The ‘But For’ Test’s Adequacy in Negligence Cases

Introduction

Negligence law is arguably the most important tort; however, it is a source of significant angst in the modern tort system. Causation in negligence law is laden with fundamental concepts of significant complexity that help to make a connection between a defendant’s wrongdoing and harm to a claimant.¹ It has to do with the fact that adjudging negligence is an extremely difficult task, especially in a case of ambiguous causation. This paper aims to critically examine the adequacy of the ‘but for’ test to establish proof of factual causation in negligence cases.

Negligence Torts

As a starting point of discussion about negligence torts, it is necessary to consider what constitutes negligence. According to Elliott and Quinn, negligence is “a breach of legal duty to take care, with the result that damage is caused to the claimant.”² It should be noted that a legal concept of a duty of care regulates whether a person should be liable for negligence. Therefore, the tort of negligence involves establishing the connection between the behavior of the defendant and the risk of causing harm to the claimant that should have been foreseen.³ In order to prove negligence, the claimant has to prove its four components: the legal duty of care, breach of the duty of care, causation between the breach and the damage suffered by the claimant, and that suffered damage was not too remote.⁴

Before establishing whether a duty of care has been breached, it is necessary to determine if the defendant owed it to the claimant. To this end, the courts recognize established duty situations such as employer to employee or determine the existence of a duty of care following the neighbor principle. The neighbors, according to Lord Arkin are “persons who are closely and directly affected by my act that I ought reasonably to have them in my contemplation as being so affected when I am directing my mind to the acts or omission which are called in question.”⁵ Therefore, Lord Macmillan stated that there is no limit to categories of negligence that could be formulated according to the neighbor principle.⁶ The neighbor principle has been reformulated to include the requirement of proximity by Caparo v. Dickman.⁷ Therefore, in order to establish proximity, it is necessary to show “that the defendant had a measure of control over and responsibility for the potentially dangerous situation.”⁸

After establishing the existence of a duty of care, it is necessary to determine whether a breach of duty has taken place. To this end, the conduct of a defendant is measured against requirements of the standard of care which is described in Blyth v. Birmingham Waterworks.⁹ The case contains a legal principle called ‘the reasonable person’ that helps to determine whether “the man in the street” would have foreseen the outcome of a situation that is being discussed in a particular case.¹⁰ The last part in the analysis of a negligence tort is showing the connection between the defendant’s breach of the standard of care and harming the claimant. The following part of the paper will examine the adequacy of the ‘but for’ test for the purpose of establishing proof of factual causation in negligence torts.

The ‘But For’ Test

The fourth element in a tort of negligence is causation. Causation helps to prove that the harm to the claimant was caused by the breach of duty perpetrated by the defendant.¹¹ After establishing that the damage occurred, it is necessary to determine whether the requirement of proximity allows holding the defendant liable for it. It helps to restrict the defendant’s liability to the point beyond which their actions could not have contributed to causing the claimant’s damage.¹²

The ‘but for the test is the first step in the causation analysis. It helps to determine whether “the defendant’s breach of duty is a factual cause of the claimant’s damage.”¹³ In other words, ‘but for’ test is a useful way of asking whether the claimant’s damage “would have occurred ‘but for’ the defendant’s negligence”¹⁴ Barnett v. Chelsea and Kensington Hospital Management Committee is an example of the application of the ‘but for’ test.¹⁵ Three patients were turned away from an emergency department managed by the defendants. The patients experienced severe poisoning by arsenic that killed one of the five hours later. Nield J concluded that the deceased would have died even if the doctor had diagnosed and treated him.¹⁶ This case beautifully illustrates that ‘but for’ tests link the notion of responsibility with the notion of negligence. However, not all scenarios are similar to the one that occurred in the case described above in that they do not present the opportunity to easily trace the event causing the eventual damage. According to Professor Stapleton, ‘but for’ test is “notoriously inadequate” due to the fact that “the law is also concerned with factors that contribute to an outcome but which are neither necessary nor sufficient for it to exist.”¹⁷ For illustration of her point, she offers a story of five members of a committee governing a club that took a unanimous vote to expel the club’s members. A majority of three votes would have sufficed for a motion to pass; therefore, the fourth vote was unnecessary. The same could be said about any of the four votes. In this situation, it is difficult to establish a factual cause of damage to the expelled club member because each individual vote was causal to the passing of the motion.¹⁸ Therefore, Stapleton rejects the idea of the adequacy of the ‘but for’ test in the situations of “duplicate necessity.”¹⁹

