BAULDERSTONE HORNIBROOK ENGINEERING PTY LTD v GORDIAN RUNOFF LTD & ORS

The word "paid" had its usual or ordinary meaning and meant funds disbursed; here, there was no payment: [282].

HIH made a recognition or admission of liability under the policy by granting indemnity to BHE: [290].

The "grant of indemnity" or admission of liability may be seen, in certain circumstances, to give rise to another agreement: [291]. Lubovsky v Snelling [1944] KB 44 Law Society v Sephton & Co [2004] EWHC 544 Thiess Pty Ltd v ERC Frankonia Reinsurance Limited [2007] QSC 4

HIH could still be held liable. The acceptance of moneys which gave partial reparation to BHE for its whole loss from a full indemnity under the AMPG policy, did not deny BHE the right to sue HIH for indemnity under its policy because there still remained loss to be the subject of reparation or indemnity: [302], [309].

Where several insurers have severally insured a risk, the insured can only receive one indemnity for its loss. Upon receipt of that full indemnity from one insurer, the others are discharged against the insured. They remain, however, liable to contribute to the insurer that has paid. They are discharged because the insured can have no claim, because it has received full reparation for its loss: [298]. Albion Insurance Co Limited v Government Insurance Office of New South Wales (NSW) (1969) 121 CLR 342

Once an insured's loss has been the subject of reparation, the insured cannot claim indemnity from another insurer in respect of the same loss, because by then there is nothing to indemnify. Subject to the terms of the policies, if there are several insurers, each is liable under its terms until the insured is provided with full reparation or indemnification for its loss: [299]. GRE Insurance Ltd v QBE Insurance Ltd [1985] VR 83

The "excess point" and the operation of exclusions 1(e), 1(o) and clause 7

Settlement of the claim was reasonable upon the information available at the time of settlement: [324], [325], [327].

Exclusions 1(e) or 1(o) or clause 7 did not engage. The design requirement of the work that the reinforced earth walls be maintenance free for 50 years was not a warranty or guarantee and did not amount to assumed contractual liability but was an inherent part of the design. BHE's liability arose by reference to the task it undertook: [328]. IN THE SUPREME COURT OF NEW SOUTH WALES COURT OF APPEAL CA 40316/06

ALLSOP P BEAZLEY JA CAMPBELL JA

BAULDERSTONE HORNIBROOK ENGINEERING PTY LTD v GORDIAN RUNOFF LTD & Ors JUDGMENT ALLSOP P: 1 Contents Introduction [2]

The nature of reinforced earth wall building technique [15]

The judgment below, the approach of the primary judge and the arguments below - a brief introduction [26]

The primary judge's reasons [43] • The primary judge's "overview findings" [46] • The primary judge's findings in relation to witnesses [47] • The primary judge's comments on the course of the hearing [56] • The width of the hand compaction zone [59] • The evidence of Mr Price [60] • The primary judge's detailed analysis of the reasons for sand loss [61] • Insurance and the causal mechanics [95] • Other aspects of the judgment below [122]

Whether BHE's arguments on appeal were run below [136]

The assessment of BHE's argument on appeal [171] • The contractual documents - the head contract [174] • The contractual documents - the subcontracts [192] • The design documents [199] • Conclusions as to the argument of mandated compaction technique [225] • The questions of design and construction and the the operation of the policies - why the appeal should be dismissed [228]

• Whether the primary judge's conclusion about the default of RE P/L and CW as to the precautionary instructions should be set aside [269] • The proper construction of special provision 2 as a so-called stand alone insuring clause [272] • The so-called Gordian attachment points [276] • The so-called "excess point" and the operation of exclusions 1(e), and 1(o) and cl 7 [311] • The lack of separate treatment of RE P/L and CW [330]

Introduction 2 This is an appeal from orders made by a Judge of the Court (Einstein J) dismissing a claim made by the appellant, Baulderstone Hornibrook Engineering Pty Ltd ("BHE"), for recovery under two polices of insurance held by it, which policies were underwritten by the first and second respondents, GIO Insurance Ltd, now known as Gordian Runoff Ltd ("Gordian") and Commercial Union Assurance Co of Australia Ltd, now known as CGU Insurance Ltd ("CGU"). (For the reasons of the primary judge, see Baulderstone Hornibrook Engineering Pty Limited v Gordian Runoff Limited (formerly GIO Insurance) & Ors [2006] NSWSC 223.) The third respondent, HIH Casualty and General Insurance Ltd (In liquidation) ("HIH") was an insurer under a policy in the programme of insurance of which the Gordian and CGU policies were also part. 3 Gordian and CGU were two of the excess of loss primary insurers in a professional indemnity programme of insurance under which HIH underwrote the bottom layer, insuring the first $20 million. Gordian provided the next $10 million above that $20 million. Thereafter, three insurers, QBE Insurance Ltd ("QBE"), CGU and a syndicate at Lloyd's provided three further layers of $6m, $5m and $9m (ie $6 million over $30 million, $5 million over $36 million and $9 million over $41 million, respectively) thereby completing a programme of $50 million of professional indemnity insurance. 4 This programme of insurance was put in place by the principal of the design and construction contract to which BHE was a party as head contractor, originally the Federal Airports Corporation ("FAC") and later, after the privatisation of Sydney Airport, Sydney Airports Corporation Ltd ("SACL"), for the benefit of all consultants undertaking responsibilities in the project, which was the design and construction of the third runway at Sydney Airport in the 1990s. 5 Speaking broadly, the policies in question were intended to cover professional indemnity risks and not construction risks. That is an inadequate identification of the problem to be solved in this appeal, as it masks some of the more precise difficulties in the resolution of the appeal. It does, however, sufficiently identify the area of discourse and debate at this point. 6 BHE had its own policy of professional indemnity insurance underwritten by AMP General Insurance Co Ltd ("AMPG"). This policy provided for insurance for breach of professional duty up to $20 million for any one claim, and $40 million in the aggregate for more than one claim. Before AMPG was joined to the proceedings that the primary judge disposed of, it agreed to indemnify BHE for $20 million; and this sum was paid in January 2005. BHE claimed a further $20 million from AMPG in relation to what was said to be a second claim. AMPG was joined to the proceedings by BHE seeking to vindicate this assertion of a second claim. This claim against AMPG was dismissed by the primary judge, from which dismissal there was no appeal. 7 BHE was the head contractor. It engaged a number of subcontractors in effecting its responsibilities. Relevantly for these proceedings, these subcontractors included Reinforced Earth Pty Ltd ("RE P/L"), which was retained in relation to the design of the reinforced earth walls on the project and Connell Wagner NSW Pty Ltd ("CW"), which was retained in relation to the provision of engineering services in relation to the third runway, including the reinforced earth walls. 8 A dispute arose between FAC, and later SACL, on the one hand, and BHE, on the other, about the adequacy of the construction of the reinforced earth walls forming the perimeter of the third runway and on an adjacent area known as the Millstream. Work had commenced in early 1993 and practical completion occurred in August 1994. By late 1996, subsidence of backfill behind the facing panels of the reinforced earth walls was discovered. BHE notified its insurers. In June 2002, SACL sued BHE for damages in the Building and Construction List of the Equity Division of the Court. In June 2004, those proceedings were settled, the settlement involving BHE undertaking to rectify the defects at a likely estimated cost of a sum exceeding the levels at which both the Gordian and CGU policies would fully respond. BHE sought indemnity from the insurers participating in the professional indemnity programme to which I have referred. Proceedings were commenced against the insurers in December 2003. Initially those proceedings did not include AMPG, but after Gordian and CGU raised defences that the proceedings by SACL involved two claims, BHE sued AMPG claiming the second $20 million. Issues now remain for decision only in relation to the policies underwritten by Gordian and CGU; though, because of the terms of the Gordian and CGU policies, it is necessary to discuss the terms of the HIH policy. 9 On the evidence before him, the primary judge concluded that neither the Gordian nor the CGU policy responded, because the liability of BHE for SACL's claim against it under the design and construction contract could be relevantly described for the purposes of the policies as arising out of (uninsured) construction risks and not arising out of (insured) professional indemnity risks. 10 The appellant challenged this conclusion by the primary judge not by an attempted re-litigation of all the factual issues that were before his Honour, but by accepting many (though not all) of the factual findings of the primary judge and the evidence upon which they were based. 11 The essential argument of the appellant was that when one examined the substance of the findings of the primary judge and the substance of the evidence of the expert witness called by Gordian, Professor Leshchinsky, and of the senior officer of RE P/L, Mr Boyd, the only proper conclusion was, and is, that the defaults by BHE, RE P/L and CW, were of a character of engineering or design, as elaborated in the contractual documents and the terms of the policies, and not of a character of construction. 12 In rebutting these claims, Gordian and CGU, asserted that the appellant was putting a case not propounded before the primary judge at the trial, being a case which, if put below, would have led to a different course being taken by them at the trial and which could have been met by evidence that was not called. Thus, it was said that the essential argument of the appellant should not be entertained. In any event, Gordian and CGU submitted that, by reason of the way the case was presented to the primary judge for disposition, his Honour was permitted to make the findings that he did, and, in that context, the findings mandate the conclusion that the relevant insurance policies do not respond, either because the events fall outside the primary insuring clauses, or because they fall within exception clauses within those policies. 13 In my view, the arguments put on appeal by BHE were encompassed in the case put at the trial. Nevertheless, I would dismiss the appeal for the following reasons. 14 At the outset, it is helpful to describe the nature of reinforced earth walls and how that technique was used here.

The nature of reinforced earth wall building technique 15 It is appropriate to begin by explaining the technique employed to build the wall that was placed around the perimeter of the runway to protect it from erosion by the waters of Botany Bay and the wall which was adjacent to a channel diversion just to the north of the runway, called the Millstream Channel Diversion. 16 Reinforced earth walls comprise compacted earth in association with reinforcing materials. Typically, granular earth and linear tensile reinforcements are combined to form a block of reinforced earth which acts monolithically as a mass gravity retaining structure. The technique used by RE P/L was said to be proprietary technology referred to in the contemporaneous documents as "Reinforced Earth". 17 The first report of Professor Leshchinsky gave an explanation of the subject walls. The reinforced earth walls were constructed using prefabricated concrete panels on the external, marine side of the reinforced compacted earth. The width of the wall measured by the length of the reinforcing strips from the inside of the external facing panels was a little over five metres. The precast panels in the Seawall (that is on the perimeter of the runway) had horizontal and vertical joints about 15-20 mm wide, and were "Double T" type panels. On the Seawall the horizontal joints were sealed. The Millstream walls employed cruciform shaped panels which, when assembled, had vertical and horizontal joints of the same width. Both the vertical and horizontal joints on the Millstream walls were unsealed. 18 The unsealed joints (vertical in the Seawall and horizontal and vertical in the Millstream walls) were covered on the inside of the panels with strips of non-woven needle-punched polyester geotextile material. The geotextile was intended to be semi-permeable, allowing water to flow through (in and out) while retaining the compacted earth. At the Seawall, the geotextile was placed around no-fines blocks that were placed behind the vertical joints. 19 The reinforced earth was placed behind the concrete panel walls. This earth was compacted in a manner to be discussed in due course. For present purposes, it is sufficient to understand that at prescribed heights on the internal face of the panels metal strips were affixed to the panels, and laid over compacted earth. A further layer of earth was placed and compacted over these strips. Thus, by a process of force, tension and friction, the strength of the compacted earth as a mass gravity retaining structure was increased by this reinforcement. 20 The compaction of the reinforced earth was an important aspect of the integrity of the structure. It is important to understand that compaction took place, relevantly, in two areas and by two different methods. First, in the area not immediately adjacent to the inside face of the concrete panels, a heavy motorised roller was used to compact the earth. The use of such equipment too close to the panels might have placed sideways force on the panels and thereby threatened their integrity. So, in the area adjacent to the inside face of the panels, hand-held compaction equipment was used to compact the earth. 21 The configuration of structures inside the concrete panels also necessitated compaction close to the panels. There were buttresses behind the Seawall panels. Also, no-fines blocks were placed behind the vertical joints in the Seawall. Thus the compaction that was required had to be effected around these structures which broke the straight line of the inside of the concrete panels. 22 The adequacy of the compaction in this latter area (to which I will sometimes refer as the hand compaction zone or area, and sometimes as the one metre zone or area) and the adequacy (or more relevantly, inadequacy) of any testing of compaction immediately adjacent to the inside face of the panels were issues central to the resolution of the case at trial. The method of compaction in this area and the relationship between construction and design responsibilities insofar as they bear on that method of compaction were issues central to the resolution of the appeal. 23 This introduction about the technique of constructing reinforced earth walls can be helpfully supplemented by reference, first, to the so-called "Recommendations and Rules of the Art" for reinforced earth structures issued (in translation) by the French Ministry of Transport, upon which RE P/L's proprietary technology was based; and, secondly, to RE P/L's "Reinforced Earth Wall Construction Manual". In the former document, the construction process was described as follows: The construction process of the reinforced earth is quite similar to the construction of the classical embankments, it is realized by successive layers. It includes the following stages: - Placing of the facing elements (fig 1.1.1) - Backfilling with the soil layer, eventually under compaction (fig 1-2) - Placing of the reinforcement strips and their attachment to the facing by the bolts (fig 1-3) - This construction can be entirely realized from the internal side of the facing and therefore it does not require any scaffolding or any external arrangement. The reinforced earth mass is thus constituted from soil layers placed between the uniformly spaced horizontal layers of the reinforcement strips. It is generally, externally limited by the vertical facing (fig 1-4). 24 In the latter, the following introduction was given: Reinforced Earth is a composite material formed by the association of a frictional soil and reinforcing strips. In concept, it is like reinforcing concrete; that is, it is an economical means of improving the mechanical properties of a basic material, earth, by reinforcing that material with another, steel. Stresses produced within the soil mass are resisted by the strips. The stresses are transferred to the strips by friction. … Concrete facing panels are used at the face of the reinforced volume to prevent erosion of the backfill and to provide an attractive, finished appearance. 25 Before concluding this introduction to the reinforced earth technique, it should be noted that the structure had three vertical zones: the deepest or lowest zone which was always submerged (sometimes referred to as zone 1); an intermediate zone which was regularly submerged by the operation of the tide, swell and waves (sometimes referred to as zone 2 or tidal zone 2); and a higher zone which was only intermittently subjected to water by wave action (sometimes referred to as zone 3).

