CLA: dangerous recreational activity & statistical evidence of the risk

Campbell v Hay [2014] NSWCA 129.


Meagher JA at [1];
Barrett JA at [7];
Ward JA at [10]

110 For the purposes of division 5 of the Civil Liability Act, a “dangerous recreational activity” is defined in s 5K as meaning a recreational activity that involves “a significant risk of physical harm”.

Relevant activity

111 It is not disputed that Mr Campbell, when taking flying lessons from Mr Hay, was engaged in a recreational activity. However, Mr Campbell contends that his Honour erred in considering the relevant activity in general terms, as simply flying a, or in a, light aircraft (as opposed to flying in a light aircraft with a highly experienced instructor). The distinction is relevant because Mr Campbell contends that there was no basis for his Honour to conclude that flying a light aircraft under the supervision of an instructor with Mr Hay’s experience was an activity that carried with it a significant risk of physical harm.

112 There is no basis for the criticism that his Honour wrongly characterised the relevant activity. At [59], his Honour expressly took into account the fact that there was an experienced pilot in the aircraft, when describing the activity as “flying with an experienced pilot in a single engine light aircraft”. Therefore, implicit in his Honour’s conclusion that Mr Campbell was engaged in a dangerous recreational activity must be that the presence of a highly experienced flying instructor in the aircraft did not objectively lessen the risk of physical harm involved in that activity to one that was trivial or not significant for the purposes of the s 5K definition.

113 The complaint by Mr Campbell in this regard is that his Honour did not properly take into account, as part of considering the relevant circumstances of the activity in question, the particular level of experience of Mr Hay as a flying instructor. In Fallas v Mourlas [2006] NSWCA 32; (2006) 65 NSWLR 418 at [37]), Ipp JA emphasised the need to take into account the particular circumstances of the case when determining whether the activity in question was a dangerous recreational activity (having noted, at [36], that “[f]actors such as time, place, competence, age, sobriety, equipment and even the weather may make dangerous a recreational activity which would not otherwise involve a risk of harm” and recognised that the converse may also be the case). In the present case, Mr Campbell’s contention is, in effect, that whatever the general risk of flying in a light aircraft, in the present case, there was not a significant risk of physical harm because of Mr Hay’s experience (and, presumably, the level of competence that was indicated thereby).

114 Mr Campbell contends that the contrary conclusion reached by his Honour leads to the surprising result that anyone who flies in a single engine aeroplane is engaging in a dangerous recreational activity. Mr Hay submits that even if that were to be the effect of his Honour’s conclusion, such a result would not have the surprising consequences to which Mr Campbell adverts, having regard to the small proportion of recreational light aircraft flights compared with the number of commercial and general aviation flights disclosed in the statistical material before his Honour.

115 Mr Hay further contends that the conclusion reached by his Honour is not surprising insofar as the nature of the activity in which Mr Campbell was engaged is such that even a highly experienced pilot exercising reasonable care may be confronted with dangerous circumstances (such as engine problems) in the demanding and difficult environment of flying a single-engine light aircraft.

Relevant test

116 Consideration was given to the meaning of “significant risk of physical harm” in Falvo v Australian Oztag Sports Association [2006] NSWCA 17 and Fallas, from which a number of propositions can be drawn. The definition of “dangerous recreational activity” must be read as a whole (Ipp JA, with whom Hunt AJA and Adams J agreed, at [28] in Falvo). Regard must be had both as to the nature and degree of harm that might be suffered, on the one hand, and the likelihood of the risk materialising on the other (Falvo per Ipp JA at [28]). The expression constitutes one concept with the risk and harm mutually informing each other (such that the risk of physical harm may be significant if the risk is low but the potential harm is catastrophic or if the likelihood of both the occurrence and the harm is more than trivial; but the risk of physical harm may not be significant if, despite the potentially catastrophic nature of the harm, the risk is very slight) (Falvo per Ipp JA at [31]). “Significant risk” has been said to mean more than trivial and does not import an “undemanding” test of foreseeability (Fallas per Ipp JA at [14]); it does not mean a risk that is likely to occur (Fallas per Ipp JA at [16]) but lays down a standard lying somewhere between a trivial risk and a risk likely to materialise (Fallas per Ipp JA at [18]); and as a general guide, it means a risk that is “not merely trivial, but generally speaking, one which has a real chance of materialising” (Tobias JA at [90]-[91], this being the test his Honour applied later in Jaber v Rockdale City Council [2008] NSWCA 98 at [54]). An objective test is required in determining whether a recreational activity was dangerous (Fallas per Ipp JA at [13]; Basten JA at [136]).

