Rundle v State Rail Authority of New South Wales

[2002] NSWCA 354

Reported Decision: (2002) Aust Torts Reports 81-678

New South Wales

Court of Appeal

CITATION:	Rundle v State Rail Authority of New South Wales [2002] NSWCA 354
FILE NUMBER(S):	CA 40855/01
HEARING DATE(S):	12 August 2002
JUDGMENT DATE:	23 October 2002
PARTIES :	Kane Rundle by his next friend Gail Rundle (Appellant) State Rail Authority of New South Wales (Respondent)
JUDGMENT OF:	Heydon JA at 1; Young CJ in Eq at 64; Foster AJA at 65

LOWER COURT JURISDICTION :	Supreme Court
LOWER COURT FILE NUMBER(S) :	SC 20386/97
LOWER COURT JUDICIAL OFFICER :	McClellan J

COUNSEL:	Mr D E Grieve QC/Ms D M Coulton (Appellant) Mr C R R Hoeben SC/Mr P Biggins (Respondent)
SOLICITORS:	George Sten & Co (Appellant) Dibbs Barker Gosling (Respondent)
CATCHWORDS:	Tort - negligence - duty of care - plaintiff injured while spraying graffiti on roof of moving carriage - plaintiff travelling with upper body squeezed through upper window of carriage - contravention of delegated legislation - application of principles in Modbury Triangle Shopping Centre Pty Ltd v Anzil (2000) 205 CLR 254 - D
CASES CITED:	Bone v Commissioner of Stamp Duties [1972] 2 NSWLR 615 Businessworld Computers Pty Ltd v Australian Telecommunications Commission (1988) 82 ALR 499 Crimmins v Stevedoring Industry Finance Committee (1999) 200 CLR 1 Donoghue v Stevenson [1932] AC 562 Henwood v Municipal Tramways Trust (South Australia) (1938) 60 CLR 438 Modbury Triangle Shopping Centre Pty Ltd v Anzil (2000) 205 CLR 254 Reeves v Commissioner of Police of the Metropolis [2000] 1 AC 360 Ryan v State Rail Authority of New South Wales [1999] NSWSC 1236 Sullivan v Moody (2001) 183 ALR 404 Tame v New South Wales [2002] HCA 35 Wyong Shire Council v Shirt (1980) 146 CLR 40
DECISION:	See paragraph 63

CA 40855/01
SC 20386/97

HEYDON JA
YOUNG CJ in Eq
FOSTER AJA

23 October 2002

KANE RUNDLE by his next friend GAIL RUNDLE v

STATE RAIL AUTHORITY OF NEW SOUTH WALES

Tort – negligence – duty of care – plaintiff injured while spraying graffiti on roof of moving carriage – plaintiff travelling with upper body squeezed through upper window of carriage – contravention of delegated legislation – application of principles in Modbury Triangle Shopping Centre Pty Ltd v Anzil (2000) 205 CLR 254

On 6 May 1994 the plaintiff, then aged 15, was badly injured while travelling on one of the defendant’s double-decker silver carriages. The injuries caused the plaintiff to have no memory of the accident. At the time of the accident the plaintiff had been engaged in spraying graffiti on the roof of the carriage while travelling with the top half of his body squeezed through an upper window of the carriage. The plaintiff had engaged in such activity daily for several months.

The plaintiff sued in negligence for damages for personal injury. He failed at trial and appealed as to the scope of the duty. Both parties agreed that the defendant owed the plaintiff a duty of care to take reasonable steps to provide a safe railway carriage. The defendant said that the duty was to exercise care in relation to dangers likely to arise from the ordinary use of the carriage which might reasonably be expected, while the plaintiff said that it was to exercise care in relation to dangers which might reasonably be expected, having regard to the defendant’s actual or imputed knowledge of the behaviour of all passengers.

(Heydon JA, Young CJ in Eq and Foster AJA agreeing), dismissing the appeal,

1. As to duty of care:

a. The defendant owed the plaintiff a duty to provide a safe railway carriage by exercising care in relation to dangers likely to arise from the ordinary use of the carriage which might reasonably be expected: [51]-[52].

Donoghue v Stevenson [1932] AC 562; Tame v New South Wales [2002] HCA 35, considered; Modbury Triangle Shopping Centre Pty Ltd v Anzil (2000) 205 CLR 254, distinguished; Henwood v Municipal Tramways Trust (South Australia) (1938) 60 CLR 438, applied.

b. When the plaintiff projected his body outside the window, he did not do so instinctively, impulsively, forgetfully or as a result of illness or other physical condition. Any duty on the defendant to prevent that conduct was fulfilled by provision of a window only capable of being opened 235 mm. Further, what the plaintiff did was not part of the “ordinary use” of the carriage, and the danger to him was not “likely to arise out of the ordinary use of the carriage”. The defendant had taken precautions against dangers arising from the ordinary use of the carriage, for example by passengers desiring to put their heads out of the window for a better view or air or to relieve illness, or by passengers desiring to put their arms out. The defendant’s duty did not extend so far as to oblige it to take precautions against passengers seeking to sit on the window edge with their entire torso and head outside with a view to damaging the defendant’s property: [53].

Bone v Commissioner of Stamp Duties [1972] 2 NSWLR 615; Businessworld Computers Pty Ltd v Australian Telecommunications Commission (1988) 82 ALR 499, considered; dicta of Dixon and McTiernan JJ in Henwood v Municipal Tramways Trust (South Australia) (1938) 60 CLR 438, applied.

c. The conduct of the plaintiff went very close to being deliberately self-inflicted harm. It was certainly recklessly self-inflicted. None of the limited areas where a duty of care exists to prevent persons form inflicting harm on themselves applies. In particular, though the plaintiff was aged 15, he cannot relevantly be said not to have been “of full age” or “too young to appreciate” the risks of what he was doing: [58].

Reeves v Commissioner of Police of the Metropolis [2000] 1 AC 360; Sullivan v Moody (2001) 183 ALR 404, considered; Ryan v State Rail Authority of New South Wales [1999] NSWSC 1236, distinguished.

2. As to breach of duty of care:

a. Though if the risk came home the consequential injuries could be very serious, the risk was reasonably foreseeable but not probable: the degree of reasonable foreseeability was at the low end of the scale. The incidence of injury had been very low considering the number of times the plaintiff and others had engaged in the practice in question. To overcome the problem with the carriage windows would either call for very large expenditures or would create major disruptions to the entire Sydney train system to deal with a possibility which was reasonably foreseeable, but only just. The trial judge correctly found there to have been no breach of duty: [62].

Wyong Shire Council v Shirt (1980) 146 CLR 40, considered.

O R D E R S

1. The appeal is dismissed.

2. The appellant is to pay the respondent’s costs of the appeal.

CA 40855/01
SC 20386/97

HEYDON JA
YOUNG CJ in Eq
FOSTER AJA

23 October 2002

KANE RUNDLE by his next friend GAIL RUNDLE v

STATE RAIL AUTHORITY OF NEW SOUTH WALES

Judgment

1 HEYDON JA: This is an appeal by an unsuccessful plaintiff against orders made by McClellan J on 5 October 2001 that there be a verdict and judgment for the defendant. The trial took place on 3-6 September 2001.