The judgment below, the approach of the primary judge and the arguments below - a brief introduction 26 It is important to appreciate that the primary judge heard the case for 46 days from 11 October 2005 to 3 March 2006. Thirty five of those days were consumed by the opening and the evidence and 11 by final addresses. I remark on these matters, not merely to pay tribute to the primary judge's despatch in the resolution of a long, complex, hard-fought technical case, but to emphasise that the way the trial was fought and the arguments that were put (including their clarity or obscurity) were fresh to the mind of the primary judge, who plainly exerted considerable effort to produce a judgment as promptly and as meticulously as he could. 27 In his introduction to the judgment, the primary judge identified the central issue as the true cause of the sand loss behind the walls. Once one understood the cause of that loss, one could characterise the claims made for indemnity under the policies. No complaint was made about this approach. His Honour identified BHE's contentions as containing three elements: (1) that the excessive sand loss was caused by defects in the design of the walls; (2) that the construction work was not defective in any way, in particular the specified degree of compaction to 80% Density Index, (to which I will refer as "80% DI") was attained in the area one metre from the panel walls; and (3) the construction conformed with a flawed design. The primary case of BHE was that it had compacted the earth behind the walls to the requisite degree of compaction. It identified the cause of the sand loss as the inappropriately specified geotextile material that was placed over the unsealed joints. This geotextile material was said to have holes that were too large, such that water was permitted not only to enter and leave (as intended), but also sand was permitted to escape (not as intended). This primary case was supported by an expert, Professor Ingold. The various directional changes and iterations of Professor Ingold's evidence were dealt with by the primary judge. 28 In contrast, Gordian and CGU contended that there was no design defect; that the cause of excessive sand loss was the failure of BHE properly to construct the walls by its failure to compact uniformly to 80% DI and by its failure to test and certify that fact; that in the alternative, if inadequate compaction was not the sole cause of the sand loss, it was a substantial contributing cause, which alone would have been sufficient to have caused the need to undertake the rectification of the work; and, further in the alternative, that the sand loss was caused by defective affixation (in construction) of the geotextile material, permitting sand to be trapped between the panel wall and the geotextile creating a path for sand to escape. 29 These contentions were supported by Professor Leshchinsky. The main thrust of his views were ultimately accepted by the primary judge as a reliable explanation for the loss of sand. To a degree, however, Professor Leshchinsky's views developed as the hearing progressed (though not to the degree that those of Professor Ingold did). 30 It is important to appreciate that the above contentions are a summary of the issues that were in contest at the commencement and during a significant part of the course of the trial. The causal mechanisms identified by Professor Leshchinsky in his reports and cross-examination, came to form the foundation of an alternative case propounded by BHE. It is the extent to which this alternative case of BHE, as run at the trial, can be seen to encompass the argument put by BHE on appeal that lies at the heart of the debate about the legitimacy of the approach BHE has taken on appeal.

The insurance policies 31 Before examining the primary judge's reasons, it is appropriate to identify the relevant clauses of the insurance policies in question. Each of the Gordian and CGU policies adopted the terms of the HIH underlying policy, to the extent the HIH wording was not inconsistent with their respective wordings. Thus, it is convenient to begin with the HIH policy. 32 The named insureds in the HIH policy included BHE and CW (being specifically named) and RE P/L (as falling within the class described as "all consultants"). The "profession" was stated in the policy Schedule to be "engineers and as defined herein". The primary insuring clause (to which I will refer as "insuring clause 1" or "the insuring clause") was in the following terms: To indemnify the Insured against any claim or claims which may be made against them or any of them and which are notified to the Company during the period specified in the Schedule, for breach of professional duty in the profession stated in the Schedule, by reason of any act, error or omission whenever committed unless limited by the retroactive date stated in the Schedule or wherever the same was or may have been committed or alleged to have been committed on the part of the Insured in the conduct and execution of the Professional Activities and Duties as defined herein. 33 The "Professional Activities and Duties" were defined in the policy and endorsement as follows: The Professional Activities and Duties in respect of which coverage is provided is deemed to be those services as set out in the Schedule as well as those defined hereunder. This form of the clause was found in the endorsement which deleted a paragraph in the standard terms, and substituted the above clause. In the standard terms, after the clause that was replaced, there appeared three numbered paragraphs entitled "1. Advisory Services", "2. Predesign and Feasibility Studies" and "3. Miscellaneous Services (ancillary to the main design criteria) Related to the Design and Construction Phases" . Each of these had various sub-paragraphs. Under the third paragraph ( "3. Miscellaneous Services" etc) the following sub-paragraphs were included:

… (i) Checking of designs and plans prepared by others (j) Preparing detail drawings …

The endorsement also contained the following paragraph under the heading "Professional Activities": Engineering, project management, surveying, designing, geotechnical, environmental monitoring, construction management, certification and as defined in the policy wording. 34 Thus, the HIH policy can be seen to relate to engineers and the professional activities and duties described above. 35 The second (and presently irrelevant) insuring clause dealt with defence costs. 36 A number of clauses were then set out under the heading "Exclusions". Relevantly, the following appeared: 1. This Policy shall not indemnify the Insured in respect of any claim made against them: … (e) for any liability of whatsoever nature imposed upon the Insured pursuant to the Law of Contract but only when such legal liability would not have been imposed upon the Insured pursuant to the Law of Tort. … (o) arising out of the provision by the Insured of any express warranty or guarantee which increases the Insured's measure of liability but this exclusion shall not apply to liability which would have attached to the Insured in the absence of such express warranty or guarantee to the extent that such is attributable breach of professional duty otherwise insured hereby. (p) arising out of construction work performed involving the means, methods, techniques, sequences, procedures and use of equipment, of any nature whatsoever which are employed by the Insured's contracting staff or others in executing any phase of any Project. 2. There shall be no liability hereunder in respect of any claim for which the insured is entitled to indemnity under any other insurance required to be effected by or under any law or indemnity or under any statutory fund or fidelity fund, except in respect of any amount beyond that which is payable or would have been payable under such other insurance, statutory fund or fidelity fund, had this policy not been effected. … 7. This policy shall not indemnify the Insured in respect of any claim made against them which relates to any contractual or other duty or obligation assumed by the Insured which is not assumed in the normal conduct of the Insured's profession as stated in the Schedule. 37 After the section entitled "Exclusions", there was a section headed "Special Provisions" under which the following two provisions appeared (the second having been altered by the deletion of a paragraph by a clause of the endorsement): 1. It is understood and agreed that, where the Insured act in more than one capacity in the same project that is where they are, at the same time: (a) Building or Project Owners and Contractors and Designers, or (b) Contractors and Designers, or (c) Building or Project Owners and Designers and the Insured in any of these capacities incur loss, damage or expense arising from any act, error or omission in connection with the professional activities as defined herein, the Company agrees to consider it as if a claim had been made against the Insured provided that such loss, damage or expense is not insured by any other policy. "Designers" in the above context shall extend to include all those professional activities and duties defined in and insured by this policy. 2. The coverage provided by this Policy is extended to indemnify the Insured, subject to the Policy's terms and conditions, against their legal liability for claims (and/or against loss, damage or expense as defined in Special Provision 1) arising out of any act, error or omission in the conduct of professional activities or duties committed by specialist designers or consultants acting on the Insured's behalf pursuant to any contract for service and for whom the insured are responsible. The Company's total liability for all claims under this Policy including this Special Provision, shall not exceed the sum stated on the Certificate representing the limit of indemnity. 38 There then followed a section entitled "Conditions" containing six clauses which included a claims notification clause (condition 4), in the following terms: If during the subsistence hereof the Insured shall become aware of any circumstance which may subsequently give rise to a claim against them for breach of professional duty as specified in the Schedule by reason of any act, error or omission and shall during the subsistence hereof give written notice to the Company of such occurrence, then any such claim which may subsequently be made against the Insured arising out of that act, error or omission shall for the purposes this policy be deemed to have been made during the subsistence hereof. 39 There was a trade practices and related legislation extension. 40 In the Gordian policy, the following appeared as clause C: The GIO shall only be liable in respect of the indemnity herein given after the Underlying Insurers have paid or have admitted liability or have been held liable to pay the full amount of their indemnity provided always that the liability of the GIO under the indemnity herein given shall be limited to the amount in respect of any one claim and in the period of insurance stated in the Policy Schedule. It is a condition of this Policy that the Underlying Insurances shall be maintained in full effect during the currency of this Policy. 41 It is unnecessary to set out any of the separate CGU policy wording. 42 As will be discussed more fully in due course, the principal insurance issues that arose on appeal from these policies were:

(a) Was insuring clause 1 engaged so as to indemnify BHE against SACL's claim against it?

(b) Was special provision 2 engaged so as to indemnify BHE against SACL's claim against it?

(c) Was exclusion 1(p) operative to deny indemnity?

(d) Was Gordian clause C, dealing with attachment of the Gordian policy, satisfied?

The primary judge's reasons 43 It is necessary to examine the primary judge's reasons in some detail. From time to time, for convenience and as an aid to comprehension, reference will be made in this part of these reasons to some of the evidence and arguments, though these will be dealt with more fully later. 44 The primary judge commented upon the difficulty of making factual findings in what he described as a "very complex environment". This was a reference to the changes in approach of the witnesses (in particular Professor Ingold), the changes in direction of BHE (in particular after Professor Ingold's cross-examination), the difficulty caused by BHE's late assertion of the width of the hand compaction area as 1.5m, the technical complexity of the issues and the volume of material placed before the court in evidence. In the light of the debate about what was, and was not, run at the trial these (entirely legitimate) comments of the primary judge should be borne in mind. 45 Also, it is unnecessary to deal with many of the issues resolved by the primary judge.

The primary judge's "overview findings" 46 The primary judge commenced by making what he referred to as "overview findings". Relevantly for understanding the issues in the appeal, these were:

(a) The hand compaction zone extended one metre from the back of the Seawall and the Millstream walls.

(b) The requisite compaction density of 80% was not achieved in the zone one metre from the back of the Seawall.

(c) There were no records of any testing within the one metre zone from the back of the Seawall and no testing took place there.

(d) There were some records of testing in the one metre zone behind the Millstream walls early in the work. After this testing in the early stage of the Millstream walls work a conscientious approach to testing by BHE fell away.

(e) The requisite compaction density of 80% was also not achieved in the zone one metre from the back of the Millstream walls, except in a few positions.

(f) There was no particular appreciation by BHE of the significance of achieving the 80% compaction (implicitly in the one metre zone, see heading above [76] of the reasons).

(g) The sand loss was caused by operations of mechanisms one and two as described by Professor Leshchinsky which were as follows: i. the first being that if the fill had not, when originally laid, been compacted to the 80% Density Index, it would tend to settle under the influence of tide in the area adjacent to the joints to a distance of 100mm… ii. the second being that if the sand was originally compacted to the 80% Density Index in 660mm layer lifts, the compactive forces involved would induce stresses into the geotextile causing it to be stretched and possibly fold. The primary judge's findings in relation to witnesses 47 The opinions of Professor Leshchinsky were preferred to those of Professor Ingold whose opinions were found to be unreliable. Professor Ingold was found not to be objective as a witness. 48 The primary judge discussed in detail lay witnesses who were called: Mr Caretti, a project-based BHE site engineer from late 1992 to mid 1994; Mr Ashton, a civil engineer employed by BHE on the job from 1992 to 1994 who was, from June 1993 until August/September 1994, the section manager responsible for overseeing construction of the Millstream walls; Mr Moore, who was a project manager at BHE and who had no involvement in the construction of either the Seawall or the Millstream walls; Mr Scott, a project-based BHE site engineer from late 1992 to early 1994.

Mr Caretti 49 Mr Caretti was relevantly involved in the project on the Millstream walls as part of a team of foremen and engineers responsible for supervising the construction of the reinforced earth structure. His responsibilities included placement of the geotextile over the joints and compaction of the backfill. He had no involvement in, or knowledge of, the compaction in relation to the Seawall. 50 The primary judge reviewed Mr Caretti's evidence in detail, in particular that relating to compaction. The primary judge noted that the burden of an extensive cross-examination was as to Mr Caretti's assertion of testing of compaction in the one metre zone. The primary judge rejected Mr Caretti's evidence about adequacy of compaction and in particular that the references to "1m" in testing documents could be understood as within one metre; but he was prepared to accept "a deal of matters which Mr Caretti recalled having occurred".

Mr Ashton 51 Mr Ashton's role as section manager responsible for overseeing construction of the Millstream walls was a management role which gave him responsibility for, amongst other things, the quality of works and the achievement of the construction plan. 52 The primary judge referred to the cross-examination directed to the geotextile and its placement and the importance of the 80% compaction. The primary judge accepted Mr Ashton as reliable, his early noticing of sinkholes and his view that they were not matters revealing any significant problem.

Mr Moore 53 Mr Moore investigated defects in the project including sand loss and was in 2003 Project Manager for the rectification work. Mr Moore's investigations gave substantial support to the primary judge's findings that (apart from early in the Millstream walls' construction) there was no record of, and so no testing for, compaction in the hand compaction zone. Importantly, Mr Moore said that in his investigations he had asked site engineers, including Mr Scott, whether there had been testing in the one metre zone and that they (including Mr Scott) had said that they did not recall the requirement to test each layer in the one metre zone.

Mr Scott 54 Mr Scott, who was a recently graduated engineer, was involved in the Seawall construction for the first 22 (out of 77) lots - 2km (out of 7km). His evidence concerned, amongst other things, compaction in the hand compaction zone, and compaction testing in that area. 55 The primary judge rejected Mr Scott's evidence as unreliable, in particular that given by him that satisfactory compaction testing was done in the one metre zone and that the hand compaction zone was 1.5 metres. The primary judge found that the hand compaction zone was one metre from the base of the internal wall of the panels.

The primary judge's comments on the course of the hearing 56 Shortly after dealing with this evidence from Mr Scott and others about the width of the hand compaction zone, the primary judge made some comments as to how BHE had run the trial. For many weeks of the hearing, BHE held back its important contention that the hand compaction zone was 1.5 metres from the back of the Seawall in the knowledge that the other parties to the litigation were proceeding on the assumption that the hand compaction zone was one metre. Inferentially at least, the primary judge's findings in this respect were that BHE not only knew that the other parties were under a misconception as to its (BHE's) case, but that BHE knew that they (the other parties) assumed it was common ground that the hand compaction zone was one metre. 57 The primary judge said the following at [238] of his reasons: … the litigious process is not furthered but is hindered where a crucial issue is not highlighted as such at an early stage in the proceedings, so as to permit the matter to be properly tested through appropriate witnesses, in the full light of day. Here BHE stepped perilously close to, but did not overstep the relevant line. (emphasis in original) 58 No occasion arises on the appeal to consider the correctness of this conclusion, including in particular, his Honour's conclusions in the last sentence quoted. It will, however, be necessary to deal with a similar issue as to how BHE now puts its case on appeal.

The width of the hand compaction zone 59 After an exhaustive review of the evidence, the primary judge concluded that on the Seawall the hand compaction zone was one metre from the back of the wall. There was no issue about this with the Millstream walls which had no buttresses and no no-fines blocks. No challenge to these conclusions was made on appeal.

The evidence of Mr Price 60 Mr Price was the solicitor on the record for BHE until May 2004. He conducted a detailed investigation and gave detailed evidence of the claim and its history. This was particularly relevant to the assertions and counter assertions by BHE and the insurers of a lack of good faith of the other. It is unnecessary to discuss at length his Honour's treatment of Mr Price's evidence. Two aspects are, however, relevant. First, it became obvious to Mr Price in the course of his investigations that BHE had conducted no compaction testing close to the back of the panels. Secondly, Mr Price gave evidence about his advice to BHE in the settlement of the proceedings brought by SACL. This evidence formed the basis of his Honour's conclusion that the settlement was reasonable.

