Fourth, the defendant quite clearly arrived at theatre the next day under the impression that the plaintiff was there for a caesarean section and a tubal ligation. It was not suggested to him in cross-examination nor submitted to us in argument that the defendant performed the sterilisation knowing that the plaintiff did not want it. Apart from the absence of financial or other incentive, he would have known such an act to be grossly improper and one that would land him in trouble with the law and with his professional body. Even if it be supposed that an experienced gynaecologist could in consultation absent-mindedly write a referral letter incorrectly stating that his patient was to be admitted for a tubal ligation when she had moments ago clearly said no, that could not on the plaintiff’s version have been an error as to what was actually in his mind. On the plaintiff’s version, if the defendant had been asked by a hypothetical bystander at the end of the consultation whether the plaintiff had consented to the sterilisation his answer would have been an unequivocal ‘no’. It is difficult, then, to see how he could have arrived at the theatre barely 24 hours later under a different impression. Yet quite obviously he was under the impression on 4 November 2004 that she had consented.

52. It might be said that an assessment of the inherent probabilities must also take into account that as a fact the plaintiff did on the following day require the hospital’s consent form to be amended to delete reference to sterilisation. This might be said to point to a conclusion that this was also her attitude the previous day. However, if the matter of sterilisation was first raised (as the defendant testified) on 3 November 2004, it is not far-fetched that the plaintiff may have reconsidered the matter overnight. The procedure was not a medical necessity. Sterilisation is a very personal choice and conflicting emotions may come into play.

53. Another factor which might be said to favour the plaintiff’s version is that she admittedly did not tell the defendant in theatre on 3 November 2004 that she had changed her mind. This might be said to indicate that her attitude was in truth consistently against sterilisation and that she had no reason to believe the defendant thought otherwise. One possibility is that although she remained awake during the operation, the combined effects of the spinal anaesthetic and of the emotions and tensions associated with imminent surgery and child-birth caused her not to act in the logical way one might have expected. If she was still alert and thinking logically, it is on any reckoning peculiar that she said nothing to the defendant on his arrival in theatre. Even on her version of the consultation, she must have realised from her interactions with Samsodien regarding the consent form that there was some misunderstanding, because Samsodien clearly only filled out the consent form in the way she did because of the content of the defendant’s referral letter. A patient who was firmly against sterilisation but aware that her surgeon might be labouring under a misapprehension would ordinarily be anxious to bring this to his attention. If, on the other hand, the performance of the sterilisation raised conflicting emotions within her she may have said nothing. I do not think it is possible to reach any firm conclusion. I regard her failure to raise the matter with the defendant in theatre as a puzzling but neutral consideration.

54. Although the inherent probabilities in my opinion point in favour of the defendant’s version that on 3 November 2004 the plaintiff orally consented to the sterilisation, other aspects relevant to credibility must also be assessed. The trial judge considered that the defendant’s credibility was undermined by supposed contradictions between his letter of 4 June 2007 and his oral testimony. As will appear, I do not regard the supposed contradictions as being of much moment. It should be remembered that the defendant was writing the letter more than 2½ years after having performed the operation, with no intervening complaint from or contact with the plaintiff. He was relying on his memory and his own consultation notes – he had not yet seen the hospital records. The matters mentioned by the trial judge in relation to this letter are the following:

[a] In the letter the defendant said that the plaintiff was accompanied to the consultation by her ‘husband’. In his oral testimony he said that the plaintiff was accompanied by a man whom he saw that day for the first time and whom she did not introduce. His explanation in cross-examination was that after the mistaken performance of the sterilisation came to light Venter and another nurse told him that the man was the plaintiff’s husband. Prinsloo, it will be recalled, was at the hospital on 4 November 2004 and met the nurses. There is nothing implausible about this explanation nor was it shown how it would have advanced the defendant’s case falsely to claim that Prinsloo was not introduced to him at the consultation on 3 November 2004.

[b] In the letter the defendant referred to the date of the operation as 4 October 2004 whereas the correct date was 4 November 2004. This was an obvious typographical error since the defendant correctly stated in the letter that the last consultation was on 3 November 2004 and that the operation was performed the next day.

[c] In the letter the defendant said that when he came to the point of performing the tubal ligation he asked the scrub sister ‘whether I am proceeding with the sterilisation’ and she replied ‘yes’. According to the trial judge, the defendant’s oral testimony was that before starting the tubal ligation he asked the scrub sister ‘whether or not consent for tubal ligation was obtained’. Even if the trial judge’s recordal of the evidence were correct, it would not be a material discrepancy, because inherent in the former version of the question is the matter of consent – there would be no other purpose for asking it. But as a fact the defendant did not in his oral testimony say that he formulated his question to Venter by referring to the obtaining of consent – his oral evidence was consistent with the letter [see 13/1232 and 13/1262-4].

[d] In the letter the defendant stated that he learnt of the fact that the sterilisation had been erroneously performed from Venter while he was attending to the plaintiff in the recovery room where she was dozing off. In his oral evidence he testified that Venter informed him of the error while he was in the change room or as he was coming out of the change room and that he then followed her to the recovery room. This seems to me to be a minor difference of detail. It is true that in his letter he did not add that in the recovery room he spoke with the plaintiff by expressing regret and saying he would need to find out where things had gone awry. He explained in cross-examination that his letter was not intended to set out every detail. Perhaps his memory on this aspect was only jogged during subsequent consultation. It seems inherently likely that upon learning of the error he would have expressed regret though if the plaintiff was dozing off she may not have heard it or may have forgotten.

