BLUE & WHITE BARRA PTY LTD v SOLLEY No. SCCIV-00-1034 [2001] SASC 194

PERRY J. The appellant was the plaintiff in proceedings in the District Court in which it sued the respondent to the appeal, Robert James Solley for damages for negligence.

At the relevant time, the appellant was the owner of a fish farm which it conducted in a large shed at Murray Bridge. On 9 May 1998 a fire occurred which substantially damaged the shed and the equipment in it associated with the fish farming operation. It was common ground at the trial, and the learned trial judge so found, that the fire began in an electrical switchboard.

The respondent is, and was at the time, an electrician carrying on business at Murray Bridge under the name "Robert Solley Electrical". When the fish farm was established by the appellant, the respondent performed all the electrical work, including the installation of the switchboard and the wiring associated with the conversion of the premises, which had previously been used for other purposes. Subsequently, on the instructions of the appellant, the respondent modified some of the electrical installation, and from time to time attended to maintenance work. No other electrician performed any electrical work in the premises once they had been converted for use as a fish farm.

Although the precise cause of the fire remains unknown, the appellant contended that on the balance of probabilities it must have been occasioned by faulty electrical work performed by the respondent.

The learned trial judge upheld the respondent's defence of the proceedings and dismissed the appellant's claim. In case his finding as to liability was disturbed on appeal, notwithstanding his dismissal of the claim, the learned trial judge proceeded to assess damages.

The appellant appeals against the learned trial judge's findings on the issue of liability and also as to the assessment of damages. The respondent cross-appeals on the issue of damages alone.

In order to understand the arguments raised on the appeal, it is necessary to explain the manner in which the fish farm was set up, more particularly the nature of the electrical work performed by the respondent.

The appellant company was incorporated in 1996 by Maxwell White and Preston Nourse, both of whom became directors of the company.

Mr White had experience as a builder and manufacturer. He was described by the learned trial judge as a "practical, hands-on man". When he perceived what he thought to be an opening for the development of an aquaculture business for the farming of barramundi, he persuaded Mr Nourse to invest much of the initial capital. On the evidence, it appears that of the two directors it was Mr White who had most to do with the physical setting up of the business and the management of its operation.

In 1996, the appellant leased the shed at Murray Bridge in which the business was established. Previously it had been used as a poultry shed.

To adapt it for use in the fish farming operation, the appellant re-lined the shed and installed a number of water tanks in which to hold the fish. The tanks were quite large; about five metres in diameter and about a metre deep. They were constructed either of masonry or steel.

The operation involved rearing fish bought as fingerlings, about 70 millimetres long, over a period of about six months until they were of the order of 600 grams in weight. They were then sold.

At first, the appellant installed six tanks, but later, before the fire, put in a further four, making ten in all. This enabled the appellant to hold fish at various stages of development so that they could be introduced as fingerlings and sold off at the desired level of maturity, in a continuous cycle.

It was necessary to maintain each tank at a constant temperature, which was achieved by using heaters. The first six tanks were heated by electric immersion heaters, and the other four by gas.

The water had to be kept clean and free of the accumulation of ammonia which was generated by excreta and other waste. To maintain the condition of the water, each tank was fitted with two filter systems, operated by separate, electrically powered pumps.

One filter system, which operated rather like a swimming pool filter, aerated and cleaned the water.

The other filter system was described as a "bio-filter", which was suspended above the tank. Water from the tank was pumped up to the bio-filter and percolated through it back into the tank. That process removed the ammonia.

Three-phase electric power was drawn from the mains, supplied from a pole in the street connected to a junction box on the outside of the northern end of the shed. At that point three fuses, known as cartridge fuses, protected the mains supply, one for each phase. The power was then connected to a switchboard inside a small office, built as a lean-to and situated on the outside of the western wall.

There were in fact two switchboards inside the office, but at the time of the fire only one of them was being used. This was known as switchboard "A". Another switchboard, known as switchboard "B" had been installed with a view to supplying power to additional tanks at some time in the future.

At the time of the fire, switchboard A was supplying power by separate circuits to the heaters on the six tanks which were heated electrically, to various power points, to lights and to switchboard B.

