His Honour Judge Havelock-Allan Q.C. :

1. The claim in this action arises out of a Shelltime 4 charterparty, dated 14^th June 1996, (“the charterparty”) whereby Carmine Shipping Limited (“the owners”) let their vessel “Eleuthera” to Hanjin Shipping Limited (“the charterers”) for a period of 3 years, 30 days more or less in charterers’ option. The Eleuthera was delivered to the charterers on 1^st August 1996 and was redelivered to the owners on 2^nd July 1999. During the second and third years of the fixture, the charterers made deductions from the amount of hire which they paid each month. These amounted in total to a sum of US$1,156,956.59. After the vessel was redelivered, the charterers admitted that most of the deductions had been made on an incorrect basis, but claimed that a substantial part of them could still be justified by the fact that the vessel had failed to achieve her warranted speed and fuel consumption in the relevant period. The charterers reimbursed a sum of US$427,000 to the owners, but maintained that they were entitled to withhold the balance. A dispute arose over the final hire statement. By the beginning of the trial that dispute was worth US$590,583.39. A number of items of off-hire and expense were agreed shortly before the trial. That left the vessel’s performance and the treatment of interest on sums wrongly deducted from hire as the two main bones of contention. These are the two matters addressed in this judgment. The parties concentrated their arguments on certain issues of principle. I have not been troubled with the rival figures. The financial implications of my decision on the issues of principle will be for the parties and their advisers to work out.

The terms of the charterparty

2. Four clauses of the charterparty are of particular relevance. Two of them, clauses 9 and 24, are modified versions of the printed text of the Shelltime 4 form. It will help if I reproduce them here exactly as they appear in the charterparty, using italics to identify the parts inserted by the parties into the printed text, and including those parts of the printed text which the parties deleted. I shall also cross-reference clause 24 to the numbered lines of the Shelltime 4 form. The third clause, clause 44, is a rider clause drafted by the parties to contain the speed and consumption warranties. It is to be read in conjunction with clause 24. Clause 82 is also a rider clause. Its significance is explored later in this judgment.

3. The clauses read as follows:

“Payment of Hire

9. Subject to Clause 3(iii), payment of hire shall be made in immediately available funds to: Carmine Shipping Ltd. Account No. 5-840-51338-481-0179 Euromerchant Bank S.A., 75 Akti Miaouli Street, 18537 Piraeus, Greece. Ref: M/T ELEUTHERA .... per calendar month in advance, less:

(i) any hire paid which Charterers reasonably estimate to relate to off-hire periods, provided Owners’ consent has been previously given, and

(ii) any amounts disbursed on Owners’ behalf, any advances and commissions thereon, and charges which are for Owners’ account pursuant to any provision hereof; and

(iii) any amounts due or reasonably estimated to become due to Charterers under Clause 3(ii) or 24 hereof,

any such adjustments to be made at the due date for the next monthly payment after the facts have been ascertained. Charterers shall not be responsible for any delay or error by Owners’ bank in crediting Owners’ account provided that Charterers have made proper and timely payment.

In default of such proper and timely payment,

(a) Owners shall notify Charterers of such default and Charterers shall within seven days of receipt of such notice pay to owners the amount due including interest, failing which Owners may withdraw the vessel from the service of Charterers without prejudice to any other rights Owners may have under this charter or otherwise; and

(b) Interest on any amount due but not paid on the due date shall accrue from the day after that date up to and including the day when payment is made, at a rate per annum which shall be 1% above the U.S. Prime interest Rate as published by the Chase Manhattan Bank in New York at 12:00 New York time on the due date, or, if no such interest rate is published on that day, the interest rate published on the next preceding day on which such a rate was so published, computed on the basis of a 360 day year of twelve 30-day months, compounded semi-annually.

Detailed Description and Performance

24. (a) Owners guarantee that the speed and consumption of the vessel shall be as follows: See Clause 44

The foregoing bunker consumptions are for all purposes except cargo heating and tank cleaning and as described in main details of vessel as given and shall be pro rated between the speeds shown. [lines 308-309]

The service speed of the vessel is 13.0 knots laden and 14.0 knots in

ballast and in the absence of Charterers’ orders to the contrary the vessel shall proceed at the service speed. However, if more than one laden and one ballast speed are shown in the table above Charterers shall have the right to order the vessel to steam at any speed within the range set out in the table (the “ordered speed”). [lines 310-313]

If the vessel is ordered to proceed at any speed other than the highest speed shown in the table, and the average speed actually attained by the vessel during the currency of such order exceeds such ordered speed plus 0.5 knots (“the maximum recognised speed”) then for the purposes of calculating any increase or decrease of hire under this Clause 245 the maximum recognised speed shall be used in place of the average speed actually attained. [lines 314-318]

The vessel to perform at speeds within the range given. However, if during the currency of the Charter Party the vessel can steam at a higher speed, then Owners will advise Charterers immediately of her new capability.

Minimum speed about 9 knots in ballast on 51 MT HFO + 2.5 MT MDO and about 8 knots laden on 55 MT HFO + 2.5 MT MDO

For the purposes of this charter the “guaranteed speed” at any time

shall be the then-current ordered speed or the service speed, as the case may be. [lines 319-320]

The average speeds and bunker consumptions shall for the purposes. of this Clause 24 be calculated by reference to the observed distance from pilot station to pilot station on all sea passages during each period stipulated in Clause 24(c), but excluding any time during which the vessel is (or but for Clause 22(b)(i) would be) off-hire and also excluding “Adverse Weather Periods”, being (i) any periods during which reduction of speed is necessary for safety in congested waters or in poor visibility (ii) any days, noon to noon, when winds exceed force 5 on the Beaufort Scale for more than 12 hours. [lines 321-326]

(b) If during any year from the date on which the vessel enters service (anniversary to anniversary) the vessel falls below the performance guaranteed in Clause 24(a) then if such shortfall results

(i) from a reduction in the average speed of the vessel, compared to the speed guaranteed in Clause 24(a), then an amount equal to the value at the hire rate of the time so lost or gained, as the case may be, shall be deducted from or added to the hire paid;

ii) from an increase in the total bunkers consumed, compared to the total bunkers which would have been consumed had the vessel performed as guaranteed in Clause 24(a), an amount equivalent to the value of the additional bunkers consumed or the bunkers saved, as the case may be, based on the average price paid by Charterers for the vessel's bunkers in such period, shall be deducted from or added to the hire paid. [lines 327-336]

The addition to or deduction from hire so calculated for laden and ballast mileage respectively shall be adjusted to take into account the mileage steamed in each such condition during Adverse Weather Periods, by dividing such deduction by the number of miles over which the performance has been calculated and multiplying by the same number of miles plus the miles steamed during the Adverse Weather Periods, in order to establish the total addition to or deduction from hire to be made for such period. [lines 337-341]

Reduction of hire under the foregoing sub-Clause (b) shall be without prejudice to any other remedy available to Charterers. [lines 342-343]

(c) Calculations under this Clause 24 shall be made for the yearly periods terminating on each successive anniversary of the date on which the vessel enters service, and for the period between the last such anniversary and the date of termination of this charter if less than a year. Claims in respect of reduction of hire arising under this Clause during the final year or part year of the charter period shall in the first instance be settled in accordance with Charterers' estimate made two months before the end of the charter period. Any necessary adjustment after this charter terminates shall be made by payment by Owners to Charterers or by Charterers to Owners as the case may require. [lines 344-350]

Payments in respect of increase of hire arising out of this Clause shall be made promptly after receipt by Charterers of all the information necessary to calculate such increase. [lines 351-352]

44. SPEED/BUNKERCONSUMPTION

Owners guarantee the vessel's speed and consumption in metric ton per day, in moderate weather conditions up to and including Beaufort Scale Five (5), shall be as follows:-

Ballast

12.50 knots on 59 MT HFO + 2.50 MT MDO

13.00 knots on 62 MT HFO+2.50 MT MDO

13.70 knots on 67 MT HFO + 2.50 MT MDO

14.00 knots on 69 MT HFO + 2.50 MT MDO

Laden

12.50 knots on 65 MT HFO + 2.50 MT MDO

13.00 knots on 69 MT HFO + 2.50 MT MDO

Loading : 23 MT HFO + 2.5 MT MDO per 24 hours

Discharging : 65 MT HFO + 2.5 MT MDO per 24 hours

Idle in port : 15 MT HFO + 2.5 MT MDO per 24 hours

Tank cleaning : About 250 MT HFO per operation

Ballasting : About 30 MT HFO

Deballasting : About 25 MT HFO

Crude oil washing : 15MT HFO basis 8 Hours

82. MINIMUM SPEED (SLOW DOWN/SPEED-UP)

The vessel is only to be operated within the range of speeds as given. Owners to warrant that vessel is capable of maintain minimum 9 Knots in ballast, and 8 Knots in laden.”