[e] In the letter the defendant said that due to other emergencies he had not seen the defendant on Sunday 7 November 2004 prior to her discharge. In the event, the hospital records reflected that the defendant saw the plaintiff at 11h00 on that day, about an hour before she discharged herself (he had intended she should only be discharged on the Monday). The defendant explained that when he wrote his letter he had not seen the hospital records and that he did indeed have emergencies over that weekend. During his evidence in chief he was taken to the Nursing Progress Report and could then recognise the instructions recorded there by the nurse as having emanated from him. There is nothing about this which reflects adversely on his credibility.

55. Before dealing with the defendant’s consultation notes, it is convenient here to mention that the plaintiff’s own evidence was by no means unblemished. In general, my impression from the transcript is that of an intelligent person, deft at sparring with her cross-examiner and with the undoubted ability to put the best gloss on her case. Among the points of criticism are the following:

[a] In cross-examination the plaintiff stated that the pregnancy was planned as a last attempt to save the marriage [4/399-401]. This is at odds with what she told Lewis, namely that although she only divorced Prinsloo at the end of 2005 ‘she had planned before this to get divorced from him, but then discovered that she was pregnant with her youngest child’ [3/224; see also 8/715, 720-721 and 754-7]. This latter version is more consistent with her evidence in chief, given in a different context, that she was on contraception prior to the pregnancy [4/351]. The suspicion arises that her evidence that the pregnancy was planned was intended to neutralise a contention that she had decided to have no further children.

[b] Although she said that the pregnancy was seen by her and Prinsloo as a last chance to make their marriage work, she testified that as at November 2004 she did not intend to have any more children with Prinsloo but that she would still have wanted to be able to have children with someone else who could make her happy [4/392]. It strikes one as strange, this being the case, that on 3 November 2004 she would – as she claimed at one point in her evidence – have sought Prinsloo’s view at the consultation on the question of sterilisation.

[c] She was adamant that her first consultation with the defendant was a consultation concerning her pregnancy. It is clear, though, from the defendant’s notes of the first consultation on 9 March 2004 and from her GP’s referral letter that she saw him on that date for other reasons and was not yet known to be pregnant. She came across as rather obstreperous on this aspect [5/416-428].

[d] She initially testified that when the question of sterilisation was discussed (on three occasions, according to her) the defendant did not ask whether she intended to have more children [5/461] and did not mention any risks associated with further pregnancy, instead advising that she could have two or three more caesarean sections because her skin was resilient and healed quickly [4/352-3; 4/403-11]. It seems most unlikely that the defendant would not have mentioned the risks. The expert evidence shows that the subject of sterilisation would only have been raised by him because of the increased risks of complications associated with further pregnancies following a third caesarean section. It would also have been standard practice in the context of a discussion on sterilisation to ask whether the patient intended to have more children. Later in cross-examination the plaintiff conceded that the defendant did explain certain risks to her but she insisted that he told her that she herself would not face these risks [5/464-5]. The defendant denied that as an experienced gynaecologist he would ever have expressed the view that a patient was not at risk because of her healthy skin [13/1226], and the plaintiff’s own expert, Dr Rosemann, testified that he would not expect a gynaecologist to express such an opinion [9/849-50].

[e] There is some inconsistency in the plaintiff’s version as to how her husband was brought into the conversation about sterilisation. She did not in her evidence in chief mention his involvement in the discussion or decision. At one stage in cross-examination she testified that when the defendant raised the matter she turned to her husband and said that the defendant was asking about sterilisation – in other words, she sought his view [5/439-444]. Later she said that Prinsloo overheard the discussion about sterilisation and intervened to ask what it was about because this was the first time he was hearing about it [5/446; 5/462-3].

[f] According to the plaintiff, the question the defendant asked her on 3 November 2004 was whether she was still having the sterilisation, which made her irritable because she had already said no on prior occasions [5/445]. The question she attributed to the defendant requires one to accept not only that he was repeatedly raising the matter but also that he erred on 3 November 2004 by implying that on the prior occasions she had expressed a wish to be sterilised. This strikes one as an implausible scenario. I may add that the defendant’s consultation notes for the period prior to 3 November 2004, the authenticity of which has not been questioned, made no mention of any discussion of sterilisation.

[h] The defendant’s evidence was that initially the intention was to perform the caesarean section in the next week but that the plaintiff was feeling uncomfortable and wanted to have it sooner. He said he phoned the hospital in the presence of the plaintiff and Prinsloo on 3 November 2004 and ascertained that there was a slot for a caesarean section and tubal ligation at 13h15 the following day. This, according to him, is when he obtained the time mentioned in his referral letter. The plaintiff’s evidence in cross-examination on this aspect does not read particularly well [6/514-529]. She denied that the defendant himself phoned the hospital, claiming that he buzzed his secretary and asked her to book the plaintiff into the hospital [5/469-70; 6/520-3]; but she also said that the sealed referral letter (which we know specified the time of the operation) had already been given to her by the time the defendant buzzed his secretary [6/517].

