DOCUMENTS

ConCourt judgment upholding Dudley Lee's appeal

In a split decision Nkabinde J rules that DCS must pay damages after prisoner contracts TB

CONSTITUTIONAL COURT OF SOUTH AFRICA

Case CCT 20/12

[2012] ZACC 30

In the matter between:

DUDLEY LEE - Applicant

and

MINISTER FOR CORRECTIONAL SERVICES - Respondent

and

TREATMENT ACTION CAMPAIGN

WITS JUSTICE PROJECT

CENTRE FOR APPLIED LEGAL STUDIES - Amici Curiae

Heard on: 28 August 2012

Decided on: 11 December 2012

JUDGMENT

NKABINDE J (Moseneke DCJ, Froneman J, Jafta J and Van der Westhuizen J concurring):

Introduction

Before this Court is an application for leave to appeal against a decision of the Supreme Court of Appeal[1] overturning a decision of the Western Cape High Court, Cape Town[2] (High Court). The High Court declared the respondent liable for the delictual damages suffered by the applicant as a result of contracting tuberculosis (TB) while in detention. Having rejected the applicant's claim on a narrow factual point on the application of the test for causation, the Supreme Court of Appeal upheld the respondent's appeal and absolved her from liability.

[1] Primarily, the case concerns whether the applicant's detention and the systemic failure to take preventative and precautionary measures by the Correctional Services authorities caused the applicant to be infected with TB while in detention. The complaint is that the unlawful detention and specific omissions violated the applicant's right to freedom and security of the person and the right to be detained under conditions consistent with human dignity, and to be provided with adequate accommodation, nutrition and medical treatment at state expense.[3] The question is whether the causation aspect of the common law test for delictual liability was established and, if not, whether the common law needs to be developed to prevent an unjust outcome. 

[2] The relief sought in this Court is essentially an order:

(a) granting leave to appeal, upholding the appeal and setting aside the order of the Supreme Court of Appeal; and

(b) reinstating the High Court order.

In the alternative to (a) and (b) above, the applicant seeks an order allowing him to amend his particulars of claim to include a claim for constitutional damages based on alleged unlawful detention and infringement of certain constitutional rights.

Parties

[3] The applicant is Mr Dudley Lee (Mr Lee or plaintiff). The respondent is the Minister for Correctional Services (respondent or responsible authorities),[4] who is cited in her capacity as the Minister responsible for the conduct of the Department of Correctional Services pursuant to certain provisions of the Correctional Services Act.[5] The Treatment Action Campaign, Wits Justice Project and Centre for Applied Legal Studies, jointly represented by SECTION 27, whose written submissions have been most helpful and for which we are grateful, were admitted as friends of the court (amici).

Facts

[4] The facts are set out elegantly in the judgment of the Supreme Court of Appeal. I repeat those that are necessary for the purpose of this judgment and set out also the parties' agreed statement of factual findings with specific reference to portions in the High Court Judgment and the Supreme Court of Appeal Judgment.[6] 

[5] The applicant was incarcerated in the admission section at the maximum security prison at Pollsmoor[7] (Pollsmoor) from 1999 to 2004,[8] but was released on bail for a period of approximately two months in 2000. He attended court on no fewer than 70 occasions. When inmates were transported for court attendance, they were stuffed into vans like sardines. At court they were placed into cells which were jam-packed. Those who appeared before the regional court were taken to a separate, smaller cell which was not overly full.[9]

[6] For most of his incarceration Mr Lee was housed in E-section of the maximum security prison at Pollsmoor, a cell designed for occupation by one person but which he shared with two other inmates.[10] At one stage, inmates at E-Section, including the applicant, were moved to the Medium B prison where they were detained in a communal cell with about 25 inmates for a period of time.[11] On being moved back to E-Section the applicant was held in a communal cell until he was placed in a single cell again.[12]

[7] The following appeared from the Statement of Agreed Factual Findings: the applicant was not infected with TB when he arrived at Pollsmoor; the responsible authorities were "pertinently aware of the risk" of inmates contracting TB;[13] TB is an airborne communicable disease which spreads easily especially in confined, poorly ventilated and overcrowded environments; Pollsmoor is notoriously congested and inmates are confined to close contact for as much as 23 hours every day - this providing ideal conditions for transmission;[14] on occasion, the lock-up total was as much as 3052 inmates and single cells regularly housed three inmates; communal cells were filled with double and sometimes triple bunks;[15] the responsible authorities relied on a system of inmates self reporting their symptoms upon admission to the prison and during incarceration; and the control of TB at Pollsmoor depends upon effective screening of incoming inmates, the isolation of infectious patients and the proper administration of the necessary medication over the prescribed period of time.[16]

[8] The Standing Correctional Orders[17] (SCOs), as summarised by the High Court, require inmates to be subjected to the effective screening as set out in clauses 4.1(a),[18] 4.4(a),[19] 6.1,[20] 6.2,[21] 14 and 15[22] of Chapter 3 of the SCOs. 

[9] During his incarceration the applicant regularly underwent sputum tests, the results of which were negative until June 2003.[23] He was diagnosed with TB after three years of his incarceration.[24] Despite this diagnosis and the possibility that he would remain contagious for at least another two weeks, the applicant was returned to his cell where he was confined for up to 23 hours with at least one other person. After his release in 2004, the applicant instituted an action for damages against the respondent in the High Court.

High Court

[10] At the commencement of the trial the parties asked the High Court to separate the issues relating to liability from those relating to the quantum of damages. The issue relating to liability was decided first, hence the declaratory order,[25] while quantum stood over for later determination.[26]

[11] It is important to mention, with reference to the pleadings, how the issues were defined by the parties in the High Court. The applicant's damages claim is said to have arisen as a result of the responsible authorities' negligent conduct, alternatively dolus eventualis because they knew that their conduct placed inmates, including the applicant, at risk of TB infection.[27] It is pleaded that but for that unlawful conduct on the part of the responsible authorities, the applicant would not have been exposed to inmates who were actively infected with TB and further would have been treated and cured earlier.[28] In the premises, the applicant pleaded that it was the conduct of the responsible authorities which caused his active infection with TB.[29]

[12] The pleadings establish that the applicant was imprisoned at Pollsmoor and that during the period of his imprisonment the responsible authorities failed to take adequate, or any, steps to protect him against the risk of TB infection; failed, once he was diagnosed as actively infected with TB, to provide him with adequate medical treatment and medication to cure and prevent further spread and to adhere to his numerous requests for adequate treatment of TB. The said conduct and omissions, it is pleaded, thus violated the applicant's rights, including:

(a) under the common law, specifically his right to respect for and protection of his physical integrity;

(b) in terms of the Bill of Rights, specifically the invasion of the rights to:

(i) human dignity under section 10;[30]

(ii) life under section 11;[31]

(iii) freedom and security of the person under section 12(1);[32] and

(iv) be detained in conditions that are consistent with human dignity, including at least exercise and the provision, at state expense, of adequate accommodation, nutrition, medical treatment in section 35(2)(e);[33] and

(c) section 2(a) and (b) and section 12 under the Act[34] and its regulations.

[13] The High Court upheld the claim. It ruled in the applicant's favour regarding the evidence pertaining to the break-down of the health care system at Pollsmoor and the inadequacy of nutrition which played a role in the development and uncontrollable spread of TB during the applicant's incarceration. It held that the responsible authorities' omission(s) constituted a negligent breach of its constitutional and statutory duty to protect the applicant's rights.[35] The High Court held that the evidence tendered established that TB could be curtailed by introducing certain measures including: (i) early identification of persons who are deteriorating and who may accordingly become vulnerable to contracting TB; (ii) early diagnosis of the disease; and (iii) effective treatment and proper nutrition.

[14] The High Court said that those measures would translate into-

(a) the proper screening of incoming inmates, inclusive of a physical chest examination;

(b) separating those who had or were suspected of having TB or were obviously undernourished and vulnerable to TB;

(c) the provision of adequate nutrition to those who were undernourished and otherwise vulnerable to TB;

(d) regular and effective screening of the prisoner population, inclusive of examinations by means of x-rays and/or physical chest examinations by means of a stethoscope, to identify possible TB infection; and

(e) isolation of infectious inmates and effective implementation of the DOTS[36] system over the prescribed period of time.

[15] These measures, the High Court said, were dependent on sufficient numbers of nursing staff and doctors available to perform the various tasks. However, the shortage of nursing staff had been a major problem at Pollsmoor. The High Court concluded that-

"as staff shortages remained a problem throughout the time of the [applicant's] incarceration . . . a reasonable person in the [respondent's] position would have realised that adequate staffing was the key to the prevention and control of TB and would have taken steps to ameliorate the staff shortages . . . [and] would have isolated all of the persons who were in the infectious stage of TB."[37] 

The Court declared the respondent liable to the applicant in delict and ordered her to pay costs.

Supreme Court of Appeal

[16] On appeal, the Supreme Court of Appeal confirmed the High Court's findings regarding the responsible authorities' failure to have taken reasonably adequate precautions against contagion and held that such failure ought indeed to be categorised as wrongful.[38] It held that:

"A person who is imprisoned is delivered into the absolute power of the state and loses his or her autonomy. A civilised and humane society demands that when the state takes away the autonomy of an individual by imprisonment it must assume the obligation to see to the physical welfare of its prisoner. We are such a society and we recognise that obligation in various legal instruments. One is s 12(1) of the [Act], which obliges the prison authorities to ‘provide, within its available resources, adequate health care services, based on the principles of primary care, in order to allow every inmate [of a prison] to lead a healthy life'. The obligation is also inherent in the right given to all prisoners by s 35(2)(e) of the Constitution to ‘conditions of detention that are consistent with human dignity'."[39]

It went further and held that:

"Prisoners are amongst the most vulnerable in our society to the failure of the state to meet its constitutional and statutory obligations. It seems to me that there is every reason why the law should recognise a claim for damages to vindicate their rights. To find otherwise would altogether negate those rights."[40]

[17] Regarding the question of negligence, the Supreme Court of Appeal relied on the classic test expressed in Kruger v Coetzee.[41] In that case the Court heldthat negligence arises if a reasonable person: (a) would have foreseen the reasonable possibility of his conduct injuring another person and causing him harm; (b) would have taken reasonable steps to guard against such occurrence; and (c) the defendant failed to take such steps. The Supreme Court of Appeal then concluded:

"The prison authorities were well aware that prisoners might contract [TB] if reasonable steps were not taken to prevent it. I think I have made it clear earlier . . . that the evidence establishes convincingly that to the extent that any system existed at all for the proper management of the disease its application in practice was at best sporadic and in at least some respects effectively non-existent. . . . I need only say that I agree with the court below that the prison authorities failed to maintain an adequate system for management of the disease and in that respect they were negligent."[42]

[18] The last issue was causation. The Supreme Court of Appeal cautioned that it ought not to be overlooked that recognition of a delictual remedy will not impose obligations on the state that will be too onerous to fulfil and that what is required is no more than reasonable conduct on its part.[43] It held that for the applicant to succeed he "must establish that it is probable that the negligent conduct caused the harm"[44] and that the test in this regard is "whether but for the negligent act or omission of the [respondent] the event giving rise to the harm in question would have occurred."[45] 

[19] The Supreme Court of Appeal relied on the test formulated in International Shipping Co (Pty) Ltd v Bentley[46] for its use of a substitution exercise to determine factual causation in terms of the but-for test. It applied a reasonableness test and held that "proof alone that reasonable precautions were not taken to avoid foreseeable harm, and that the harm occurred, does not establish that the former caused the latter."[47] It remarked, in relation to the standard for determining culpability, that-

"it cannot be gainsaid that a consistent system of some kind at least was required to screen prisoners, isolate any that were found to be contagious, and administer treatment. I have already found that if any system existed at all its application in practice was at best sporadic and in at least some respects non-existent. On any standard that falls short of what ought reasonably to have been done. But while that failure on any standard is sufficient to find that the prison authorities were negligent it is not sufficient for determining whether the harm was caused by the omission. What needs to be established in addition is what the prison authorities ought to have done: only from there can one proceed to the enquiry whether that would have prevented Mr Lee being infected."[48]