The vessel’s performance

4. There is no issue that the vessel performed according to the warranties in clause 44 of the charterparty during the first year i.e. between August 1996 and July 1997. The allegation that the vessel under-performed arises only in respect of the second and third years of the charterparty and concerns voyages 9 to 23. The first dispute is whether and to what extent the Master was ordered by the charterers to proceed at a speed other than the service speed on these voyages.

5. According to clause 24 of the charterparty, the vessel’s guaranteed speed, in default of any express order to proceed at a different speed, was to be her service speed. The guaranteed consumption was the consumption corresponding to the service speed, or to the ordered speed if different. If the ordered speed was a speed within the range set out in clause 44 but not expressly specified in the clause, the guaranteed consumption was to be calculated pro rata from the consumption for the next higher speed and the next lower speed given in clause 44.

6. Issues 1 to 4 in the parties’ agreed list of live issues on the performance claim are framed in the following terms:

“1. Whether abstract fixture telexes sent to the aster constitute orders to the Master, in particular for purposes of:

(a) Instructions on speed “about 13 knots”;

(b) Instructions on laycan.

2. Whether references to laycan constitute an order to the Master to arrive within that laycan period.

3. If so, where there were instructions to meet a particular laycan (whether in abstract fixture telexes or voyage instructions/orders), whether those periods should be included in any assessment of the average speed and consumption in good weather conditions.

4. If those periods are to be taken into account, whether the vessel’s performance is to be judged according to:

(i) Whether she met the first moment of the laycan date (i.e. arrival at any later time than commencement of laytime constitutes under performance); or

(ii) Whether she arrived during the laycan period (i.e. that she only under performs if she misses the last moment of the laycan period).”

7. The background to issues 1 and 2 is that on every laden passage from voyage 9 to voyage 22 the Eleuthera was carrying one or more cargoes of crude oil under a voyage charterparty concluded for that voyage. In the majority of cases, the charterers sent the Master details of the terms of the voyage charterparty. These usually took the form of a telex summarising the principal terms of the fixture or reproducing the recap of the fixture exchanged between the charterers and the voyage charterers or between their respective brokers. The telex (“the recap telex”) was usually sent to the Master during or shortly after the discharging of the previous cargo. The recap telex contained two provisions of significance in the present context. First, it informed the Master of the laycan date or spread of laydays by which, or within which, the vessel had to arrive at the next port of loading and tender notice of readiness. If the vessel failed to do so the charterers risked the voyage charterparty being cancelled by the voyage charterers. Second, it informed the Master in most cases but not in every case, that it had been agreed in the voyage charterparty that the vessel was to perform the loaded passage at a speed of “about 13 knots”. Thus, to take one example, the vessel completed voyage 13 by discharging the second of two consignments of crude oil at Kaohsiung in Taiwan on 4^th April 1998. She then waited off Kaohsiung for further instructions. On 7^th April 1998, the charterers telexed to the Master a copy of the recap of the fixture they had concluded with Chinese Petroleum Corporation for voyage 14. The fixture was for loading at one or two safe ports in the Arabian Gulf and specified a laycan date of 20^th April 1998, basis Ras Al Khafji. According to the recap telex, the voyage charterparty was to be on the ASBA II Tankvoy Form with the voyage charterers’ rider clauses amended in various respects. One amendment was to specify in rider clause 16 that the vessel was to perform the voyage at “about 13 knots”.

8. The owners’ case, quite simply, is that in every instance where voyage charterparty details of the above kind were sent to the Master they constituted an order (1) to perform the laden voyage at a speed of about 13 knots; and (2) to proceed on the voyage to the loading port at whatever speed would permit the vessel to meet the cancelling date. It is agreed that the word “about” in respect of a vessel’s speed imports a tolerance of half a knot. So an order to proceed at about 13 knots would be an order to proceed at not less than 12.5 knots. That would be an order to steam at less than the laden service speed of 13 knots. Further, if the vessel could travel the distance to the port of loading in time to meet the cancelling date at a speed less than the relevant service speed, an order to meet a laycan date would be an order to proceed at less than the service speed. The charterers’ argument is equally simple. They say that the recap telexes were sent to the Master for information only. They did not purport to be and did not in fact contain any orders.

9. The picture is complicated by the fact that in every case the recap telex was followed by at least one other message from the charterers to the Master which was headed “Voyage Order” or “Voyage Instruction”. Frequently these messages were silent as to the vessel’s speed, although invariably they repeated the laycan date or laydays at the first and any subsequent ports of loading. Thus the first question is whether, in default of an express instruction in the message headed “Voyage Order” or “Voyage Instruction” as to the speed to be adopted on the laden passage, the Master was entitled to treat the reference to speed in the recap telex as an order to follow that speed. If he was not, the only alternative is that he was obliged to follow the service speed of 13 knots provided in the charterparty.

10. An additional feature of the exchanges between the charterers and the Master is that, on each of the relevant voyages, the Master was accustomed to sending a message to the charterers on sailing from the last or only port of loading, which was in the nature of a sailing report. It gave details of the cargo which had been loaded and of the time taken to load it. It also confirmed the Master’s understanding of the instructions he was to follow in proceeding to the port of discharge. The Master throughout this period was Captain Ioannis Iliopoulos, save on one voyage. On voyage 16, Captain Iliopoulos was on leave and Captain Psarros was in command. In all but one of the sailing reports which Captain Iliopoulos sent to the charterers, he stated the ordered speed as being “13 knots”, even where the recap telex had indicated that the speed inserted in the voyage charterparty was “about 13 knots”. The exception was voyage 19, where the sailing report made no mention of speed. However on that occasion, Captain Iliopoulos had already received voyage orders from the charterers which expressly instructed him to perform the laden voyage at 12.5 knots.

11. In deciding what interpretation to place on the recap telexes, the starting point to my mind is that it is incumbent on a time charterer to give instructions to the Master which are clear and which can readily be understood. If there is room for doubt, the Master should seek clarification: but he is entitled to the benefit of the doubt if he interprets the communications sent to him in a way which is plausible and not obviously wrong. Here there was room for doubt. The Master was sent a recap telex which specified a speed different from the service speed in the charterparty. He was then sent voyage instructions which in most cases were silent as to the speed at which the laden voyage was to be performed. In the oil trade the time of arrival of a cargo at the port of discharge is often price sensitive. It is no doubt for that reason that it is common for tanker voyage charterparties to include an agreement between the owners or the disponent owners and the voyage charterers as to the speed at which the voyage is to be performed. There is a strong case for saying that where, as here, the latter speed is expressed as the time charterparty service speed qualified by the word “about”, it should be apparent to the Master that all the charterers are doing is to cover themselves in the voyage charterparty for any shortfall in performance under the time charterparty, but that the time charterparty service speed remains the contractual obligation as between the charterers and the owners. That certainly seems to have been the impression gained by Captain Iliopoulos. I reject the evidence given in his statement that he omitted the word “about” from the speed acknowledged in his sailing reports because they were only brief reports. They were precise in all other respects, even if brief, and Captain Iliopoulos was well aware of the difference between “13 knots” and “about 13 knots”. However it was apparently not the impression gained by Captain Psarros. The recap telex which he was sent indicated that the speed agreed in the voyage charterparty for voyage 16 was about 13 knots. His subsequent voyage instructions said nothing about speed and in his sailing report Captain Psarros, relying on the recap telex, recorded the ordered speed as being 12.5 knots.