[i] She maintained that the operation was performed on a Friday and that she was discharged on Monday. She claimed to remember having asked the defendant why the operation was being done on a Friday, given that it was mosque time. She conceded in cross-examination, however, that the calendar showed that the date of the operation (4 November 2004) was a Thursday and that the date of discharge (7 November 2004) was a Sunday.

[j] Dr Rosemann stated in his third report that according to the plaintiff the admissions nurse (presumably Samsodien) opened the referral letter in the plaintiff’s presence and said to her that she was booked in for a caesarean section and tubal ligation [3/212]. He confirmed in cross-examination that the plaintiff told him this [9/894-5; 9/899-900]. In her oral testimony the plaintiff first said that she handed the sealed letter to the admissions nurse who went off with it to prepare the consent form. In cross-examination she said she could not remember if the letter was opened in her presence [6/544-5] but later became quite definite that it was not [6/567; 6/573-5].

[k] Although she claimed not to have heard the term ‘tubal ligation’ from the defendant on 3 November 2004 (or before), her statements to Dr Rosemann and her evidence as to the scratching out of this term from the form on 4 November 2004 tended to indicate knowledge of the term prior to her arrival at the hospital.

[l] She was, as I have mentioned before, vague and inconsistent as to the date of her divorce from Prinsloo and as to the beginning, end and duration of her relationship with Davids.

56. I do not suggest that each of these features is individually of great moment but overall I do not believe that a careful consideration of the record justifies a conclusion (leaving aside the inherent probabilities) that the plaintiff was a palpably credible witness while the defendant was not. One must also bear in mind that a rejection of a witness’ testimony on some aspects does not necessarily lead to a rejection of the witness’ evidence on all material issues.

57. Turning to the defendant’s consultation notes (which cover the period 9 March to 8 November 2004), a question mark hangs over the authenticity of the notes for the period as from 3 November 2004. The notes for 9 March to 20 October exist in one form only; they were written on both sides of two pages of thin paper, being the defendant’s ordinary paper for consultation notes. Their authenticity is not questioned. However, two versions exist of the notes entered against the dates 3, 4 and 8 November 2004: at 2/134-5 (version A) and 3/292-3 (version B). Apart from the fact that the layout of the writing in the two versions differs, the texts differ in the following respects: version B has the additional words ‘also L/W’ (‘also labour ward’) at the end of the note for 3 November; and version A’s note against the date 8 November has at the end an additional commentary by the defendant as to why he raised the issue of sterilisation on 3 November and as to what she should have done on 4 November if she had changed her mind.

58. Even on the defendant’s version, the notes written by him against the dates 4 and 8 November must have been written after he learned of the error in sterilisation (he says he was told of the problem shortly after completing the procedure and before he would have had a chance to write these particular notes). However, his evidence was that his note for 3 November was written at the end of the consultation (ie virtually contemporaneously). He testified that the note he wrote on that day was version B; that after he faxed a copy thereof to his attorney (this would have been several years later), the latter said it had not come through clearly; that he then rewrote the note on thicker paper and faxed this to his attorney because he thought the poor fax quality was attributable to the thin paper of the original; and that version A was the rewritten notes. In cross-examination it was put to him that his explanation was false, that version A was the original and that version B was a rewritten version; and that it was not rewritten for the purpose claimed by the defendant but to present it as if it were the original note.

59. The significance of this is the following. In both versions the note for 3 November includes the words: ‘For C/S + T/L mane’ (medical shorthand meaning ‘For caesarean section and tubal ligation tomorrow’). However in version A the text ‘+ T/L’ appears to be squeezed in between ‘C/S’ and ‘mane’ whereas in version B the elements of the phrase are regularly spaced. The defendant’s evidence was that when he prepared the rewritten version (which he said is version A) he must have initially omitted the text ‘+ T/L’ and then inserted it before sending it to his attorney (though I do not read his evidence as claiming that he distinctly remembers this); and that he must also have mistakenly omitted to rewrite the text ‘also L/W’ which appears in version B. It was put to him in cross-examination that version A was the original version of the note and that in contemporaneous form the note for 3 November ended with the text ‘For C/S mane’; and that everything else thereafter was written after he already knew of the problem and that he then also inserted the text ‘+ T/L’ to make it look as if the plaintiff had consented to the tubal ligation when this was not in truth the case.

60. In my view the probabilities are firmly against the defendant’s explanation, and I would reject it:

[a] He said his attorney asked for a clearer version of his notes. If that were true, it must have been obvious to him that what the attorney wanted was a better copy of his actual note, not a rewritten version. He could have had the original or a good photocopy delivered to his attorney; in the meanwhile he could, if necessary, have arranged for his secretary to type a transcript of the note so that his attorney could see exactly what he had written. I cannot accept that an experienced professional person would imagine that his attorney was wanting a non-contemporaneous rewritten version of the notes. And since the defendant claimed to have had to find time between consultations to produce the rewritten version (hence his supposed errors) it is doubly unlikely that he would have followed this cumbersome procedure in the first place.

[b] If the attorney wanted a better version of the notes, why did the defendant only rewrite the notes for 3, 4 and 8 November? There is nothing to show that the fax quality of the earlier pages would have been better – they were all written on the same thin paper. It is suspicious that only the two controversial pages were rewritten.