[20] The Supreme Court of Appeal agreed with the High Court's sentiments about ‘reasonable measures' that ought to have been taken to prevent and control TB[49]but found that the question in each case is what would have been reasonable.[50] It held that "while proper screening procedures for incoming [inmates] are no doubt required, that begs the question what procedures might reasonably be expected in a large and congested prison."[51] The Court then dealt with the factors that would need to be balanced against one another to determine what might reasonably be expected in a large prison.[52] It held that the responsible authorities cannot "reasonably be expected to examine some 4000 prisoners with such regularity and thoroughness that [TB] will always be detected before the prisoner becomes contagious."[53] Self-reporting, the Supreme Court of Appeal said, "will necessarily be the only means for its detection in many cases."[54]

[21] The Supreme Court of Appeal found, that "it is just as likely as not that Mr Lee was infected by [an inmate] who the [responsible] authorities could not reasonably have known was contagious."[55] It concluded that the applicant's difficulty is that-

"he does not know the source of his infection. Had he known its source it is possible that he might have established a causal link between his infection and specific negligent conduct on the part of the [responsible] authorities. Instead he has found himself cast back upon systemic omission. But in the absence of proof that reasonable systemic adequacy would have altogether eliminated the risk of contagion, which would be a hard row to hoe, it cannot be found that but for the systemic omission he probably would not have contracted the disease. On that ground . . . the claim ought to have failed."[56]

In upholding the appeal, the Supreme Court of Appeal found that the applicant failed on a narrow factual point on the application of the but-for test, but was successful on all other elements of the delictual claim, including that Mr Lee was probably infected by a prisoner who had active TB while he was incarcerated.[57]

In this Court

[22] The issue relates primarily to the approach adopted by the Supreme Court of Appeal to the question of causation. It is contended by the applicant that the Supreme Court of Appeal was wrong in holding that causation had not been established and that it failed to follow the approach in Minister of Safety and Security v Van Duivenboden[58] regarding the standard of proof required of an applicant in establishing the existence of a causal nexus. 

[23] In the alternative to the delictual claim as pleaded, the applicant sought to amend his pleadings to introduce a claim based on unlawful detention for constitutional damages in the amount of R200 000 based on the same facts regarding the unlawful conduct. It was contended that the Supreme Court of Appeal misdirected itself by failing to import both restraints of sections 12 and 35 of the Constitution into the concept of wrongfulness against which negligent causation of harm stands to be tested in any claim for delictual damages. The applicant argued that he was denied an effective substantive remedy under section 34 of the Constitution.

[24] The amici applied to introduce two documents in accordance with Rule 31 of the Constitutional Court Rules. The first document relates to the report of the Judicial Inspectorate setting out, inter alia, TB fatalities in prisons and information on overcrowding. The second document is an expert scientific study which confirms and quantifies the nature and extent of the risk created by the conditions of imprisonment to which the applicant was subjected and to which many inmates continue to be subjected.

[25] They contended further that the approach adopted by the Supreme Court of Appeal is flawed in two respects: first, the application of the but-for test is inconsistent with the articulation of the test in Van Duivenboden and second, if on the correct application of the test the applicant will not be entitled to relief, the but-for test should be developed in accordance with the spirit, purport and objects of the Bill of Rights, specifically sections 27, 34 and 35 read with sections 172 and 173 of the Constitution, as well as the norms and values which underlie the Constitution. In addition, the amici do not reject the "prudent and fair" approach that the common law test is meant to engender. Rather, it is concerned with what appears to be a mechanical application of the test which results in an injustice.

[26] The respondent contended that the amendment to introduce an alternative claim for constitutional damages should be disallowed because that raises a new cause of action for the first time on appeal. It was contended that the applicant has failed to provide a basis for the amendment and that, in any event, constitutional damages have only been awarded where the remedy at common law would ordinarily not have been available. The respondent contended that she would be prejudiced if the amendment is allowed without affording her an opportunity to raise a defence of prescription, alternatively that the matter was governed by a binding judicial decision (res judicata). In the circumstances, she may be compelled to adduce new evidence to defend her position. 

[27] As regards the application to introduce new evidence under Rule 31, the respondent argued that leave should not be granted because the evidence sought to be introduced is not common cause or otherwise incontrovertible and capable of easy verification and is, in certain respects, at odds with the factual conclusions of the High Court and the Supreme Court of Appeal. The respondent denied that section 34 is violated. It was argued that the Supreme Court of Appeal's decision must be upheld. 

Issues

[28] The preliminary issues relate to: (a) leave to appeal; (b) introduction of new evidence; and (c) an application for an amendment to introduce a claim for constitutional damages based on unlawful detention (alternative residual relief). Issues on the merits relate to: (i) whether the negligent conduct of the responsible authorities was the cause of harm suffered by the applicant; if not, (ii) whether the common law regarding causation should be developed to give effect to the spirit, purport and objects of the Bill of Rights; and (iii) the determination of costs.

Leave to appeal

[29] This matter falls within the jurisdiction of this Court. The applicant sought to vindicate his right to freedom and security of the person under section 12(1) and right to be detained under conditions that are consistent with human dignity, including at least to be provided with adequate accommodation, nutrition and medical treatment under section 35(2)(e) of the Constitution.[59] In addition, based on the state's inherent constitutional obligations,[60] the constitutional norms of accountability and responsiveness[61] are, in my view, implicated.

[30] The matter is of importance, not only to the parties, but also to other inmates and the health sector generally. It is thus in the interests of justice that leave to appeal should be granted.

New evidence

[31] The amici sought leave to introduce new evidence of two documents under Rule 31. It is alleged that the evidence is relevant for the determination of the issues. The first document is an extract from the Annual Report 2010/2011: Treatment of Inmates and Conditions in Correctional Centres (Annual Report). The Annual Report provides information on ten Correctional Centre clusters reporting the most deaths. In seven of these, TB was the number one cause of natural deaths. It also contains information on overcrowding, which according to the agreed statement of facts, is one of the primary drivers of TB transmission. The Annual Report, it was argued, will enable this Court to develop a better understanding of the impact its decision will have beyond the parties. The second document is a scientific study entitled Tuberculosis in a South African prison - a transmission modelling analysis (Study) on the risk of TB contagion under prevailing conditions at Pollsmoor and what it would take to reduce the risk of TB transmission. The Study also shows why it is impossible to establish, with precision, the source of a TB infection.

[32] Rule 31 provides:

"(1) Any party to any proceedings before the Court and an amicus curiae properly admitted by the Court in any proceedings shall be entitled, in documents lodged with the Registrar in terms of these rules, to canvass factual material that is relevant to the determination of the issues before the Court and that does not specifically appear on the record: Provided that such facts-

(a) are common cause or otherwise incontrovertible; or

(b) are of an official, scientific, technical or statistical nature capable of easy verification.

(2) All other parties shall be entitled, within the time allowed by these rules for responding to such document, to admit, deny, controvert or elaborate upon such facts to the extent necessary and appropriate for a proper decision by the Court." [62]

[33] The evidence in the Annual Report sought to be introduced should, in my view, be disallowed because it is not relevant to the determination of the issues. Similarly, the evidence contained in the Study should be disallowed because its introduction does not meet the threshold requirement of being incontrovertible or common cause or capable of easy verification as envisaged in Rule 31. Accordingly, I would dismiss the application to introduce further evidence.

Alternative residual relief

[34] Part of the controversy on appeal related to the shift in the cause of action from the relief initially sought. The applicant sought permission to amend his particulars of claim to introduce an alternative claim for constitutional damages on appeal as a result of his unlawful detention. He pleaded that the detention was the cause of his contracting TB. It is noteworthy that the applicant accepted that this new claim was not raised in the High Court.[63]

[35] The granting of an amendment and allowing a claim for constitutional damages at this stage of appeal proceedings will be prejudicial to the respondent. New evidence may have to be presented. Apart from the fact that the raising of a new claim on appeal raises procedural and evidential difficulties, it is generally not in the interests of justice for this Court to sit as a court of first and final instance on appeal.[64] Accordingly, the application should be dismissed.

Causation

[36] The Supreme Court of Appeal dealt with the elements of a delictual claim and confirmed the High Court's finding regarding wrongfulness in relation to the responsible authorities' failure "to have reasonably adequate precautions against contagion, which was the foundation of the claim."[65] I agree with the Supreme Court of Appeal that there was a negligent breach on the part of the responsible authorities for failing to maintain an adequate system for management of TB.[66] The next prong of the inquiry is, however, whether the negligent omission caused the applicant harm - in becoming infected with TB. This is so because it is only causal negligence that can give rise to legal responsibility.[67]

[37] The point of departure is to have clarity on what causation is. This element of liability gives rise to two distinct enquiries. The first is a factual enquiry into whether the negligent act or omission caused the harm giving rise to the claim. If it did not, then that is the end of the matter. If it did, the second enquiry, a juridical problem, arises. The question is then whether the negligent act or omission is linked to the harm sufficiently closely or directly for legal liability to ensue or whether the harm is too remote.[68] This is termed legal causation.

[38] This element of liability is complex and is surrounded by much controversy. There can be no liability if it is not proved, on a balance of probabilities,[69] that the conduct of the defendant caused the harm. This is so because the net of liability will be cast too wide. A means of limiting liability, in cases where factual causation has been established, must therefore be applied. Whether an act can be identified as a cause depends on a conclusion drawn from available facts or evidence and relevant probabilities. Factual causation, unlike legal causation where the question of the remoteness of the consequences is considered, is not in itself a policy matter but rather a question of fact which constitutes issues connected with decisions on constitutional matters as contemplated by section 167(3)(b) of the Constitution.[70] 

[39] Although different theories have developed on causation,[71] the one frequently employed by courts in determining factual causation, is the conditio sine qua non theory or but-for test.[72] This test is not without problems, especially when determining whether a specific omission caused a certain consequence. According to this test the enquiry to determine a causal link, put in its simplest formulation, is whether "one fact follows from another."[73] The test-

"may involve the mental elimination of the wrongful conduct and the substitution of a hypothetical course of lawful conduct and the posing of the question as to whether upon such an hypothesis plaintiff's loss would have ensued or not. If it would in any event have ensued, then the wrongful conduct was not a cause of the plaintiff's loss; [otherwise] it would not so have ensued. If the wrongful act is shown in this way not to be a causa sine qua non of the loss suffered, then no legal liability can arise."[74]

[40] In the case of "positive" conduct or commission[75] on the part of the defendant, the conduct is mentally removed to determine whether the relevant consequence would still have resulted.[76] However, in the case of an omission the but-for test requires that a hypothetical positive act be inserted in the particular set of facts, the so-called mental removal of the defendant's omission. This means that reasonable conduct of the defendant would be inserted into the set of facts.[77] However, as will be shown in detail later, the rule regarding the application of the test in positive acts and omission cases is not inflexible. There are cases in which the strict application of the rule would result in an injustice, hence a requirement for flexibility. The other reason is because it is not always easy to draw the line between a positive act and an omission.[78] Indeed there is no magic formula by which one can generally establish a causal nexus. The existence of the nexus will be dependent on the facts of a particular case.[79]

[41] As is evident from the statement of agreed facts, the applicant was not infected with TB when he was admitted to Pollsmoor. It is common cause that, on the evidence on record, it is more probable than not that Mr Lee contracted TB in prison, rather than outside it. The Supreme Court of Appeal Judgment proceeded on an acceptance of this probability,[80] but it non-suited Mr Lee on the basis that he failed to prove that reasonable systemic adequacy would have "altogether eliminated" the risk of contagion, that he does not know the source of his infection and that had he known the source it is possible that he might have been able to establish a causal link between his infection and the specific negligent conduct on the part of the responsible authorities.[81]

[42] In my respectful view the Supreme Court of Appeal erred in adopting that approach. The reasons for this are twofold. First, it was not necessary for the substitution of reasonable alternative measures to determine factual causation because our law allows for a more flexible approach. Second, even if the use of a reasonable alternative substitution was necessary in the circumstances, our law does not require evidentiary proof of the alternative, but merely substitution of a notional and hypothetical lawful, non-negligent alternative. The purpose of the exercise is to evaluate the evidence presented by a plaintiff, not to require more evidence. If the substitution exercise is done in this way, probable factual causation is established.