The primary judge's detailed analysis of the reasons for sand loss 61 After dealing with the making of the claim and the resolution of the underlying dispute with SACL, the primary judge, over more than 80 pages, examined the evidence about the reasons for the sand loss. He treated this (and there is no argument about this) as an immediately antecedently relevant question to the issues of design and construction and so as relevant to the response of the Gordian and CGU policies. 62 After explaining necessary terms and concepts, the primary judge turned first to Professor Ingold's evidence. His Honour noted that Professor Ingold's first proposition was that geotextile should never have been used and that all joints should have been sealed. At the heart of this theory was that the pores of the geotextile used (indeed of any geotextile that could have been used) were too large. This led to sand loss, the formation of cavities and, by one of two ways, the geotextile coming away from the wall and large quantities of sand then escaping. 63 Compaction was not irrelevant to Professor Ingold's first theory, as the better compacted the sand the greater the resistance to being washed away. He also expressed the view that if there were folds in the geotextile, the loosely compacted sand might settle against the geotextile pressing it against the wall. 64 The terms of Professor Ingold's theory, as applied to the Millstream walls and the Seawall, were carefully set out by the primary judge. Early in his analysis of Professor Ingold's theory, the primary judge rejected one operative element of it - the growth of cavities from behind the centre of the geotextile to the edge of the geotextile. This left as the operative element of the theory the cavity formation causing a pulsation effect on the geotextile, pulling the geotextile away from the wall. 65 The primary judge found Professor Ingold to lack objectivity and his Honour was unwilling to place any trust in his evidence. He rejected Professor Ingold's theory. 66 The primary judge noted the various developments in the views of Professor Ingold, in particular as he was cross-examined. The first such development (made early in his oral evidence) was that the designers should only have chosen the geotextile if "laminar flow" marine conditions were present. Professor Ingold said that only "turbulent flow" conditions were present at the walls. This assertion (not in his reports) required the primary judge (and counsel and witnesses at the trial) to deal with this issue. His Honour did so. The primary judge concluded that the flow was laminar and so rejected Professor Ingold's criticism of the use of the geotextile in this regard. 67 The primary judge also rejected Professor Ingold's view that the formula used by the designers to assess the geotextile was inappropriate. 68 The primary judge then considered what he called Professor Ingold's second development, being his statement, on the last morning of cross-examination by Mr Smith, that cavities would not grow from the joint to the edge of the textile. This was not a development of his hitherto expressed views, but a view contradictory to that which he had hitherto espoused. This then placed all the weight for his theory on the likelihood of the oscillation of the geotextile pulling it away from the wall. This development required the analysis of tests conducted by the Water Research Laboratory of the University of NSW. 69 The primary judge then considered what he called Professor Ingold's third development, his revision of the size of the smallest sand particle. On this revision, a much larger proportion of particles would be lost before the sand loss mechanism based on cavities came into play. The primary judge rejected this view. 70 The primary judge then dealt with Professor Leshchinsky's evidence. His Honour identified Professor Leshchinsky's two theories (or mechanisms, as they came to be called) of sand loss as set out earlier. 71 The primary judge noted that Professor Leshchinsky expressed the view that the two mechanisms (set out earlier) could combine and together be responsible for folding deformation of the geotextile, allowing a path for escape of sand. 72 The primary judge then examined each of these mechanisms. The first mechanism was that the fill behind the walls was not adequately compacted and the geotextile during sand laying and compacting remained against the wall, not displaced by folds or wrinkles. Professor Leshchinsky's view on mechanism one can be seen from [7.9] of his report of 5 September 2005: The combination of the absence of records demonstrating that in the 1m zone immediately behind the sea wall the required density was met, together with the circumstance that the compaction method was not mechanical but hand held, causes me to conclude that the fill (as each 660mm layer was placed) did not meet the required density requirement through the entirety of each layer. I am confident that this is the case.

73 The primary judge recorded the vigorous attack made on this conclusion by BHE, in furtherance of BHE's case that the 80% DI compaction requirement was achieved. 74 It is necessary to understand how the first mechanism was said to operate. In particular, this is so in connection with the central debate on appeal as to whether BHE raised below what it argued on appeal. The following explanation of mechanism one was set out at [728] of the primary judge's reasons: a. Water flows through the joints, past the geotextile and into the sand; b. In the second zone, as the sand is loose, it will become more dense as it becomes submerged in water. This will occur also in the first zone when the temporary dams are first removed, where this sand is also not properly compacted; c. This loose sand, being submerged in water, will compact and settle. One might imagine settlement as being represented by the overall height of the submerged sand decreasing. As the sand compacts, it will exert a force against the geotextile and the wall; d. This increased pressure against the geotextile will cause it to stretch; e. As the tide rises, water flows into the sand; f. As the pressure exerted by the water is acting in a direction opposite to the directions of the pressure which the sand is exerting, the pressure holding the geotextile against the wall is temporarily decreased; g. The pressure, which earlier caused the geotextile to stretch, decreases, while the pressure pushing the geotextile into the body of sand increases. This allows the geotextile to relax by forming a fold; h. Sand is allowed to enter this fold, and as the tide decreases, and water starts flowing out into the sea, sand is carries [sic] through this fold and out into Botany Bay. 75 Thus, once the inadequately compacted sand settles, forces are set up with the entry of water, in particular in the tidal zone 2, causing folds in the geotextile. 76 The primary judge noted that Professor Ingold agreed that compression of sand vertically by tidal forces could induce drag on the geotextile and produce folds, that it was elementary to any engineer that a function of compacting earth is to reduce settlement as much as possible, and that if the sand had not originally been compacted to 80% then tidal forces could lead to the creation of folds. 77 In this context of Professor Leshchinsky's mechanism one, the primary judge then resolved the factual issue about compaction in the hand compaction zone. His Honour found that in the one metre zone behind the walls the sand had not been compacted to 80% DI and that to the extent that there was compaction it had been compacted only to a Density Index which permitted settlement to occur. This finding was based on a number of matters set out by the primary judge at [747] of his reasons which included the findings:

(a) that it would have been "extraordinarily difficult" to achieve 80% DI density by compacting layers of backfill of 660mm with hand-held devices;

(b) that the BHE testing records were not a reliable indication of compaction (implicitly because they were not in relation to the hand compaction zone); and

(c) that the early (satisfactory) tests in the construction of the Millstream walls were an unreliable indication of compaction in later work. 78 The primary judge then analysed Professor Leshchinsky's reasoning in detail, to which it is unnecessary to descend here. In doing so, however, his Honour made a number of factual observations that are relevant to the appeal. In rejecting a line of attack on Professor Leshchinsky in cross-examination by Mr Finch, senior counsel for BHE at the trial and on appeal, the primary judge said that photographic evidence showed that the sand was not placed uniformly before compaction was applied. The primary judge also found that compaction testing did not take place across the vast bulk of the Millstream walls and Seawall in respect of the hand compaction zone. At [758] of his reasons, the primary judge set out a number of facts relevant to this conclusion (at [761]) that Professor Leshchinsky's "first mechanism" had been shown to be one of the mechanisms by which sand loss occurred. It is unnecessary to recount all of these, but the following should be noted: i. in the Millstream the layer lifts in the 1m zone were 375mm … and in the seawall compaction was undertaken in 660mm layer lifts. ii. the Millstream wall involved compacting a 375mm layer lift with a self-propelled compactor and in the case of the seawalls, a 660mm layer lift with the same machine and a jumping jack/whacker in between the buttresses. iii. Compacting a 660mm layer lift to 80% DI is "very difficult" as Mr Moore said …; or "impossible" as Professor Leshchinsky said. Mr Moore said up to 100 passes may be necessary… iv. Each agreed that it would be necessary to undertake a test program to demonstrate that a 660mm layer lift could be compacted with a whacker. So did Mr Scott. v. BHE has no results of testing whether a 660mm layer lift could be compacted in the 1m zone. vi. BHE's engineering change proposal of 13 May 1993 included test results to support the change from compacting 330mm layer lifts to 660mm layer lifts. But none of those were at a point closer to the wall than 1m. Nothing in those tests indicates whether or not the compaction devices used in the 1m zone would achieve an 80% DI on a 660mm layer lift. vii. In terms of contemporaneous records of tests carried out within the first metre from the walls, twenty-five records exist in regard to the Millstream walls (accounting for the first approximate 300 metres of it), and no records in regard to the Seawall. viii. The October 1997 SMEC results which Mr Adler obtained, the BHP ground radar penetration tests which preceded those results and the SMEC tests of 38 holes between the buttresses obtained in September 2000, all point to loose zones of sand behind the walls at depth remote from the vertical joints. All of those reports indicate that a potential cause of that looseness was inadequate compaction when the fill was laid, as a number of witnesses agreed. ix. BHE has called no witness to say that it was possible to compact a 660mm layer lift with the self-propelled compactor or the jumping jack in the case of the buttresses in the seawall to 80% DI. It was open to it to do so if it thought it could establish that the compaction techniques which were adopted in the 1m zone, would result in the fill being compacted by using the self propelled roller in the case of the Millstream wall and the self propelled roller and the jumping jack in the case of the seawalls to achieve the required 80% density. x. BHE itself thought that the sand loss was caused by inadequate compaction (amongst other things). It instructed Mr Peck to proceed on that basis. (emphasis in original) A number of matters need be said about these findings. First, care needs to be exercised in understanding the phrase "layer lift" . The primary judge appears here to have been using the phrase with the meaning of layer of uncompacted backfill, rather than compacted layer. Importantly, the phrase, or a like phrase, was not always so used in the evidence. Secondly, a backfill layer of 660mm was difficult, if not impossible, to compact with hand-held equipment to 80% DI. Thirdly, there was no discussion as to whether the use of a 660mm layer of uncompacted backfill was required as opposed to permitted by a change proposal submitted by BHE in May 1993, and approved by RE P/L and CW, subject to adequate testing. The proposition that it was required was central to BHE's arguments on appeal. The reference in [ix] to "compaction techniques which were adopted in the 1m zone" may tend against the proposition of any compulsory use of layers 660mm of backfill to be compacted in that zone. 79 Having found that mechanism one was one of the mechanisms of sand loss, the primary judge turned to Professor Leshchinsky's second mechanism. This mechanism involved two elements or assumptions: first, that compaction to 80% DI level was achieved; and, secondly, that the process of compaction used to achieve that 80% DI created folds in the geotextile and thus passages for the escape of sand. This second element was not limited to circumstances where 80% DI compaction was achieved. The folds in the geotextile caused during the compaction process could occur during compaction to less than 80% DI. 80 The primary judge found the second mechanism also to be an operative cause of sand loss. 81 At the foundation of this finding was the following encapsulation of evidence at [769] of the primary judge's reasons: As already observed, both Professor Ingold and Professor Leshchinsky state, to varying degrees, that it would have been difficult to compact the sand immediately behind the walls without adversely affecting the geotextile. Mr Boyd gave evidence that the application of compacting effort to a "significant" height of backfill would also likely induce folds and wrinkles. 82 Though his Honour was dealing with the second mechanism, for the purposes of the appeal it is important to understand this evidence, since BHE says it contains the elements of its design case touching mechanism one said to have been overlooked by the primary judge. 83 The primary judge examined Professor Ingold's evidence. It contained, as his Honour described (at [770]), some prevarication and variation. The primary judge at [770] took from Professor Ingold the following: … whilst it would be possible for the inducement of wrinkles and folds to have been avoided, it would have been of the highest order of difficulty to do so; indeed, so difficult that measures additional to those actually adopted, would be required [transcript 1016.4-15]. The "measures additional to those already adopted" last mentioned in this extract were, at the transcript passage cited by the primary judge, the affixation of the geotextile in a manner more comprehensive than spot gluing. Also, during his cross-examination by Mr Smith to the effect that compacting a 660mm layer of backfill to 80% DI with hand-held equipment was very difficult, Professor Ingold stated that the "normal practice" would be to compact backfill as it was shovelled into a narrow zone such as the one metre zone. Professor Ingold said: Q. And if you had men with whackers trying to meet this 80 per cent density index with a 660 millimetre layer lift that is correct was a big ask, would you agree? A. The only thing I would say is that normally in compacting something like that, when the fill is actually brought onto the site what happens is that is then tipped in a pile and then spread quite often with a dozer actually towing the roller. That could not be compacted - that could not be spread in that one metre wide zone. That would have been shovelled across, and usually what happens is the man with the whacker is walking up and down with that all the time the material is being shovelled in. So I don't think it is the case that a 660 millimetre loose layer would have been placed and then the whacker would have been put on top. A man would be pushing that along all the time. Q. Well, you don't know one way or the other, do you?

A. No, I don't, but I am saying is normal practice.

This evidence was given in the context of Mr Smith seeking to show in his cross-examination that compaction of 80% DI was not achieved. He was doing so by pointing out the difficulty of compacting a thick layer of backfill with hand-held equipment. Professor Ingold was answering this saying that normal practice would be to hand compact shallower layers shovelled in. This is precisely what Mr Boyd of RE P/L also said. It lies at the heart of the failure of BHE to exercise reasonable competence as a constructor in carrying out the work. 84 As is discussed elsewhere, the evidence appears to indicate that BHE placed layers of uncompacted backfill in the order of 660mm (at the Seawall) and 375mm (at the Millstream walls) for compaction. The question as to whether this was required by the documents provided to BHE for construction, or whether another technique for hand compaction could have been used was not discussed in the primary judge's reasons. It is central to the resolution of the appeal. 85 Mr Boyd, the managing director of RE P/L, was responsible for the development of the design of the reinforced earth wall structures. The primary judge at [772] of his reasons stated the following about his evidence (in this section of the reasons on the second mechanism): Mr Boyd's evidence is that performing compaction on layer lifts of "significant" height would induce folds and wrinkles [Ex D3(iv) at [59]], however, it was possible to avoid such problems if the constructor had experimented with its techniques and layer lifts [Ex D3(iv) at [61]]. His evidence that experimentation would have been required stems from his stated opinion that "the choice of optimum layer depth is not a theoretical question and requires on-site experimentation and testing" [Ex D3(iv) at [61]]. 86 The references to Ex D3(iv) were to a statement of Mr Boyd dated 9 December 2005. In this statement, Mr Boyd was dealing (in advance) with the matters in the proposed Second Further Amended Summons (about which argument took place, and judgment was handed down, on 15 December 2005). In that statement Mr Boyd dealt with the following:

(a) He recognised that the subcontract required design and specification as to, amongst other things, construction method.

(b) He understood that this encompassed the steps necessary to be taken in the construction and erection of the reinforced earth walls and the manner and sequence of those steps to ensure the structure would perform in accordance with the design.

(c) He recounted what Mr Burns, the BHE project manager with prior experience of reinforced earth walls, had said to the effect that BHE was comfortable with the construction of reinforced earth walls, had experience in such work, and would not require involvement of RE P/L, except to review any documents produced relating to construction.

(d) He recounted what Mr Jean-Louis Betbeder, BHE's project director, told him during the tender process to the effect that he (Mr Betbeder) was aware of the techniques of reinforced earth walls and was involved in the construction of one at Réunion Island (in the Indian Ocean, south west of Mauritius).

(e) He dealt with methods of compaction and achieving the specified density. He said that BHE never asked for assistance about compaction or methods of construction, and that when he asked Mr Ashton how the compaction was going "close to the walls" , he was told "No problem, we are achieving the fill density over the full depth" . He expressed the view that compaction of uniform sand as on this job could be difficult and the appropriate technique involved experimentation and depended on such matters as the nature, weight and method of operation of the equipment, the layer thickness of backfill, and the use of water with mechanical compaction. He said that he thought it obvious that achieving 80% DI would be more difficult in the hand compaction zone than in the zone where the motorised roller was used and that he understood that an appropriate construction technique would be evolved by BHE to "match the depth of the layer carefully with the capacity of the equipment which had been selected to compact it" .

