England and Wales High Court (Commercial Court) Decisions

IN THE HIGH COURT OF JUSTICE
QUEEN'S BENCH DIVISION
COMMERCIAL COURT

		Royal Courts of Justice Strand, London, WC2A 2LL
		16/06/2008

MR JUSTICE CHRISTOPHER CLARKE :

1. I have before me for decision five preliminary issues ordered by Walker, J on 16^th November 2007, in an action in which Mopani Copper Mines PLC ("Mopani") claims against the defendant on behalf of itself and other Lloyd's Underwriters and Company Insurers an indemnity in respect of damage done to an electrostatic precipitator.

The background

2. Mopani is a Zambian company. In 2004 it was making arrangements to construct new copper smelting facilities at its site at Mufulira in Zambia, The project was to comprise a number of freestanding units consisting of: (a) an oxygen plant; (b) an Isasmelt furnace (sometimes referred to as the TSL furnace); (c) an electronic precipitator ("ESP"); (d) a matte settling electric furnace ("MSEF"); and (e) an acid plant. The function of the ESP is to process exhaust gases emitted by the Isasmelt furnace by removing copper dust particles from the gas and allowing cleaned gas to be transferred to the acid plant^[1]. The particles of dust separated out by this process add to the total production of copper.
3. Zambian law required the project to be insured with a Zambian insurer. For that reason the form which the insurance took was that the project was insured with, as it turned out, Zigi Insurance Company Limited, a Zambian insurer, and reinsured in London. In substance the risk was placed in the London market. The terms were negotiated with the reinsurers, the risk being broked to them by Cooper Gay and Company Ltd ("Cooper Gay"). The identity of the insurance company was only determined at the end of the negotiations. Subsequently the arrangements became subject to a Memorandum of Understanding ("the MOU") by which it was agreed, inter alia, that monies due to the insured in respect of claims as agreed by the reinsurers would be paid by the reinsurers to the brokers for onward transmission to Mopani.
4. The reinsurance policy is, so far as presently relevant, contained in a slip, originally scratched by the defendant on 29^th November 2004, being an amended and retyped version of a slip scratched on 12^th November 2004, with further provisions agreed on 18^th January 2005, together with an endorsement ("Endorsement No 1") scratched on 11^th February 2005 which varied the previous agreement.
5. There has been considerable debate before me as to what material is admissible as a guide to construction of the slip, and as to what use that material may be for that purpose. In the version of the slip as scratched on 18^th January 2005 certain words have been deleted. I consider hereafter the extent to which it is possible, or helpful, to look at the words deleted. It is common ground (i) that, in construing the contract, the Court must look at the circumstances surrounding its making in order to see what was the objective that the parties had in view; (ii) that a prior agreement is at least admissible as a guide to construction; and (iii) that the following are inadmissible (a) the content of prior negotiations; (b) communications between the insured and its broker which were not communicated to the reinsurers and (c) the subjective views of the parties as to what they thought they had achieved. The fact that, in the summary of events which follows, I set out both communications that did and those that did not cross the line between Mopani and reinsurers does not mean that I have forgotten or intend to elide the distinction between the two types of communication.

The placement of the risk

6. A preliminary meeting took place between the defendant, Millennium Underwriting Ltd ("Millennium"), and Cooper Gay to discuss the risk as early as 24^th February 2004. But it was not until September 2004 that Millennium was asked to provide, and did provide, indicative terms. In an e-mail of 20^th September 2004 Mr Stephen Coward, the lead underwriter for Millennium, set out terms for EAR insurance i.e. Erection All Risk insurance for a period of 18 months from October 2004, including 60 days T & C (Testing and Commissioning) plus 12 months extended maintenance. This was subject to, inter alia, satisfactory details of physical separation from existing plant.
7. On 24^th September 2004 Mr Coward scratched a slip drawn up by Cooper Gay which described the risk as follows:

"TYPE:	Construction/Erection All Risks including Third Party Liability Reinsurance as original
PERIOD	Whole contract period estimated to be 18 months from 27th September 2004 including 30 days Testing and commissioning plus 12 months extended maintenance period thereafter".

8. Mr Coward scratched the slip for a 25% line with certain amendments and the addition of the words "Subject receipt of o/s information referred to in email 20/9/04". Munich Re UK and Liberty scratched the slip for a 25% line each on the same day, below Mr Coward's scratch.
9. The information needed was provided in a letter of 28^th October from Glencore UK Ltd ("Glencore"), who acted for Mopani, to Cooper Gay. Cooper Gay e-mailed that letter to Mr Coward on 1^st November 2004 and in April 2005 he scratched it as "Contents Noted". In that letter Ms Went of Glencore wrote as follows:

"We want to make sure that we do not, have, or incur at a later date, any gaps in cover and therefore, please would you confirm when each party (sic) of the cover is intended to trigger i.e. When does the construction all risks cover cease and the 30 day testing and commissioning period commence? Also when does the extended maintenance period start and also how does this work in relation to Mopani purchasing all risk of physical damage insurance (incl BI) for the plant as the plant will be up and running".

10. On 12^th November 2004 Cooper Gay presented a slip to Mr Coward which he modified slightly and scratched on behalf of Millennium for a 25% line. By now the Risk Details included:

"TYPE	Construction/Erection All Risks & DSU [2] including Third Party Liability Reinsurance as original. Wording to be agreed MLM.
…	…
REINSURED	As agreed by MLM
PERIOD	Whole contract period estimated to be 18 months from 1st November 2004 including 30 days Testing and commissioning plus 12 months extended maintenance period thereafter.
INTEREST	Section 1 All Risks All Contract Works, whether permanent or temporary, materials incorporated or for incorporation therein, Temporary Buildings and their contents, and all other property or equipment of whatsoever nature (other than Constructional Plant and Equipment) the property of the Original Insured or for which they are responsible, whilst at the contract site(s) or elsewhere in the territorial limits, including whilst in transit other than by sea or air as may be more fully described in the Original Policy. Section 2 – Third Party Liability [Various provisions were then set out] Section 3 – Delay in Start-Up ……."

The individual pieces of plant were set out in the Information sheets with their respective values.

11. The slip scratched by Mr Coward constituted a contractual obligation on behalf of Millennium for a 25% line and was treated as such by the brokers: see their e-mail of 22^nd November 2004. The slip was not, however, taken round the market because in the afternoon of Friday 12^th November 2004 Mr Colin Chappell of Glencore telephoned Cooper Gay to confirm the order, but then called back to say that Third Party Liability cover was not required, and, later still, to say that the price needed to come down to $ 714,500, a further reduction of about 6%.
12. The concern expressed by Ms Went in her letter of 28^th October 2004 about gaps in cover was apt. The type of cover then under consideration was Construction/Erection All Risks. The risks covered by such a policy are the risks associated with constructing or erecting one or more buildings or structures. Such risks are to be contrasted with those insured under Property or Operational Insurance, where the risks are the risks associated with the ownership and use of constructed or erected buildings or structures.
13. As is apparent from the description of the perils in the two slips of 12^th November (see above) and 29^th November 2004 (see below) the insurance was to cover 30 days (but no more) of testing and commissioning. Cover for testing and commissioning is a form of operational cover often provided by a CAR/EAR policy. What is covered is the operation of the plant for the purpose of seeing whether it works as it should, an activity which takes place during the contractor's period of responsibility for the plant and before it is handed over to the owner.
14. The slips also provided cover for an extended maintenance period. Maintenance cover is limited and relates, in effect, to manifestations of the construction risk during the maintenance period. During that period the plant will be in operation, and its operational risks will be covered by the operational policy. It is not suggested that Mopani is entitled to claim under that heading.
15. If a project involves the construction of several buildings in sequence the insured or reinsured runs the risk of finding itself without CAR/EAR cover in respect of any building which has been constructed, tested and commissioned and has begun to operate, even though other buildings are still being constructed or commissioned. Even if there is only one building, the insured may find itself without cover if the building has been erected, tested and commissioned, or if the testing and commissioning lasts longer than 30 days (if that is the testing & commissioning period). Further the reference to 30 days testing and commissioning leaves open the question whether that means 30 days testing and commissioning for each item or an aggregate for all of them.
16. On 2^nd November 2004 Mr Coward sent an e-mail to Cooper Gay which included the following:

"The comment about avoiding gaps in cover is fair enough and if MLM is involved all aspects then this is easily achieved. Problems are likely to arise if you "mix and match". It may be worth setting up a meeting with Samantha Went so that I can give a simple guide to construction insurance and the passing from one phase of risk to another e.g. construction – testing – maintenance".

Such a meeting did not take place.

