Causes, Duties and Risks

October 2, 2007

Complex IT projects frequently overrun the programmed project period. Overruns cause loss. Loss caused by culpable overruns should be payable by the culpable party, whether employer or contractor. Establishing culpability requires causation to be shown.


As a practical matter the identification of the causes of delay and disruption is difficult in any complex project because an interaction of events will have lead to disruption and delay. It is even more difficult in complex IT projects. Most large IT projects involve a due diligence or requirements gathering phase. Particularly in that phase, but even after that phase, mutual co-operation is required between the parties. Identifying culpable non-co-operation is difficult. Further, the progression of the work will not necessarily be sequential beyond an elementary level and so the link between event and effect may be obscure. The causes may be a number of events, possibly running into hundreds, which operate over many months and which interact with other events so as to render the causative effect of any one event apparently opaque. Further, there may be a considerable elasticity in the ability of the project to accept increased resource and so the need for any overrun may not be obvious.


Analytical programmes are available, by which a mass of information can be ordered and the overall causes and effects of delay broken down into smaller manageable chunks.[1]


But quite apart from the practical difficulties it is necessary to consider the question of what in law counts as cause and effect and how is it demonstrated. How are causes that carry legal responsibility in any particular case to be differentiated from those that do not? How is legal responsibility for loss to be attributed among several causes? These are ‘the lawyer’s main perplexities’[2] How are they to be resolved? By commonsense or by something else?


There is a special ‘sense in which the word “cause” is used in law’.[3] In law causation is not causation as it is usually understood either in the sense that a logician or philosopher would give to it or in the sense that a member of the public would give to it.


In preparing proof of causation it will be necessary to analyse causation in fact sufficiently rigorously to show that the event of interest is within the class of the jointly sufficient causes (broadly the work done by analytical programmes). But it will also be necessary to analyse the duties owed by the respective parties. This analysis may involve some analysis of issues on which most contracts are relatively unclear, such as whether there is a duty to mitigate time or expense where events occur which have a tendency to increase either and, if so, whether that duty is or is not accompanied by a right to recover the costs of that mitigating action.


If, for example, the contractor is hopelessly and culpably in delay but nonetheless unable to complete integration testing due to one outstanding decision by the employer, is the contractor entitled to an extension of time? If the employer delays the contractor but the resources that can be deployed are elastic so that the delay can be caught up, is the contractor entitled to maintain resources at programmed levels and insist on an extended period for the execution of the works?



What Counts as Proof?


In every contested case there will be a claimant and a defendant on opposite ends of each claim. For example, the employer will claim unliquidated or possibly liquidated damages for delay. The contractor will claim damages or loss and expense in the form of the cost to it of its extended period on site or its less efficient time on site or both. The court will have to determine those claims.


Issues for the parties to such disputes include; ‘What counts in the courts as proof of a cause of delay so that the other party is liable?’ ‘What evidence will I need? What advice as to prospects can be given on the basis of that evidence?’


The current edition of Chitty on Contracts[4] states at para 26-029,


          Requirements of a causal connection. The important issue in remoteness of damage in the law of contract is whether a particular loss was within the reasonable contemplation of the parties, but causation must first be proved: there must be a causal connection between the defendant’s breach of contract and the claimant’s loss. The claimant may recover damages for a loss only where the breach of contract was the “effective” or “dominant” cause of that loss. The courts have avoided laying down any formal tests for causation: they have relied on common sense to guide decisions as to whether a breach of contract is a sufficiently substantial cause of the claimant’s loss. The answer to whether the breach was the cause of the loss or merely the occasion for the loss must “in the end” depend on “the court’s commonsense” in interpreting the facts. So where a company continued to trade after a negligent audit by the defendant failed to reveal the true financial position of the company, the Court of Appeal held that the auditor’s breach of contract gave the company “the opportunity … to incur … trading losses: it did not cause those trading losses”. The trading losses flowed from trading, not auditing’[5].


