These appeals from a final decree in admiralty require us to untangle a near-collision case in which respondent's tanker, although it managed to miss the three other vessels on the scene, struck and damaged libellant's submarine cable.1

Libellant runs two of its telephone cables over the floor of the Delaware River between Deepwater Point, New Jersey, and Pigeon Point, Delaware. The south cable having failed, libellant began repair operations on April 19, 1949. In order to carry out the repairs, the services of three vessels were engaged: the tug Adriatic, the barge Acco (under bareboat charter), and the derrick barge Contractor. The tug Adriatic was owned by respondent Martug Towing Company, and her services were obtained through respondent P. F. Martin, Inc., her operating agent. She was to perform towage and transportation services, while the Contractor was to lift the cable and lay it across her own and the deck of the Acco so that libellant's repair crew could work on it.

At about 7:30 p. m. on April 20, 1949, the Acco was lying broadside in the middle of the upbound fairway in the Cherry Island Range, with the uplifted cable chained along her deck. From the deck of the Acco, the cable dipped back into the water and touched bottom about one hundred feet to the east. The Contractor was lying out of the channel about six hundred feet east of the Acco, and the cable rose from the bottom and was chained across her deck. From the Acco and the Contractor, the cable sloped to the bottom and exended to the Delaware and New Jersey shores, respectively. Neither the Acco nor the Contractor were anchored because libellant did not allow repair vessels to drop their anchors for fear of fouling the other cable on the bottom, but both vessels were kept fairly stationary by the cable on their decks. The Adriatic was north of, and had her bow against, the Acco and used her engines intermittently in order to counteract the effect of the flooding tide and thus take the tension off the cable. She was not anchored but had a line fast to the Acco.

The district court found that these three vessels were brilliantly illuminated. Finding of Fact No. 6 reads, in part, as follows:

"The `Acco' had twelve red kerosene lanterns around her guard rail, one red lantern on top of a nine foot cable reel, and two five hundred watt floodlights shining on her deck. The `Contractor' had three red lights in a vertical line in her rigging, with the lowest approximately twenty-five feet above her deck-house, two floodlights shining on her deck and deck lights around her house. The `Adriatic' had one white mast light, her red and green running lights and lights around her deck."

The loaded tanker W. C. Yeager, owned and operated by respondent The Atlantic Refining Company, was proceeding upriver at about 13½ knots. She went to half speed when she ported at the intersection of Deepwater Point and Cherry Island Ranges. At this point she was first sighted by those aboard the Adriatic.2 Naturally, the versions of the witnesses for the vessels differ. Libellant's witnesses and those of respondent P. F. Martin, Inc., testified that shortly after the Yeager was first sighted, the Adriatic blew a combination signal: four short blasts and two long ones, intending to indicate danger on the Adriatic's port side. Hearing no answer from the Yeager, the Adriatic repeated the combination signal six or seven times. The Yeager blew one blast and continued ahead. Those on the Acco, believing a collision to be imminent, scrambled aboard the Adriatic and cast off her line so that she drifted free of the Acco. The Yeager, with the Acco on her port side, cleared the latter with about fifteen to twenty feet of open water, but she struck the cable which sloped off the Acco, and the impact whipped the Acco around so that she scraped along the tanker's side. Substantially, this was the version which the district court accepted.

The captain of the Yeager, on the other hand, testified that he had seen the lights of the repair vessels even before he reached the intersection of the Deepwater Point and Cherry Island Ranges. The Yeager's captain, first mate, and helmsman all testified that the Adriatic was showing white lights in a vertical line and a red running light. Consequently, to them she looked like a tug and tow3 proceeding diagonally across the river from New Jersey to Delaware. They planned, therefore, to go under her stern. To carry out this intention, the Yeager, when she made the Cherry Island Range, steadied on a course of 20° rather than 17°, the correct course for that range. When about one-half mile from the flotilla, the Yeager sounded one blast and went to slow ahead. Those on the Yeager heard no response. Shortly thereafter, the Adriatic sounded a danger signal and turned out her running lights. This was the first indication to the Yeager that the Adriatic was not a moving vessel. Immediately upon the Adriatic's danger signal, the Yeager's engines went to stop and her rudder hard left. Having lost steerageway, however, she did not respond to the helm. Therefore, her engines were put at full astern, she sounded three blasts to indicate that maneuver,4 and she went to hard right rudder. She was unable to break her headway, and she drifted by the Acco and struck the cable, as described above.