[43] The Supreme Court of Appeal approached the matter on the basis that in the case of an omission the issue of factual causation by definition involves an obligation on a defendant to initiate reasonable action, which a plaintiff needs to establish as an alternative in order to determine what would have happened if that had occurred.[82] The Court found that Mr Lee failed to do so.[83] It went further and applied a rigid deductive logic that necessitated the conclusion that, because Mr Lee did not know the exact source of his infection, he needed to show that "reasonable systemic adequacy would have altogether eliminated the risk of contagion".[84]

[44] I emphasise that our law requires neither the inflexible application of a substitution exercise in the application of the but-for test, nor the inflexible kind of logic used by the Supreme Court of Appeal in its application of that test. In addition, the wrong done to Mr Lee is not treated as a mere omission. In what follows I will attempt to justify these propositions. Like other jurisdictions our courts have also struggled to come to terms with the difficulties of causation. It is not necessary to chart that development in our law, as it is now settled and, as stated, sufficient and flexible enough to dispose of this case. This flexibility has a long history, and has never been discarded.

[45] In Kakamas[85] it was stated that "[c]ausality often raises difficult legal questions which cannot always be answered by strict adherence to logic. Recourse may sometimes be had to what [the House of Lords] called the law's ‘empirical or common-sense view of causation'".[86] In Siman[87] the minority judgment noted that "[f]inally, as in other problems relating to causation in delict, in applying the ‘but-for' test the Court should not overlook the importance of applying common sense standards to the facts of the case".[88]

[46] The most recent, post-constitutional affirmations of that flexibility are to be found in Van Duivenboden[89] and Gore.[90] In Gore the approach adopted in discharging the onus in relation to factual causation was described thus:

"With reference to the onus resting on plaintiff, it is sometimes said that the prospect of avoiding the damages through the hypothetical elimination of the wrongful conduct must be more than 50%. This is often followed by the criticism that the resulting all-or-nothing effect of the approach is unsatisfactory and unfair. A plaintiff who can establish a 51% chance, so it is said, gets everything, while a 49% prospect results in total failure. This, however, is not how the process of legal reasoning works. The legal mind enquires: What is more likely? The issue is one of persuasion, which is ill-reflected in formulaic quantification. The question of percentages does not arise (see to this effect Baroness Hale in Gregg v Scott).[91] Application of the ‘but for' test is not based on mathematics, pure science or philosophy. It is a matter of common sense, based on the practical way in which the ordinary person's mind works against the background of everyday-life experiences. Or, as was pointed out in similar vein by Nugent JA in Minister of Safety and Security v Van Duivenboden:

‘A plaintiff is not required to establish the causal link with certainty, but only to establish that the wrongful conduct was probably a cause of the loss, which calls for a sensible retrospective analysis of what would probably have occurred, based upon the evidence and what can be expected to occur in the ordinary course of human affairs rather than metaphysics.'"[92]

[47] It is useful to have a close look at what the Appellate Division said inSiman[93] regarding the application of the substitution exercise:

"As was pointed out by this Court in Minister of Police v Skosana 1977 (1) SA 31 (A) at 34F-35D, 43D-44F, causation in the law of delict involves two distinct enquiries: (i) whether the defendant's wrongful act was a cause in fact of the plaintiff's loss; and (ii) if so, whether and to what extent the defendant should be held liable for the loss sustained by the plaintiff (this latter enquiry often being referred to as the question of the remoteness of damage). In Joubert The Law of South Africa vol 8 title "Delict" (by Prof J C van der Walt) paras 47-49, the terms ‘factual causation' and ‘legal causation' are used to denote the concepts underlying these two enquiries. They seem to be convenient labels.

In the present case the problem revolves essentially round the question of factual causation . . ..

The enquiry as to factual causation generally results in the application of the so-called ‘but-for' test, which is designed to determine whether a postulated cause can be identified as a causa sine qua non of the loss in question. This test is applied by asking whether but for the wrongful act or omission of the defendant the event giving rise to the loss sustained by the plaintiff would have occurred. In a case such as the present one, which is uncomplicated by concurrent or supervening causes emanating from the wrongful conduct of other parties . . . the but-for or, causa sine qua non, test is, in my opinion, an appropriate one for determining factual causation.

In order to apply this test one must make a hypothetical enquiry as to what probably would have happened but for the unlawful act or omission of the defendant. In some instances this enquiry may be satisfactorily conducted merely by mentally eliminating the unlawful conduct of the defendant and asking whether, the remaining circumstances being the same, the event causing harm to plaintiff would have occurred or not. If it would, then the unlawful conduct of the defendant was not a cause in fact of this event; but if it would not have so occurred, then it may be taken that the defendant's unlawful act was such a cause. This process of mental elimination may be applied with complete logic to a straightforward positive act which is wholly unlawful. So, to take a very simple example, where A has unlawfully shot and killed B, the test may be applied by simply asking whether in the event of A not having fired the unlawful shot (ie by a process of elimination) B would have died. In many instances, however, the enquiry requires the substitution of a hypothetical course of lawful conduct for the unlawful conduct of the defendant and the posing of the question as to whether in such case the event causing harm to the plaintiff would have occurred or not; a positive answer to this question establishing that the defendant's unlawful conduct was not a factual cause and a negative one that it was a factual cause. This is so in particular where the unlawful conduct of the defendant takes the form of a negligent omission. In The Law of South Africa (ibid para 48) it is suggested that the elimination process must be applied in the case of a positive act and the substitution process in the case of an omission. This should not be regarded as an inflexible rule. It is not always easy to draw the line between a positive act and an omission, but in any event there are cases involving a positive act where the application of the but-for rule requires the hypothetical substitution of a lawful course of conduct (cf Prof A M Honoré in 11 International Encyclopaedia of Comparative Law c 7 at 74-6)."[94]

[48] What we may glean from Siman is that substitution as part of the application of the but-for test may not be apposite where there may be concurrent or supervening causes; that it should not be applied inflexibly; that drawing the line between a positive act and an omission is not always easy to do; and, finally, that even in the application of the but-for test common sense may have to prevail over strict logic.[95] What is more, the majority in Siman considered it wrong or inappropriate to apply the substitution exercise of a hypothetical course of lawful conduct for unlawful conduct.[96] The minority characterised[97] examples in other South African cases[98] and a New Zealand case[99] as instances of the substitution process of reasoning. In LAWSA,[100] the authors think that it might be better to characterise them as illustrating the point that "common-sense standards should be used where the but-for test proves to be inadequate".[101]

[49] Our existing law does not require, as an inflexible rule, the use of the substitution of notional, hypothetical lawful conduct for unlawful conduct in the application of the but-for test for factual causation.[102]

[50] There is, in my view, an additional reason why one should be cautious when applying the substitution test in our law of factual causation. The substitution exercise of determining hypothetical lawful conduct involves an evaluation of normative considerations. The determination of a question of fact, although it is also an evaluative exercise, cannot depend on social and policy considerations.[103] Even though the purpose of using the normatively determined lawful conduct as an alternative is not primarily aimed at making an "is" question an "ought" question, it seems to me that it inevitably makes it at least a mixed question of fact and law. The distinction between factual and legal causation made in our law becomes unnecessarily less clear.

[51] It is apparent from the minority judgment in Siman[104] and the judgment in International Shipping,[105] that delictual liability for omissions, still a recent development then,[106] was a matter of some concern. In the former the Court made it clear that the substitution test might find particular application in relation to omissions and in the latter it referred explicitly to the danger of limitless liability that needs to be kept within reasonable bounds by "giving proper attention to, inter alia, the problem of causation".[107]

[52] But since then our law relating to the development of wrongfulness as a criterion for determining the boundaries of delictual liability has moved on. InEwels[108] it was held that our law had reached the stage of development where an omission is regarded as unlawful conduct when the circumstances of the case are of such a nature that the legal convictions of the community demand that the omission should be considered wrongful.[109] This open-ended general criterion has since evolved into the general criterion for establishing wrongfulness in all cases, not only omission cases. The imposition of wrongfulness under this enquiry is determined with reference to considerations of public and legal policy, consistent with constitutional norms.[110] The general criterion of "reasonableness" in the wrongfulness enquiry concerns the reasonableness of imposing liability on the defendant and not the reasonableness of the defendant's conduct, which is an element of the separate negligence enquiry.[111] There is thus no pressing need to contaminate the factual part of the causation enquiry with these kinds of normative considerations based on social and policy considerations.

[53] As noted earlier in Siman, it "is not always easy to draw the line between a positive act and an omission".[112] The wrong Mr Lee complains of is, in our law, based on being detained in conditions that infringe upon his dignity. That was the position under our pre-constitutional common law,[113] and that has received constitutional support in section 35(2)(e) of the Constitution[114] that provides that detained persons have the right "to conditions of detention that are consistent with human dignity".[115]

[54] There was thus nothing in our law that prevented the High Court from approaching the question of causation simply by asking whether the factual conditions of Mr Lee's incarceration were a more probable cause of his tuberculosis, than that which would have been the case had he not been incarcerated in those conditions. That is what the High Court did and there was no reason, based on our law, to interfere with that finding.

[55] Even if one accepts that the substitution approach is better suited to factual causation, the preceding discussion shows that there is no requirement that a plaintiff must adduce further evidence to prove, on a balance of probabilities, what the lawful, non-negligent conduct of the defendant should have been. All that is required is "the substitution of a hypothetical course of lawful conduct and the posing of the question as to whether upon such a hypothesis the plaintiff's loss would have ensued or not".[116] What is required is postulating hypothetical lawful, non-negligent conduct,[117] not actual proof of that conduct. The law recognises science in requiring proof of factual causation of harm before liability for that harm is legally imposed on a defendant, but the method of proof in a court room is not the method of scientific proof. The law does not require proof equivalent to a control sample in scientific investigation.

[56] Postulating hypothetical lawful, non-negligent conduct on the part of a defendant is thus a mental exercise in order to evaluate whether probable factual causation has been shown on the evidence presented to court. It is not a matter of adducing evidence, as the Supreme Court of Appeal appears to have found.[118] I accept that the postulate must be grounded on the facts of the case, but that is not the same as saying that there is a burden on the plaintiff to adduce specific evidence in relation thereto.

[57] What was required, if the substitution exercise was indeed appropriate to determine factual causation, was to determine hypothetically what the responsible authorities ought to have done to prevent potential TB infection, and to ask whether that conduct had a better chance of preventing infection than the conditions which actually existed during Mr Lee's incarceration. Substitution and elimination in applying the but-for test is no more than a mental evaluative tool to assess the evidence on record. In my view, this hypothetical exercise shows that probable causation has been proved.

[58] That there is a duty on Correctional Services authorities to provide adequate health care services, as part of the constitutional right of all prisoners to "conditions of detention that are consistent with human dignity",[119] is beyond dispute. It is not in dispute that in relation to Pollsmoor the responsible authorities were aware that there was an appreciable risk of infection and contagion of TB in crowded living circumstances. Being aware of that risk they had a duty to take reasonable measures to reduce the risk of contagion.