The trial judge’s findings

2 On 6 May 1994 the plaintiff, then aged 15, was badly injured while travelling on one of the defendant’s double-decker silver carriages. The injuries have caused the plaintiff to have no memory of how they eventuated and there was no witness to what happened. But the trial judge made the following findings:

“By the time of the accident, the plaintiff was accustomed to disfiguring railway property with graffiti. He gave evidence that he started writing graffiti on trains in about Year 8. He appears to have regularly engaged in the activity and failed to attend much, if at all, to his school studies from this time onwards. Initially, he used ‘textas’ to write on the inside of trains, but over time worked his way outside where he used spray cans. He mostly wrote words, usually one of his ‘tags’. He commonly wrote the word ‘Oops’ and sometimes ‘fork’.
In 1994 the Sydney metropolitan railway network had a total of 775 double-decker silver carriages. The plaintiff was travelling in one of these carriages when he was injured. By the time of the accident the plaintiff had taken to squeezing the top half of his body out of an upper window of the train to spray graffiti on the side of the train, and above the window, with a ‘pressure pack’ can. The window was located just below the gutter at the low point of the roof. As a consequence graffiti painted above the window was on the roof of the train that curved at this point.
The upper window consisted of four panels of glass, two on either side of a fixed mullion. Each of the panels could slide to allow ventilation but the window was designed so that the distance any panel could travel was confined. As designed the window would only have allowed a horizontal opening of 115 millimetres, the vertical opening was a constant 425 millimetres. The design of the window included a rubber gasket that, if it was in place, had the consequence that the window could not be opened beyond its designed width. It would have been impossible for any person to manoeuvre through such a narrow opening.”

3 The trial judge did not accept the plaintiff’s evidence that it was “quite easy” to get out of the window. He said: “I have no doubt the plaintiff became adept at the manoeuvre but it must be remembered that he practised it on many occasions.”

4 The trial judge continued:

“Simon Walker is a friend of the plaintiff. He commonly joined the plaintiff in spraying paint on to trains. His tag was the word ‘abuse’. He described the process in a similar manner to the plaintiff:
‘You would first find a window that would open to be wide enough, and then sort of put your head out and twist your hips out and then sit on the outside of the train and do it on the top.’
Because the side of the train tapered, and the roof curved away, he said by staying close to the outside of the window he could use his spray can without hitting anything. He said he was careful not to lean out too far.”

5 Turning to the events of 6 May 1994, the trial judge said:

“it is apparent that the plaintiff, having boarded a train, manoeuvred himself out of an upper window of the double decker carriage that did not have the restraining rubber gasket and commenced to write the word ‘oops’ using a spray can. After the accident, the train was found to have the two oo's’ and part of the ‘p’ completed on the roof. As he was in the course of writing the word, and, while the train was moving, the plaintiff's head contacted an object, probably a stanchion or signal, beside the track, causing him severe injuries. The external object which struck the plaintiff's head was never identified, but there was blood found inside the carriage at the position of the window.
The plaintiff gave evidence that it had been his usual practice to board a train at Revesby, and travel to East Hills. Once on the train he would look for an empty carriage and test the windows to see if he could find one in which the gasket had been removed so that he could use it to access the outside of the train. It is likely that this was the sequence of events on 6 May 1994.
Both the plaintiff and Simon Walker said that the plaintiff had been putting his tag on the roof of trains for maybe six months before the accident. The plaintiff and his friends were not the first to spray graffiti on to the roofs. They merely copied others who had done it before them.”

6 The trial judge then moved to the question of whether the window through which the plaintiff had manoeuvred himself could have been made secure:

“Mr Colin Simpson is an experienced engineer with particular expertise in mechanical engineering problems. His evidence was not a matter of significant dispute. In his opinion the train window could have been modified so that it would not open far enough to allow a fifteen year old to squeeze through it. Either by the use of screws passing through the window frame and rubber gasket or by applying adhesive to the rubber to glue it to the window frame, the opening could have been confined to its original design. Although adhesive alone may have failed, if both adhesive and screws were used the window would have been most unlikely to have failed.

Although Mr Simpson accepted that a teenager could not accidentally pass through the fully opened window, he was of the opinion that some people could easily fit through it when fully opened. However, he accepted that only a person of limited size could do this and agreed that a person ‘would have to manipulate their body in a way particularly designed to get them through.’

Mr Cowling, who is also an engineer, gave evidence in the defendant's case. Although not employed by the defendant, it is clear he has spent a great deal of time examining issues related to the safety and performance of the railways and has given advice to and evidence on behalf of insurers of the defendant. His knowledge of the design and performance of trains on the city network is considerable.
He accepted that the defendant could have attempted to glue the gasket to the window frame and could have applied screws to each window. Indeed, after the accident, as an interim measure, the gaskets were glued. Mr Cowling did not believe that this alone would be an effective long term solution. Apparently the gasket will naturally deteriorate and with time the window could be forced out to its maximum opening unless otherwise mechanically restrained.
Mr Cowling examined many records and carried out extensive research into the details of railway accidents. He indicated that he had not found any record which suggested that people were squeezing through the windows prior to the accident. There had been incidents where people spraying graffiti when hanging out of carriage doors had been injured and killed but none which involved the use of windows.

Carriages of the silver double-decker type were first constructed and brought into service in the 1970s. By the late 1980's it was recognised that there was a need to upgrade them to provide locking doors and other modifications. Together with a program to fund the provision of a new type of train known as the Tangara, funds were provided to upgrade the silver double-deckers. It became known as the ‘CityDecker’ upgrading program. Although it was intended that the program would commence in 1990 there were some problems. By May 1994, although the 775 carriages had been fitted with automatic doors, the rest of the program had not been implemented. This locking door program alone cost $4.5 million.

In May 1994 the CityDecker project was restarted at a cost of about $148,000 per car for the first 200 cars, making a budget for this batch of about $28.7 million.
The project involved upgrading with improved lighting, vandal resistant seating, repainted ceilings and walls, glass screen partitions between vestibules and saloons, slip resistant floors, tinted inward-opening hopper windows, high-visibility paint treatment to external ends of carriages, passenger ‘help-points’ in vestibules and improved communications.

The implementation of the intended program was made difficult by problems with the Tangara. At about the time that the CityDecker program was to commence, an unprecedented increase in damage by vandals to the glazing of the Tangara carriages occurred, creating a major crisis in the availability of those carriages.

The design of the Tangara is such that although passenger safety was not at risk, the large external areas of glass offered vandals an easy target for missiles such as rocks and slingshots.

