Nautical Challenge Ltd v Evergreen Marine (UK) Ltd

On 11 February 2015 a collision occurred between a laden VLCC, ALEXANDRA 1, owned by Nautical Challenge Ltd., a company registered in the Marshall Islands, and a laden container vessel, EVER SMART, owned by Evergreen Marine (UK) Ltd., a company registered in the UK. The collision between these two large vessels occurred just outside the dredged channel by which vessels enter and exit the port of Jebel Ali in the United Arab Emirates. The collision was at night but there were clear skies and good visibility. The damage suffered by ALEXANDRA 1 (in way of her bows) is said to amount to over US$32 million. The lesser damage suffered by EVER SMART (to her port bow) is said to amount to almost US$4 million.

Both vessels carried a Voyage Data Recorder (devices which record navigational data of use in accident investigations, see Marsden and Gault, Collisions at Sea 14^thed. Appendix 18). Accordingly the parties were required at the Case Management Conference to exercise their best endeavours to agree the track of each vessel. The parties were able to do so. In consequence the parties have also been able to agree a schedule showing the course, heading and speed of the vessels from C-26 until collision. The schedule also shows the engine orders (taken from the engine loggers of each vessel) and the helm orders issued on each vessel (taken from the audio record of what was said on the bridge of each vessel). In the result, save for one disputed helm order, there is little factual dispute concerning the navigation of either vessel (apart from issues relating to the perception of those on the bridge of one vessel of what the other vessel was doing). This is now common in collision actions; see Samco Europe and MSC Prestige [2011] 2 Lloyd’s Rep. 579 at paragraph 2 and Nordlake and Seaeagle [2016] 1 Lloyd’s Rep. 656 at paragraph 7. The prevalence of electronic data which records the navigation of each vessel has led the court, with the assistance of the Admiralty Bar and the Admiralty Solicitors Group, to propose (i) the early disclosure and inspection of such data so that the navigation of vessels in collision can be agreed at an early stage and (ii) in the event of there still being a dispute as to liability, the adoption of trial procedures designed to achieve a speedy and cost effective resolution of that dispute. The required changes to CPR 61 and PD 61 came into force on 28 February 2017; see the Civil Procedure (Amendment) Rules 2017 SI 95 of 2017 and the 88^th update to the Practice Directions. The effect of those changes are summarised on the court’s web site and will be included in Section N of the Admiralty and Commercial Courts Guide.

In this case, notwithstanding the agreed track of each vessel, there was a substantial dispute as to liability and in particular as to the relevance and applicability of the "crossing rule", rule 15 of the Collision Regulations. At trial there were short openings and no oral evidence was called. Counsel made oral submissions as to liability over 2 days. The trial was, however, disrupted by the need for the parties to agree a transcript and, where necessary, a translation of the audio record of what was said or heard on the bridge of each vessel. This ought to have been agreed long before the start of the trial. It is important for parties to inform the court at the case management conference of what electronic records there are so that appropriate orders can be made so as to ensure that there is an agreed interpretation of all relevant records before the commencement of the trial. Audio records from the bridge of a vessel are obviously relevant and important records. In the present case some translations were not agreed and so I was given two translations. However, it was not suggested that the sense or tenor of the two translations differed.

In the light of the electronic data recording the vessel’s navigation it is unnecessary to dwell upon the failure of the owners of ALEXANDRA 1 to retain her original working chart, course record trace, log or engine logger print out or upon the reasons why the owners of EVER SMART produced as their original working chart a different working chart from that which they had supplied to the MAIB. It is also unnecessary to dwell upon the account given by the master of ALEXANDRA 1 of his vessels’ navigation in his Letter of Protest which was in some respects untrue (as shown by the electronic data) or upon the account given by the master of EVER SMART in his log and statements of fact which was also in some respects untrue (as shown by the electronic data).

The navigation of both vessels has been helpfully illustrated by the parties. A copy of the agreed track of each vessel is annexed to this judgment. In addition the ALEXANDRA 1 interests prepared a video of radar screen shots taken from the VDR of EVER SMART and the EVER SMART interests prepared a number of videos using the data from the VDR of each vessel showing a bird’s eye view of the collision and also the view from the bridge of each vessel. Those videos, whilst interesting to view, did not contain any more information regarding the navigation of the vessels than was apparent from the agreed track and schedule of navigation.

ALEXANDRA 1 is a VLCC built in 1997 of 79,779 grt and 48,796 nrt, some 269 m.

On the bridge at the time of the collision were her Russian master, her Russian third officer as officer of the watch and her Georgian helmsman. The second officer, also Russian, joined them on the bridge shortly before the collision.

EVER SMART is a container ship built in 2005 of 75,246 grt and 39,564 nrt, some 299 m. in length and 42 m. in beam. She is powered by a Mitsubishi Sulzer engine developing 54,900kW through a fixed pitch, right handed propeller. She was at the time of the collision laden with 48,564 mt of containerised general cargo and was drawing 12.7 m. even keel. She was outbound from Jebel Ali. She has a laden full ahead manoeuvring speed of 15 knots at 52 rpm. Her full sea speed was stated in her particulars to be 25 knots at 105 rpm. She is equipped with a full range of modern navigational aids. Her manoeuvring speed at half ahead was 10.7 knots, at slow ahead 9 knots and at dead slow ahead 7 knots. She exhibited the correct masthead, side and stern lights.

On the bridge at the time of the collision were her Taiwanese master, her Filipino third officer as officer of the watch and the helmsman.

The dredged channel leading out from Jebel Ali lies on an axis of 315/135 degrees (true) and is about 8.5 nautical miles in length and slightly less than 2 cables in width. It is marked by lateral buoys, from buoys no.1 at the seaward end to buoys no. 12 at the outer breakwater, and is dredged to 17 meters. At the seaward end is the designated pilot boarding area, a circular area with a 1 nautical mile radius. The limit of the circular area as shown on the chart is about 3 cables beyond buoys no.1.

There is no dispute that the dredged channel was a narrow channel for the purposes of rule 9 of the Collision Regulations.

As already stated there were clear night skies and good visibility of 10-12 miles. There was a wind of force 3, said by ALEXANDRA 1 to be easterly and said by EVER SMART to be east north easterly. The parties are agreed that nothing turns on the precise direction of the wind. There was a SW setting current of about 1 knot according to ALEXANDRA 1 but no current was particularly noted by EVER SMART. The agreed track takes account of a SW setting current.

The collision occurred at 2342:22 on 11 February 2015 about 5 cables west north west of the no.1 buoys and so outside the dredged channel but within the pilot boarding area. EVER SMART had disembarked her pilot during the course of her passage along the dredged channel and ALEXANDRA 1 was waiting to embark that same pilot. The port bow of EVER SMART struck the starboard bow of ALEXANDRA 1 at an angle of about 40 degrees (on the agreed plot 42.7 degrees) leading aft on EVER SMART. At collision the speed of EVER SMART was 12.4 knots over the ground and the speed of ALEXANDRA 1 was 2.4 knots over the ground.

ALEXANDRA 1 had arrived off Jebel Ali anchorage at 0500 on 11 February 2015 when she anchored about 1.7 nautical miles from the western edge of the pilot boarding area. At about 1600 she raised her anchor and was instructed by Port Control to wait 2.5 miles from buoy no.1. At about 2118 Port Control told ALEXANDRA 1 that no pilot would be available before midnight. Her master decided to return to the anchorage rather than continue drifting. ALEXANDRA 1 anchored at about 2200 but at about the same time was instructed by Port Control that a pilot would be boarding at 2315 and that ALEXANDRA 1 should be “at buoy no.1” at that time. The master replied that at 2314 he would be “at buoy no.1”. By 2247 ALEXANDRA 1 was underway. At 2254 Port Control informed ALEXANDRA 1 that the pilot was on the outbound EVER SMART passing buoy no.12 and that EVER SMART would continue “up to buoy no.1”. Port Control advised that “once EVER SMART was clear then you can enter the channel.” The master replied that he was “coming close 1 nautical mile close to buoy No.1; will wait until another the vessel leave the channel, thank you.”

At 2315 ALEXANDRA 1 was within the pilot boarding area about 1.4 miles to the west north west of buoys no.1. The master of ALEXANDRA 1 said in his witness statement that he observed EVER SMART by radar when she was at about the no.8

The audio record on the bridge of ALEXANDRA 1 suggests that at about 2318 or C24 the master of ALEXANDRA 1 was irritated that he had been told that the pilot would board at 2315 when in fact EVER SMART was still proceeding along the channel in the vicinity of the no.6 buoys. At this time the engines of ALEXANDRA 1 were stopped and her helm was amidships. Her speed over the ground was 2.3 knots and her course made good was 126 degrees but her heading was 110 degrees. It is possible that at such a low speed and with an easterly wind her heading was falling off to port. At 2320 or C-22 the audio record on ALEXANDRA 1 records the words “Hard starboard” (in English), followed by “right full rudder" (in Russian). It was submitted on behalf of ALEXANDRA 1 that at this time her helm was put hard starboard and remained so until collision. This was not accepted by counsel on behalf of EVER SMART who pointed out that the words were spoken in a conversational tone and were followed by whistling. It was suggested that it was not a command. However, the heading was falling off to port and so it would make sense to put the helm hard starboard. It is difficult to identify a meaning for the recorded words apart from a helm order. I therefore accept that it is likely that there was a hard starboard helm order, notwithstanding that the heading of ALEXANDRA 1 in fact continued to fall off to port. However, the master makes no reference in his statement (or in any other document) to the helm of his vessel being at hard starboard for some 20 minutes before the collision which suggests that the helm change was no more than temporary. There is no audio record of a helm “amidships” order but I consider that there must have been one sometime after C-22. The helmsman, in a statement dated 19 September 2016, gives no evidence of helm orders until 2334 or C-8 when he says he was instructed to put the rudder amidships. Quite how he recalled in September 2016 that that order was given at 2334 I do not know. I therefore do not find his evidence as to the time of the order persuasive though his evidence that there was an amidships order is consistent with the probabilities. It is difficult to draw reliable conclusions from the heading and course made good data of ALEXANDRA 1 as to when the amidships order might have been in circumstances where she was proceeding at less than 2 knots.

