Viasystems (Tyneside) Ltd. v Thermal Transfer (Northern) Ltd & Ors

On 28^th July 1998, there was a flood at the claimants’ factory at Eldon Street, South Shields. The flood was caused by the foolish negligence of a young fitter’s mate, Darren Strang. The flood caused extensive and expensive damage, the cost of which it is now agreed the claimants are entitled to recover in contract from the first defendants, Thermal Transfer (Northern) Limited. The claimants also claimed against the second defendants, S. & P. Darwell Limited, and the third defendants, Troy Hall and Christopher Day, trading as CAT Metalwork Services. The first defendants likewise claimed an indemnity against the second or third defendants.

The question which His Honour Judge Walton heard and determined in the Newcastle-upon-Tyne County Court on 1^st December 2004 was whether it was the second defendants or the third defendants who were vicariously liable to the claimants for Darren Strang’s foolishness. It was not then suggested that they might both be vicariously liable. The judge decided on the facts which he found that the third defendants were so liable. This is the third defendants’ appeal, with leave of Chadwick LJ, against the judge’s decision.

In July 1998, the claimants engaged the first defendants to install air conditioning in their factory. The first defendants subcontracted ducting work to the second defendants. The second defendants contracted with the third defendants to provide fitters and fitters’ mates on a labour only basis. One such fitter was Mr Megson. His mate was Darren Strang. They were installing the ductwork under the instruction or supervision of Mr Horsley, a self-employed fitter contracted to the second defendants. Both Mr Megson and Darren Strang were thus employed by the third defendants.

At the time of the accident, the men were working in a roof space. Access was by crawling boards using the roof purlins. Mr Megson needed some fittings and sent Darren Strang to get them. Darren was away for a few minutes, during which Mr Horsley was helping Mr Megson with the ducting. Mr Horsley naturally expected Darren to return by a sensible route, but he did not so return. On the contrary, he attempted to return by crawling through some sections of ducting that were in place. These moved and came into contact with part of a fire protection sprinkler system. The relevant part of this system fractured – hence the flood. The judge had no difficulty in finding that Darren was negligent, as he obviously was.

It was the third defendants’ case before the judge that Darren Strang did what he did on the express instruction of Mr Horsley. The judge rejected this and there is no appeal against this finding. Indeed the third defendants now accept the judge’s primary findings of fact.

The leading relevant authority is Mersey Docks and Harbour Board v Coggins & Griffith (Liverpool) Limited [1947] AC 1, which the judge considered with other authorities. This was a case in which someone was injured by a negligently driven crane. The crane had been let by the Harbour Board to a firm of stevedores for loading a ship, together with the crane driver who was employed by the Harbour Board. The stevedores had immediate control of the relevant operation which the crane was performing, but had no power to direct how the crane driver should control the crane. The House of Lords upheld decisions of lower courts that the Harbour Board, being the crane driver’s general employer, retained responsibility for his negligence.

The opinions make clear that decisions of this kind depend on the particular facts and that many factors may bear on the result (Lord Porter at 17). In assessing the facts, certain considerations will or may be relevant. These include:

The factors relied on by the second defendants as indicating vicarious responsibility by the third defendants are:

The factors relied on by the third defendants are:

The judge did not read the Mersey Docks case as saying that consent and control were determinative preconditions, although they were very significant. The contractual arrangements did not suggest a transfer of Darren Strang’s employment. Mr Hall was able to substitute employees without reference to the second defendants. It would be strange if employees were bouncing in and out of employment. He then said at page 9 of the transcript:

“While Mr Horsley did instruct Mr Strang to perform tasks from time to time, Strang was under the immediate control of his fitter and there is no instance in evidence of Mr Horsley giving an instruction as to the manner the fitter or the fitter’s mate should work. Yes, he told them what required doing, but he himself worked at another location and left Mr Megson and Mr Strang to get on with the task of fitting the ducting sections together. When he went over to see them it was to see how they were doing and to lend a hand if necessary.”

The judge held that Mr Horsley was not in entire and absolute control of Darren Strang. He concluded that Darren’s employment was not transferred. He also accepted that Mr Horsley was not himself negligent. Of this he said at page 10:

“It was not put to him that he was careless in failing to supervise Strang to an appropriate standard, and had that been done I have no doubt he would have repeated what, in effect, he had already said, that Mr Strang had been sent on a straightforward task, that he had not shown any inclination to take a route through the ductwork and that he, Horsley, had no reason to anticipate he would do that in circumstances where it was not really necessary.”

The second defendant’s initial case in this court was that the judge properly weighed the evidence and came to a decision of fact with which this court should not interfere. The third defendants contended that the judge gave too little weight to the factors on which they relied, and too much weight to the question whether, in formal terms, Darren Strang’s employment was transferred. Although Darren considered that he was still employed by the third defendants, his other evidence was that the second defendants were in charge of the particular job. The critical question was rather whether there had been a transfer of control. The question was who was entitled to give orders as to the way in which the relevant work should be done. The third defendant’s ability in theory to replace one workman with another was of limited significance.

The parties’ cases and the judge’s decision were predicated on an assumption that only one of the second or third defendants could in law be vicariously liable for Darren Strang’s negligence, not both. Before we began hearing the appeal, counsel drew our attention to the discussion on page 156ff of Atiyah Vicarious Liability in the Law of Torts (1967), in a chapter entitled “The Borrowed Servant”, where it is suggested that it is strange that the courts have never countenanced what might be an obvious solution in some cases, namely holding both the general and the temporary employer vicariously liable for an employee’s negligence. We considered that this interesting possibility should be examined in this appeal. If it were permissible in law to hold both the second and third defendants vicariously liable, the facts of this appeal might be a paradigm example of a case in which the court should do so. So we adjourned the hearing of the appeal to enable counsel to research and make further submissions. They have done so, and we are most grateful.

In subsequent submissions, Mr Prynne submitted that dual vicarious liability was not a legal possibility. He maintained that the second defendants alone should be liable. Mr Field QC maintained that on the facts the third defendants alone should be liable, but that, if that was not the court’s view, dual vicarious liability was a legal possibility and should be found in this case.

It is first necessary to consider authority in addition to the Mersey Docks case.

In Denham v Midland Employers’ Mutual Assurance Limited [1955] 2 QB 437, the issue was which of two mutually exclusive liability insurance policies covered damages which an employer was liable to pay to the widow of an employee, who was killed while he was working under the specific direction of engineers engaged by the employer to do work on their land. Denning LJ said at page 443 that much of the difficulty arose out of the 19^th century conception that a servant of a general employer may be transferred to a temporary employer so as to become for the time being the servant of the temporary employer. The conception was a very useful device to put liability on the shoulders of one who should properly bear it, but it did not affect the contract of service itself. No contract of service could be transferred without the servant’s consent. In none of the transfer cases cited to the court had the consent of the man been sought or obtained. The general employer had simply told the employee to go and do some particular work for the temporary employer and he had gone. The supposed transfer was nothing more than a device. Lord Denning referred to the Mersey Docks case. He said that such a transfer rarely takes place when a man is lent with a machine. But a transfer does sometimes take place when a man is lent to help with labouring work.