I understand that the glass replacement cost as a result of breakage was in the region of $400,000 per annum, which raised the initial need for consideration of a separate upgrading of the Tangara glazing system.
Apart from the direct glass replacement costs, this also meant that more of the non-Tangara rollingstock were required to make up for the shortfall in Tangara availability, in order to maintain the daily timetable requirements. The result was that the CityDecker upgrading could not proceed at the intended rate.
Up to June 1996 a total of 136 suburban cars and 38 intercity cars had been upgraded for a total cost of over $25 million.
Since then approximately a further 150 suburban cars have been upgraded each year under the CityDecker project for a total approximate cost of about $111 million.

If the defendant had believed it necessary, it may have been possible to modify the windows of the silver double-decker carriages before the accident as a separate task rather than replace the windows as part of the total upgrade of each carriage. However, because of the need to maintain fleet requirements for effective daytime operation, the work could only have been done in the evening when the carriage was undergoing its nightly cleaning.

Mr Cowling indicated that two people working together could have glued the rubber gasket at a rate of two carriages per night. The task could have been achieved as part of the normal maintenance operation but would have taken many months to complete.

If, however, in addition to the adhesive, two screws were inserted, this would have involved drilling and tapping, and although this may have been an effective solution to the problem, it would have taken four persons one night to complete the modifications to a carriage if the task was done as part of the normal maintenance procedures. At this rate the total program would have occupied some years.
As part of the general upgrade, the defendant embarked upon a major program to replace the windows on the double deck carriages. The new window was differently designed being an inward opening hopper with the consequence that there was no possibility of a person squeezing through it. The cost involved in modifying the window in this manner was obviously considerable, although because it was part of the general upgrade, it was not separately identified.
Mr Cowling observed many of the replaced windows. His evidence was that by the time they were being replaced many of the windows had reached the end of their useful life. In particular by 1996, windows were found without rubber gaskets or where the gasket had worn to the point where it no longer secured the window. If this was the case by 1996 it is obvious that many gaskets would have worn at an earlier point making it likely that many were in this state in 1994. As the windows were first constructed in the 1970s, this may explain why the plaintiff and his friends readily found windows which were no longer restrained by the rubber gasket.”

7 The trial judge next dealt with the problems which have been caused to the defendant by writers of graffiti.

“From at least 1989 graffiti was a significant issue for the railways. Apart from the damage caused by the severe disfiguring of the inside and outside of carriages, by 1989 deaths had occurred when young people were hanging out of the doors of trains to spray graffiti. For this and other reasons there was a concerted program to provide automatic door closures. Even this modification did not stop a determined youth from blocking the door with a foot and leaning out of the opening.

Graffiti was common on the outside of single deck carriages, especially the older type known as ‘red rattlers’. Graffiti was put on the carriages when they were parked overnight and was also sprayed on when the carriages were moving. Some of this graffiti was located around the windows although it was mostly around the doors. Mr Cowling had observed this graffiti but said that when associated with windows it was generally poorly formed, giving him the impression that paint was sprayed by someone putting his or her hand through the window rather than leaning out from it.

Evidence given by the defendant's employees indicated that prior to this event they had observed graffiti in various places including above the windows of trains. The strategy adopted to deal with graffiti was to inspect and clean the carriages every night. Apparently a major motivation for the graffiti vandal was the pleasure in observing his or her ‘tag’ on the train. If it could be cleaned immediately the pleasure was lost and the desirability of applying graffiti diminished. This was the approach adopted by the authorities in New York. It was successful there and largely successful in Sydney.

As part of the nightly inspection the windows were also visually inspected for breakage. However it was not usual to carry out a close inspection of a window and no effort was made to determine whether the rubber gasket remained in place. To do this would have involved a greatly increased staffing requirement and a consequential significant cost.
Mr Phillip McColl, the Fleet Manager of the Flemington maintenance centre of the defendant, gave evidence that in 1994 he would, in the course of his duties, observe up to 160 to 180 carriages per day. He commonly observed graffiti on the inside of carriages and on the outside, within about one arm's length of the doors, windows and ends of trains. Although he said he could not recollect seeing graffiti on roofs, I do not accept this evidence. It is plain from answers he gave in cross-examination, that he had seen graffiti above the upper windows of double-decker carriages and the consequence is that the graffiti must have been on the roof. The only means by which this could be done would be from the upper level window. However, I accept that he did not appreciate that youths were squeezing through the windows and hanging out of the train to spray the roof.
Mr Michael Logue, the manager of fleet presentation at Central station at the time of the accident, also gave evidence. He said that amongst other places he had seen graffiti above the windows of double deck carriages. He confirmed that he had observed it on the roof of trains in a position similar to the plaintiff's ‘tag’ by the time of the accident.
Although I am of the view that the defendant was aware that graffiti was being sprayed on to the roof of double-decker carriages before the accident, I am also satisfied that the defendant was not aware that young people were squeezing through windows and hanging their bodies out in order to spray the roof. If its officers had given the matter careful thought they would have realised this was occurring but because the opening, even without the gasket, was so small, the possibility did not occur to them.
Evidence was given about the design of railway carriages and the clearance for fixed structures beside the track. The minimum separation allowing for movement of the train through roll and wearing of mechanical parts is 525 mm. A youth hanging out of the train could readily extend beyond this safety margin and contact a fixed structure as happened in the present case.”

8 The trial judge then held that although by hanging out of the window the plaintiff had breached a provision in delegated legislation, that did not of itself debar his claim.

9 The trial judge then found that what had happened was reasonably foreseeable:

“the possibility that young people might attempt to hang out of the windows of trains to spray graffiti was foreseeable by the defendant. The evidence of the graffiti on exterior parts of the train including the roof, make it apparent that young persons were leaning out of the carriage to some extent. It must have been obvious that at least their hands and arms were outside the train. I am satisfied that it was foreseeable that some would be able to squeeze their upper bodies through the window if the rubber gasket had been removed or was ineffective. The fact that it may have been difficult to squeeze through does not mean that it was not foreseeable. The risk was neither ‘far fetched or fanciful’ … .”

10 The trial judge then turned to the content of the defendant’s duty. He said:

“Although it was foreseeable that some young people might squeeze through a faulty window, I am satisfied this would only occur with difficulty. By the time of the plaintiff's injury, the defendant was not aware of any accident involving a person hanging out of an upper level window. Although it was possible that an injury might occur in this manner it was not at all probable. Most people, aware of the size of the window opening (and a cut out was tendered in evidence) would, in my view, although accepting the possibility of a determined person squeezing through, believe this to be most unlikely. …
… the risk was only created by a deliberate act of the plaintiff in circumstances where a window was found to be defective. Although by providing opening windows in the carriages, the defendant created the elements which allowed the risk to occur, the injury eventuated because of the plaintiff's act.
The window was originally designed and constructed in such a manner that it cold not be opened far enough to allow a person to pass the upper part of his or her body through it. Whether this was the intention of the designer, or whether that intention was confined to avoiding involuntary passage by small children, is not clear. However, if as the plaintiff suggested, there had been a rivet and the rubber gasket was glued, it is most unlikely that the window would have been able to open beyond its design width. Of course, the window itself might also have been of a wholly different design which prohibited passage. This is now the situation.
By providing the window which it did, even with the gasket removed, the defendant had ensured that considerable agility and determination would be required of a young person intent on hanging the upper part of his or her body out of the window. Even without the rubber gasket, the window opened only a short distance (235mm). I do not believe it would be reasonable to find that the defendant owed a duty to the plaintiff to ensure that the window was always confined to an opening of 115mm.