The Degree of Probability of Damage

In case of the balance of probabilities, the degree of probability of damage as proof of causation has to be established. In order for the claimant to succeed, they have to show that “the defendant has materially increased the risk of damage occurring by his negligent conduct.”²⁰ The case of McGhee is a perfect illustration of this approach to the ‘but for’ test.²¹ The claimant worked in hot and dusty conditions of brick kilns owned by the defendant. He was habitually exposed to brick dust, which resulted in him contracting a skin disease. The claimant argued that had the defendant provided their workers with showers he would not have contracted the disease. However, there was no medical evidence confirming that the lack of washing facilities caused the damage to the claimant. Therefore, the House of Lord decided to alter the ‘but for’ test to prevent the defendant from escaping liability because of the lack of conclusive medical evidence.²² McGhee established a new approach to the ‘but for’ test that was used in Page v Smith in which the claimant had to show that the actions of the defendant had significantly contributed to the recurrence of his physical illness.²³ The same modification of the ‘but for’ test was applied to Bailey v Ministry of Defence in which the defendant was held liable for the material contribution to the brain damage suffered by the claimant.²⁴

Criticism of the ‘But For’ Test

According to Clarke, every instance of analyzing the relevant circumstances in which a particular phenomenon occurred includes an inference. ²⁵ Therefore, the ‘but for’ test “fails to satisfy an essential requirement for any test” because it cannot be used as “an independent method of verifying of confirming what one wishes to test.”²⁶ Clarke’s argument is based on the following premise: the process of judging causation using the ‘but for’ test involves asking if an event would have occurred if it had not been preceded by another event; therefore, it does not help to uncover new information that could help to answer that question. Clarke states that in order to establish causation one has to operate with “observed patterns of the concurrence of types of event” and evidence suggesting that the case that is being analyzed coincides with one of those patterns.²⁷ Therefore, the ‘but for’ test cannot serve the purpose of establishing causation as it does not help to uncover new evidence but rather assists in making inferences by drawing conclusions based on previous experience. Clarke contests the adequacy of the ‘but for’ test stating that science is known to create theories explaining connections between two types of events where there is none.²⁸ Taking into consideration the fact that one precondition is rarely enough for the occurrence of a particular event, it is impossible to confine the list of necessary preconditions without sliding into “countless distinct, converging causal chains.”²⁹ It is necessary to draw a distinction between conditions for the occurrence of an event and its causes. However, Clarke argues that the ‘but for’ test cannot serve as a normative filter helping to select legally relevant causes because it is too inclusive. At the same time, it is too exclusive in that it fails to include simultaneous over-determination. Clarke states that regardless of what condition one chooses as sufficient for the occurrence of a particular event “it may not, in fact, be necessary because a sufficiently similar effect could have resulted from a different set of conditions.”³⁰ Sienkiewicz v. Greif is a classic example of a case that defies a basic rule of causation because of the number of competing factors each of which could have been a real cause of the claimant’s harm.³¹ The case involves a worker who died of mesothelioma due to tortious exposure to asbestos during her employment at a factory of the defendant. However, she was also exposed to asbestos in the atmosphere; therefore, it was impossible to establish what brought about the illness. Contrary to the statistical evidence of 85 percent chance that the worker died from a cause that was not tortuous, the court held that “her estate should recover in full from the defendants as if they were the sole cause.”³² The exceptional rule of causation was applied in Sienkiewicz as well. The language of a contribution that materially increased the risk of damage that was discussed in the previous section of the paper originated in Duke of Buccleuch v. Cowan.³³ The case involved numerous pollutants of a river all of whom were responsible for a spoiled state of the water. The language was later applied in Wardlaw v. Bonnington Castings in a context similar to that of Sienkiewicz.³⁴ The claimant’s harm was brought about by the inhalation of factory dust that originated from a breach of duty and one that was not a result of negligence. The House of Lords held that the dust resulting from the breach of duty produced the material contribution to the claimant’s pheumoconiosis. According to Steel and Ibbetson, the decision departed from a “common-sense application of the normal rules” in that it stretched “the conceptual component of causation.”³⁵ On the other hand, it cannot be argued that the presence of non-tortuous causes prevents one from being liable for the production of tortuous causes. Therefore, even if the defendant did not create a cause in a but-for sense, he still produced the cause in the sense of contribution. However, an important distinction between creating a case and creating a risk of damage has to be made. Unlike Wardlaw, the decision in McGhee substantially departed from the orthodox application of the rules of causation.³⁶ Essentially, it equated a material increase of risk with a material contribution to damage. Therefore, it could be argued that probabilistic explanations for a necessary condition of the occurrence of an event is a significant downside of the ‘but for’ test.