The Galoo Test : ‘By the application of the court’s common sense’


In Galoo v Graham Bright Murray[6] the defendant accountants were sued by their former clients, the claimants, for negligence and breach of contract in connection with the auditing of the claimants’ accounts. It was alleged that that audit had failed to detect a fraud whereby the claimants’ assets were substantially overstated. The claimants’ case was that had the audit detected the overstatement the claimant’s would have ceased trading and not incurred the losses incurred in subsequent trading. Therefore the trading losses were caused by the defendants’ breach of duty comprised by the negligent audit. This claim was struck out as disclosing no reasonable cause of action on the grounds that the defendants’ breach of duty gave the opportunity to the claimants to incur trading losses but ‘it did not cause those trading losses, in the sense in which the word “cause” is used in law substantially because the trading losses were caused by the factors such as the prudence of the trading, the market conditions which caused the trading to result in losses rather than profits.’[7]


The Court of Appeal was considering a specific example of a case in which Event A creates the conditions in which Event B causes loss. The loss would not occur without Event A (had the audit been accurate the company would have ceased to trade and there would have been no trading losses). Equally however the loss would not occur without Event B (had the trading been sufficiently shrewd as to be profitable the loss would not have occurred). The question was whether as a matter of ‘the sense in which the word “cause” is used in law’[8] Event A caused the loss.


However, the citation from Chitty suggests that Galoo is regarded as of general application.


Five cases constituted the immediate background to the Galoo decision.


In Monarch Steamship Co. Ltd v Karlshamns Oljefabriker (A/B)[9] Event A was the defectiveness of the ship’s boilers which delayed her voyage. But for the defect in the boilers the ship would have arrived in Sweden by July 1939. Due to the delay the ship was at sea when the Second World War broke out and was ordered by the British Admiralty to divert to Glasgow (Event B). That diversion caused loss in the form of the extra cost of the transhipment of the cargo. The defendant shipowner contended that the defect in the boilers did not cause the diversion of the ship – that was caused by and only by the Admiralty order. In a sense this was true because the order was the ‘proximate’ cause. However the only reason that the ship was at sea when war broke out and therefore exposed to the peril of the Admiralty’s order was the prior delay caused by the defective boilers. The House of Lords held the shipowners liable.


In Stapley v Gypsum Mines Ltd[10] two miners, Stapley and Dale were instructed to take down a roof with the corollary that they were not to work under it in the meantime. After an unsuccessful attempt to get the roof down they abandoned the effort by a joint decision and proceeded with their normal work under the roof. The roof collapsed killing Stapley. Stapley’s widow’s claim against his employers succeeded before the judge on the basis that Stapley’s death was caused by both his and Dales’ actions. The Court of Appeal reversed this holding that the death was caused by Stapley’s actions alone. The House of Lords by a majority allowed an appeal holding that Dales’ actions were a contributory cause so enabling the employers to reduce the damages.


In Quinn v Burch Bros. (Builders) Ltd[11] the defendant main contractor in breach of contract failed to supply the claimant sub contractor a step ladder to carry out his work (Event A). The sub contractor used a folded trestle as if it were a ladder (Event B), the foot of it slipped and he fell and broke his foot. The claim was dismissed by the judge and the Court of Appeal on the grounds that Quinn’s decision to use the trestle rather than wait for a ladder was the cause of his injury.[12]


In Alexander v Cambridge Credit Corporation Ltd[13] auditors failed to note that appropriate provisions had not been made in the accounts of the claimant company (Event A). Had the note been made the company would have been put into receivership and would not have continued to trade. During the period in which it did continue to trade its suffered an increase in the deficiency of its assets (Event B). A majority of the Court dismissed the claim.