Furthermore, I am not persuaded that the defendant was obliged to have responded to the possibility of injury to a determined youth by the implementation of an intensified maintenance regime, or accelerated window modification, or replacement program. Even a nightly maintenance program which fixed every window would not have proved a barrier to a determined youth. Such a program would have added significantly to the cost of managing the rail network.

Apart from intensified maintenance, it is obvious that the defendant could have modified the window. However, this would also have involved a significant cost and would, if the efficient running of the metropolitan network was to be maintained, have taken months, if not years to complete. In circumstances where a program was already in place, which, subject to funding being available, would provide for complete replacement of the windows in the near future, to require a special maintenance or modification program to guard against the limited possibility of a determined youth squeezing through a defective window would not be reasonable.
Although the defendant, as the body responsible for the Sydney metropolitan railway system, may, in some circumstances, owe a duty to people who may be breaking the law, I do not believe, in the present circumstances, that duty extended to preventing a young person, intent on disfiguring the train, from deliberately squeezing through the narrow window opening.”

11 The trial judge held that if he were wrong in denying a duty sufficiently wide to conform to the plaintiff’s interests, and if that duty was instead to provide and maintain a window which only opened approximately 115 mm or which did not allow a person to pass the upper part of his body through it, that duty was breached, and the breach was the cause of the plaintiff’s injuries.

The plaintiff’s arguments on appeal

12 The plaintiff submitted that by 6 May 1994 many of the windows in trains of the type on which the plaintiff travelled that day had no rubber gaskets, or had gaskets which were worn and provided no security, and that the defendant must have been aware of those facts.

13 That submission was based on the following matters.

14 By 1989 writers of graffiti had severely damaged the inside and outside of carriages and had been killed while hanging out the doors of trains. The defendant had therefore embarked upon what the trial judge called a “concerted programme to provide automatic door closures”.

15 It was said to follow that the defendant should also have been conscious of the risk of injury to youths attempting to spray graffiti by extending their arms or upper bodies through the windows of trains.

16 Employees of the defendant had observed graffiti above the windows of trains, and in particular Mr McColl and Mr Logue. Though Mr Cowling was not an employee, he too had observed graffiti on the roofs of trains from 1993.

17 Counsel for the plaintiff pointed to various newspaper articles discussing the incidence of graffiti in trains, and the death of some of those responsible in the course of leaning out of trains.

18 Counsel for the plaintiff then attacked the trial judge’s finding that though the defendant was aware that graffiti was being sprayed onto the roofs of double-decker carriages, it was not aware that young persons were squeezing through windows and hanging their bodies out.

19 Counsel for the plaintiff said that the trial judge’s finding that the possibility that young persons were squeezing through windows and hanging their bodies out was foreseeable should have compelled the trial judge to the conclusion that the defendant owed the plaintiff a duty of care to take the “comparatively simple and inexpensive steps” described by Mr Simpson, and his failure to do so revealed an inconsistency.

20 Counsel for the plaintiff supported the submission by referring to the plaintiff’s evidence that it was “very easy” to get out of a window. He said the trial judge accepted that evidence.

21 Counsel for the plaintiff submitted that the relevant windows were designed to be opened only 115 mm, and inspections carried out daily around the time of the accident must have caused the defendant’s employees to realise that the windows were not functioning in accordance with their original design. It was submitted that it was not unreasonable for the law to require a public authority which knew or ought to have known of the conduct in question to take the “comparatively simple protective measure” of ensuring that the windows were brought into conformity with their original design.

22 Counsel for the plaintiff pointed out that there was evidence that in 1984-5 the cost of repairing vandalism-caused damage and cleaning graffiti on trains was estimated at $4.76 million, and submitted that this considerable cost continued thereafter. By 1993 it had been decided to clean trains nightly. It was submitted that the money so employed would have been better directed to effectuation of Mr Simpson’s method for preventing graffiti writers from squeezing through the windows, which indeed was effectuated after the accident.

23 Counsel for the plaintiff advanced detailed submissions on points of distinction between the instant case and Modbury Triangle Shopping Centre Pty Ltd v Anzil (2000) 205 CLR 254. Attention was drawn in particular to the causation difficulties in that case and the lack of them in this, and to the lack of foreseeability in that case and the allegedly high degree of foreseeability in this.

Factual and other preliminary controversies

24 Was there in truth an inconsistency between (a) the finding that it was foreseeable that a person could squeeze through the window and that such a person was at risk of sustaining a very serious injury, and (b) the conclusion that this could only occur with difficulty, and was not probable? There is no inconsistency. The window was very small and the conduct was unusual. The finding that there was a “risk” of a very serious injury is not inconsistent with a finding that it was foreseeable but not probable.

25 The plaintiff’s contention that the trial judge should have found that there was a high degree of foreseeability because of his evidence that it was “quite easy” to get out the window, and that the trial judge accepted this evidence, overlooks the difficulty that he in fact rejected it. He rejected it on the basis that the manoeuvre was intrinsically difficult, and though the plaintiff became adept at it it was only because he had done it on many occasions.

26 The plaintiff’s submission that since at the time of the accident many of the defendant’s trains were without gaskets or had worn gaskets, they “provided no security” is a tendentious way of describing the opportunity afforded to the plaintiff and others to do what they did. Even without the gaskets, the windows opened only 235 mm, and the plaintiff’s manoeuvre required considerable agility and determination, according to findings of the trial judge which have not been challenged. There was also an unchallenged finding that to determine whether a rubber gasket remained in place would require a close inspection of each window during nightly cleaning, and this could not be done without “a greatly increased staffing requirement” and “a consequential significant cost”.

27 That point also bears on the plaintiff’s submission that the defendant “must be taken to have been aware” of the deficiencies in the gaskets, because the finding negates awareness of the risk for which the plaintiff contended, ie that “many” of the trains had the deficiency.

28 The plaintiff’s criticism of the trial judge’s failure to find that the defendant was aware of the manoeuvre employed by the plaintiff and others fails. There was evidence that Mr McColl and Mr Logue saw graffiti above windows on the roofs of trains, but in positions where it could have been placed by a person standing inside a train and moving only his arm outside the window. Mr Cowling, who had examined many of the defendant’s documents and other materials, found nothing to indicate that people were squeezing their bodies through windows to spray graffiti before the accident. The trial judge did not reject this evidence and no error has been demonstrated in his refusal to do so.

