The respondents, who owned the Crystal Jewel, claimed damages against the appellants, who owned the British Aviator, in respect of a collision between the vessels in the English Channel on 23 September 1961 in dense fog. The vessels were initially on almost opposite courses. Each vessel had spotted the other on radar. Both vessels were sounding their whistle long blasts for fog; and as they drew closer, each vessel heard the signals of the other. Both vessels were proceeding at half speed ahead. Both vessels admitted that, having heard fog signals from a vessel whose position had not been ascertained, they failed to stop their engines and to navigate with caution. Instead, the Crystal Jewel made repeated small alterations to starboard, while the British Aviator made a large alteration to port shortly before the collision. Ultimately the vessels collided at about a right angle.

At first instance, Cairns J apportioned blame for the collision as to the Crystal Jewel, two-fifths, and as to the British Aviator, three-fifths: see The British Aviator [1964] 2 Lloyd's Rep 403. Cairns J held that both vessels were in breach of the principles of good seamanship particularised in the Ministry of Transport Notice No M 445 (Notice); that both vessels were to blame in that they used radar information as if it were visual information and as an excuse for maintaining speed; that the alterations of course of the Crystal Jewel were bad navigation, but that the British Aviator was more seriously negligent in altering course to port. The defendants appealed and argued that both vessels were equally blameworthy. The defendants argued that there was no sufficient ground on which Cairns J could properly differentiate between the helm action taken by both vessels. The difference in money between that apportionment and an apportionment of both equally to blame was no less than GBP 60,000.

Held: Appeal allowed.

Both parties were equally to blame. The main faults were faults common to both vessels. First, excessive speed in fog. Secondly, failure to make adequate or appropriate use of their respective radar equipment. There was a bad look-out. Courses were not plotted. Neither vessel was interpreting correctly information from its own radar. Had they done so, there would have been no collision. Thirdly, there was the failure, on hearing a fog signal of another vessel forward of the beam, continuing and repeated, to stop their engines and navigate with caution. This was a breach of rr 16 (a) and 16(b) of the COLREGs and bad seamanship. These main faults leave the scales level.

The collision was in the main caused by faulty manoeuvres undertaken by both vessels on the strength of defective radar observations. The collision would not have happened if the ships had not been equipped with radar. Ironically, it was the helm action taken by both vessels that made it possible for them to collide. The helm action is the only basis on which any differentiation between the degrees of fault of the two vessels can be made. The successive alterations to starboard by the Crystal Jewel were in breach of good seamanship and disregarded the advice to be found in the Notice. The recommendations contained in that Notice conform with the standards of good seamanship: The Verena [1961] 2 Lloyd's Rep 127 (CA). That Notice recommended that if an alteration of course is to be made it should be such that it will be most easily possible for those on board the other vessel to appreciate it from the observations on their own radar. The alterations made by the Crystal Jewel 'contributed to bringing the Crystal Jewel into a position fraught with great danger to both ships': The British Aviator 413. Meanwhile, the last-minute hard-a-port helm action of the British Aviator was 'seriously negligent and an important contributory cause of the collision': The British Aviator 414.

Apportionment of liability for collisions at sea is dealt with by s 1 of the Maritime Conventions Act 1911 (UK) (MCA), which says:

Where, by the fault of two or more vessels, damage or loss is caused to one or more of those vessels, to their cargoes or freight, or to any property on board, the liability to make good the damage or loss shall be in proportion to the degree in which each vessel was in fault:

Then follow certain provisos, proviso (a) being:

Provided that-

(a) if, having regard to all the circumstances of the case, it is not possible to establish different degrees of fault, the liability shall be apportioned equally ...

Subsequent decisions have thrown more light on the meaning of the proviso, but it is only necessary to refer to the early case of The Peter Benoit (1915) 114 LT 147 (HL). In order to allocate different levels of blameworthiness between two vessels, there must be 'a clear preponderance of culpability' on one as compared with the other: The Peter Benoit 151. In assessing degrees of fault, regard must be had both to the blameworthiness of the conduct alleged and also to its causative potency as a factor contributing to the collision and damage: The Peter Benoit 152. The Peter Benoit 152 stated that:

The conclusion that it is possible to establish different degrees of fault must be a conclusion proved by evidence, judicially arrived at, and sufficiently made out. Conjecture will not do: a general leaning in favour of one ship rather than of the other will not do: sympathy for one of the wrongdoers, too indefinite to be supported by a reasoned judgment, will not do. The question is not answered by deciding who was the first wrongdoer, nor even of necessity who was the last. The Act says, 'having regard to all the circumstances of the case.' Attention must be paid not only to the actual time of the collision and the manoeuvres of the ships when about to collide, but to their prior movements and opportunities, their acts, and omissions. Matters which are only introductory, even though they preceded the collision by but a short time, are not really circumstances of the case but only its antecedents, and they should not directly affect the result. As Pickford, L.J. observes: '"The liability to make good the damage or loss shall be in proportion to the degree in which each vessel was in fault." That must be in fault as regards the collision. If [it] was in fault in other ways, which had no effect on the collision, that is not a matter to be taken into consideration.'