There has been no challenge to the learned trial judge's finding that no fault attached to the respondent with respect to his selection of the materials, including the switch boards used in the electrical installation, or in his design or layout of the work which he performed. As the trial judge put it:

"Apart from inferences I am asked to draw from the occurrence and location of the fire, there is no specific evidence of faulty workmanship by the defendant."

The work associated with the initial electrical installation was commenced by the respondent in February 1997. That work mainly involved the installation of the two switch boards (or boxes, as they were sometimes described) inside the office, together with cabling along the walls for the heaters used in the tanks, and power points from which to operate the pumps.

The heaters for the six tanks which were first installed each had elements which were submerged in the water, together with a relay box, timer and thermostat attached to them. Also there was a linkage to a security system which was activated if the pumping of water stopped for any reason.

Initially, the power supply within the premises was 240 volts.

When the appellant found the 240 volt heaters to be too expensive, it took advice from the company which supplied the heaters, as a result of which it replaced the heaters with three-phase heaters which had three elements in them for each tank and which could run at night on a night-time tariff. This involved an increase in voltage from 240 volts to 415 volts.

The respondent effected the necessary changeover of the heaters and a later modification of them which occurred when it was found that the three elements which were part of the conversion to three-phase heaters were still drawing too much current. In each tank one of them was removed, leaving two elements in each tank. The reduction to two elements occurred in about August 1997.

The four additional tanks, which were heated by gas, were installed between December 1997 and January 1998. The requirement for power for the pump motors for those tanks was attended to by the respondent, who extended the power supply to the side wall near those tanks for that purpose.

So it was then that between January 1998 and May of that year when the fire occurred, the appellant had ten tanks in use.

In January 1998 the respondent was responsible for some other work. This involved the installation of a large diesel powered generator. The generator was wired in by the respondent so that in the case of an emergency, if there was a failure of the mains power supply, the diesel generator would cut in and take over.

The wiring-in of the diesel generator involved modification of the internal wiring of switchboard A. At that time Mr White observed Mr Solley fitting additional wires into the switch box. Mr White obtained the impression that Mr Solley was able to accommodate the additional wiring in the box only with some difficulty. Mr Solley denied that this was so.

At all events, it appears that Mr Solley next attended at the premises on 28 April 1998. On that day Mr White was sitting at his desk in the office when he noticed that a circuit breaker had gone off. He rang Mr Solley, who came over.

When he arrived, Mr Solley tried to re-set the circuit breaker by flicking it back into position, but it "threw out" again. This suggested that there was a problem in the tank heater which it serviced.

Mr Solley deduced that the circuit breaker serviced tank 5. He went to tank 5 and felt the elements under the water with his fingers, to see if there was any hole or irregularity, but he was unable to detect any problem in that respect.

He then took the lid off the junction box leading to the elements and removed the wiring to one of the elements, which effectively disconnected one of the two elements.

There is a divergence between the account given by Mr Solley on the one hand and the account given by Mr White on the other, as to what then transpired. According to Mr White, Mr Solley:

"... fiddled with the wires and then asked me if I could go back and switch on the circuit breaker in the office, which I did. As soon as I flicked the thing, it went off with a real loud bang and there was a flash out of it, so I went back to him and I said, 'That went off with a hell of a bloody bang' and I probably swore 'and a flash' and he said to me 'Well, that must be the other element'."

According to Mr White, Mr Solley then did some further work to the junction box at the tank, following which Mr White returned to his office. According to him, Mr Solley spoke to him a little later to say that the problem had been fixed.

According to Mr Solley, when he first removed the wiring to one of the elements, he, rather than Mr White, returned to the switch board where he successfully re-set the circuit breaker without it blowing out again. He then returned to the tank and rendered safe the wiring which had been exposed when he disconnected the faulty heater. After reporting what he had done to Mr White, during the course of which he suggested that the timer for the tank be re-set in view of the fact that it was then working with one heater only, he left.

The fire occurred ten days later, on 9 May 1998. Mr Solley did not attend the premises during the intervening time, and there is nothing to suggest that there was anything untoward during that period in the functioning of the electrical system.