29 There is a qualitative difference between the manoeuvres by which youths extended their arms out of doors, windows and the ends of carriages, and the manoeuvre employed by the plaintiff, with the whole of his torso out of the window and his buttocks resting on the window edge. The relevant witnesses admitted to, or were found to have, the former awareness. They denied the latter awareness, and no argument was advanced which demonstrated that the trial judge was wrong in accepting the denials. But awareness of the latter manoeuvre cannot be inferred from knowledge of the former.

30 The reliance on newspaper reports to demonstrate the defendant’s knowledge of the practice engaged in by the plaintiff fails. Not all the articles relate to the period before the plaintiff’s accident. In general they are summaries without precise sources being ascribed. They deal with hanging out of doors as well as hanging out of windows. It is not clear that the windows referred to are windows in carriages of the type in which the plaintiff was injured: there were other types of carriage in which it was possible to lean out of the windows.

31 The defendant placed considerable reliance on various things said in Modbury Triangle Shopping Centre Pty Ltd v Anzil (2000) 205 CLR 254, while the plaintiff was at pains to distinguish that case. Thus the defendant pointed to what was said in that case about identifying the nature of the harm suffered, the relevance of lack of control by the defendant, the relevance of criminal conduct, the lack of reliance by the plaintiff on the defendant, the fact that the defendant had not assumed responsibility for the plaintiff, and the relevance of whether a special relationship existed. Save to the extent that the Modbury case deals with the law of negligence generally, it is desirable to put it on one side in the instant case. The Modbury case was sharply focused on the problems that arise from criminal conduct by third parties not by plaintiffs. It is true that in principle the difficulties it described in relation to injuries caused to plaintiffs by the criminal conduct of others are likely to have parallels in relation to injury caused to plaintiffs by their own criminal conduct. The plaintiff contended that it had not been demonstrated that he had engaged in any crime at the time of his accident. His arguments in that respect are not convincing, but it is not necessary to reach a conclusion about them. Since it is possible to decide the instant case without using the specific reasoning in the Modbury case, it is undesirable to consider whether that case has any impact on the liability of defendants for injuries caused by the illegal conduct of plaintiffs as distinct from third parties.

32 There was a controversy between the parties as to the necessity for proving knowledge in the defendant of the plaintiff’s conduct. The defendant said there was. The plaintiff said that constructive notice sufficed. Each side claimed to gain support from Crimmins v Stevedoring Industry Finance Committee (1999) 200 CLR 1 at [101]-[102] per McHugh J. The defendant could have obtained comfort from the fact that particular (viii) in the Particulars of Negligence in the Amended Statement of Claim based the alleged negligence on the defendant’s “knowledge”. The controversy appears to go only to foreseeability: if the defendant knew that youths were engaging in the conduct on which the plaintiff was engaged, it would increase the foreseeability of the risk. Despite the way particular (viii) is framed, however, if all other obstacles to recovery were overcome, the plaintiff would not necessarily fail on the ground that he did not prove that the defendant knew various things, so long as he proved that the defendant ought to have known them.

33 The plaintiff’s argument that the defendant ought to have employed the funds spent in removing graffiti each night in remedying the windows, which measure would have prevented the placing of the graffiti, must fail. It was not put to any witness at the trial. The argument assumes that most of the graffiti was applied by people using the windows, whereas there is evidence that they used doors and the ends of carriages as well, and Mr Cowling said that the bulk of the graffiti was actually applied at night while the carriages were “stabled” in carriage sheds and workshops.

34 Counsel for the plaintiff rightly identified a crucial difference between the positions of the parties. He said that while it was common ground that reasonable foreseeability of injury alone did not create a duty, the defendant conceded a duty: the controversy was as to its scope. The parties were agreed that the duty was to take reasonable steps to provide a safe railway carriage. But while the defendant said that the duty was to exercise care in relation to dangers likely to arise from the ordinary use of the carriage which might reasonably be expected, the plaintiff said it was to exercise care in relation to dangers which might reasonably be expected, having regard to the defendant’s actual or imputed knowledge of the behaviour of all passengers. That is the central issue in the case and it is now necessary to turn to it.

Conclusions on duty of care

35 The plaintiff has suffered a calamity and it is understandable that he, his family and their advisers should seek to use the law of tort to overcome the financial consequences of that calamity. But there are fundamental difficulties in the plaintiff’s case. These emerged in several ways.

36 First, difficulties emerged in the Particulars of Negligence pleaded in the Statement of Claim filed in 1997. These were as follows:

“(i) Failure to prevent the windows of the train from being opened more than approximately ten centimetres such that they would not allow a person particularly a child to exit through any of those windows.
(ii) Failure to ensure that the windows of the train would not open sufficiently to stop a person particularly a child from putting their head and/or other bodily part through any window.

(iii) Failure to warn or adequately warn the plaintiff of the danger of climbing or attempting to climb through any window of the train and of putting his head or other part of his body through any window in it.
(iv) Failure to prevent the plaintiff from putting his head or any other part of his body through any window in the train.
(v) Failure to adopt proper and reasonable precautions to ensure the safety of the plaintiff then being a child of fifteen years of age while he was a passenger on the train.
(vi) Failure to staff the said train adequately so as to ensure that the infant plaintiff was properly controlled and managed.
(vii) Failure to staff the said train adequately so as to ensure that the infant plaintiff did not put himself in a position of danger by putting his head and any part of his body out of the window of the train.”

While the trial judge’s understanding was that particulars (vi) and (vii) were not pressed at the trial, as a whole the particulars are significant. It is not correct, in the factual context under consideration, to view the plaintiff, as these particulars do, as a “child” or as an “infant”. A normal fifteen year old like the plaintiff is as fully able to understand the dangers of what he did as the most intelligent of mature adults. The trial judge specifically found that the plaintiff “would have been able to appreciate that there were risks involved when leaning out of the window on a moving train. He was not an infant or child who needed protection from his own involuntary or unthinking acts”. Further, the idea that the defendant was negligent in failing to warn the plaintiff of the dangers is unconvincing. They were dangers which were fully known to the plaintiff, and it was the fact that they were dangerous which made his conduct attractive to him. The suggestion that it was the defendant’s job to provide staff who could “control” and “manage” the plaintiff and “ensure” that he did not put himself in a position of danger is also unconvincing. One reason why it is unconvincing is the deliberate search the plaintiff habitually conducted to find suitable windows. Another is the fact that he had done what he was doing on 6 May 1994 every day for six months. One can imagine the plaintiff’s reaction to endeavours by the defendant’s staff to counsel him against his conduct. Particulars of negligence which might have been appropriate to a very young child injured while seeking a better view or feeling the wind on its hand are irrelevant in relation to the present plaintiff.

37 Perhaps because of the realisation of the unsatisfactory nature of the above particulars, a further particular was added when the Statement of Claim was amended on 8 December 1999 (though the existing seven particulars continued to appear):

“With knowledge in the defendant that for some time prior to the accident on 6 May 1994 persons of similar age and stature to the plaintiff were in the habit of extending themselves through windows of trains in the Sydney metropolitan area operated by the defendant for the purpose of spraying graffiti on those trains, its failure and neglect to take all such steps as were reasonably necessary to prevent such persons (including the plaintiff) from doing so.”