117 In Fallas, Basten JA (at [131]), read the word “significant” as not requiring a particular level of physical harm but as requiring an assessment of the “risk of physical harm”, which would import a consideration of the seriousness of the harm that might occur. His Honour considered that if the harm is potentially catastrophic, then a very low level of risk may be regarded as significant, whereas if the potential harm is not serious at all, then the risk may not be considered significant until it reaches a much higher level (see, similarly, the observation of Ipp JA at [30] in Falvo to the effect that it is difficult to see how a recreational activity could fairly be regarded as dangerous where there is no more than a significant risk of an insignificant injury.

118 Relevantly, in the context of the statistical evidence relied upon in the present case, Basten JA in Fallas (at [134]-[136]) sounded the need for caution in adopting the approach in Rogers v Whitaker [1992] HCA 58; (1992) 175 CLR 479 at 489. His Honour recognised that, while it would be tempting to say that even a very low chance of potentially devastating harm could be significant, when considering the defence under s 5L of the Civil Liability Act, the relevance of the existence of a significant risk is that such a finding could deprive an injured party of a right of action. His Honour noted that, unlike as was the case in Rogers, the test of significant risk was not dependent upon the expectations of a person in a particular relationship with another.

Statistical evidence

119 In Fallas, Ipp JA noted that the degree of likelihood of a risk occurring might be established in many ways ([20]), including by the application of logic, common sense or experience to the particular circumstances of the case ([22]). In the present case, Mr Hay relied upon statistical and anecdotal reports as to the incidence of accidents involving recreational light aircraft of the kind I have identified above. Mr Campbell disputes the usefulness of that evidence (on which his Honour appears to have placed little weight in any event, largely determining the question by reference to common sense).

120 In particular, Mr Campbell argues that the statistics relied upon by Mr Hay did not differentiate between accidents occurring during flight training and those occurring generally, and did not take into account the level of experience of the pilots involved in the accidents. It is further contended that the number of fatal accidents in a relevant period of itself says nothing as to the risk of physical harm, without a means of comparing that number with the number of flights over the relevant period.

121 Leaving aside the media reports of light aircraft accidents, which at best might inform a common sense view of the risks attendant upon flying in light aircraft, the two sets of statistical information were the ATSB’s Aviation Occurrence Statistics for the period 1999 to 2009 in its Aviation Research and Analysis Report and the statistics contained in Mr Brandon’s article relating to “Recent [RAA] fatal accident history”, obtained by Mr Hay’s solicitor by use of a Google search.

122 The ATSB report, as its heading suggests, provides aviation occurrence data for the period 1 January 1999 to 31 December 2009, being data as to accidents or incidents involving two categories of aircraft operation (commercial air transport and general aviation). The report makes clear that the general aviation category, defined as all flying activities outside of scheduled RPT (high capacity regular public transport) and non-scheduled (charter) passenger and freight operations, does not include Australian non-VH registered aircraft. Although that category was said to include “flying training”, it seems to have been accepted that statistics for the relevant period in relation to Jabiru light aircraft of the kind involved in the present incident would not have been included in the data in this report. Mr Evans confirmed that there is no requirement to report non-fatal accidents in recreational aviation registered ultralight aircraft to the ATSB. The report did, however, include a table based on occurrences involving non VH-registered Australian aircraft by state over the period 1999-2009, which appears to include all aircraft types including ultralights and gliders, that disclosed 42 occurrences in New South Wales in 2007 (that number varying from as low as 30 in 2006 to as high as 58 in 2000.

123 The ATSB report uses the number of aircraft departures (as opposed to the hours flown) as the denominator figures for the calculation of accident and fatal accident rates for all air transport operation types, on the basis that these are considered to be a more appropriate measure of the exposure or opportunity for an event to occur. Hours flown data, by contrast, is used in the report to calculate accident and fatal accident rates for different operation types (the report noting that for some operation types within general aviation there is a higher risk of accident outside of the approach/landing and takeoff phases of flight such as due to low flying requirements.

124 Reliance is placed on the ATSB report by Mr Hay as indicating (by comparison with the RAA statistics referred to in Mr Brandon’s article) that the volume of flying activity for general and commercial aviation over the relevant period was much higher than that for light aircraft. The schedule (at Blue 2/407), discloses that for 2007, a total of 1,722 thousand hours were flown in the commercial air transport category and 1,832 thousand hours flown in the general aviation category. (By comparison, the article based on RAA statistics, to which I will refer shortly, indicated that average flying hours for RAA registered aircraft were in the order of about 50-55 hours per year.)