[59] Although I accept that a reasonably adequate system may not have "altogether eliminated the risk of contagion",[120] I do not think that the practical impossibility of total elimination is a reason for finding that there was no duty at least to reduce the risk of contagion. It seems to me that if a non-negligent system reduced the risk of general contagion, it follows - or at least there is nothing inevitable in logic or common sense to prevent the further inference being made - that specific individual contagion within a non-negligent system would be less likely than in a negligent system. It would be enough, I think, to satisfy probable factual causation where the evidence establishes that the plaintiff found himself in the kind of situation where the risk of contagion would have been reduced by proper systemic measures.[121]

[60] In postulating the hypothetical non-negligent conduct needed for the substitution exercise, the SCOs[122] provide helpful guidance.[123] The minimisation of risk of contagion is largely dependent on the effective screening of incoming prisoners and the isolation of infectious patients. The SCOs provide for screening for medical problems by a registered nurse on admission and for a medical examination within 24 hours by a registered nurse, medical officer or medical practitioner. They also make special provision for the immediate reporting of communicable or contagious diseases and for the isolation of persons with these diseases. There is nothing on record to suggest that this kind of screening, examination and isolation in terms of the SCOs would not have been effective in reducing the risk of infection and contagion of a disease like TB - indeed the case the authorities initially sought to present was that the process was properly followed.[124] It does not require much imagination to postulate that adherence to the SCOs may constitute the non-negligent conduct necessary for the substitution exercise.

[61] The Supreme Court of Appeal acknowledged that an effective programme did not exist during Mr Lee's incarceration, as evidenced by superficial initial screening and the failure to isolate inmates who had TB. If the proper process had been followed, this would not have happened. In my view, it is legitimate to draw the inference that this is also probably how Mr Lee contracted the disease. As I understand the logic of the Supreme Court of Appeal's approach, it is not possible to make this kind of inference of likely individual infection from the fact that a non-negligent system of general systemic control would generally reduce the risk of contagion. I do not agree.

[62] The implication of that kind of inexorable logic is that factual causation under our law can never be proved where the specific incident or source of infection cannot be identified. This means that even wrongful and negligent conduct of correctional facility authorities can by no means, in those instances, lead to delictual liability. Fortunately, in my view, our law relating to factual causation does not require that kind of inflexibility in determining factual causation.

[63] In Van Duivenboden,[125] the Supreme Court of Appeal correctly remarked that the state has a positive constitutional duty under section 7(2) of the Constitution[126] to act in protection of the rights in the Bill of Rights.[127] The Supreme Court of Appeal in that case went on to say that where a state, as represented by the persons who perform functions on its behalf, acts in conflict with its constitutional duty to protect those who are "delivered into [its] absolute power"[128], the norm of accountability must of necessity assume an important role in determining whether a legal duty ought to be recognised in any particular case.[129] I cannot agree more with these sentiments.

[64] It is indeed so that "[p]risoners are amongst the most vulnerable in our society to the failure of the state to meet its constitutional and statutory obligations",[130] and that "a civilised and humane society demands that when the state takes away the autonomy of an individual by imprisonment it must assume the obligation . . . inherent in the right . . . to ‘conditions of detention that are consistent with human dignity'."[131] I thus agree that "there is every reason why the law should recognise a claim for damages to vindicate [the prisoners'] rights".[132] To suggest otherwise, in circumstances where a legal duty exists to protect Mr Lee and others similarly placed, will fail to give effect to their rights to human dignity, bodily integrity and the right to be detained in conditions that are consistent with human dignity under the Constitution, including at least exercise and the provision, at state expense, of adequate accommodation, nutrition, and medical treatment. I stress that on the approach adopted by the Supreme Court of Appeal it is unlikely that any inmate will ever be able to overcome the hurdle of causation and further, no effective alternative remedy will be available to a person in the position of the applicant.

[65] In any event, the Supreme Court of Appeal accepted the following as true, that:

"[reasonable measures] [in casu] would translate into the proper screening of incoming [inmates], inclusive of a physical chest examination; separating out those who had, or were suspected of having TB, or who were obviously under nourished and vulnerable to TB; the provision of adequate nutrition to those who were undernourished and otherwise vulnerable to TB; regular and effective screening of the prisoner population, inclusive of examinations by means of X-Rays and/or physical chest examinations by means of a stethoscope, to identify possible TB infection; isolation of infectious inmates and effective implementation of the DOTS system over the prescribed period of time."[133]

[66] Although the respondent agreed that control of TB at Pollsmoor depended upon effective screening of incoming inmates, the Supreme Court of Appeal held that self-reporting will necessarily be the "only" means for TB detection. The Supreme Court of Appeal went further to hold that the applicant's circumstances are an example of the time that can elapse before the risk of contagion is detected. It may, however, be deduced from the statement of agreed facts that that will be the case if no effective screening is in place for incoming inmates. In my view, the fact that Mr Lee was alive to the risk of contracting TB and sensitive to the need for early diagnosis is, with respect, neither here nor there. The fact of the matter, according to the agreed facts, is that Mr Lee was not infected with TB when he was first admitted at Pollsmoor.

[67] Having found that a causal link exists, for completeness, the next enquiry regarding legal liability becomes relevant even though legal causation was not anissue before us. It involves the question whether the defendant should be held liable. There must be a reasonable connection between the breach and the harm done.[134] This serves to limit liability because the consequences of an act or omission might stretch into infinity. The respondent did not suggest that the harm was too remote. It bears mentioning that the Supreme Court of Appeal correctly rejected the respondent's contentions for resisting liability in that "it would impose an inordinate burden on the state", and will "expose it to indeterminate liability" and that "there are means other than a claim for damages that enable prisoners to vindicate their rights."[135]

[68] Furthermore, the democratic values of state accountability, responsiveness and the rule of law are important aspects of the Constitution which are implicated here. To borrow the words of Madala J in Nyathi v MEC for Department of Health, Gauteng and Another,[136] these values are "pillar-stones of our democracy" and "must be observed scrupulously."[137] If not, particularly in light of the injunctions of the Act and the rights sought to be vindicated, "we have a recipe for a constitutional crisis of great magnitude."[138] The Supreme Court of Appeal pertinently remarked:

"Mr Lee has certainly had a hard time of it. For four years he was imprisoned while the state mustered its case against him and then the state failed. Meanwhile Mr Lee knew that he was at risk of contracting [TB] in a prison where the health-care regime was breaking down. When it occurred he had to manipulate and cajole at times to ensure that he consistently received medication, conscious that he would suffer adverse consequences if he failed to do so. He had good reason to feel aggrieved when he left prison but his troubles were not yet at an end.

When he vented his grievance by suing the state he was met with a defence on every leg of his claim. The state contested that Mr Lee had been infected in prison with no substantial grounds for doing so. It contested the allegations of an inadequate health-care regime when it must have known that it was defending the indefensible. The failing [health-care] regime had been repeatedly reported by its medical doctors at high level, various reports on the situation had been circulated, newspapers had reported the position, a report of an inspector from the office of the Inspecting Judge that had been prepared some four years before the matter came to trial disclosed that [TB] management was virtually non-existent, and so on. Yet the state persisted . . . [not acknowledging any] responsibility towards Mr Lee at any time.

Mr Lee set out to vindicate an important statutory and constitutional right and has done so substantially. It is true that his claim has failed but only on a narrow factual point. The state has important responsibilities to its citizens. It might not always be able to fulfil them but then it ought properly to recognise where it has failed."[139] (Emphases added.)

[69] The responsible authorities' function is to execute its duties in accordance with the purposes of the Act which include detaining all inmates in safe custody whilst ensuring their human dignity and providing adequate health care services for every inmate to lead a healthy life. The rule of law requires that all those who exercise public power must do so in accordance with the law and the Constitution.[140] This, including the requirements of accountability and responsiveness, provides ‘additional' reasons for finding in favour of the applicant and imposing delictual liability.[141] This would enhance the responsible authorities' accountability, efficiency and respect for the rule of law.

[70] In the circumstances, there is a probable chain of causation between the negligent omissions by the responsible authorities and Mr Lee's infection with TB. 

I would therefore uphold the applicant's claim, as the High Court did.

Development of the common law

[71] I am indebted to Cameron J for the exposition of developments in the Anglo-American common law, which illustrates the solutions that other jurisdictions have arrived at in relation to similar problems of causation.[142] I draw comfort from the fact that the injustice of an inflexible legal approach to factual causation is also recognised in foreign jurisprudence. The common law may well have to develop from time to time in this area, as in others. But in the circumstances of this case - particularly the nature of the omission - I do not consider that our law needs to be developed in accordance with the casuistic approach endorsed by the cases referred to. 

[72] The underlying cause for the development of exceptions in relation to causation in these foreign cases appears to be the inflexibility of the application of the but-for test. Our law has always recognised that the but-for test should not be applied inflexibly. A court ultimately has to make a finding as to whether causation was established on a balance of probabilities on the facts of each specific case. Causation will not always follow whenever a wrongful and negligent omission is shown.

[73] The concern that a flexible approach to factual causation and the relaxation of the but-for test in appropriate cases may lead to limitless liability, especially in relation to omission cases, has been addressed by the development of the test of reasonableness in the wrongfulness enquiry. That enquiry now concerns the reasonableness of imposing liability on a defendant, and is not restricted to the reasonableness of the defendant's conduct, which is an element of the separate negligence enquiry in our law.[143] The wrongfulness requirement in our law thus provides a normative mechanism, in addition to the negligence enquiry, to decide whether delictual liability should be extended or restricted. On my understanding of the duty of care in other common law systems, it does not allow for that kind of extended judicial policy control of imposing liability in tort, hence the perceived need to do so through exceptional relaxation of the but-for test of causation.

[74] In conclusion, I would grant leave and uphold the appeal.

Costs

[75] The applicant seeks costs including costs of three counsel. The explanation proffered does not justify the award of costs for three counsel in the circumstances of this case. Costs of two counsel as between party and party should suffice.

Order

[76] In the event, the following order is made:

1. Leave to appeal is granted.

2. The application to introduce new evidence under Rule 31 of the Constitutional Court Rules is dismissed.

3. The application to amend the particulars of claim is dismissed.

4. The appeal is upheld.

5. The order of the Supreme Court of Appeal is set aside.

6. The respondent is declared liable to the applicant in delict.

7. The case is remitted to the Western Cape High Court, Cape Town for a determination on quantum.

8. The respondent is ordered to pay the applicant's party and party costs in this Court, the Supreme Court of Appeal and the High Court including, where applicable, costs of two counsel.

CAMERON J (Mogoeng CJ, Khampepe J and Skweyiya J concurring):

[77] I have had the benefit of reading the judgment of my colleague Nkabinde J, and gratefully adopt her exposition of the facts and issues, though I differ from her reasoning and conclusion.

[78] In my view, on the facts before the High Court[144] and the Supreme Court of Appeal,[145] it cannot be said that it is more probable than not that ‘but for' the negligence of the prison authorities, Mr Lee would not have contracted tuberculosis (TB). The only conclusion possible on the evidence is that the prison authorities' negligent conduct increased the overall risk that Mr Lee would contract TB. But that should not lead to defeat for Mr Lee. In my view, our law should be developed to compensate a claimant negligently exposed to risk of harm, who suffers harm. But the delicate and difficult task of undertaking that development should start in the High Court, and should involve full assessment of the intricacies of a system of risk-based compensation. The case should therefore be remitted.

[79] The central question is whether Mr Lee established at the trial that his infection with TB would probably have been averted if his jailers had taken reasonable measures to protect him from it. We know from the agreed statement of facts, which we required the parties to submit instead of the trial record, that Mr Lee probably contracted TB as a result of being locked up in Pollsmoor at a time when, despite standing orders, TB control measures at the prison were virtually non-existent.[146] So he proved negligence and harm. But did he establish a link between the two?[147]

[80] The Supreme Court of Appeal upended the High Court's finding in Mr Lee's favour on the principle of causation. It found that he had failed to show that the prison's negligence caused his injury. The Court's reasoning is dense and in places elliptic. But it centres on the fact that Mr Lee did not know precisely how he acquired TB. It found in his favour that he probably became infected while in Pollsmoor.[148] And it assumed in his favour, without making a definite finding, that he was probably infected by a fellow prisoner who had active (and thus transmissible) TB while under the control of the prison authorities.[149] But Mr Lee's claim now confronted a two-fold problem. First, what measures ought the prison authorities reasonably to have taken to prevent TB transmission? And second, had the prison authorities taken these measures, would they more probably than not have saved Mr Lee from getting TB?