12. Miss Birch, who appeared for the owners, submitted that whatever interpretation Captain Iliopoulos or Captain Psarros may have placed on the communications he received from the charterers cannot determine what instruction he in fact received as to the vessel’s speed. I agree. But it does not mean that the sailing reports are irrelevant. The charterparty plainly stated that in the absence of a contrary instruction, the vessel was to perform all laden sea passages at 13 knots. Unless there was a clear contrary instruction, 13 knots was the speed to be achieved. In my judgment, the recap telexes did not contain a clear contrary instruction. They were communications which provided the Master with details of the vessel’s next fixture, rather than instructions as to what he was to do. Whilst the format of the recap telexes was not uniform, the most that can be said of any of them was that it had the potential to cause doubt in the mind of the Master as to whether the ordered speed was to be less than the service speed. I might have been prepared to give the owners the benefit of such doubt if the recap telexes had in fact caused confusion in the mind of Captain Iliopoulos. But I do not think that they did. His sailing reports state clearly that he regarded the ordered speed as being 13 knots. Evidence was given by the charterers’ expert, Mr Robert Millard of Burls Gordon Ltd., that the consumption figures in the vessel’s engine logs also showed that this was the speed which Captain Iliopoulos tried to achieve on every laden passage where no contrary order was given. I place less weight on this evidence than on the sailing reports because the consumption figures may be imprecise, and the load settings on the main engine do not necessarily corroborate that the vessel was striving to maintain 13 knots. Nevertheless I am satisfied that 13 knots is the speed which Captain Iliopoulos believed that the vessel should achieve. My conclusion is that, save where other orders were expressly given in other messages from the charterers, the required speed on laden passages was the laden service speed of 13 knots and not the speed given in the recap telexes. In other words, my answer to issue 1(a) is in the negative. However, I hold that the vessel’s performance on the laden passage on voyage 16 should be judged by reference to a speed of 12.5 knots. Captain Psarros was misled by the recap telex into thinking that that was the required speed. He stated as much in his sailing report and the charterers did nothing to disabuse him.

13. The pattern of communications between the charterers and the Master was exactly as I have described in relation to voyages 9, 11, 13, 14 and 20. There were departures from the pattern in respect of other voyages which, in my judgment, make no difference to the conclusion I have already expressed. As to voyage 10 the Master received an express order “… to perform voyage on laden passage (after last loading port) at 13.0 knots, weather and safe navigation permitting”. The order went on to say “Vessel not to speed up unless instructed by charterer”. I am unable to accept the owners’ submission that this meant that the originally ordered speed was less than 13 knots. I think that the inclusion of this phrase is attributable to the charterers’ mistaken assumption (reflected in the terms as to speed which they agreed in the voyage charterparties for voyages 19 and 21) that they were entitled to order the vessel to proceed at more than 13 knots. In fact the vessel was not asked to speed up on voyage 10. Accordingly, 13 knots was the required speed on the laden passages of voyage 10 because it was the speed expressly ordered. The telex of voyage instructions given to the Master for voyage 12 merely said “Speed: as per c/p”. I would regard that as equivocal. As between the owners and the charterers it was most appropriate to construe it as a reference to the time charterparty: but in context it could have been read as referring to the voyage charterparty. In his sailing report, Captain Iliopoulos described the ordered speed as 13 knots rather than about 13 knots. In the circumstances I consider that 13 knots was the required speed.

14. The voyage order given for voyage 15 instructed the vessel to proceed “at maximum speed under the agreed terms of the charterparty” after completion of loading at the second loading port. That was plainly an express order to proceed at 13 knots on voyage 15L2; but it gave no guidance as to how the vessel was to perform between the two loading ports i.e. on voyage 15L1. The telex recap had indicated that it was agreed in the voyage charterparty that the vessel should perform the laden passage at about 13 knots. An express instruction had been given to proceed more slowly than that on the ballast voyage to the first loading port. Captain Iliopoulos recorded in his sailing report that the ordered speed was 13 knots, but that report was not sent until the vessel was fully loaded. There was more than one port of loading on all of the voyages save voyage 11, voyage 19 and voyage 20. So the question of the required speed between loading ports arises in respect of all the other voyages as well. My view is that, absent an express order, the position in respect of part laden passages is in principle the same as that for laden passages. The laden service speed applied. Whilst the sailing report does not state what the Master thought was the ordered speed when the vessel was only part laden, he can hardly have been confused into thinking that it was less than the speed fully laden. It follows that 13 knots was the required speed on voyage 15L1 and on all the other part laden passages where the fully laden speed was 13 knots. By the same token, the required speed on voyage 16L1 was 12.5 knots.

15. Other variations in the pattern of communications with the Master occurred on voyages 17, 18, 19, 21 and 22. The copy in the trial bundles of the telex recapping the voyage charterparty for voyage 17 is incomplete. It is not clear whether it referred to any agreement between the charterers and the voyage charterers as to the vessel’s speed. However the voyage charterparty itself was in evidence. It contained amongst the rider clauses a “Speed Clause” which expressly provided: “Vessel to perform the laden voyage at about 13.0 knots weather and safe navigation permitting”. Whether or not the Master was sent a copy of the voyage charterparty, which I doubt, I am prepared to assume that he did receive the full version of the recap telex and that it mentioned the speed clause. Nevertheless, for the reasons already given in paragraph 12 of this judgment, I think that 13 knots was the laden speed required under the charterparty. The voyage instructions given in the next message which the Master received said nothing about the vessel’s speed, and the Master recorded in his sailing report that the ordered speed was 13 knots.

16. On voyage 18 the vessel was expressly instructed to proceed at a reduced speed of 12.5 knots to the first port of loading. The order was given on 20^th October 1998 when the vessel had already passed the Singapore Straits and was heading in ballast towards the Arabian Gulf. The following day the Master received a message giving details of the voyage charterparty. The message was brief. It referred to the range of loading and discharge ports, the laycan and the laytime. It did not say whether any speed had been agreed with the voyage charterers. The message ended by saying “Otherwise as per Eleuthera/Hanjin CP dd. 18 Aug 1998”. That was a reference to the immediately preceding voyage charterparty and to its rider clauses, including the speed clause. The subsequent communications between the charterers and the ship were similar to those on voyage 17. I do not think that a telex recap referring back to the previous voyage charterparty was any more an instruction to the Master than a telex recap which expressly set out all of the principal terms of the fixture including any terms agreed between the charterers and the voyage charterers as to the vessel’s speed. Once the vessel had loaded cargo she was required to proceed at the service speed of 13 knots. She actually achieved a greater speed between the first and second port of loading. So no complaint is made by the charterers as to voyage 18L1. On sailing from the second port of loading the Master acknowledged in sailing report that the ordered speed was 13 knots, which I find was the case. A claim does arise in respect of the vessel’s performance on the fully laden passage (voyage 18L2).