(f) He dealt with the BHE engineering change proposal of May 1993 concerning the proposal to allow backfill to be compacted using a 660mm layer of backfill. In his statement, he noted that in the double layer trial, which referred to a single layer of 660mm being pushed out with a dozer and being compacted with vertical rammers between the buttresses and a "walk behind plate" in the one metre zone, there was nothing said about the construction technique used with this equipment.

(g) He stated that he, as an experienced engineer, understood that applying compaction effort to a "significant height" of backfill directly behind a geotextile could cause it to fold or wrinkle. He expressed the view (as recounted by the primary judge at [772]) that it was a matter for on site experimentation by the constructor to take care in the choice of layers of uncompacted backfill to use with the particular equipment employed close to the panels.

This evidence of Mr Boyd, supported as it was by that of Professor Leshchinsky (and even by the comments of Professor Ingold of "normal practice") clearly placed the choice of the height of uncompacted backfill in the hand compaction zone as a matter for the constructor and its experience and skill, as a matter of construction practice and technique. This ultimately was the primary judge's view. It was his view as to the cause of the failure to compact to 80% DI and as to the cause of folds or wrinkles in the geotextile even if 80% DI was achieved. It was firmly anchored in the evidence. The evidence of both Professor Leshchinsky and Mr Boyd was that compaction of backfill was a common feature of contractors' work. 87 The evidence of Mr Boyd in this regard raised a distinction that was important - being that between construction method and construction technique. His evidence was that "construction method" meant specification of steps necessary to be taken in construction and erection of the walls and the manner and sequence of such steps necessary to ensure that the reinforced earth walls were constructed in the right order so that the structure would perform in accordance with its design. His understanding was that specification of construction method required a statement as to how each element in the construction and erection process had to be incorporated in the right place and in the correct sequence. Construction technique concerned the matters left to the builder to develop and employ using, if necessary, trial and error and testing. It was not appropriate for a designer to pre-empt the constructor in these respects. It depended on equipment and personnel. These were views gained from his experience as an engineer. They lay at the foundation of the primary judge's approach. In circumstances where it was clear that BHE had significant experience in building reinforced earth walls and where it had told RE P/L that it did not need advice as to technique, the placing of responsibility on BHE's inadequate construction technique and lack of testing is to be regarded as firmly anchored in that evidence. In cross-examination, Professor Leshchinsky supported the proposition that some aspects of construction technique were matters for the constructor in empirical testing, not for design. 88 Mr Boyd's statement was directed to the adequacy of the specification for which RE P/L was responsible as to compaction and causing folds, that is Professor Leshchinsky's second mechanism. It contained, implicitly, however, the proposition that BHE was not required to use layers of uncompacted backfill of 660mm near the panels and that good construction practice would have involved experimentation with shallower layers of fill with the hand-held equipment. Together with appropriate testing, such experimentation would have led BHE to being able to compact to 80% DI and without disturbing the geotextile. 89 Mr Boyd was cross-examined immediately prior to the debate on 15 December 2005 about the Second Further Amended Summons, including the debate which then took place about the so-called new case of BHE. It is sufficient for present purposes to observe that Mr Boyd's evidence was directed in part to the inadequacy of BHE's compaction methods in the hand compaction area. 90 On appeal, Mr Boyd's evidence was accepted by BHE, though subject to its argument about characterisation and proper weight being given to the contract and design documents. 91 The primary judge also referred to Professor Leshchinsky's opinion that it would have been difficult to compact to 80% DI without inducing folds and wrinkles. The process of attempting to achieve 80% DI compaction using thick layers of backfill to compact would have produced the folds and wrinkles. It is this opinion which underpins BHE's argument on appeal. 92 As the primary judge said at [773] of his reasons, implicit in the opinions of Professors Leshchinsky and Ingold and in the evidence Mr Boyd was the fact that BHE applied layer lifts (using that phrase to refer to a layer of uncompacted backfill to be compacted) in excess of 150mm. The primary judge said at [773], referring to layers of uncompacted backfill of over 600mm: … The evidence of these witnesses is that layer lifts of such a height are too high if one is attempting to compact the whole height of the lifts to 80% while at the same time maintaining the integrity of the geotextile. 93 The primary judge found (at [774]) that BHE did not use layer lifts (meaning, probably, though it is not entirely clear, uncompacted backfill to be compacted) which at the time were considered too high by any of its advisers (RE P/L or CW) or by industry standards. It is not entirely clear what the primary judge meant by this finding. Whatever its precise content, on a reading of the whole of the judgment, it should not be seen as a qualification of his view that BHE exhibited inadequate construction technique. At [775] and [776] of his reasons, the primary judge made some important findings, in which his Honour's use of the phrase layer lifts continued to be a reference to layers of placed uncompacted backfill. At [775] and [776], he said: Based on what was said in evidence by the experts, it appears that the use of 600mm layer lifts was inappropriate for the task that BHE was required to achieve. And it was the adoption of layer lifts of this height which likely substantially contributed to the induction of the wrinkles and folds in the geotextiles during the construction process. That is to say, the task facing the workers performing the compaction job within the first metre was already difficult enough. Hence the adoption of 600+mm layer lifts only increased the skill and care that would have been required of the workmen. The Court's finding is that it would have been extraordinarily difficult to avoid the inducement of folds and wrinkles without some additional precautionary or preventative measures [transcript 1998.28-53]. Both expert's [sic] readily accepted that it would have been extraordinarily difficult to achieve proper compaction levels without very real difficulties being encountered by way of displacement in the form of folds/wrinkles during the compaction processes. Indeed all of the evidence before the court seemed to me to strongly support these opinions as being correct. (emphasis in original)

The measures referred to in the emphasised portion of this paragraph included using 150mm layers of backfill for compaction. 94 The primary judge then turned to the causal relationship between the two mechanisms. At [779] of his reasons, the primary judge accepted Professor Leshchinsky and found that both the first and second mechanisms caused the sand loss and it was not practicable or possible to attribute a share of responsibility to one or the other. At [781] of his reasons, the primary judge said that the sand loss was occasioned by a combination of both the first and second mechanisms. These are findings adequate for the proposition that each mechanism was responsible for the sand loss sufficient to cause the problems that manifested themselves and that founded SACL's claim against BHE. That is, these findings enable the conclusion that each mechanism was sufficient to give rise to the claim by SACL. The contrary was not asserted on appeal.

Insurance and the causal mechanisms 95 At this point the primary judge turned to the question of insurance. 96 In examining the HIH policy, the primary judge concluded (for the purpose of insuring clause 1) that the activity of carrying out the construction work was not an activity within the field of a consultant engineer. 97 The primary judge introduced the analysis of the operation of the insurance policies by examining how the two found mechanisms of sand loss fell within the operation of the terms of the policies and the arguments of the parties. 98 At [877] of his reasons, the primary judge recorded the following concession of counsel for BHE (Mr Finch) that was important to his Honour: … BHE accepted that if and to the extent that it had been shown to have failed to compact leading to the operation of the first mechanism as a cause of sand loss, to that extent Exclusion 1 (P) would apply. [transcript 2966.40.]

99 On appeal, counsel for BHE (Mr Finch) submitted that the primary judge had misconstrued, or ignored the context of, the concession that he had made. This error led, it was submitted, to the primary judge ignoring a case that was put to him and failing to address the essential case, being that which was put on appeal. 100 In short, (and I will come to it in more detail later in these reasons) BHE submitted on appeal that it had, and has, a fall back position should it be found (as it was) that it had not compacted adequately. That fall back position was that there was a design error in the documents setting out the construction method, being the stated requirement in the design documents for construction to compact uncompacted backfill layers of a depth of 375mm (at the Millstream walls) and 660mm (at the Seawall). In effect, if compaction was not achieved, it was not achieved because the design documents and method statements required a method that was nigh on impossible to achieve (at least without leading to mechanism two). 101 The importance of the concession in the treatment of the case by the primary judge can be seen immediately. His Honour did find that BHE failed to compact to 80% DI leading to operation of the first mechanism as a cause of sand loss. Thus, on the concession as understood by the primary judge, it was only necessary to address exclusion 1(p) to see if it operated in respect of mechanism one being a sufficient independent cause of the sand loss: see [877] of the primary judge's reasons set out above. 102 The primary judge therefore focused on mechanism two, without any further factual consideration of mechanism one, or any fall back argument to mechanism one. In approaching mechanism two, the primary judge, at [883] of his reasons, quoted extensively from BHE's final written submissions. Two aspects of these submissions of BHE should be noted. First, they directed criticism principally, but not only, to RE P/L and CW in respect of design, in an attempt, apparently, to invoke the extended indemnity in special provision 2 of the HIH policy. They direct little attention (albeit some attention) to any asserted failure of BHE itself in respect of design. Secondly, the submissions sought to rely on both Professors Ingold and Leshchinsky. The submissions reflect the tactical accommodation made by BHE in running its case to embrace, as far as possible, the views of Professor Leshchinsky in the propounding of a fall back position. This occurred at a point in the trial when the lack of reliability and acceptability of Professor Ingold and his evidence was becoming apparent to those running the case. Extensive extracts of these submissions were recited over six pages of the primary judge's reasons at [883] of his reasons. Particularly relevant for the argument on appeal are the following extracts from those submissions. (I have interpolated comments to highlight the relevance of particular factors.) (c) Read in context, by the documents described in (b) above RE and/or CW prescribed, recommended, or specified: (i) the use of geotextile as a drainage filter to retain the fine uniform sand backfill behind the reinforced earth walls. (ii) spot gluing of the geotextile to the back of the reinforced earth walls rather than continuous gluing all along the edge of the geotextile; (iii) windrowing of the backfill at an angle to the backface of the reinforced earth wall; (iv) prescription that compaction lift layers of backfill for the Millstream Walls should not exceed 375mm unless instructed by the superintendent; (v) prescription that compaction lift layers of backfill for the Seawall should not exceed 660mm, following Engineering Change Proposal No. 79 dated 11 May 1993, reviewed and approved by RE and CW; (vi) compaction of the backfill to a Density Index of 65% - which subsequently changed to 80%. (The structure of these paragraphs reflects reliance on the primary case as developed from the evidence of Professor Ingold and reliance on a fall back case embracing Professor Leshchinsky's evidence. It is also to be noted that (iv) and (v) were not expressed as requiring layers of backfill of 375mm and 660mm, but prescribing that they not exceed those thicknesses: that is stating that layers of backfill could be placed to that thickness, but no more.) (d) RE and CW: (i) failed to design the reinforced earth walls to retain the fine uniform sand backfill placed behind the reinforced earth walls; (ii) failed to design the reinforced earth walls to be impermeable in the absence of prescribing any or any adequate measures, to prevent the loss of sand around the geotextile; (iii) designed the reinforced earth walls for the Millstream Walls and Seawall to be permeable in the absence of any, or any adequate measures, to prevent the loss of fine uniform sand through the joints; ... (vi) failed to specify as part of the design compaction of the fine uniform sand backfill in layers lower than 150mm in height to reduce the risk of inducing wrinkles, folds or creases into the geotextile; (vii) failed to specify as part of the design a quality control and quality assurance system for compaction by experimentation and trial, to determine the appropriate layer lift heights, to which compaction of the backfill could be achieved to a Density Index of 80%, without inducing wrinkles, folds or creases, into the geotextile (including the placement of sheet metal between the sand and the geotextile during the compaction process); (viii) failed to design the reinforced earth walls against the risk that wrinkles, folds or creases would be induced into the geotextile, likely leading to sand loss around the geotextile, by compaction of the backfill in the absence of any, or any adequate, special measures being developed, by a system of experimentation and trial, to avoid the induction of wrinkles, folds or creases into the geotextile (including the placement of sheet metal between the sand and the geotextile during the compaction process); ... (x) specified that the layer lifts in the reinforced earth block of fine uniform sand backfill should be compacted by hand held equipment in layer lifts in excess of 150mm, inducing wrinkles, folds or creases into the geotextile leading to sand loss; (xi) failed to specify a trial in regard to construction, to determine the appropriate layer lifts of fine uniform sand backfill to be compacted by hand held equipment, to: ... (bb) verify that each geotextile strip was attached free of folds, creases or wrinkles; (cc) after placement and compaction of a backfill layer, excavate to expose geotextile strips at selected locations to ascertain that no folds, wrinkles or creases were apparent; (xii) failed to specify any quality control measures in regard to inspection or testing, to determine whether the geotextile once placed behind the joints in the Millstream Walls and Seawall would be adversely affected by the compaction of the backfill; (xiii) failed to specify any quality assurance test to determine how to place the geotextile and compact the adjacent fill without inducing wrinkles, folds or creases leading to sand loss; (xiv) failed to specify quality assurance (QA) tests and quality control (QC) tests to be conducted during construction to ensure that the construction method selected rendered the geotextile free of wrinkles, folds and creases; (xv) failed to specify quality assurance (QA) tests to ascertain the maximum layer thickness by which the backfill could be compacted without inducing wrinkles, folds and creases in the geotextile; (xvi) failed to specify adequately, or at all, in regard to the compaction of the fine uniform sand fill behind the joints of the reinforced earth walls the measures that should be taken to avoid the introduction of creases, wrinkles or folds into the geotextile; … (xx) failed to specify, in respect to the compactive effort (Density Index) to be applied, or the manner by which compaction of the fine uniform sand fill behind the reinforced earth walls was to be achieved, the precautions which should be taken to avoid the introduction of wrinkles, creases or folds into the geotextile in the absence of gluing the geotextile strip all along its length on either side of the joint; (e) RE and CW failed to provide any warning, advice or instruction to the effect that: … (vi) the design and specification for the Reinforced Earth Walls (including the construction method described in the RE Construction Manual, Technical Specifications, Method Statements and ITP's attached to the Method Statements) was premised on an assumption made by Mr Boyd of Reinforced Earth that BHE knew and understood that compaction of the fine uniform sand backfill in layers exceeding 150mm would likely induce wrinkles, folds or creases into the geotextile, impugning the integrity of the design to retain the fine uniform sand backfill behind the reinforced earth walls; (vii) the design and specification for the Reinforced Earth Walls (including the construction method described in the RE Construction Manual, Technical Specifications, Method Statements and ITP's attached to the Method Statements) was premised on an assumption made by Mr Boyd that BHE knew and understood that there was an inherent risk that compaction of the fine uniform backfill, would induce wrinkles, folds or creases into the geotextile, in the absence of BHE developing a construction method for compaction (not described in the construction method outlined in the RE Construction Manual, Technical Specifications, Method Statements and ITP's attached to the Method Statements) by trial and experimentation to determine the appropriate layer lift heights, to which compaction of the backfill could be achieved to a Density Index of 80% without inducing wrinkles, folds or creases into the geotextile (including the placement of sheet metal between the sand and the geotextile during the compaction process). (Parts of (d) and (e) can be seen to be directed to the case supported by Professor Leshchinsky. It is to be noted that in all this verbiage there is no simple statement that CW or RE P/L or BHE failed to specify 150mm (or some other) uncompacted backfill layer to enable compaction to occur to 80% DI. But, in fairness, a number of these paragraphs came close to such a proposition: see (d)(x) (though directed at RE P/L and CW, and not self-directed to BHE). On one view, the addition of "without inducing wrinkles" is the link that is necessary between mechanism one and mechanism two. These submissions did complain (in (d)(vi), (vii), (x) and (xi)) that layers of backfill lower than 150mm should have been specified; and, indeed (d)(x) complains of a compulsory specification of layers of backfill of greater than 150mm.) 103 After these extensive quotations, the primary judge (at [884] of his reasons) noted that BHE, in "[c]ontinuing its focus on mechanism 2", placed reliance on Professor Leshchinsky's evidence as to the failure of the design documents to deal with methods to avoid folds or wrinkles in the geotextile, including (particularly relevantly for this appeal) his dealing with the fact that: [t]here was no instruction on how to compact without inducing folding.

