Appeal from the District Court of the United States for the Southern District of New York.
Before L. HAND, AUGUSTUS N. HAND, and CHASE, Circuit Judges.
L. HAND, Circuit Judge. At just ten twenty-four p. m. on the night of September 25, 1925, the steamer, City of Rome, bound from Savannah to Boston, collided with and sank the submarine S.51, then engaged upon an "availability" test, with the loss of all but three of her officers and crew of thirty-six men. This occurred about twelve miles east of Block Island on a clear night, the vessels being within sight of each other for over twenty minutes. The City of Rome had been off the South Light of Block Island at nine eighteen, and the distance between her position at that time and the spot where the wreck was found was eleven and six-tenths miles. Her average speed may thus be determined exactly; it was ten and forty-five hundredths miles an hour; her course was east, northeast. The survivors of the submarine, being below and asleep, could not say what was her course and speed, so that these must be determined wholly by inference.
At two minutes after ten the Rome's lookout reported a white light four points on her starboard bow which he estimated to be at a distance of between five and six miles; the third officer, who was on the bridge, also saw it and thought it about three or four miles off. The master came upon duty shortly thereafter, looked at it through his night glasses and estimated it at five or six. The Rome kept the light under observation and soon saw that its bearing did not change. After fifteen minutes they concluded that it had become brighter, its wake being seen upon the water. Thereupon the master, supposing that he was overtaking a vessel bound in the same general direction, eased his course to port without signal. At this time he was about two hundred yards away and a very short time thereafter, which he puts at only twenty seconds, a red light appeared close beside the white. The Rome at once hard aported, blew a backing signal, and gave the order full speed astern, but so close to collision that it was not executed until after the vessels struck. The angle of collision was variously estimated at between fifty degrees and seventy-two degrees; the stem of the Rome struck the port side of the submarine about thirty feet forward of her bridge and did such damage that she sank two minutes later.
The submarine carried her mast light at the forward end of her "chariot" bridge, eleven feet above her hull; though she had a beam of over twenty-five feet. About three feet abaft this light and seven and a half above the hull was her port running light. It had no proper fore and aft screen, but was apparently so screened as to be shut off, like the mast light, at an angle of two points abaft the beam. She also carried a stern light set in the after part of the bridge more than a hundred feet from the stern. The evidence as to the screening of this light is not clear, but it is probable, if not certain, that it showed over an arc of more than twelve points, though by how much it is impossible to say. If so, there was a zone, less than two points abaft her beam, where all three lights were visible at the same time. The submarine had a maximum speed of between eleven and twelve miles an hour; she would, unless some engine trouble had developed, keep at that speed during her surface test of twelve hours. Nothing is known of her movements for some time before the collision; the drowned officers whose representatives are claimants in the limitation proceeding were not however on duty.
When the Rome saw the white light four points on her starboard bow, we agree that she was justified, indeed required, to understand that it was the stern light of a vessel under way. As the bearing remained constant, her only conclusion must be that this vessel was either on a course converging upon her own by two points or less, that she was on a parallel course at the same speed, or that she was diverging at greater speed. By proper attention she should have observed the approach of the light, as indeed she eventually did, though not in season as it turned out. Being an overtaking vessel her duty was to keep out of the way and to avoid the risk of collision. If in doubt as to whether the other ship was converging or parallel, she was bound to assume the first; ambiguities must be resolved in favor of safety.