[81] The Supreme Court of Appeal noted that the law does not require the prison authorities to "guarantee" that a prisoner does not get TB.[150] Pollsmoor did not have to do "everything that would have avoided" TB being transmitted.[151] The prison authorities only had to take reasonable steps to protect Mr Lee against it. What reasonable prison authorities should have done depended on medical best-practice, security issues, financial resources, personnel, space, the prevalence and incidence[152] of the disease, "and other factors besides."[153] On all this, Mr Lee had led only "scant evidence", which went "nowhere" towards showing what reasonable prison authorities should have done.[154]

[82] But in any event, the Court found, because Mr Lee could not say precisely how he became infected, he was unable to prove that what the prison authorities did not do, but should have done, caused his injury. Here the particular nature of TB infection placed an "insuperable hurdle" in Mr Lee's path.[155]

[83] I pause here to emphasise the unique nature of TB. It was accepted in the High Court and the Supreme Court of Appeal that transmission can occur by breathing in just one airborne TB mycobacterium.[156] A human being, at any one time, may be carrying any number of bacterial cells,[157] from any possible number of sources. Indeed it was also common cause that in some cases TB will take hold but remain dormant, while in others it will "multiply and manifest in active disease".[158] What is more - and this is perhaps the biggest difficulty the prison authorities face - a person in whom TB has progressed from "dormant" to "active" will not always immediately show symptoms. The symptoms may manifest only as the disease progresses. To add to the complexity, it is possible to be tested for TB, but be screened as "negative", even though the bacterium is in fact present.[159] This means that identifying the presence of the bacterium and indeed the number of cells carried by the host is extremely difficult. Thus, we know exactly what agent causes TB (the mycobacterium tuberculosis), but science cannot identify which one of innumerable exposures was the probable source of infection in this case. 

[84] Hence, in a prison setting, particularly a large prison like Pollsmoor, the disease cannot always be diagnosed immediately.[160] So prisoners with active TB are contagious to others before the prison can reasonably be expected to diagnose, treat and if necessary isolate them. The result was that even if Mr Lee had shown what steps the prison authorities should reasonably have taken, the course and ferocity of the disease meant that he would always be at risk of contagion from undiagnosed fellow prisoners, whatever prison management did. Since reasonable measures could not eliminate this risk,[161] and since Mr Lee could not pinpoint who had infected him, it was "just as likely as not" that he was infected by a prisoner whom the prison authorities could not reasonably have known might pass the disease on to him. It was therefore not possible to find that a negligent omission by the prison authorities probably caused his infection.[162]

[85] To this the Supreme Court of Appeal added an observation. Since Mr Lee could not track a precise trail of negligent acts by the prison authorities that caused his infection,[163] to make his case he found himself cast back unavoidably on proving a "systemic omission"[164] - namely that lack of adequate overall management measures caused his injury. But without knowing exactly from whom he contracted TB, this was impossible. To prove that a failure to implement a reasonable system of TB controls had probably caused him to contract TB, Mr Lee had to show that such a system "would have altogether eliminated the risk of contagion".[165] Because he could not do this, his claim had to fail.

[86] The finding that Mr Lee had to show that an adequate system of TB prevention would have "altogether eliminated" the risk of contagion entreats further reflection. It proceeds from the fact that it was impossible for Mr Lee to show that his infection arose from a source to which he was negligently exposed, as opposed to one to which he was non-negligently exposed.[166] Given the indeterminacy of his source of infection, taking reasonable measures could therefore reduce his overall risk of becoming infected, in general terms - but he could never show that in the specific instance of his own infection those measures would probably have saved him from TB. This is because his own infection might equally well have resulted from a non-negligent exposure. It was this factor that made it necessary for him to show that reasonable systemic measures would have excluded all risk of infection.[167]

[87] In other words, to prove probable causative negligence, Mr Lee had to show that a system of reasonable precaution would have eliminated all possibility of infection. Only this would enable the Court to conclude that, since he did in fact become infected, albeit from an unknown source, he was probably infected by a negligent lapse in those precautions. This was because, if the system had been adequately applied, there would have been no risk of transmission at all.

[88] On the existing but-for test for causation, the Court's logic is not at fault. It entails that to establish liability a claimant who contracts TB would have to show that, had a reasonable system been in place, the risk of contagion would have been zero. Ordinarily, where the source of infection is known, we would say a claimant need show merely that the measures reduced the risk to below 50%. However, where the claimant cannot specify the source of his infection, this cannot apply on the existing but-for test for causation. This reasoning warrants three comments. 

[89] First, the formulation seems, paradoxically, to be at odds with the Supreme Court of Appeal's own earlier point, namely that the law does not demand a state institution or employer to "guarantee" absence of risk.[168]

[90] Second, because it requires elimination of systemic risk in order to make a finding of probable causative negligence in cases of undetermined transmission, it rejects the possibility of risk-based recovery. If a claimant cannot prove that his or her injury was attributable to the defendant's negligence on the established more-probably-than-not test, he or she must fail. This entails that the Supreme Court of Appeal's reasoning axiomatically rejects the notions of both apportioned blame and recovery on the basis of substantial increase in risk.

[91] Third, a corollary of the approach the Supreme Court of Appeal took seems to be that, where a diseased claimant can prove negligence but not the source of infection, duties and standards of care in effect become redundant. Where the source of infection is unknown, the plaintiff's obligation is to show, in the hypothetical, a total elimination of risk. This has an obviously bad spin-off, in that it may reduce incentives for state institutions and employers to reduce risk in circumstances where they cannot reasonably be expected to eliminate it.

[92] To bring the matter starkly back to the case before us, because a claimant in the position of Mr Lee may never be able to trace the source of his or her infection, the prison authorities would on the but-for test for causation have no legal incentive, at least in principle, to reduce the risk of contagion. This is because they need show only a small chance that a claimant like Mr Lee may have contracted TB anyhow, even if reasonable systemic measures had been applied. Because the risk was not altogether eliminated, the claim will fail. And this will almost always be the case. Hence claimants in Mr Lee's position will, on the existing test, almost never be able to succeed.

[93] These problems all arise from the rigidity of the common law test for causation, which requires claimants to prove more probably than not that the defendant's negligence caused their injury. They are not unique to our legal system. They have caused other jurisdictions to grapple with new approaches to the test for causation in cases where the claimant is unable to pinpoint the source of his injury, or to indicate that his injury was probably caused by a defendant who contributed to it, or where the defendant exposed the claimant to a risk of the injury in fact suffered.

[94] The jurisprudence of the United Kingdom in particular has responded with innovation to the problem Mr Lee's case presents. The claimants in Fairchild[169]contracted mesothelioma because successive employers negligently exposed them to asbestos dust. Inhaling asbestos dust and fibres plays some part in causing the disease, but the precise way in which a mesothelial cell is transformed into a mesothelioma is unknown. All that was certain was that the greater the exposure, the greater the overall risk of contracting mesothelioma. The claimants' difficulty was that the condition can be caused by being exposed to a single asbestos fibre - so it was impossible for them to prove precisely which employer had negligently caused their injury. It was also impossible for them to prove that the negligence of any one employer had more probably than not caused their exposure. No claimant could prove that ‘but for' the negligence of any specified employer, the injury would probably not have occurred. Thus, though any one of the employers might have caused the disease, the claimants could not prove which of them probably did. The House of Lords, approving its previous decision in McGhee,[170] held that the claimants were entitled to recover from any of the negligent employers.

[95] In its causal sequence or aetiology, TB is analogous to a "single fibre" disease.[171] As explained earlier,[172] TB may be caused by any one of innumerable exposures to the mycobacterium tuberculosis, some of which occur because the defendant was negligent and some not. Indeed, we cannot know how many of the exposures are due to negligence and how many are ‘innocent'. While we know the agent in general, we do not know, and cannot know, whether the particular agent stemmed from a prisoner, from a guard, from a visitor, and so on - any one of whom may be carrying any number of detectable or undetectable bacteria. And while the Supreme Court of Appeal was prepared to assume in Mr Lee's favour that a fellow prisoner infected him, the problem remained that the prisoner who (presumably) passed TB on to him may not have been detectably contagious when he did so.

[96] It is these unique characteristics of TB that ran headlong into the fact that our law of delict demands proof of probable cause of harm. And it is those characteristics that led the Supreme Court of Appeal to non-suit Mr Lee. The question is whether the result - that Mr Lee is without remedy - is acceptable in our constitutional state. The answer is No. But how do we get to a just outcome? It seems to me that our common law of recovery for negligent injury should be developed to allow for recovery in cases where a plaintiff can prove that the defendant negligently exposed him or her to the risk of harm, and the harm eventuated. But development is an intricate task. It seems most appropriate in the limited cases of "single fibre"[173] diseases where one can identify a single agent, but the disease, by its very nature, defies the but-for test.

[97] Subsequent case law has further refined Fairchild.[174] The exception applies only where it is not possible to establish causation in the normal way.[175] InBarker,[176] which also concerned death from mesothelioma caused by exposure to asbestos dust, it was held that a defendant is not liable in full, but only in proportion to the relative increase in risk for which it is responsible.[177] And crucially, the exception to the rules of causation is confined to "single agent" cases, as opposed to those where multiple differing causes[178] may have given rise to the condition.

[98] American courts have also grappled with the inadequacy of the but-for test for causation.[179] Some have even considered reversing the onus of proof for certain classes of injury.[180]

[99] What seems clear is that in some classes of claim the traditional common law but-for test is just not enough to deliver adequately just outcomes. This is particularly so in the case of Mr Lee. As the main judgment rightly points out, prisoners are particularly vulnerable and there are special reasons for imposing liability to ensure accountability and responsiveness.[181] The Supreme Court of Appeal in its judgment gave feelingful expression to the same point. It noted that Mr Lee was delivered into the absolute power of the state, lost his autonomy and was entitled to measures ensuring his physical welfare.[182] It said there was "every reason why the law should recognise a claim for damages" to vindicate prisoners' rights.[183]

[100] All this indicates that the common law but-for test for causation is an over-blunt and inadequate tool for securing constitutionally tailored justice in cases where prisoners have proved exposure to disease because of negligence on the part of the prison authorities, but cannot pinpoint the source of their injury. The change in the United Kingdom jurisprudence has been justified frankly, on grounds of simple justice, since it would be wrong for employers to avoid liability for wrongdoing because of causal indeterminacy. The English courts did this without constitutional imperative. With us, that imperative is there. In Mr Lee's case, his vulnerability as a prisoner, which meant he was unable to put himself out of harm's way, together with the lack of proper care on the part of the prison authorities, makes a similarly powerful case for developing our common law.

[101] This is not to say that normative considerations bearing on what wrongs should be compensable in our constitutional system play into determining factual cause and effect - but, rather, these considerations make the case for relaxing the over-rigid strictures of but-for factual inferences. Should the common law be developed, the causation inquiry remains a question of fact, though now the question is a wider one, that is easier to answer in cases like Mr Lee's: would reasonable measures have reduced the overall risk of infection? And, then, should the extent of risk to which the defendant's negligent conduct exposed the claimant lead to recovery for the injury that was suffered?

[102] What the parties' agreed statement of facts tells us about the epidemiology of TB underscores how inapt the blunt application of the common law test is. TB is a "formidable infectious disease" that constitutes a serious public health problem, and South Africa "has one of the highest incidence rates of tuberculosis in the world" - which means that the rate of transmission in this country is one of the highest anywhere. And prisons are a "favourable environment for contracting tuberculosis". The authorities at Pollsmoor were "pertinently aware of the risk to prisoners of contracting tuberculosis" - yet they failed to observe their own standing orders.