Conclusion

The ‘but for’ test can rarely help to establish causation in the law of torts with a high degree of certainty without resorting to a deterministic worldview. Even though it can be successfully applied to numerous cases, its conceptual limitations lead to the creation of explanatory connections between, sometimes, unrelated phenomena. It fails to distinguish between a material increase of risk and a material contribution to damage; therefore, its application can be philosophically justified only for simple cases with an extremely limited number of independent explanatory elements.

Footnotes

1. Kirsty Horsey and Erika Rackley, Tort Law (4^th edn, Oxford University Press 2015) 268
2. Catherine Elliott and Frances Quinn, Tort Law (10^th edn, Pearson 2015) 17; Kirsty Horsey and Erika Rackley, Kidner’s Casebook on Torts (13^th edn, Oxford University Press 2015) 17.
3. Emily Finch, Law Express (1st edn, Pearson Education Limited 2016) 3.
4. ibid 3.
5. Donoghue v Stevenson [1932] AC 562
6. ibid 562.
7. Caparo Industries plc v Dickman [1990] 2 AC 605 (HL)
8. Finch (n 3) 8.
9. Blyth v Birmingham Waterworks (1856) 11 Exch 781
10. Hall v Brooklands Auto-Racing Club [1933] 1 KB 205 (CA)
11. Chris Turner, Unlocking Torts(4^th edn, Routledge 2014) 2; Nicholas McBride and Roderick Bagshaw, Tort Law (5^th edn, Pearson 2015) 4.
12. John Cooke, Law of Tort (12^th edn, Pearson 2015) 167
13. ibid 167.
14. Jenny Steele, Tort law (1^st edn, Oxford University Press) 177
15. Barnett v Chelsea and Kensington Hospital Management Committee [1969] 1 QB 428
16. Barnett v Chelsea and Kensington Hospital Management Committee (n 15).
17. Sarah Green, Causation in Negligence (Bloomsbury) 19
18. ibid 19.
19. Ibid.
20. Cooke (n 12) 167.
21. McGhee v National Coal Board [1973] 1 WLR 1
22. ibid [1].
23. Page v Smith [1996] AC 155 (HL) 181
24. Bailey v Ministry of Defence [2013] UKSC 41
25. Desmond Clarke, ‘Causation and Liability in Tort Law’ (2014) 5 J 217
26. Clarke (n 25) 225.
27. ibid 225.
28. ibid 226.
29. ibid.
30. Clarke (n 25) 227.
31. Sienkiewicz v Greif [2011] UKSC 10
32. Sandy Steel and David Ibbetson, ‘More Grief on Uncertain Causation in Tort’ (2011) 70 CLJ 451
33. Duke of Buccleuch v Cowan [1866] HL 14
34. Wardlaw v Bonnington Castings [1956] AC 613
35. Sandy Steel and David Ibbetson (n 31) 451.
36. ibid 451.