Tag Archives: causation

Causation, causation, causation: NSW v Mikhael

In a NSW Court of Appeal decision handed down last week, New South Wales v Mikhael [2012] NSWCA 338, the causation issue once again proved to be the plaintiff’s Achilles’ heel.

This is the second time this year in which the Court of Appeal has overturned a first instance decision, whilst still upholding the trial judge’s decision on duty of care and breach (cf: Garzo v Liverpool/Campbelltown Christian School Ltd [2012] NSWCA 151, in respect of which there is a pending special leave application to the High Court).

It’s as if the Court is trying to send a message: just as in real estate terms it is important to buy a property in a good spot (‘location, location, location’) so too is causation something that plaintiffs need to consider closely before litigating.

The facts in Mikhael are interesting. A schoolboy sued the State of NSW regarding an incident in which another student (named “T” in the judgment) violently assaulted him after class. This had occurred following an argument between the plaintiff and T about 10 minutes before the end of the period. The teacher determined that by the end of the lesson things seemed to have cooled down between the two boys so did nothing further about it.

Only a few weeks prior, T had punched another student after football practice. So the duty of care in this case required the school to implement reasonable measures to avoid a foreseeable risk of T doing a similar thing again; it was unsuccessful in so doing.

The school did have in place a system for dealing with this kind of problem. Pursuant to this system, teachers at the school were in fact alerted to T’s involvement in the football-training assault. However, it was significant in the Court’s eyes that whilst the system had been followed and teachers advised, they were not told that the football practice assault occurred after only slight provocation of T. This is information with which the Court thought the teachers should have been armed but were not: breach of the school’s duty of care.

The question remained- what would the teacher of the class have done if armed with the additional piece of information that T was easily provoked? The plaintiff (as the respondent on appeal) filed a notice of contention setting out the kinds of measures the teacher could have taken: checking if T was hanging around after class waiting for a fight, escorting the plaintiff student to a position of safety and asking the plaintiff if he had any fear about T.

However, the Court found that even if these steps were taken, the assault would not in all likelihood have been averted (at [110]). For example, there was no description of where or what a position of safety might be and whether this required the plaintiff to be kept separate from other students for a certain period and if so for how long. The problem was compounded by the fact that the teacher in question had not been cross-examined at the trial about whether in her opinion, these kinds of measures would have stopped T from assaulting the plaintiff. She was adequately experienced and knowledgeable, and indeed in the best position, to give such a comment.

In re-stating the relevant legal principles, the Court confirmed that the ‘but for’ test was the applicable test when considering causation under section 5D of the Civil Liability Act; even in cases of negligent omission such as the instant. Guidance was found in the High Court’s decision in Adeels Palace v Moubarak (2009) 239 CLR 420 in which an angry restaurant patron had left premises and returned with a gun that once inside, he discharged to injure the plaintiff. The plaintiff contended that the restaurant should have had more security guards on duty but the HCA found that this would not have averted the shooting in any event. They would not or could not have stopped him.

But the most important ‘take-away’ from Mikhael is that the Court (at [96] and [107]) said that the ‘but for’ test is not concerned with possibilities about what might have happened had the necessary preventative action/measures been put in place. In this case, Beazley JA (with whom Allsop P and Preston CJ of the LEC both agreed) said that the plaintiff’s causation case amounted to ‘no more than a series of possibilities’. Instead, he needed to prove that any such steps would, on the balance of probabilities, have averted the harm suffered.

An “Achilles heel” is a fatal weakness despite overall strength. A perception of overall strength will often be somewhat justified when the case on duty and breach is strong. Nevertheless, a weak case on causation can have fatal consequences.

1 Comment

Filed under causation, Negligence

High Court Decision in Strong v Woolworths – ‘probability theory’ vs common experience


Plaintiff personal injury lawyers are all agog over the recent High Court decision of Strong v Woolworths [2012] HCA 5. But has it really moved things much in their favour or was it simply a case of the best possible response to a difficult set of facts?

The facts themselves are straightforward: in an unfortunate accident, an amputee woman on crutches slipped on a hot chip dropped on the sidewalk of a shopping centre, which was property under the care and control of Woolworths. On the evidence, there was no way of determining when the chip had been dropped but it was agreed the accident took place at 12.30pm.

Woolworths conceded that it had no proper system of inspection in place to prevent this kind of accident, but argued that even if it did, the errant chip would not have been detected in time to prevent the accident.

This argument raised the question of causation under section 5D of the Civil Liability Act: was Woolworths’ negligence in failing to have a proper system of inspection and cleaning in place the cause of the harm suffered?

A key step in the reasoning process adopted by the majority of the High Court was to determine that a reasonable system of ‘inspection and removal’ would require intervals between checks of not more than 20 minutes.

Therefore, for Woolworths to succeed the Court would need to find that the chip was dropped after the time of the last hypothetical inspection at 12.10pm. If so, the argument holds true that the chip would not have been detected by the time of the accident at 12.30pm, regardless of whether or not there was a proper system of inspection in place.

Conversely, if the Court determined that the chip was dropped before 12.10pm the Woolworths argument would be defeated and the plaintiff would be entitled to succeed.

Given the supermarket had been open since the start of normal trading hours, the plaintiff’s counsel argued that by sheer dint of comparing the time frames involved, the chip was more likely to have been dropped in the period between 8am and 12.10pm rather than the much shorter period between 12.10 and 12.30pm.

This argument had been defeated in the NSW Court of Appeal decision by one based more upon common sense and experience than numbers. In the court below Woolworths’ counsel, Mr Maconachie QC argued that as hot chips are more likely to be eaten at lunch, and there are more people around at that time generally (among whom, there are presumably those eating hot chips), the chip was more likely to have been dropped in the period closest to lunchtime (ie- between 12.10 and 1230pm). That’s the beauty of the common law, it’s perhaps one of the last jurisdictions left where common sense and experience have a chance to have their say.

Each of these two reasoning processes (common experience vs numbers) have their difficulties but the NSW Appeal Court judges chose common experience. However, the majority of High Court overturned the Court of Appeal and went for the odds (at [38]):


“The probabilities favoured the conclusion that the chip was deposited in the longer period between 8am and 12.10pm and not the shorter period between 12.10pm and the time of the fall.”


Perhaps the High Court felt that to permit Woolworths to escape liability in the face of no proper inspection system would prevent justice being done.

In his dissenting judgment Justice Heydon took a more hardline approach – as the plaintiff hadn’t positively discharged the burden of proving her case she should lose. This seems in keeping with the requirements of section 5E of the Civil Liability which states that the plaintiff bears the onus of proof.

In other words, it was for the plaintiff to prove via the evidence that the chip was dropped before the moment at which the last hypothetical inspection prior to the accident would have taken place, and that her recourse to “probability theory” in attempting to do this was “unconvincing”. His Honour thought the plaintiff needed more than just favourable odds to make out her case.

It is rare that there will be no evidence at all before the court on a matter so crucial. Both the NSW Court of Appeal and High Court were faced with a difficult choice between what some might consider the lesser of two evils; both common sense and probabilities are an inexact means of deciding such a serious and unfortunate situation.

If there was anything more substantial in evidence the case would not have gone nearly this far, which begs the question whether the decision will be treated as one limited to its facts or whether it will see the unlikely proliferation of reasoning based on “probability theory” in our court system. Only time will tell.


Leave a comment

Filed under causation, Negligence