[103] The test for causation the Supreme Court of Appeal applied, while adequate for cases where a claimant can identify the source of his injury, may not be sufficient where he or she cannot know the source. On the existing test, the Court correctly found that because Mr Lee cannot specify the source of his infection, he cannot show on balance of probabilities that the prison authorities caused his injury. But this approach rejects all prospect of recovery even where the court finds that the defendant's negligence increased the risk of injury. This seems unjust. And it may be constitutionally unsustainable. The law has never required perfection in imposing standards of care. It has only ever required reasonable conduct. Reasonableness permits a defendant to fall short of perfection. This means that there are legally countenanced levels of risk of harm - as well as levels of risk that the law should not countenance.

[104] My colleague holds that probable factual causation is established where a plaintiff can show that contagion is less likely to occur in a non-negligent system than in an existing negligent one. She finds that if a non-negligent system reduced the risk of general contagion, it can properly be inferred that specific individual contagion within a non-negligent system would be less likely than in a negligent system. "It would be enough", my colleague finds, "to satisfy probable factual causation where the evidence establishes that the plaintiff found himself in the kind of situation where the risk of contagion would have been reduced by proper systemic measures."[184] My disagreement with this finding is three-fold.

[105] First, it is not possible to infer probable factual causation from an increase in exposure to risk by itself. By corollary, where the actual origin of the injury cannot be traced (as with Mr Lee's TB), it is impossible to say that infection was probably caused by a negligent exposure to risk, as opposed to an exposure that no amount of care on the prison authorities' part could have avoided.

[106] Second, the very nature of negligent conduct is that it increases risk and thus makes harm more likely to occur. To infer probable factual causation merely from increased likelihood of harm is to suggest that probable factual causation follows from every finding of negligence. But increased likelihood, or an overall increase in risk, still does not tell us whether the negligent conduct was more probably than not the cause of the specific harm.

[107] Third, the approach entails that factual causation may be inferred from any increase in risk. This is because my colleague's approach leaves no room for assessment of the amount of risk exposure that occurred, how much of it was attributable to the negligence of the defendant, and what level of risk exposure should lead to recovery of compensation. There is thus no room for risk appreciation or possible blame apportionment. The evidence the amici and the prison authorities tendered on appeal illustrates this. Though their respective experts conflicted on certain points, they agreed that, even if minimum South African national standards of incarceration had been implemented, there would still have been a more than 50% risk of transmission of TB in Pollsmoor.[185]

[108] The intricacies of this area of law, in my view, require accommodation of these complexities. Thus, it would seem that on my colleague's approach, if, before proper measures were in place, a prisoner had a 90% chance of contracting TB, but proper measures reduced this to 85%, the fact that the prisoner was negligently exposed to an increase in risk would by itself render the defendant liable. In other words, even if the harm were likely to result despite reasonable measures being taken, the defendant would still be liable because the risk was increased, even if only nominally.

[109] The approach of the Supreme Court of Appeal also shut out consideration of any of these avenues of development. In my view, it erred in doing so. Denying recompense, and insulating the prison authorities from responsibility, merely because Mr Lee was unable to pinpoint the source of his infection, may have trenched upon his constitutional right to security of the person.[186] It may also have impinged on his right to conditions of detention that are consistent with human dignity.[187]

 

[110] The Supreme Court of Appeal, having concluded that the High Court judgment in Mr Lee's favour was unsustainable on the basis of the traditional but-for test for causation, was therefore obliged to consider developing the common law. InCarmichele,[188] this Court held that "where the common law deviates from the spirit, purport and objects of the Bill of Rights the courts have an obligation to develop it by removing that deviation."[189] The Court also required judges to "remain vigilant" and "not hesitate to ensure that the common law is developed to reflect the spirit, purport and objects of the Bill of Rights."[190]

[111] This case is very different from Everfresh,[191] in which a private law litigant asked this Court to develop the common law of contract "where a clause grants both parties the discretion to negotiate"[192] so as to require those parties to do so reasonably and in good faith, but failed to raise the constitutional argument before either the High Court or the Supreme Court of Appeal. The majority of this Court[193] decided that it was not in the interests of justice to entertain the appeal because there were no special circumstances requiring this Court to undertake the development the litigant sought as a court of first and last instance. On the contrary, it would be unfair to the opposing litigant to have to meet a new case, the termination of which would result in no dire consequences for the claimant.[194]

[112] Mr Lee's position is radically different. He seeks development of the common law affecting a vulnerable group to whom our system of constitutional protections owes particular solicitude. In this his plight echoes that of the claimant in Carmichele, whose protection and right to a remedy raised urgent questions of gender justice.[195] And the position of the defendant from whom Mr Lee seeks recompense is also different. The country's interest in the development of a sound system of incarceration, in which risk of exposure to pathogens is minimised as much as is reasonably possible, suggests there may be a need to develop the common law of causation.

 

[113] What the results of that development might be it is not possible or desirable for this Court to say. The defendant strongly argued that Mr Lee had failed to establish his case even if the touchstone of liability were risk-aversion, and not probable causation. This was because Mr Lee was a smoker (which placed him at greater risk of contracting TB),[196] and because he was better-off than most prisoners. For most of his incarceration he enjoyed the special privilege of single cell-accommodation (which he shared with two other prisoners). This meant that he was less exposed to overcrowding, smoke and other prisoners' sputum and coughing.

[114] In my view, it is not possible for this Court, particularly given the truncated record before it, to consider properly and justly all the avenues of possible development, and their implications for the parties' respective cases. In Carmichele, this Court noted the value of "close and sensitive interaction between, on the one hand, the High Courts and the Supreme Court of Appeal which have particular expertise and experience in this area of the law and, on the other hand, this Court."[197] It also noted that litigants may be disadvantaged because they have not had "the opportunity of reconsidering or refining their respective arguments" in the light of previous Supreme Court of Appeal jurisprudence.[198] These considerations apply equally here.

[115] In the circumstances, the just order would be remit the matter to the trial court, for it to consider, in the light of the findings of the Supreme Court of Appeal, and this judgment, the manner in which the common law ought to be developed. It may have been appropriate, depending on the trial court's conclusion on this, to permit the parties to lead further evidence.

For the Applicant:

For the Respondent:

For the Amici

Advocate M Donen SC, Advocate M Bridgman and Advocate L Dzai instructed by Jonathan Cohen & Associates.

Advocate I Jamie SC, Advocate Karrisha Pillay, and Advocate Dheshni Pillay instructed by the State Attorney.

Advocate A Hassim instructed by SECTION27.

Footnotes:


[1] Minister of Correctional Services v Lee 2012 (3) SA 617 (SCA) (Supreme Court of Appeal Judgment).

[2] Lee v Minister of Correctional Services 2011 (6) SA 564 (WCC) (High Court Judgment).

[3] The first-mentioned violation relates to conduct (an act) limiting the applicant's liberty by being detained and the second involves omissions.

[4] The applicant was unable to identify all the people responsible for his treatment while incarcerated at the maximum security prison at Pollsmoor and thus refers to them as the responsible authorities. That includes the Minister for Correctional Services and officers of the Department of Correctional Services responsible for the control and management of the prison during the period of his incarceration.

[5] 111 of 1998 (Act).

[6] The statement of agreed factual findings was lodged following this Court's directions dated 30 May 2012, setting the matter down for hearing on Tuesday, 28 August 2012 and which stated, in relevant part:

"2. The parties are directed to submit by Friday, 22 June 2012 an agreed statement based on the factual findings of the High Court and the Supreme Court of Appeal, setting out the facts that are necessary for this Court to determine the issues before it.

3. Should the parties be unable to agree on a statement of facts as envisaged, they are required to lodge an affidavit setting out-

(a) precisely why they could not agree;

(b) the draft statement proffered by each party;

(c) the facts that are common to each statement;

(d) the factual findings about which they could not agree;

(e) the precise portions of the record that relate to the disputed findings."

[7] Pollsmoor is a prison complex consisting of five different prisons: the admissions centre which is also known as the maximum security prison, the women's prison, the juvenile prison and the medium security prisons B and C for sentenced prisoners.

[8] On charges of, among others, counterfeiting, fraud and money laundering.

[9] High Court Judgment above n 2 at para 50.

[10] Id at para 52.

[11] Id.

[12] Id.

[13] Supreme Court of Appeal Judgment above n 1 at para 18.

[14] Id at para 11.

[15] High Court Judgment above n 2 at para 223.

[16] Id at para 214.

[17] The SCOs have been compiled to give effect to the Act.

[18] Clause 4.1(a) provides that all persons-

"admitted to prison, should be seen on admission by a registered nurse for, inter alia, medical problems, whether acute or chronic."

[19] Clause 4.4(a) provides:

"[A]ll admissions must be screened by a registered nurse on admission using the screening form."

[20] Clause 6.1 provides:

"[F]ollowing screening at the reception, all admissions must be taken to the prison health facility by the unit manager or reception manager within 24 hours, for a medical examination by the registered nurse or medical officer/practitioner as prescribed."

[21] Clause 6.2 provides:

"[a]t prisons where there are primary health care clinics at the housing units, the medical examination may be performed at such clinics."

[22] Clauses 14 and 15 of Chapter 3, in relevant part, make provision for communicable and contagious diseases:

"14.6.1 Whenever there is a suspicion that a prisoner . . . could be suffering from a communicable, or contagious disease. . . [t]he case must immediately be brought to the attention of the Supervisor: Nursing and the attending medical officer/practitioner.

. . .

14.6.3 If the registered nurse or attending medical officer/ practitioner deems it necessary to isolate/segregate the prisoner . . . suspected to be suffering from a communicable, or contagious disease, the recommendations or prescriptions must always be adhered to.

. . .

15.1 All prisoners with communicable conditions must be isolated in strict accordance with the medical officer's/practitioner's and registered nurse's orders issued in each case.

. . .

15.3 Each prison must have written orders on infection control which must be monitored and reviewed annually."

[23] High Court Judgment above n 2 at para 42.

[24] Id at para 232.

[25] The High Court made the following order:

"1. That the [respondent] is declared to be liable to the [applicant] in delict pursuant to the [applicant] having become ill with TB whilst he was incarcerated in the Maximum Security Prison at Pollsmoor.

2. That the Registrar is requested to set the matter down for hearing, in consultation with the Judge President, in order for the parties to lead evidence pertaining to the quantum of the [applicant's] damages in respect of his illness with TB as aforesaid and the sequelae thereof.

3. That [respondent] is to pay the [applicant's] costs of suit as between party and party."

[26] High Court Judgment above n 2 at para 2.

[27] Id at para 6.

[28] Id.

[29] Id.

[30] Section 10 provides that "[e]veryone has inherent dignity and the right to have their dignity respected and protected."

[31] Section 11 provides that "[e]veryone has the right to life."

[32] Section 12(1) provides that:

"Everyone has the right to freedom and security of the person, which includes the right-

(a) not to be deprived of freedom arbitrarily or without just cause;

(b) not to be detained without trial; 

(c) to be free from all forms of violence from either public or private sources;

(d) not to be tortured in any way; and

(e) not to be treated or punished in a cruel, inhuman or degrading way."

[33] Section 35(2)(e) of the Constitution provides:

"Everyone who is detained including every sentenced prisoner, has the right to conditions of detention that are consistent with human dignity, including at least exercise and the provision, at state expense, of adequate accommodation, nutrition, reading material and medical treatment".

[34] Section 2 provides in relevant part:

"The purpose of the correctional system is to contribute to maintaining and protecting a just, peaceful and safe society by-

(a) enforcing sentences of the courts in the manner prescribed by this Act;

(b) detaining all inmates in safe custody whilst ensuring their human dignity".