The channel at the point of mishap is about eight hundred feet wide. The night was dark but clear.

It is admitted that neither the Acco nor the Adriatic displayed the three red lights in a vertical line, required by the Pilot Rules to be shown on vessels engaged in cable laying.5

On the points where the testimony was conflicting, the district court found that the Adriatic was the first to blow; that the Yeager ignored those signals which were ample warning of danger on the Adriatic's port side; and that the Yeager would have had sufficient time to avoid the cable if she had stopped or passed to starboard of the Adriatic. As to the absence of the three red lights on the Acco and Adriatic, the court concluded that there was already such a multitude of lights on the two vessels that three more could not possibly have changed the result. Finally, the decree dismissed the libel as against P. F. Martin, Inc., and held The Atlantic Refining Company solely liable for the damage to the cable. These appeals followed, Atlantic being the appellant in No. 10,867 and libellant having taken a protective appeal in No. 10,877 from the dismissal as to P. F. Martin, Inc.

Jurisdiction in admiralty is based on the Admiralty Extension Act.6

There are two major phases to the case: The question of Atlantic's liability to libellant and, depending upon the disposition of that problem, the question of the liability of P. F. Martin, Inc. We shall first treat libellant's claim against Atlantic.

The basic findings upon which Atlantic was held solely liable are that the Yeager disregarded the Adriatic's danger signals and did not stop although she had ample time to do so. These findings have support in the evidence, and we accept them.7 We cannot accede, however, to the conclusion drawn from them. The Yeager owed a duty to the Acco and the Adriatic to observe the rules and to navigate prudently. The difficulty in sustaining libellant's position is that the undisputed facts show that the Yeager fulfilled that duty. She did not collide with the Acco or Adriatic.8 She struck the cable, of whose presence she was unaware because of the failure by the Adriatic and Acco to display the requisite lights. The district court deemed that failure immaterial, reasoning that, because the flotilla was already lit up like a Christmas tree, the absence of the three red lights called for by the Pilot Rules could not have contributed to the accident. We would agree, if the only purpose of the three red lights was to provide illumination. The other lights that were showing told the Yeager that the flotilla was there, and, as to it, the Yeager navigated successfully. Had the required lights been displayed, the Yeager would have been told, not only that the flotilla was there, but also what it was doing there. Only then do we think the Yeager had a duty to the cable which, in the absence of the warning lights, was really an obstruction to navigation. By no means do we say that the cable was a trespasser in the river. It had a right to be there. But, correlatively, the Yeager had a right to be told that it was there. The Pilot Rules have established a specific light signal to indicate to an approaching vessel that the lighted vessel is engaged in underwater cable work or, at least, that it might have cables or pipes extending out under water which would embarrass the navigation of the approaching vessel. When not inconsistent with the statutory rules, the Pilot Rules have the force of law.9

In our view of the case, it is irrelevant that the Yeager failed to stop in the face of a danger signal, that she crossed the Adriatic's signal, and that she was also negligently navigated, because she did not collide with the Adriatic or Acco. The lights on the flotilla and the whistles by the Adriatic were successful. Because of them the Yeager was enabled to avoid collision. Maybe it was more a matter of good luck than good navigation but, in view of the final result, that should not affect the application of the classic remark, "Proof of negligence in the air, so to speak, will not do."10 The various acts and omissions charged against the Yeager were negligent toward the Acco and Adriatic and not towards libellant's cable. We are not to be understood as holding that a vessel must actually collide with another before she can be held. There are many cases holding a vessel in fault and liable even though not herself in collision.11 But in those cases the noncolliding vessel was negligent toward the colliding and damaged vessel. That is not our case. Here we have the unforeseeable libellant, a maritime instance of the landlubber's unforeseeable plaintiff.12 The scope of the Yeager's duty is not to be determined by what we know now but by what those aboard her knew or should have known as they approached the flotilla. They knew that there were one or more vessels ahead of them and that any unseamanlike navigation on their part would subject those vessels to an unreasonable risk of harm. Thus, they had a duty to navigate so as not to create such risk, and they did. They could not possibly have foreseen that poor navigation would have subjected libellant's cable to an unreasonable risk of harm because they neither knew nor should have known that the cable was there. Thus, as to the cable, there was no duty and, consequently, could be no negligence.13 Had the flotilla displayed the required three red lights in a vertical line, the risk of harm would have been foreseeable, thus giving birth to a duty of prudent navigation toward the cable.