At 2327 or C-15 the engines of ALEXANDRA 1 were put to dead slow ahead. Her speed through the water was 1.3 knots and began to increase. She was now about 1 mile west north west of buoys no.1. At 2328 or C-14 the master overheard a conversation between Port Control and the tugboat ZAKHEER BRAVO. The tug was towing a barge and requested permission to pass the pilot station from west to east en route to Jumeirah. She was asked if she could see a waiting tanker and she replied that it was on her starboard bow. Port Control advised her to proceed at least one mile astern of the tanker. The master of ALEXANDRA 1 mistakenly thought that Port Control was speaking to EVER SMART. This caused him concern because he did not understand how EVER SMART could pass one mile astern of ALEXANDRA 1. He feared that if ALEXANDRA 1 went around buoy no.1 there would be a “fucking crunch” at the entrance to the channel. At 2331 or C-11 ALEXANDRA 1 stopped her engines but then put them to dead slow ahead again at 2332 or C-10. By this time she was about 9 cables west north west of buoys no.1. Her speed over the ground was 1.8 knots, her course made good was 101 degrees and her heading was 97 degrees.

The master said that at about 2335 or C-7 the speed of EVER SMART had fallen and that he could see the pilot vessel alongside the portside of EVER SMART. He judged that EVER SMART was approaching buoys no.2. The audio record suggests that he was still irritated by the delay in embarking the pilot. He appeared to compare the two vessels in terms of speed and manoeuvrability. EVER SMART was a “Mercedes” whilst ALEXANDRA 1 was a “hog on ice, with no skates”. The officer of the watch was the third officer. He said that he observed EVER SMART by radar and visually and saw the pilot vessel leaving her.

By 2337 or C-5 ALEXANDRA 1’s speed over the ground was 2.1 knots and she was making good a course of 106 degrees, though her heading was again falling off to port and was now 92 degrees. The audio record has the master saying “it is time for us to make a turn.” He must have had in mind a turn to starboard for the channel. The plots provided by the parties showing the progress of ALEXANDRA 1 do indeed show her at C-5 as approaching the point at which one would expect her to turn to starboard so as to line up on the starboard side of the approaches to the channel. But the course made good and heading changes between C-5 and C-4 do not suggest that there was any starboard helm action.

At about 2338 or C-4 when her course made good was 102 degrees and her heading 90 degrees the engines of ALEXANDRA 1 were put to slow ahead. The audio record again suggests that the master was concerned that it was about time to turn into the channel between buoys no.1. “And now student [or “grasshopper”] we need to move between these two buoys, red on the left, green on the right …………Don’t even think of smashing into it” [or “and not to slam into something”]. Counsel for EVER SMART submitted that this was an instruction to the helmsman to steer into the channel but the third officer is more likely to be the “student” than the helmsman. The audio record does not contain a starboard helm order, none of those on the bridge of ALEXANDRA 1 refer in their statements to a starboard helm order and no such helm order is pleaded by ALEXANDRA 1. The heading of ALEXANDRA 1 did not appreciably change. I therefore do not consider that there was a helm order. The master did however consider that it was time to turn to starboard so as to turn into the channel. The effect of not turning to starboard was that ALEXANDRA 1 continued heading so as to cross the approaches to the channel.

At about 2340 or C-2 the master observed that EVER SMART was abeam of buoys no.1. The audio record shows that he was concerned that EVER SMART was not turning to port as he expected her to do in the light of his mistaken understanding of the conversation between Port Control and ZAKHEER BRAVO. “He’s not changing fucking course. Is he an idiot?” The engines of ALEXANDRA 1 were put to full astern. At the time her speed over the ground was 2.3 knots. The engine order had little effect on her speed. By 2341 or C-1 her speed over the ground was 2.5 knots. At that time the master called Port Control and advised it that EVER SMART was not changing course and that there would be a collision.

At 2341.52 or C-30 seconds the engines of ALEXANDRA 1 were put to dead slow astern. Very shortly afterwards the master told EVER SMART by VHF to go hard starboard and switched on the deck lights of ALEXANDRA 1.

The starboard bow of ALEXANDRA 1 collided with the port bow of EVER SMART about 5 cables west north west of buoys no.1. The speed of ALEXANDRA 1 over the ground was 2.4 knots and her course made good was 104 degrees with her heading at 101 degrees. The change of course and of heading to starboard would appear to have been the result of the transverse thrust caused by her right handed propeller being operated at full astern. There is no evidence of any helm action and none is pleaded.

Less than a minute after the collision the master said to Port Control “he’s not following your rules………….you told him to go to my stern.”

At about 2230 EVER SMART left the Container Terminal with a pilot on board. As she proceeded down the channel her engines were at full ahead (manoeuvring), her speed over the ground was 12.9 knots and her course made good was of 313 or 314 degrees. EVER SMART passed two inbound vessels port to port in the channel. When EVER SMART passed buoys 6, 5 and 4 she was in about mid channel.

At about 2331 or C-11 her engines were reduced to half ahead and at 2332 or C-10 her engines were further reduced to slow ahead. At this time EVER SMART was passing buoys no.3 and was slightly to port of mid-channel. It may be that it was the effect of the easterly wind on the high sided container vessel which caused her to be set to port.

At about 2333 or C-9 the pilot, before leaving the bridge to disembark, advised the master to proceed at 10 knots and to keep a course of 314 degrees over the ground. He advised that there was a vessel to port and that the master should take care. At this time the speed of EVER SMART over the ground was 12.2 knots (or 11.1 knots through the water) and her course made good was 312 degrees. The third officer accompanied the pilot to the port side ladder.

At 2334 or C-8 the master ordered that a course of 319 degrees (rather than 316 degrees) be steered and that the engines of the vessel be reduced to dead slow ahead. At the same time her radar which had been on relative motion was switched to north up. It is possible that the alteration of course was ordered to achieve a course made good of 314 degrees and to bring the vessel back to the starboard side of the channel (though the latter does not appear to have been achieved). It is likely that that the engines were reduced to dead slow ahead to enable the pilot to be disembarked. It also seems likely that the radar was switched to north up to assist with the keeping of a good radar lookout.

The master does not however say in his unsigned witness summary (service of which was permitted by an order of this court) that he studied the radar echo of ALEXANDRA 1. Indeed, it is accepted in the preliminary act (part 1 of the Statement of Case) of EVER SMART that the echo of ALEXANDRA 1 was not particularly observed. The master does say in his witness summary that ALEXANDRA 1 was seen visually bearing 10 degrees on the port bow of EVER SMART distant about 1.5

At 2336 or C-6 the pilot had disembarked and the pilot vessel was clear. The pilot vessel proceeded ahead of EVER SMART to meet ALEXANDRA 1.

At 2337 or C-5 (when the course made good was 314 degrees and the speed over the ground was 9.5 knots or 8.3 knots through the water) the engines of EVER SMART were put to half ahead and at 2338 or C-4 her engines were put to full ahead (manoeuvring). 30 seconds later, at C-3½ her engines were put to full sea speed (when according to the third officer the engine rpm had reached 80rpm). Thus the vessel’s speed over the ground and through the water increased. By 2341 or C-1 the speed of EVER SMART was 11.8 knots over the ground or 9.8 knots through the water.

At about C-½ Port Control contacted EVER SMART by VHF. The master replied saying “Jebel Ali port. Ever Smart. Good morning”. Port Control asked whether EVER SMART was clearing to starboard. At the same time the pilot (whose vessel had come round the stern of ALEXANDRA 1 and was now on her starboard side aft) instructed EVER SMART by VHF to go hard to starboard. Port Control advised EVER SMART that there was a tanker coming to enter the channel. ALEXANDRA 1 also advised EVER SMART by VHF to hard starboard. The master of EVER SMART ordered “hard to starboard” and the helmsman repeated “hard starboard”.

Very shortly before collision (about 3 seconds before) the master of EVER SMART said “what’s that ?”. It is likely that he said that having seen the deck lights of ALEXANDRA 1 switched on.

At collision the course of EVER SMART was 316 degrees and her speed over the ground was 12.4 knots or 10.3 knots through the water.

Less than 2 minutes after the collision the master said (apparently to the officer of the watch and helmsman) “both of you…have you seen it or not ?”. He then said (it is suggested to himself) “how come you didn’t see it ?”.