In March v E. & M.H. Stramare Pty Ltd[14] the defendants parked their truck straddling the centreline of a six lane road (Event A). The drunken plaintiff travelling at night drove his car into the back of the defendants’ truck (Event B). The High Court of Australia allowed the claim. The judgment of Mason CJ is a lucid survey of the difficult issues in this area. As to the test for causation he said (at paras 5 and 27):


          ‘5.      It has often been said that the legal concept of causation differs from philosophical and scientific notions of causation. That is because “questions of cause and consequence are not the same for law as for philosophy and science”, as Windeyer J. points out in The National Insurance Co. of New Zealand Ltd . Espagne (1961) 105 CLR 569, at p 591. In philosophy and science, the concept of causation has been developed in the context of explaining phenomena by reference to the relationship between conditions and occurrences. In law, on the other hand, problems of causation arise in the context of ascertaining or apportioning legal responsibility for a given occurrence. The law does not accept John Stuart Mill’s definition of cause as the sum of the conditions which are jointly sufficient to produce it. Thus, at law, a person may be responsible for damage when his or her wrongful conduct is one of a number of conditions sufficient to produce that damage: see McLean v Bell (1932) 147 LT 262, per Lord Wright at p 264; Sherman v Nymboida Collieries Pty Ltd (1963) 109 CLR 580, per Windeyer J at pp 590-591.




           27.    As a matter of both logic and common sense, it makes no sense to regard the negligence of the plaintiff or a third party as a superseding cause or novus actus interveniens when the defendant’s wrongful conduct has generated the very risk of injury resulting from the negligence of the plaintiff or a third party and that injury occurs in the ordinary course of things. In such a situation, the defendant’s negligence satisfies the “but for” test and is properly to be regarded as a cause of the consequence because there is no reason in common sense, logic or policy for refusing to so regard it.


Having reviewed these cases Glidewell LJ concluded:[15]


          ‘The passages which I have cited from the speeches in Monarch Steamship Co Ltd v Karlshamns Oljefabriker A/B [1949] AC 196 make it clear that if a breach of contract by a defendant is to be held to entitle the plaintiff to claim damages, it must first be held to have been an “effective” or “dominant” cause of his loss.  The test in Quinn v Burch Bros (Builders) Ltd [1966] 2 QB 370 that it is necessary to distinguish between a breach of contract which causes a loss to the plaintiff and one which merely gives the opportunity for him to sustain the loss, is helpful but still leaves the question to be answered ‘How does the court decide whether the breach of duty was the cause of the loss or merely the occasion for the loss?’


          The answer in my judgment is supplied by the Australian decisions to which I have referred which I hold to represent the law of England as well as of Australia, in relation to a breach of a duty imposed on a defendant whether by contract or in tort in a situation analogous to breach of contract. The answer in the end is ‘By the application of the court’s common sense’.


          Doing my best to apply this test, I have no doubt that the deputy judge arrived at a correct conclusion on this issue. The breach of duty by the defendants gave the opportunity to Galoo and Gamine to incur and to continue to incur trading losses; it did not cause those trading losses, in the sense in which the word ‘cause’ is used in law.’


Practical Difficulties arising out of the Galoo Test


Certain descriptive terms used by the courts such as ‘the effective cause’[16] ‘the dominant cause’[17] or the distinction between the ‘occasion’ and the ‘opportunity’ and the ‘cause’[18] may really be different ways of stating the conclusion as to the legal cause rather than a description of how that conclusion is reached. That is the point made by Glidewell LJ[19] and drives his search for a better formulation.


However perhaps the Galoo answer as shortly stated itself suffers from the same defect. In its simplest terms what Glidewell LJ appears to be saying is that in law the cause is the event that, ‘By the application of the court’s common sense’, is the cause.


It is worth emphasising that the common sense is the court’s common sense and not anyone else’s. A consideration of the cases considered by Glidewell LJ shows that that may be problematical. As Glidewell LJ observed, noting the dissent of Glass J.A. in the Alexander case on the application of the facts to the law:


          ‘The fact that Glass JA did not agree with his colleagues in the result shows that, inevitably, not all judges regard common sense as driving them to the same conclusion.’[20]


The fact that different judges can come to different conclusions even in the same cases is a significant objection to the Galoo answer as it is stated. If the court’s common sense is simply the common sense of the individual judge and if there is no shared common sense across all judges explained or exemplified by transparent criteria then the outcome in the complex cases that are litigated will be difficult to predict.


What Would Count as Persuasive?