[c] If version A is the rewritten version, it is most peculiar that it is more untidy and less legible than the supposed original, version B. For example, the jotting down of the blood pressure reading in version A is indecipherable (except no doubt to the defendant) but is reasonably legible in version B. In version A the defendant’s writing of his observation ‘urine clear’ is rendered as an abbreviation ‘Ur Cl’ or perhaps as an extremely rapid and illegible rendition of the full words, and is in keeping with the way he recorded the same observation in five of his earlier and admittedly authentic notes; whereas in version B the full phrase is written out in legible form. It would be odd if a version rewritten for the benefit of the attorney were less legible than the original (even if the writing came out more clearly in a fax). These considerations point to a conclusion that version A is the original.

[d] There is also force in Mr Bhoopchand’s submission that the concluding paragraph of the note purportedly made on 3 November is the first entry in all of the defendant’s notes in narrative style rather than his usual abbreviated and epigrammatic manner. This paragraph has a defensive quality about it, something written after the defendant knew there was a controversy.

[e] A further suspicious element is that whereas the defendant was able to produce the original of the first two double-sided pages of the notes (covering the period 9 March to 20 October) and of the double-sided page of version B, all written on his thin consultation note paper, he was not able to produce the original of version A, stating that it had been mislaid by one or other of his secretaries. The importance of this is that he claimed that version A was produced by him on thicker paper to improve the fax quality. He could have proved this by producing the original of version A. But if the original of version A was, like the original of version B, written on his usual thin paper his explanation would have been exposed as undoubtedly untrue.

[f] I also find it most implausible that in a rewritten version prepared for his attorney he would, when initially rewriting it, mistakenly have omitted what was, in context, the most crucial element of the note for 3 November, namely ‘+ T/L’.

[g] There was also inconsistency in the defendant’s testimony as to who had the idea of rewriting the note. He initially testified that the suggestion was his secretary’s [13/1212] yet later he asserted that he was the one who came up with the idea [14/1361]. (He did not call his secretaries as witnesses.)

[h] I am not particularly impressed by the argument on behalf of the defendant that his innocent explanation is established by the fact that both versions of the notes were discovered. If the defendant’s attorney received both versions, he would have been under a professional duty to include both of them in the discovery affidavit prepared for the defendant. The latter may not have appreciated that this would happen when he faxed the one version and then the other to his attorney.

61. For these reasons I consider that Mr Bhoopchand’s submission must be accepted that the authentic contemporaneous version of the note for 3 November is version A; that it ended with the text ‘For C/S mane’; and that the insertion ‘+ T/L’ and the concluding paragraph of the note for that date were, like the notes for 4 and 8 November, written after the defendant knew there was an issue about the performance of the sterilisation. In particular, I think it probable that after the controversy came to light he inserted ‘+ T/L’ into his original consultation note (version A) and that a copy of this version was probably sent to his attorney; but that when he later realised that the production of the original would reveal the insertion even more obviously than the copy, he rewrote the note as version B.

62. This conclusion naturally reflects adversely on the defendant’s credibility. But there are two possible explanations for his discreditable behaviour: that he wanted it to appear that the plaintiff had orally consented to tubal ligation when this was not the case; or that although she did consent he was embarrassed subsequently to find, when the problem arose, that his notes did not reflect this. It is by no means unknown for a witness who has truth on his side to attempt falsely to improve his case.

63. On balance, the inherent probabilities discussed earlier point to the second of these explanations. The referral letter, which was also contemporaneous and quite possibly written before the consultation note, expressly mentions tubal ligation. This letter and the defendant’s conduct in performing the sterilisation the next day are compatible with the second of the explanations for the alteration of the notes but not with the first.

64. Even if the probabilities were equipoised, the onus on this question in so far as it bears on the defendant’s alleged negligence (as distinct from justification of prima facie wrongfulness) rested on the plaintiff. I thus proceed on the basis that the plaintiff orally consented to the sterilisation on 3 November 2004.

65. The events surrounding the plaintiff’s admission the next day were the subject of conflicting evidence by the plaintiff and Du Plessis. It is common cause that before the plaintiff was transferred from the maternity ward to the labour ward she had decided not to have the sterilisation and that the consent form had been altered to reflect this. The main point of controversy is whether she originally signed the unamended consent form and only later changed her mind or whether from the outset she refused to sign it in unamended form.

66. I do not find it inherently unlikely that overnight the plaintiff changed her mind and arrived at the hospital intending not to have the sterilisation. Of course, her version was that she had at no stage consented to the sterilisation and that there was no change of mind. I have rejected that version, so a finding that she changed her mind is inevitable. I think it less likely that the change of heart occurred within the relatively short space of time which would, on Du Plessis’ version, have passed between the signing of the unamended form and the initialling of the changes. The plaintiff was administratively admitted on the 8^th floor at 10h56 but there is no indication that the consent form was presented to her at that stage. It was in the maternity ward on the 11^th floor that she dealt with Samsodien. By 11h30 Samsodien had recorded in the Nursing Progress Report that the plaintiff had been admitted to the maternity ward and that she and her husband had ‘verbalised’ that she did not want a tubal ligation. The same note records certain medical observations made in the maternity ward. The note was likely to have been made after the interactions concerning consent form.