That particular failed at the trial because the trial judge found that there was no relevant knowledge. But it does highlight the unrealistic nature of the other particulars.

38 The difficulties in the plaintiff’s case manifested themselves in an argument advanced on behalf of the plaintiff that what happened to the plaintiff was analogous to a passenger being injured on leaning against a defective window, and that it was therefore wrong to view the essence of the harm which the plaintiff suffered as an injury the direct and immediate cause of which was being struck on the head by an external object while the upper part of his body from the hips was hanging or suspended outside the window. The substantive answer to that argument is that there is no true analogy between an injury suffered by a passenger using the carriage in the way it was intended and an injury suffered by the present plaintiff while carrying out a dangerous, foolhardy and reckless act in the course of which he was damaging the defendant’s property, and thereby using the window for a purpose for which it was not intended.

39 A further difficulty in the plaintiff’s case revealed itself in arguments advanced in reaction to certain phrases in the defendant’s written submissions to this Court. The defendant submitted that the plaintiff was contending for “a duty … to take reasonable care to protect persons in the position of the appellant from their own conduct, including reckless and criminal conduct”. The defendant also argued that the Modbury case “identified as a consideration the unpredictability of criminal behaviour. That was in the context of criminal behaviour by third parties. The respondent submits there is no real difference when that criminal behaviour is performed by the injured party.”

40 For reasons given above, it is unnecessary to rely on the specific reasoning in the Modbury case to deal with the instant case. But what is presently significant is that counsel for the plaintiff objected to the court entertaining any contention that the plaintiff had been engaging in criminal conduct, eg malicious damage to property, on the ground that the issue had not been raised at the trial, and, if it had, evidence could have been given which could possibly have prevented the contention from succeeding. This evidence was not identified. Nor was it explained how it could have been tendered in view of the plaintiff’s lack of recollection and a want of any other witnesses, beyond the contention that the youth, immaturity and want of intelligence of the plaintiff might have deprived him of mens rea. On the other hand, the defendant submitted that the fact that the plaintiff had been engaged in criminal acts had never been in issue at the trial, and that fact had been treated as a matter relevant to foreseeability. The defendant pointed to two passages in its written submissions to the trial judge which corresponded word for word with the passages in its written submissions on appeal of which the plaintiff complained. It also pointed to another passage which described the plaintiff as having “knowingly performed an illegal and dangerous act”. To that may be added the fact that By-Law 20 prohibits damaging the defendant’s property whether wilfully or negligently: this appears to leave very little room for mens rea to operate.

41 Unpersuasive though the plaintiff’s contention on this point is, it is not necessary to arrive at a conclusive resolution of the controversy. What is plain is, as counsel for the plaintiff conceded, that the plaintiff behaved recklessly, and engaged in inherently dangerous activity, in putting his head out the window far enough for it to be hit as it was. It is also plain that the plaintiff was committing a tort against the defendant’s property. Either he was committing the tort of trespass to goods, or, if the argument about mens rea meant that his conduct could not be described as intentional interference, he was committing the tort of negligence in relation to the defendant’s chattel, the carriage. It is true that while the defendant pleaded contributory negligence in its Defence, it did not plead either of these torts, but they are inherent in what the plaintiff was doing. And the trial judge must have thought that there was no difficulty about the plaintiff’s mens rea at least as far as civil law is concerned. He said:

“Although the defendant … may, in some circumstances, owe a duty to people who may be breaking the law, I do not believe, in the present circumstances, that duty extended to prevent a young person, intent on disfiguring the train, from deliberately squeezing through the narrow window opening.”

It would appear likely that by “breaking the law” the trial judge meant the criminal law as well as the civil law, but it is not necessary to decide that either. It must also be remembered that the plaintiff admitted in cross-examination that he had been apprehended by the police in the first three months of 1994 for applying graffiti to a bus shelter, for applying silver and red spray paint to a public toilet, and for spraying graffiti on the walls of an industrial warehouse. He said he could not remember being detected spraying graffiti inside a railway carriage, running away and being caught. The trial judge found:

“The plaintiff had on three occasions, been apprehended by the authorities for his graffiti activities. On the last occasion he was sentenced to a term of community service.”

42 The presentation of the plaintiff’s case does not face up to the fact that what happened was not some sudden piece of misjudgement, but part of a stream of conduct, namely writing graffiti, in which the plaintiff had engaged for years on exposed surfaces other than trains and for six months on a daily basis on trains. It was conduct which the plaintiff knew the authorities opposed, since they had apprehended him three times and punished him once. The particular manoeuvre was known to him to be dangerous. Indeed, that was part of its charm: he placed graffiti on the roof “so more people can see it” and because it is “braver or bolder” to place graffiti on the outside of the train rather than the inside. Although it is not necessary to decide whether the particular acts of the plaintiff on 6 May 1994 were criminal or not criminal for want of mens rea, because it is not necessary to decide this case by reference to that aspect of the Modbury case, the plaintiff’s arguments that his acts on 6 May 1994 were not in themselves criminal highlight the difficulties which they raise.

43 To some extent the present case is unusual. In unusual cases it is convenient to return to first principles.

44 The fundamental starting point must be that the tort of negligence rests on reasonableness. In Tame v New South Wales [2002] HCA 35 at [8], Gleeson CJ said that the tort:

“involves a failure to protect the interests of someone with whose interests a defendant ought to be concerned. A definition of the ambit of a person’s proper concern for others is necessary for a decision about whether a defendant’s conduct amounts to actionable negligence. The essential concept in the process of definition is reasonableness. What is the extent of concern for the interests of others which it is reasonable to require as a matter of legal obligation, breach of which will sound in damages?”

45 Gleeson CJ then referred to the observation of Lord Atkin in Donoghue v Stevenson [1932] AC 562 at 580:

“Who, then, in law is my neighbour? The answer seems to be – persons who are so closely and directly affected by my act that I ought reasonably to have them in contemplation as being so affected when I am directing my mind to the acts or omissions which are called in question.”

Gleeson CJ then said (at [9]):

“It is the reasonableness of a requirement that a defendant should have certain persons, and certain interests, in contemplation, that determines the existence of a duty of care.”

Gleeson CJ also pointed out that one theme in the authorities turns on an analysis of which relationships create duties. He referred to what Lord Macmillan said in Donoghue v Stevenson at 619:

“In the daily contacts of social and business life human beings are thrown into, or place themselves in, an infinite variety of relations with their fellows; and the law can refer only to the standards of the reasonable man in order to determine whether any particular relation gives rise to a duty to take care as between those who stand in that relation to each other.”

46 Gleeson CJ then said at [13]:

“What a person is capable of foreseeing, what it is reasonable to require a person to have in contemplation, and what kinds of relationship attract a legal obligation to act with reasonable care for the interests of another, are related aspects of the one problem. The concept of reasonable foreseeability of harm, and the nature of the relationship between the parties, are both relevant as criteria of responsibility.”