At about 2348 or C+6 the master reported the collision to his owners saying “We hit her…because she stopped outside waiting, we were leaving the port, we did not see that ……I saw the light, but I didn’t know she was transverse, so we knock against her bow.”

The principal dispute in this case is whether rule 15 of the Collision Regulations, the crossing rule, applied. It is necessary to resolve this dispute before considering the respective alleged faults of each vessel.

Rule 15 of the Collision Regulations, entitled Crossing Situation, provides:

“When two power-driven vessels are crossing so as to involve risk of collision, the vessel which has the other on her own starboard side shall keep out of the way and shall, if the circumstances of the case admit, avoid crossing ahead of the other vessel.”

On behalf of EVER SMART Mr. Turner QC submitted that with EVER SMART proceeding along the dredged channel on a course of 314 degrees and ALEXANDRA 1 proceeding towards the entrance channel on an east south easterly course the two vessels were crossing so as to involve risk of collision. ALEXANDRA 1 had EVER SMART on her starboard bow and so, it was submitted, was under a duty to keep out of the way of EVER SMART pursuant to rule 15 of the Collision Regulations. On behalf of ALEXANDRA 1 Miss Selvaratnam QC submitted that authority establishes that the crossing rules have very limited, if any, application to questions of navigation in and around a narrow channel and in particular do not apply to a vessel in a narrow channel and a vessel navigating towards that channel in preparation for entering it. Further, she submitted that the “crossing” situation had only come about by reason of EVER SMART’s failure to navigate on the starboard side of the channel after the pilot had disembarked and in such circumstances EVER SMART was not entitled to invoke the crossing rule. She also submitted that the crossing rule did not apply because EVER SMART was not on a sufficiently defined course. Finally, she submitted that if the crossing rule otherwise applied it was ousted by rule 18(a)(ii) which requires a power driven vessel underway to keep out of the way of a vessel restricted in her ability to manoeuvre. ALEXANDRA 1 was, she submitted, restricted in her ability to manoeuvre because she was waiting to embark the pilot.

I consider first the question whether the crossing rules apply in the vicinity of a narrow channel. This question has attracted the attention of Admiralty judges since at least 1866. I was referred to The Leverington (1886) 11 PD 117, The Kaiser Wilhelm Der Grosse [1907] P.36 and 259, The Treherbert [1933] 47 Ll.L. Rep. 274, The Empire Brent (1948) 81 Ll. L. Rep. 306, The Canberra Star [1962] 1 Lloyd’s Rep.24, The Glenfalloch [1979] 1 Lloyd’s Rep. 247, Kulemesin v HKSAR [2013] 16 HKCFA 195 and The Nordlake and Seaeagle [2016] 1 Lloyd’s Rep. 656. The issue is discussed in Marsden and Gault, Collisions at Sea 14^th.ed. at paragraphs 5-302, 5-303 and 5-391.

It is apparent from The Leverington (1886) 11 PD 117 that where a vessel in one channel is approaching the junction between that channel and another channel and there is a second vessel in that other channel approaching the same junction the crossing rule can apply. Those are not the circumstances of the present case. In The Kaiser Wilhelm Der Grosse [1907] P.36 and 259 the court had to determine whether the crossing rule applied as between a vessel exiting a port between two breakwaters and a vessel preparing to enter the port through the same breakwaters. Sir Gorrell Barnes, President, in a judgment given a little more than three weeks after the collision had occurred, thought, “having regard to the locality and the difficulties there are in applying [the crossing rule] the probability is that [the crossing rule] is not applicable” (see p.44). The Court of Appeal agreed. Lord Alverstone CJ said that he had “very little doubt” that the crossing rule did not apply. He considered that the narrow channel applied to the entrance into the port through the breakwaters (see p.263). I am not persuaded that it would be safe to apply that decision, which involved a narrow channel through breakwaters, to the present case which involves a long dredged channel. The Treherbert [1933] 47 Ll. L. Rep. 274 does not assist because it was held that there was no narrow channel and the crossing rule applied.

In The Empire Brent (1948) 81 Ll. L. Rep. 306 Willmer J. expressed a clear view, albeit obiter, as to the non-application of the crossing rule when two vessels are navigating in a narrow channel. He said:, at p.312:

“As I understand the principles which apply in narrow channels it has been laid down for many, many years that, although the crossing rule does from time to time have to be applied in narrow channels (when, for instance, a vessel which is crossing the channel has to act in relation to a vessel which is proceeding up or down the channel), nevertheless, when vessels are approaching each other, navigating respectively up and down the channel, it is Article 25 of the Collision Regulations [then the narrow channel rule] which applies and which applies exclusively. There is no room in such a situation for applying the provisions of the crossing rule at the same time as the provisions of the narrow channel rule, because the requirements of the rules are different.”

The facts of the present case do not involve either a vessel crossing a narrow channel or two vessels navigating respectively up and down a narrow channel and so Willmer J.’s statement of principle as to the applicable rules in such circumstances does not assist, at least directly, in resolving the dispute in the present case.

In The Canberra Star [1962] 1 Lloyd’s Rep.24 Hewson J., said

“In the particular circumstances of this case where vessel A, proceeding down river outside the channel, intending to enter it, sees an upcoming vessel B approaching in the next reach, bearing on her starboard side, on a main channel course which, if followed into the reach in which A is navigating, will or should enable the two vessels to pass safely port to port by reason of the fact that B should keep to her own starboard side, the crossing rule does not, in my opinion apply. ”

The circumstances of that case are arguably comparable to the circumstances of the present case which concerns a vessel navigating within a channel and another vessel approaching the entrance to the channel with a view to embarking a pilot and then entering the channel. The statement of principle by Hewson J. is therefore some support for the case of ALEXANDRA 1.

In The Glenfalloch [1979] 1 Lloyd’s Rep. 247 Brandon J. held that where one vessel was proceeding along a channel and another vessel was proceeding along the same channel but in a contrary direction the crossing rule did not apply. He said:

“In my view, where one ship is proceeding along a narrow channel in one direction, and another ship is proceeding along the same channel in the other direction, even though their courses are crossing so to involve risk of collision, the narrow channel rule governs the case, and not the crossing rules; The Kaiser Wilhelm Der Grosse [1907] P. 259, The Heranger

None of the circumstances mentioned by Brandon J. replicate the circumstances of the present case. In The Nordlake and Seaeagle [2016] 1 Lloyd’s Rep. 656 I agreed with what Brandon J. had said in The Glenfalloch about the crossing rules not applying where two vessels are proceeding along a narrow channel in opposite directions. But those circumstances are not the circumstances of the present case.

The relationship between the crossing rules and the narrow channel rule in circumstances where one vessel is navigating along a narrow channel and another vessel is preparing to enter the channel was considered in the Final Court of Appeal in Hong Kong in a judgment delivered by Lord Clarke of Stone cum Ebony; see Kulemesin v HKSAR [2013] 16 HKCFA 195. This was an appeal in a criminal case arising out of a serious collision off Hong Kong as a result of which there was much loss of life. The masters of both vessels had been convicted of a criminal offence and each appealed. One of the questions certified for hearing on the appeal was whether the crossing rules apply when a vessel is approaching a channel on a crossing course involving risk of collision with another vessel navigating in the channel; see paragraph 142. There was however another question, which Lord Clarke described as “the most important issue” in the case namely, whether the buoyed channel was a narrow channel within rule 9 of the Collision Regulations; see paragraph 150. It was held that it was; see paragraph 198. Lord Clarke further held that the vessel (N67) which was navigating in the narrow channel was bound by the narrow channel rule and not by the crossing rule. He said that the observations of Willmer J. in The Empire Brent were “plainly correct”; see paragraph 201. He also said that the observations of Hewson J. in The Canberra Star seemed “good sense”; see paragraph 217-8. When dealing with the navigation of the other vessel (YH) which was approaching the entrance to the narrow channel he said this, at paragraph 225:

“…vessels approaching a narrow channel and intending to proceed along it are not bound by the crossing rule but must enter the channel and, as they do so, keep as near to the starboard side as is safe and practicable in accordance with r.9. It seems to me to follow that a vessel shaping to enter the channel should, as a matter of good seamanship, navigate in such a manner that, when she reaches the channel, she is on the starboard side of the channel in accordance with r.9.”

The circumstances of that case appear to be similar (though not of course identical) to the circumstances of the present case. N67 was proceeding along the narrow channel (on a course of 90-92 degrees) and could see the green light of YH. YH was proceeding towards the entrance to the narrow channel (on a course of about 260 degrees making alterations to starboard so as to take a down channel course by the

The statements of principle in both the Canberra Star and Kulemesin v HKSAR support the case of ALEXANDRA 1 that the crossing rule did not apply in this case. But I am not bound by the decision of Hewson J. in the Canberra Star because it is a decision at first instance. Nor am I bound by the decision of the Final Court of Appeal in Hong Kong in Kulemesin v HKSAR because it is a decision of a foreign court. However, both decisions are of considerable persuasive authority bearing in mind the experience and knowledge of collision actions possessed both by Hewson J. (a naval officer, an Admiralty specialist, author of a treatise upon navigation and Admiralty Judge from 1958 until 1966) and by Lord Clarke (whose expertise in this field is too well-known to require explanation). I consider that I ought to follow their statements of principle unless I consider that they are wrong.