It is undoubtedly a difficult matter to ascertain the distance of a light upon the water, but that very difficulty should be the measure of the necessary caution. The Rome had twenty minutes within which to observe the light, which was in fact all the time coming nearer; we do not see how she can be absolved from responsibility for failing to notice its approach before she in fact did so. The only question open is whether she delayed action too long. Two masters of very wide experience testified that in such a situation they would give such a light a berth of a full mile, and that it was hazardous to get so close that you could see the wake of the light upon the water. A third said that he considered a berth of a hundred yards enough and that it was safe to wait till you could see the wake of the light. Without necessarily accepting the larger estimate as essential to safety, it seems to us not only that the smaller is far too little, but that the Rome approached too near. A master in such a position may indeed assume that the other vessel's lights are in order, but he must also consider that she may be upon the most converging course permissible, two points. In the case at bar the Rome was to suppose that her own speed was greater, because she was apparently overhauling the other ship. If both chanced to be in fact upon courses converging at two points, the Rome was only therefore about five hundred yards from collision when the ships were two hundred yards apart. As her speed was more than fifteen feet a second, she had only about a minute and a half to avoid collision. A hard over helm does not at once affect the bow, but at first throws the stern in upon the course on the side which the vessel is to leave; it is only after she has advanced from two to three lengths that her stern finally leaves her original course. The Rome was three hundred and fifty feet long and even under a hard astarboard helm her stern would have been within say six or seven hundred feet of contact with the other vessel before it left the course. She had the open sea upon either hand and she accepted an unnecessary risk of collision in getting so near. Indeed, she did not even hard astarboard when she did act, so that for about half a minute more she was bearing down upon the other with only a slight change of course. This left her still less leeway; how much no one can say. A constant bearing is a sure sign of danger, made so by the rules at their very outset; a danger signal which should put every mariner on guard. To be sure, it may not call for an immediate change of course or speed, but it must always be remembered that it is the risk of collision, not the collision itself, that masters must avoid. The Maverick, 84 F. 906 (C.C.A. 2); Wilder's S.S. Co. v. Low, 112 F. 161 (C.C.A. 9); Le Boyteaux, Rules of the Road at Sea, 106. We cannot absolve the Rome from shaving the submarine too close.
A second, though a lesser fault, was her failure to blow when she starboarded. It is indeed very doubtful whether this would have altogether avoided collision, but the result might have been very different. We are to suppose that the submarine must have seen the Rome steadily bearing down upon her and have been waiting as the holding on vessel for some action by her. As we shall show, she was on a crossing course, though at what angle it is impossible to know. Had this been only fifty degrees a hard aport wheel by the submarine when the Rome starboarded would have substantially changed her heading. It is the natural response of every master and would probably have been that of the submarine had she learned that the Rome was bearing away to cross her bows. How much it would have changed the result nobody can tell. If the courses were at right angles or more, a hard astarboard helm by the submarine when the Rome starboarded accompanied by immediate backing, might have at least brought the point of collision further forward or advised the Rome against a hard aport helm. True, this is all pure speculation, but the doubts must be resolved against the Rome, which violated the rule; though she was already close aboard it is impossible to say beyond reasonable doubt that this fault did not contribute.
As to the submarine, her fault depends upon her course, since upon that in turn depends the contribution to the collision of the unlawful position of her mast light. The evidence as to the position of the red light when it appeared is not the most satisfactory, but according to the better part of it, it was to the right of the white light, and three lights never appeared. To this the testimony of Boll, the engineer, is an exception, who somewhat faintly suggests that he saw two white lights and one red. He was further astern than the bridge and the submarine's stern light, if not properly screened, might have shown along with the mast light when she hard aported as she did. It therefore seems most probable that it was the masthead light which had been visible all the time, as the witnesses asserted, who were sure that the white light which was to the left of the red had not been substituted at the last moment in place of any other. Such testimony is indeed not conclusive, but the District Judge, who saw some of these witnesses, relied upon it. It must certainly count in the scale.
Again, when first observed the white light appeared to be some four or five miles away. As we have already said, it is hard to measure the distance of a light upon the water and the estimate is not to be taken very seriously, but, unless it can be disregarded in toto, it necessarily follows that it was the mast light which showed, for the stern light, if of regulation power, had a range of only two miles. Moreover, the stern light could not have been anything like five miles off, unless the vessels were on crossing courses. If the courses were converging, its distance would vary with the angle between them, the extreme limit being two points. Twenty minutes before the collision the Rome was about three and a half miles from the point of collision, and a light four points on her starboard bow was less than a mile and a half away. Perhaps so great an error was made, but the judgment of the experienced men on the bridge is not to be wholly ignored. Moreover, the speed of the submarine if on that course would not have been more than eight miles, which could not have been the case unless something had gone wrong with her engines. The smaller the angle of convergence, the greater the mistake as to the distance, and while in that case the speed of the submarine could be greater, even then we are obliged to take it as much less than the Rome's.