125 The ATSB report also included statistics for aircraft involved in flying training occurrences, though again those statistics do not appear to have included non-VH registered aircraft. In 2007, for example, there were 41.2 accidents per million flying training hours but no fatalities. The report included a description of some of the flying training fatal accidents over the period from 2002 to 2008 but, other than describing the flight instructor in one such incident as an experienced flight instructor, it did not give details of the level of experience of the instructors.

126 Turning to the RAA statistical information in Mr Brandon’s article, this is relied upon by Mr Hay as demonstrating that in 2007, on an analysis of fatal accidents involving RAA registered aircraft, there was a fatality involved in 1 out of every 500 light aircraft flights. It was noted that the average annual hours flown by RAA aircraft for the past 15 years, including flying school machines, ranged from 44 to 60 but that in most years was between 50 and 55).

127 Mr Hay points to the comparison made in the Brandon article as to fatal accidents per 100,000 hours as between all RAA aircraft and the general aviation, training, private and business categories (the average for the former in 2007 being 5.8, the average for the latter in that year being 1.9), from which it is submitted a comparison can be made as to the safety risk or danger involved in flying light aircraft as opposed to general aviation. Mr Hay submits that for the years prior to 2007, the performance of light aircraft can be seen to have been significantly more dangerous than that of general aviation by reference to the comparative statistics there set out.

128 Mr Hay submits that these figures are indicative of the level of activity of ultralight and other recreational light aircraft flying (which, by virtue of the number of members of the RAA and the number of aircraft, is very small compared to the rest of the aviation industry). Mr Hay submits that, on its own, the statistical information that indicates on average 1 fatal accident per 500 light aircraft flights, having regard to the average hours per aircraft of around 50 hours per year, demonstrates that there is a significant measure of risk proportional to usage and the length of the activity (as compared with the ATSB information per 100,000 hours flown in other categories of aviation).

129 Against this, Mr Campbell places weight on the fact that Mr Hay had flown for about 56 years and had, over that period, been in a position where he had made 9 forced landings in none of which had any injury been suffered. Mr Campbell submits that the statistical evidence referred to above did not enable a conclusion to be drawn by his Honour that this was a dangerous recreational activity because the statistics did not enable a conclusion to be drawn as to the experience of the pilots involved in the aircraft incidents the subject of those statistics. It is submitted that general aviation statistics and “undifferentiated” anecdotal material concerning plane crashes do not provide a reliable indication from which to assess whether or not the particular activity in which Mr Campbell was engaged was a dangerous recreational one.

130 His Honour recognised that there were limitations on the use that might be made of the statistical information, to which I have earlier referred, and accepted (at [68]) that they were not “entirely apposite”. The reliance placed by his Honour on that material seems to have been that they indicated that from time to time accidents occurred involving light aircraft, as substantiated by the incidents described in the media reports to which his Honour had been referred ([68]) and perhaps also that they indicated matters that might cause or contribute to something “going wrong” in the operation of an aircraft in flight and in safely landing it.

131 His Honour’s conclusion as to this aspect of the defence was in large part based on the common sense approach to which Ipp JA had referred in Fallas and which his Honour considered had been demonstrated by the examples given by Ipp JA in that case at [26] and [27] and Ipp JA’s approach in Falvo; as well as the approach of Harrison J in Vremen and Morris v Albury City Council (2011) NSWSC 39.

132 This approach was criticised by Mr Campbell as being one that was said to conceal both the source of knowledge of the fact and the nature of the fact said to be determined by its application. It was accepted that one could list hypothetical causes of something going wrong in flight but that there was no basis on which to use common sense to form a view as to the relative frequency with which those things occurred.

133 Mr Campbell also placed weight on the statement in Mr Brandon’s report that the RAA membership had “perhaps been achieving near-reasonable safety results – taking into account the continuing introduction of faster, heavier, more complex and less docile aircraft, together with a marked reduction in the average years of experience of the RA-Aus pilot base…” as pointing to the significance placed by the author of this report on pilot experience/inexperience as a contributing factor towards the incidence of accidents. That said, the report also went on to refer to “very puzzling instances, where those who might be regarded as very experienced and knowledgeable, expose themselves to extreme risk”.