17. According to the telex recap, the voyage charterparty for voyage 19 contained a clause in the following terms: “Base speed 13 knots WSNP – with ref clause 37 C/P charterers’ option to increase or decrease base speed by 0.5 knots free of charge”. No issue arises as to whether the charterers were entitled, under the charterparty, to order the vessel to steam at more than 13 knots fully laden because, in the event, no such order was given. What happened is that, on 23^rd December 1998, as the vessel was proceeding in ballast to the port of loading, the Master received voyage orders expressly instructing him to proceed on the laden passage at 12.5 knots and informing him that the “target discharge date” was 0000 hours on 12^th January 1999 at Ain Sukhna. He was asked to confirm the ordered speed and the average speed required to meet the target date and time at noon that day in each of his regular Thursday and Sunday reports to the charterers after sailing from the port of loading. The only example of such a report which was in evidence was the report of 3^rd January 1999. At noon on that day the average speed required to meet the target date was only 10.6 knots. In the event the vessel reached Ain Sukhna on 10^th January 1999, some 31 hours and 30 minutes early. She had performed the voyage at an average speed of 12.37 knots. Had she achieved an average speed of 12.5 knots she would have been even earlier. This raises, in the context of a laden passage, an issue very similar to that which arises on most of the ballast passages where there was a laycan date or spread of laydays which the vessel was expected to meet. To my mind, the contractual position on voyage 19L is straightforward. The vessel was ordered to steam at 12.5 knots. The target date was for information purposes only. It did not override what was a clear express order as to the speed. No doubt the request by the charterers that the Master inform them in his reports of the average speed required to meet the target date was to enable them to assess how early she would arrive if the ordered speed was left unchanged and to decide whether, and if so by how much, they should order the vessel to reduce speed further. However the vessel never was ordered to reduce speed further. The ordered speed remained throughout a speed of 12.5 knots. If, in accordance with calculations performed in accordance with clause 24 of the charterparty, the vessel failed to achieve that speed on voyage 19L, the charterers are entitled to compensation. A separate question is whether it is arguable that the charterers suffered no loss because the vessel met her target date. That point was not argued specifically in relation to voyage 19L and there was no investigation of the factual circumstances following the vessel’s arrival at Ain Sukhna. A similar point was expressly taken in respect of those ballast voyages where the vessel arrived at the port of loading before the laycan date or before the first of the laycan dates. It will be clear from what I have to say about the ballast voyages that I do not think that the point affords any answer to a claim for compensation under clause 24.

18. The recap telex in respect of the voyage charterparty for voyage 21 is not relevant because the Master was given an express order about speed in the voyage orders subsequently sent to him. The vessel loaded cargo at three ports on voyage 21. In the course of the ballast voyage to the first of these ports, the Master received voyage orders which included the following: “Vessel to perform voyage on laden passage (after last loading port) at about 13.5 knots, weather and safe navigation permitting. Vessel not to speed up unless instructed by charterer”. The owners’ expert, Mr Tony Watson of BMT Murray Fenton Ltd., assumed that this must have been a mistake for “about 12.5 knots” because 13.5 knots was in excess of the maximum laden speed in Clause 44 of the charterparty. I do not think that that is a safe assumption. The terms of the voyage charterparty concluded for voyage 19 demonstrate that the charterers were prepared to fix the vessel on terms which represented (erroneously) to the voyage charterers that they could require the vessel to steam at more than 13 knots when laden. The voyage charterparty for voyage 21 was not produced. It is quite possible that it too held out the prospect to the voyage charterers that the vessel could be ordered to steam at more than 13 knots. However Clause 24 of the charterparty, read in conjunction with Clause 44, is clear. The vessel was to perform at a maximum laden speed of 13 knots. That was the maximum guaranteed speed. Any order to proceed at a greater speed was not a legitimate order. It was not an order with which the vessel was bound to comply and the charterers could have no complaint if, the Master having agreed to try to comply with it, the vessel failed to do more than an average of 13 knots. On 22^nd March 1999, which was about 5 days into voyage 21L, the charterers requested the Master to “speed up more as possible as you can”. But an analysis of the Logs suggests that she had not achieved more than about 12.75 knots up to that point. So the order does not indicate whether the charterers wanted her to achieve more than 13 knots. In his sailing report, the Master recorded the ordered speed on voyage 21L as 13 knots. In my view he was right.

19. Nothing was said about the laden speed on voyage 22, either in the recap telex or in the voyage instructions. In default of any mention of speed at all, the service speed of 13 knots would have been the required speed. The vessel was initially ordered to proceed to Mina al Ahmadi in Kuwait as the first port of loading and then to Mina al Fahal in Oman as the second port of loading. This order was given before she had reached the port of discharge on voyage 21. After discharging at Yosu, the vessel proceeded back to the Arabian Gulf in ballast. On 10^th May 1999 the Master received fresh orders. The first was to proceed to Ash Shihir in Yemen and await further orders. The second was in these terms: “Pls proceed to Ash Shihir, Yemen for Order at economic speed. Pls adv us of ETA off Ash Shihir by full spd n economic speed respectively”. These orders arrived at or about the time that the vessel actually reached Ash Shihir. The vessel was then ordered to load at Ash Shihir before proceeding to Mina al Fahal to load a second consignment. She sailed from Mina al Fahal on 29^th May with a full cargo. There can be no doubt that the required speed on the fully laden passage from Mina al Fahal (voyage 22L2) was 13 knots. The vessel achieved this speed and also achieved it on the part laden passage from Ash Shihir (voyage 22L1). So the charterers make no complaint about voyage 22. The only issue raised by the owners was whether the order to proceed at “economic speed” applied to the ballast voyage from Singapore to Ash Shihir (voyage 22B2) or to voyage 22L1. I think that it applied to the former rather than to the latter, but was received so late in the voyage that it was of no practical effect.

20. My definitive answer to issue 1(a) is that the required laden speed on all of the voyages was 13 knots, with the exception of voyages 16 and 19 where it was 12.5 knots for the reasons stated. I now turn to issues 1(b), 2, 3 and 4. The significance of laycan dates arises in respect of all of the ballast passages preceding voyages 10 to 22, and the part laden passages on those voyages on which there was more than one port of loading (all of the voyages from voyage 9 to 22, except voyages 11, 19 and 20).

21. At the outset I should say that I am quite satisfied that the correct answer to issue 2 is “No”. The answer to issue 1(b) is less straightforward. The primary function of the recap telexes was to give the Master details of the vessel’s next employment. For the reasons already outlined I am not persuaded that telling the Master the terms of the voyage charterparty constituted an order to him as to how the next voyage was to be performed. Voyage orders performed that function and in every case a message containing voyage orders followed. Where that message was silent about speed, the charterparty rather than the voyage charterparty was the governing document. The one qualification I acknowledge is where the recap telex was the only communication which told the Master where the vessel was to proceed after completion of discharge. In those cases I agree that the recap telex must be understood as an order to the Master to proceed to the place or port in question. If that is so, the owners say that it was surely also an order to reach that place or port by the stated laycan date or within the spread of laydays. Let me accept, for a moment, that that proposition is correct. In my judgment it still does not mean that the vessel was being given an order to reduce speed if she could meet the laycan at the named destination by steaming at less than the service speed, laden (on part laden passages) or in ballast.