According to Mr White, at 10 to 4 am on the day of the fire he received a call from the monitoring service to say that the alarm had gone off. He went straight to the premises. He saw flames coming out of the top of the office and smoke pouring out of the ridge capping along the full length of the building and also out of the doors at the front end.

The only door which he was able to open was to the office. When he did so he could see flames coming out of switchboard A and going up the wall from it.

The Trial Judge's Findings as to the Incident of 28 April

After the fire, investigations were carried out by James Watson, a technical officer employed by the Department of Primary Industries and Resources, which has the management of the Electricity Act and the Gas Act, and by Henry Kutek, a forensic scientist. They interviewed Mr White. According to them, he did not mention the "flash-bang" incident. It was not until sometime later that he first is recorded as having said anything about it.

It is important to note what the learned trial judge held as to this incident.

I have explained the differences in the evidence as between Mr White and Mr Solley.

The learned trial judge reached the following conclusion:

"I must decide between the two versions. Each was clear about it. It is a matter far more likely to be recalled by Mr White, who was shocked by it, than by the defendant, who might be expected to be more used to it. As against that, Mr White's version was not documented until a good while after the fire and, given his belief that it was significant, that is odd. There is some force in the defendant's assertion that he just would not allow another person to activate circuits on which he was working. I prefer the defendant."

That finding was not placed under challenge by the appellant on the hearing of the appeal.

As I have already noted, it was common ground that the fire began in switchboard "A".

The evidence was that the faceplate or escutcheon of switchboard "A" had a hole burned through it. The learned trial judge found that hole was caused by arcing of electricity between that part of the escutcheon and some nearby source of live power. There were further holes burnt into the rear of the metal box in which the switchboard was mounted. The learned trial judge found that they, too, had resulted from arcing.

Both the front and back holes were low down in the switchboard. Smoke and heat damage to the escutcheon suggested that the fire started near the hole in the escutcheon and spread towards the holes in the back of the box. Photographs were tendered which show that after the fire the switchboard had been reduced to a tangle of uninsulated wires, the insulation having been burnt off by the fire. Some of the wires were still connected to circuit breakers or other components in the switchboard.

Investigations after the fire did not reveal that any of the electrical installations in the shed, more particularly the heaters, the pumps, the lights or other appliances, had malfunctioned in any way. The learned trial judge found that more likely than not, when the fire broke out, pumps and some of the heaters were operating.

One of the expert witnesses, Professor Hobson, a highly qualified electrical engineer, offered the opinion that when the contact points in a circuit breaker while under load are tripped or opened, a spark or arc is generated, with the result that there is a deposit of carbon inside and outside the circuit breaker. That can eventually lead to the development of electrical shorting across the carbon deposit, which creates something of an electrical path or track.

A stage may eventually be reached where a track is close enough to an earthed component, such as the front of the escutcheon plate, that arcing could then result in a hole being burnt into the nearby component. Professor Hobson believed that some such sequence of events could explain the holes burnt in the escutcheon plate and in the box.

This theory led to close consideration of the circumstances of the so-called "flash/bang" incident described by Mr White as having occurred on the visit to the premises by the respondent on 28 April 1998. The theory gained an extra dimension from Mr White's evidence that the circuit breaker for tank 5, which had caused the problem on 28 April 1998, was in the bottom row of circuit breakers adjacent to the hole burnt in the escutcheon.

The suggested proximity of the circuit breaker linked with No 5 tank was challenged by the respondent at the trial. During the course of the hearing he visited the premises and was able to see the plastic cover to the junction box on tank 5 which housed the electrical connections to that tank. The junction box, which was apparently in the same condition as was the case before the fire, still carried a small plastic tag which indicated that it was connected to a circuit breaker numbered 3.

In the switchboard there were three rows of circuit breakers, and No 3 was in the top row, some distance away from the hole burnt into the escutcheon. It followed that the circuit breaker connected with the heaters attached to tank 5 was unlikely to have been responsible for the arcing which caused the hole in the escutcheon. The learned trial judge concluded that he was not satisfied that tank 5 had anything to do with the fire.