17. On 3^rd November 2004 Ms Went e-mailed Mr Javier Medina of Cooper Gay as follows:

"..As discussed with Andrew, I would like to sit down with Stephen Coward in the near future but in the meantime please would you kindly provide a summary of when each of the covers as per the final point of our letter would be triggered in general. We really do need something from Cooper Gay on this.

Given that we have now provided further information on the separation of the construction site and the existing property, please confirm that the subjectivities have been lifted on the quote".

18. On the same day Mr Cave-Jones of Cooper Gay e-mailed Ms Went as follows:

"Javier is going to discuss the various outstanding issues with Stephen Coward, but in response to your questions regarding gaps in cover we would respond as follows.

Handover of the different stages of a policy usually occur (sic) when the principal signs off and accepts the previous stage having been completed, this information then being passed on to insurers. It is usual practice to insure all the stages with one insurer, partly because to place them separately may not always be possible but also to allow continuity of cover. Standard operational cover should be organised to coincide with handover of the completed testing stage. Maintenance period then running concurrently with the Operational All Risks policy. Insurers normally cover for the whole contract period and we do not normally have to declare each new stage for this stage to be in effect, i.e. moving from construction/erection to testing.

The contract period would normally be fully set out from the start of the project, which would include construction phase, testing phase and the subsequent maintenance period. What you might need to do is extend contract periods when approaching testing as it is quite usual for projects to over-run, thereby allowing yourself cover for the full (over-run) construction phase followed by "x" weeks of testing plus maintenance thereafter.

Hope this explains what you need …"

19. The explanation given in that e-mail left open whether the operational cover that should be arranged should be in respect of each building after the completion of the testing of that building, or in respect of all buildings after the completion of testing of all of them.
20. On 22^nd November 2004 Mr Andrew Stammers of Cooper Gay e-mailed to Ms Went as follows:

"Tried to see Stephen Coward but he is away all week on hols – bad timing.

However as his line is effectively bound anyway we have cover in force and will need to sort out with him next week as I prefer to do this rather than fuss around with his deputy.

However have had a long chat with Mike Robertson at Liberty, who also write the contract. There is indeed a grey area over the gradual handover of the plant and this only really applies as far as the ALOP is concerned. If there is no ALOP, each sector of the plant – acid plant, oxygen plant etc can be handed over to the operational policy post testing. However is (sic) a loss occurred at the plants handed over early which caused an ALOP loss at the main item being the smelter, there would be no cover if this had already been handed over. The solution seems to be to review the bar chart and try to obtain the market agreement to allow a certain amount of operational cover post T & C for individual items. Mike has offered to review this and give us his assessment. Will bind up the remaining lines tomorrow."

21. In this e-mail Cooper Gay recognised either the need or desirability for each part of the plant to be handed over to the operational policy post testing in the case of a policy not covering Advanced Loss of Profit^[3] (ALOP); together with a problem in this respect if there was ALOP cover. The problem identified, albeit somewhat opaquely, was that ALOP cover was dependent on there being insured damage to a plant covered under the policy. If one part of the project was handed over and a loss occurred at that plant causing ALOP there would be no cover.
22. On 23^rd November Ms Went replied:

"I think this needs to be discussed further as don't like grey areas. Also assume inception date was y'day".

23. On 24^th November she suggested a form of wording to be agreed by underwriters viz:

"It is hereby agreed that the final handover of the last piece of equipment will be deemed to be the handover of the project notwithstanding that parts of the plants/construction project may be handed over prior to the final handover. As such, all equipment/plant/machinery is deemed to be under construction until the final over (sic) hand over certificate is issued".

The brokers' attempt to deal with the grey area

24. On 29^th November 20004 Cooper Gay e-mailed to Glencore as follows:

"Stammers and I have discussed the best way to present the preferred coverage for yourselves in respect of the above. We think that the attached slip covers the angles to make sure there is continuity of cover between construction and operational phases of the project – i.e. we're making it clear that this is a Constructional and Operational insurance as well as reference under conditions of our slip."

25. Mr Chapple of Glencore e-mailed back on the same day :

"We're happy with your suggested changes to the slip, which should now provide cover for those items already built (i.e. in an operational stage) even though the construction project has not concluded".

26. The new version of the slip ("the 29^th November slip") contained the following material terms:

"TYPE:	Construction/Erection and Operational All Risks including Advanced Loss of Profits Reinsurance as original."
PERIOD	Whole contract period estimated to be 18 months from 22nd November 2004 including 30 days Testing and Commissioning plus 12 months extended maintenance period thereafter.
INTEREST	As before"

27. Under the heading "Conditions," the following had been inserted by Cooper Gay:

"The whole contract period is deemed to include all projects incorporated into the Mopani Smelter Upgrade project and cover extends to include Operational All Risks for completed phases of the project up until the final handover certificate is issued."

28. Mr Coward scratched this slip for a 25% line on 29^th November 2004. But when he did so he put a line in pencil through the condition set out in the previous paragraph ("the condition") and added the acronyms "TBA MLM" indicating thereby that this condition was yet to be agreed by Millennium. In December the 29^th November slip was subscribed (in fact oversubscribed) by the following market.

Quotes for operational cover

29. On 1^st December 2004 Mr Coward e-mailed to Cooper Gay the following quote from reinsurers:

"AP	to provide all risks cover on items brought into commercial use in advance of overall handover/startup will be calculated at rate 0.105% on the replacement values involved, provided period for any one item does not exceed 3 months. Deductible = USD 250,000…"

30. Under the slip the deductibles for Section 1 All Risks were as follows:

"USD	500,000 each and every loss in respect of Testing and Commissioning and maintenance in respect of furnace.
USD	250,000 each and every loss in respect of Testing and Commissioning and maintenance in respect of all other equipment and maintenance and losses in respect of wind, storm, flood, water damage and hail.
USD	100,000 each and every loss for all other perils."

The deductible of $ 250,000 in the quote was, thus, the same as for testing and commissioning of all equipment save the furnaces.

31. Ms Went replied to the brokers, who sent her the quote, on the same day, in an e-mail which was not forwarded to reinsurers:

"Andrew – this is typical but unfortunately also unacceptable. All the information we disclosed and insurers were aware that parts of the project would be completed before others and as such, we understood the cover to be there. It was only due to Glencore/Mopani's request for confirmation of coverage that has raised this issue and should not be allowed to be a way in which insurers can try and get more premium.

Suggest you rediscuss before I am back on Friday as I can only imagine the response from Mopani particularly given you told them this was covered!"

32. Mr Stammers replied:

"I accept your point about the response from lead.

I think that there was a greater degree of understanding of the contract obtained at the meetings in Mopani. We should discuss this further on Friday and will review with Coward in the meantime. Either way the assets need to be insured at all stages in the project and we will have to find a solution that is acceptable all round

All of this only becomes an issue if the ALOP is purchased"^[4]

33. On 7^th December 2004 Mr Coward gave a further quote as follows:

"Agree to provide up to 3 months operational "all risks " cover i.r.o Oxygen plant and/or Smelting Furnace and/or Acid Plant for up to 3 months until overall handover …AP USD 50.000, adjustable at expiry on pro rata basis (period/values) subject to minimum retained premium for this element of cover of USD 20,000 (100%)".

34. On 14^th December 2004 Cooper Gay e-mailed to Glencore stating that the market had now agreed to the revised slip. What in fact had happened was that the 29^th November slip, including the pencilled out condition and the annotation "TBA MLM" had been taken round the market and subscribed, as to 120%. But the subjectivity inherent in the annotation had not been lifted. The e-mail continued:

"Understanding that there is still a slight issue over the premium charge proposed by Millennium which is our situation to handle, the market have agreed to cover the operational aspect of the contract from the point of individual handover until the final handover. This as we have discussed does not become an issue if Mopani opt not to buy any Advance Loss of profits cover as the individual plant units can be handed over to the operational policy as they are completed".

35. The e-mail then went on to ask for more details as to how the plant would be operated post completion of the individual units.
36. On 15^th December 2004 Ms Went responded:

"I am afraid I do not agree with your opening paragraph regarding the additional cost only being an issue in relation to ALOP. It has never been considered or agreed by Glencore/Mopani that the all risk insurance of the individual units will transfer to Mopani's property policy once the construction of same is complete. In fact it has always been intended that the entire project will fall under the CAR policy until the final handover certificate is issued.

However, notwithstanding this we will ask Mopani for further information and will revert as soon as possible".