“The temporary employer can then no doubt tell the labourer how he is to do the job. The labourer becomes so much part of the organisation to which he is seconded that the temporary employer is responsible for him and to him.” (page 444)

Lord Denning then applied the principles to the facts and said:

“These results are achieved in law by holding that Clegg became for the time being the temporary servant for Le Grands. There is no harm in thus describing him so long as it is remembered that it is a device designed to cast liability on the temporary employer. The real basis of the liability is, however, simply this: if a temporary employer has the right to control the manner in which a labourer does his work, so as to be able to tell him the right way or the wrong way to do it, then he should be responsible when he does it in the wrong way as well as the right way. The right of control carries with it the burden of responsibility.”

In my view, Denham, applying the principles in the Mersey Docks case, relevantly states and illustrates those elements of principles most relevant to the present appeal. To look for a transfer of a contract of employment is, in a case such as this, no more than a distracting device; in the present case a misleading one. Darren Strang’s employment was not transferred. The inquiry should concentrate on the relevant negligent act and then ask whose responsibility it was to prevent it. Who was entitled, and perhaps theoretically obliged, to give orders as to how the work should or should not be done? In my view, “entire and absolute control” is not, at least since the Mersey Docks case, a necessary precondition of vicarious liability.

In the present appeal, Mr Prynne QC, for the third defendants, correctly formulated the question to determine vicarious liability, substantially as I have outlined it, as who was entitled to exercise control over the relevant act or operation of Darren Strang. He submitted, again I think correctly, that consent to any transfer was not a determinative factor in this case. He suggested that the judge looked at what Mr Horsley did, rather than what he was entitled to do.

The relevant negligent act was Darren Strang crawling through the duct. This was a foolish mistake on the spur of the moment. I have said that a central question is: who was entitled, and perhaps in theory obliged, to give orders as to how the work should or should not be done? Here there is no suggestion, on the facts found by the judge, that either Mr Horsley or Mr Megson had any real opportunity to prevent Darren’s momentary foolishness. The judge specifically acquitted Mr Horsley of personal negligence: and we should proceed on the footing that Mr Megson was not personally negligent either. Vicarious liability is liability imposed by a policy of the law upon a party who is not personally at fault. So the core question on the facts of this case is who was entitled, and in theory, if they had had the opportunity, obliged, so to control Darren as to stop him crawling through the duct. In my judgment, the only sensible answer to that question in this case is that both Mr Megson and Mr Horsley were entitled, and in theory obliged, to stop Darren’s foolishness. Mr Megson was the fitter in charge of Darren. Mr Horsley was the foreman on the spot. They were both entitled and obliged to control Darren’s work, including the act which was his negligence. The second defendants, through Mr Horsley, would, I think, have qualified for vicarious liability, if it had been Mr Megson who foolishly crawled through the duct. It makes no difference to a sensible analysis that it was Darren who was negligent, and that Mr Megson in some respects was interposed. But neither is there any good sense in saying that, because Mr Horsley was relevantly entitled to control Darren, Mr Megson was not: and vice versa.

All this presupposes that it is possible in law in these circumstances for both the second and third defendants to be vicariously liable. If, contrary to the view I express below, this is not permissible in law, so that a judicial choice has to be made between the second and third defendants, I would not disturb the judge’s finding. I would accept Mr Field’s submission that it was a finding of fact which the judge was entitled to make. But, on the facts of this case, it would be an artificial choice required by an inflexible rule of law.

As Professor Atiyah pointed out, it has been assumed since the early 19^th century to be the law that, where an employee, who is lent by one employer to work for another, is negligent, liability must rest on one employer or the other, but not both. But the foundation on which this rests is a slender one and the contrary has never been properly argued. I agree that the House of Lords implicitly proceeded on this assumption in the Mersey Docks case; but the point was not argued, and I do not consider that the facts would have sustained a finding that both employers were vicariously liable.

The pedigree of the assumption goes back to 1826. Laugher v Pointer (1826) 5 B. & C. 547 was an appeal heard by 11 judges of the court of King’s Bench. The owner of a carriage hired a pair of horses for a day to draw the carriage. The owner of the horses also provided the driver, by whose negligence a horse belonging to a third party was injured. It appears that the majority of the court held that the owner of the carriage was not liable for the injury. Abbott CJ and Littledale J gave judgments with which an unrecorded majority of the court must have agreed. Bayley and Holroyd JJ gave dissenting judgments. Littledale J said at page 558 that the coachman or postillion cannot be the servant of both the owner of the horses and the traveller.

“He is the servant of one or the other, but not the servant of one and the other; the law does not recognise a several liability in two principals who are unconnected.”

One reason for this was that to allow two principals to be severally liable would tend to a multiplicity of actions, an objection which modern procedure does not find unduly troublesome. Professor Atiyah points out that at this time the courts were generally reluctant ever to admit that two parties might both be liable for the same damage in tort, unless they were joint participants in the tort. Things have moved on since then. It was accepted before us that technically these remarks of Littledale J were dicta, not part of the ratio, and the point does not feature in any other recorded judgment in the case.

Jones v Scullard [1898] 2 QB 565, a considered decision of Lord Russell of Killowen CJ, is perhaps representative of cases in which it is implicitly assumed that only one employer can be vicariously liable. It was a carriage case in which Lord Russell decided that a borrowed driver was acting as the servant of the defendant owner of the carriage and horses so as to make the defendant liable for the driver’s negligence. The decision turned on the particular facts. The judgment considered Laugher v Pointer; Quarman v Burnett (1840) 6 M. & W. 499; Brady v Giles (1835) 1 MOO & R 494; Rourke v White Moss Colliery 2 C.P.D. 205; and Donovan v Laing [1893] 1 QB 629. The reference to Laugher v Pointer is to the main ground of its decision. There is no reference to Littledale J’s dicta. Murphy v Caralli (1864) 3 H. & C. 462, referred to as another example by Professor Atiyah, is in the present context unremarkable. Likewise Sykes v Millington [1953] 1 All ER 1098, which concerned a prosecution for an offence under section 2(3) of the Road and Rail Traffic Act 1933.

Donovan v Laing. was another case where the plaintiff was injured by the negligence of a crane driver. The defendants had contracted to lend the crane with its driver to a firm who were loading a ship. It was held that, although the crane driver remained the general employee of the defendants, they had parted with the power of controlling him with regard to the matter on which he was engaged and were not liable for his negligence while he was so employed. Lord Esher MR said at page 632:

“For some purposes, no doubt, the man was the servant of the defendants. Probably, if he had let the crane get out of order by his neglect, and in consequence anyone was injured thereby, the defendants might be liable; but the accident in this case did not happen from that cause, but from the manner of working the crane. The man was bound to work the crane according to the orders and under the entire and absolute control of Jones & Co.”

Bowen LJ said at page 634 that the court only had to consider in whose employment the man was at the time the acts complained of were done, in the sense that by the employer is meant the person who has a right at the moment to control the doing of the act. The question was whether the defendants had parted with the power of controlling the operation on which the man was engaged.

Although Bowen LJ’s formulation and application of the test in Donovan v Laing was approved and applied in the Privy Council in Bain v Central Vermont Railway Company [1921] 2 AC 412 and Bull v West African Shipping Agency [1927] AC 686, it was subjected to critical consideration in the Mersey Docks case, from which it scarcely survived (see Lord Simon at page 11, Lord MacMillan at page 14, Lord Porter at page 16, and Lord Simons at page 18).