Section 12 provides:

"(1) The Department must provide, within its available resources, adequate health care services, based on the principles of primary health care, in order to allow every inmate to lead a healthy life.

(2)(a) Every inmate has the right to adequate medical treatment but no inmate is entitled to cosmetic medical treatment at state expense.

(b) Medical treatment must be provided by a correctional medical practitioner, medical practitioners or by a specialist or health care institution or person or institution identified by such correctional medical practitioner except where the medical treatment is provided by a medical practitioner in terms of subsection (3).

(3) Every inmate may be visited and examined by a medical practitioner of his or her choice and, subject to the permission of the Head of the Correctional Centre, may be treated by such practitioner, in which event the inmate is personally liable for the costs of any such consultation, examination, service or treatment.

(4)(a) Every inmate should be encouraged to undergo medical treatment necessary for the maintenance or recovery of his or her health.

(b) No inmate may be compelled to undergo medical intervention or treatment without informed consent unless failure to submit to such medical intervention or treatment will pose a threat to the health of other persons.

(c) Except as provided in paragraph (d), no surgery may be performed on an inmate without his or her informed consent, or, in the case of a minor, without the written consent of his or her legal guardian.

(d) Consent to surgery is not required if, in the opinion of the medical practitioner who is treating the inmate, the intervention is in the interests of the inmate's health and the inmate is unable to give such consent, or, in the case of a minor, if it is not possible or practical to delay it in order to obtain the consent of his or her legal guardian."

[35] Supreme Court of Appeal Judgment above n 1 at para 15.

[36] Directly Observed Therapy Short-Course. Some authors describe DOTS as a good therapy for pan-susceptible TB but that its success depends on the efficacy of the antibiotics used. See in this regard Farmer Pathologies of power - health, human rights, and the new war on the poor, (University of California Press, California 2005) at 124.

[37] High Court Judgment above n 2 at paras 249-50.

[38] Supreme Court of Appeal Judgment above n 1 at para 35.

[39] Id at para 36.

[40] Id at para 42.

[41] 1966 (2) SA 428 (A) at 430E-F.

[42] Supreme Court of Appeal Judgment above n 1 at para 44.

[43] Id at para 38.

[44] Id at para 46.

[45] Id at para 46. See also Minister of Police v Skosana 1977 (1) SA 31 (A) at 35C-D.

[46] 1990 (1) SA 680 (A) (International Shipping). Supreme Court of Appeal Judgment above n 1 at paras 47-50.

[47] Supreme Court of Appeal Judgment above n 1 at para 55.

[48] Id at para 57.

[49] See [15] above.

[50] Supreme Court of Appeal Judgment above n 1 at para 59:

"All that is true but in each case it begs the question what would have been reasonable. So, for example, while proper screening procedures for incoming prisoners are no doubt required, that begs the question what procedures might reasonably be expected in a large and congested prison. And while regular and effective screening of inmates will clearly reduce the risk of contagion, what is reasonably regular and effective when applied to some 4 000 prisoners? It might be tempting to answer those questions by saying that what ought to have been done was everything that would have avoided tuberculosis being transmitted but that would be fallacious. I have already indicated that the prison authorities are not required to guarantee that transmission will not occur: only to take reasonable steps to prevent it."

[51] Id.

[52] Id at para 60.

[53] Id at para 61.

[54] Id.

[55] Id at para 63.

[56] Id at para 64.

[57] Id at paras 53-4.

[58] 2002 (6) SA 431 (SCA). The case concerns a negligent omission on the part of policemen causing physical harm to the claimant. One B, fond of alcohol and guns, shot the respondent in an accident that resulted in B shooting and killing both his wife and daughter and shooting the respondent in the ankle and shoulder. The police had prior knowledge that B had a drinking problem and had threatened to harm not only himself and his wife and children but also the police themselves with the guns that were licensed to him. Section 11 of the Arms and Ammunition Act 75 of 1969 provides that a person like B may be deprived of his gun and licence for a period not less than two years following an enquiry. The police were in the circumstances required to reduce the information in their possession to writing, under oath, and forward that information to the relevant person in charge of the enquiries. They failed to do so and this was the basis for the claim of the negligent omission causing the harm to the claimant.

[59] See also sections 2 and 12 of the Act, quoted in full above n 34.

[60] Section 7(2) of the Constitution provides:

"The state must respect, protect, promote and fulfil the rights in the Bill of Rights."

[61] Section 1(d) of the Constitution provides:

"Universal adult suffrage, a national common voters roll, regular elections and a multi-party system of democratic government, to ensure accountability, responsiveness and openness."

[62] In S v Shaik and Others [2007] ZACC 19; 2008 (2) SA 208 (CC); 2007 (12) BCLR 1360 (CC) at paras 18-9, this Court, relying on Prince v President, Cape Law Society, and Others [2000] ZACC 28; 2001 (2) SA 388 (CC); 2001 (2) BCLR 133 (CC), interpreted the provisions of Rule 31 and held that evidence sought to be adduced under Rule 31 must be incontrovertible or easily verifiable to be admissible. See in this regard Rail Commuters Action Group and Others v Transnet Ltd t/a Metrorail and Others[2004] ZACC 20; 2005 (2) SA 359 (CC); 2005 (4) BCLR 301 (CC) (Rail Commuters) at para 37. See also In re Certain Amicus Curiae Applications: Minister of Health and Others v Treatment Action Campaign and Others [2002] ZACC 13; 2002 (5) SA 713 (CC); 2002 (10) BCLR 1023 (CC) at para 8.

[63] This Court, in Everfresh Market Virginia (Pty) Ltd v Shoprite Checkers (Pty) Ltd[2011] ZACC 30; 2012 (1) SA 256 (CC); 2012 (3) BCLR 219 (CC) (Everfresh) at para 63, held that-

"while there may be cases where the interests of justice require that a constitutional complaint be raised for the first time before this court, these would be rare and exceptional. In Lane and Fey NNO this court set out the proper approach in the following terms:

‘Where the development of the common law is the issue, the views and approach of the ordinary courts, and particularly the SCA, are of particular significance and value. Save in special circumstances, this court should not consider this kind of matter as a court of first instance. No relevant factors have been raised by the applicants that would constitute such special circumstances.'" (Footnotes omitted.)

See also Phillips and Others v National Director of Public Prosecutions [2005] ZACC 15; 2006 (1) SA 505 (CC); 2006 (2) BCLR 274 (CC) at para 44.

[64] See Betlane v Shelley Court CC [2010] ZACC 23; 2011 (1) SA 388 (CC); 2011 (3) BCLR 264 (CC) at para 22; Satchwell v President of the Republic of South Africa and Another [2003] ZACC 2; 2003 (4) SA 266 (CC); 2004 (1) BCLR 1 (CC) at para 6; andVan der Spuy v General Council of the Bar of South Africa (Minister of Justice and Constitutional Development, Advocates for Transformation and Law Society of South Africa Intervening) [2002] ZACC 17; 2002 (5) SA 392 (CC); 2002 (10) BCLR 1092 (CC) at para 19.

[65] Supreme Court of Appeal Judgment above n 1 at para 35.

[66] Id at para 44.

[67] Skosana above n 45 at 34D-E.

[68] Id at 34E-H.

[69] Van Duivenboden above n 58 at para 12.

[70] Rail Commuters above n 62 at para 52.

[71] These theories include the foreseeability theory, adequacy theory and the direct consequences theory. See Neethling et al The Law of Delict 5 ed (LexisNexis Butterworths, Durban 2006) (Neethling) at 160.

[72] Some authors refer to it rather as conditio cum qua non. See Van Oosten De Jure(University of Pretoria, Pretoria 1982) at 257. 

[73] Neethling above n 71 at 160.

[74] International Shipping above n 46 at 700F-H.

[75] Neethling above n 71 at 162. 

[76] See Van der Walt and Midgley Principles of Delict 3 ed (LexisNexis Butterworths, Durban 2005) at 199-201.

[77] The facts in S v Van As en ‘n Ander 1967 (4) SA 594 (AD) (Van As) and other cases including Skosana above n 45 and Siman & Co (Pty) Ltd v Barclays National Bank Ltd 1984 (2) SA 888 (AD) (Siman) are illustrative. In Van As, the policemen neglected to search for children who had fled into the night and later died of exposure. The question was whether the children's death was caused by the omission to search for them. The Court tested the causal connection between the omission and the death by asking whether a reasonable search would have prevented the children's death. It inserted a positive act in the place of the omission. Some authors (Neethling above n 71) are of the view, and I agree, that the insertion of the reasonable conduct of the defendant into the set of facts is said to have the potential to cause confusion between factual causation and negligence. What must be determined first is whether the wrongdoer could have done anything to prevent the relevant consequence (and only then, whether the reasonable person in the position of the defendant would have prevented the consequence).

[78] Siman id at 914F-915H.

[79] Id.

[80] Supreme Court of Appeal Judgment above n 1 at paras 52-4.

[81] Id at paras 57, 60 and 64.

[82] Id at para 47.

[83] Id at paras 48-63.

[84] Id at para 64.

[85] Kakamas Bestuursraad v Louw 1960 (2) SA 202 (A) (Kakamas) at 220B-C.

[86] Id. See also Smith, Hogg and Company, Limited v Black Sea and Baltic General Insurance Company, Limited 1940 AC 997 at 1003-4.

[87] Siman above n 77.

[88] Id at 917H-918A.

[89] Van Duivenboden above n 58.

[90] Minister of Finance and Others v Gore NO 2007 (1) SA 111 (SCA).

[91] [2005] 4 All ER 812; [2005] UKHL 2 at para 202.

[92] Gore above n 90 at para 33.

[93] Siman above 77.

[94] Siman above n 77 at 914F-915G. 

[95] Id. See also Kakamas above n 85 at 220B-C. 

[96] Id at 907E.

[97] Id at 916A-917H.

[98] Kakamas above n 85 and Van As above n 77.

[99] Smith v Auckland Hospital Board [1965] NZLR 191.

[100] Van der Walt and Midgley "Delict" 8(1) The Law of South Africa 2005 (LAWSA).

[101] Id at para 130 in fn 15.

[102] Corbett JA acknowledged this again in International Shipping above n 46 at 700G: "This enquiry may involve the mental elimination of the wrongful conduct and the substitution of a hypothetical course of lawful conduct and the posing of the question as to whether upon such an hypothesis plaintiff's loss would have ensued or not." (My emphasis.)

[103] Minister of Safety and Security and Another v Carmichele 2004 (3) SA 305 (SCA) at para 59.

[104] Siman above n 77.

[105] International Shipping above n 46.

[106]Administrateur, Natal v Trust Bank van Afrika Bpk 1979 (3) SA 824 (A) and Minister van Polisie v Ewels 1975 (3) SA 590 (A).

[107] International Shipping above n 46 at 701F-G.

[108] Ewels above n 106.

[109] Id at 597A-C.

[110] F v Minister of Safety and Security and Others [2011] ZACC 37; 2012 (1) SA 536 (CC); 2012 (3) BCLR 244 (CC) at para 119 and the cases cited in fn 46.

[111] Id at para 119 and the cases cited in fn 47.

[112] Siman above n 77 at 915F-G.

[113] Minister of Justice v Hofmeyr 1993 (3) SA 131 (AD) at 141C-142C and 146C-D.

[114] Quoted in full above n 33.

[115] See also section 2 of the Act, the relevant part of which is quoted above n 34.

[116] International Shipping above n 46 at 700G.

[117] Id at 702A.

[118] Supreme Court of Appeal Judgment above n 1 at paras 63-4.

[119] Section 35(2)(e), quoted in full above n 33.

[120] Supreme Court of Appeal Judgment above n 1 at para 64.

[121] Cameron J points out, at [84] below, that there can never be scientific certainty in identifying the source of infection in the case of TB. That may be so. However, that standard of absolute certainty is too exacting and is not the one that our law requires.

[122] The relevant clauses of the SCOs are set out in full above n 18 to n 22.

