Factual causation: One size does not fit all

December 1st, 2013

By Jerome Veldsman

The judgment of Lee v Minister for Correctional Services 2011 (6) SA 564 (WCC) may be an opportunity missed by both the Supreme Court of Appeal (SCA) (see Minister of Correctional Services v Lee 2012 (3) SA 617 (SCA)) and the Constitutional Court (CC) (see Lee v Minister of Correctional Services 2013 (2) SA 144 (CC)) to develop the South African law of delict with regard to factual causation. In the light of four starkly different judgments, namely, one in the High Court; one in the SCA; and two in the Constitutional Court (Nkabinde J and Cameron J), in relation to the same (mainly undisputed) facts, such development seems necessary or at least expedient.

A distinction between two different factual casual patterns is a useful start to this discussion:

  • In the majority of delictual claims, the facts are such that the actual factual cause of the plaintiff’s injury can be determined on a balance of probabilities (ie, straightforward cases).
  • However, in some delictual claims, there are two or more competing but independent possible causes for the plaintiff’s injury, only one of which is a result of the defendant’s negligence; and the others are caused by some other factor unrelated to the defendant’s (negligent) conduct. Further, there is an admitted lack of sufficient evidence to determine on a balance of probabilities which of the possible causes actually caused the plaintiff’s injury. In ‘Ambiguous Cause-in-Fact and Structured Causation: A Multi-Jurisdictional Approach’ (2003) Texas International Law Journal vol 38: no 249, Erik S Knutsen terms such factual patterns ‘ambiguous cause-in-fact cases’ (http://papers.ssrn.com, accessed 4-11-2013).

A reason for the distinction in certain other common law jurisdictions between straightforward cases and ambiguous cause-in-fact cases is to treat an ambiguous cause-in-fact case differently with regard to the determination of factual causation. In an ambiguous cause-in-fact case the ‘but-for’ test inevitably non-suits the plaintiff, as he or she (by definition) fails to prove on a balance of probabilities that the defendant’s (proven) negligence was a causa sine qua non of his or her injury. The distinction is not generally overtly made in South Africa, but was alluded to in Siman & Co (Pty) Ltd v Barclays National Bank Ltd 1984 (2) SA 888 (A) at 915.

The Lee case is a ‘classic’ ambiguous cause-in-fact case. The applicant was arrested in 1999 and detained in prison awaiting trial. Save for a four-month period out on bail in 2000, the applicant was detained until his acquittal in 2004. In 2003 the applicant became ill and was diagnosed as suffering from pulmonary tuberculosis (TB). He instituted a delictual action against the Minister of Correctional Services for damages on the basis that the poor prison health management caused his infection with TB.

The negligence of the Department of Correctional Services was apparent – it had breached its duty of care towards the applicant with regard to proper prison health management – and its conduct was wrongful. However, many prisoners become infected with TB in prison, even under proper prison health management. In other words, either the Department of Correctional Services’ negligence, or an unavoidable ‘fact-of-life’ risk of contagion, caused the applicant’s infection; and medical science cannot reliably predict whether or not proper prison health management would have prevented the applicant’s infection.

The High Court applied the common law ‘but-for’ test to determine factual causation, and found in favour of the applicant: ‘On the totality of the evidence, I am accordingly satisfied that it is more probable than not that the plaintiff contracted TB as a result of his incarceration in the maximum security prison at Pollsmoor’ (at para 236). In brief, the High Court (as to the facts, incorrectly) dealt with the matter as a straightforward case rather than as an ambiguous cause-in-fact case, but achieved a just outcome.

On appeal, the Supreme Court of Appeal (SCA), in a unanimous judgment, applied the standard common law ‘but-for’ test to determine factual causation, and found in favour of the Minister: ‘The difficulty that is faced by Mr Lee 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 prison 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 I think that the claim ought to have failed’ (at para 64). In brief, the SCA (as to the facts, correctly) dealt with the matter as an ambiguous cause-in-fact case, but achieved an unjust outcome.

All the judges in the Constitutional Court (CC) disagreed with the unjust outcome in the SCA, but the two judgments agreed on little else.

Cameron J, writing for the minority in the CC, held that the SCA had correctly applied the common law ‘but-for’ test (implying that, in his view, such test is the be all and end all for factual causation), and that the common law ought to be developed to prevent the unjust outcome of the SCA judgment. Cameron J, having conducted an analysis of some foreign judgments dealing with ambiguous cause-in-fact cases, would have ordered that the matter be remitted to the trial court to develop the common law.

Nkabinde J, writing for the majority in the CC, found in favour of the applicant, and held that the SCA had incorrectly applied the common law ‘but-for’ test (and also implied that such test is not the be all and end all for factual causation), and that the case does not require any development of the common law. However, as is argued below, Nkabinde J did indeed, at least obiter, develop the common law in respect of ambiguous cause-in-fact cases.

Nkabinde J endorsed the High Court finding as follows: ‘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’ (at para 55).