37. On 17 December 2004, Ms Went provided information by e-mail as to the use of the plant following completion. The e-mail enclosed a project schedule, with accompanying bar chart, showing completion dates for each section of the plant. The details in the schedule included the following:

ITEM	START	FINISH
PLANT STARTUP AND COMMISSIONING	24 Oct 05	22 Mar 06
OXYGEN PLANT	02 Nov 05	20 Dec 05
Start-up and Commissioning	02 Nov 05	06 Dec 05
Hot Commissioning	07 Dec 05	20 Dec 05
ISASMELT FURNACE	14 Dec 05	15 Mar 06
Start-up and Cold Commissioning	14 Dec 05	25 Jan 06
Heat-up Furnace	26 Jan 06	01 Feb 06
Hot Commissioning	03 Feb 06	15 Mar 06
MATTE SETTLING FURNACE	23 Nov 05	13 Jan 06
Start-up and Commissioning	23 Nov 05	22 Dec 05
Hot Commissioning	23 Dec 05	13 Jan 06
SULPHURIC ACID PLANT	01 Dec 05	22 Mar 06
Start-up and Cold Commissioning	01 Dec 05	09 Jan 06
Hot Commissioning	23 Feb 06	22 Mar 06

38. The e-mail, which was not shown to Millennium at this stage but was seen by it subsequently, stated:

"After further discussions with Mopani, we can provide the following information:

……

You will note that the oxygen plant will be started up first and operated for testing and operator traing [sic] etc. The oxygen is only required for the Isasmelt Furnace and all oxygen produced prior to 26 Jan 2006 will be vented to atmosphere and a small amount will be stored in the buffer storage tank. Thereafter, the Matte Settling Furnace will be commissioned. Both the Oxygen Plant and the Matte Settling Furnace need to be operational to allow commissioning of the Isasmelt Furnace to proceed. Once the Isasmelt furnace hot commissioning has commenced then sufficient off-gas (furnace exhaust gas) will be produced to allow commissioning of the acid plant and acid production to commence.

I hope the above clarifies and that we will not be charged additional premium for cover of the individual units, notwithstanding that we may have hand over certificates for same, under the CAR policy until the final project hand over certificate is issued."

39. On 12^th January Cooper Gay sent Mr Coward the revised project plan and bar chart, including the details set out in paragraph 37 above, but not the accompanying e-mail of 17^th December. It was apparent from this plan, and was apparent to Mr Coward, that the commissioning of the Oxygen Plant was to be completed on 20^th December 2005, well before the completion of the commissioning of the plant as a whole. Cooper Gay's e-mail said that they would appreciate Mr Coward's "prompt response regarding the possible operation extension as per the bar chart attached".

17^th January 2005

40. Mr Coward e-mailed Cooper Gay on 17^th January:

"In order to conclude my thoughts on the revised programme, please clarify the operating status of the Oxygen Plant between 20/12/05 and 22/3/06. Its T & C will have been completed but will it stand idle for 3 months or will it be brought into operation during this period?"

41. Cooper Gay replied on the same day:

"The position as it stands now is that the Oxygen and Acid plants will be tested and commissioned and then shut down until the final handover of the entire project. Initially they had discussed operating these plants prior to the overall handover although this has now been changed. Therefore would hope that you can leave these items under the EAR policy without charging any additional premium."

42. There is nothing on Cooper Gay's files which explains where they got this information from. It is inconsistent with the information contained in the e-mail of 17^th December 2004; and, in relation to the Acid Plant, not easy to follow. Under the plan the commissioning of the whole plant was to be completed on 22nd March 2006 which is the date for the completion of commissioning of the Acid Plant.
43. Mr Coward replied:

"Noted and confirm all property to remain covered under EAR policy until handover on 22/3/06. No AP."

18^th January 2005

44. On 18^th January 2005 Cooper Gay took the slip, in its 29^th November form, to Mr Coward, together with a sheet of paper which contained Cooper Gay's message of 17^th January and Mr Coward's reply of the same date set out in paragraphs 41 and 43 above ("the e-mails document").
45. Mr Coward made certain amendments to the slip, of which the most important are these. Firstly he inserted Zigi Insurance Ltd as the insurer. Secondly he removed the pencil strike out from some of the condition and struck out the remainder permanently so that it read as follows:

"The whole contract period is deemed to include all projects incorporated into the Mopani Smelter Upgrade project and cover extends to include Operational All Risks for completed phases of the project up until the final handover certificate is issued."

46. Mr Coward also scratched the e-mails document.
47. The information that had been provided to the defendant included the following (i) that oxygen from the Oxygen Plant would be piped to the Isasmelt furnace; and (ii) that the MSEF would be connected to the Isasmelt furnace so as to feed it. In other words, so far as those three items were concerned, there was going to be an interdependent system.
48. That information appears in a document headed "SMELTER UPGRADE PROJECT" provided in September 2004 as follows:

"MATTE SETTLING FURNACE

…The matte settling furnace will separate copper matte from discard slag for the treatment of 850,000 tpa concentrate in the TSL furnace …"

OXYGEN PLANT

….The TSL furnace will initially treat 650,000 tpa of dry copper concentrate and for this duration, the furnace will require a tonnage oxygen supply at a daily rate of circa 500 tonnes. The planned increase in concentrate treatment rate to 850,000 tpa will be achieved without any changes to the TSL furnace by increasing the degree of oxygen enrichment, which makes a modular expansion of the oxygen plant imperative"

49. It also appears in the Information attached to the 29^th November 2004 slip:

"The Oxygen Plant is stand alone ….Oxygen will be piped to the new Isasmelt furnace.

…

The two new furnaces, the Isasmelt and the matte Settling Electric Furnace (MSEF) will be connected as the molten copper matte will be transferred from one to the other via a launder…."

50. The same message appears in the e-mail of 17^th December 2004 (which reinsurers were not to see until later), which refers to (i) the oxygen being "only required for the Isasmelt Furnace " and being vented to the atmosphere prior to 26^th January 2006 (the then date for the heat up of the Isasmelt Furnace) and (ii) the fact that the MSEF will thereafter be commissioned and will (like the Oxygen Plant) need to be operational to allow commissioning of the Isasmelt Furnace to proceed.

Events after 18^th January 2005

51. On 9^th February 2005 the parties agreed Endorsement No 1 to the contract, which I shall consider in more detail below.
52. Thereafter there was a series of endorsements to the policy which extended cover until 14^th March 2007. In respect of all those extensions save the last one an additional premium was charged based on the whole contract value rather than the value of the uncompleted Acid Plant. The additional premium under the last endorsement (of 14^th February 2007 i.e. after the claim had arisen) was calculated based on a 28 day extension for the Acid Plant only.
53. By September 2006 all the items at the plant had been tested and commissioned save for the Acid Plant, which was not tested and commissioned until 14^th February 2007. These items became operational, albeit at less than their full intended capacity. On 25^th September 2006 the ESP was switched on and began to operate in a limited way. Because the Acid Plant had not been commissioned the cleaned gas had to vent to the atmosphere rather than pass to the acid plant. But, as things turned out, the plants other than the Acid Plant were not shut down until the final handover of the entire project (as indicated in Cooper Gay's e-mail of 17^th January 2005), nor were the Oxygen Plant and the MSEF used only for commissioning of the Isasmelt Furnace. Instead those three items were taken into general use.
54. The ESP did not work correctly. There was a temporary resolution of the problem in November but in December all three fields failed and the ESP was operated as a drop down unit to remove as much material in the gas as possible.
55. On 6^th February 2007 reinsurers were notified of a claim. On that day the ESP was damaged, allegedly as a result of the generation of excessive heat caused by the presence of combustible material in the feed from the Isasmelt furnace which accumulated in the ESP, ignited, and sintered, causing damage to the whole of the ESP's internal structure. The claim is not based on any defect in construction or any constructor's fault.
56. It is apparent from the narrative that I have described that at the time of the events giving rise to the claim (a) testing and commissioning of the ESP plant had long since completed; and (b) the ESP plant was in operation, albeit not fully. The reinsurers contend that the CAR/EAR reinsurance and insurance do not (save to an extent not presently material) provide cover to Mopani for loss occurring after testing and commissioning and whilst the plant is in operational use.
57. As it happens, in February 2007 there was a retrospective post loss declaration of the completed items, including the ESP, to Mopani's operational reinsurance policy, which was for 12 months from September 2006. Those reinsurers accepted the declaration and have paid, or are in the process of paying, their share of the claim. But the deductible on the operational policy of $ 1,000,000 is far larger than the deductible on the EAR policy. So what is at issue, commercially, in the present case is an amount equivalent to the $ 1,000,000 deductible under the operational policy (less the deductible under the reinsurance in suit). The loss is something of the order of $ 1,200,000 and Mopani has or will have recovered $ 200,000 from the property reinsurance.