There is an oddity about Donovan v Laing. Counsel’s diligent researches have found 7 reported versions of this Court of Appeal decision in addition to the report in the Official Law Reports ([1893] 1 QB 629). These are in the Solicitors’ Journal 37 SOL JO 324; the Weekly Reporter (41 WR 455); The Law Times (68 LT 512); Justice of the Peace (57 JP 583); Times Law Reports (9 TLR 313); Law Journal (63 LJ QB 25); and Reports (4 R 317). A version reproduced in the All England Reprint 1558-1935, a series collated in the 1960s by Butterworths using as a source for late 19^th century judgments reports in the Law Times, is representative of the versions in the Law Times, the Weekly Reporter and the Reports as having passages which do not appear in the Official Law Reports’ version.

Reference to The Law Times version at page 513 shows that counsel for the appellant plaintiff submitted that, if the employee was not the servant of the defendants alone, he was the servant of Jones & Co and the defendants jointly. Lord Esher MR is reported at page 514 as saying:

“It has also been argued that Wand was the servant both of Jones and Co. and the defendants. In one sense, he was; he was the general servant of the defendants, and at a particular moment he was the servant of Jones and Co.; he was the servant of both at the same moment, but not with regard to the same acts of service. As to the mode of working the crane he was under the sole control of Jones and Co., and he was therefore not the servant of the defendants as regards that working of the crane, whatever he might be as regards other matters. The defendants are clearly not liable for the negligence of Wand, both on principle and on authority, and I think this appeal should be dismissed.”

Lindley LJ is reported as saying at the end of his judgment:

“As to the alleged joint responsibility of the defendants and Jones and Co., I think the answer may be found in the words of Littledale J in Laugher v Pointer (5 B & C 558).”

Lindley LJ then cites the passage to which I have referred earlier in this judgment.

There are further oddities in that (a) the version in the Official Reports is a heavily edited version which does not correspond sentence by sentence with The Law Times version; and (b) Bowen LJ said at page 634 of the Official Reports that the argument for the plaintiff was founded on what might be called the carriage cases, identified as Laugher v Pointer and Quarman v Burnett, “but they really have nothing to do with the point presented in this appeal”.

I have no doubt but that we should in this instance look at and consider the unofficial reports – see Practice Direction (Court of Appeal: Citation of Authority) [1995] 1 WLR 1096. I am content also to take it from the unofficial reports that the Court of Appeal’s oral judgments in substance contained the passages relied on by Mr Prynne in his muted submission that, by virtue of Donovan v Laing, Littledale J’s dicta are binding on this court. I reject that submission because:

Academic text book writers have acknowledged the prevailing assumption derived from Littledale J’s dicta, but recognise that cases where there is in fact dual control could sensibly result in both employers being vicariously liable – see Salmon & Heuston on the Law of Torts, 21^st edition at page 440; Fleming, the Law of Torts, 9^th edition page 418. Cardozo, writing in 1921 in the Harvard Law Review, volume 35 page 113 at page 121, said that the law that defines or seeks to define the distinction between general and special employers is beset with distinction so delicate that chaos is the consequence. No lawyer can say with assurance in any given situation when one employment ends and the other begins.

Clerk & Lindsell on Torts, 18^th edition paragraph 5-21, states that, in some cases, the solution that both employers are liable might be preferable to the solution that it must be one or the other, but that it is doubtful whether such a solution is open to the courts except where both employers are engaged in a joint venture. The editors cite the passage in Atiyah to which I have referred; Karuppan Bhoomides v Port of Singapore Authority [1978] 1 WLR 189; McKee v Dumas (1976) 70 D.L.R. (3d) 517; and Esso Petroleum v Hall Russell [1988] 3 WLR 730 at 758.

Karuppan Bhoomides was a decision of the Privy Council on appeal from the Court of Appeal in Singapore. A member of a gang of stevedores was killed in an accident caused by the negligence of a fellow workman while the gang was loading a ship. The gang was employed by the Port Authority. The plaintiff, who was the administrator of the deceased’s estate, brought an action against the Port Authority for damages. The only issue was whether the Port Authority was exempted from liability under the terms of a by-law which provided that “serangs and labourers employed in discharging and loading vessels shall be under the superintendence of the ship’s officers; the board undertake no responsibility as stevedores”. The Privy Council, reversing the decisions of lower courts, held that the by-law did not exclude or transfer to the shipowners any part of the Port Authority’s vicarious liability at common law as employer for the negligent acts of its employees.

Lord Salmon, giving the judgment of the Board, first considered what the legal position would have been at common law had the by-law never existed. An orthodox application of the Mersey Docks case to the facts would entirely fail to shift the responsibility of the Port Authority as general employer for the negligence of their employee. It was not argued otherwise. As to the by-law, it “falls far short of putting the servants of the port authority under the entire and absolute control of the ship” (page 194A). Here again it is implicit that one or other must be liable, but not both. The possibility of dual vicarious liability was not argued or considered; nor would it have been a realistic outcome on the facts.

In Esso Petroleum v Hall Russell, a tanker was being berthed at a jetty at Sullom Voe oil terminal in the Shetland Islands when a coupling blew out of a hydraulic pipe above the starboard exhaust of one of the three tugs attending her. The tanker was then not under full control and various damage was caused. At the time of the accident, a pilot in the general employment of the harbour and pilotage authority was compulsorily on board the tanker. One issue was whether the authority was vicariously responsible for acts or omissions of the pilot. The House of Lords held that a pilot was an independent professional man who navigated a ship under pilotage as a principal, not as a servant of his general employer and, moreover, section 15(1) of the Pilotage Act 1913 made a compulsory pilot the shipowners servant for all purposes connected with navigation. Accordingly, the employer of a qualified licensed pilot was not as a general rule vicariously responsible to the owner of a ship under pilotage damaged by his negligence.

Lord Jauncey of Tullichettle gave the leading opinion with which the other members of the House agreed. So far as is material to the present appeal, Lord Jauncey said at page 758E:

“My Lords, nothing that has been said on behalf of Hall Russell persuades me that the rationale of the line of authority to which I have referred was wrong or that there is any exception to the general application of section 15 of the Act of 1913 to damage suffered by a ship under pilotage. Subject only to what I have to say in the context of Hall Russell’s second submission, the pilot is to be considered for all purposes the servant of the owner. I would only add that if Hall Russell’s argument were correct there would follow the curious result that the doctrine of respondent superior would apply to two different masters in respect of two different claims of damage arising out of a single act of negligence. It is a well recognised principle, exemplified in cases involving crane drivers, that a servant in the general employment of A may, for a particular purpose, be treated as in the pro hac vice employment of B. However, there is no principle which permits a servant to be in the de jure employment of two separate masters at one and the same time.”

This passage, at first blush, might be read as an authoritative restatement of Littledale J’s dicta. But it has to be read in the context of the argument which it was addressing. That is to be found at page 755H as follows:

“Hall Russell advanced two reasons for their contention that they had relevantly averred vicarious responsibility on the part of [the pilotage authority] for the negligence of the pilot. In the first place it was said that although earlier authorities had appeared to impose liability on a shipowner for the negligence of a voluntary pilot on the basis of a master and servant relationship this rationale was no longer correct. The true position was that the shipowner owned a non-delegable duty to third parties to have his ship navigated carefully and that he could not discharge that duty by delegating it to an independent contractor in the shape of a pilot. When section 15(1) was looked at against that background there was nothing in it which altered the common law position obtaining between the ship-owner and pilot which remained that of employer and independent contractor. It followed that quoad the ship-owner the pilot remained the servant of [the pilotage authority].”