67. Although a version was put to the plaintiff as to what Samsodien would say, the defendant did not call her as a witness. There is thus nothing directly to refute the plaintiff’s evidence concerning her initial discussion with Samsodien. The latter’s note in the Nursing Progress Report does not hint at the signing of a an unamended consent form and a subsequent change of mind. Du Plessis claimed that when Samsodien came to her the unamended form had already been signed and that Du Plessis went back with Samsodien to the plaintiff at which point the initialled deletions were effected. The plaintiff denied that she signed anything in a second nurse’s presence. What tells against Du Plessis’ version is that although she claimed that she made crosses at the places where the plaintiff was to initial and was present when the initialling was done, the plaintiff as a fact only initialled against one of the three crosses made by Du Plessis. If Du Plessis was present, she would have made sure that the plaintiff signed in all three places. There is the further point that Du Plessis’ signature appears in full as the first witness at the foot of the form, and Samsodien’s as the second witness. This indicates that there were no witness signatures on the form when the matter was first reported to Du Plessis. That being so, Du Plessis’ version requires one to find that the plaintiff’s original signing of the unamended form was never witnessed, and that witness signatures were only added when she changed her mind. That strikes me as unlikely.

68. I would thus not disturb the trial judge’s credibility finding in favour of the plaintiff on this aspect. On the probabilities, I consider that the plaintiff declined from the outset to sign the unamended form and that the matter was reported by Samsodien to Du Plessis because of a conflict between the plaintiff’s attitude and the referral letter from the defendant. I should add, though, that I doubt that it matters when it comes to the wrongfulness of the performance of the sterilisation (though an acceptance of Du Plessis’ version would strengthen the conclusion that the plaintiff gave her oral consent on the previous day). If the plaintiff originally signed the unamended form, she clearly retracted that decision within a short space of time, well before her transfer to the labour ward. Both at common law and in terms of the Sterilisation Act a patient may retract consent. On any reckoning this occurred more than one and a half hours before the defendant arrived at the labour ward intending to see the plaintiff pre-operatively and about two hours before he began the sterilisation. No consent as required by law existed at that stage. But as I have said, my conclusion, insofar as it is relevant, is that there never existed a signed consent encompassing sterilisation.

69. Du Plessis said that she told Samsodien to tell the nurse in the labour ward (presumably Solomons) that the plaintiff had changed her mind. That seems likely, and the plaintiff in the nature of things could not dispute it. There is no evidence as to whether Samsodien complied with Du Plessis’ instruction.

70. Du Plessis also said that she told the plaintiff to tell the defendant about her change of mind. This again brings her into conflict with the plaintiff. A rejection of Du Plessis’ evidence regarding the events surrounding the signing of the consent form does not necessarily mean that her other evidence must be rejected. She might have seen the plaintiff being taken to the labour ward and told her at that stage, because Du Plessis by then would on any reckoning have known about the change. Nevertheless, I do not think there is a sufficient basis to reject what I take to be the trial judge’s implicit finding against Du Plessis on this aspect. Again, though, it would not affect my overall conclusion if in truth Du Plessis gave the alleged advice to the plaintiff.

71. As to what happened in and shortly before theatre, I accept that the hospital staff did not notify the defendant that the plaintiff had refused or retracted written consent to sterilisation. The defendant arrived too late to see the plaintiff pre-operatively in the labour ward. Had he arrived a little earlier, the matter may have come to light. I do not attach any particular significance to the defendant’s discussion with Solomons in the labour ward shortly after 13h10. The defendant on his version was simply getting clarity regarding the identity of the patient who had been taken into theatre – he was not asking Solomons whether the plaintiff had provided the necessary written consent to sterilisation.

72. In theatre the defendant admittedly did not himself check the plaintiff’s folder to see that her written consent was in place. He arrived in theatre before scrubbing, so he could without risk of contamination have done so.

73. The plaintiff admittedly did not tell the defendant in theatre that she no longer wanted the sterilisation. (This is, as I have previously remarked, on any reckoning a puzzling feature of the case.)

74. There was no evidence contradicting the plaintiff’s testimony as to what the junior theatre nurse said after looking at her folder or as to Venter’s conduct in holding up the bottle after completion of the sterilisation. I do not have reason to reject her evidence in these respects. Dr Whitehead, on the other hand, denied speaking to the plaintiff as alleged. Although the trial judge said that Dr Whitehead seemed to distance himself from anything to do with the matter, he did not specifically find that Dr Whitehead’ evidence was false. Having read the transcript, I cannot say that there is a basis for rejecting Dr Whitehead’s evidence. In the event, though, the remarks and conduct attributed by the plaintiff to the junior theatre nurse, Venter and Dr Whitehead are not of great significance to the outcome of this appeal. At most they are relevant to show that if someone had looked at the plaintiff’s folder in theatre, he or she would have noticed that the plaintiff had decided not to have the sterilisation. I think one can make that assumption even in the absence of evidence that somebody did actually look at the folder and see the amended consent form.