The fuller formulation by Mason CJ in the Stramare case may assist. He said, ‘In law, … problems of causation arise in the context of ascertaining or apportioning legal responsibility for a given occurrence.’[21] He applied that proposition by holding at paragraph 27 of his judgment that the determinative fact was that the event which happened fell within the class of events which was the target of the relevant duty.


In short it may be that a legal cause must be both a member of the set of jointly sufficient causes[22] and within the class of events which the relevant duty aimed at preventing.


The facts of Stapley and Stramare illustrate this. In Stapley the decision of Dale to give up on dismantling the roof was a jointly sufficient cause and the employer’s duty extended to protecting Stapley from Dale’s lack of care. In Stramare both the defendant’s negligent parking and the claimant’s drunkenness were jointly sufficient causes and the negligent parking was a breach of a duty aimed at avoiding precisely what occurred.


If this approach is right then the focus should be not on distinguishing between ‘occasions’ and ‘causes’ or characterising the conclusion as ‘common sense’ but on the relevant duty as the key by which to unlock the question of legal cause.[23]


There is a similarity between this approach and the approach taken by the House of Lords to the question of duty in South Australia Asset Management Corporation v York Montague Ltd[24] In that case the claimants sued surveyors employed by them to carry out pre-purchase valuations of certain properties in respect of which the claimants proposed to make loans. The borrowers defaulted and the claimants had to rely upon the secured properties to recover the amount of the loans. The properties were insufficiently valuable for that purpose. The claimants’ loss was markedly increased due to a fall in the property market which occurred after the purchases had gone through. The claimants alleged that had the surveyors’ reports not been negligent it would not have made the loans. Therefore the negligence was a sufficient cause of both the loss due to the undervaluation and of the loss due to the fall in the property market (although in the latter case jointly sufficient with that fall). The House of Lords disallowed recovery of the loss due to the fall in the market holding that the scope of the duty defined the loss in respect of which the duty was owed and therefore the nature and amount of the loss that could be recovered and the fall in the market was outside the scope of that duty.


Lord Hoffman has subsequently drawn attention to the similarities of this to the Court’s approach to the issue of causation.[25]


Therefore in many cases the same sort of test may deal with both the issue of the scope of the duty in law and also with the issue of the legal cause.


Lord Hoffman’s well-known example of the doctor advising the mountaineer is a good illustration. The doctor’s negligent advice that the mountaineer’s knee will come to no harm causes the mountaineer to climb the mountain where he suffers an accident unrelated to his knee. The doctor’s advice is Event A. It is clearly a member of the set of jointly sufficient causes. The subsequent accident is Event B. It too is a member of the set of jointly sufficient cases. But because the scope of the duty is confined to knee related injuries the other accident does not fall within ‘the very risk of injury that befell the appellant’.[26]


Delay Analysis in IT Projects


If a value judgment based on legal responsibility is required to consider questions of causation in relation to IT projects it is likely to be based on a consideration of the respective rights and obligations of the parties, defined, of course, by their contract.


Suppose for example that a variation instruction is issued late but then the contractor takes longer than ‘reasonable’ to implement it. The delay is caused both by the instruction of the variation and by the contractor’s decision to carry it out within the original and extended period rather than within the original period. Is the instruction simply a jointly sufficient cause both of the ‘reasonable’ and of the ‘unreasonable’ delay (employer pays all contractor’s loss) or is it both that and the legal cause of the ‘reasonable’ delay but not of the ‘unreasonable’ delay? Does it follow that the contractor’s ‘slowness’ in carrying out the additional work is the legal cause of the unreasonable delay (contractor pays employer’s loss rateably)?


If the answer depends upon the duty then it is necessary to examine the specific relevant duty owed by (in this case) the contractor. What, for example, is ‘reasonable’ progress? Is it progress at the cost to profit ratios of the accepted tender (the contractor can carry on the work without acceleration or adjusting its resources) or at a higher ratio (contractor increases resources to accelerate to complete the increased work in the same time as allowed for the original work)?