Thus it was argued that the shipowner would be liable to third parties for the pilot’s negligence, but that the pilot’s general employer could still be liable to the shipowner for the same negligence of the same pilot. The pilot would be the employee of the shipowner in the first instance, but the employee of the general employer in the second. That is not the present case, which is concerned with a single original liability, not two. Nor is it concerned exclusively with Darren Strang’s de jure employment, rather than who had the right to control, and theoretically prevent, his relevant negligent act.

There is one first instance English authority, Mileham v St Marylebone Borough Council and Latter (1903) 67 JP 110, in which each of two defendants was held vicariously liable for the negligence of the employee of one of them. The plaintiff was injured and his cab damaged when the cab collided with an upright key left standing in a water stop-cock in the roadway. The key had been so left by the driver of one of the council’s water carts. The driver was the employee of the defendant, Latter, who had contracted with the council to supply them with horses and drivers for their water carts. The council’s surveyor had general control of the drivers who, under the terms of the contract, acted under the direction of the council’s surveyor or superintendent. The jury found that the driver was negligent; that he was the servant of Latter; but that at the time of the accident he was acting under the control of the council.

The plaintiff argued that each of the defendants should be held liable. Each of the defendants argued that the other should be held exclusively liable. Channell J gave judgement against both defendants “… on the authority of Jones v The Mayor of Liverpool [(1985) L.R. 14 QBD 890], and on my construction of the agreement. I think that the defendant Latter remains liable as master of [the driver], and that the Borough are liable on the ground that they authorised the use of the stop-cock, and they could only do so with the liability to have the public protected.”

The report of Mileham is brief and I am unclear how Channell J derived his decision of dual vicarious liability from Jones v The Mayor of Liverpool. In that case, the defendant Corporation hired by the day from Mrs Dean a driver and horse to draw a water cart belonging to them. The plaintiff claimed damages for injuries caused by the driver’s negligence while he was in charge of the water cart. The driver was paid by Mrs Dean and was not under the defendants’ direction or control otherwise than because their inspector directed him which streets to water. The Queen’s Bench Divisional Court, reversing the first instance decision, held that the defendants were not liable. The court applied Laugher v Pointer – but without reference to Littledale J’s dicta – and Quarman v Burnett. The essence of the decision in Jones appears to have been that the defendants’ inspector had no relevant control over the driver. As Grove J said at 593:

“If [the inspector] had interfered when the accident happened by directing the driver of the cart what to do, the case would be different, but he did no more than point out the streets to be watered.”

In Mileham, the council had argued that Jones was a case in which the contractor only was held liable. It looks as if Channell J regarded the terms of the contract between the council and Latter as giving the council’s surveyor a degree of control over the driver sufficient to make the council vicariously liable without thereby exonerating Latter. But the reported reasoning is not transparent. I regard Mileham as interesting, but of little authoritative weight.

I conclude that technically there is no authority binding this court to adopt Littledale J’s dicta. There is a general persuasive point that the assumption has not been questioned judicially for 180 years. In a modern context, it is not obvious why. So far as counsels’ enquiries have revealed the point has never been argued in this jurisdiction in modern times, other than as an unreported coda to Donovan v Laing and in Mileham. I do not consider that the Mersey Docks case precludes this court from considering the point. It was not an issue in that case.

My view that the Mersey Docks case does not preclude us from considering the point has some support from Brennan J in the High Court of Australia in Oceanic Crest Shipping Company v Pilbara Harbour Services Pty Limited [1986] HCA 34; (1986) 160 CLR 626, who said at paragraph 32 of his judgment:

“The rule to be derived from Mersey Docks and [McDonald v The Commonwealth [1945] 46 SR(NSW) 129] is not that two persons cannot be vicariously liable for the same damage or that an employee cannot be the servant of two masters, but that two employers of the same servant who negligently causes damage will not both be liable for the damage if one rather than the other has what Jordan C.J. called “the relevant control”.”

See also Deutz Australia Pty Limited v Skilled Engineering Limited [2001] VSC 194, a decision of Ashby J in the Supreme Court of Victoria, at paragraph 105. Ashby J observed at paragraph 106 that Brennan J had dissented in Oceanic Crest in the result. But he did not consider that that in any way diminished his analysis of the ratio of the cases to which he referred.

A leading authority in the United States Supreme Court is Standard Oil v Anderson (1909) 212 US 215. The reasoning and statements of principle approximate to those later adopted by the House of Lords in the Mersey Docks case. We were referred to Gaudet v Exxon Corporation (1977) 562 F. 2d 351 as exemplifying the current approach of the US Federal Courts of Appeal. We understand that a number of States follow the US Supreme Court, but that others do not. Examples, referred to by Atiyah, are Gordon v Byers (1932) 309 Pa. 453 and Siidekum v Animal Rescue League of Pittsburg (1946) 353 Pa. 408, both decisions of the Supreme Court of Pennsylvania. We were also referred to Morgan v A.B.C. Manufacturer (1997) 710 Fo. 2d 1077, a decision of the Supreme Court of Louisiana.

In each of the Pennsylvania cases, both the lending and borrowing employers were held to have a requisite control for each to be liable for the employee’s negligence. In the Siikekum case, Stern J said that it was well recognised that under some circumstances both the lender and the borrower may have the right of control over the employee so as to render each of them liable for his conduct. There was also identified an element of common enterprise or common purpose. Professor Atiyah says that the Siikekum decision seems to be eminently sensible, but that it was doubtful if such a decision would be open to an English court on similar facts. However that might be, the judgment articulates a sensible principle which recognises the possibility in Pennsylvania of dual control. I do not, however, consider that common enterprise or common purpose is a workable guide, nor that it accords with the principles in the Mersey Docks case.

In Morgan v ABC Manufacturer, a case involving negligence of a borrowed crane driver in a scrap yard, it was said that, since liability is based on the right of control, rather than actual control at the time of the accident, it is unreasonable to choose between two employers when each shares the right to control the employee’s actions.

Treacy v Robinson [1937] IR 255 is a majority decision of the Irish Supreme Court in two appeals which mainly concerned analysis of some complicated facts. The plaintiffs were injured by a car driven by a Mr Beggan, who was an employee of a garage in Dublin. The car had been bought by Robinson & Son for use by one of their employees when he was on business in Ireland. There was a question whether Beggan had been authorised by Robinson’s employee to drive the car. The plaintiffs claim damages against the garage and Robinson, as well as against the driver. It was conceded that the driver was liable. It was also conceded by the plaintiffs that only one of the other defendants could be liable for the driver’s negligence. Fitzgibbon J, who was in the majority, said at page 276:

“It has been settled law since Laugher v Pointer and Reedie v London and North Western Railway Co. [4 Ex. 244 at 257] that law does not recognise a several liability in two principals who are unconnected; if they are jointly liable you may sue either, but you cannot have two separately liable. This doctrine is one of general application, irrespective of the nature of the employment.”