Having regard to the clear terms of rule 15 on which Mr. Turner relies it is reasonable to ask why the crossing rules do not apply to two vessels which are crossing so to involve risk of collision - so that the circumstances appear to be squarely within rule 15 of the Collision Regulations - but where one of them is navigating in a narrow channel and the other is navigating in preparation to enter that narrow channel. The answer, in my judgment, is that given by Lord Clarke in Kulemesin v HKSAR at paragraph 167:

“Safety requires a vessel approaching the channel so as to proceed along it to navigate so that if the vessels pass in the channel they will pass port to port. This will be achieved if the narrow channel rule applies. If it does not, there is considerable scope for confusion. ”

A similar point was made by Willmer J. in The Empire Brent when he explained why the crossing rule did not apply to two vessels navigating a narrow channel. He said that there was no room for applying the crossing rule at the same time as applying the narrow channel “because the requirements under the rules are different”. To have two sets of rules with different requirements applying at the same time is of course unsafe and cannot have been intended by those who drafted the Collision Regulations. Similarly, where one vessel is within a narrow channel and has a vessel on her port bow on a crossing course outside the channel but proceeding towards it in preparation for entering it, the vessel in the narrow channel cannot be under a duty (pursuant to the crossing rules) to maintain her course and speed and at the same time under a duty (pursuant to the narrow channel rule) to keep to the starboard side of the channel since the two duties may, depending upon the circumstances, require different action. As Lord Clarke said there would be considerable scope for confusion.

These considerations strongly suggest that in the interests of safety, which of course is the foundation of the Collision Regulations, the crossing rules cannot have been intended to apply where one vessel is navigating along a narrow channel and another

I have therefore decided that I should follow the statements of principle by Hewson J. and Lord Clarke. Indeed, I respectfully agree with them.

Mr. Turner submitted that the facts of Kulemesin v HKSAR were distinguishable because “YH was broadly end on to N67 …….. and shaping to enter the channel” whereas “until the last moments before the collision ALEXANDRA 1 was not

YH and N67 were not end on. YH could see the green light of N67 and N67 could see the red light of YH. The angle between their respective courses may have been less than the angle between the respective courses of ALEXANDRA 1 and EVER SMART but that does not appear to me to be a material distinction.

I accept that ALEXANDRA 1 had not altered her course to starboard in order to turn into the channel. Mr. Turner suggests that in the last moments before the collision she was shaping to enter the channel. For the reasons I have given I do not consider it probable that there was any starboard helm order before the collision. The change of heading to starboard is more likely to have been the result of the engines of ALEXANDRA 1 being put full astern at C-2. But the circumstances before that were that ALEXANDRA 1 had been proceeding very slowly towards the channel entrance whilst waiting to embark her pilot in order to proceed down the channel. It does not appear to me that that is a material distinction from the facts of Kulemesin v HKSAR such as to render the crossing rules applicable at the same time as the narrow channel rule was applicable to EVER SMART. That would be unsafe for the reasons indicated by Lord Clarke and Willmer J. and so would not be consistent with the principles underlying the Collision Regulations.

Mr. Turner further submitted that the facts of the instant case are “four square within the Albano” and that the crossing rule applied. The Albano [1907] AC 139 was a decision of the Privy Council on appeal from the Supreme Court of Canada. In that case Albano was proceeding at about 9 knots towards the entrance of a port where she intended to embark a pilot. Parisian was proceeding to the same spot at 14-15 knots, also with the intention of embarking a pilot. Albano was on the starboard bow of Parisian. Their courses were crossing. However, although Albano was seen at a distance of about 6 miles, those on board Parisian were “strangely oblivious” of her until their attention was directed to her again when Albano blew three short blasts immediately before the collision. As the vessels approached each other and the spot at which they were to take a pilot each vessel reduced her engines. Parisian’s engines were stopped shortly before the collision and her speed was about 1 knot as the pilot cutter proceeded to her. The engines of Albano were also stopped but she still had way on. Shortly before the collision her engines were put full astern and she sounded three short blasts. Parisian then put her engines full ahead. At that time Parisian was crossing the bows of Albano. She was struck by Albano at nearly right angles. At collision, the speed of Albano was no more than 3 knots.

It was contended on behalf of Albano that the vessels were crossing so as to involve risk of collision and that it was the duty of Parisian having Albano on her starboard side, to keep out of her way. It was contended on behalf of Parisian that the crossing rules were not applicable. Parisian was practically a stationary vessel at the time when Albano was approaching her and Albano ought to have avoided her.

The judgment of the Privy Council was delivered by Sir Gorell Barnes (the first Admiralty judge to have been a common lawyer rather than a civil lawyer but nevertheless an Admiralty specialist). It was held that the crossing rules applied. Sir Gorell Barnes said at p.205:

“The consideration of the situation must be carried further back to the time when these vessels were approaching towards the spot where the collision took place, and would if they continued doing what each of them was respectively doing, arrive at that spot so as to involve risk of collision. ……….They were in fact converging on a spot on courses and at speeds which would probably bring them to that spot so as to present a danger of collision when they reached it, which each of them would in the course of her navigation, and their Lordships are of the opinion that in these circumstances the vessels were vessels crossing so as to involve risk of collision and that arts 19, 22 and 23 [the crossing rules] were applicable.”

It was held that Parisian was solely responsible for the collision. There was not a sufficient basis for concluding that Albano ought to have taken action to avoid the collision earlier than she did.

It is immediately to be noted that there was no narrow channel. Thus, the essential reason why the operation of the crossing rules may be excluded when one vessel is navigating within a narrow channel and another vessel is approaching that channel with a view to entering the channel is absent.

The facts of the present case are therefore not “four square within the Albano.” Since there was no narrow channel the decision and reasoning in that case does not assist in resolving the principal dispute between the parties as to the applicability of the crossing rules. The Albano provides no reason for not following Lord Clarke's statement of principle in Kulemesin v HKSAR.

I have therefore concluded that rule 15 of the Collision Regulations, the crossing rule, did not bind ALEXANDRA 1 when she approached the dredged channel leading to Jebel Ali and so she was not under a duty to keep out of the way of EVER SMART. Her duty, as a matter of good seamanship, and as formulated by Lord Clarke, was to navigate in such a manner that, when she reached the channel, she would be on the starboard side of the channel in accordance with rule 9. She required to embark a pilot but that circumstance did not, in my judgment, absolve her from that duty.

In the light of my decision it is unnecessary to deal with Miss Selvaratnam’s further submissions as to why the crossing rule did not apply on the facts of the present case. I shall therefore deal with them as shortly as I can.

First, it was said that when the pilot had disembarked from EVER SMART the only reason that the vessels were crossing so as to involve risk of collision was that EVER SMART had allowed herself to navigate on the portside of the channel. Had she navigated on the starboard side of the channel there would have been no collision (as shown by a "counterfactual" plot). The authorities show that a vessel cannot claim the benefit of rule 15 of the Collision Regulations when the vessel brings about the crossing situation by her own fault; see The Spyros [1953] 1 Lloyd’s Rep.501 at p.509 and The Forest Pioneer [2007] EWHC 84 (Comm) at paragraph 39. However, rule 15 applies when vessels are crossing so to involve a risk of collision. Rule 7(d)(i) provides that risk of collision shall be deemed to exist if the compass bearing of an approaching vessel does not appreciably change. The agreed schedule shows that the bearing of ALEXANDRA 1 from EVER SMART did not appreciably change. Although EVER SMART was not on the starboard side of the channel it is unlikely that the bearing would have appreciably changed had she been on the starboard side throughout her transit of the channel. The counterfactual plot (showing that there would have been no collision had EVER SMART navigated on the starboard side of the channel) shows a passing distance of only three quarters of a cable, about 450 feet. Thus the risk of collision was not brought about by the fault of EVER SMART. It was brought about by her proceeding out of the channel on a course of 314 degrees. That course was, in the language of Steel J. in The Forest Pioneer “necessitated by external circumstances", namely, the direction and width of the channel.

Second, it was said that EVER SMART was not on a sufficiently defined course. In order for the crossing rule to apply each vessel must be on a sufficiently defined course as explained in The Alcoa Rambler [1949] AC 236 at p. 249 per Lord Wright:

“The test always is: was what was being done open and notorious to a seaman on the other ship in the ordinary course of navigation……The ordinary idea of a course is a sufficiently constant direction of a ship on the same line or heading. This will enable a navigator when he sees the other vessel to know if she is on a crossing course. He can often only become aware of that if he can keep the other vessel under observation for sufficient time to ascertain if she is or is not changing her heading…… ”

It was said that after her pilot had disembarked EVER SMART did not “achieve ‘an open and notorious’ course or speed: she accelerated and moved a little to starboard indicating that she was perhaps intent on returning to her correct water, but particularly after she passed No.1 buoys there could be no certainty about that.” However, there can, in my judgment, be no doubt that EVER SMART, when proceeding along the channel, was on a sufficiently constant heading to be on a course. The agreed schedule of the vessels’ navigation shows that her course made good from C-26 was between 312 and 315 degrees (and that her heading varied between 315 and 319). It was or ought to have been clear to ALEXANDRA 1 that she was on a course which took her along the line of the dredged channel. That surely was a sufficiently constant direction for the purposes of the crossing rule.