134 In at least one respect, Mr Brandon’s analysis suggests that some attempt had been made to analyse the factors contributing to fatal general aviation accidents in that reference was made to a preliminary study attributing percentages to the numbers of accidents in which matters such as aircraft handling errors, flight planning or fuel starvation had been a factor and Mr Brandon’s comment was that there was no reason to believe recreational aviation experience differed markedly, noting that the likelihood of engine failure was higher in sports and recreational aviation than in general aviation (though commenting that, in itself, engine failure should not cause a serious accident).

135 Leaving aside Mr Brandon’s observations, the criticism based on use by his Honour of a common sense approach to this question (somewhat curious when resort is made in the appellant’s own submissions to common sense without any apparent empirical basis, namely the assertion that flying light aircraft is safer than travelling by motorcycle) fails to acknowledge the force of his Honour’s observation that if something were to go wrong with the operation of the aircraft in flight (particularly, as in this case, with the engine) then a forced landing would be inevitable (unless the problem were able to be resolved in flight).

136 Mr Campbell contends that his Honour was not there stating the correct test. However, his Honour had earlier set out the statements as to what was meant by significant risk of physical harm for the purposes of the definition; his Honour had referred to the statistical and anecdotal evidence as indicative that “from time to time” incidents did occur involving light aircraft (by which I understand his Honour to be accepting that there was a real or not trivial chance that in any given flight a problem of the kind to which he had adverted might materialise) and, coupled with the risk of serious injury or death if such an incident did occur, concluded that the test posed in Falvo and Fallas had been met.

137 Had it been necessary for the purposes of this appeal to determine the issue, I would have concluded that his Honour did not err in concluding that flying a light (or ultralight) aircraft for recreational purposes constituted a dangerous recreational activity in which Mr Campbell was engaged notwithstanding that a flying instructor with Mr Hay’s experience was present.

138 Common sense indicates that, if there is complete engine failure in a single-engine light aircraft, then there is a risk that a forced landing will result and a risk that such a landing will result in serious injury or death. That risk exists whether or not the pilot is an experienced pilot, since it cannot be assumed that all engine failures can be resolved without forced landings in which there is a risk of serious injury or death. Furthermore, although one would certainly hope that the more experienced the pilot the less likely the risk that there will be pilot error contributing to the ultimate fate of the aircraft and its passenger(s), everyday human experience would lead one to conclude that even experienced people can make mistakes particularly when under the stress of an emergency or unexpected event.

139 This is not a case where the potential risk of physical harm arising from an engine failure in flight could be described as trivial in the manner of that considered in Falvo. In assessing the likelihood that such a risk would materialise (i.e., where the risk falls on the spectrum between a trivial risk and one likely to materialise, to use the terminology of Ipp JA in Fallas), then the statistical evidence that 1 in 500 light aircraft flights in 2007 ended in a serious accident means that this cannot be seen as a trivial risk. While account is to be taken of the experience and qualifications of the pilot or instructor as part of the overall circumstances of the particular activity in question, one must also take into account that the risk (as here, of engine failure) could occur in circumstances where even an experienced pilot would not be in a position to avoid an outcome involving serious injury or death.

140 His Honour likened the activity in which Mr Campbell was engaged to that of a parachute diver. Reference was made in submissions to Echin v Southern Tablelands Gliding Club [2013] NSWSC 516 where Davies J was prepared to conclude that the evidence demonstrated that, as a recreational activity considered generally, gliding involved a significant risk of physical harm because, although the risk of an accident was low, the potential harm was catastrophic (though his Honour went on to conclude that, even if gliding generally could not be considered a dangerous recreational activity, the act of performing a landing over powerlines as had there been attempted was).

141 In the present case, if there were to be a problem with the single engine (which could not be discounted as a trivial risk), it might be one that the flying instructor would not be able to rectify such that, even without any negligence on the part of the flying instructor, a forced landing might have to be made, carrying with it the obvious risk of death or serious injury. The presence of an experienced flying instructor was a factor to be taken into account in assessing whether there was a significant risk of physical harm. However, in my view his Honour did not err insofar as he implicitly concluded that that fact did not reduce the assessment of risk of physical harm arising from a forced landing due to engine failure to a risk that was not significant for the purposes of the definition in s 5K.

142 Ground 1 of the notice of appeal is not necessary to determine having regard to the conclusion I have reached on the grounds raised in the notice of contention. Had it been necessary to do so I would have concluded that this ground was not made good.


Reproduced with permission from Sheldon SC


David Cormack – Brisbane Barrister.

Related Posts

Recent Comments