The factual matrix surrounding the 18^th January slip

58. Mr Colin Edelman, QC, for Millennium submits that an important part of the factual matrix lies in the fact that there is a clear distinction between a CAR/ EAR policy and Property or Operational cover. They constitute different risks, and attract different rates of premium, the risk under the latter being usually regarded as greater than under the former, and thus attracting a higher rate of premium and a higher deductible. The inclusion of cover for testing and commissioning means that a very limited degree of cover for operational risks is provided (with a higher deductible) but that is the limit of the operational cover. The insured and its brokers recognised or came to recognise this distinction, as is apparent from the decision to seek a change from "Construction/Erection All Risks" insurance to "Construction/Erection and Operational All Risks" insurance; and the way in which the brokers treated with the underwriter for a further amount of operational cover.
59. Mr Alastair Schaff, QC, for the Insured submits that the distinction between CAR/EAR insurance and operational or property insurance is not as clear cut as the reinsurers assert it to be, and that I should not, in the absence of market evidence, for which no permission has been given, start with some a priori assumption as to what does and does not come within the confines of CAR/EAR insurance. He accepts, however, that the wording of the 12^th November 2004 slip with its reference to Construction/Erection All Risks insurance did not provide for operational cover beyond that inherent in the 30 days T & C - not because the wording was defective for any such purpose but because the parties thereafter set out to negotiate for additional operational cover.
60. In view of the latter concession it may not matter greatly whether, even without it, a distinction is to be drawn between Construction/Erection All Risks insurance and Operational/Property All risks insurance. But there is, as it seems to me, a clear distinction between the two types of risk, even though the terms of particular policies may blur that distinction; and that I do not need expert market evidence to hold that that is so.
61. I find support for that conclusion in the judgment of Morison J in Swiss Reinsurance Co & Ors v United India Insurance Co Ltd [2005] EWHC 237 (Comm). In that case there was a reinsurance policy covering a large power project on which on 18^th June 2001 work had been suspended. One of the questions was whether the cedant was entitled to a return of premium. The title of the policy was "Project Insurance, Construction All Risks and Delay in start up etc". One element of the policy covered construction and erection occurring at the Contract site until the issue of a Taking Over Certificate. Condition 12 provided:

"12. Should the Work insured or any part thereof be entirely stopped by any cause whatsoever and the Insured give notice thereof, the cover under this Policy shall continue without interruption up to a maximum period of six months without additional premium with any further extension of this period to be agreed by the Insurer…

62. Morison J said this:

"32 On the basis of the words used, I take the view that Condition 12 is there to avoid the risk that on a cessation of work the Insurance would lapse due to a material alteration in the risk. The insurance was directed to cover a substantial building contract with its attendant risks. When a project is abandoned by the contractors, then the nature of the risk alters. The site is no longer an active building site; rather it becomes a warehouse or repository for the various equipment and installations which were more or less completed at the time of abandonment. The nature of the hazards has changed. Instead of there being risks associated with a building project with a contractors' all risks cover, the emphasis is now on the dangers of fire, deterioration owing to climate and damage due to inclement weather. The site is passive and not active. What was insurance in relation to a substantial building project has become insurance resembling property insurance. CAR policies are essentially different from property insurance, although some of the risks may be the same in each. But for condition 12.1 I take the view that as at June 18 2001 Swiss Re would have been entitled to say that the Policy was at an end as the underlying risk had altered. If work resumed during the six-month period then the Cover would remain in place although the assureds would no doubt have to consider asking for extensions to the completion dates and would have to pay additional premium accordingly. But if the works did not recommence within the six-month period then in the absence of any agreement as to a further period, the contract would come to an end. As it seems to me, that is how the condition is worded. It would make no commercial sense to argue as do UII that after the six months the policy continues in force for the balance of the 44 months, whether extended by six months or not, even though no further extension was agreed under condition 12. I ask why should a contractor's all risk policy continue to bind the parties when there were no contractors and no work being done.

……

42 As I have said, Swiss Re were not offering a property insurance policy which is a different contract from that originally entered into. On a proper construction of the policy, the new situation which obtained after the work ceased was not one which the reinsurers had agreed to cover. Condition 12 describes the only circumstances in which the cover continued when no work was going on and that lasted for a period of six months unless the parties to the policy agreed a further extension. Therefore on the basis of the wording of the policy and by virtue of the principle enunciated by Lord Justice Saville, Swiss Re were entitled to treat the policy as defunct as from December 2001."

63. The principle to which Morison, J, referred was that "there would be no cover where the circumstances had so changed that it could properly be said by the insurers that the new situation was something which, on the true construction of the policy they had not agreed to cover": Kauser v Eagle Star Insurance Co Ltd [2000] Lloyd's Rep IR 154.
64. As Morison J's judgment makes clear, CAR risks are essentially different from property insurance and insureds may find themselves without cover if, in the events which have happened, there is no longer any construction going on, even though the period of the policy has not expired. That does not means that a CAR/EAR insurance could not cover operational risks, beyond testing and commissioning, if that was what was agreed. But, given that there is, and was appreciated to be, a material distinction between CAR/EAR and operational/property insurance one would expect to find that any application for operational insurance specified the items for which such cover was sought, and for how long, and envisaged the payment of some premium for that cover.
65. The second matter to take into account is the agreements made between the parties prior to the 18^th January 2005 slip. Negotiations which have not matured into an agreement are inadmissible since they are nothing more than negotiations. But prior agreements are admissible: HIH Casualty and General Insurance Ltd v New Hampshire Insurance Co [2001] 2 Lloyd's Rep 161; KPMG LLP v Network Rail Infrastructure Ltd [2007] EWCA Civ 363. In the former case, which concerned a slip followed by a policy, Rix, LJ expressed the view (at paragraph 83) that, where a later contract was intended to supersede an earlier one, the exercise of looking at the earlier agreement might cast no light on the meaning of the later one, since, if the two said the same, the later agreement added nothing, and, if they differed, the later one was prima facie to be regarded as a deliberate departure from its predecessor. In KPMG Blackburn, J, had applied that approach, considering that an earlier agreement for a lease was of little, if any, assistance in construing the lease in its final form. The Court of Appeal regarded this as too narrow a view and treated an agreement for a lease as an important part of the background and a permissible aid to the construction of the lease in its final form.
66. In the present case it seems to me that the sequence of agreements does cast some light on the meaning of the final one. The parties started with the slip of 12^th November 2004 which was limited to Construction/Erection All Risks. They then moved to discussing, but not agreeing, the inclusion of operational cover beyond the limited operational cover provided by the T & C provision. I call this "additional operational cover". The effect of Mr Cowards' pencilling out of the condition and adding the acronyms "TBA MLM" on 29^th November signified the parties' acceptance that any additional operational cover had yet to be agreed. Whatever may be the general position as to the admissibility of deleted words as an aid to construction, I regard it as legitimate, in a case such as this, to read the added acronyms and the pencilled over words to establish what the parties have, in and by their contract, expressly agreed is yet to be agreed.
67. The third relevant matter is the subject matter of the negotiations between the parties which led up to the scratching of the slip on 18^th January 2005. The content of those negotiations is not admissible; but what they were about is. The subject matter of those negotiations was whether there should be, because there needed to be, additional operational cover and, if so, for how long and at what price. It is also relevant that those negotiations took place in the context of a concern, communicated to the reinsurers, as to whether cover would extend until final handover, or whether there might be a gap in the cover in respect of individual units that had been tested and commissioned prior to final handover: see Cooper Gay's e-mail of 22^nd November 2004.

The 18^th January contract.

68. The documents having contractual force are the two documents scratched on 18^th January, namely the amended slip and the e-mails document.

The Insured's submissions

69. Mr Schaff submits that these two documents taken together provide ample basis for concluding that Mopani is entitled to additional operational cover for all the items of Equipment down to 22^nd March 2006 (subsequently extended to 14^th March 2007). The 17^th January 2005 e-mail confirms that "all property" is to remain covered under the policy until final handover on 22^nd March 2006. The slip policy is in terms described as covering "Operational All Risks" and the condition in its final form reads:

"The whole contract period is deemed to include all projects incorporated into the Mopani Smelter Upgrade project until the final handover certificate is issued"

This condition was included in circumstances where all concerned knew that some "projects" would be completed well before others. There is no basis for any implied exclusion of operational cover (cp the different policy exhibited to Mr Coward's second statement which does have such an exclusion). The effect of any such implication would be that "Operational" did not mean what it says and "all" did not mean "all". If no reference is made to the deleted words the answer is clear.