Libellant makes much of the fact that the location is marked on the chart as a cable area, contending that this is notice enough that the flotilla was engaged in cable work. We are not prepared to say that the marking on the chart to the effect that there are submarine cables there is sufficient notice that any vessels in that area have raised the cables to their decks, when those vessels are not showing the required lights. It is admitted that libellant made no effort to publish the fact of their repair work in the "Weekly Notice to Mariners," even though it knew three weeks beforehand that they were going to work on the cable. But, says libellant, the Contractor did display the three red lights in a vertical line, thus giving notice of cable work. The Contractor was, however, six hundred feet away from the flotilla — much too far away to help the Acco and Adriatic. Proper lights on one vessel do not satisfy the duty of every other vessel in the river. Furthermore, the rule says that the lights shall be displayed on cable-laying "vessels," not one vessel out of three.

Respondent Martin interjects the following contention, which bears on its liability to libellant, but must be discussed here also: The Pilot Rules require three red lights in a vertical line for cable-laying vessels which are "moored or anchored." The argument is that the Adriatic was "under way"14 and not moored or anchored and, thus, had no duty to show the three red lights, but was obliged to show her running lights. We disagree. True, the Adriatic was not anchored or aground. She was, however, made fast to the Acco which was, in turn, made fast to the shore because of the cable chained to her deck. The cable extended from shore to shore and, hence, the Acco was just as effectively moored as if she were tied up at the bank, and the Adriatic was fast to her. Therefore, the Adriatic was moored even though she used her engines now and then to thrust against the flooding tide. She should have displayed the three red lights and not her running lights. The Charles C. Lister, 2 Cir., 1910, 182 F. 988, cited by Martin, does not stand in the way of our holding on this issue. There the tug was held to be under way and thus properly showing her running lights even though she merely kept the tow stationary and made no way over the ground. There was no line or cable to shore, however, as we have here. We have found no case holding a vessel to be under way when it is tied to the shore, as the Acco and Adriatic were here.15

The Adriatic would justify her showing of running lights and failure to show lights indicating cable work by the fact that her duties called for her to run back and forth between the Acco and Contractor and to and from the Marine Terminal in Wilmington. That may be, and, while doing so, she was properly showing her running lights. The testimony, however, had her tied up to the Acco for from ten minutes to an hour before the accident.16 Even supposing the shorter period of time to be the fact, it is certainly not unreasonable to require her to change her lights in ten minutes.

Having concluded that the Yeager was not at fault, we must now resolve the dispute between libellant and respondent Martin.

Tied up to the Acco, as she was, the Adriatic thus became a vessel "engaged in laying cables" and was, therefore, required to show the three red lights. Granting that she ran back and forth between the barges and also to the Marine Terminal in Wilmington and was obliged to display running lights then, she also attended the Acco by making fast to her, at which times she needed the three red lights. She contracted to perform just that dual service and, hence, assumed the obligation to apparel and outfit herself properly so that she could carry out both phases prudently. Consequently, she cannot escape because she did not have proper lights or sockets.