22. There are a number of reasons for this conclusion. The first is that the Master is only entitled to reduce the vessel’s speed if he is clearly told to do so. The giving of a laycan date is not a clear instruction to reduce speed. At most it is an implied instruction that the speed may be reduced. The implication is one which depends upon the circumstances. One circumstance to which Miss Birch drew attention was the fact that most of the voyage charterparties under which the Eleuthera was operating contained a term to the effect that the vessel was not authorised to tender Notice of Readiness at the port of loading before the laycan date or the first of the laydays if there was a spread. She submitted that the Master could infer from this provision that there was no advantage to the charterers in the vessel reaching the port of loading before the first layday or the laycan date. I do not think that that follows or that the inference would be a proper one to draw in the absence of further evidence. It is not uncommon for arrangements at loading terminals to be altered if a vessel arrives early. Stems can sometimes be advanced. An earlier stem might result in the voyage being completed sooner and might allow for more voyages or longer voyages to be accomplished within the period of the charterparty. And in cases where there was no clause precluding the tender of an early Notice of Readiness (the voyage charterparty for voyage 22 is an example), it might well be in the charterers’ interest to have the vessel loaded early or have her wait at the port of loading and earn demurrage. Second, the Master cannot anticipate what may happen on the voyage, so as to be able with any certainty to calculate what reduction in speed he may safely adopt in order to arrive at the named destination on the due date. Planning of that kind is not an exact science. The Master does not know what weather conditions or other hazards he may encounter which will spoil the calculation. He cannot always foresee what the charterers have in mind to do with the vessel. They might be planning to order her to stop on route for bunkers or to deviate for some other purpose. So if an order to meet a particular laycan is an order to reduce speed to the minimum speed necessary to meet that laycan, the required reduced speed will remain uncertain throughout the voyage. The problems of calculation are compounded in the case of a part laden passage between two ports of loading since the vessel’s performance capability will be different according to the size and distribution of the cargo in her tanks. This leads me to the third reason. The uncertainty to which I have just referred would make it very difficult to assess the vessel’s performance to warranty under clause 24 of the charterparty in respect of any voyage on which the Master had been told meet a particular laycan date. The truth is that if the giving of a laycan date is an order to arrive by that date, it is an order which is subject to the qualification that it should be possible to meet the date at the maximum guaranteed speed. The charterers have no right to order the Master to meet a laycan date if that would require the vessel to exceed the maximum guaranteed speed. That being so, any order to meet a laycan date does not override the service speed, it is an additional order which is subject to the service speed, or subject to any lesser speed which has been expressly ordered. Viewed in this light I would hold that a request to the Master to proceed to a particular destination by a laycan date is not an order as to speed. It may certainly be an order but its validity, or to put it another way, the extent of the owners’ potential liability for failing to comply with it, depends on whether the date can be met by steaming at the service speed or at any lesser speed which has been expressly ordered.

23. In the circumstances, issues 3 and 4 do not arise. The service speed of 14 knots was the required speed on all the ballast passages, except on those voyages where an express order for a lesser speed was given. The service speed of 13 knots was the required speed between ports of loading. The voyages where an express order was given for a lesser speed on the ballast passage were voyages 12B1, 13B, 15B, 18B and 22B2. The order given on voyage 13B on 10^th February 1998 was to slow down in order to reach the loading port of Djeno on 21^st February. The Master responded that he would reduce the vessel’s speed to an economical speed of a minimum 12.5 knots in good weather in order to reach Djeno on the morning of 21^st February. On voyage 18B the vessel was ordered to proceed to Umm Said at a reduced speed of 12.5 knots from 1600 hours on 20^th October 1998. On those two voyages I have no difficulty concluding that the required good weather speed was 12.5 knots from the time the order was received. Voyages 12B1, 15B and 22B2 are controversial because the order given by the charterers in the course of those ballast passages was to reduce the vessel’s speed to an “economic speed” (voyages 12B1 and 22B2) or a “minimum speed” (voyage 15B). There is a dispute as to what that meant and as to how an order of that kind related to the laycan date. This is the subject of issues 5 and 6. I will come to issues 5 and 6 later.

24. In case I am wrong about issue 2, I will give my answer to issues 3 and 4 briefly. On issue 3 there was a difference between Mr Millard, who favoured the view that the passages should be included in the calculation of the annual average under clause 24 by classifying them as reduced speed passages, and treating the actual average speed achieved as the guaranteed speed, and Mr Watson who left them out of account. I favour the former approach, although if the achieved speed is equated with the guaranteed speed, the effect on the average speed is neutral. The same, however, will not necessarily be true of the bunker consumption. There was also one voyage (voyage 21B) where the vessel missed the laycan date. Here both experts rated her performance against the service speed of 14 knots. The vessel’s under-performance on this passage would affect the annual average if it was included. Another approach would be to calculate the minimum average speed which, in the light of the conditions actually experienced on the passage, it was necessary for the vessel to achieve in order to reach the destination at the beginning of the laycan date or on the first of the laydays, and then to judge her performance by reference to that speed. This would result in over-performance on all the ballast passages save voyage 21B. The reason why Mr Watson considered it inappropriate to include any of these passages in the calculation of the annual average was that in his view it was wrong to penalise the vessel for steaming at less than the service speed if in fact she arrived by the laycan date or within the spread of laydays. The rationale for this view was that the charterers will have suffered no loss in those circumstances. Whether there was financial loss might be relevant to a claim for damages for breach of the charterparty e.g. for failing to comply with a legitimate order. I do not think that it is relevant to a claim under clause 24 of the Shelltime 4 form. Clause 24 contains a complete code for compensating charterers (or owners, if the provision for rewarding over-performance had not been deleted as it was here) for time lost and excess consumption measured against the service or ordered speed and corresponding guaranteed consumption rates. The clause is in the nature of a liquidated damages provision. If the vessel is ordered to proceed to a port of loading at the service speed, and proceeds there at less than the service speed, the clause provides a mechanism for assessing the time lost, regardless of whether the vessel may have arrived before the laycan date. The charterer may have had his reasons for not expressly ordering a reduced speed. He is entitled to have the vessel arrive after the time the passage would take at the service speed, even if he has to pay hire whilst the vessel waits idle at the port of loading until the laycan date. Any alternative construction would result in owners second-guessing the charterer’s intentions as to the vessel’s employment and reducing speed by an ill-defined and variable amount in the course of the voyage. This gives rise to the problems already highlighted in paragraph 22 of this judgment.

25. I would answer issue 4 in terms of 4(i). For the reasons given in paragraph 22, it was in the charterers’ interests to present the vessel as early as possible on the laycan date or within the laydays.

26. Issues 5 and 6 are in these terms:

“5. Whether orders to proceed at a “minimum speed” or an “economic speed” are orders to proceed at 12.5 knots.

6. Where there is an order to proceed at a “minimum speed” or “economic speed” as well as a laycan or target date, whether the latter supersedes the former.”

27. The orders given to the vessel to reduce speed on voyages 12B1, 15B and 22B2 were given by Mr Chris Yeo, a vessel operator in the charterers’ tanker team at the material time. He was not called to give evidence but provided a witness statement in which he said that when he referred to “economic speed” or “minimum speed” he meant 12.5 knots. His evidence was that he gave the order to reduce speed to an economic speed on voyage 12B1 by speaking to the Master on the telephone. When he did so he told Captain Iliopoulos that an economic speed meant 12.5 knots. Captain Iliopoulos is ill and was not called to give evidence either. His statement did not deal with any conversations with Mr Yeo. But there is some support for Mr Yeo’s version in the fact that, when the vessel was again ordered to reduce speed on voyage 13B, Captain Iliopoulos reported back that he would reduce speed to an economical speed of 12.5 knots.

28. There was some debate between the experts as to the test of what is economical in this context. Mr Watson’s view was that an economical speed was one which minimised fuel consumption. Mr Millard suggested that consumption was not the only factor. Time was important because of the hire cost. His evidence was that the Eleuthera was most economic at the fastest possible speed, although this proposition was disputed by Mr Watson on the figures. In my judgment an “economical speed” meant a slower speed, which minimised fuel consumption. I do not think that the Master was expected to do a time/consumption analysis and calculate the speed which produced the cheapest fuel cost per mile. Moreover all of the orders to proceed at an economical speed were given in a context where, plainly, the charterers wanted the vessel to slow down. The only question is whether an “economical” ballast speed meant a speed of 12.5 knots or some lower speed. If one leaves aside Mr Yeo’s evidence of his conversation with the Master, there is much to be said for the view that an “economical speed” meant a speed less than 12.5 knots. Although the experts were in agreement that the relationship between speed and bunker consumption was not linear below 12 knots and the saving in fuel dropped as the speed was reduced below 12 knots, there would still be a saving. On the other hand the logic of that view is that the minimum ballast speed of 9 knots would be the most economic. Yet the charterers clearly differentiated between an “economical speed” and “minimum speed” when giving their orders, even though Mr Yeo’s evidence was that he intended both terms to mean the same thing. In truth the term “economical speed” is ambiguous. For that reason I do not find Mr Yeo’s evidence of his conversation with the Master as unbelievable as Miss Birch urged me to hold that it was. On voyage 13B the Master equated an economical speed with a speed of 12.5 knots. My conclusion is that an economical speed meant 12.5 knots in the present case.