70. Mr Schaff further submits that authority binding on me precludes reference to the deleted words for the purpose of construing the words that remain. It is necessary to construe the contract of 18^th January 2005 as if the document had never contained the deleted words. Even if the authorities that deal with deletions do not compel the court to that course, examination of the deleted words shows that they are simply part of negotiations towards the final agreement and are inadmissible on that ground alone.
71. Further, even if it is legitimate to look at the deleted words, the exercise would produce no different result. It would still be necessary to construe the contract. The court could not say that by deleting the words the parties must have intended to agree the converse (i.e. that cover should not extend to include Operational All Risks for completed phases of the project up to final completion). The result of looking at the deleted words would be to discover that the parties had reached a compromise whereby they kept in part of the condition and excised the rest. Why they did so cannot be known. It may be that the words excised were thought superfluous, because the policy was described as covering operational risks; or that the parties had different views about the effect of the deletion and were prepared to settle for a half way house. There may be other explanations. Mr Coward's evidence that his failure to cross out "and Operational" was a mere oversight is inadmissible (as indeed it is). The e-mails document does not detract from the provisions of the slip, which is the primary document. There is no basis for assuming that by 18^th January 2005 the negotiations had changed from negotiations about additional operational cover into negotiations about some other form of cover such as shutdown cover.
72. Powerfully though those submissions were developed, I do not accept them. The parties were considering whether there needed to be additional operational cover in order to ensure that there was no gap. The upshot of those negotiations, which the parties chose to have recorded as part of their contractual documentation, was, as it seems to me, that such cover was not needed because, as matters stood, there would be no operations after commissioning before final handover. The message of Cooper Gay's e-mail of 17^th January was that after commissioning the Oxygen and Acid Plants would be shut down. The e-mail contained no reference to the Isasmelt furnace and the MSEF. But, in the light of the information that had been provided as part of the broke, that omission did not, in my judgment, signify that additional operational cover was still being sought for two items about which Mr Coward had not chosen to inquire; but rather that such cover was not needed. The question is not whether there was an implied exclusion of operational cover; but what was the cover upon which the parties agreed.
73. The information conveyed had been that the Isasmelt furnace needed the Oxygen plant in order to start up. If the Oxygen plant was to be shut down until final handover, the Isasmelt furnace could not start up until then. The MSEF was to supply the Isasmelt furnace. So the contemplated sequence was (i) testing and commissioning of Oxygen and Acid Plants followed by their being shut down; (ii) final handover of the entire project; (iii) Oxygen Plant starts up; (iii) Isasmelt furnace and MSEF furnace begin operations, the Isasmelt furnace needing the Oxygen Plant, and the function of the MSEF being to supply the Isasmelt furnace.
74. Mr Edelman submitted that, if I was in any doubt about the significance of the e-mails document and the slip, I should construe the documents contra proferentem, the reinsured being the proferens for this purpose both because the slip was produced by its broker and because it was it that was seeking to take the benefit of the wording. Mr Schaff submitted that the contra proferentem doctrine applied to contracts and not to factual matrices and that there was no general principle that, in the event of ambiguity, the insured always lost if it, or its broker, was the party seeking to rely upon an unclear clause.
75. The e-mails document is not simply part of the factual matrix. It is a contractual document as is the slip. I do not, however, propose to base myself on the contra proferentem rule, not least because the wording both of the slip and in the e-mails document is the product of a dialogue and subsequent negotiation. Rather it seems to me that, having regard to the subject matter of the negotiations, and the distinction, perceived by all concerned, between additional operational risks and CAR/EAR insurance, it would be a misconstruction of the 17^th January 2005 e-mails to regard them as signifying that the brokers were still seeking additional operational cover either in respect of the Oxygen and Acid plants or of the two furnaces and the associated ESP.
76. If what the brokers were seeking was additional operational cover for the entire project (or, at least, all items other than the Acid Plant) it seems to me that the reinsurers could legitimately expect the brokers' e-mail to make some mention of this, and of the size of premium to be charged for such cover, particularly having regard to Millennium's earlier quotes. Instead the e-mail invites reinsurers to leave the Oxygen and Acid plants under the EAR policy without charging any additional premium because, as it explains, after those items have been commissioned, they are going to be shut down, i.e. not operated, until final handover. Such an e-mail is disingenuous if, what reinsurers are in fact being invited to do is to provide additional operational cover for the Oxygen Plant from commissioning until final handover of the last item of plant and also for two furnaces, and associated equipment, of which the broker makes no mention, and which, on the information previously provided, were to operate only when the Oxygen Plant had started up (which as the e-mail explains is to be after final handover). If additional operational cover was being provided, the reinsured would also be expected to pay an additional premium for it.
77. That consideration encourages me in the construction that I propose to adopt. I decline to interpret Cooper Gay's e-mail as inviting the reinsurers to grant additional operational cover for all items, but with no additional premium on the basis that there was no risk for two items and unspecified risk for the other; or to regard reinsurers' e-mail response as proceeding on that or any similar basis. I note also that both Cooper Gay and Mr Coward referred to the EAR policy without reference to "Operational All Risks" and that Mr Coward confirmed that all property was covered until handover on 22^nd March 2006 i.e. the date after which the Acid Plant would become operational, as opposed to 22^nd April 2006, the end date of the estimated "whole contract period" of 18 months from 22^nd November 2004. If the reinsurance policy provided additional operational cover it is difficult to see why Mr Coward's confirmation should stop in March.
78. Mr Schaff places considerable reliance on the confirmation by Mr Coward gave that all the projects were to remain covered until the final handover certificate was issued. It does not seem to me, however that that advances the position. The fact that all the projects were covered until then begs the question as to what they were all covered for. Further, one of the known concerns was whether there would be a gap in the cover such that before final handover one piece of equipment might go off cover because it had not been declared to the property policy (see the 22^nd November 2004 e-mail). There is room for doubt in the case of a CAR/EAR policy with a T & C period whether cover extends to an item of plant which had been constructed, commissioned and handed over, but which is then shut down until final handover of the whole project several months later. The effect of Mr Coward's confirmation was that the insured could be assured that such property would be covered. It also removed any problem that might arise because the testing and commissioning, however calculated, lasted for more than 30 days.
79. I turn then to consider the provisions of the slip. The strongest point in Mopani's favour is that the reinsurance is still described as Construction/Erection and Operational All Risks. Mr Edelman submits that that is, even without reference to Mr Coward's inadmissible evidence or the background, an obvious mistake. It should have been crossed out; but by a slip was not. He points out that when the words "and Operational" appeared in the 29^th November 2004 slip they did not signify that operational cover was being afforded because the TBA notation against the condition indicated that such cover was to be agreed. There can, therefore, be no assumption that the words had acquired some greater significance by 18^th January 2005. The words were put there for the purposes of a possible extension of cover. In effect the words are only part of a heading. If the heading of a clause is left in but the operative provisions are removed, as where a draft arbitration clause is crossed out but the heading "Arbitration" remains, the heading has no contractual force.
80. Mr Schaff submits that the analogy is inapposite. There has not been a deletion of a whole clause with only a heading remaining. The type of insurance has remained unaltered; but there has been a compromise as to what the condition should say. There is, in those circumstances, no justification for ignoring the words "and Operational" in the "Type" clause. Those words were to be part of the means whereby additional operational cover was to be provided. The fact that such cover was not provided by the 29^th November 2004 slip is irrelevant once the TBA subjectivity is removed. The words "and Operational" then fulfil their intended function of providing operational cover. It is, moreover, illicit to construe those words, in the context of the 18^th January slip, by reference to what they may or may not have signified at an earlier stage of the negotiations for additional operational cover.
81. Mr Schaff further submits that, unless you adopt the illegitimate approach of looking at the deleted words for the purpose of interpreting the meaning of the words that remain, it does not begin to be obvious that there was any error in leaving "and Operational" in the "TYPE" clause, much less that the error was mutual. Nor, if there was a mistake, is it clear what the mistake was.
82. Mr Edelman referred me to authorities that hold, or appear to hold, that, at any rate in some circumstances, it is legitimate to look at the deleted words in order to understand what the remaining words meant. He submits, however, that, even if it is not legitimate to look at the deleted words for that purpose, it is legitimate to look at the deleted words in order to discover whether or not there has been a clerical error in not crossing out "and Operational". If you are concerned to discover whether there is an error on the face of the document it is necessary to look at what on its face the document says. He relies on the principle, vouched by the cases summarised in Lewison on The Interpretation of Contracts, 4^th edition (2007) at pages 345 and 349, that:

"As part of the process of construction, the court has power to correct obvious mistakes in the written expression of the intention of the parties. Once corrected, the contract is interpreted in its corrected form …

This principle enables the court to take background facts into account in deciding whether a mistake has been made and, if so, what it is".