75. What is undoubtedly important is whether the defendant, as he testified, directed a query to Venter, at the end of the caesarean section, as to whether they were proceeding with the sterilisation and whether she answered in the affirmative. The plaintiff said that she did not hear such a discussion but she conceded that it may have occurred without her hearing it, and the circumstances of the operation make that a distinct possibility (the defendant and Venter were probably shielded from the plaintiff’s head by a screen, and the plaintiff had no particular reason to follow what were probably quiet remarks passing between medical personnel). If Venter knew the true facts, she would obviously not have answered the defendant’s question in the affirmative. However extraordinary her ignorance may seem, there are other circumstances to suggest that she was not aware that the plaintiff had rejected the sterilisation. Firstly, the defendant’s evidence was that he used a special silk thread for performing the sterilisation and that this was laid out in readiness for him and then handed to him with other necessary instruments when performing the sterilisation [13/1232-3]. Even if no question had been posed to her, Venter would not have failed to intervene if she knew the defendant was about to sterilise a patient contrary to the latter’s express wishes. There is the further circumstance that upon completion Venter showed the plaintiff the bottle containing the excised portions of her fallopian tubes, presumably because Venter thought the plaintiff would or might be interested to see the evidence of the procedure. This would have been be an implausibly callous act if Venter believed the plaintiff had not consented to the procedure.

76. The defendant’s evidence was that his practice was to rely on the scrub sister to check that the necessary written consent was in place. There was evidence from Dr Van Helsdingen, the defendant’s very experienced expert witness, that this was also his practice and that he did not know of any surgeons who themselves checked the written consent form. It thus seems not unlikely that the defendant would have obtained brief confirmation from the scrub sister.

77. Venter was deceased by the time of the trial so could not be called to confirm or refute the defendant’s evidence. Dr Whitehead could not recall hearing the exchange but said that his tasks would not have required him to attend to the discussion between the defendant and Venter. Possibly the evidence of the junior theatre nurse and Drs Kriel and Zieff could have cast light on the matter. On the assumption that such witnesses could more plausibly have been called by the defendant than the plaintiff, there is nevertheless insufficient reason to reject the defendant’s evidence, despite the fact that in my opinion his credibility on the question of his consultation notes was badly tarnished. His consultation note against the date 8 November 2004 records his query to the scrub sister. That note, if not written on 8 November 2004, was probably written fairly shortly afterwards. It seems that there was a preliminary investigation at the hospital on Monday 8 November 2004. It is doubtful that the defendant would have recorded in his note a version that he knew would be contradicted by the theatre nursing staff or his professional colleagues. It is also not irrelevant that the defendant repeated his version in his letter to the plaintiff’s then attorneys on 4 June 2007. Venter died more than two years later, and the defendant would thus have written his letter in the belief that Venter would be able to contradict his statement if it were untrue (and she would have an interest in doing so, since it reflected negatively on her and the hospital).

78. The trial judge did not make a specific factual finding as to whether the defendant directed the alleged query to Venter. In my view, and insofar as it bears on the question of negligence, I do not think the plaintiff established on a balance of probability that the defendant did not make the query or receive an affirmative answer.

79. The expert evidence established that a scrub sister in Venter’s position should in accordance with usual practice have checked the plaintiff’s folder on the latter’s arrival in theatre to confirm that the written consent was in place. Had she done so, the question the defendant asked her would have been sufficient to attract an answer that the plaintiff had decided not to have the sterilisation, and the defendant would then not have performed it. One is driven to conclude that Venter did not check the plaintiff’s folder. If (as the defendant said) the theatre slot was booked on 3 November 2004 and the booking specified that the patient was being admitted for a caesarean section and tubal ligation, and if Samsodien or the nursing staff in the labour ward failed to communicate the plaintiff’s changed decision to the theatre staff, Venter may have assumed that everything was proceeding as normal without bothering to look at the folder.

80. Venter was no longer alive to speak in her own defence. The hospital, to judge by what we were told by counsel, was defensive of its own position during the parties’ trial preparations. The hospital and its advisers were presumably aware of the version the defendant intended to advance. Their employee, Du Plessis, was called as a witness by the defendant. The hospital nevertheless seems not to have volunteered any evidence to the plaintiff to refute the defendant’s version. I think one must conclude, at least for purposes of this case, that one or more of the hospital’s nurses were negligent in failing to ensure that the plaintiff’s decision was not brought to the defendant’s attention. That was the view of the plaintiff’ expert, Dr Rosemann, in his first report and also the view of Dr Van Helsdingen.

The defendant’s liability

81. On the basis of these factual findings, is the defendant liable to the plaintiff for any damages suffered in consequence of the performance of the sterilisation? I have already expressed the view that a lawful consent to sterilisation must be in writing as required by the Sterilisation Act. No such written consent was obtained in this case. The prima facie conclusion that the performance of the sterilisation was wrongful was not negatived by consent as required by law.

82. As to causation, this is not a case where, at the time the operation was performed, the patient was intending to undergo a sterilisation after having given fully informed oral consent, with the absence of written consent a mere technical oversight. The plaintiff did not by that time wish to be sterilised and had refused to sign the required written consent. I thus consider that the wrongful sterilisation caused whatever damages she suffered.

83. This leaves negligence. The evidence did not establish that it was impossible or even practically difficult for a gynaecological surgeon to take a written consent in his or her rooms (whether on the day of the operation or at an earlier consultation) or to check pre-operatively that the written consent was in place (either in the ward or on arrival in theatre). Indeed, there was evidence that since 2006, and quite possibly because of what happened in this very case, the hospital group of which the CMBH forms part has required surgeons using its facilities to take their own written consents. Dr van Helsdingen said that his practice was to see his patient pre-operatively in the ward (though such visits were not for the specific purpose of checking consent). In this very case the defendant intended to see the plaintiff in the ward but arrived there too late. Nevertheless, the fact that the defendant could have personally checked whether a written consent was in place does not mean that he is liable if the obtaining of written consent was properly a function of the hospital staff and if he is not liable for their negligence. If he could, without personal responsibility, leave it to the hospital staff to take the written consent and to tell him if the patient refused to sign the consent, it might suffice for him briefly to have confirmed with the scrub sister that they were still proceeding with the sterilisation, since her affirmative answer would bring him under the impression that the required written consent was in place. Indeed, even a query to the scrub sister may not have been necessary if he could properly rely on the hospital staff to bring any problem with written consent to his attention.