We are convinced also that the Acco herself should have displayed the three red lights. Libellant would avoid this conclusion by contending that it is not a maritime concern, had only cable-repair personnel aboard the Acco, and left all navigational decisions to the captain of the Adriatic. Those assertions appear to be true, but libellant had bareboat chartered the Acco and was, thus, her owner pro hac vice. Libellant became, to that extent, maritime. Furthermore, when it became apparent that the work could not be completed before dark and discussion was had between libellant's supervisor and the captain of the Adriatic as to what warning signals should be displayed, the latter informed libellant's supervisor of the proper precaution. The supervisor, however, decided not to erect a makeshift mast on the Acco from which to hang the three red lights because that would interfere with "snaking the cable" on the Acco's deck. We think libellant, therefore, assumed responsibility for the failure to show the lights. This made the Adriatic's compliance more imperative, but did not wholly relieve libellant. Thus, both the Acco and the Adriatic should have displayed the lights.

We come now to causation. As stated above, the district court felt that the failure to show the three red lights was not causative because many other lights were shown and were seen. In the first phase of the case, we did not reach the issue of causation, holding that the absence of the lights related more properly to the element of duty on the part of the Yeager. We must meet that issue here.

It is axiomatic that we will not disturb findings of fact in admiralty when those findings are based on testimony given by witnesses whom the district court saw and heard. There are several factors present in this case which lessen our usual reluctance to disagree with the trial court. In the first place, as to one or two rather important issues, the court made no finding of fact with which we could agree or disagree. Furthermore, the testimony of the Yeager's captain and first mate (the very witnesses whose testimony would have to be weighed in order to decide the issues upon which there are no findings) was taken by deposition, so that we are in as good position to assay this testimony as was the district court. Moreover, the court's finding that the absence of the three red lights was not a causal factor was a conclusion of ultimate fact, and, while we disagree, our disagreement goes only to that conclusion of ultimate fact. We accept the basic facts as found by the trial court.

There is a finding, which is well supported, that the Acco had twelve red kerosene lanterns around her rail, a like lantern on top of a nine foot cable reel, and two five-hundred watt floodlights shining down on her deck. The Adriatic was showing a white light in her mast and her running lights. Of all these, the Yeager's witnesses said they saw only the Adriatic's mast light and her red running light and that they navigated accordingly. Libellant, in No. 10,867, and respondent Martin argue that, the Yeager's witnesses having seen only two of these many lights, it would be absurd to say that they would have seen the three red lights had they been displayed. We cannot say that they would not have seen and heeded those lights had they been shown. They saw the lights which were meant as navigational signals and which were apparently of the prescribed strength. The only ones they did not see were not navigational signals at all but were kerosene lanterns, twelve of the thirteen being only one foot above the Acco's deck. We think it unlikely that kerosene lanterns, placed as these were, would throw as strong a light as the lights prescribed by the rules. Lights indicating cable work are required to have a visibility of at least two miles on a dark night with a clear atmosphere.17 Thus, had those lights been shown, the Yeager could have seen them from about the intersection of the Deepwater Point and Cherry Island Ranges. Certainly, those on the Yeager saw the flotilla long before that, but, we think, with the additional information which the three red lights would have given her, she would have been enabled to avoid the cable. Because the Yeager did not see kerosene lanterns, it does not follow that she would not have seen lights which must be visible for two miles. In fact, the contrary would seem more logical.

Of course, we recognize the rule relied on by the district court to the effect that the failure to carry proper lights will not render a vessel at fault when she was in fact seen by the other in time to avoid collision and the other thereafter fails to take proper avoiding action. We think that rule is inapplicable here. It would be dispositive of this case had the Yeager collided with the flotilla, which, though improperly lighted, was in fact seen in time for the Yeager to avoid it. The Yeager struck the cable which was not seen at all. Thus, we think that the failure to show lights disclosing the presence of the cable was a cause of the accident. This would be true even without the rather stringent burden cast upon libellant and Martin by The Pennsylvania, 1873, 19 Wall. 125, 22 L. Ed. 148, to show that their violation of the rule could not have been a contributory cause of the collision with the cable.