If the contractor engages new staff to carry out the varied and original work in the original time at a higher cost than the cost to it of carrying out the original and varied work in the original and extended time can it recover that higher cost? Is that higher cost legally caused by the variation or by the contractor’s decision to carry out the original and varied workscope in the original time? On the other hand if the contractor carries out the original and varied work in the original and extended period when by lesser extra expenditure it could have carried out the original and varied workscope in the original period can the employer claim damages for delay and/or deprive the contractor of the ‘unnecessary’ extra increment of any loss and/or expense?


In the case of varied work which may or may not extend the contract period, whether the contractor should recover the higher cost of completing the work within the original period may depend upon whether the contract requires the contractor to take all steps to reduce delay. If it does then the contractor may be bound (a) to carry out the varied work and (b) to reduce the delay. If he does so then he might recover the extra cost of carrying out the original and varied work within the original period. If he does not do so, then he might be liable for damages for late completion although the employer may have to give credit for the saving made in not having to pay the contractor the higher cost of completing the varied and original work within the original period.


In a contractual context, responsibility is defined by the express and implied terms of the contract. Therefore the answer to such questions depends upon the terms of the relevant contract.


Conclusions about whether loss caused by delay and disruption is recoverable involve issues of duty and risk as well as questions of occasions, conditions and causes. An analysis of duty and risk will be carried out by reference to the terms of the contract and in appropriate cases may involve consideration of the commercial context, not least the contractor’s pricing of the original undisrupted, un-delayed project.


Nicholas Baatz QC is from Atkin Chambers ( . He has been in practice specializing in Technology and Construction law since 1980. He has also acted as an arbitrator, as an adjudicator, and as a mediator.



[1]               Such programmes are considered by Jonathan Westwell and Stuart Wilks in Computers & Law, June/July 2006, p 18.

[2]               See Hart and Honore Causation in the Law 2nd Edn (1985) p 24. ‘the lawyers’s main perplexities…these arise when, after it is clearly understood how some harm happened, the courts have, because of the form of legal rules, to determine whether such harm can be attributed to the defendant’s action as its consequence, or whether he can properly be said to have caused it. These may be called attributive enquiries’


[3]               Galoo v Bright Graham Murray [1994] 1 WLR 1360, considered below, per Glidewell at 1375B.

[4]               29th Edition 2004.

[5]               A reference to the Galoo case.

[6]               [1994] 1 WLR 1360.

[7]               See [1994] 1 WLR 1375A/B and 1374E-G.

[8]               Per Glidewell at 1375B.

[9]               [1949] AC 196.

[10]             [1953] AC 663

[11]             [1966] 2 QB 370.

[12]             As Salmon LJ noted, at p 394, the claimant relied in part on the skilful cross examination of the foreman to show that since Quinn was a conscientious workman it was foreseeable that he would attempt to get on with the work rather than, for example, play cards. The argument failed. The parallels to, and contrast with, Stapley’s case are clear.

[13]             (1987) 9 NSWLR 310.

[14]             (1991) 171 CLR 506

[15]          At p 1374G-1375B


[16]             Lord Reid in Stapley v Gypsum Mines Ltd [1953] A.C. 663 at 681.

[17]             Lord Wright in the Monarch Steamship case at [1949] A.C. 196 at page 228.

[18]             Drawn by Salmon and Danckwerts LJJ in Quinn v Burch Bros (Builders) Ltd [1966] 2 QB 370.

[19]             [1994] 1 WLR 1370 at 1394H.

[20]             At [1994} 1 WLR 1370 at 1372C

[21]             At paragraph 5

[22]             Used in the sense of including occasions and conditions and so not necessarily ‘dominant’.

[23]             Of course, that may shift the real discussion to the question of duty. In the Quinn case, was the employer’s duty to be defined by its appreciation of what Quinn as a conscientious employee would do? But at least this is normative, not an opaque form of ‘legal’ causation.

[24]             [1997] AC 191.

[25]             In an address to the London Common Law and Commercial Bar Association on 20 June 2005.

[26]             To adopt the words of Mason CJ in the Stramare case at paragraph 28. Hart and Honore give a similar example of the man who shoots at his wife intending to kill her. She takes refuge in her parents’ house where she is injured by a falling tile, see Causation in the Law p 12.