McKee v Dumas (1976) 70 DLR (3d) 517 is a first instance decision of Holland J in the Ontario High Court. There was a collision between a train and a tractor pulling two trailers owned by Mr Daoust. The tractor driver was a Mr Dumas. An engineer was killed and Mr McKee, an employee of the train company, was injured. The judge held that the accident was caused solely by the negligence of Dumas. Daoust, as owner of the tractor, was liable for the negligence of Dumas under the provisions of a Canadian statute. It was, however, further alleged that Eddy Forest Products, Dumas’ general employers was vicariously liable for his negligence. The judge decided on the facts that they were not so liable upon an orthodox application of the Mersey Docks principles. He said that Eddy Forest selected Dumas, paid his wages and fringe benefits and had a qualified right to suspend or dismiss him. He then said at page 525:

“The only question over which there is any dispute is the question of control. Control itself is to some extent the justification for imposing liability, for if the master can control the servant as to the manner in which he shall do his work and the servant does it in such a way as to cause damage then it is natural to think that the master should be responsible.

I can find no case in Canada or England which would indicate that both the general employer and the particular employer or hirer can be liable for the negligence of the same servant arising out of a single occurrence. It must be one or the other. In the present case we are dealing with control in the operation of this tractor-trailer on the highway.”

There is thus an explicit reassertion of Littledale J’s dicta, but without reference to them and without explanation.

In summary, therefore, there has been a long-standing assumption, technically unsupported by authority binding this court, that a finding of dual vicarious liability is not legally permissible. An assumption of such antiquity should not lightly be brushed aside, but the contrary has scarcely been argued and never considered in depth. This is not surprising, because in many, perhaps most, factual situations, a proper application of the Mersey Docks principles would not yield dual control, as it so plainly does in the present case. I am sceptical whether any of the cases from this jurisdiction which I have considered would, if they were re-examined, yield dual vicarious liability. Even Mileham is not transparent.

I conclude below in considering contribution that, if the relevant relationships yield dual control, it is highly likely at least that the measure of control will be equal. An equal measure of control will not often arise. Dual vicarious liability is most unlikely to be a possibility if one of the candidates for such liability is also personally at fault. It would be entirely redundant, if both were.

Academic commentary tends to favour the possibility of dual vicarious liability, but feels that authority constrains it. Other jurisdictions have reacted variously, giving no clear lead. Their decisions range from articulating the assumption to favouring or adopting dual liability.

In my judgment, there is, in a modern context, little intrinsic sense in, or justification for, the assumption. Multiplicity of claims is not a modern impediment. A contest between two defendants, where only one could be liable, is just as likely as a claim against the same two defendants, if both could be liable. The underlying basis for the assumption appears to be the notion, exposed as a device by Denning LJ in Denham, that, to find a temporary employer vicariously liable, you have to look for a transfer of employment. Although the nature and incidence of the employee’s employment is plainly material, I do not read Mersey Docks as saying that these are the determinative matters in all cases. If, on the facts of a particular case, the core question is who was entitled, and in theory obliged, to control the employee’s relevant negligent act so as to prevent it, there will be some cases in which the sensible answer would be each of two “employers”. The present is such a case. In my judgment, dual vicarious liability should be a legal possibility, and I would hold that it is. It follows that I would allow this appeal to the extent of holding each of the second and third defendants vicarious liable for Darren Strang’s negligence.

The second defendants had orthodox contribution proceedings against the third defendants, but Mr Field QC confirmed to us that these were not pursued. But since, as I think, the second and third defendants should each be held to be vicariously liable, the question of contribution between them needs to be considered.

By section 1(1) of the Civil Liability (Contribution Act) 1978:

“… any person liable in respect of any damage suffered by another person may recover contribution from any other person liable in respect of the same damage (whether jointly with him or otherwise).”

The relevant vicarious liability is liability to the claimants. If each of the second and third defendants are vicariously liable to the claimants, they are each severally liable in respect of the same damage, so that, in my judgment, contribution is available to each of them under the section. This does not affect the antecedent question whether each may be vicariously liable by virtue of their dual control. But it does afford a fair distribution of the liability between each of them.

Section 2 of the 1978 Act provides that the amount of recoverable contribution “shall be such as may be found by the court to be just and equitable having regard to the extent of that person’s responsibility for the damage in question”. In the present case, I consider that the just and equitable division of contributory responsibility should be equal, so that each of the second and third defendants should, as between each other, contribute 50% of their several liability to the claimants. For dual vicarious liability, equal contribution may, depending on the facts, be close to a logical necessity. This is because vicarious liability is a policy device of the law to redistribute the incidence of loss from a supposedly impecunious employee, who is personally at fault, to one or more supposedly solvent and insured employers, who are not personally at fault. The court is not, therefore, concerned to look for personal responsibility in the employer based on what might otherwise have been direct responsibility – for example, whether Mr Hall trained Darren Strang adequately or not. Vicarious liability derives from the relationship between the employee and the employers, the critical relationship being the employers’ right (and theoretical obligation) to control the relevant activity of the employee. If the relationships yield dual control, it is highly likely at least that the measure of control will be equal, for otherwise the court would be unlikely to find dual control.

For these reasons I would allow the appeal in part and hold that each of the second and third defendants should be vicariously liable to the claimants. As between each other, there should be equal contribution.

I agree. I gratefully adopt Lord Justice May’s account of the facts and of the authorities and can therefore proceed directly to the issue of whether two separate employers can both be vicariously liable for the negligence of a single employee. I agree with Lord Justice May’s conclusions, and with his reasons for thinking that such a result is possible and, on the facts of this case, desirable. I add some words of my own only because we are departing from a long-standing assumption that such dual vicarious liability is not possible.

The concept of vicarious liability does not depend on the employer’s fault but on his role. Liability is imposed by a policy of the law upon an employer, even though he is not personally at fault, on the basis, generally speaking, that those who set in motion and profit from the activities of their employees should compensate those who are injured by such activities even when performed negligently. Liability is extended to the employer on the practical assumption that, inter alia because he can spread the risk through pricing and insurance, he is better organised and able to bear that risk than the employee, even if the latter himself of course remains responsible; and at the same time the employer is encouraged to control that risk. For these purposes, issues have naturally arisen as to when the relationship of employer and employee, as distinct from that of employer and independent contractor, exists; or as to the doctrine of the course of employment, which seeks to set the scope and limits of the employer’s liability. Over the years, the tests which have been adopted to answer these issues have developed in a way which has gradually given precedence to function over form.

One of the issues which has arisen in this context is the case of the lent, borrowed or transferred employee, illustrated on the facts of this appeal. In such a case, which employer is liable, the lending employer, often referred to as the general employer, or the borrowing or temporary employer? Or may there be a dual vicarious liability? As Lord Justice May’s analysis of the authorities has shown, the last possibility has very rarely been argued: in this jurisdiction it has been decisively relied on, and then in favour of dual liability, only once, at first instance, in Mileham v. St Marylebone Borough Council and Latter (1903) 67 JP 110.

It may take a legal historian to explain why the authorities have developed in the way they have. The background appears to arise in the so-called carriage cases, of which Laugher v. Pointer (1826) 5 B&C 547 is an influential example. In such cases, where the borrowing employer is merely the owner of the carriage and as such very possibly not in business at all, as distinct from the lending employer, who appears to be in business as a hirer of drivers, or at any rate as a supplier of drivers as part of the hiring of horses, it is not surprising that it should have been difficult to shift the employment from that of the borrower to the lender. That it could be done on particular facts, however, is illustrated by Jones v. Scullard [1898] 2 QB 565, where the defendant borrower was the owner of the horses as well as the carriage and the driver was in the owner’s livery. It was regarded as a jury question on the facts as to when “the general employer of the driver, ceased to be responsible and the temporary hirer became so” (at 571).