104 The primary judge then noted (at [885] of his reasons) BHE's submissions based on this evidence. Over two pages of these submissions were set out. They dealt with BHE's assertions as to the inadequacy of one category of documents produced during construction called "Method Statements". The submissions addressed, amongst other things, the absence of any reference to using layers of backfill of less than 150mm to compact in the hand compaction area in any documents prepared by BHE, RE P/L and CW. These submissions of BHE concluded as follows: The documents prepared by RE, CW and by BHE all proceeded on the assumption that the reinforced earth walls would be constructed in accordance with the usual construction method recommended by RE, that is, the standard earthwork construction methods described in the RE Construction Manual and Terre Armee Rules of the Art document, by backfilling and compacting in layers of 330mm for the Seawall (until changed with the approval of RE and CW to 660mm), and 375mm for the Millstream Wall. 105 The primary judge, at [886] of his reasons, summarised BHE's contentions as to the characterisation of the acts, errors or omissions as the subject of insurance under each contract as designing, engineering or construction management. It is clear from the terms of the primary judge's reasons, including the recited submissions of BHE, that these acts, errors and omissions, so characterised, were said by BHE to have been committed by it (BHE), RE P/L and CW.

106 At [911] of his reasons, the primary judge found that BHE had not established any breach of duty, by BHE itself conducting any activity. (This is challenged.) Thus, the primary judge concluded that insuring clause 1 was not engaged. 107 The primary judge then dealt with special provision 1 in a manner not relevant to the appeal. In the context of this clause, the primary judge found (at [925] of his reasons) that BHE was not a designer of the reinforced earth wall and did not commit an act, error or omission in that capacity. This conclusion is challenged on appeal, but reliance was not placed on appeal on this term of the policy as a foundation for indemnity. 108 The primary judge then turned to special provision 2. His Honour examined it and the arguments in relation to it at [928]-[951] of his reasons. He rejected the submission of Gordian and CGU that special provision 2, when engaged, continued to require BHE to have been in breach of its professional duty, albeit that the mechanism or means by which it is found to be in breach is the act, error or omission of the consultant. His Honour accepted BHE's construction of special provision 2 that it was a stand-alone indemnity clause not requiring breach of professional duty by BHE, but in the absence of such breach by BHE, it covered loss, damage or expense arising out of any act, error or omission in the conduct of the defined Professional Activities and Duties committed by a specialist designer or consultant acting on BHE's behalf for whom BHE was responsible. 109 The primary judge concluded, however, that although RE P/L and CW did commit an act, error or omission of a relevant design character, BHE's liability to SACL did not "arise out of" that act, error or omission. So, his Honour held, special provision 2 was not engaged. This conclusion was based on reasoning similar to that which led the primary judge to conclude that exclusion 1(p) applied. There was no issue (being conceded by BHE) that exclusion 1(p) qualified the operation (if engaged) of special provision 2. The primary judge's conclusion about special provision 2 was contained in [955] of his reasons: In my view notwithstanding the finding [repeated below in dealing with Exclusion 1(p)] that RE/CW should have given an instruction requiring adoption of a construction practice [of the type the subject of the findings later set out in these reasons], the finding that BHE, in any event: i. was dealing with an 'on-site' general construction issue requiring trial and error in its construction procedures;

ii. was upon notice of the fact that the geotextile was an integral part of a complex reinforced earthwall design;

iii. ought by reason of i and ii to have been on notice of the importance of taking special care to ensure that the geotextile was carefully and properly placed against the walls in fashion such that sand was not allowed to be caught between the geotextile and the walls, and that the compaction process did not affect the integrity of the geotextile

has a particular present consequence. The consequence is that BHE's liability to SACL did not 'arise' [within the meaning of Special Provision 2] out of RE/CW relevant act, error or omission. This is because BHE's legal liability for claims (and/or its loss, damage or expense as defined in Special Provision 1) did not sufficiently spring out of the relevant act, error or omission of RE/CW. (emphasis in original) 110 This finding should be read together with [972] and [973] of the primary judge's reasons dealing with the act, error or omission of RE P/L and CW, in which he said: [972] Putting to the side for one moment the fact that BHE had accepted liability for design, I accept that as a matter of fact one material act, error or omission was that of RE/CW, namely the failure to specify, with respect to the manner by which compaction of the fine uniform sand fill behind the reinforced walls was to be achieved, the precautionary methods which should be taken to avoid the introduction of wrinkles, creases or folds into the geotextile.

[973] The finding is that in the very special circumstances which obtained, the specification/detail to be found in the design drawings [which simply showed geotextile to be placed up against the Millstream and Seawalls] should have been accompanied by instructions as to the importance of adopting appropriate precautionary methods to avoid the introduction of wrinkles, creases or folds into the geotextile. 111 The primary judge then turned to the central analysis in his reasons - the application of exclusion 1(p), which he saw as central for the following reasons set out in [957] of his reasons: Ultimately whether or not: i. BHE's liability to SACL did or did not 'arise' [within the meaning of Special Provision 2] out of Reinforced Earth's relevant act, error or omission;

ii. Special Provision 2 is or is not correctly characterised as a stand-alone Insuring Clause,

is not determinative of these proceedings. This is because, as was conceded by BHE, Special Provision 2 is subject to the exclusion clauses, and in particular to Exclusion Clause 1 (p). (emphasis in original) 112 Before coming to the question of construction of the terms of exclusion 1(p), it is important to appreciate the nature of the factual findings made by the primary judge insofar as they relate both to the lack of engagement of special provision 2 and exclusion 1(p). They also have a bearing on the asserted engagement of insuring clause 1. These findings, his Honour said at [962] of his reasons, were "pertinent only to the mechanism 2 issue". This was because of the finding as to the operation of mechanism one and of the perceived effect of the concession by senior counsel for BHE to which I have referred. As I have said earlier, the legitimacy of this approach to mechanism one is central to the appeal. It was asserted that the primary judge should have dealt with BHE's fall back case on mechanism one, as well as the fall back position that concerned mechanism two. The primary judge concluded that RE P/L was the responsible designer and that BHE had not acted as designer in respect of the reinforced earth wall. Importantly, the primary judge said the following at [971] of his reasons: There is a deal of evidence that BHE's employees reviewed the design process undertaken by RE/CW, but that evidence falls short of establishing that BHE itself was a designer.

On appeal, BHE challenged this conclusion about its own role. 113 It was in this context that the primary judge made the findings about RE P/L's and CW's design failing, set out above. From [973] of his reasons the primary judge elaborated upon the failure (act, error or omission) of RE P/L and CW, saying at [976]: If one concentrates on the internal division between designer and builder, the party/parties responsible for design should have given an instruction requiring adoption of a construction practice which ensured that the geotextile was so held against the walls prior to the compacting of each layer lift of the sand and the laying of the following layer lift:

• so as to prevent sand from being caught between the geotextile and the walls; and/or

• so as to prevent wrinkles or folds forming in the geotextiles. 114 The primary judge elaborated upon this failure at [978] and [979] of his reasons as follows:

[978] The requisite instruction would have highlighted the need for the above described construction practice and would have emphasised the need for particular caution in relation to: i. the unrolling of the geotextile; ii. the mistreatment of the geotextile [as for example standing on it in a fashion which may lead to a fold]; iii. any combination of mistreatment of the geotextile and/or a practice which would permit sand to be caught between the geotextile and the wall [as for example during the bulldozer operation and/ or by reason of the effect of compaction ].

[979] Without being exhaustive, appropriate instructions [although there may have been other instructions to achieve the same end], may have included the 'moving sheet metal' device explained by Professor Leshchinsky [at transcript 1998.50-1999.18; 2071.52-2072.28]. A requirement for the gluing of the geotextile strip all along its length on either side of the joint may have been another alternative. The spot gluing site initiated practice may or may not have been viable depending upon the strength of the bonding [transcript 1937.30]. (emphasis added) 115 This failure by RE P/L and CW had to be placed in the context that BHE, in any event, was dealing with on-site construction issues, knew that the geotextile was an integral part of the design and ought to have been on notice of the importance of taking special care to ensure that the geotextile was properly placed against the walls and to ensure that the compaction process did not affect the integrity of the geotextile: [980] of the reasons of the primary judge. At [980] to [986] of his reasons, the primary judge placed responsibility on BHE, as constructor, by its failure in its construction technique to achieve 80% DI without disturbing the geotextile. In the context of the evidence, in particular that of Professor Leshchinsky and Mr Boyd, these criticisms of BHE at [981] to [986] of the reasons of the primary judge can be seen as implicitly based on the proposition that the height of layers of uncompacted backfill to be compacted was a matter of construction technique for BHE. At [1022] of his reasons, the primary judge said: The precise poor workmanship of BHE may be identified as the failure to adopt a construction practice which ensured that the geotextile was so held against the Millstream and Seawalls both before as well as after the compacting of each layer lift of the sand:

• so as to prevent sand from being caught between the geotextile and the walls; and/or

• so as to prevent wrinkles or folds forming in the geotextiles. 116 The primary judge concluded at [1012] of his reasons that: In the instant proceedings, the claim made against the Insured is, in context, aptly described as sufficiently springing from the undertaking of construction work of the character identified. It is clear and conceded that if the sand loss was only caused by mechanism 1, the exclusion clause is engaged. 117 His Honour continued at [1019] and [1020] of his reasons: [1019] The finding is that upon the proper construction of the exclusion considered in the context of the whole of the HIH policy, the exclusion is engaged if the construction was a significant cause of the claim made against the Insured in respect of the sand loss.

[1020] The finding [of special relevance to the operation of mechanism 2] is that on the balance of probabilities, a significant cause of the claim made by SACL against BHE, was poor workmanship. That finding underpins the holding that the reach of Exclusion 1 (p) extends to cover the subject claim made by SACL against BHE. The position is a fortiori with respect to the operation of mechanism 1 which simply involves inadequate compaction in the critical areas close to the panels. (emphasis in original) 118 At [1026]-[1033] of his reasons, the primary judge dealt in a little more detail with what he identified as BHE's faults in construction. These paragraphs are important in the understanding of the submissions on appeal and in the resolution of the appeal, because they contain detailed findings about the nature of the failings of BHE in its employed construction technique, in particular insofar as it concerned testing and compaction. The primary judge noted (at [1027] of his reasons) BHE's central contention that it could not be inferred that it was or should have been aware of any danger associated with compaction of layers of backfill of a height greater than 150mm. In response to this, the primary judge made an express finding (at [1028] of his reasons) that it was BHE's task to experiment in order to ensure that, in the course of compaction at whatever layer lifts (using the phrase as referable to layers of uncompacted backfill), the geotextile would not be interfered with by the compaction process. Importantly, the primary judge rejected BHE's submission that it should have received instructions from the relevant "design arm [here, in BHE's submissions on appeal: BHE, RE P/L and CW] of the dangers [in terms of inducing faults into the geotextile] associated with compaction of sand fill in layers in excess of 150 or 300 or 600mm and as to the precise precautions which it should have taken to avoid those dangers" ([1028]). The primary judge found that even without the better instructions from the designers, BHE should have, as part of its role as a constructor, experimented and tested to find, assess and adopt a suitable construction technique including choice of appropriate layers of backfill to achieve 80% DI compaction with the equipment being used and to ensure the geotextile was not interfered with. These were findings plainly based on Professor Leshchinsky's and Mr Boyd's evidence. They can be seen even to be reflected in the views of Professor Ingold. They are findings that go to the heart of the debate about the responses of the policies because they are conclusions as to the relevant causal effect of inadequate construction technique which, when combined with the absence of testing as a required part of the construction method and process, can be seen to engage exclusion 1(p). 119 Importantly for the argument on appeal, these findings can be seen to be made on the premise that the use of any particular layer of uncompacted backfill was not compulsory or mandatory. This is reinforced by the primary judge's discussion (at [1029]-[1030] of his reasons) of Mr Boyd's evidence (as to which see above) that: (a) BHE understood the risks associated with placing backfill behind the panels; (b) that the risk of the geotextile folding was obvious; and (c) that the method of compaction including the thickness of layers of backfill for compaction was for BHE to decide upon depending on method, equipment and testing. 120 Thus, the primary judge found (in respect of the operation of a clause in the HIH policy not relevant on appeal) at [1042] of his reasons that the substance of the claim made by SACL against BHE: …was comprised of the claims pursued in terms of BHE failing to construct the Works and to exercise standards of workmanship in conformity with the provisions of the Contract and in a manner suitable for its purpose and consistent with the nature and character of the Works. 121 This was another way of putting the central conclusion that neither insuring clause 1 nor special provision 2 was engaged; and, in any event exclusion 1(p) would be engaged.

Other aspects of the judgment below 122 The above is sufficient description of the approach of the primary judge in order to determine the appeal. It will be necessary to deal with further aspects of the judgment towards the end of these reasons in dealing with issues that only become relevant should I be wrong in relation to the substance of the appeal.