83. In my judgment it is necessary to take the e-mails document and the slip together. The e-mails document shows not only the stage which the agreement had reached by 17^th January but the basis upon which the parties had agreed the reinsurance. Otherwise there was no point in having it scratched. Taking into account the admissible background, it seems to me that on 17^th/18^th January 2005 the parties were proceeding on the footing that the potential problem in relation to a gap in the cover, and the potential need for additional operational cover with an additional premium, was solved because there was not going to be a period between completion of commissioning of individual units and final handover (when cover would in any event cease) during which additional operational cover would be needed, provided the insurers confirmed, as they did, that cover would remain until final handover even though some of the equipment would have been both constructed and commissioned and would thereafter remain idle. It was the fact that the Oxygen and Acid plants were not going to be operational that was the basis put forward for not having to pay an additional premium for additional operational cover ("Therefore would hope that you can leave these items under the EAR policy without charging any additional premium").
84. The parties recorded the outcome of their negotiations in a contractual document which the underwriter scratched so as to confirm his agreement with it. The words "and Operational" had been introduced as part of a yet-to-be-agreed means of providing additional operational cover. But, as the e-mails document showed, the conclusion was that it was not needed. Since, in a CAR/EAR contract covering Testing and Commissioning, operational activities are covered to a very limited extent, it is not necessary to go to the length of saying that the words "and Operational" must be struck out as an obvious error. But, in those circumstances, the words should not be interpreted as having acquired a status that they did not possess when they appeared in the contract of 29^th November 2004, so as to signify that additional operational cover was in fact to be provided. I accept that the words of the 18^th January 2005 contract must be looked at in their 18^th January 2005 context but that context included not only the 29^th November 2004 slip but the e-mails document, which showed, with some specificity, the basis upon which the parties were proceeding.
85. In reaching this conclusion it is not necessary to use the deleted words for the purpose of interpreting those that remain. My conclusion can be reached by interpreting the words that remain in the light of the admissible factual background, including the subject matter of the negotiations (whether there should be any additional operational cover and for how long), the contract of 29^th November and the contractual e-mails document.
86. Reference to the deleted words does, however, fortify the conclusion that I have reached without them. For the reasons set out below I regard it as open to the court, in this particular case, to look at the deletion which the parties agreed upon as indicating, when taken with the e-mails document, what it is that the parties did not agree and why. In some cases a deletion may be an uncertain guide. But the combination of the e-mails document and the deletion underscores the fact that the parties were not agreeing additional operational cover on all items until final handover, an agreement which would, also, appear to render the provision for cover for testing and commissioning unnecessary.
87. Thus construed the contract makes perfect sense. The condition, in its amended form, ensures that cover is provided, from 18^th November until 18 months later, even if items of the plant have been constructed, tested and commissioned, and then shut down. Operational cover for completed items is not provided because it is not needed.
88. Such a construction also avoids an interpretation which would cut across the rationale for the deductible. The 18^th January 2005 slip provided, as had its predecessors, for a deductible of either $ 500,000 or $ 250,000 in respect of losses in respect of testing and commissioning. The obvious rationale for such a deductible lies in the perceived increase in risk once operations, even if limited to testing and commissioning, start. If, however, Mopani's construction be right there would be cover in respect of losses incurred when the equipment was operating after its commissioning, with only the entry level deductible of $ 100,000.

Endorsement No 1

89. On 20^th January Cooper Gay sent Glencore a copy of the 18^th January slip and the signed MOU and the e-mails document and asked for confirmation that all was in order.
90. On 21^st January 2005 a telephone conversation took place between Mr Chapple of Glencore and Mr Stammers of Cooper Gay. It looks as if it then dawned on Glencore and the brokers that the information given in Cooper Gay's e-mail of 17^th January 2005 was misleading in stating that there would be no operational activity at the Oxygen Plant between commissioning and the final handover. The upshot of the conversation, as recorded by Mr Chapple was an agreement that Glencore's e-mail of 17^th December 2004 (see paragraph 38) should be attached to the cover note and it and the schedule i.e. the bar chart referred to in paragraph 37 above should be scratched by the underwriter. The note records that the underwriter "has received the aforementioned documents". If that was intended to be a reference to the e-mail of 17^th December 2004, it was wrong. The note also records, referring to the terms of Cooper Gay's e-mail of 17^th January 2005, that "these items may not be "shut down" – OK understood".

The brokers' return to the underwriter

91. On 9^th February 2005 Mr Stammers returned to Mr Coward at the box as a result of which he agreed Endorsement No 1, the material parts of which read:

"PREMIUM	Additional Premium USD 15,000 in respect of coverage for projects incorporated into the Mopani Smelting Project until the final handover certificate is issued.
ALL OTHER TERMS, CLAUSES AND CONDITIONS REMAIN UNALTERED
INFORMATION	Glencore e-mail dated 17th December 2004 seen."

92. Mr Coward was shown the 17^th December 2004 e-mail. He wrote against the second of the paragraphs set out in paragraph 38 above the words:

"OK

AP $ 15,000

Deductible

= $ 250,000".

He then put his scratch against those words, and also at the bottom of the whole e-mail.

93. Mr Schaff submits that Endorsement 1 supports the construction for which he contends. The e-mail of 17^th December 2004 necessarily assumes that there is additional operational cover and that the inclusion of "and Operational" in the "Type" paragraph was not a mistake, and provides additional information relevant to an assessment of the risk arising from such cover. The source of that cover cannot be the document provided by way of Information. Such cover must, therefore, derive from the 18^th January 2005 slip. The Information was what enabled the underwriter to assess the additional premium. But it did not define the scope of the cover; nor was the wording apt for the purpose. Further, even if no additional operational cover was provided by the 18^th January contract and the words "and Operational" are to be ignored or somehow written down, the contract as varied by Endorsement No 1 was made in a new context, where there is no justification for any qualification of the words "and Operational" because, as the Information e-mail of 17^th December 2004 showed, operational risks were clearly now to be covered.
94. Mr Edelman submits that, on the contrary, Endorsement No 1 supports his construction of the 18^th January 2005 slip. Since Endorsement No 1 is, on any view, a variation of the previous contract, it is both legitimate and necessary to construe the contract as varied. The copy of the 17^th December e-mail provided by way of Information must have contractual force, once scratched, since it is only in that document that there is reference to a deductible of $ 250,000. It is wrong, therefore, to approach the Endorsement as if any additional operational cover must arise from something other than the scratched 17^th December e-mail.
95. The significance of that e-mail is, firstly, that it records the underwriter's agreement to provide cover for the operational activities specified in it, namely

(i) the production of oxygen between the conclusion of the commissioning of the Oxygen Plant and the commencement of heat-up of the Isasmelt furnace scheduled for 26^th January 2005 (during which time the plant would be operating to produce oxygen to be vented into the atmosphere or, as to a small amount, for storage in a buffer tank);

(ii) the subsequent commissioning of the MSEF;

(iii) the operation of the Oxygen Plant and the MSEF to allow commissioning of the Isasmelt furnace; and

(iv) the operation of the Isasmelt furnace after the start of commissioning in order to allow its exhaust gases to permit commissioning of the Acid Plant.

This is limited operational cover in relation to, or for the purpose of, commissioning: not operational cover for the period after commissioning.

96. Such a conclusion seems to me to be reinforced by the fact that, as I have held, the 18^th January contract did not provide additional operational cover because, as it appeared, it was not needed. In that context Endorsement No 1 is more naturally to be read as an expression of the limited additional operational cover that was now sought than a grant of cover for any and all operations.
97. Secondly, the terms of the document sit very uneasily with the proposition that additional operating cover had already been granted by the 18^th January 2005 slip. If that was so, there would, prima facie, be no need for an additional premium or an increased deductible.
98. As to the latter, under the 18^th January slip, if it covered operating the various items of plant after they had been tested and commissioned, there would only be a deductible in excess of $ 100,000 in the case of losses in respect of wind, storm, flood, water damage and hail: see paragraph 88 above. But under the Endorsement the deductible was $ 250,000 for the operations described in it, which include operating the Oxygen Plant and the MSF for the purpose of commissioning the Isasmelt furnace.
99. I accept Mr Edelman's analysis, and for the reasons given in paragraphs 94-98 I conclude that the operational cover granted by the 18^th January contract as varied by Endorsement No 1 was limited to the operational activities specified in the 17^th December 2004 e-mail.