29. That still leaves open the question of what Mr Yeo meant by his order on voyage 15B that the vessel was to proceed to the Arabian Gulf “at min. speed” from 1800 hours on 20^th May 1998. “Minimum speed” was defined in clause 24 and clause 82 of the charterparty as meaning a ballast speed of 9 knots or about 9 knots and a laden speed of 8 knots or about 8 knots. I would have thought that an order to proceed on a ballast voyage at “minimum speed” was an order to proceed at 9 knots or about 9 knots, unless it was obvious that the charterers had no right to order the vessel to perform at that speed. Both clause 24 and clause 82 made it quite clear that the vessel was only to be operated at speeds within the given range. Were the minimum speeds in clause 24 and clause 82 part of that “range”? On one view it is hard to see why reference was made in the charterparty to the minimum speeds if they were not intended to mark the lower end of the permitted range. The evidence of Captain Iliopoulos was that the minimum speeds were the lowest speeds at which the vessel could maintain steerage. Even so, their inclusion in the charterparty would appear to serve no purpose if the charterers were not permitted to order the vessel to proceed at such low speeds. However the structure of clause 24 suggests that the minimum speeds were not intended to be part of the range. According to clause 24 (lines 311 to 313): “… if more than one laden and one ballast speed are shown in the table above Charterers shall have the right to order the vessel to steam at any speed within the range set out in the table (“the ordered speed”)”. Only ordered speeds were guaranteed speeds. At the beginning of clause 24, where the table ought to be, the parties inserted the words: “See clause 44”. The minimum speeds were inserted in clause 24 after line 313 and after the reference to the “table above”. In the circumstances, I consider it an inescapable inference that the range of speeds in the table only included those in clause 44. So Mr Yeo was right when he said that he took “minimum speed” and “economical speed” to mean one and the same thing. In this case, the minimum ballast speed which the charterers were permitted to order was 12.5 knots. Nevertheless an order to proceed at “minimum speed” was potentially ambiguous. If Captain Iliopoulos had thought that he was being ordered to reduce speed to 9 knots, I might have been willing to hold that there was no guaranteed speed on the passage in question. But Captain Iliopoulos does not say anything about voyage 15B in his statement and the vessel’s logs show that, after the order was given, the vessel maintained an average speed in excess of 12 knots on an average fuel consumption slightly in excess of that guaranteed in respect of a ballast speed of 12.5 knots. My conclusion, therefore, is that the order to proceed at minimum speed meant, and was understood by the Master to mean, that the vessel was to reduce speed to 12.5 knots.

30. The answer to issue 5 is “Yes”. The answer to issue 6 is “No”. An express order as to speed overrides any order to be derived from the giving of a laycan date, especially if the order to slow down was given after the laycan date was communicated rather than vice versa. An order to proceed at an “economical speed” might be qualified by an express order, given at the same time, to meet a particular arrival date. But no dual order of that kind was given in this case. The nearest was the order given on voyage 13B, which is dealt with in paragraph 23 of this judgment.

31. I turn to issue 7, which is as follows:

“7. Whether, having calculated an average speed based on good weather performance [during each voyage] under clause 24, the calculation of performance of the vessel is to be made by reference to the whole of the steaming distance over the year (anniversary to anniversary) or whether it is to be calculated on a voyage by voyage basis.”

32. I have added the words in brackets in order to highlight the nature of the question. It turns on the construction of lines 321 to 324 of the Shelltime 4 Form which provide that: “The average speeds and bunker consumptions shall for the purposes of this Clause 24 be calculated by reference to the observed distance from pilot station to pilot station on all sea passages during each period stipulated in Clause 24(c), but excluding any time during which the vessel is (or but for Clause 22(b)(i) would be) off-hire and also excluding “Adverse Weather Periods” …”. The question is whether the averages are to be calculated as single averages in respect of laden and ballast passages for each accounting period, by taking the total distance steamed in each condition in good weather in that period and dividing it by the total time taken (“the cumulative average method”), or whether the averages are to be calculated voyage by voyage (“the individual voyage average method”). The two methods produce a different result in terms of over or under-performance even if the ordered speed was a constant speed throughout the year. This is because the average speed over the whole year may be higher or lower than the average speed on a particular passage. The disparity in results between the two methods becomes greater still if the ordered speed varied during the year. A cumulative average is then a much blunter tool with which to measure performance. It will deny a claim for under-performance on a particular passage on which the ordered speed was high, but not as high as the year’s average, and promote a claim in respect of a particular voyage on which the ordered speed was low, but not as low as the year’s average. Judged by reference to the average speed on the passage in question, the pattern of over or under-performance could be significantly different.

33. The owners (with one modification) argued in favour of calculating cumulative averages for each accounting period: the charterers for calculating the averages voyage by voyage. The owners’ modification was that a separate cumulative average should be calculated for each speed ordered during the year in question (“the modified cumulative average method”). Aside from the dispute as to the calculation of averages, there was a good deal of common ground between the experts. It was agreed that the accounting periods over which the vessel’s performance was to be assessed were the 12 month periods between anniversaries of the date of the charterparty, save for the final period which ran from 1^st August 1998 to 2^nd July 1999. It was also common ground that any compensation under clause 24(b)(i) would be payable at the rate of hire applicable at the date of the voyage on which the time was lost (being in this case the rate of hire for the relevant accounting period since the rate only changed on anniversaries of the charterparty) and that any compensation under clause 24(b)(ii) would be payable at the average of the bunker prices paid during the relevant accounting period.

34. Neither of the rival methods of calculating the average is entirely satisfactory. But to my mind the owners’ modified cumulative average method is more consistent with the aim of clause 24, which, in its unamended form, is to compensate for an overall under-performance or over-performance when each year or part year of the charterparty is examined as a whole. Mr Millard said that it was customary to compensate charterers, or to reward owners, on a voyage by voyage basis, taking the average performance on each voyage. I acknowledge Mr Millard’s considerable experience in this field, but have no evidence other than his to suggest that this is the generally accepted practice. Mr Watson disagreed with Mr Millard and said that in the past two years he had done numerous performance calculations under Shelltime 4 charterparties using single averages derived from performance on all passages for a given ordered speed in each accounting period.

35. The cumulative average method has the effect of smoothing out the peaks and troughs of actual performance and establishing a trend. It is simpler and gives proportionate weight to longer sea passages because the overall average is derived from dividing the total distance steamed by the total steaming time. Mr Millard accepted that longer sea passages tended to be a better test of performance because time spent speeding up and slowing down at either end of the voyage assumes less importance. The one disadvantage of the cumulative average method is that, in its basic form, it does not cater adequately for orders to proceed at different speeds where the charterparty contains a range of guaranteed speeds and rates of consumption.. To avoid a distorted result, it is either necessary to exclude from the calculation of the average and from the regime of compensation, passages on which the vessel was ordered to proceed at a speed other than the service speed or the generally ordered speed, or to calculate separate averages for passages performed under different speed orders. The latter (the modified cumulative average method) is the better solution even if it results in a hybrid approach which is part way between the cumulative average method and an individual voyage average method.