The argument of BHE on appeal 123 The Further Amended Notice of Appeal (the "Notice of Appeal") raised three primary propositions. It was said that the primary judge erred in holding:

(c) that exclusion 1(p) was engaged. 124 A fourth proposition contained in the Notice of Appeal was that the primary judge erred in holding that clause C of the Gordian policy was not satisfied. This point is only reached if BHE is successful on (a) and (c), (b) and (c) or (a), (b) and (c) above (which in my view it is not). 125 The above three primary propositions carried with them the following challenges to what were called "subsidiary findings". Some of the following is taken from the text of the primary judge's reasons, some is paraphrasing. The numbers refer to the relevant paragraphs in the primary judge's reasons. (i) The activity of carrying out the construction of the reinforced earth walls was not an activity within the field of a consultant engineer: [833]. (ii) BHE had not established that any breach of professional duty giving rise to SACL's claim was committed (ie occurred) in the course of BHE itself conducting one of the professional activities defined in the HIH policy. Its relevant act, error or omission was as contractor, not designer. This is not an act, error or omission in the conduct and execution of the Professional Activities and Duties, as defined. The insured professional activities and duties do not include liability as a contractor, nor liability arising from the supervision of construction techniques: [911]. (iii) BHE has not established that: 1. it was in fact a designer of the reinforced earth walls; 2. in that capacity, it committed an act, error or omission in design which would have made it liable to itself on some legal basis had its contractor and designer capacities been separate legal entities: [925]. (iv) No relevant breach by BHE of any professional duty was proven: [926]. (v) The matters set out in [955] above to the effect that because of BHE's construction faults, the claim did not arise from RE P/L's and CW's acts, errors or omissions. (vi) There was no evidence which established that BHE employed any designers itself, particularly any reinforced soil design engineers, or that BHE itself acted as a designer in respect of the reinforced earth walls: [969]. (vii) There was a deal of evidence that BHE's employees reviewed the design process undertaken by RE/CW and CW, but that evidence falls short of establishing that BHE itself was a designer: [971]. (viii) The findings at [980]-[986] that notwithstanding the errors and omissions of RE P/L and CW, BHE in any event was or ought to have been on notice of the need to take special care to ensure that the geotextile was placed carefully and not interfered with by the compaction process. BHE should have used trial and error in the construction process, including proper testing and including experimentation as to layer lifts (using that term to refer to uncompacted backfill). (ix) Exclusion 1(p) for the purpose of determining the question before the court required a sufficient springing from the matters identified: [1004]. (x) The character of construction work was broad and exclusion 1(p) was to be read as "means employed", "methods employed", "techniques employed", "sequences employed" and "use of equipment": [1005]. (xi) The claim made against BHE was in the context aptly described as sufficiently springing from the undertaking of construction work of the character identified. It was clear and conceded that if the sand loss was only caused by mechanism 1, the exclusion clause was engaged: [1012]. (xii) On the evidence it was clear that if the sand loss was only caused by mechanism 2, the construction work materially contributed to the claim made against BHE. Hence in terms of the finding that the sand loss was contributed to by each mechanism it follows that the exclusion clause is also engaged: [1013]. (xiii) Upon the proper construction of the exclusion 1(p) considered in the context of the whole of the HIH policy, the exclusion is engaged if the construction was a significant cause of the claim made against BHE in respect of the sand loss: [1019]. (xiv) On the balance of probabilities, a significant cause of the claim made by SACL against BHE, was poor workmanship. That finding underpins the holding that the reach of exclusion 1(p) extends to cover the subject claim made by SACL against BHE. The position is a fortiori with respect to the operation of mechanism 1 which simply involved inadequate compaction in the critical areas close to the panels: [1020]. (xv) Although the finding of the court is that the design should have embodied an instruction requiring adoption of construction practice described elsewhere in the judgment, even in the absence of such an instruction, a responsible contractor charged with the task of construction, should and would have been upon notice that the geotextile was an integral part of a complex reinforced earth wall design and of the importance of taking special care to ensure that the geotextile was carefully and properly placed against the walls in fashion such that sand as placed and/or as compacted was not allowed to be caught between the geotextile and the walls. BHE had notice of all these matters: [1021]. (xvi) The precise poor workmanship of BHE may be identified as the failure to adopt a construction practice which ensured that the geotextile was so held against the Millstream wall and Seawalls both before as well as after the compacting of each layer lift of the sand: • so as to prevent sand from being caught between the geotextile and the walls; and/or • so as to prevent wrinkles or folds forming in the geotextiles: [1022]. (xvii) The SACL claim may for the purpose of exclusion 1(p) be seen as arising out of the construction work performed by BHE involving the specified matters, even had the construction conformed with the plans and specifications for the performance of the work: [1024]. (xviii) As Gordian submitted, the policy was not intended to cover the construction process. Whenever a claim arises out of construction work there is an anterior event or cause which leads to the implementation of the construction work. It is irrelevant whether the anterior cause is incompetence, ignorance, malice or bad specialist advice. BHE's submission qua mechanism 2 requires the court to select the anterior event as the cause of the claim and to treat the claim as not arising out of the more proximate cause and implementation of the construction work: [1025]. (xix) The task of the constructor of the walls was to experiment in order to ensure that in the course of its compaction at whatever layer lifts, the geotextile would not be interfered with by the compactive process. It was obvious that the geotextile was an integral part of the construction. The fact that the geotextile would play an important role in ensuring that the retaining walls retained the sand has to have been known to BHE as constructor. BHE's submission that it should have received instructions from the relevant design arm of the dangers (in terms of inducing faults into the geotextile) associated with compaction of sand fill in layers in excess of 150 or 300 or 660 millimetres and as to the precise precautions which it should have taken to avoid those dangers, eschews an acceptance of its (by no means unusual) role as constructor: [1028]. (xx) The apparent acceptance by the primary judge at [1031] of the evidence of Mr Boyd that the rate at which the compaction would occur being subject to the equipment and processes of BHE were matters for BHE as were the number of layers which would effect the timing and costing of the job; and that construction method effectively meant the order of events; and that it was up to the contractor (BHE) to employ whatever method it required to hold the geotextile to the wall while the compaction construction process was undertaken: [1031]. (xxi) The subject construction method used by BHE had shortcomings and clearly involved the means, methods, techniques, sequences, procedures and use of equipment by BHE's construction staff employed in the executing of a phase of the project. The SACL claim made against BHE clearly arose out of the defective construction work performed by BHE: [1033]. (xxii) The substance of the claim made by SACL against BHE was comprised of the claims pursued in terms of BHE failing to construct the Works and to exercise standards of workmanship in conformity with the provisions of the contract and in a manner suitable for its purpose and consistent with the nature and character of the Works: [1042]. 126 The above three general propositions (reflected in grounds 1, 4 and 7 of the Notice of Appeal) and the above particular challenges to the subsidiary findings were addressed in written and oral submissions. The written submissions were detailed and exhaustive. The oral submissions of senior counsel for BHE were precise and confined, though without limiting the text and substance of the written submissions. 127 The substance of the argument has two particular aspects to it, both of which must be made out for it to succeed. 128 The first essential proposition within the argument is that his Honour was in error in failing to recognise the important design role for BHE in the reinforced earth wall construction. This is particularly so, it was said, in relation to the change in the design of the construction from the use of 330mm layer lifts in the Seawall to 660mm. Mr Finch spent some time identifying the obligations of BHE in the contractual documents as reflective of obligations falling within the scope of professional activities and duties in the policy. 129 The second essential proposition within the argument and one that lies at the heart of the criticism by BHE of the primary judge's reasons is that the contract documents, and the design documents and directions prepared and issued under the contract required or mandated adherence to instructions, with which BHE as constructor had to comply, which were fundamentally flawed and incapable of bringing about a structure by way of reinforced earth wall without the problems associated with mechanism one and mechanism two identified by Professor Leshchinsky. It was said that if the design instructions were followed the walls would necessarily fail in the manner that they did fail. This was said to be so because BHE (as constructor) was required to construct the walls following design and construction drawings prepared and reviewed by BHE (as designer) and RE P/L and CW (as designers) which, it was said, required the placement of uncompacted backfill at layers of 375mm (at the Millstream walls) and of 660mm (at the Seawall, after the relevant change) in the hand compaction area behind the panels. Thus, it was said, there was no room to reason, as the primary judge did, that BHE (as constructor) failed to exercise proper construction technique in ascertaining the correct layers of backfill to hand compact near the walls to ensure compaction to the requisite 80% DI, without causing wrinkles or folds. This was so because BHE (as constructor) was required to follow the specification and design and construction documents to place layers of uncompacted backfill of the above identified depths before attempting to hand compact those layers by the hand-held equipment. 130 In other words, because of the mandatory terms of the relevant design and construction documents, even if one accepted Professor Leshchinsky, as the primary judge did, it can be seen that (on this argument) the required design would lead ineluctably to the very problems that manifested themselves. It was wrong, therefore, to criticise BHE as constructor for failing to experiment and choose an appropriate layer height of uncompacted backfill in order to achieve compaction without causing folds or wrinkles in the geotextile (as found by the primary judge and as reflected in Mr Boyd's evidence) and even for failing to test, because BHE was required to follow the relevant documents and hand compact with the relevant equipment using, and only using, layer heights of uncompacted backfill of 375mm in the Millstream walls and (after the change) 660mm in Seawall construction. 131 The criticism of the primary judge underlying this argument was that he failed to deal with the argument that the documents, including in particular the technical specifications and method statements, required the compaction to be carried out as it was. Given that both Professor Leshchinsky and Mr Boyd said that it was very difficult, if not impossible, to compact to the 80% DI without causing folds and wrinkles in the geotextile, BHE said that the design deficiencies were therefore central to the problem that occurred. 132 Thus BHE said that the primary judge failed to address these crucial aspects of the case. 133 The centrality of the asserted mandatory nature of the instructions as to layer height to the argument on appeal was recognised expressly by Mr Finch on a number of occasions in address on the appeal. Though, in fairness to him, he did also argue that success of the appeal would be possible if it were found that there was no mandatory direction as to use of layers of uncompacted backfill of thickness of 375 and 660mm, but that there was a design error in failing to advise that layers of backfill in the hand compaction zone should be no thicker than about 150mm and that exclusion 1(p) operated to exclude claims arising only out of construction flaws and did not operate to exclude claims arising out of both construction flaws and design flaws. I will return to the important application of exclusion 1(p) in due course. 134 The insurers, on the other hand, argued that the case about the mandatory nature of the direction as to layer heights in the design documents was never run at the trial. The insurers accepted that there was a fall back case based on the operation of mechanism two. However, they said that there was no fall back position based on the operation of mechanism one. In this regard, they pointed to the concession made by Mr Finch in relation to mechanism one to which I have already made reference. In other words, the insurers accepted that BHE had a fall back position to the effect that if 80% DI compaction had been achieved, but that the process of compaction had caused wrinkles or folds to occur in the geotextile, a design fault existed in the documents because BHE should have been instructed as to achieving compaction without causing wrinkles or folds. The insurers said, however, that BHE had never run a case that it was required to compact layer heights of uncompacted backfill of 375mm and 660mm in the Millstream walls and Seawall, respectively, and that it was not told in the design documents of the appropriate way to compact at layer heights of uncompacted backfill of no more than 150mm. 135 It is necessary therefore to ascertain what can be legitimately seen to be before the primary judge by way of issues.

Whether BHE's arguments on appeal were run below 136 The consideration of this issue requires an analysis of how the case was pleaded and run. This is no easy task given the size of the case. The matters to which I will make reference are by no means exhaustive. A convenient starting point is the application made by BHE at the conclusion of the evidence to amend the summons. The application was opposed, but allowed: see the primary judge's extempore reasons: [2005] NSWSC 1339. 137 The Further Amended Summons prior to the amendment in December 2005, only pleaded insuring clause 1. There had been pleaded, however, in an amended reply filed by BHE during the hearing, a claim under special provision 2. It was located in a part of the reply which made its terms unresponsive to the relevant iteration of the defence. It might be thought to have been an attempt to tack on to the proceedings an issue to be run, however disconnected its context. The Second Further Amended Summons, brought forward late in the trial in 2005, made a claim under special provision 2. Schedule 3 to the Second Further Amended Summons claimed indemnity under special provision 1 (not now relevant), special provision 2, insuring clause 1, the costs extension (insuring clause 2) and extension 6. The particulars in Schedule 3 referred to BHE's outline of submissions at the commencement of the trial. In those submissions, there was clearly reliance on special provision 2. These submissions reflected, in part, Professor Ingold's evidence, and in part Professor Leshchinsky's; they contained the following: (e) the failure on the part of the designer to identify that the introduction of creases, wrinkles or folds into the geotextile could lead to sand loss through, and around, the geotextile in the absence of gluing the geotextile strip all along its length on either side of the joint. (f) the failure to specify, in respect to the compactive effort (Density Index) to be applied, or the manner by which compaction of the fine uniform sand fill behind the reinforced walls was to be achieved, the precautions which should be taken to avoid the introduction of wrinkles, creases or folds into the geotextile in the absence of gluing the geotextile strip all along its length on either side of the joint. 138 BHE relied on these passages (in particular (f) above) to support the proposition that the argument on appeal was before the primary judge. 139 Further, these submissions set out what was referred to as a "secondary case". This secondary case made references to a number of asserted design flaws and then referred to certain "background matters" which included the following assertion: RE and CW failed to advise BHE that there was a risk of sand loss, through and around the geotextile affixed to the rear of the wall panels of the Millstream Walls and Seawall, by the introduction of folds, creases or wrinkles into the geotextile by: (a) compaction to a density less than 80%. (b) compaction to the specified density of 80% by the use of hand held equipment in the 1 m zone immediately behind the wall panels. 140 None of these references states the matter with the clarity with which it was stated on appeal - that the design mandated the layer heights of uncompacted backfill. 141 Some light is thrown on the debate in the transcript of the amendment application on 15 December 2005. The insurers argued that the design case based on mechanism 2 had not been in the case before the Second Further Amended Summons came forward. The primary judge rejected this. During this debate Mr Robb QC, Senior Counsel for RE P/L and CW said: We would have been presented with the rights that the defendants are entitled to, to actually investigate properly what is involved in the claim that RE had the obligation to tell this immense contracting company BHE as to how it should compact sand. 142 Nevertheless, the primary judge allowed the amendment. No appeal was brought from this. 143 Before looking at what was said in addresses at the conclusion of the hearing, it is useful to note what was put in the various written submissions. I have already referred to parts of BHE's submissions at the commencement of the trial that were picked up in the amendment application in December in Schedule 3 to the Second Further Amended Summons. Also, I have referred to some of the detailed final written submissions set out by the primary judge in his reasons. 144 In BHE's skeleton outline of submissions served after the oral evidence, an outline of argument was given which was said to be based on: … [t]he evidence given by Professor Ingold, Professor Leshchinsky and Mr Boyd as to the cause of sand loss around the geotextile applied to the rear of the joints of the reinforced earth walls, [and which was] to explain how it is that each of the insuring clauses contained within the HIH Contract responds to the common integers of the evidence given by each of those witnesses. In that outline of argument the following propositions were put: Wrinkles, folds or creases were introduced into the geotextile by the compaction of the sand backfill in layer lifts which exceeded 150mm. There is such common ground between each of Professor Ingold, Professor Leshchinsky and Mr Boyd that there is no need for the Court to prefer the evidence of one witness over another - the relevant task is to consider which theory, or which combination of them, best fits with what has been observed about sand loss in regard to the Millstream Wall and Seawall as a matter of fact. Irrespective of whether the Court concludes that sand is lost around the edge of the geotextile consistent with Professor Ingold's theory, or Professor Leshchinsky's theory, or a combination of them both, each of Gordian and CGU is liable to indemnify BHE under the Gordian Contract and CGU Contract respectively. 145 Also, in final written submissions in chief in dealing with the alleged inadequate compaction and exclusion 1(p), a seven staged argument was put, which concluded as follows: Seventhly, even if the Court were to accept that Professor Leshchinsky's first mechanism explained sand loss at one or more locations behind the reinforced earth walls (which locations have never been identified, and which theory should be rejected for the reasons previously stated), the Court would find that his first mechanism was not the proximate cause of the claim made by SACL against BHE, in circumstances where sand loss was inevitable if the reinforced earth walls were constructed in accordance with the design. (emphasis added) 146 In submissions in reply in answer to a claim that there had been a "u-turn" in BHE's case, the following was put in writing: At the commencement of its Submissions, Gordian criticises BHE for a "forensic u-turn". There has been no forensic u-turn on the part of BHE (Gordian's Closing Submissions, at para [4(a)]). BHE's case has always been from the outset that, if the Court accepted Professor Leshchinsky's evidence, the Court would find that there was an act, error or omission in design, engineering or construction management, and that the defence based on Exclusion 1(p) would be rejected: BHE's Outline of Submissions dated 11 October 2005, at paras [127]-[138], [445]-[448], and [506]-508]. … There is ample evidence to prove that BHE constructed the reinforced earth walls in accordance with the design prepared by RE, CW, and supervised and reviewed by BHE. So much is evident from the thousands of documents tendered in the case, as summarised in BHE's Evolution of Design Document. … On the case put forward by Gordian and CGU, it was impossible to achieve compaction of sand to an 80% Density Index with "thick" layers in excess of 150mm using hand held equipment. The design, which included the specification of the construction method, prescribed compaction in thick layers using hand held equipment. It follows that the design was faulty. It can not be said that construction work was a cause of sand loss. Furthermore, on the evidence propounded by CGU and Gordian, compaction to an 80% Density Index would have led to wrinkles, folds and creases in the compacted layers, which in turn would have led to sand loss. Therefore, their case is that compaction to 80% Density Index would have led to more sand loss, not less. In other words, regardless of whether there was compaction to 80% Density Index or not, the sand loss would have eventuated. It was inevitable. It follows that design was the effective cause of loss, not the construction work carried out to construct the walls in accordance with the design. The sand loss does not, therefore, "arise out of" construction. 147 Further illumination comes from final addresses. In final oral address before the primary judge, Mr Finch put the following: And it is important for our case for the Court to appreciate that when read carefully there is no necessary inconsistency between the fundamental elements, as opposed necessarily [to] all the bells and whistles, but the fundamental elements of the theories of Professor Ingold and the second mechanism of Professor Leshchinsky , and reassuringly for the Court, those two approaches, which we say are fundamentally reconcilable are also consistent with the observations of Mr Boyd, both on site and deploying the expertise that he had and the earlier remarks which we have put, as your Honour has seen, in some appendices of Messrs Phillips and Shirley. … [I]t is our position that Professor Leshchinsky's second mechanism , Professor Ingold's theory, Mr Boyd's observations and the observations of Messrs Shirley and Phillips are all of them consistent with all of these following propositions. … Now, pausing there, there appears to be great excitement in the defendants' camp that we should adopt any part of Professor Leshchinsky's evidence, presumably because it came from one of their witnesses. The answer to this excitement of course is simple: if your Honour accepts Professor Leshchinsky's second mechanism , then that is simply evidence of a defective design like any other evidence, it doesn't matter who thought it up, nor is there any forensic U turn involved in appreciating and describing that evidence in a way which suits our case. If new evidence comes up which happens to be squarely in line with our theory we will accept it with gratitude and add it to our case, but we do not abandon other elements of our case, as has been indicated by the summary of the mechanisms of sand loss that I have just gone through. … More importantly perhaps, and in any event, Professor Leshchinsky's second theory is merely one of the particulars in which the design was defective and there are a whole range of these particulars, but it is important, as we have already said to observe there was a more fundamental design defect: that is, when constructed as designed in accordance with the instructions and specifications, the wall did not retain sand. That is the fundamental design defect. It is meant to be a retaining wall. It isn't. It doesn't and didn't retain sand. That is the defect. … I am still marching in the same direction and boringly in some of the same footsteps that I did some months ago. But that fundamental design defect is important to recall because it leads to the more important appreciation that the various experts are really simply attempting to give some reasons for the symptoms that one observes, but the design defect is that the wall doesn't retain sand and it is undoubtedly the fact that all of the witnesses say that the reason it doesn't retain sand is because it has been so designed that sand can and does go around the edge of the geotextile and out the joint. Where one sees the flowering of imagination is in the various mechanisms by which that highway is travelled by the sand. None of them, apart from the suggested lack of compaction mechanism, involve anything other than a design defect. A major part of today's exercise will be devoted to attempting to persuade your Honour why that single mechanism of lack of compaction as a suggested mechanism can't possibly be accepted by the Court. Once that occurs, all other theories are species of design defect. … Because we have shown your Honour that there were objectively observable circumstances in which there was sand loss which was at least associated with areas where folds were observed, there is still room for the question well how did those folds or such folds as may equivalent in whatever numbers they do exist get there? As we have been at pains to emphasise since I started my remarks, it is very important to put that question in its context. That is not the answer to the case. This case is not about the answer to the question how do folds get there. There is only one possible circumstance in which the answer to that question can have any effect upon your Honour's judgment and that is if folds were created caused by a lack of compaction in accordance with the design instructions, that being a matter entirely unattended by any proof, even if your Honour did accept that my questions we say, should come back to haunt all of us, if that is right, that is the compaction either did or could - lack of compaction either did or could result in folding, how much folding was there? How much did that contribute to the loss of sand compared to the places where we have seen there was no folding and loss of sand? (emphasis added) 148 As Mr Robb conceded on appeal, by the time of oral address at the trial, Mr Finch was at least developing (though not without ambiguity and a certain opacity) a holistic way of putting the claim - that whatever the cause of the problem was, it was to be characterised as a design failure. BHE's primary case was, however, that 80% DI compaction was achieved; and Mr Finch's emphasis on the alternative case was on mechanism two. It was said that the above revealed that BHE's case in address was: mechanism one did not happen and mechanism two was a design defect. There is real force in that submission. 149 Later in his address, Mr Finch addressed the primary judge on the basis that a lack of adequate compaction (which of course, was denied) was also a result of a design defect. He said the following on 21 February 2006, in address: Fourthly, it is undoubtedly the fact that the instructions as to what to do in respect of compaction stipulated layer departments in excess of what is now suggested as appropriate . One only needs to recall the cross-examination of Mr Boyd about the repeated instructions to that effect to appreciate that that is so. Even if there were some lack of compaction resulting from layer lifts in excess of 150 millimetres, that results from compliance with the manual, not departure from it, but of course our fundamental point is: Who said that happens?