Third, it was submitted that ALEXANDRA 1 was restricted in her ability to manoeuvre and so, pursuant to rule 18 of the Collision Regulations, the crossing rule did not apply and instead EVER SMART was under an obligation to keep out of the way of ALEXANDRA 1. The term “restricted in her ability to manoeuvre” is defined in rule 3(g) of the Collision Regulations as meaning a vessel "which from the nature of her work is restricted in her ability to manoeuvre as required by these Rules and is therefore unable to keep out of the way of another vessel". The term expressly includes (see rule 3(g)(iii)) a vessel engaged in transferring persons. But had the crossing rule applied and ALEXANDRA 1 had been obliged to keep out of the way of EVER SMART she could have done so, for example at C-6, by putting her engines astern and so permitting EVER SMART to cross ahead in safety or by turning to starboard. She was not restricted in her ability to take such action by reason of embarking the pilot because that work had not commenced (and never did commence). Her turning circle data shows that she could have turned 180 degrees within half a mile.

In her oral submissions Miss Selvaratnam also submitted that the vessels were not crossing because “everybody knew that ALEXANDRA 1 was about to board the pilot to enter the channel and so would not be expected to cross ahead of EVER SMART”. Mr. Turner responded by saying that it was sufficient, as explained in the Albano, if the vessels would, if they continued doing what each of them was respectively doing, arrive at the same spot so as to involve risk of collision. However, I am very doubtful that ALEXANDRA 1 was on a sufficiently constant heading to be on a course. The agreed schedule shows that from C-26 until C-23 her course over the ground varied from 119 to 127 degrees (altering to starboard), from C-23 until C-12 her course over the ground varied from 127 to 81 degrees (altering to port), from C-12 until C-7 her course over the ground varied from 81 to 115 degrees (altering to starboard) and from C-7 until C-1.5 her course over the ground varied from 115 to 91 degrees (altering to port). During that same period her heading also varied; from C-26 until C-14 her heading varied from 112 to 84 degrees (altering to port), from C-14 until C-8 her heading altered from 84 to 100 degrees (altering to starboard) and from C-8 until C-3 her heading altered from 100 to 90 degrees (altering to port). This variation of course made good and of heading was no doubt caused by the circumstance that ALEXANDRA 1 was proceeding very slowly (about 1-2 knots over the ground). At such a slow speed ALEXANDRA 1 was not very manoeuvrable. Nevertheless she made progress in a broadly east south easterly direction towards the entrance of the channel as she waited to embark the pilot. Was she on a sufficiently constant direction or heading to be on a course? I do not consider that she was. Her "course made good" varied between 81 and 127 degrees (and her heading varied between 84 and 112 degrees). It is difficult to describe that as "a course" (though her preliminary act describes her as being on an east south easterly course). I would describe her as maintaining a broadly east or east south easterly heading as she waited for the pilot vessel to approach. That required her to have some, but not very much, way on. I would describe ALEXANDRA 1 as waiting for the pilot vessel to arrive rather than being on a course. Had a good lookout been kept on board EVER SMART from C-21 until collision it would have been apparent that ALEXANDRA 1 had moved less than a mile. It would or ought to have been obvious that she was waiting to embark a pilot.

In summary, the crossing rules do not apply for the reasons given by Lord Clarke in Kulemesin v HKSAR and in any event ALEXANDRA 1 was not on a sufficiently

Since the Owners of ALEXANDRA 1 opened the case it is convenient to consider their allegations of fault against EVER SMART first.

Rule 9 of the Collision Regulations, entitled Narrow Channels, provides that a vessel proceeding along the course of a narrow channel or fairway shall keep as near to the outer limit of the channel or fairway which lies on her starboard side as is safe and practicable. It is accepted that the exit channel was a narrow channel. From about C10 when EVER SMART was passing buoys no.3 she was no longer navigating in mid channel but was to port of mid channel. EVER SMART was 42 metres in breadth. The channel was a little under 2 cables in width. It was therefore safe and practicable for EVER SMART to navigate well to starboard of mid channel. She never did so after C-10 and was therefore in breach of rule 9. This was probably the result of the wind on the high sided container vessel.

It was submitted on behalf of EVER SMART that she was not in breach of rule 9 in circumstances where ALEXANDRA 1 was not within the narrow channel. I am unable to accept that submission. EVER SMART was or ought to have been aware that ALEXANDRA 1 was proceeding towards the channel intending to embark the pilot and then proceed down the channel. Certainly the pilot was aware of that when on board EVER SMART. In those circumstances it was incumbent upon EVER SMART to comply with rule 9 so as to ensure that if the vessels met in the channel they would pass safely port to port. No particular submissions were made in this regard with respect to the instruction by Port Control to ALEXANDRA 1 that she should proceed down the channel after EVER SMART was clear. But even if this had been overheard by EVER SMART the two vessels might still meet in the approaches to the channel beyond buoys no.1 and so it remained necessary for EVER SMART to navigate on the starboard side of the channel so that if they met in the approaches they would pass safely port to port. It was further submitted that the breach of rule 9 was not causative. I am unable to accept that submission. The effect of the breach was that it created a risk that the vessels might not pass safely port to port. That is sufficient to be a material cause of the collision. The master’s order to steer 319 degrees was likely to have been ordered with a view to bringing EVER SMART back towards the starboard side. But she never returned to the starboard side. When she passed buoys no.1 she was still on the port side of the channel. She passed the green buoy on her portside at a distance of just over half a cable, whereas the red buoy on her starboard side was passed at a distance of almost a cable.

It was submitted that the radar and visual lookout on board EVER SMART were defective. Indeed it was submitted that the master of EVER SMART never observed ALEXANDRA 1 by radar or visually.

It is admitted that the echo of ALEXANDRA 1 was not particularly observed. Neither the master in his witness summary nor the third officer in his statement suggest that they observed ALEXANDRA 1 by radar. It is also apparent from the radar stills recorded in the VDR that her echo was never acquired as an ARPA target. It was submitted on behalf of ALEXANDRA 1 that it ought to have been acquired as an ARPA target, certainly when the pilot left the bridge, at about C-9, having advised the master that there was a vessel to port and that he should take care. Rule 7(b) of the Collision Regulations provides that “proper use shall be made of radar equipment……and radar plotting or equivalent systematic observation of detected objects.” It was submitted on behalf of EVER SMART that there was insufficient time at C-6 (after the pilot had disembarked) to capture the echo of ALEXANDRA 1 as an ARPA target. Reliance was placed on the master’s note of his interview with the MAIB in which he said that it would take 2-3 minutes to acquire the echo as an ARPA target and reliance was also placed on the explanation by the court (Geoffrey Brice QC) in The Contship Success [1998] 2 Lloyd’s Reports 488 at p.494 that international standards require that within one minute a relative target motion trend shall be provided and that within 3 minutes the true motion of the target with 95% probability shall be provided. This submission raises a matter of seamanship and an understanding of ARPA so both counsel said that this is a matter upon which I should seek the advice of the Elder Brethren. In formulating the appropriate question on this matter (and on others) I invited the comments of counsel. In accordance with the guidance of the Court of Appeal in The Bow Spring and Manzanillo [2005] 1 Lloyd’s Reports 1 at paragraphs 57-61 I also provided counsel with an opportunity to comment on the advice given by the Elder Brethren.

I asked the Elder Brethren the following question:

“Did good seamanship require EVER SMART to acquire ALEXANDRA 1 as an ARPA target and, if so, when should that process have been initiated having regard to the time it would take to capture the echo of ALEXANDRA 1 as a reliable ARPA target ? Was it required when the pilot was on the bridge advising the master, when he left the bridge, or when he had disembarked the vessel? How long would it take to capture the echo as a reliable ARPA target?”

I received the following advice:

“Despite the duties and obligations of the pilot, his presence on board does not relieve the master or officer in charge of the watch from their duties and obligations for the safety of the ship. As a matter of accepted best practice in Bridge Team Management and good seamanship, particularly when under pilotage, the master and officer of the watch of EVER SMART should have acquired ALEXANDER 1, amongst other vessels, as soon as it had been detected on radar as a target. This would have been at least 5 miles before the departure position of the pilot. Furthermore, the use and monitoring of ARPA (either manually acquired or automatically acquired) should have been assigned to a single officer as part of the Bridge Team. The automatic detection of radar screen displayed AIS targets is not a substitute.

ARPA is designed to provide continuous, accurate and rapid situation evaluation. The performance standards and accuracy of ARPA are a matter of an IMO Resolution based on steady state tracking and straight line course and constant speed of both own ship (EVER SMART) and target ship

(ALEXANDER 1). In summary the longer the steady state tracking the more accurate the displayed information. In this case, in our opinion, a minimum of 3 minutes should have been sufficient to obtain increasingly valuable data and the warning of a distinct risk of collision (at zero or very low CPA).”