36. Individual voyage averages solve the problem of different ordered speeds. An average is calculated for each passage at each speed. This approach rewards under-performance and over-performance more precisely than a single cumulative average, but in doing so it accords equal significance to the performance of the vessel on each mile of a short passage as it does to her performance on each mile of a long passage. Moreover in the case of a charterparty like the present one, where clause 24 has been amended so as to give monetary compensation to the charterers for under-performance but no monetary compensation to the owners for over-performance, individual voyage averages give the owners no credit for any passage on which the average speed or consumption exceeded the speed or consumption guaranteed. On the face of it, the voyage by voyage approach could be used to pick the duff voyages from the plum ones. This is not a characteristic of the cumulative average method because all passages make up the average. Any over-performance results increases the average speed or decreases the average consumption for the period in question, even if there is no entitlement to a monetary reward. Anticipating this objection, Mr Millard conceded in his report that credit should be given for those passages on which the Eleuthera performed better than warranted. His calculations struck a net balance due to the charterers, after the owners had been credited with the financial value of the savings made on the passages where the vessel over-performed. Without that concession, the individual voyage average method would have operated unfairly in this case.

37. The other problem with the individual voyage average method is that it is difficult to apply to sea passages performed entirely in Adverse Weather Periods. Lines 337 to 341 of clause 24 of the Shelltime 4 form require the addition or deduction from hire per mile for good weather periods to be applied to Adverse Weather Periods as well. This process of “grossing-up” presents no difficulty where cumulative averages have been calculated for good weather performance on all sea passages, laden and in ballast, for all ordered speeds. The addition or deduction will have been assessed by reference to the relevant cumulative average. The resulting monetary figure can readily be applied to adverse weather periods on equivalent passages (i.e. passages performed in the same condition and under the same orders as to speed), employing the formula in lines 337 to 341 of the clause. Where, however, a sea passage is performed wholly in adverse weather conditions (such as voyage 12B or voyage 13L in the present case) it is not possible to calculate any average good weather performance for the particular voyage in question. What average is to be used to arrive at the appropriate addition or deduction for the purposes of the “grossing-up” formula? Mr Millard’s evidence was that, in the circumstances I have described, he is accustomed to using an average taken from the combined averages of the vessel’s performance on the immediately preceding and immediately subsequent good weather passages. But those could be very short passages, and therefore less representative of the vessel’s performance as a whole. Alternatively one could be a short passage and one could be a long one. This would require an average to be calculated which took account of the total miles steamed and the total time taken on the two passages. The calculation of a single average from the statistics of more than one voyage is a concession, albeit a very limited one, to the cumulative average methodology. I should make clear that the problem just mentioned is not unique to the individual voyage average method. It could cause difficulty even where the modified cumulative average method was adopted, if the passage or passages performed wholly in adverse weather conditions were the only passages performed under that particular order as to speed. But those circumstances would be likely to be rare under a time charterparty of more than minimal duration.

38. In my judgment the wording of clause 24 is not of great assistance in resolving which method of calculating the averages is the correct approach. The clause refers, in line 321, to speeds and bunker consumptions in the plural. That might be thought to point towards the individual voyage average method. But it is equally consistent with the cumulative average method because, even if the vessel was only ever ordered to perform at the service speed, there would be one set of averages for the laden condition, and one for the ballast condition. The phraseology of clause 24(b)(i) is capable of supporting either method, provided that it is recognised that if a range of performance is guaranteed there will have to be averages calculated for each ordered speed. However clause 24(b)(ii), may be thought to incline more towards the cumulative average method where it refers to “the total bunkers consumed”, In the end the choice comes down to which method most fairly gives effect to the scheme of compensation or reward contained in the clause in all of the eventualities likely to arise. For the reasons I have given, I think that the modified cumulative average method espoused by the owners most nearly performs that function.

39. Issue 8 is this:

“8. Whether clause 44 provides a guarantee of consumption outside the range of speeds indicated and, if so, how it is to be calculated.”

40. The dispute is as to whether the charterparty provides a guarantee of consumption below 12.5 knots. On the conclusions I have already reached, this question only arises in relation to voyage 15B. The answer to the question must be in the negative, for two reasons. The first, is that the charterers were not entitled to order the vessel to proceed at less than 12.5 knots. They could only request that she do so. The charterers were limited to ordering the vessel to operate within the range of speeds in clause 44. Only speeds and corresponding rates of consumption within that range were the subject of any guarantee. Second, even if the charterers were entitled to order the vessel to proceed at speeds below 12.5 knots (but not less than the minimum speed) (a) no rates of consumption were given in the charterparty for any speed below 12.5 knots, other than for the minimum speeds, because the pro-rating provision in clause 24 (lines 308 to 309) applied only to “the foregoing bunker consumptions” which included those in clause 44 but not those for the minimum speeds, and (b) so far as the minimum speeds are concerned, the only warranty was that contained in clause 82 and that clause did not include a warranty as to rates of consumption. It would not have been difficult for the parties to have stipulated in the charterparty for guaranteed speeds and rates of consumption between the minimum speeds and 12.5 knots. But they did not do so.

41. Issue 9 raises a discreet question in relation to voyage 21L. The experts agreed that the Log books showed that during this voyage between loading ports inerting of the cargo tanks was carried out and that 23 tons of intermediate fuel oil (IFO) were used for that purpose. The question is:

“9. Whether fuel oil consumed during cargo tank inerting is to be treated as consumption for the purpose of calculating the vessel’s performance under clauses 24 and 44 of the charter.”

42. Mr Watson excluded the additional fuel oil consumption from his calculations of the vessel’s performance on the passage in question because he considered it most probable that the tanks had been gas-freed for inspection, gauging or sampling rather than for maintenance purposes. Clause 59 of the charterparty provided: “If Charterers so require, Owners shall arrange for the vessel’s tanks to be de-inerted to facilitate inspection, gauging and sampling. Any time taken in de-inerting, inspection, gauging, sampling and re-inerting thereafter shall count as on-hire”. There was no evidence of any request by the charterers that any of the tanks should be gas-freed, but Mr Watson believed that 23 tons of fuel oil was far more than would be required to re-inert a single cargo tank. By his calculation the Eleuthera would burn around 38 tons of fuel oil to inert all of the cargo tanks. So 23 tons was enough to inert at least 50% of the tanks. In his view it was unlikely that maintenance work would have needed to be done in half or more than half of the cargo tanks at one and the same time. I did not understand Mr Millard seriously to challenge Mr Watson’s calculations. His view that the additional consumption had to be taken into account in assessing the performance claim rested on the fact that the consumption of bunkers for the purpose of inerting tanks was not separately addressed in clause 44. Any such consumption had therefore to be accommodated within the guarantees in clause 44 and if, by adding that consumption to the main engine consumption for steaming or for other purposes, the guaranteed rate was exceeded, the charterers were entitled to claim.

43. I am prepared to infer that the inerting of the tanks was required for charterers purposes rather than for maintenance. The vessel loaded approximately 50% of her total cargo at the first two ports of loading (Jebel Dhanna and Ruwais). The other 50% was loaded at the third port of loading (Mina al Fahal). The inerting operation was carried out in the interval between loading at Ruwais and loading at Mina al Fahal. At that time about 50% of the vessel’s cargo tanks would have been empty. Although there is no evidence that the charterers required these tanks to be gas-free at Mina al Fahal for inspection, gauging or sampling, I think, for the reasons given by Mr Watson, that that is a much more likely scenario than that the owners wanted to carry out maintenance work. Nevertheless, clause 59 of the charterparty said nothing about the charterers having to pay for the fuel required to re-inert the cargo tanks in addition to the fuel required for any other purpose while the vessel was on-hire. Clause 24 expressly provided (in lines 308 to 309) that “The foregoing bunker consumptions [i.e. those in clause 44] are for all purposes except cargo hearing and tank cleaning …”. It follows in my judgment that fuel consumed to inert the cargo tanks is subject to the guaranteed rates of consumption in clause 44 and must be accommodated within those rates or the charterers will have a basis of claim. I therefore give an affirmative answer to question 9.