… The question is why does the fault happen and let's assume for a moment we use Professor Leshchinsky's theory too, why does it happen? It happens because firstly there is Geotex which allows the mechanism to start. Secondly, because folds are induced as an inevitable concomitant of complying with the design instructions of how to build this thing. They didn't just leave it up to us. They said do it in these layer lifts, do it to this level of compaction using these sorts of machines in these areas . (emphasis added) 150 But, on the same day, Mr Finch said: Another aspect of this point six of ours is of course that we accept that it is arguable that if there was a lack of compaction and if there was evidence that such lack of compaction could lead to subsidence of the sort which would induce folds and if there was evidence that such subsidence leading to folds might lead to accelerated sand loss that might be something which fell within the ambit of arising out of construction because, I ask myself why, because you departed from what the design said. The design said compact. If you didn't and it gave rise to that problem then you might be exposed. 151 The following day, Mr Finch continuing with his address, said: Now if, having got them in our hands, we built them badly or wrongly - I am not suggesting negligently, but we simply made an error in building them so that it didn't conform to those instructions, then we would concede that there would be room for 1(p) to operate … 152 Later on this day, when Mr Smith SC (for Gordian) was addressing, he said: Now, Professor Leshchinsky's first mechanism is very important because my learned friend's case concedes that if your Honour finds that there was a cause of sand loss, then his case that he has indemnity must fail because he accepts that there was a failure in the means of construction, ie a failure to compact.

To which Mr Finch responded: That's not quite right, your Honour. Your Honour knows that we say that if and to the extent that there is such a failure, then to that extent an exclusion would apply. 153 This is what was taken by the primary judge as a concession about mechanism one. 154 Finally, Mr Smith in oral address also put the following in an exchange with the primary judge about the engineering change proposal to use layer lifts of 660mm: Well, my point is, stepping back a bit, my learned friend's case is that we were instructed to compact in 660 millimetre layer lifts. My simple point is they were not. They chose to compact in 660 millimetre layer lifts for the obvious reason that, if they could do it with a 10 ton roller, it would as the Scott explanation indicates assist in ease of construction and consequently one would infer reduced time and cost. 155 On appeal Mr Finch, rejected the proposition that there was no fall back position in relation to mechanism one. On appeal he said the following: Right from the start of this case we had a staged approach. The first was as my learned friend Mr Smith correctly said we achieved 80% compaction. And that the design error in question was a failure to specify particular characteristics of the geotex associated with a failure - cutting it short - to seal the wall so that there was in fact no ingress of water and no egress of sand. The fallback position was if that's wrong then there were two parts to our case. First, if Professor Leshchinsky is right then M2 involves a design fault and we endeavoured to persuade the judge that M2 was preferable as an explanation to M1. The final fallback was but even if all of that's wrong then M1 and M2 still involve a design fault. 156 He accepted that the last aspect developed over time and was subject to the primary case that there had been compaction to 80% DI. He also accepted that what he had said to the primary judge was less than clear. 157 All these matters of submissions and pleadings were taking place in the context of the evidence. From as early as 5 September 2005 (about a month before the hearing commenced) Professor Leshchinsky put forward a report which drew attention to the difficulty in compaction using layers of 660mm of uncompacted backfill and recommended 150mm layers. 158 There was a degree of looseness in the language used by Professor Leshchinsky in his evidence. He said that the expression "layer lift" was well known in geotechnical engineering to mean depth or thickness of the layer of fill material once compacted. He did use the expression, however, in his evidence, in particular in cross-examination (often picking up the language of the cross-examiner) to encompass layers of backfill for compaction. As will become evident in due course, the former meaning is contained in various documents such as the RE P/L and CW technical specifications, to which I will come. Taking Professor Leshchinsky's and Mr Boyd's evidence as a whole, it is clear that they were both of the view that the specification and other documentation directed the constructor to the thickness of compacted layer lifts (not layers of backfill to be compacted) and that it was for the constructor using good practice (including experimentation and testing) in the hand compaction zone to achieve this, by, amongst other things, choosing the layer of backfill to achieve compaction. 159 I have set out at length some of the history of the hearing. It is most unfortunate that a long and hard-fought commercial case in the Commercial List should descend on appeal, to an argument about whether a point was run below. 160 Giving due weight to the realities of life in running a long and complex trial and the vicissitudes of the appreciation of the evidence given, it cannot be emphasised too strongly that it is the responsibility of the parties, through their legal representatives, to exercise a degree of co-operation to express the issues for trial before and during the trial. Such co-operation can now be taken as an essential aspect of modern civil procedure in the running of any civil litigation, including hard-fought commercial cases. The need for clarity, precision and openness as part of this co-operation has been emphasised in the context of ambush or surprise: White v Overland [2001] FCA 1333 at [4], expressly approved in Nowlan v Marson Transport Pty Limited [2001] NSWCA 346; 53 NSWLR 116 (Heydon JA, with whom Mason P and Young CJ in Eq agreed); Glover v Australian Ultra Concrete Floors Pty Limited [2003] NSWCA 80 at [59]-[60] (Ipp JA, with whom Sheller and Hodgson JJA agreed); Sutton v Erect Safe Scaffolding (Aust) Pty Ltd [2006] NSWCA 265 at [4] (Bryson JA with whom Basten JA agreed); and Hooker v Gilling [2007] NSWCA 99 at [52] (McColl JA, with whom Ipp and Basten JJA agreed). 161 The need for clarity, precision and openness in the conduct of litigation and the responsibility of parties and their legal representatives therefor flows most clearly from the statutory duty of a party and his, her or its legal representatives in civil proceedings to assist the court to further the overriding purpose to facilitate the just, quick and cheap resolution of the real issues in dispute and to participate in the processes of the Court to that end: see Civil Procedure Act 2005 (NSW), 56. It may be that the provision no more than restates the proper approach of the modern law of procedure reflected in cases such as Nowlan v Marson Transport. It places the proper approach, however, on a firm statutory foundation. These principles can be seen to be reflected in the longstanding rules of pleading requiring any matter that may cause surprise to be pleaded. 162 An enhanced requirement of clarity and disclosure in modern civil litigation can be seen in Australia and England from at least the early 1990s: see the discussion of the "cards on the table" approach by Ipp J (as his Honour then was) in Boyes v Colins [2000] WASCA 344; 23 WAR 123 (with whom Pidgeon and Wallwork JJ agreed), citing Mercer v Chief Constable of the Lancashire Constabulary [1991] 1 WLR 367 and Khan v Armaguard Ltd [1994] 1 WLR 1204. Indeed, from the late 1970s and early 1980s, the Commercial List of this Court (in which List this hearing took place) has been sought to be run on the strict basis of the clear and full enunciation of issues for trial, in a way that has always demanded the fullest co-operation among parties and legal practitioners to delineate and illuminate the real issues in dispute. 163 The clear statutory duty to assist the Court, and, in a practical way, to co-operate to bring forward the real issues in dispute, encompasses the requirement to be clear and precise in the illumination of the issues for trial. The occasion for this is not merely pleading (using the word broadly to encompass the modern commercial list summons and defence), it extends to all aspects of the engagement in the Court's processes. For similar responsibilities in the conduct of references, see Bellevarde Constructions Pty Limited v CPC Energy Pty Limited [2008] NSWCA 228 at [55]-[56]. 164 This does not deny the possibility, as occurs in real life in litigation, that issues will develop. Litigation is a dynamic human activity. Changes in how a case is put can be expected. This often occurs in large commercial cases. Such change, and the potential for it, makes it, however, all the more important that legal practitioners and parties ensure that the clear enunciation of issues keeps pace with that growth and change. This responsibility will encompass parties and their legal representatives making clear what is being put and also what they regard as not legitimately part of the controversy, if it is apparent to them that an issue not pleaded or presented is being relied on. 165 Here, the argument put on appeal did come forward at the trial, but only gradually. By the time of final addresses, there was still ambiguity, but the final submissions were wide enough to be understood as encapsulating the argument that was put on appeal. The submissions recounted by the primary judge in his reasons reflect this. 166 That said, the fact that the primary judge did not appreciate the width of the submission can be laid at the feet of BHE. There was an ambiguity and opacity, which permitted the judge to take the course that he did. Indeed that ambiguity was founded on the fact that, at times, Mr Finch expressed himself in a manner well able to be understood in a way reflected by his Honour's approach. 167 In the end, however, I am prepared to permit the argument as presented on appeal to be run as sufficiently encompassed by the way the case was presented below. Having said that, I would only add that it is perfectly understandable that the primary judge dealt with the issues as he did, given the ambiguity and opacity of the presentation of BHE's case. 168 It should not be taken that my view is one which can be taken as a statement of principle that ambiguous and opaque presentation of issues below will always permit argument on appeal on a point the subject of such ambiguity or opacity. The judgment will be a factual one in each case dependent significantly on assessing the fairness of permitting the issue to be pressed on appeal. 169 Here, in the light of Professor Leshchinsky's and Mr Boyd's evidence, and the terms of the submissions and addresses below, I think the arguments pressed on appeal were in the arena below, albeit in the shadow of BHE's primary case that it compacted to 80% DI. Further, it is common ground that the fall back position to mechanism two was before the primary judge. As can be seen from BHE's submissions below, the place of the requirement of a layer of backfill of no greater than 150mm to avoid folds or wrinkles provides an intertwined conceptual link between mechanisms one and two: the failure of the design to direct the compaction process in a way that would bring about 80% DI compaction without folds and wrinkles. 170 It is therefore necessary to address the legitimacy of the arguments that his Honour did not address. The assessment of BHE's argument on appeal 171 Working on the basis that the documents before the primary judge and the conduct of the case before him was such as to permit BHE to argue the appeal in the way it did, it is necessary to assess the validity of that argument. 172 I will deal with the two essential propositions of BHE's argument referred to at [127] to [130] above together. This is convenient because, whilst I think that the primary judge did err in his conclusions that BHE was not a designer of the reinforced earth walls, the relevant documentation did not require BHE as constructor to compact only after specified layers of uncompacted backfill had been placed. Further, it is helpful, as Mr Finch did in address, to examine the relevant documents from the more general to the more specific, with an eye to both the question as to the responsibility for design and the question as to the direction given as to compaction by individual documents. 173 It is also appropriate to examine the relevant contractual documents, not to give content to the words of the insurance policy (their contextual relevance being doubtful, and not pressed), but in order to understand the respective responsibilities of the parties (including BHE) which can be assessed by reference to the terms of the policies.