One reason why the law may have developed as it has done is simply the accident of which defendant the claimant has chosen to sue. Thus in Laugher v. Pointer the claimant sued the carriage owner, not the stable keeper, and failed. The argument addressed on behalf of the claimant (see at 551) made no reference to the possibility that the driver was the servant of both or that there was a dual vicarious liability: he sought to show that the driver, whom the carriage owner had paid directly and for whom he had not paid the stable keeper a separate fee, was the carriage owner’s rather than the stable keeper’s servant. Thus when Littledale J opined as he did, negativing the possibility of a dual liability, he was certainly speaking by the way.

Another possible reason for the law’s development is that the matter of control, the right to control the method of doing work, has long been an important and sometime critical test of the master/servant relationship, although the inadequacy of the test as a necessary condition of such a relationship, has also, over time, been demonstrated especially in cases concerning employees who have to exercise their own skill and judgment in their work (see Clark and Lindsell, 18^th ed, 2000, at paras 5-06/11. One can see this test at work in the interesting case of Donovan v. Laing [1893] 1 QB 629. The claimant there again chose to sue only one defendant, on this occasion the lending or general employer, who had hired both employee and crane to a firm of stevedores. It was held that the defendant was not liable, but only because the injury in question arose as a result of the crane-driver’s working of the crane, and that had been put under “the entire and absolute control” of the stevedores. In the critical part of Lord Esher’s judgment (at 22/23), he said this:

“The ordinary mode of using a crane for loading a ship is well known…How far the crane is to be swung, and how much the chain is lowered, depends on what part of the ship the goods are to be placed in, and every act in connection with the working of the crane must be done according to the orders of those who are directing the loading…For some purposes, no doubt, the man was the servant of the defendants. Probably, if he had let the crane get out of order by his neglect, and in consequence anyone was injured thereby, the defendants might be liable; but the accident in this case did not happen from that cause, but from the manner of working the crane. The man was bound to work the crane according to the orders and under the entire and absolute control of Jones & Co.”

Lord Esher therefore inferred that for relevant purposes the crane driver was the servant of the stevedores and “no longer” the servant of the lending employer (at 632). Similarly, Bowen LJ said (at 633/4):

“We have only to consider in whose employment the man was at the time when the acts complained of were done, in this sense, that by the employer is meant the person who has a right at the moment to control the doing of the act.”

These remarks are instructive for present purposes. On the one hand, Lord Esher, with whom Lindley LJ agreed, accepted that a person may be the servant of two masters, and, depending on the circumstances, may cause either employer to incur vicarious liability. On the other hand, the court seems to have considered that employment would switch, at any rate for the doctrine of vicarious liability, if the entire control of the employee’s relevant working switched. It is not clear why that should be so. Even if it be granted that the test of control is a sufficient test to make the borrowing employer liable, it does not follow that it is also a necessary test of employment so as to render the lending employer not liable. However, the court in Donovan regarded the non-liability of the lending employer as concluded by earlier authority in this court in Rourke v. White Moss Colliery Co Ltd 2 CPD 205: see Donovan at 632, 633, and 634.

It is reasonably clear from the Law Times version of the reports of Donovan that it had been argued on behalf of the claimant that the crane driver was the servant of both the stevedores and of the defendant; but, although that submission was in one sense accepted by Lord Esher, it did not suffice to render the general employer liable, on the ground that for relevant purposes the employee was “no longer” his employee.

In Rourke the defendant colliery contracted with a contractor, Whittle, to do engineering work in the pit, and for these purposes supplied the contractor with equipment and an engineer in the colliery’s own employment and pay. The claimant was an employee of the contractor, who had been injured by reason of the engineer’s negligence. Under the work contract, the engineer was to be under the control and orders of the contractor. The court of appeal differed in the reasons for its decision that the colliery was not liable for its engineer’s negligence. Cockburn CJ put it on the general basis that an employee lent to a temporary employer for a particular purpose “must be dealt with as the servant of the man to whom he is lent” and that therefore the claimant was a “fellow-servant” and fell foul of that era’s doctrine of common employment, namely that one employee could not recover damages from his employer for the negligence of his fellow-employee (at 209). Mellish LJ and Baggallay JA, however, held that, although the engineer remained the general servant of the colliery, it was not liable for his negligence because the fact that he had been put under the control and orders of the contractor meant that his acts were “the acts of Whittle and not of the defendants” (at 211). They therefore considered that they did not have to deal with the doctrine of common employment. Bramwell JA, however, adopted the ratio of common employment (at 211/212). It seems to me that on either ratio the court considered that for relevant purposes either the whole status of the borrowed employee was transferred to the temporary employer, or at any rate that his acts were to be regarded as those of the employer under whose orders he was working. There was, again, no submission that it might be possible to view both employers as liable, but an assumption that a choice had to be made.

If the line of authority had stopped there, it might be arguable that Donovan had implicitly, even if not expressly, decided that dual vicarious liability could not exist where the entire and absolute control of the relevant working of the employee rested in one or other of the two employers. In my judgement, however, in that respect Donovan merely followed Rourke, which did not determine the part but proceeded on an assumption. Moreover, even if that argument were accepted, that would not determine the answer where the relevant control was divided between the employers. And there is something highly unsatisfactory about the relationship of employer and employee switching backwards and forwards between two employers depending upon what the employee is doing at any moment or what the precise causation of the injury is. In truth, if for one purpose one employer has entire control and for another purpose the other employer has entire control, control is shared: and what then? In any event, it would seem doubtful that the court in Donovan could really be talking about a true relationship of employer and employee, since that can hardly depend on the precise work which an employee is doing at any moment or the precise causation of the injury. It appears rather that the court has imperceptibly moved from using the test of right of control as determinative of the relationship of employer and employee, to using it as the test of the vicarious liability of a defendant.

In Mersey Docks and Harbour Board v. Coggins & Griffith (Liverpool) Limited [1947] AC 1, however, these matters were revisited. That case again concerned the loan of a crane and a crane driver to a firm of stevedores. On this occasion, the claimant sued both the stevedores and the Board, which was the lending or general employer. Donovan was distinguished by Lord Simon on the ground that there the entire and absolute control was in the hands of the stevedores (at 11): but in truth their Lordships took a different view from this court in Donovan of the relationships involved, distinguishing the orders of the stevedores as to what work was to be done from the crane driver’s discretion as to how to perform that work, and pointing out that the crane driver was still performing work according to a discretion vested in him by his general employer, the Board (see Lord Macmillan at 13, and Lord Porter at 16, and compare Lord Esher’s purely general description of the loading of ships in Donovan at 632). In such circumstances, the entire and absolute control as to the method of performing the crane driver’s work had not been transferred, and the responsibility for the crane driver’s negligence remained with the Board.

Their Lordships, while making use of Lord Esher’s test of control, have advanced the analysis in at least three respects. First, they emphasise that the question of where responsibility lies is not to be decided by a single universal test, but depends on all the facts of the case and a variety of tests (see Lord Justice May’s summary at para 7 above). Secondly, they emphasise that the question ultimately is as to where vicarious responsibility for the negligence of the employee is to be placed, rather than strictly as to whose employment the employee is in: for they point out that employment cannot formally be transferred without the consent of the employee (at 15, 22). Thirdly, however, they also emphasise that where one starts with the fact that the negligent actor is the employee of the lending or general employer, the burden of showing that there is a transfer of responsibility is a heavy one. In practice, therefore, where, as in Mersey Docks, there was no transfer of entire and absolute control to the other employer, liability was likely to remain with the general employer.