It is impossible accurately to determine the angle of collision. After the first impact the submarine swung side to side to the Rome and it is possible, if not probable, that this somewhat widened the cut, though one surveyor thought not. So far as we can tell the angle was between fifty and sixty degrees, but in any case more than twice two points, enough to make the situation a crossing case, if it properly indicated the courses. The United States argues that it did not, because of the Rome's hard aport helm, which began some thirty seconds before the impact. During that period she would travel about four hundred and fifty feet except for the substantial reduction of her speed due to her hard right rudder. If we assume that she went a length and a half, her heading would change not more than two points, or twentytwo degrees. This, however, ignores not only her reduction in speed, but her original starboarding, which at least put her on a swing which must be overcome and probably had thrown her off to some extent. Nor should we forget that the submarine was found with her own rudder hard right, showing that she had hard aported at some time before the collision, how long we cannot tell. It is pretty clear from all this that it was quite as likely that the angle of collision was less than that of the courses as that it was greater; and that at any rate it is most unlikely that the courses originally converged by as little as two points.
Thus it seems to us that the evidence by several approaches leads to the same conclusion, which is that it was a crossing case. The strength of the other argument is the Rome's failure to see the red light until the ships were so close together. While it is of course possible that the red light was not switched on until the Rome saw it, that appears to be a very dubious conclusion; it would have been so gross a fault of navigation. It is not necessary to assume so. Experiments in the Delaware River under more favorable conditions showed that the bright mast light masks the running light at distances greater than a mile and a half. A naval officer of unquestioned credibility had in one instance been unable to pick up the running lights of a submarine until she got half a mile off; a ship's master had failed to see them until within two hundred yards.It is not difficult to understand why this should be. The mast light, only four feet away, is much stronger and it is white; its rays would effectively obscure the red until the angle subtended by their distance became large enough upon the vessel's approach. Even at two hundred yards that angle is less than half a degree. Possibly the distance was greater at which the submarine's red light appeared but to us it does not seem incredible that it could be detected only at that distance. At any rate it seems to us impossible that it appeared from behind its screen when first observed. If so, the mast light appeared at the same time, and it was much brighter. Even if the stern light was then extinguished, which we do not believe, it is incredible that all this would not have remained in the memory of the witnesses. We are not therefore disposed to differ from the learned District Judge, who held that the submarine was upon a crossing course.If so, the acknowledged fault in the position of the mast light was a contributing cause of the collision.
There remains only the contention that the submarine was not subject to the ordinary rules of the road. It does not appear that it was impossible for her to comply with these, but it would make no difference if it did. The International Rules (section 61 et seq. title 33, U.S.C. [33 USCA § 61 et seq.]), apply to "all public and private vessels of the United States." Article 13 (section 83, title 33, U.S.C. [33 USCA § 83]) provides that "nothing in these rules shall interfere with the operation of any special rules made by the Government of any nation with respect to additional station and signal-lights for two or more ships of war or for vessels sailing under convoy." It is thus apparent that the rules regulating lights were meant to apply to ships of war; article thirteen would be conclusive if the preamble alone were not enough. If unfortunately it is impossible to equip submarines properly, they must take their chances until some provision has been made for them by law. We have no power to dispense with the statute, nor indeed has the Navy. As they now sail they are unfortunately a menace to other shipping and to their own crews, as this unhappy collision so tragically illustrates. We cannot say that they are not to be judged by the same standard as private persons. The safety of navigation depends upon uniformity; only so can reliance be placed upon what masters see at night.
We shall not take up the question of marshalling the proceeds of the Rome as between the United States and the private claimants. This is properly a matter for the commissioner appointed by the interlocutory decree. All we decide for the present is that both vessels were correctly found at fault, and that the submarine should recover half damages against the fund and the claimants ...