Even beyond this, the Adriatic showed misleading lights in violation of Article I of the Inland Rules.18 She was not under way, but her lights misled the Yeager into believing that she was. The district court found that the Adriatic was showing her running lights, but made no finding as to how long before the accident those lights were showing or whether they actually misled the Yeager, as her witnesses stated. In any event, the testimony shows that they were showing long enough to cause the Yeager to believe that the Adriatic was under way and that the former could easily pass under her stern after she cleared to the Delaware side of the river. Therefore, the Adriatic's misleading lights were also a causal factor.

In summary, we hold that the Yeager was not at fault because she owed no duty of prudent navigation to the cable; the Adriatic was at fault for showing misleading lights and failing to show the lights indicating cable work; the Acco was at fault for failing to show the required lights; these faults were causative; therefore, libellant and Martin must share the damages.

The decree of the district court will be vacated. The libel will be dismissed as to The Atlantic Refining Company, with its costs to be paid, one-half by libellant and one-quarter each by P. F. Martin, Inc., and Martug Towing Company. A decree will be entered in favor of libellant and against P. F. Martin, Inc., and Martug Towing Company in the amount of $1,592.68,19 each of these parties to bear its own costs.

HASTIE, Circuit Judge (dissenting in part).

The court has concluded, correctly I think, that blameworthy conduct of three vessels, the Adriatic, the Acco and the Yeager, combined to cause one of them, the Yeager, to collide with and damage libellant's cable. But I think the court is mistaken in its view that the damage in question, though a consequence of faults including the Yeager's failure to navigate properly, is not any injury for which the Yeager should share legal liability with the Adriatic and the Acco.

The court reaches its result by applying the familiar Palsgraf doctrine1 to maritime tort. I too would apply the Palsgraf doctrine to maritime causes. Indeed, The Eugene F. Moran, 1909, 212 U.S. 466, 29 S. Ct. 339, 53 L. Ed. 600, seems to require that this be done. But that doctrine leads me to a different result in this case.

In the Palsgraf case itself Chief Judge Cardozo thus explained the conception limiting actionable negligence:

"Negligent the act is * * * only because the eye of vigilance perceives the risk of damage. * * * The risk reasonably to be perceived [which] defines the duty to be obeyed * * * is risk to another or to others within the range of apprehension." 248 N.Y. at page 344, 162 N.E. at page 100.

After the doctrine is stated the problem remains in each case to define "the range of apprehension." Applying this conception to our facts the court is saying that when the Yeager proceeded on course without diminishing speed or determining the situation ahead in disregard of the warning represented by four short blasts of the Adriatic's whistle — which it admittedly heard — the reasonable range of apprehension was narrowly restricted to the possibility of damaging the signalling tug and tow to the exclusion of anything else technically distinct from the signalling vessel, however near to it physically and however intimately connected with its presence and activity. With this I disagree.

I would say there was negligence as to the signalling vessel, its gear, its tow, any launch it might have alongside, any member of its crew who might be taking a dip alongside, or any cable or other structure, whether technically a part of the vessel or not, alongside of or extending out from it, with the limitation that these things all have some significant connection with the signalling vessel and be physically close enough to it so as to be within the area which would be avoided by correct and prudent navigation of a passing ship.

Here the Yeager followed up its disregard of a signal, which had put it on notice of possible danger ahead, with an actual passing within 20 feet of the Acco, the very kind of close shaving which most obviously constitutes a serious fault in navigation. Once it appears that the Yeager thus wrongfully ignored a signal and passed much too near to the Acco, I have no difficulty in holding it responsible for colliding with something closely connected with and extending out from the Acco, without regard to the undoubted distinctions in fact and in law between the ship itself and a submarine cable it has raised and placed across its deck for repair. Fundamentally, I think we are trying to define and apply a policy limitation on the legal concept of negligence. We are trying to avoid taxing even the person whose conduct has been blameworthy with liability for hurtful consequences beyond what seems fair and socially desirable. I think my solution of the present problem does that and at the same time gives deserved protection not only to a ship itself but also to things around it and related to it.

I would require the interests represented by the Yeager, the Adriatic and the Acco to share the burden of this mishap equally.