84. The expert evidence as to a gynaecological surgeon’s duties regarding the obtaining of written consent was not entirely satisfactory. As I have said, the defendant’s expert, Dr van Helsdingen, testified that it has never been his practice to take or personally check written consents and he was not aware of any of his colleagues who did so. The taking of a written consent in his view was a hospital function. If a patient changed her mind after giving the gynaecologist informed oral consent, the hospital staff, in particular the scrub sister, should bring this to the gynaecologist’s attention.

85. Dr van Helsdingen’s evidence as to his own practice and as to the practice of his fellow professionals is not altogether consistent with a work entitled Basic Principles of Gynaecological and Obstetrical Surgery which is used inter alia in the teaching of medical students in South Africa and to which Dr van Helsdingen himself contributed a chapter. In that chapter there is a step-by-step summary of the procedures to be followed in the ward on the day of an operation. Step 6 is ‘Check the consent form’ [record 3/336]. Dr van Helsdingen’s attempts to explain this part of the book in cross-examination [at 11/1047-66] were unsatisfactory, and he was somewhat evasive and argumentative. For example, he said the book ‘carries no legal power’ and was ‘merely a manual of suggestions’. He accepted that the book nevertheless represented good practice. He also said it was good practice for the gynaecologist to visit the patient in the ward prior to the operation and that there would then be opportunity to look at the patient’s folder. But when pressed with the passage in the book, he said it was ‘written in abbreviation’ and denied that it contradicted the practice he followed. He said that his cross-examiner did not know ‘how these things work’. He insisted that it was not practical for the surgeon to obtain the written consent ‘as he stands next to the patient’, that there may be emergencies and that it depends on considerations of timing. This was all beside the point. It was not being suggested to him that the gynaecologist should take a written consent while standing next to the patient in theatre or even that the gynaecologist should take the written consent himself nor was he being asked about the special case of an emergency – what was being put to him was that, as the book apparently indicated, the surgeon should check the consent form.

86. Nevertheless, Dr Van Helsdingen’s evidence as to the actual practice of most gynaecological surgeons was not disputed by Dr Rosemann. The latter said that he considered that the surgeon should check the consent form and that he himself did so, but he could not speak as to the practice of others. Widespread practice might thus differ from step 6 in the book. Given the summarised style of the six steps listed on the relevant page, it would not be right to place too much weight on it as evidence of appropriate professional standards. Although the book is directed at gynaecological and obstetrical surgeons, step 6 could perhaps mean that somebody must check the consent form, not necessarily the surgeon.

87. It is also fair to point out that Dr Rosemann’s evidence was not without blemish on this aspect. He testified that given the importance of a patient’s choice regarding sterilisation it was imperative for the surgeon to inspect the consent form, and Dr Rosemann said that he himself did so. He also said that the HPCSA guidelines in booklet 15 [record 1/75-6] so required. Although three reports and an addendum in respect of Dr Rosemann’s evidence were filed by the plaintiff, these views were not recorded in any of the reports or in the addendum. Dr Rosemann’s views as expressed in oral evidence were, furthermore, inconsistent with the conclusion he reached in his first report. At the time of writing his first report he assumed that the facts were as the defendant alleged – remarkably, he appears not as yet to have been given the plaintiff’s version. On the basis of the facts he assumed, Dr Rosemann said in his first report that the negligence lay with the hospital staff and that the defendant could not be held liable. If it was his consistent and firm opinion that the surgeon should always personally check the consent form, one would not have expected him – even on the defendant’s version of the facts – to exonerate the defendant in the first report. Initially in cross-examination Dr Rosemann reverted to the first report’s conclusion by conceding that if the facts were as the defendant alleged he could not be held responsible; but later in cross-examination, after an overnight adjournment, he said that he should have included reference in his first report to the surgeon’s duty of inspection and that this omission was attributable to the rush in which his first report was prepared. It was put to him with some justification that this was an afterthought. Under further lengthy questioning, some of it by the trial judge which in turn prompted renewed cross-examination, Dr Rosemann see-sawed between opposing opinions as to whether – if the defendant’s version of the facts were correct – the defendant could be faulted. The last of the views offered by Dr Rosemann was a concession that it was sufficient for the defendant to have asked the scrub sister about consent [10/980].

88. Dr Rosemann’s reliance on the HPCSA’s booklet 15 was also somewhat argumentative and misplaced. Para 3 of the booklet does not deal specifically with the written consent required for sterilisation but with informed consent in general. The paragraph refers to the need to obtain informed consent and states that the explanations to the patient and the obtaining of the consent can be designated by the doctor to a suitably qualified person but that the doctor will remain responsible for ensuring, before he starts any treatment, that the patient has been given sufficient time and information to make an informed decision and has given consent. Here the defendant himself explained the position to the plaintiff and obtained her oral consent on 3 November 2004. The question in the present case concerns a change of mind and the legal requirement for written consent. That is a question which the booklet appears to me not to address.