It followed that the theory advanced by Professor Hobson that the known activation of the circuit breaker connected to the heaters to tank 5 on 28 April 1998 had given rise to arcing and the development of tracking which ultimately caused the fire was no longer tenable.

On the hearing of the appeal, Mr Trim QC convincingly argued that there was no reason for this Court to interfere with that finding. I agree.

It remains then to consider what possible causes there might have been for the fire and what conclusion that process leads to with respect to the issue of liability.

The trial judge had the assistance of three expert witnesses. The appellant called Mr Kutek and Mr Watson. The respondent called Professor Hobson.

Mr Watson's evidence tabulated the possible causes of failure of the switchboard assembly giving rise to arcing as follows:

"Failure in service due to one or more of the following:

Of those causes, the learned trial judge held on the evidence of both the appellant and that called by the defendant that switchboard "A" "was clean and free from flammable rubbish, moisture and rodents". He went on:

"I cannot exclude the possibility of what counsel called 'a single kamikaze rat' - but it is most unlikely."

Mr Watson himself excluded ageing of the insulation, given that the materials used were new. He also excluded corrosion, contact wear and lack of ventilation as likely causes. He thought that component fatigue or breakage was unlikely but could not be ruled out. Failure due to a loose connection was possible, but he had no evidence from which he could conclude that that was the cause.

Mr Kutek, when asked in cross-examination to consider the various possible causes suggested by Mr Watson, had difficulty in separating "pollution" from "overheating", in that pollution itself would not cause a fire but it might cause overheating. Subject to that reservation, he agreed "... it is not really possible to point to any of these causes being more likely than the others".

Mr Watson concluded that he simply did not know what caused the fire.

Professor Hobson expressed the following view:

"Following consideration of the new evidence from Mr Solley [explaining the location of the circuit breaker connected to tank 5] and Mr Watson's report it is my view that the cause of the fire can only be a matter for conjecture. Mr Watson has identified a number of possible causes with which I concur. Circuit breaker failure remains a distinct possibility, but of the various other main causes of fires in electrical installations, no higher degree can be attached to any one."

On the hearing of the appeal, Mr Beazley for the appellant strongly contended that of the various possibilities, having regard to the expert evidence, the possibility that the fire was caused by a loose connection in the switchboard should have been accepted by the learned trial judge as the most likely cause.

But in my view, having reviewed the expert evidence, this is a case where, of the various possible causes, it is simply not possible to identify any one as being more likely than another.

A loose connection would have given rise to what was described in evidence as a "non-Ohmic joint". During the course of his reasons, the learned trial judge commented:

"I am left with the possibility of a non Ohmic joint consequent on failure to tighten a terminal properly. I am left with the possibility of a manufacturing defect in a component, presumably one of the circuit breakers. Could tracking, leading to arcing have developed? Is there some other explanation altogether?

I am unable to say whether the remote chance of trouble from a non Ohmic joint is greater than that of some manufacturing defect in a component, or greater than the possibility of some other cause to which the evidence has not adverted. Mr Watson, with many advantages over me, could not say how the fire started. Neither could Professor Hobson, or Mr Kutek. Neither can I.

To the extent that the res ipsa loquitur doctrine is different from the course of logic I have already followed, I note that I am not able to be sure (on balance) what went wrong. It is possible that the defendant is responsible. He may have left a terminal loose, he may have dislodged a wire in the course of work on the switchboard. He may have done something else. I accept that it is very unlikely that anyone other than he had anything to do with installing or connecting up the switchboard. If anyone did something wrong in connection with it, it was probably him.

On the evidence, it is not possible for me to deduce, as a matter of common knowledge, that the only probable cause of the fire was some wiring defect, and that any such defect could only have occurred as a result of negligence on the part of the defendant. I cannot exclude component failure of some kind. I cannot exclude the possibility that Professor Hobson's explanation about tracking and arcing is correct - and, if so, that that involved no negligence on the part of the defendant. I cannot exclude the possibility of there being some other cause within the switchboard not adverted to in evidence and not now ascertainable because of the destruction of the switchboard.

Given the failure of the experts to come up with the solution, at least with any reasonable degree of certainty, I cannot do so as a matter of lay, common sense.