The Court disagreed with the respondents' argument that the real meaning of the proviso is that a court must resort to equality only in cases where the paucity or uncertainty of the evidence does make it actually 'impossible' to establish different degrees of fault. The respondents' argument is contrary to the way in which the proviso has habitually been applied over the years. In plain language what the proviso means is that, whatever the ground may be for attributing different degrees of fault, it must be a ground which is proved by cogent evidence, and if that is not so, then the old admiralty rule of equal division continues to prevail.

When the court below has decided on the apportionment of blame, normally an appellate court would only interfere if it thinks that the court below has been guilty of some error of principle or has taken a wrong view of some of the material facts: The Peter Benoit; The Macgregor [1943] AC 197 (HL). In The Macgregor 200, Lord Wright quoted Scrutton LJ in The Karamea [1921] P 76 (CA) 89 as follows:

If the Court of Appeal agrees with the findings of fact and law of the learned judge below, and the only difference is that it attaches more importance to a particular fact than he did, it would require an extremely strong case to alter the proportions of blame which the learned judge below has attributed to the ships.

Lord Wright went on to say that

the judge's apportionment might not be interfered with by an appellate court, but I do repeat that it would require a very strong case to justify any such review of or interference with this matter of apportionment where the same view is taken of the law and the facts. It is a question of the degree of fault, depending on a trained and expert judgment considering all the circumstances, and it is different in essence from a mere finding of fact in the ordinary sense. It is a question, not of principle or of positive findings of fact or law, but of proportion, of balance and relative emphasis, and of weighing different considerations. It involves an individual choice or discretion, as to which there may well be differences of opinion by different minds. It is for that reason, I think, that an appellate court has been warned against interfering, save in very exceptional circumstances, with the judge's apportionment.

The respondents argued that all that the Court was being asked to do was to attach greater weight to the helm action of one of the vessels than was attached by Cairns J below. The respondents argued that, even if the Court were to think that the Crystal Jewel's helm action was more blameworthy than Cairns J seemed to think, that would be only a matter of individual opinion, with which, having regard to The Macgregor 200, an appellate court is not entitled to interfere. The Court rejected this argument. This is a case in which Cairns J's apportionment of blame was properly open to review. Cairns J never did appreciate the seriousness of the fault of the Crystal Jewel in altering course to starboard. Cairns J took a wrong view of the facts. It was not a mere expression of individual opinion.

The passage in which Cairns J criticised the helm action of the Crystal Jewel was a masterpiece of understatement. Based on evidence given on behalf of the Crystal Jewel, but for the Crystal Jewel's successive alterations to starboard, both vessels would have safely passed each other. The Court referred to a very useful diagram plotted by the appellants' advisers that was undisputed by the respondents. Those alterations led to 'a close quarters situation'. The Crystal Jewel's fault in starboarding has much greater significance than was attributed to it by Cairns J. Not only was it more blameworthy than Cairns J appears to have thought it, but also that as a matter of causation it had more effect in bringing about this collision than any other single factor. In the last two minutes before the collision, the fault on the part of the Crystal Jewel was not a lesser fault than the British Aviator's turn to port. Cairns J did not have any good ground for differentiating between these two vessels.

Sellers LJ added that the proviso would apply where the evidence was too meagre, vague, or uncertain to enable facts to be found on which to assess different degrees of fault, although of course it must be sufficient to establish negligence and liability. There may be other cases where the facts can be found with sufficient reliability but on those facts it is not possible to establish different degrees of fault. Where negligence contributing to the collision is found or admitted, then the burden is on the one side or the other to justify a departure from a decision of both being equally to blame. There must be a preponderance of fault on the one side to disturb the balance. Only fault which affected the collision must be assessed, and fault must be measured by blameworthiness, causation and by the damage caused by the negligent conduct. The decision calls for an appreciation and a comparative assessment of those ingredients of the respective faults found against each party.

The Court must approach these respective helm actions and a consideration of their importance and an assessment of their effect in the light of what was said in The Macgregor and other cases. The most apt passage is Lord Wright's citation (The Macgregor 200) of Scrutton LJ in The Karamea. The Court might on those directions have been persuaded to leave the matter where it is and not disturb Cairns J's decision on the gravity of these matters respectively. But the issue is the heart of this dispute between the parties, and the amount involved is considerable. The Court has a duty to review and, in this case, vary the decision.

Cairns J did not have the advantage of a diagram plotting the courses from the collision back to the period of the vessels sighting each other and admitted 'some uncertainty as to how it came about': The British Aviator [1964] 2 Lloyd's Rep 403, 407. In contrast, the Court had such a diagram, which the parties did not dispute. Without the courses being ascertained with some degree of reliability it would be difficult to appreciate the effect of the courses sufficiently to apportion liability. The diagram would have provided clarity on the gravity of the Crystal Jewel's conduct to Cairns J.