89. I may add, without attaching too much weight to it, that the regulations promulgated under the Sterilisation Act specify, as the prescribed consent form for sterilisation, the standard consent form for use in the health facility in question. It may be questioned whether this is altogether apt for the case where the surgeon is not a doctor employed by the hospital but nevertheless those responsible for framing the legal requirements for a lawful sterilisation consent appear to have taken for granted that the written component of the consent requirement would be attended to by the hospital where the surgery was to be performed. They presumably did so because that is how thing in general are done.

90. I thus consider that the defendant’s conduct on the day of the operation did not depart from the practice which prevailed among most professionals in his field as at November 2004 and that his profession would regard his behaviour as acceptable and reasonable. This does not in itself mean he was not negligent – that is a question of law on which the views of experts will be relevant but not necessarily decisive (see, eg, Michael & Another v Linksfield Park Clinic (Pty) Ltd & Another 2001 (3) SA 1188 (SCA) paras 36-40). Particularly on matters concerning consent as distinct from medical science, the views of the profession, even if unanimous, might properly be rejected by the court (cf Castell at 421D). Nevertheless, the onus was on the plaintiff to prove that the defendant was negligent. Given the expert evidence summarised above, I am not satisfied that the defendant’s conduct, which appears to have been in accordance with the standards of his profession, was in law negligent. He had performed the most significant component of obtaining informed consent by his oral explanations to the plaintiff the preceding day. The completion of that consent by its commitment to writing was, and would generally be, a mechanical exercise which could be left to others. There was nothing to alert the defendant to the fact that anything had changed or that the hospital staff had not done their job properly. Venter was known to him as an experienced theatre nurse. She had prepared the table for the performance not only of a caesarean section but also a tubal ligation. The defendant went further and checked with Venter before doing the tubal ligation that the sterilisation was still going ahead – a query which could have had no other purpose than to make sure there had been no change from the plaintiff’s side. If he asked the question in a perfunctory manner, that was only because he had no reason to think the plaintiff had undergone a change of heart. The plaintiff said nothing to him along those lines in theatre.

91. It is clear that the hospital staff were negligent in not communicating to the defendant that the plaintiff no longer wanted the sterilisation and had refused to sign the consent required for sterilisation. However, the plaintiff did not plead that the defendant was vicariously liable for the hospital staff’s negligence. The trial judge does not seem to have been asked to consider the case along those lines; nor were we. There is no doubt that vicarious liability must be pleaded. In the present case the plaintiff’s pleaded case was that the defendant himself had been negligent. A cause of action based on vicarious liability would have required the plaintiff to allege the negligence of one or more of the hospital staff; to allege a relationship between himself and the hospital or its staff which in law could give rise to vicarious liability; and to allege that their negligent conduct was performed within the scope or course of their relationship with or authority from him (and cf Harms Amler’s Precedents of Pleadings 7t Ed at 390; Stadsraad van Pretoria v Pretoria Pools 1990 (1) SA 1005 (T) at 1007H). It is possible that if vicarious liability had been pleaded the defendant may have wished to adduce additional evidence or that he would have taken steps to join the hospital as a third party. Furthermore, the legal questions that would arise in relation to vicarious liable in this setting are by no means straightforward, since one is not dealing with a classic relationship of vicarious liability established by existing authority.

Conclusion

92. I thus consider that the defendant should not have been held liable in the court a quo. This conclusion makes it unnecessary to consider the question of damages. I would uphold the appeal with costs and substitute for the trial judge’s order an order dismissing the plaintiff’s action with costs, such costs to include the reasonable qualifying expenses of Dr van Helsdingen.

BLIGNAULT J

93. I concur in the judgment of Rogers J.

94. It would not be fair to conclude this matter without reverting to the trial judge’s very critical remarks about the defendant, in which inter alia he branded the defendant an ‘outright liar’. On appeal we have found that in one respect the defendant acted discreditably – that is in relation to the two versions of his consultation notes. In the main, though, we have accepted his evidence and it is on the basis of such evidence that his appeal succeeds. To describe him as an ‘outright liar’ is obviously not in keeping with the findings in this court’s judgment. I venture to suggest that in civil matters, where factual disputes often have to be resolved with reference to inherent probabilities, caution should be shown in expressing credibility findings in such strong terms. The fact that on balance one factual version is preferred over another does not normally justify leaving the losing party with the stigma of having been labelled by a court as an outright liar.

95. The appeal is allowed with costs. The order of the court a quo is set aside and replaced with an order as follows: ‘The plaintiff’s action is dismissed with costs, such costs to include the reasonable qualifying and attendance costs of the defendant’s expert Dr van Helsdingen.’ 

SALDHANA J:

96. I concur in the judgments of Rogers J and Blignault J.

______________________

BLIGNAULT J

______________________

SALDHANA J

______________________

ROGERS J

APPEARANCES

For Appellant: Mr MA Albertus SC

Instructed by:

Albertus Attorneys

148 Lansdowne Road

Cape Town

For Respondent: Mr Ajay Bhoopchand

Instructed by:

De Klerk & Van Gend Inc

Cape Town

---