I accept the advice of the Elder Brethren that within not less than 3 minutes of detecting the echo of ALEXANDRA 1 that vessel ought to have been acquired as a reliable ARPA target. (That is my understanding of the phrase “a minimum of 3 minutes should have been sufficient”.) The pilot departed EVER SMART as she was approaching the no.2 buoys. 5 miles before that EVER SMART would have been approaching buoys no. 8. Thus the advice given by the Elder Brethren suggests that in their opinion ALEXANDRA 1 ought to have been detected when EVER SMART was approaching buoys no.8. (In this regard it is to be noted that the master of ALEXANDRA 1 observed EVER SMART by radar when that vessel was at buoys no.8.) However, Mr. Turner in his comments upon the advice of the Elder Brethren has said that it was apparent from the radar images that ALEXANDRA 1 was only detected by radar at about 2322 or C-20 when EVER SMART was passing buoys no.6. If that is so then ALEXANDRA 1 ought to have been acquired as a reliable ARPA target by about C-17 which was when EVER SMART was passing buoys no.5. Thus it is clear that, allowing 3 minutes to acquire a reliable ARPA target, those on the bridge of EVER SMART ought to have had a reliable ARPA target of ALEXANDRA 1 several minutes before the pilot disembarked. They did not do so. There was a failure to keep a good radar lookout.

The master of EVER SMART maintains (in his unsigned witness summary) that he saw ALEXANDRA 1 at a distance of about 1.5 miles about 10 degrees on the port bow after the pilot left the bridge. It was submitted on behalf of ALEXANDRA 1 that this cannot be true because his question “what’s that ?” three seconds before the collision and his question to himself within two minutes after the collision “how come you didn’t see it ?” are inconsistent with the master having previously seen ALEXANDRA 1. I accept that these contemporaneous comments are powerful pieces of evidence suggesting that he had not seen ALEXANDRA 1.

The master’s explanation for the words “what’s that” is that until the deck lights of ALEXANDRA 1 were switched on he did not appreciate how far across the channel exit ALEXANDRA 1 had come or her angle of approach. This explanation is said to be untrue. It is given in his unsigned witness summary. (The “summary” was an account of what the master told the solicitors acting for EVER SMART on 19 February 2015 and in subsequent telephone calls on 12 July 2016 and 24 August 2016. It was unsigned because of a dispute concerning his pension.) Although it was unsigned I was invited to read it and so it was put in evidence. But of course it cannot have the weight of a signed statement. Moreover, the vessel’s log, signed by the

Thus the master’s visual lookout was seriously defective. Masthead lights are visible at a minimum distance of 6 miles and sidelights are visible at a minimum distance of 3 miles (see rule 22 of the Collision Regulations). Had the master kept a good visual lookout he would have seen not only the green side light of ALEXANDRA 1 but also the orientation of the masthead lights which would have indicated to him from about C-4 that ALEXANDRA 1 was not set to pass EVER SMART port to port but was heading across his bows or, as he told his owners, "transverse". Had he acquired ALEXANDRA 1 as an ARPA target with a low CPA and paid attention to the radar he would have been alerted to the risk of collision by a CPA alarm and the heading of ALEXANDRA 1 would also have been apparent so that the master would have appreciated that ALEXANDRA 1 was not turning to starboard. Instead he continued to assume that the vessels were set to pass port to port. The master failed to use either his eyes or his radar to confirm his assumption. If he had done so he would have realised that his assumption was mistaken. I therefore accept that the master’s lookout was seriously defective. He ought to have been assisted by the officer of the watch, the third officer. But the third officer had left the bridge to escort the pilot to the portside ladder and when he returned to the bridge he went into the chart room to “update the log book” and did not concern himself, as he ought to have done, with his duty to keep a good lookout.

Thus there was a breach of rule 5 of the Collision Regulations which requires a proper look out to be kept “so as to make a full appraisal of the situation and of the risk of collision” and a breach of rule 7 which provides that assumptions shall not be made on the basis of scanty information.

At about C-6, as buoys no.2 were passed, the pilot was disembarked. The engines of EVER SMART were at dead slow ahead and her speed over the ground was falling from over 11 knots to less than 10 knots. ALEXANDRA 1 was fine on the port bow of EVER SMART distant about 1.5 miles. EVER SMART was aware, or ought to have realised, that ALEXANDRA 1 was preparing to embark the pilot. At C-5 the master increased the engines to half ahead, at C-4 he increased the engines to full ahead (manoeuvring) and at C-3½ he increased the engines to full sea speed. The result of these engine movements was that at collision EVER SMART had a speed over the ground of 12.4 knots. Thus from C-4 until collision the speed of EVER SMART increased from 9.6 to 12.4 knots over the ground, or from 7.9 to 10.3 knots through the water.

It was submitted on behalf of ALEXANDRA 1 that this speed was unsafe and that the vessel’s engines ought to have remained at dead slow ahead so that the speed of the vessel did not exceed 10 knots. What is a safe speed is a question of good seamanship and so I asked the Elder Brethren the following question:

“Assuming that a good lookout was kept on board EVER SMART and that EVER SMART complied with the narrow channel rule what engine orders ought to have been given on board EVER SMART when the pilot vessel was clear of her at about C-6. Was it in accordance with good seamanship to increase the speed of the engine successively from C-5 to C-3 to full sea speed ?”

I received the following advice:

“EVER SMART, in decreasing speed to land the pilot while still in the buoyed channel, was allowed to set and drift to her port. This resulted in the vessel being on the wrong side of the channel by C5. Increasing speed as soon as the pilot vessel was clear would have reduced the effect of the wind (drift) on this large high-sided container vessel and therefore an increase of speed through the water in combination with an alteration of heading to starboard was the correct manoeuvre. Full Ahead Manoeuvring (57 RPM) would, in our opinion, be a reasonable action.

To immediately order Full Sea Speed (increasing to 80 RPM) and therefore increasing to 20 to 21 knots through the water without being on immediate engine manoeuvring status was not good seamanship. This should only have been ordered once the vessel was clear of the narrow channel, the pilot boarding and landing area and any concentrated traffic. At the earliest, once clear of the Fairway Buoy would be a reasonable position to consider Full Away.”

This advice considers the question of speed in the context of EVER SMART being on the wrong side of the channel but seeking to regain the right side of the channel in accordance with the narrow channel rule. In such circumstances and assuming that a good lookout revealed an incoming vessel shaping to proceed down her starboard side

Mr. Turner sought to make a virtue out of EVER SMART’s excessive speed by pointing out (with the assistance of a plot) that had she kept to a lesser speed of 9.45 knots EVER SMART would have collided with ALEXANDRA 1 in way of a cargo tank and that the consequences of such a collision, involving spillage of cargo, would have been much worse. However, the court is concerned with the effect of EVER SMART’s excessive speed on the damage which in fact occurred. There can be no doubt that the causative potency of her excessive speed with regard to the damage which in fact occurred was substantial.

EVER SMART applied hard starboard helm very shortly before the collision. It was too late to avoid collision. She ought to have taken avoiding action before this and her failure to do so is a further fault. But, like her excessive speed, it was caused by the failure to keep a good lookout.

It was suggested that 5 short blasts ought to have been sounded by EVER SMART or that a light should have been flashed, pursuant to rule 34 (d) and rule 36 of the Collision Regulations. Neither was done because those on board were not aware that ALEXANDRA 1 was heading across the approaches to the channel. So this failure was also caused by the failure to a keep a good lookout.

Mr. Turner QC submitted that as a matter of good seamanship ALEXANDRA 1 ought not to have got to within 1 nautical mile of the entrance to the channel until EVER SMART had cleared the channel. In the event the bows of ALEXANDRA 1 got to about 5 cables from buoys no.1 and rather less (about 2 cables) from the end of the pecked channel as marked on the chart. Her heading was athwart the channel; she was heading 101 degrees at collision when the direction of the channel was 135 degrees. Her bow at collision was on about the centre line of the channel (projected forward from buoys no.1).

Mr. Turner's submission receives apparent support from statements of the master of ALEXANDRA 1 in his Letter of Protest and in his Statement of Facts that he was instructed by Port Control to proceed to 1 nautical mile off buoy no.1. However, the

I therefore asked the Elder Brethren the following question:

“Did good seamanship require ALEXANDRA 1 to keep a certain minimum distance from buoys no.1 so long as EVER SMART was still in the dredged channel and if so what was that distance ? ”

They advised as follows:

“Subject to a good aural and visual lookout, it would be reasonable and good seamanship for the Master of Alexander 1 to have approached the first pair of buoys keeping close to her own side of the entrance channel.”

I accept that advice. It is consistent with the opinion of Lord Clarke in Kulemesin v HKSAR as to how a vessel should approach a narrow channel. An incoming vessel which is approaching the dredged channel can expect to pass an outgoing vessel safely port to port if it approaches the channel in such a way that she keeps to the starboard side of the approaches so that on entering the channel she is on the starboard side of it. Keeping a certain distance from the channel whilst another vessel is in the channel is not required in order to achieve a safe port to port passing.

Mr. Turner, in his comments upon this advice, relied upon the fact that Port Control had advised ALEXANDRA 1 to wait until EVER SMART was “clear” before entering the channel. However, ALEXANDRA 1 did not enter the channel whilst EVER SMART was still in the channel. EVER SMART passed buoys no.1 (the seaward end of the channel) at C-2 at which time ALEXANDRA 1 put her engines full astern. She never left the pilot boarding area before the collision.