However, that is not exactly what the High Court did (refer to the quotation from the High Court finding above). The High Court did not find factual causation based on an increase in risk to the applicant; it found factual causation per se (I submit, on the facts, incorrectly so). Nkabinde J does not dispute the findings of the SCA and Cameron J that medical science cannot reliably predict whether or not proper prison health management would have prevented the applicant’s infection.

One of the solutions in certain other common law jurisdictions for the inadequacy of the ‘but-for’ test in ambiguous cause-in-fact cases is to substitute a ‘material increase in risk test for the “but-for” test’. In the case of Clements v Clements 2012 SCC (at paras 13 and 14) the Supreme Court of Canada made the following judgment: ‘Exceptionally, however, courts have accepted that a plaintiff may be able to recover on the basis of “material contribution to risk of injury”, without showing factual “but for” causation. … [T]his can occur in cases where it is impossible to determine which of a number of negligent acts by multiple actors in fact caused the injury, but it is established that one or more of them did in fact cause it. In these cases, the goals of tort law and the underlying theory of corrective justice require that the defendant not be permitted to escape liability by pointing the finger at another wrongdoer. Courts have therefore held the defendant liable on the basis that he materially contributed to the risk of the injury. “But for” causation and liability on the basis of material contribution to risk are two different beasts. “But for” causation is a factual inquiry into what likely happened. The material contribution to risk test removes the requirement of “but for” causation and substitutes proof of material contribution to risk.’

In Sienkiewicz v Greif (UK) Ltd; Knowsley MBC v Willmore [2011] UKSC 10 the United Kingdom (UK) Supreme Court applied the ‘material contribution to risk’ test as a substitute for the ‘but-for’ test to a single wrongdoer (the Sienkiewicz case is a ‘follow-up’ to Fairchild v Glenhaven Funeral Services Ltd [2002] UKHL 22, discussed by Cameron J). The simplified facts were that the plaintiff died of mesothelioma, a rare form of cancer commonly caused by exposure to asbestos. In common with other inhabitants of the local area, she would have been exposed to a low level of asbestos in the general atmosphere. However, the manufacturing process at her place of employment released asbestos dust into the atmosphere. The court held that the contribution to risk was sufficient to constitute factual causation.

In the United States of America (USA), the equivalent ‘material contribution to risk rule’ in Rutherford v Owens-Illinois Inc (1997) 16 Cal. 4th 953 – the plaintiff’s exposure to a particular product was a substantial factor in causing or bringing about the disease if in reasonable medical probability it contributed to the plaintiff’s risk of developing cancer – has been applied to a single wrongdoer (see Kennedy v Southern California Edison Co, 219 F. 3d 988 (9th Cir. 2000)).

Nkabinde J appears to endorse the ‘material contribution to risk’ approach in respect of a single wrongdoer: ‘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’ (at para 60). Cameron J (in paras 105 – 108) disagreed with the material increase in risk approach as enunciated by Nkabinde J. His criticism includes the concern that a miniscule increase in an already substantial risk would result in (full) liability. Such matter is dealt with in the Sienkiewicz case as follows: ‘I doubt whether it is ever possible to define, in quantitative terms, what for the purposes of the application of any principle of law, is de minimis. This must be a question for the judge on the facts of the particular case’ (at para 108).

Technically, ‘… the material contribution to risk exception to “but for” causation is not a test for proving factual causation, but a basis for finding “legal” causation where fairness and justice demand deviation from the “but for” test’ (the Clements case at para 45). Note the criticism of Nkabinde J (at para51) on blurring of the distinction between factual and legal causation.

The Clements case concerned a motorcycle accident, and caution was expressed about the general extension of the Sienkiewicz case (which is restricted in the UK to mesothelioma cases). The Sienkiewicz case contains the following statement: ‘Although, therefore, mesothelioma claims must now be considered from the defendant’s standpoint a lost cause, there is to my mind a lesson to be learned from losing it: the law tampers with the “but for” test of causation at its peril’ (at para 186).

Plaintiff lawyers, especially in the USA, view the ‘material contribution to risk’ approach with glee. It would perhaps be wrong, albeit plausible, to seize on the Lee case as introducing this approach into South African law. However, there may be a risk that the material contribution to risk approach could open the floodgates of litigation.

In the law of delict/tort, subtle differences between different legal systems make reliable borrowings and comparisons challenging. In the light of the complexity in and volume of relevant foreign judicial authority, our law of delict could benefit from academic assessment of factual causation in ambiguous cause-in-fact cases in a South African context, including consideration of –

  • the ‘material increase in risk’ approach;
  • the ‘reversal’ approach; reversing the burden of proof of causation to the defendant; and
  • the ‘inference’ approach; inferring causation from the facts of the case (similar to the ‘common sense’ approach alluded to by Nkabinde J in the Lee case).

An ideal legal position would be more predictable than an imprecisely formulated ‘material increase in risk’ approach or ‘common sense’ approach.

Jerome Veldsman BCom LLB (Stell) is an attorney at Walkers Inc in Cape Town.

This article was first published in De Rebus in 2013 (Dec) DR 32.