Deleted Words

100. I turn then to consider the authorities on reference to deleted words.
101. The cases on this topic show, as Diplock, J put it in Louis Dreyfus & Cie v Parnaso Cia Naviera S.A. "The Dominator" [1959] 1 Q.B. 499, 513, "a pleasant diversity of authority on this subject". That diversity is helpfully summarised in Lewison, from which the following account partly derives.
102. In Inglis v Buttery [1878] 3 App.Cas. 552, on which Mr Schaff relies, the House of Lords held that a deleted sentence in a contract for the overhaul and repair of a ship could not be looked at for the purposes of construing the agreement. Lord Hatherley said at p 569:

"When I turn to the deleted words and find that in spite of a line being drawn through them I can read the words….. it appears to me that, those words being deleted, and a marginal note affixed shewing that they were deleted before the contract was finally concluded, it is not in the power of any Court to look at words, for any purpose whatever connected with the construction of that contract of which they form no part whatsoever…… It is to my mind perfectly immaterial whether the instrument was torn up and rewritten, written out again with those words no longer contained in it, or whether the course was taken of running through those words as they stood in writing…"

Lord O'Hagan said words to the same effect, expressly differing from the reasoning of the Lord Justice Clerk, and Lord Blackburn agreed. Lord Gordon expressed no opinion directly on the point.

103. In that case the words in question were :"Ironwork: - The plating of the hull to be carefully overhauled and repaired, but if any new plating is required the same to be paid for extra…Deck beams, diagonal ties, main and spar dock stringers, and all iron work, to be in accordance with Lloyd's rules for classification". The words underlined were deleted before the agreement was made. Not entirely surprisingly, the House held, ignoring the deleted words, that the obligation to overhaul and repair required the shipbuilder to supply without extra charge any new plates required to enable the vessel to be classed 100 A 1 at Lloyd's. The ground for excluding reference to the deleted was that the words were part of the antecedent negotiations, although Lord Hatherley indicated that the same result would follow if the words had been deleted by agreement after execution.
104. In Sassoon (MA) & Sons Ltd v International Banking Corp [1927] AC 711, Viscount Sumner referred to there being "a good deal of authority, now old, about the effect of deleting words in a printed form of mercantile contract….[Their Lordships] take it to be settled in such a case as this that the effect is the same as if the deleted words had never formed part of the print at all".
105. Nevertheless in two cases in the Court of Appeal presided over by Lord Esher, MR, in neither of which was Inglis v Buttery cited, the Court held that it had a right to look at what had been struck out of a printed form: Baumvoll Manufactur Von Scheibler v Gilchrest & Co [1892] 1 QB 253 and Caffin v Aldridge [1895] 2 QB 648. In the latter case Lord Esher treated the deletion of the words "and complete" after cargo in a printed form as showing that the remaining word "cargo" did not mean a complete cargo. These cases appear inconsistent with Inglis v Buttery.
106. In City and Westminster Properties (1934) Ltd v Mudd [1959] Ch 129 Harman J doubted the validity of Lord Esher's approach and suggested that it must be confined, if legitimate at all, to commercial cases where the words struck out appeared on the face of the signed document. As Sir Kim Lewison points out it is difficult to see why a contract should be treated as in a different category because it can be labelled "commercial".
107. In the same year, in Louis Dreyfus & Cie v Parnaso Cia Naviera S.A. "The Dominator" [1959] 1 Q.B. 498, 513, Diplock, J, as he then was, decided that it was unreal to suppose that when contracting parties struck out from a standard form of words familiar to commercial men a provision dealing with a specific matter but retained other provisions they intended to effect any alteration other than the exclusion of the words struck out. He said:

"I cannot prima facie, at any rate, ascribe to them any intention of altering the meaning of words in the provisions which they have chosen to retain. I say "prima facie" because there may be added or substituted words which drive one to the conclusion that they did intend to ascribe to the words retained a meaning modified by the added or substituted provisions; but while I think that I must first look at the clause in its actual form, without the deleted words, if I find the clause ambiguous, I think that I am entitled to look at the deleted words to see if any assistance can be derived from them in solving the ambiguity, bearing in mind the prima facie rule I have indicated."

108. Diplock J held that the words "no loss or damage or delay" in the second paragraph of the Gencon exceptions clause were ambiguous; and had recourse to the third paragraph as an aid to the true construction of the second, even though the third paragraph had been deleted and another sentence substituted. The sense of this decision is plain; but it does not seen to me consistent with Inglis v Buttery. The case is not on all fours with the present case, where no question arises of interpreting one paragraph of a standard form clause by reference to its original
109. In Compania Naviera S.A. v Tradax Export S.A [1965] 1 Lloyd's Rep 198, 204 Mocatta J was persuaded that, in the confused state of the authorities, he should pay no attention to a deleted clause.
110. In two cases in the seventies, in neither of which does Inglis v Buttery appear to have been cited, the House of Lords accepted the admissibility of deleted words at least in some circumstances. In Mottram Consultants Ltd v Sunley (Bernard) & Sons Ltd [1975] 2 Lloyd's Rep 197, 209 Lord Cross of Chelsea said:

"When the parties use a printed form and delete parts of it one can, in my opinion pay regard to what has been deleted as part of the surrounding circumstances in the light of which one must construe what they have chosen to leave in".

Lord Cross then proceeded to do just that.

111. In Timber Shipping Co S.A v London & Overseas Freighter Ltd [1972] AC 15 Lord Reid had explained, obiter, the rationale for looking at words in a printed form differently from other cases of deletion thus:

"There is a controversy as to whether one can ever look at deleted words in an agreement. If words were first inserted by the draftsman of the agreement and then deleted before signature then I have no doubt that they must not be considered in construing the agreement. They are in the same position as any other preliminary suggestion put forward and rejected before the final agreement was made. But it appears to me that striking out words in a printed form is quite different. The process of adapting a printed form to make it express the parties' intentions requires two things to be done. Those parts which are not part of the agreement are struck out and the words are inserted to complete the rest of the form so as to express the agreement. There is no inference that in striking out words the parties had second thoughts: the words struck out were never put there by the parties or any of them or by their draftsman. I must not express a concluded opinion because for some reason this question was not argued by counsel on either side. But fortunately in this case the result is the same on any view."

112. Despite the undoubted distinction to which Lord Reid refers, it might not be thought entirely logical for a party to be allowed to use a deletion from the printed form to establish that the words struck out never formed part of the agreement; when, in the absence of such a deletion, he would not be able to give evidence that a particular provision or term never formed part of the negotiations, or that once suggested, it was instantly rejected.
113. In The Golden Leader [1980] 2 Lloyd's Rep 573 Lloyd J declared that he did not belong to any school of thought which regarded it as inadmissible to look at deletions in a printed clause:

"Thus the use of a word or a phrase in the deleted part of a clause may throw light on the meaning of the same word or phrase in what remains of the clause: see The Dominator [1959] I Q.B. 498. But it seems to me quite another thing to say that the deletion itself has contractual significance; or that by deleting a provision in a contract the parties must be deemed to have agreed the converse. The parties may have had all sorts of reasons for deleting the provision; they may have thought it unnecessary; they may have thought it inconsistent with some other provision in the contract; it may even have been deleted by mistake. Even if I had thought that the first paragraph of the printed clause was ambiguous I would have got very little, if any help from the deletion of the second…"

114. In 1993 in the Court of Appeal case of Jefco Mechanical Services v London Borough of Lambeth 24 B.L.R 1, 8 Slade LJ accepted that it was in principle open to the court in an appropriate case to deduce parties' intentions from deletions in a standard form of contract but derived no assistance from the adoption of one formula in a standard form rather than another one.
115. In The C Joyce [1986] 2 Lloyd's Rep 285 Bingham, J, briefly considered the authorities. Whilst treating the dicta of Lord Reid and Lord Cross as highly persuasive he regarded Inglis v Buttery, Sassoon & Sons Ltd v International Banking Corporation, and Compania Naviera S.A. v Tradax Export S.A [1965] 1 Lloyd's Rep 198, 204 as of such persuasive weight that he could not disregard them and decided, therefore to pay no attention to the terms of a clause deleted from the standard printed form bills of lading.
116. In Wates Construction v Frantham Property [1991] B.L.R 23 the Court of Appeal considered the effect of the deletion of a clause. Bingham, L.J referred to Mottram Consultants; Sassoon; Inglis v Buttery; and Timber Shipping and said:

"It may well be that exceptionally, in the case of a standard form of contract contained in a printed document, the fact that a particular deleted provision may assist in the resolution of an ambiguity in another part of the agreement justifies looking at the deleted part. But I have no doubt that the general rule is that stated by Viscount Sumner, Lord Hatherley and Lord O'Hagan. In any event, even if there may be exceptional cases, in my view, they are of no assistance in construing the agreement in this case".