In Mersey Docks however the issue was ultimately between the stevedores and the Board, each of whom argued for the sole responsibility of the other. It was not argued that there could be a dual vicarious liability. Although Lord Simon referred in general to “the carriage cases” (at 11), Laugher v. Pointer was not mentioned by name. Rourke, which this court regarded as binding it in Donovan, was mentioned only by Lord Porter, and then as a case which did not give much assistance: Lord Porter pointed out that it concerned an employee who had left his ordinary employment for another employer, as distinct from one who continued to do his ordinary work while subjected from time to time to the directions of a third party (at 16).

In such circumstances, although the background of Mersey Docks may have reflected an assumption that total vicarious liability was to be found either in one employer or the other and could not be found in both, it is not a decision to that effect. As for Donovan, to the extent that it may suggest that entire and absolute control over an employee can be easily transferred, its authority is very much weakened if not extinguished by Mersey Docks. Moreover, to the extent that Donovan might be read as implying that, if entire and absolute control is transferred, then it must follow that vicarious liability is to be found exclusively in the transferee employer, its authority has also been very much weakened if not extinguished: both because entire and absolute control, although an important test of where vicarious responsibility lies, is not a sufficient let alone a necessary one, and also because Rourke was treated as being derived from a different factual situation.

It was at this point that Denning LJ’s analysis in Denham v. Midland Employers’ Mutual Assistance Limited [1955] 2 QB 437 exposes most clearly and helpfully the distinction to be drawn between a transfer of employment (which depends on contract and needs the employee’s consent – indeed, it would seem to me to need the consent of all three parties) and the imposition of vicarious liability on the borrowing or temporary employer. The distinction needed to be made because the litigation arose, not as usual out of a claim by a third party against one or other of the employers, but out of a claim by the employee’s widow against the insurers of the temporary employer. The defendant insurers had issued an employers’ liability policy which covered the temporary employer’s liability to “any person under a contract of service…with the insured”. So the question was whether the deceased had had a “contract of service” with the borrowing or temporary employer, and he plainly had not. No issue of vicarious liability arose in that case concerning the acts of the borrowed employee. Nevertheless Denning LJ’s observations covered both situations. He said (at 443/444):

“Much of the difficulty which surrounds the subject arises out of the nineteenth century conception that a servant of a general employer may be transferred to a temporary employer so as to become for the time being the servant of the temporary employer. That conception is a very useful device to put liability on the shoulders of the one who should properly bear it, but it does not affect the contract of service itself. No contract of service can be transferred from one employer to another without the servant’s consent…The supposed transfer is nothing more than a device – a very convenient and just device, mark you – to put liability on the temporary employer; and even this device has in recent years been very much restricted in its operation. It only applies when the servant is transferred so completely that the temporary employer has the right to dictate, not only what the servant is to do, but also how he is to do it: see Mersey Docks…The labourer becomes so much part of the orgnization to which he is seconded that the temporary employer is responsible for him and to him.

Applying these principles to the facts found by the arbitrator, I have myself no doubt that if a third person had been injured by the negligence of Clegg [the employee] in the course of his work, then Le Grands [the temporary employer] and not Eastwoods [the general employer] would have been liable to the third person…These results are achieved in law by holding that [the employee] became for the time being the temporary servant of Le Grands. There is no harm in thus describing him so long as it is remembered that it is a device designed to cast liability on the temporary employer. The real basis of the liability is, however, simply this: if a temporary employer has the right to control the manner in which a labourer does his work, so as to be able to tell him the right way to do it, then he should be responsible when he does it in the wrong way as well as in the right way. The right of control carries with it the burden of responsibility…”

Romer LJ put the matter thus (at 447):

“There is a clearly marked distinction between the transfer of a servant on the one hand and the transfer of only his services on the other…It is true, as Denning L.J. has pointed out, that for some purposes Le Grands undoubtedly assumed the obligations and liabilities of a master in relation to Clegg; for example they would have been answerable to a third party who was injured as a result of Clegg’s negligence while working for them. This consideration does not, however, in my opinion affect the real issue in this case, which is whether Clegg at the time of his death was a person under a contract of service with Le Grands…”

Birkett LJ agreed with both judgments (at 445).

There is again, at any rate in Denning LJ’s judgment, the assumption that vicarious liability cannot be dual (“and not Eastwoods”), at any rate on the basis that the transfer of services is so complete that the temporary employer has the right to control (he does not, however, say the sole right to control) how the employee is to do his work. Nevertheless, that assumption is patently obiter.

I come next to Esso Petroleum Co Ltd v. Hall Russell & Co Ltd [1989] AC 643, with which Lord Justice May has dealt in detail at paras 33/34 above. The critical passage in Lord Jauncey’s speech is as follows (at 685H/686B):

“I would only add that if Hall Russell’s argument were correct there would follow the curious result that the doctrine of respondeat superior would apply to two different masters in respect of damage arising out of a single act of negligence. It is a well recognised principle, exemplified in cases involving crane drivers, that a servant in the general employment of A may, for a particular purpose, be treated as in the pro hac vice employment of B. However, there is no principle which permits a servant to be in the de jure employment of two separate masters at one and the same time.”

I would respectfully make the following observations about this passage. In the first place, it is not rejecting a submission that there was in that case dual vicarious liability. Hall Russell did not so submit. Its submission was that section 15(1) of the Pilotage Act 1913 (which provided that a compulsory pilot was the shipowners’ servant for all purposes) reflected the common law position that a pilot was the shipowners’ independent contractor not servant (albeit shipowners owed a non-delegable duty to have their ships navigated safely) and therefore it was open for a pilot to render his general employer vicariously liable for his negligence. Secondly, the “no principle” final sentence is not denying the possibility of dual vicarious liability, but stating that an employee cannot formally (de jure) have two separate master/servant employment relationships at the same time. That formal position is contrasted with the informal or perhaps metaphorical doctrine whereby the employee of one master may be treated as in the employment of another master for certain purposes. Thirdly, Hall Russell’s submission was an impossible one because by reason of the statute the pilot was to be considered the servant of the shipowner for all purposes. Fourthly, Lord Jauncey observed that, if Hall Russell’s submission, viz that the general employer was liable, were correct, then there would be dual vicarious liability (because the statute, contrary to Hall Russell’s submission about it, made the pilot the shipowners’ servant): and that that would be a “curious result”. But he did not say that it was, in principle, an impossible result. Fifthly, Lord Jauncey’s observations are introduced by the phrase “I would only add”, which to my mind indicates that they are not intended to be part of any ratio of the case. Presumably, however, his observations reflect the historical assumption against dual vicarious liability.

The jurisprudence of other jurisdictions, as shown by Lord Justice May, encompasses both reflections of Mersey Docks but also in places the acceptance of the possibility of dual vicarious liability. Academic opinion favours the latter possibility, but, as Lord Justice May puts it, feels the restraint of authority.