It must follow that the plaintiff has not established any breach of the duty owed by the defendant."

I deal separately with that part of His Honour's reasoning which deals with the doctrine res ipsa loquitur.

Apart from that, it appears to me that the learned trial judge correctly concluded that none of the experts were able to identify any possible cause as being more probable than another. Given that at least one possible cause (component failure) would not be associated with negligence on the part of the defendant, the case in negligence was not made out.

At the trial reliance was placed, at least in part, upon this doctrine. On the appeal, Mr Beazley specifically eschewed any such reliance. He did so in recognition of the decision of the High Court in Schellenberg v Tunnel Holdings Pty Ltd.[1]

In that case, in their joint judgment, Gleeson CJ and McHugh J observed:[2]

"Piening v Wanless[3] and Anchor Products Ltd v Hedges[4] as well as other cases in this court make it clear that a plaintiff may rely on res ipsa loquitur event though he or she has also pleaded particular acts or omissions of negligence on the part of the defendant provided that the tribunal of fact concludes that:[5]

1. there is an 'absence of explanation' of the occurrence that caused the injury;

2. the occurrence was of such a kind that it does not ordinarily occur without negligence; and

3. the instrument or agency that caused the injury was under the control of the defendant."

Here it might be argued that requirement 3 was not satisfied in that the switchboard was not under "the control" of the respondent. Against that if, as was established on the evidence, he had installed the switchboard and was the only person to have access to it since then, that element might arguably be taken to have been satisfied.

However, the problem for the appellant is that as to the second of the elements referred to, an occurrence will not be regarded as of a kind that "does not ordinarily occur without negligence" unless it can be said to be so "within the common knowledge and experience of mankind". As to that aspect of the matter, in Schellenberg, Gleeson CJ and McHugh J, after referring with approval to dicta of Barwick CJ in Piening v Wanless[6] and of Dixon CJ in Frankland v Victorian Railways Commissioners,[7] went on to observe:[8]

"In this case, where the occurrence is outside the experience of the lay person, and the evidence, expert or otherwise, does not establish that such an occurrence ordinarily does not occur without negligence, res ipsa loquitur is inapplicable."

Here, quite properly, Mr Beazley conceded that the workings of switchboard "A" were not such as would endow "a lay person" with sufficient understanding to infer from the occurrence of the fire that it would ordinarily not occur without negligence.

So that for the purposes of this case, the principle of res ipsa loquitur may be put on one side.

That does not mean to say that inferential reasoning has no part to play: see Schellenberg per Gleeson CJ and McHugh J:[9]

"... The court has affirmed time and again that res ipsa loquitur is merely a mode of inferential reasoning and it not a rule of law. ...... The fact that a plaintiff falls outside the 'proper scope' of the rule does not mean that he or she may not avail himself or herself of inferential reasoning. There is therefore no need to subsume the maxim into the general body of tort law: it is already fully consonant with it."

However, in this case, the process of inferential reasoning pursued outside of the potential application of the maxim res ipsa loquitur does not advance the matter any further. The evidence still fell short of identifying a cause attributable to fault on the part of the respondent which was more likely than other possible causes, at least one of which would not have involved fault on his part.

I would dismiss the appeal insofar as it challenges the learned trial judge's conclusion that the appellant had not made out its case on liability.

In those circumstances, it is unnecessary to deal with the appeal and cross-appeal on the question of damages.

82. WILLIAMS J. I agree with the orders proposed by Perry J for the reasons which he has given.

83. MARTIN J. I agree with the orders proposed by Perry J for the reasons which he has given.

[1] [2000] HCA 18; (2000) 170 ALR 594.

[3] [1968] HCA 7; (1968) 117 CLR 498.

[4] [1966] HCA 70; (1966) 115 CLR 493.

[5] See Balkin and Davis, Law of Torts (2nd ed 1996) pp 287-96; and Fleming, The Law of Torts (9th ed 1998) 353-9.

[6] [1968] HCA 7; (1968) 117 CLR 498 at 508.

[7] [1959] HCA 48; (1959) 101 CLR 197 at 204.