[123] Compare Olitzki Property Holdings v State Tender Board and Another 2001 (3) SA 1247 (SCA) at para 12.

[124] Although the respondent sought to argue that the SCOs were not applicable under the previous legislation, the repealed Correctional Services Act 8 of 1959 (1959 Act), it is common cause that the SCOs in terms of the Act are identical to those that applied in terms of the 1959 Act. See in this regard the High Court Judgment above n 2 at para 68.

[125] Van Duivenboden above n 58.

[126] Section 7(2) provides:

"That the state must respect, protect, promote and fulfil the rights in the Bill of Rights."

[127] Van Duivenboden above n 58 at para 20.

[128] Supreme Court of Appeal Judgment above n 1 at para 36.

[129] Van Duivenboden above n 58 at para 21.

[130] Supreme Court of Appeal Judgment above n 1 at para 42.

[131] Id at para 36.

[132] Id at para 42.

[133] Id at para 58.

[134] Neethling above n 71 at 160.

[135] Supreme Court of Appeal Judgment above n 1 at para 37.

[136] [2008] ZACC 8; 2008 (5) SA 94 (CC); 2008 (9) BCLR 865 (CC).

[137] Id at para 80.

[138] Id.

[139] Supreme Court of Appeal Judgment above n 1 at paras 66-8.

[140] Sidumo and Another v Rustenburg Platinum Mines Ltd and Others [2007] ZACC 22; 2008 (2) SA 24 (CC); [2007] 12 BLLR 1097 (CC) at para 232. See also Van Duivenboden above n 58 at para 21.

[141] F v Minister of Safety and Security and Others above n 110 at para 123 in fn 53.

[142] At [94]-[102] below.

[143] International Shipping above n 46 at 701F-G.

[144] Lee v Minister of Correctional Services 2011 (6) SA 564 (WCC) (High Court judgment).

[145] Minister of Correctional Services v Lee 2012 (3) SA 617 (SCA) (Supreme Court of Appeal judgment).

[146] The parties agreed that, to the extent that any system existed at all for managing TB, "its application in practice was at best sporadic and in at least some respects effectively non-existent" (echoing the Supreme Court of Appeal judgment id at paras 44 and 67 that TB management was virtually "non-existent"), and that the spread of TB was "facilitated by the prevailing conditions in the prison". The parties agreed further that it "is more probable than not that [Mr Lee] contracted TB as a result of his incarceration" in Pollsmoor (echoing the Supreme Court of Appeal judgment id at paras 52-3).

[147] As the Supreme Court of Appeal judgment observed id at para 55, "[p]roof alone that reasonable precautions were not taken to avoid foreseeable harm, and that the harm occurred, does not establish that the former caused the latter."

[148] Id at paras 52-3.

[149] Id at para 54.

[150] Id at para 56 ("[l]ife has many hazards that will not be avoided even when reasonable steps are taken to do so") and at para 59.

[151] Id at para 59.

[152] The Supreme Court of Appeal judgment id at para 60 mentioned incidence, which is the rate at which a disease is transmitted in a vulnerable population. To this should be added prevalence, which is a current snapshot of how many in that population have the disease.

[153] Id at para 60.

[154] Id.

[155] Id at para 61.

[156] According to the parties' agreed statement of facts: "[t]ransmission in human beings occurs by inhalation of the organism".

[157] According to the parties' agreed statement of facts: "[s]ome persons who are ill with tuberculosis shed more bacteria than others and are known as ‘super shedders'." (Footnote omitted.)

[158] Supreme Court of Appeal judgment above n 145 at para 7.

[159] According to the parties' agreed statement of facts: "[a] negative test confirms the absence of organisms from the sample, but the sample may not be representative of the host, which means that a negative test is not confirmation of the absence of the disease."

[160] The parties' agreed statement of facts stipulated that "whatever management strategies are put in place there will always be a risk of infection if only because diagnosis is a precursor to intervention, and the disease might often be diagnosed well after the prisoner has become contagious." (Footnote omitted.)

[161] Supreme Court of Appeal judgment above n 145 at para 61 ("I do not think the prison authorities can reasonably be expected to examine some 4000 prisoners with such regularity and thoroughness that tuberculosis will always be detected before the prisoner becomes contagious"). The parties' agreed statement of facts echoes the High Court and the Supreme Court of Appeal judgments' findings on this score and others.

[162] Supreme Court of Appeal judgment above n 145 at para 63.

[163] The parties' agreed statement of facts noted that had Mr Lee known the source of his infection, "he might have established a causal link between the infection and specific negligent conduct on the part of the authorities."

[164] Supreme Court of Appeal judgment above n 145 at para 64.

[165] Id.

[166] Id at paras 62-3.

[167] Id at para 64.

[168] Id at paras 56 and 59.

[169] Fairchild v Glenhaven Funeral Services Ltd and Others; Fox v Spousal (Midlands) Ltd; Matthews v Associated Portland Cement Manufacturers (1978) Ltd and Others [2002] UKHL 22, [2003] 1 AC 32.

[170] McGhee v National Coal Board [1973] 1 WLR 1, [1972] 3 All ER 1008 (HL). InMcGhee, the employer failed to enable employees to wash off brick dust before leaving work. The claimant contracted dermatitis, but could not prove that the employer's omission was the sole or main cause of his injury. The House of Lords held that the employer's breach of duty had "materially increased the risk" of dermatitis, and that the claim should therefore succeed.

[171] See Fairchild above n 169 at para 7D-E, where, in speaking of the development of a mesothelioma caused by asbestos exposure, it was said: "the condition may be caused by a single fibre, or a few fibres, or many fibres: medical opinion holds none of these possibilities to be more probable than any other, and the condition once caused is not aggravated by further exposure."

[172] [84] above.

[173] Fairchild above n 169.

[174] See Peel "The Law of Tort: Recent Developments and Current Issues" (unpublished Judicial Studies Board guidance, May 2012).

[175] Sanderson v Hull [2008] EWCA Civ 1211 (CA).

[176] Barker v Corus UK Ltd; Murray v British Shipbuilders (Hydrodynamics) Ltd and Others; Patterson v Smiths Dock Ltd and Another [2006] UKHL 20, [2006] 2 WLR 1027.

[177] It is important to note that the immediate effect of apportionment underBarker id was reversed in England by section 3 of the Compensation Act 2006 but only in relation to mesothelioma cases. See Peel above n 174.

[178] Compare Wilsher v Essex Area Health Authority [1988] AC 1074 (HL).

[179] See Herskovits v Group Health Cooperative of Puget Sound 99 Wash.2d 609, 664 P.2d 474 (1983) at para 477. Here, the court affirmed the but-for test in general, but held that a patient, with a less than 50% chance of survival on admission, had a cause of action against the hospital and its employees if they were negligent in diagnosing a lung cancer that reduced his chances of survival by 14%, because otherwise there "would be a blanket release from liability for doctors and hospitals any time there was less than a 50 percent chance of survival, regardless of how flagrant the negligence."

[180] See Summers v Tice et al 33 Cal.2d 80, 199 P.2d 1 (1948). Here, a claimant was injured by a bird shot discharged from a shotgun, but could not show which of two defendants had fired the shot that actually caused the injury. It was held that the onus shifted to the two defendants. In Sindell v Abbott Laboratories et al; Rogers v Rexall Drug Company et al 26 Cal.3d 588, 607 P.2d 924, 163 Cal.Rptr. 132 (1980), it was said that the burden reverses only where there is a substantial likelihood that the defendant is the guilty tortfeasor. However, the California Supreme Court rejected reversal entirely in Rutherford v Owens-Illinois Inc 941 P.2d 1203 (Cal.1997) holding that the burden of proof for causation remains with the plaintiff even in cases of cancer caused by asbestos.

[181] See [65] and [69] above. See Olitzki Property Holdings v State Tender Board and Another 2001 (3) SA 1247 (SCA) and Minister of Safety and Security v Van Duivenboden2002 (6) SA 431 (SCA).

[182] Supreme Court of Appeal judgment above n 145 at para 36.

[183] Id at para 42. Van Zyl Smit and Snacken Principles of European Prison Law and Policy: Penology and human rights (Oxford University Press, Oxford 2011) at 147 rightly consider health care "one of the most important determinants of the general conditions of imprisonment."

[184] [60] above.

[185] The amici tendered a study of the trial court record by Professor Robin Wood and others, published as Johnstone-Robertson et al "Tuberculosis in a South African prison - a transmission modelling analysis" (2011) 101 SAMJ 809. The main judgment holds that this evidence does not meet the threshold requirements of being incontrovertible or common cause as envisaged in Rule 31. However, it usefully illustrates the difficulties that arise when risk exposure is telescoped into proof of factual causation. Professor Wood and his co-authors found that conditions prevailing in Pollsmoor prison are extremely conducive to TB transmission, resulting in exceptionally high annual TB transmission risks of 90%. However - and this is the point - implementing current national cell occupancy recommendations alone would reduce transmission probabilities by 30%; implementation of active case finding, together with the implementation of current national cell occupancy recommendations, would reduce transmission by 50%. In other words, even upon implementing good practice, there would still be a high risk of TB transmission. Implementation of active case finding, along with international recommendations, could reduce transmission by as much as 94%. The study did not examine the reasonableness of the measures assessed. The expert on behalf of the respondent, Professor Dheda, stated that even if the minimal South African standards of incarceration were implemented, thereby leading to substantially less over-crowding, "there would still be an appreciable risk of transmission in the order of more than 50%." Accordingly, he found, "it cannot be said that it is more likely than not, if the South African minimal standards of incarceration had been applied, that Mr Lee would not have had TB transmitted to him." (At para 8 of the report submitted to the Court by Professor Dheda.)

[186] Section 12(1) of the Bill of Rights.

[187] Section 35(2)(e) of the Bill of Rights.

[188] Carmichele v Minister of Safety and Security and Another (Centre for Applied Legal Studies Intervening) [2001] ZACC 22; 2001 (4) SA 938 (CC); 2001 (10) BCLR 995 (CC).

[189] Id at para 33.

[190] Id at para 36. The Court also stressed at para 39 that-

"the obligation of Courts to develop the common law, in the context of the section 39(2) objectives, is not purely discretionary. On the contrary, it is implicit in section 39(2) read with section 173 that where the common law as it stands is deficient in promoting the section 39(2) objectives, the Courts are under a general obligation to develop it appropriately. We say a ‘general obligation' because we do not mean to suggest that a court must, in each and every case where the common law is involved, embark on an independent exercise as to whether the common law is in need of development and, if so, how it is to be developed under section 39(2). At the same time there might be circumstances where a court is obliged to raise the matter on its own and require full argument from the parties."

This construction of section 39 of the Bill of Rights has been strenuously contested - see Fagan "The secondary role of the spirit, purport and objects of the Bill of Rights in the common law's development" (2010) 127 SALJ 611; Davis "How many positivist legal philosophers can be made to dance on the head of a pin? A reply to Professor Fagan" (2012) 129 SALJ 59; and Fagan "A straw man, three red herrings, and a closet rule-worshipper - a rejoinder to Davis JP" (2012) 129 SALJ 788.

[191] Everfresh Market Virginia (Pty) Ltd v Shoprite Checkers (Pty) Ltd [2011] ZACC 30; 2012 (1) SA 256 (CC); 2012 (3) BCLR 219 (CC) (Everfresh).

[192] Id at para 61.

[193] Judgment of Moseneke DCJ, in which Ngcobo CJ, Cameron J, Jafta J, Khampepe J, Nkabinde J and Van der Westhuizen J concurred.

[194] Everfresh above n 191 at paras 64-6.

[195] Carmichele above n 188 at para 62.

[196] Stipulated by the parties in their agreed statement of facts.

[197] Carmichele above n 188 at para 55 (footnotes omitted).

[198] Id at para 59.

Click here to sign up to receive our free daily headline email newsletter