117. In Punjab National Bank Ltd v de Boinville [1992] 1 WLR 1138 Staughton, L.J. reviewed the authorities. He pointed out that in the Louis Dreyfus case Diplock J looked at the deleted words in a printed form as a guide to the meaning of similar words in a neighbouring paragraph – an approach he regarded as "eminently sensible". He characterised Lord Reid's approach in London & Overseas Freighters and that of Lord Cross in Mottram Consultants as using the fact of deletion as an aid to construction, a different process to that of Diplock J, who had used the deleted words as a guide to the meaning of similar words elsewhere and, again, regarded that as eminently sensible. "The fact of deletion shows what the parties did not want in their agreement". In the instant case there was an endorsement to a policy which formed part of the contract and which stated what was to be deleted and how it was to be replaced. Staughton, LJ, regarded it as "contrary to all reason" to ignore what was deleted in such a case and held that:

"..if the parties to a concluded agreement subsequently agree in express terms that some words in it are to be replaced by others, one can have regard to all aspects of the subsequent agreement in construing the contract, including the deletions, even in a case which is not, or not wholly, concerned with a printed form".

The present case is not one falling within that category. But the underlying principle appears to be that the Court may look at what the parties have, by way of contract, deleted if the fact of deletion shows what it was that they did not want to agree.

118. In Team Service Plc v Kier Management and Design Ltd (1993) 63 BLR 76, the majority of the Court of Appeal, following "The Golden Leader", "The Dominator", Mottram v Sunley, and London & Overseas Freighters held that in construing a one off contract apparently based on a printed form it was possible to consider the parts of the relevant clause that had been deleted. Lloyd, L.J. also said that he could see no distinction between a deletion in a printed form of contract and a deletion in a printed form of clause included in a one off contract; or between a deletion and an omission saying:

"Of course, it would be necessary to show that the omission was deliberate. But if the court is satisfied then the omission is as much a surrounding circumstance as a deletion".

In that case the effect of looking at certain omitted words had a material bearing on the construction of the contract.

119. In Rhodia Chirex v Laker Vent Engineering Ltd 2004 BLR 75, 84 Auld, LJ expressed doubt as to whether a deletion was admissible as an aid to construction.

Conclusion

120. The diversity of authority, of which Diplock, J, spoke, renders it difficult for a judge of first instance to recognise when recourse to deleted words may properly be made. The tenor of the authorities appears to be that in general such recourse is illegitimate, save that (a) deleted words in a printed form may resolve the ambiguity of a neighbouring paragraph that remains; and (b) the deletion of words in a contractual document may be taken into account, for what (if anything) it is worth, if the fact of deletion shows what it is the parties agreed that they did not agree and there is ambiguity in the words that remain. This is classically the case in relation to printed forms (Mottram Consultants, Timber Shipping, Jefco Mechanical Services), or clauses derived from printed forms (Team Service), but can also apply where no printed form is involved (Punjab National Bank Ltd).
121. Support for that view may be found in the latest edition of Keating on Building Contracts which contains the following passage:

"In this confusion the second school is generally to be preferred. Where parties have made a contract in a document that contains deletions, to look at the deletions does not offend the principle discussed above which prevents reference to preliminary negotiations. The deletion is physically contained in the concluded contract. It is submitted that the court should first construe the retained words. If they are unambiguous, reference to the deletion is unnecessary. If they are ambiguous, reference to deletions from printed documents should be permitted to see whether objectively they throw light on the meaning of the retained words."

122. Even if recourse is had to the deleted words, care must be taken as to what inferences, if any, can properly be drawn from them. The parties may have deleted the words because they thought they added nothing to, or were inconsistent with, what was already contained in the document ; or because the words that were left were the only common denominator of agreement, or for unfathomable reasons or by mistake. They may have had different ideas as to what the words meant and whether or not the words that remained achieved their respective purposes.
123. Further, as Morgan, J, pointed out in Berkeley Community Villages Ltd v Pullen [2007] EWHC 1330:

"Even in the cases where the fact of deletion is admissible as an aid to interpretation, there is a great difference between a case where a self contained provision is simply deleted and another case where the draft is amended and effectively re-cast. It is one thing to say that the deletion of a term which provides for "X" is suggestive that the parties were agreeing on "not X"; it is altogether a different thing where the structure of the draft is changed so that one provision is replaced by another provision. Further, where the first provision contains a number of ingredients, some assisting one party and some assisting the other, and that provision is removed, it by no means follows that the parties intended to agree the converse of each of the ingredients in the earlier provision."

124. In the present case the parties had, in the 29^th November 2004 reached a concluded agreement which included the pencilling out of certain words and the addition of the notation "TBA MLM". In those circumstances it seems to me legitimate to look at what by and in their contract of that date they agreed that they had not agreed.
125. Against that background I also regard it as legitimate to take into account that the relevant words remained deleted in the slip of 18^th January 2005 (but now in ink) because that deletion, when taken with the (contractual) e-mails document, shows (a) what was still not agreed, and why it was not agreed, and (b) that the non-deletion of the words "and Operational" in the "Type" paragraph did not now signify that operational cover, other than that inherent in cover for Testing and Commissioning, was part of the reinsurance.
126. Accordingly the answers that I give to the third, fourth and fifth issues are as follows:

Issue 3: On the true construction of the original contract of insurance and of the contract of reinsurance, was there cover for completed elements of the project once they became operational?

127. The contract of reinsurance did not cover the completed elements of the project once they became operational. I do not understand it to be suggested that the insurance was on any different terms to the reinsurance.

Issue 4; If the contract of insurance and/or the contract of reinsurance as originally agreed did not include cover for the ESP once it had been brought into operation, on its true construction did endorsement number 1 operate so as to vary the terms of the contract of insurance and/or reinsurance and to have the effect that they would:

i as the Claimant contends, provide cover for the ESP during any operations prior to final handover of the entire project? or

ii as the Defendant may contend, add nothing to the original reinsurance policy? or

iii as the Defendant in any event contends, provide cover limited to only such operations as were described in the scratched email printout referred to in endorsement no. 1?

128. The answer to this issue is as in (iii).

Issue 5: If the answer to questions 3 or 4 (i) is yes, is it nevertheless open to the Defendant to say that commencement of the commercial operation of the ESP represented a material alteration to the risk originally insured and/or reinsured?

129. This question does not arise.

The Memorandum of Understanding

130. There was at one time an issue as to whether the MOU had been signed by Mopani, but it is now apparent that it was signed by Mopani, Zigi, Mr Coward on behalf of reinsurers, and Cooper Gay. Accordingly, the answer to Issue 1: "Were the Claimant and Zigi party to the MOU?" is Yes.
131. Clause 2 of the MOU provided:

"Monies due to the Insured in respect of claims as agreed by Reinsurers hereon and return premiums hereunder shall be paid by Reinsurers to Cooper Gay & Co Ltd for onward transmission to Mopani. Any payments made by Reinsurers shall be considered a full and proper discharge, to the extent of such payment by reinsurers of their liability to Zigi Insurance Limited."

132. This clause does not entitle Mopani or its brokers to recover in respect of any claim that has not been agreed. But Millennium accepts that Mopani is entitled to seek a declaration in respect of the coverage granted by the reinsurance and has confirmed that, if a declaration is given in Mopani's favour, the reinsurers will make such payment as is appropriate in accordance with the terms of the MOU. In those circumstances the second issue: "If so, is the Claimant entitled to sue for such sums as would, in the absence of the MOU, have been payable by the Defendant to Zigi?" is moot.
133. I invite Counsel to consider what form of order should be made to give effect to these findings.

Note 1   The precipitator accepts gas from the waste heat boiler at 350°. It contains three separate “fields” each of which removes about 85% of the dust entering the field so that the total dust removed across the three fields is about 99.6%. Each field contains 22 collector plates between which are interspersed 21 electrode assemblies. A high voltage emission from the electrodes imparts a positive charge onto the dust particles which drives them to the earthed collector plates. A rapping mechanism dislodges the dust from the plates into hoppers from which the dust is conveyed back into the smelting process.     [Back]

Note 2   Delay in Start Up.    [Back]

Note 3   Sometimes called Delay in Start Up. The cover is for the profits that would have been made but for the delay.    [Back]

Note 4   This was true only in the sense that, if there was no ALOP cover, there would be no problem, provided that the relevant individual units were handed over to the operational policy after they were tested and commissioned, and provided that it was clear what 30 days T & C did and did not cover.    [Back]