The master said in his witness statement that he first saw the echo of EVER SMART when she was at about the no.8 buoys and that he continued to monitor her movements on the radar. No.8 buoys were a little under 6 miles from the no.1 buoys. The audio record suggests that the master saw EVER SMART when she was approaching buoys no.6 at about C-24. The criticism made of ALEXANDRA 1’s radar lookout is that there is no evidence that EVER SMART was obtained as an ARPA target. However, the audio record suggests that the master was observing EVER SMART. At C-12 he (rightly) considered that there was a risk of collision at the entrance to the channel if, as he (wrongly) thought, EVER SMART had been instructed to pass astern of ALEXANDRA 1. At that time EVER SMART was between buoys 4 and 3. He said in his witness statement that he observed that the speed of EVER SMART had fallen. This observation is confirmed by the audio record at C-6 (“it is slowing down, pushing the brake pedal”) at which time the pilot had disembarked from EVER SMART and the speed of that vessel had fallen. In his witness statement the master said that it was clear to him that EVER SMART was not turning to port as he had (mistakenly) expected her to do. This is confirmed by the audio record. At C-5 it evidences the master’s frustration with EVER SMART (“kindly fuck off please ….it is time for us to make a turn”) which suggests that he was watching her. The audio record also shows that he continued to observe EVER SMART because at C-2 (when EVER SMART was passing the no.1 buoys) he notes that EVER SMART has not changed course and asks “what’s he doing ?”

Thus it appears that the master of ALEXANDRA 1, notwithstanding that he may have failed to acquire EVER SMART as an ARPA target, did keep a good visual lookout such that he was aware of EVER SMART’s progress down the channel, that she had slowed down to disembark the pilot and that she was maintaining her down channel course and was not altering to port as he (mistakenly) expected her to do at the no.1 buoys. It was submitted that the master had a “constant appreciation” of what EVER SMART was doing. That submission is supported by the audio record. Thus although there may have been a failure to use ARPA it does not appear that such failure was causative of the collision. I am not persuaded to reach any different conclusion because the master’s letter of protest contains inaccuracies which Mr. Turner submitted were lies to make things look better than they were. The audio record is cogent contemporaneous evidence of the master’s visual lookout.

However, rule 5 of the Collision Regulations also requires a good lookout by hearing. The master’s aural lookout is open to criticism. The master of ALEXANDRA 1 misheard or misunderstood the VHF conversation between Port Control and ZAKHEER BRAVO. He thought, mistakenly, that the conversation was between Port Control and EVER SMART and so concluded that EVER SMART was being instructed to pass astern of him at a distance of a mile. Since the name of ZAKHEER BRAVO was clearly stated on VHF, as was the fact that she was towing a barge from west to east, it is likely that the master was not listening carefully. ZAKHEER BRAVO was at the time of the conversation off the starboard quarter of ALEXANDRA 1 as indicated by the screen shot of EVER SMART’s radar at C-12. It was submitted that it was “not negligent to mishear a conversation”. I am not sure whether it was intended by that submission to say that the master was not under a duty to listen to the conversation. But if that was intended I think that it was his duty to listen carefully because the conversation concerned his vessel (“the big tanker waiting there”). It was after the reference to his vessel that ZAKHEER BRAVO said that the

I therefore consider that the master’s aural lookout was defective. It seems to me that it was causative of the collision. The master realised at C-5 and C-4 that ALEXANDRA 1 ought to be turning to starboard towards the channel. However, he did not order starboard helm because he mistakenly thought that EVER SMART had been directed to pass astern of ALEXANDRA 1. Had he not been under this mistaken impression it is probable that ALEXANDRA 1 would have turned to starboard at C-5 or C-4 in order to approach the channel in such a way as to enter it on a down channel course and on the starboard side of it which would entail the two vessels passing port to port in the approaches to the channel. Instead, ALEXANDRA 1 headed so as to cross the approaches to the channel so as to give room for EVER SMART to turn to port (and for the vessels to pass starboard to starboard). When the master realised at C-2 that EVER SMART was not changing course to port at the no.1 buoys, as he thought she would, he put the engines of ALEXANDRA 1 full astern. This had the effect of turning the vessel’s heading to starboard (as a result of transverse thrust from the propeller when the engines were put full astern) but at collision ALEXANDRA 1 was still heading across the entrance to the channel and had there been no collision she would have continued onto the portside of the approaches to the channel. Miss Selvaratnam stressed that ALEXANDRA 1 “remained on her own side of the notional mid line of the channel extended up into the pilotage boarding area”. That is true. But she was heading so as to cross that line and would have done so had there been no collision.

Miss Selvaratnam submitted that the fault was not causative because ALEXANDRA 1 was not “blocking” the exit to the channel and there was ample room for EVER SMART to avoid the collision by turning to starboard. Although ALEXANDRA 1 was not blocking the exit to the channel in the sense that EVER SMART could, by altering course to starboard on passing buoys no.1, have avoided collision (by crossing ahead of ALEXANDRA 1 at a distance of only about a cable, as indicated on the plot to which I was referred) it does not follow that the prior fault by ALEXANDRA 1 in the form of her poor aural lookout, which resulted in her heading across the channel rather than having altered to course to starboard so as to pass EVER SMART port to port in the approaches to the channel, was not causative. The submission that it was not causative seeks to resurrect the “last opportunity” rule which is not a proper guide to causation; see The Ouro Fino [1988] 2 Lloyd’s Reports 325 at p.329.

Mr. Turner appeared to submit that ALEXANDRA 1 was wrong to put her engines to dead slow ahead at C-10 and to slow ahead at C-4. Such action was said to have increased and not decreased the risk of collision. (Mr. Turner referred to a plot which assumed ALEXANDRA 1 had kept to a speed of 1.2 knots and showed that EVER SMART would have crossed ahead of her at a distance of more than a cable.) However, the effect of such engine movements was to increase the vessel’s speed from a little under 2 knots (C-10) to a little more than 2 knots whilst she awaited the pilot. I do not consider that such a speed can be said to have been unsafe.

ALEXANDRA 1 decided to head across the approach to the channel rather than, at C- 5 or C-4, turn to starboard so as to pass EVER SMART port to port in the approaches to the channel. It seems to me that, but for the master’s poor aural lookout, he would have altered course to starboard so as to approach the channel in such manner that when he reached it ALEXANDRA 1 would be on the starboard side of the channel in accordance with the narrow channel rule. Instead, at C-3 ALEXANDRA 1 was heading about 90 degrees so as to cross the approaches to the channel and at collision ALEXANDRA 1 was crossing the approaches to the channel on a heading of 101 degrees with her bows on the centre line of the channel (as extended from the no.1 buoys). From such position it would have been difficult to reach the channel at the no.1 buoys on the starboard side of it. Miss Selvaratnam said that because from C-2 ALEXANDRA 1 was taking avoiding action her position at collision cannot be used to suggest that she would not or could not have entered the channel on her starboard side; but it was her failure to alter course at C-4 which meant that it would be difficult for her to reach the channel on the starboard side of it. If it could have been achieved by a laden VLCC in the 5 cables available it would have entailed ALEXANDRA 1, having crossed onto the portside of the approaches to the channel, applying starboard helm to regain the starboard side of the approaches to the channel before applying port helm to adopt a down channel course by the time the starboard hand no. 1 buoy was reached. Such a snaking manoeuvre is not the sort of manoeuvre which would be expected of a vessel approaching a narrow channel. However, this was not an additional fault. Her failure to turn to starboard at C-5 or C-4 was caused by her poor aural lookout.

Rule 36 of the Collision Regulations permits a vessel to attract the attention of the other by flashing a light at her. Mr. Turner submitted that at C-2 when ALEXANDRA 1 realised that EVER SMART had not altered course to port she ought to have flashed her light at EVER SMART. Instead, at C-1, she called Port Control. I agree that it would have been better to flash a light at C-2 (and also to sound 3 short blasts indicating that she was putting her engines astern). Whilst it is possible that the attention of the master of EVER SMART might have been attracted by the flashing of

ALEXANDRA 1 was equipped with AIS (Automatic Identification System) as is apparent from a Det Norske Veritas survey report dated February (or June) 2014. Whilst AIS is intended to enhance safe navigation and, in particular, to assist in target tracking and so may be used to assist in collision avoidance decision making AIS does not replace the vessel’s radar (see IMO resolution A.917(22) paragraphs 4 and 40) and an alert and systematic visual and radar lookout remains the primary instrument for safe navigation; see The Hakki Deval [2006] EWHC 2809 (Comm) at paragraph 24 per David Steel J. The advice from the Elder Brethren in this case as noted above is to the same effect.