The argument about interest

44. The rate of daily hire payable in the first year of the charterparty was US$22,500. In the second year it was US$24,000 and in the third year it was US$25,500. The charterers paid the hire due in the first and second years of the charterparty. They did so monthly in advance and without deductions, or at least without deductions which were or are now controversial. The deductions to which the interest claim relates arose out of problems which the vessel experienced on voyage 17. The laycan date given to the Master for presenting at Kharg Island was 6^th September 1998. The charterers complained that the vessel was unnecessarily delayed on her ballast passage from the Far East to Kharg Island. Nevertheless she reached Kharg Island on 6^th September and tendered Notice of Readiness at the pilot station at 2300 hours. According to the charterers the vessel then departed from the pilot station without notifying them. She sailed to a point about 17 miles away, where she drifted in open sea in order to carry out repairs to the main engine. When, early on 7^th September, the terminal wanted to call in the vessel to load, it was reported that she was not at the pilot station. She lost her place in the queue and the voyage charterers said they would cancel the voyage charterparty unless the charterers agreed to accept a reduced rate of freight. The charterers agreed to this demand because they said that they did not know what had happened to the vessel. As a result of the freight reduction, the charterers lost income of US$744,870 on voyage 17.

45. On 16^th September 1998 the charterers sent a long telex to the owners complaining about this loss and about other aspects of the vessel’s performance, including her speed and consumption on recent passages. They followed it up with another message on 24^th September in which they proposed (i) that the loss of US$744,870 should be shared on a 50:50 basis, so that the owners should reimburse to them a sum of about $372,435, and (ii) that the daily rate of hire should be reduced by US$1,500 from US$25,500 to US$24,000 on account of the vessel’s poor performance. The owners rejected both proposals and strongly resisted any adjustment to the rate of hire. The charterers were then late in paying hire for the month of October. When finally they paid, they made two deductions from the amount remitted. By way of shorthand I shall refer to these two deductions as “the unauthorised deductions”. The first was a sum of US$1,500 from the daily rate of hire. The second was a sum of US$37,243.50, being one-tenth of 50% of the loss of freight on voyage 17. The charterers stated that they would deduct a similar sum of $37,243.50 from each of the next 10 monthly hire payments (i.e. until July 1998 which was when the vessel was due to be redelivered).

46. Under protest from the owners the charterers continued to make the unauthorised deductions until the charterparty came to an end. By that time the unauthorised deductions totalled in excess of US$830,000. As recorded at the outset of this judgment, the charterers eventually conceded that these deductions had been made on a wrong basis. They paid a sum of US$427,000 to the owners on 8^th July 1999, but continued to withhold the balance. They justified the retention of the rest of the money by their claim for breach of the performance warranties, which had now been particularised.

47. So much for the background. The argument about interest is a narrow one. The charterers admit that they are liable to pay interest in accordance with clause 9(b) of the charterparty on all amounts wrongly deducted from hire. They accept that this applies to the unauthorised deductions to the extent that they cannot substantiate that they were entitled to make the deductions under clause 9. The parties agree that, for interest purposes, the sum of US$427,000 should be treated as repaying the unauthorised deductions made earliest in point of time. The owners claim interest on the total of the unauthorised deductions down to 8^th July 1999 and on the amount by which the deductions exceeded US$427,000 since that date. I am not concerned with the calculation of the precise figure. I was told that the claim for interest amounted to around US$243,000 at the start of the trial. The charterers’ answer to this claim is to say that they are under no liability or a far smaller liability to pay interest because they can replace the deductions made on an incorrect basis with the performance claim which they have now formulated in respect of years 2 and 3 of the charterparty. The charterers can only succeed in this argument if they can bring their performance claim within the ambit of clause 9(iii) of the charterparty. To do so they have to establish that the amounts which they now claim to have been entitled to deduct from hire were “amounts due or reasonably estimated to become due to Charterers under Clause … 24 hereof” in the relevant months.

48. In my judgment the charterers face an impossible task. Clause 9 required hire to be paid in full, except for any deductions within the categories of clause 9(i), clause 9(ii) and clause 9(iii). Clause 9 required “… any such adjustments to be made at the due date for the next monthly payment after the facts have been ascertained”. For the charterers Mr Davey submitted that this phrase simply meant that sums initially deducted on the basis of an estimation could subsequently be adjusted after the facts were ascertained. This is a misreading of the words. The phrase used is “any such adjustments”. In my opinion that phrase refers back to the permitted deductions, not to a subsequent modification of those deductions. Clause 9 recognises that it may take some time before sufficient facts are known to justify any deduction at all. Timesheets and disbursement accounts have to be received. Log books may need to be inspected. Clause 9 provides simply that the deduction is to be made at the due date for the next monthly hire payment which follows the ascertainment of the facts which support the deduction. By clear implication no deduction can be made unless and until the facts justifying it have been ascertained.

49. Miss Birch pointed out that the charterers made no attempt to substantiate a claim for under-performance until March 1999. In a message to the owners dated 30^th March they referred to the fact that they were employing a worldwide weather monitoring system to check the performance of their fleet, including that of the Eleuthera. They said that according to this system the performance of the Eleuthera had been worse than that of their other vessels. But they did not put forward any figures. On 8^th April the charterers sent the owners a copy of a weather report by Oceanroutes Inc., in order to demonstrate that the Master’s reports of adverse weather conditions were exaggerated. But still they did not put forward any figures. It was not until 1^st June 1999, only some 4 weeks before the redelivery date, that the charterers put forward an explicable figure for the first time. They submitted to the owners a Statement of the performance claim together with supporting calculations and said:

“… Charterers are notifying Owners that the vessel have under-performed over the last 2 years amounted to USD478,998.7 under the clause 24 of the C/P. Accordingly, the under-performance claim amount and off-hire during the vessel’s drydock having been deducted, there is nothing to be paid on the next hire payment due on 1/06/99, but the over-paid hire of USD122,769.80 should be remitted to Charterers’ bank account in due course.”

50. It is therefore plain that the charterers did not ascertain the facts to justify the performance claim which they now advance until sometime shortly before 1^st June 1999. The first point at which they were entitled to adjust the hire to take account of their claim under clause 24 was on the due date for the next hire payment, namely 30^th June 1999. They cannot use the claim put forward on that date to justify arbitrary deductions from hire made at any earlier point in time.

51. There was some argument as to the meaning of the word “due” in clause 9(iii). I think it means “payable”. The sum must be one which can be shown to be payable in accordance with the relevant clause (clause 3(iii) or clause 24). It goes without saying that a sum cannot be shown to be payable unless the facts have been ascertained to justify the claim for payment. Miss Birch referred to the decision of the Court of Appeal in The Nanfri [1978] 2 Lloyd’s Rep. 132 and to the decision of Robert Goff J. in The Kostas Melas [1981] 1 Lloyd’s Rep. 18 for the proposition that a prerequisite for the deduction of any amount from charter hire is that the assessment or estimation of the amount should have been made in good faith and on reasonable grounds. Those were cases concerned with equitable set-off rather than with an express clause permitting deductions to be made. But the qualification applies equally to clause 9 of the Shelltime 4 form. The right to make a deduction under clause 9(iii) does not arise until facts have been ascertained which enable the charterer, acting in good faith, to make a reasonable assessment of an amount which is payable under the relevant clause. Claims under clause 24, cannot be formulated until after the end of the relevant accounting period unless the accounting period is the final year or part year of the charterparty. In that case, clause 24(c) permits the charterer to formulate a claim based on a reasonable estimate made two months before the end of the charter period. This is so as to enable the claim to be deducted from the ultimate, or possibly the penultimate, hire payment. The phrase “or reasonably estimated to become due” in clause 9(iii) caters for this situation.

52. My conclusion is that the unauthorised deductions cannot be justified retrospectively by the performance claim put forward on 1^st June 1999 because until 1^st June 1999 (i) the facts had not been ascertained with sufficient or any precision to substantiate a claim that any sum was due under clause 24 and (ii) no reasonable assessment of the sum due under clause 24 had been made by the charterers.