47 Gummow and Kirby JJ said at [185]:

“A fundamental objective of the law of negligence is the promotion of reasonable conduct that averts foreseeable harm. In part, this explains why a significant measure of control in the legal or practical sense over the relevant risk is important in identifying cases where a duty of care arises. Further, it is the assessment, necessarily fluid, respecting reasonableness of conduct that reconciles the plaintiff's interest in protection from harm with the defendant's interest in freedom of action. So it is that the plaintiff's integrity of person is denied protection if the defendant has acted reasonably. … Moreover, the assessment of reasonableness, which informs each element of the cause of action, is inherently adapted to the vindication of meritorious claims in a tort whose hallmark is flexibility of application.”

48 The plaintiff’s written submissions contain the following warning:

“Doubtless these proceedings may excite many and varying ‘subjective’ or ‘philosophical’ views on the ‘state of our society’. Among the spectrum of thought many may, quite sincerely, entertain the notion that the appellant’s claim ought, as a matter of principle, be rejected out of hand having regard to his anti-social behaviour on the day in question. But, against that, there exists the undoubted consideration that this young man has been very severely wounded and his life permanently affected. Of course, ultimately, the matter is one of law for the Court’s determination.”

That was a prudent and fair warning to give. It is not intended to suggest that the plaintiff’s claim should be rejected out of hand. But it is necessary to notice features of it which do not conform with the elements of a tort based on reasonableness.

49 It would offend criteria of reason to permit the plaintiff to recover in circumstances where:

(a) No employee or officer of the defendant had actual knowledge of the plaintiff’s conduct.

(b) The plaintiff’s conduct was not probable, and though it was reasonably foreseeable, it was only foreseeable at the low end of the relevant scale.

(c) The plaintiff’s conduct involved exploiting the failure of a design feature which, though not introduced with the plaintiff’s conduct in mind, did have the purpose of stopping passengers from injuring themselves by placing their heads outside the window for mundane purposes like obtaining a better view or some fresh air.

(d) The conduct involved using the window for a purpose for which it was never intended, namely, tortiously damaging the defendant’s property: the plaintiff’s case requires the defendant to pay him for injuries caused to him while he was causing injuries to the defendant.

(e) The plaintiff’s conduct was extraordinarily dangerous, reckless and foolhardy, and was undertaken in part for that very reason, since it proved the plaintiff’s braveness and boldness: it is unreasonable to place a duty of care on the owner of property which is being used against that owner’s interests and without that owner’s knowledge or consent in a highly dangerous way, because while the law recognises that people may be careless without losing their right to recover damages, it is not unreasonable for the law to expect them to abstain from self-destructive conduct, or, if they will not do so, to bear the sole personal responsibility for the outcome.

(f) The plaintiff’s conduct was one item in what must to the defendant have seemed like a war waged on it by a generation of youths involving vandalism and graffiti writing of all kinds, each aspect of the war costing significant sums of money, and each aspect of the war rendering it harder for the defendant to overcome the results of any other aspect.

(g) The plaintiff, having with others in his position caused the defendant considerable expense and trouble for years, seeks substantial compensation from the defendant for injuries suffered while engaged in the very conduct that caused that expense and trouble. It is true that behind the defendant lies that large section of the citizenry which uses trains, and that the economic choice in the case is between making an order that that section of the citizenry compensate the plaintiff and making an order which will have the consequence that that section of the citizenry which pays federal income tax and funds the social security payments which the plaintiff will receive if he fails against the defendant will have to do so. These sections of the citizenry are distinct, though they do overlap. It does not, however, seem reasonable that the former segment should pay unless some specific feature of the law requires it.

50 The instant case can be decided on the simple basis that the applicable law was stated in Henwood v Municipal Tramways Trust (South Australia) (1938) 60 CLR 438. A tram passenger, feeling ill, leaned over a rail out of the tram in order to vomit. His head struck two steel standards and he died. His conduct was in breach of a by-law making it a criminal offence. Napier J held that the breach of the by-law was a conclusive defence. The High Court disagreed. It is not necessary to examine the reasoning of the High Court on that point, since the defendant here does not contend that any illegality in the plaintiff’s conduct of itself debars recovery. The High Court divided equally on the question whether there should be judgment for the plaintiff, or whether the matter should be remitted to Napier J to determine whether the defendant was in breach of duty. Dixon and McTiernan JJ favoured the former course; Latham CJ and Starke J favoured the latter course. The latter course was adopted.

51 What is important for the instant case is what the High Court said about the duty and standard of care. Latham CJ said at 444:

“The duty of the carrier [of passengers] is to use due care to carry the passengers safely … .”

Starke J said at 452:

“the duty of a carrier of passengers is to carry them safely so far as reasonable care can do it.”

But neither of these two justices analysed the matter as fully as did the joint judgment of Dixon and McTiernan JJ at 455-456 and 466:

“The injury bringing about the deceased's death was sustained by him because, when he leaned over the rail to vomit, the moving tram-car brought his head into contact with the post or standard carrying the overhead wires. Should the constructional design of the tram and tramway which made this possible be treated as a condition of the premises occupied by the defendant trust in respect of which they owed a duty to persons like the deceased coming upon them as passengers? If so, the measure of the primary duty of the trust to him would depend upon the description which he filled among the well-known classification of persons coming upon the premises of another. As he paid his fare as a passenger, we should suppose that a contractual relationship existed between him and the trust. Thus, in respect of the state of the premises, he would be entitled to expect that they should be as safe as reasonable care and skill could make them. But we are disposed to think that the liability of the trust for such a thing as befell the deceased should be determined, not by the rules governing the responsibility of the occupiers of structures or premises in respect of their unsafe condition when a stranger coming upon them is hurt, but by the legal duty resting upon a carrier of passengers to exercise due diligence to carry them safely.
It has long been settled that a carrier of passengers incurs no higher responsibility than that of exercising reasonable skill and care for the purpose of carrying them safely and securely. A high degree of precaution is necessary before, in such a matter, the standard of reasonable skill and prudence is attained. But carriers of passengers, even if by statute they be common carriers, do not insure safety, and their duty in respect of the transportation of passengers is measured by no exceptional rule. …

In any case, we are of opinion that the danger to passengers who lean out or project any part of their bodies is so serious and, among the many carried, the likelihood of a passenger doing so instinctively, or impulsively, or forgetfully, or as a result of illness or other physical condition, is high enough to make mere warning insufficient to discharge the duty of care. …
We think that the obligation of the trust was to exercise due care for the safety of passengers from dangers likely to arise out of the ordinary use of the tram which might reasonably be expected and that the case should not be treated as one where the danger consists in defective premises or a dangerous structure.”