In my judgment, there is no doubt that there has been a long standing assumption that dual vicarious liability is not possible, and in such a situation it is necessary to pause carefully to consider the weight of that tradition. However, in truth, the issue has never been properly considered. There appears to be a number of possible strands to the assumption. Two are mentioned by Littledale J: the formal principle that a servant cannot have two masters; and the policy against multiplicity of actions. As for the first, even if it be granted that an employee cannot have contracts of employment with two separate employers at the same time and for the same period and purposes – and yet it seems plain that a person can (a) have two jobs with separate employers at the same time, provided they are compatible with one another; or (b) be employed by a consortium of several employers acting jointly – nevertheless that does not prevent the employee of a general employer being lent to a temporary employer. As was so clearly exposed in Denham, it is an inaccurate metaphor to say that the employment or the employee has been transferred: it is rather that the services of the employee have been lent or hired out, or borrowed or bought in, in circumstances where the temporary employee becomes responsible, under the doctrine of vicarious liability (respondeat superior) for the employee’s negligence, and does so even though the formal contract or relationship of employment has not been transferred. That demonstrates that the doctrine of vicarious liability may properly be invoked against an employer who is not really, in law, the employee’s employer; and that the use of the expression “transfer” is potentially misleading. As for the policy against multiplicity of actions: no doubt the law does not favour unnecessary complexity which may lead to the suing of unnecessary defendants. But such a policy, while it may inform the formulation of doctrine, cannot determine it; and in any event, the history of this jurisprudence has demonstrated clearly that it is not safe for a claimant to assume that he can sue one employer only. Whereas the Mersey Docks approach may have discouraged the suing of the temporary employer save in exceptional situations, the problem remains a live one.

In my judgment, if consideration is given to the function and purposes of the doctrine of vicarious liability, then the possibility of dual responsibility provides a coherent solution to the problem of the borrowed employee. Both employers are using the employee for the purposes of their business. Both have a general responsibility to select their personnel with care and to encourage and control the careful execution of their employees’ duties, and both fall within the practical policy of the law which looks in general to the employer to organise his affairs in such a way as to make it fair, just and convenient for him to bear the risk of his employees’ negligence. I am here using the expression “employee” in the extended sense used in the authorities relating to the borrowed employee. The functional basis of the doctrine of vicarious liability has become increasingly clear over the years. The Civil Liability (Contribution Act) 1978 now provides a clear and fair statutory basis for the assessment of contribution between the two employers. In my judgment, the existence of the possibility of dual responsibility will be fairer and will also enable cases to be settled more easily.

The remaining question is to attempt to define the circumstances in which the liability should be dual. It is possible that where the right to control the method of performance of the employee’s duties lies solely on the one side or the other, then the responsibility similarly lies on the same side. That reflects the significance of Lord Esher’s doctrine of entire and absolute control. If so, then it will only be where the right of control is shared that vicarious liability can be dual. I would agree that the balance of authority is in favour of this solution. On this basis, I agree with Lord Justice May’s analysis of the facts in this case as demonstrating a situation of shared control. I would go further and say that it is a situation of shared control where it is just for both employers to share a dual vicarious liability. The relevant employee, Darren, was both part of the temporary employer’s team, under the supervision of Mr Horsley, and part of the general employer’s small hired squad, under the supervision of its Mr Megson.

However, I am a little sceptical that the doctrine of dual vicarious liability is to be wholly equated with the question of control. I can see that, where the assumption is that liability has to fall wholly and solely on the one side or the other, then a test of sole right of control has force to it. Even Mersey Docks, however, does not make the control test wholly determinative. Once, however, a doctrine of dual responsibility becomes possible, I am less clear that either the existence of sole right of control or the existence of something less than entire and absolute control necessarily either excludes or respectively invokes the doctrine. Even in the establishment of a formal employer/employee relationship, the right of control has not retained the critical significance it once did. I would prefer to say that I anticipate that subsequent cases may, in various factual circumstances, refine the circumstances in which dual vicarious liability may be imposed. I would hazard, however, the view that what one is looking for is a situation where the employee in question, at any rate for relevant purposes, is so much a part of the work, business or organisation of both employers that it is just to make both employers answer for his negligence. What has to be recalled is that the vicarious liability in question is one which involves no fault on the part of the employer. It is a doctrine designed for the sake of the claimant imposing a liability incurred without fault because the employer is treated by the law as picking up the burden of an organisational or business relationship which he has undertaken for his own benefit.

One is looking therefore for practical and structural considerations. Is the employee, in context, still recognisable as the employee of his general employer and, in addition, to be treated as though he was the employee of the temporary employer as well? Thus in the Mersey Docks situation, it is tempting to think that liability will not be shared: the employee is used, for a limited time, in his general employer’s own sphere of operations, operating his general employer’s crane, exercising his own discretion as a crane driver. Even if the right of control were to some extent shared, as in practice it is almost bound to be, one would hesitate to say that it is a case for dual vicarious liability. One could contrast the situation where the employee is contracted out labour: he is selected and possibly trained by his general employer, hired out by that employer as an integral part of his business, but employed at the temporary employer’s site or his customer’s site, using the temporary employer’s equipment, and subject to the temporary employer’s directions. In such a situation, responsibility is likely to be shared. A third situation, where an employee is seconded for a substantial period of time to the temporary employer, to perform a role embedded in that employer’s organisation, is likely to result in the sole responsibility of that employer.

Finally, there is the question of contribution between two employers who are both vicariously, but only vicariously, liable. In principle, the two employers will generally be able to contract between themselves as to the basis of contribution. In practice, any such contract is likely to place responsibility entirely on the one side or the other, often accompanied by further provisions regarding insurance. In such a case, while both employers will each be fully liable to the claimant, only one will be liable inter se, for the effect of their contract will be recognised within contribution proceedings. Since the doctrine of vicarious liability is for the benefit of the claimant, and since the two employers are in general able to contract as to their liability inter se with respect to any responsibility they may incur for the negligence of the employee, it seems to me to be unnecessarily restrictive to insist on the impossibility of a dual vicarious liability, thereby forcing on the claimant the historically problematical issue of where vicarious liability ultimately lies, and also forcing on the two employers the need to fight about an all or nothing responsibility.

The 1978 Act provides, so far as is material, as follows:

“1(1) …any person liable in respect of any damage suffered by another person may recover contribution from any other person liable in respect of the same damage (whether jointly with him or otherwise).

2(1)…the amount of the contribution recoverable from any person shall be such as may be found by the court to be just and equitable having regard to the extent of that person’s responsibility for the damage in question.”

It is clear that both employers are liable to the claimants in respect of the same damage. What is the just and equitable amount that each should contribute having regard to the extent of the responsibility of each for the damage in question?

It has been established that “responsibility” includes both causative potency and blameworthiness. However, in the case of vicarious liability, the employer is liable without personal fault. The fault in question is the employee’s. The employer thus stands fully in the shoes of the negligent employee as regards both aspects of responsibility: see Dubai Aluminium Co Ltd [2002] UKHL 48, [2003] 2 AC 366 at paras 47 and 160.

Where, therefore, there is dual vicarious liability arising out of the negligence of a single employee, it follows that the responsibility of each employer for the purposes of contribution must be equal. In other words, in the absence of any personal fault on the part of either employer in respect of the same damage, and in the absence of any other negligence by another employee contributing to the same damage, as here in the absence of any negligence by either Mr Megson or Mr Horsley, the essential decision as regards contribution as well as liability occurs at the time when the court determines that there is dual vicarious liability. The realisation that dual vicarious liability means equal responsibility and equal financial liability could and probably should therefore enter into the earlier and determinative decision. The question would be whether in all the circumstances, including the important question of control, vicarious liability should be shared, on the basis that the employee in question, although not formally the employee of the temporary employer, is, at least for relevant purposes, so much a part of the work, business or organisation of both employers as to make it just for there to be dual and shared liability.