It is accepted that ALEXANDRIA 1’s AIS was not working on the day of the collision. Mr. Turner submitted that those on board ALEXANDRA 1 had switched off her AIS system and that she therefore had no ready means of identifying vessels in her vicinity. It was further submitted that had her AIS system been working ALEXANDRA 1 would have triggered an alert on EVER SMART’s radar. I was not persuaded that there was any causative negligence in this regard. First, the suggestion that the AIS system had been switched off was based upon no more than, in circumstances where it had been alleged that ALEXANDRA 1 had not been properly equipped with AIS and/or that the AIS had not been operated, the absence of evidence dealing with these allegations. It was suggested that in the light of the “yawning gap” in the evidence the “only safe conclusion” was that the AIS was capable of working and had been switched off. I do not consider that it would be safe to draw that conclusion. In an email dated 20 February 2015, that is 9 days after the collision, the master of ALEXANDRA 1 reported that the AIS system sometimes “stuck or malfunctioned” due to “vessel vibration”. Given the proximity of this email to the collision it is likely that this was the reason why the AIS was not functioning on the day of the collision. There was no investigation at trial as to whether this malfunction ought to have been prevented by the deck superintendent (before the voyage) or by those on board (during the voyage). Second, whilst the absence of AIS meant that those on board ALEXANDRA 1 were deprived of that system’s ability to identify vessels in the vicinity those on board were keeping a good visual lookout on EVER SMART and were aware of what she doing. The absence of AIS did not therefore impinge on their visual lookout. (AIS would also have identified ZAKHEER BRAVO by name but in circumstances where the attention of those on board ALEXANDRA 1 was not directed to ZAKHEER BRAVO by the VHF conversation between that vessel and Port Control it is unlikely that the absence of AIS impinged on her lookout, either visual or aural.) Third, whilst an alert or alarm would have been triggered on EVER SMART’s radar had ALEXANDRA 1’s AIS system been working and had pre-set CPA and TCPA limits been set on EVER SMART (see IMO resolution A.917(22) paragraph 28) it is not likely that those on board EVER SMART would have noted the alert or alarm since they were not keeping a good radar lookout and in particular had not captured ALEXANDRA 1 as an ARPA target. For these reasons I am unable to accept the submission that there was negligence with regard to the AIS or that, if there was such negligence, it was causative of the collision.

Before considering the apportionment of liability for the collision it is necessary to mention Mr. Turner’s submission that the master and/or third officer of ALEXANDRA 1 were intoxicated. There was documentary evidence that the crew were tested for alcohol between 1800 and 1911 on 11 February 2015 (before the collision) and between 0310 and 0425 on 12 February 2015 (after the collision). The results were negative. Mr. Turner drew attention to (i) the fact that the witness statements of the master and third officer and the statement of facts referred to the second test but not to the first, (ii) the fact that the second officer (who it is said carried out the tests) was not on duty at the time of the first test and (iii) the fact that there is no reference in the audio record to the tests. He further submitted that the badinage and chatter heard on the audio record and the master’s emotional reaction to the collision are suggestive of intoxication. Finally, he said that intoxication explains the master’s misunderstanding of the conversation between ZAKHEER BRAVO and Port Control and ALEXANDRA 1’s “dangerous” navigation.

ALEXANDRA 1, being a VLCC, was a dry ship. The use and possession of alcohol was prohibited on board. It is unlikely that the master was intoxicated. The truth is that there is no evidence that he was intoxicated. I do not regard the absence of support for the first test as indicative that the master was intoxicated. It may be that the audio record suggests that the master was at times irritated, at times excited and at times voluble. But it would be most unsafe to conclude from the audio record that he was intoxicated. I am unable to accept Mr. Turner’s submission that either he or the third officer was intoxicated.

In The Nordlake and Seaeagle [2016] 1 Lloyd’s Reports 656 at paragraphs 148-150 I summarised, with the assistance of an article by Sir Henry Brandon, the manner in which the court apportions liability for a collision pursuant to section 187 of the Merchant Shipping Act 1995. The relative culpability and causative potency of each vessel’s faults have to be assessed.

EVER SMART was in breach of the narrow channel rule and she failed to keep a good lookout. As a result of her poor lookout she proceeded at an unsafe speed, having put her engines to full ahead manoeuvring at C-4 and to full sea speed at C3½, and failed to take action to avoid the collision.

In terms of culpability these faults were very serious. Breaches of the obligations imposed in certain defined situations by the Collision Regulations are usually regarded as seriously culpable. One such rule is the narrow channel rule. In this case EVER SMART’s failure to keep to the starboard side of the channel was seriously culpable because it was known that ALEXANDRA 1 was outside the channel preparing to enter it and that the vessels might meet either in the channel or in the approaches to it beyond the no.1 buoys. It is true that as EVER SMART approached buoys no.1 she was in the process of moving towards the starboard side of the channel but she ought to have been well on the starboard side of the channel. The master’s failure to keep a good lookout was also seriously culpable. Although he was aware of

ALEXANDRA 1’s failure to keep a good aural lookout was culpable because had her master listened carefully to the conversation at C-14 between ZAKHEER BRAVO and Port Control he would have appreciated that the vessel involved was not EVER SMART but ZAKHEER BRAVO and, notwithstanding that his mistaken understanding of the conversation did not make sense in that it appeared to give rise to a risk of collision, he failed to check with Port Control or EVER SMART whether it really was EVER SMART which had been instructed to pass astern of ALEXANDRA 1. This check could have been made at C-12 or at any time thereafter. The failure to make this check, coupled with the earlier failure to listen carefully to the conversation, was culpable to a substantial degree.

In terms of relative culpability the faults of EVER SMART were, in my judgment, much more culpable than those of ALEXANDRA 1.

Each vessel was guilty of a poor lookout. But the fault of EVER SMART in this respect was much greater than that of ALEXANDRA 1. EVER SMART wholly failed to keep a lookout from C-6 until collision. To adopt the language of Sir Gorrell Barnes in The Albano she was “strangely oblivious” that ALEXANDRA 1 was heading so as to cross the entrance to the channel. By contrast ALEXANDRA 1, although she may not have kept a good radar lookout by the use of ARPA, kept a good visual lookout with the result that she followed EVER SMART's progress along the channel. Her error was in her aural lookout. She misunderstood a VHF conversation between another vessel and Port Control believing it was between EVER SMART and Port Control and failed to check that her understanding was correct. Whilst this was a substantial and significant error in her aural lookout which affected her navigation at C-5 and C-4 the error of EVER SMART in her lookout was, in my judgment, much worse. Whilst EVER SMART assumed what ALEXANDRA 1 was doing without checking by any visual or radar lookout ALEXANDRA 1 critically assessed the progress of EVER SMART down the channel by means of a good visual lookout but her assessment was in error because of her poor aural lookout.

In addition EVER SMART had allowed herself to be set to the portside of the dredged channel and failed, after the pilot had left the bridge at C-9, to take effective action to bring EVER SMART to the starboard side of the channel. ALEXANDRA 1 failed to alter course to starboard at C-5 or C-4 to head towards the buoys so as to be on the starboard side of the approaches to the channel but that was the result of her error in her aural lookout. It was not a separate and distinct fault as was EVER SMART's failure to keep to the starboard side of the channel.

In terms of causative potency, whilst EVER SMART's failure to navigate on the starboard side of the channel created a risk of collision with ALEXANDRA 1, ALEXANDRA 1's failure to alter course to starboard at C-5 or C-4 increased that risk of collision by causing ALEXANDRA 1 to head across the approaches to the channel. I do not think that one can characterise the fault of ALEXANDRA 1 as a failure “to react properly to a situation of difficulty or danger created” by EVER SMART (the wording being taken from my summary of Sir Henry Brandon’s principles or broad lines of approach in The Nordlake and Seaeagle at paragraph 149(iv)). Rather, ALEXANDRA 1’s fault stemmed from her poor lookout at C-12 and was not in substance a failure to react properly to a situation of difficulty or danger created by EVER SMART. Whilst the situation of difficulty or danger created by EVER SMART was that she was on the portside rather than the starboard side of the channel, from ALEXANDRA 1’s (mistaken) point of view the difficulty or danger was that EVER SMART was not turning further to port early enough. I therefore do not consider that this is one of those cases where one say that there is a marked difference in quality between the contribution which each vessel made to the fact that the collision occurred. But the unsafe speed of EVER SMART contributed far more to the damage resulting from the collision than the very much lower (and safe) speed of ALEXANDRA 1. This is apparent when one looks at the damage done to the bows of ALEXANDRA 1 and at the much lesser damage to the port bow of EVER SMART. It must follow that the causative potency of EVER SMART’s fault was greater than that of ALEXANDRA 1.

Thus in terms of culpability and causative potency there was, to use Sir Henry Brandon’s phrase, a “clear preponderance of fault” on the part of EVER SMART. I have asked myself, as Admiralty judges have frequently done, how many times greater the fault of EVER SMART was than that of ALEXANDRA 1. This can only be assessed “somewhat broadly and upon common sense principles” (see The Volute [1922] 1 AC 129 at p.144 per Lord Birkenhead). To conclude that the degree of EVER SMART’s fault was no more than three times as great as that of ALEXANDRA 1 would, I think, underplay the extent to which in terms of culpability and causative potency the fault of EVER SMART exceeded that of ALEXANDRA 1. In my judgment the degree of EVER SMART’s fault was at least four times as great as that of ALEXANDRA 1. I have asked myself whether it was more than four times as great but I have concluded that it was not. As I have said the culpability of ALEXANDRA 1's fault was substantial. Her master could have corrected his misunderstanding of the VHF conversation at any time after C-12. Instead, he allowed his misunderstanding to affect his navigation and in consequence it caused him to fail to turn to starboard at C-5 or C-4 when he would otherwise have done so. That failure contributed to the fact that the collision occurred. The assessment of relative liability which is required by section 187 of the Merchant Shipping Act 1995 requires that account be taken of that fault.

In my judgment EVER SMART should bear 80% of the liability for the collision and ALEXANDRA 1 should bear 20% of the liability for the collision.