52 While the remarks of Dixon and McTiernan JJ are not part of any ratio decidendi, they have never been doubted. They conform to the principles of a tort resting on reasonableness. The considered dicta of High Court judges, while not strictly binding on this Court, are “entitled to the greatest respect” and are “of the greatest persuasive authority”, particularly when one of them is Dixon J. See Bone v Commissioner of Stamp Duties [1972] 2 NSWLR 615 at 654 at 664 per Jacobs P and Hope JA; Businessworld Computers Pty Ltd v Australian Telecommunications Commission (1988) 82 ALR 499 at 504 per Gummow J. The remarks of Dixon and McTiernan JJ should be followed by this Court unless plainly wrong. The plaintiff has not pointed to any case disagreeing with what they said, nor to any error of principle in what they said. Hence what they said should be applied.

53 When the plaintiff in the instant case projected his body outside the window, he did not do so instinctively, impulsively, forgetfully or as a result of illness or other physical condition. Any duty on the defendant to prevent that conduct was fulfilled by the provision of a window only capable of being opened 235 mm. Further, what the plaintiff did was not part of the “ordinary use” of the carriage, and the danger to him was not “likely to arise out of the ordinary use of the carriage”. The defendant had taken precautions against dangers arising from the ordinary use of the carriage, for example by passengers desiring to put their heads out of the window for a better view or air or to relieve illness, or by passengers desiring to put their arms out. The defendant’s duty did not extend so far as to oblige it to take precautions against passengers seeking to sit on the window edge with their entire torso and head outside with a view to damaging the defendant’s property.

54 There are several factors pointing against any widening of the duty as formulated by Dixon and McTiernan JJ.

55 One problem in widening the formulation so as to create a duty is that if the duty existed it would be necessary to take account not only of reasonably foreseeable reckless harm by travellers using windows, but also by travellers using the doors, and no doubt in other ways.

56 Another factor is that duties of care should not be recognised if they would subject defendants “to an intolerable burden of potential liability, and constrain their freedom of action in a gross manner”: Sullivan v Moody (2001) 183 ALR 404 at [42] per Gleeson CJ, Gaudron, McHugh, Hayne and Callinan JJ. It is wrong that this defendant’s freedom of action should be constrained by having to spend even larger sums than it did to stop conduct of the type which injured the plaintiff and other reckless conduct.

57 Thirdly, in Reeves v Commissioner of Police of the Metropolis [2000] 1 AC 360 at 268 Lord Hoffmann spoke of:

“the sound intuition that there is a difference between protecting people against harm caused to them by third parties and protecting them against harm which they inflict upon themselves. It reflects the individualist philosophy of the common law. People of full age and sound understanding must look after themselves and take responsibility for their actions. This philosophy expresses itself in the fact that duties to safeguard from harm deliberately caused by others are unusual and a duty to protect a person of full understanding from causing harm to himself is very rare indeed.”

At 379-380 Lord Hope of Craighead said:

“It is unusual for a person to be under a duty to take reasonable care to prevent another person from doing something to his loss, injury or damage deliberately. On the whole people are entitled to act as they please, even if this will inevitably be to their own death or injury. …
But the duty of care may sometimes extend to preventing people injuring themselves deliberately. The person to whom the duty is owed may be unaware of the risks to which he will expose himself by his deliberate act. Or he may be too young to appreciate them. … Or he may be of unsound mind, with the result that he is at risk of doing something to himself which no rational person would do as he would appreciate that to do this would inevitably lead to injury. Or the risk that the person may commit an act of deliberate self-harm may be the result of something which the defendant has done or is doing to him.”

58 Here the conduct of the plaintiff went very close to being deliberately self-inflicted harm. It was certainly recklessly self-inflicted. None of the limited areas where a duty of care exists to prevent persons from inflicting harm on themselves applies. In particular, though the plaintiff was aged 15, he cannot relevantly be said not to have been “of full age” or “too young to appreciate” the risks of what he was doing.

59 Counsel for the plaintiff relied on Ryan v State Rail Authority of New South Wales [1999] NSWSC 1236, in which a child injured while trespassing on a train was nonetheless held to be owed a duty of care. The court treated that case as an allurement case, and counsel for the plaintiff conceded that the present case not an allurement case.

Breach

60 The trial judge considered that if there were an applicable duty, the defendant was in breach of it. The defendant challenged that conclusion by Notice of Contention. If the accident was reasonably foreseeable (as the trial judge held) and all other conditions for the recognition of a duty were satisfied (which he did not hold), it would be necessary to consider what the response of a reasonable man would be: see Wyong Shire Council v Shirt (1980) 146 CLR 40 at 46-47. That in turn depends on analysing the magnitude of the risk, the degree of probability of its occurrence, the expense, difficulty and inconvenience of taking alleviating action, and any conflicting responsibilities which the defendant had. The conclusion of this process may be that the reasonable man would make no response.

61 Here, though if the risk came home the consequential injuries could be very serious, the risk was reasonably foreseeable but not probable: the degree of reasonable foreseeability was at the low end of the scale. Indeed, the incidence of injury had been very low considering the number of times the plaintiff, his friends, and the many others of their generation had engaged in the practice in question. Even without the rubber gaskets in place, the windows opened only to the extent of 235 mm, less than 10 inches. The defendant was obliged to keep its passenger fleet in action, and 94% of available carriages were in use every day, according to Mr Cowling. While the plaintiff was correct in submitting that the actual process of modifying each window to make it safe was a simple process, the submission that the task of ensuring that all windows were safe was simple was erroneous. This is because if the daytime requirements of the defendant were to be met, the work could only be done at night. There were 745 carriages. Two people working together could have glued the rubber gaskets at the rate of two carriages per night. That would take at least a year. If screws were inserted as well, the process would take twice as long. Among the conflicting responsibilities of the defendant was a need to upgrade the silver double-decker carriages to provide automatic locking doors to prevent people hanging out of them, the cost being $4.5 million, and also to effect other improvements, to remedy damage caused to Tangara carriages by vandals attracted by their large external glass areas, and to have more non-Tangara rolling stock in action in consequence to make up for Tangaras out of action. The result was that if the identified modifications in the silver double-decker carriages were carried out, this could only be done compatibly with maintenance of the overall fleet at existing levels of service by doing the work at night over a period of years; to do the work faster would greatly increase its cost. Thus to overcome the problem would either call for very large expenditures or would create major disruptions to the entire Sydney train system to deal with a possibility which was reasonably foreseeable, but only just. The reasons given by the trial judge for not finding a duty are in fact convincing reasons for concluding that even if there was a duty it was not breached. Hence the Notice of Contention should be upheld.

Causation

62 The defendant submitted that even if there were a relevant duty and it had been breached, the plaintiff would still have failed to establish causation, because in practical terms the real cause of the injury was the conduct of the plaintiff. In view of what has been said above it is not necessary to consider this argument.

Orders

63 The following orders are proposed.

1. The appeal is dismissed.

2. The appellant is to pay the respondent’s costs of the appeal.

64 YOUNG CJ in Eq: I agree with Heydon JA.

65 FOSTER AJA: I agree.

**********