The contractual documents - the head contract 174 The contract documents were: the Formal Instrument of Agreement ("FIA"), the General Conditions of Contract ("GCC"), the Special Conditions of Contract ("SCC"), the Specification and the Drawings. 175 The contractual documents clearly identified an obligation of BHE to design (as well as to execute) the works (including the reinforced earth walls). The FIA recited the requirement of design in recital A, recited BHE's undertaking of design responsibilities in recital K; and clause 3 of the FIA contained an express obligation to "design the works in accordance with the Project Brief". 176 The GCC contained a clause on design (cl 8A). BHE was required to engage independent consultants in relation to design: cl 8A.1. BHE was responsible for adequacy of design "and the specifying of materials and methods for the construction of the Works": cl 8A.2. BHE was expressly responsible for the exercise of skill and judgment in design (by itself and others): cl 8A.3. 177 The SCC contained an obligation upon BHE (SC 4) to compile a "method statement" as part of BHE's "Quality Plan" for the information of the Principal's (FAC's, later SACL's) Representative "which fully details how the Work under the Contract is to be constructed". Such documents were also called for in cl E3.2.2 in Section E on "Marine Works" in the Specification. 178 The Specification reinforced the responsibility of BHE for all aspects of design in that in Section A, dealing with the scope of works, it made clear the assumption of responsibility by BHE for the pre-tender design work carried out by FAC and its consultants: cl A1. 179 The Specification also identified the "Permanent Works" which included the Seawall "in accordance with the details shown on the Drawings" and the Millstream walls: cl A2. 180 The Specification contained a specific undertaking by BHE for design of all relevant work: cl A4. Clause A4.1 expressed the extent of "design services" by BHE as including "investigations, designing, construction phase services and commissioning". Such phrases cannot convert activity into design activity for the purposes of the insurance policies if they do not otherwise amount to design. They do, nevertheless, reveal the intended width of the responsibilities of BHE under the contract expressed in terms of design. 181 Importantly for BHE's obligations as a constructor, BHE was required to set up and maintain an appropriate regime for inspection and testing of its work: see Specification, section B. The findings of the primary judge (not challenged on appeal) were that BHE failed comprehensively to test in the hand compaction area. This was in no way a failure of design. It was a failure to comply with a contractual regime for testing as part of the construction of the works. 182 Section D of the Specification included provisions about the design of the "Permanent Works", which included the Seawall and the Millstream walls. This section of the Specification concerned the "design services" for the "design development, detail design and documentation, and construction phases of the permanent works": cl D1. 183 The scope of the design work to be provided by BHE was set out in cl D2 - "all engineering services needed for the purpose of investigating, designing, constructing and commissioning" the runway. The phrase "engineering services" was stated to include: • "[p]repare final and complete working drawings and specifications suitable for construction…" • "[a]ttend design review meetings with the Principal…" • "[u]ndertake all survey and site investigation necessary to design the permanent works…"

In argument on the appeal, Mr Finch stressed these provisions. He submitted that they provided for the very kind of investigative work that Professor Leshchinsky and Mr Boyd said should have been done in developing the construction technique suitable for the 80% DI compaction in the hand compaction area. This proposition faces the difficulty, however, of the findings, amply supported by the evidence, that BHE failed comprehensively as constructor to adopt any effective testing regime and to adopt a construction practice after experimentation and testing that would bring about compaction to the required density, without disturbing the geotextile. BHE's necessary answer to that depends upon the mandatory character of any direction about layers of uncompacted backfill of 375 or 660mm. Even accepting, it was argued, a failure of construction practice to observe the problem, only a design change to alter the required backfill heights could lead to the problem being cured. Thus, it was said, it was always effectively a design issue, any construction failure being antecedent to a further design step. 184 All design work by BHE was to be undertaken by qualified designers and consultants: cl D4.1. Clause D4.1.5 required BHE and its "Design Consultants" (here RE P/L) to prepare a "Quality Plan" for design in accordance with Section B of the Specification. BHE was obliged to prepare a "Preliminary Design Report", including construction methodology (cl D4.1.6) and a "Detail Design Phase" report (cl D4.1.7). The latter required BHE to produce "complete designs, … drawings, specifications and other supporting efforts to the degree of detail that competent sub-contractors can clearly determine … construction work requirements". Thus, it was said BHE had to produce documentation ensuring that the appropriate technique for construction was made clear. 185 Section E of the Specification dealt with "Marine Works" of which the reinforced earth walls formed part: section E4. The work was identified as including the Seawall and Millstream walls. One of the "essential elements" identified was the construction of the reinforced earth Seawall "as shown on the Drawings": cl E1.3B(m). Clause E2.2 dealt with "Design Parameters" for Marine Works and included the following: Detailed specifications covering the design and construction of elements of the seawall … and Mill Stream channel diversion works are contained in other clauses and sections of this Specification. Where those clauses and sections do not adequately cover the design and construction of the works proposed by the Contractor … additional specifications shall be provided by the Contractor …

This last clause was also stressed by BHE. It identified, BHE submitted, the very area where BHE, RE P/L and CW failed - in producing detailed specification for compaction construction procedures. 186 Clause E3.2.2 required BHE to provide "Method Statements" being "a description of the method [it] proposes to use to execute the work". It is at least arguable that such an obligation can be seen not so much as design, but as an incident of construction. 187 Section E4 dealt with reinforced earth walls. Clause E4.1 obliged BHE to construct the reinforced earth walls "using the technique developed by [RE P/L]". BHE was obliged to liaise with RE P/L to provide complete design details. Importantly, cl E4.1 also obliged BHE to revise construction methods urgently in circumstances where the design calculations may not be conservative. Again, this can be seen as much an obligation incidental to construction as design. 188 Clause E4.8 dealt with placement of fill. It was couched in general terms that included the following: Placement of the reinforcing strips shall be in accordance with the Drawings. The RE fill material shall be deposited, levelled and compacted in layers appropriate to the reinforcing strip spacings and fill type. Compaction shall be such that each reinforcing strip can be placed on the compacted fill without the formation of voids below the strip. Compaction shall be by plan appropriate to the reinforcing strip spacings and layer thickness. RE fill shall be compacted to a minimum of 80% Density Index (AS 1289) when tested at a level 150mm below the surface of the completed layer. Testing shall be tested in accordance with Clause E4.9. Compaction shall be carried out in a direction parallel to the wall face and shall be staged to follow the erection of the facing units. Compaction shall be undertaken in such a manner so as not to damage the reinforcing strips or facing units. Plant shall not be operated over reinforced strips having less than 100mm of sand cover in position.

No specific detail was directed to the hand compaction zone. The layers of backfill are, however, required to be "appropriate" to fill type. No level of uncompacted backfill for the compaction process is here mandated. 189 Clause E4.9 dealt with testing. This clause set out BHE's obligations in this regard and included the following: In addition, the following tests shall be carried out for each 'lot' of each layer of fill compacted: (i) In situ density AS1289 - E3.1, E3.2 (ii) Density Index AS1289 - E6.1 Where the test results indicate the density index to be less than 80% or the specified grading is not obtained, the layer of fill shall be removed or recompacted as indicated by the test result.

These obligations were upon BHE as a constructor. It failed, comprehensively, to comply with these obligations in the hand compaction zone. If it had undertaken such testing, in accordance with its obligations in relation to construction, it would have been aware that it was failing to compact to 80% DI in the hand compaction zone. 190 Section E9 dealt with the geotextile fabric. The placement of the geotextile and its relationship with the backfill was expressed in very general terms, including the following: The overlaying material shall be placed so as not to damage the fabric. If the fabric has been damaged by the placing of the overlaying material, the Contractor shall remove as much of the overlaying material as necessary, so that the fabric can be replaced. 191 It is clear from the above provisions of the head contract that BHE undertook extensive obligations concerning design and documentation of design, including in relation to the reinforced earth wall, as well as obligations on construction of the works.

The contractual documents - the subcontracts 192 The subcontracts between BHE and RE P/L concerning engineering services for the reinforced earth walls, mirrored the head contract. RE P/L can be seen to have undertaken responsibility for the development of the design details and of the technique for construction. For instance, in its subcontract with BHE in respect of the Seawall, RE P/L's scope of work concerning design drawings and specifications, included the design and specification of: - handling and erection system - backfill within the Reinforced Earth block - construction method - wall construction including structure geometry, wall layout and element distribution 193 These obligations, however, are to be understood and given content and context by the evidence of Mr Boyd about BHE's self-expressed skill and expertise. Whilst not dissolving RE P/L's obligations, the nature, extent and detail of documentation dealing with construction technique would need to be assessed in the light of BHE's skill and expertise as a constructor of reinforced earth walls in a marine environment. This is of particular importance because of clause 6 of the first schedule to the Seawall subcontract and clause 7 of the Millstream subcontract with RE in which the following was stated: [BHE] shall supervise and perform the installation and erection [of the walls] on-site in accordance with [RE P/L's] Working Design Drawings and Specifications. [RE] shall make available at no cost to [BHE] , advisory personnel to consult with [BHE] on erection and construction techniques… 194 Two things are to be noted. First, there is a difference in terminology in the contract documents between construction method and technique. This distinction was reflected in Mr Boyd's evidence and, in particular, his view that it was for the contractor to experiment and test to choose a satisfactory technique that employed the appropriate layer of backfill for compaction. Secondly, here, in connection with construction technique, RE's responsibility was to consult, implicitly, if called for. Its consultation was not called for. 195 In its subcontract with BHE, CW was responsible for all engineering over the whole contract. Its obligations included detailed design and documentation, in the following terms: Undertake detailed design and documentation suitable for the construction of the works, excluding workshop drawings, cable schedules and reinforcing bar bending schedules. 196 With this undoubted design responsibility of BHE, RE P/L and CW from the contract documentation, and recalling the context of the undertaking of those obligations set out in Mr Boyd's evidence, it is necessary to examine how the work was implemented, including, in particular, examining the documents against or pursuant to which the compaction was undertaken, and especially the change to the construction method to the use of layers of 660mm. 197 The evidence disclosed that BHE personnel attended design meetings with personnel from FAC, CW and RE P/L. 198 In August 1992, BHE's Design Manager prepared a document setting out the detailed design phases of the work. The aim of the document was expressed as follows: This document will describe the proposed limits of the phases of the design process for the Parallel Runway Project. The design phases correspond with milestones or hold points defined in the Contract and incorporated into the Quality Assurance Plan for the project. Approval of each milestone in the Design process shall be by the Principal's Assistant (Design).

A "30% Technical Review" was carried out for the following purpose: [T]o consider the extent of proposed documentation for the design element. This review shall follow one week after the issue of the documents required to detail the design element development to this hold point. The documents for this technical review shall describe the design element sufficiently to authenticate the suitability of the "concept".

A "70% Technical Review" was identified and then a "100% Technical Review". The purpose of this latter review was described as follows: The purpose of this review is to carry out an Audit of the final design element documentation. The documentations for this review shall constitute the complete specification and signed final design drawings together with any construction briefing notes and construction hand-over report, prepared to provide constructors with the Designer's assumptions, restrictions on construction techniques, and areas of potential technical risk ie: foundation bearing and stability.

Mr Finch submitted that one of the "Designer's restrictions" on the constructor was the required or mandated level of uncompacted backfill to 375mm and 660mm, in the Millstream walls and Seawall, respectively. None of the documents thus far examined, however, makes good the assertion of a required or mandated level of uncompacted backfill in the hand compaction area. It is now necessary to examine further documents that are said to make good such a proposition.

The Rules of the Art 199 The first document stressed by BHE was the "Recommendations and Rules of the Art" prepared by the French Ministry of Transport in 1979, in English translation. The document gave the history of the development of the technique and a description of the technique and construction process. It described the prefabricated elements as follows: Except of the backfill material which represents the most important part in volume, the process of the Reinforced Earth uses entirely prefabricated elements. This enables an easy and a rapid execution of the structures without the necessity of specialized workmanship.

It described the construction process as earlier set out. A general sequence can be seen, but no instructions as to method of compaction. Later in the document, dealing with backfilling, reference is made to "adjusting" backfill following "unloading", the following appears: According to the type of the facing, the adjusting should be realized by layers of: 0,33m - average thickness - in the case of metallic facing, 0,375m - average thickness - in the case of concrete facing.

Layers of 330mm and 375mm are referred to. Shortly thereafter, the topic of compaction is dealt with. At this point in the document, a clear distinction is made between the areas close to, and those not close to, the facing: The compaction operation in a reinforced earth structure has essentially for [sic] an objective to prevent any ulterior settlement of the material. This objective should be really achieved for structures supporting a super structure (wall under road, bridge abutment, building foundation, etc …). For some particular retaining structures, the methodical compaction does not constitute an imperative requirement. Actually, on one hand, the materials placed are generally easy to compact and on the other hand the thickness of the layers is necessarily limited to the relatively small values previously defined, thus the sufficient compaction in these structures is obtained by the circulation of the transport engines and of the adjusting engines. The structures supporting a superstructure must be methodically compacted, separating the zone adjacent to the facing, 1 to 1.5m wide, and the main portion of the structure.

The importance of the different type of equipment used for compaction in these areas is then discussed: The previous rules cannot be applied because the use of the compactors considered in the Recommendations for Road Earthworks can lead to deformation of the facing. It is therefore necessary to impose for the compaction of this zone lighter vibrating compactors, the weight of which per one centimetre of the generator is included in the range of 60 N to 80 N.

The reference to "N" is to Newtons. 200 Thus, although the Rules of the Art identify the importance of compaction and the methods of compaction, distinguishing between compaction close and not close to the panels, and although some description of layers is given, there was no instruction as to the effective methods of compaction in the hand compaction zone and no mandating of any required level of uncompacted backfill.

A video produced by RE P/L entitled "Construction of Reinforced Earth Walls with Concrete Facings" was provided to BHE. It reveals the use of motorised rollers, but does not deal with compaction in the hand compaction zone.

The RE P/L Construction Manual 201 This was the first site-specific document relied upon by BHE. It was expressed to be a "guide for building Reinforced Earth structures". In its introduction, a general description was given of the structure which was set out earlier. A description of the equipment to be provided by the contractor is given: • Large, smooth-drum, vibratory roller or other large compactor for mass compaction. • Small manual vibrating roller (walk-behind type) or flat plate compactor not exceeding 1 tonne static weight for compaction within 1500mm of facing panels.

A basic construction procedure was identified. This included statements such as "spread and compact backfill" . These can be seen as descriptive of a body of activity and not, as often stressed in the submissions of BHE, as a strictly sequential direction. Put another way, the fact that backfill placement and compaction are stated to occur one after the other (as in a sense is obvious) does not mean that no compaction is to begin in the hand compaction zone until all backfill to be compacted has been placed. The need for hand compaction equipment is made clear in such passages as: Step B-9 : Begin backfilling. Place approved backfill up to the bottom row of panel tie strips (figure 21a). Backfill can be compacted to within 1500mm of the panels by a large vibratory roller. A small hand operated compactor must be used within 1500 mm of the panels to avoid undue panel movement. 202 Also, in reading these documents, one must bear in mind that Professor Leshchinsky gave evidence that the expression "layer lift" had a well-known meaning of layer thickness once compacted. This was not challenged on appeal. Thus, expressions such as "lifts of backfill" should be understood, unless the context clearly demonstrates otherwise, as lifts of compacted backfill. One must also bear in mind that these are not legal documents, but documents to be read by builders and their subcontractors. In this respect, it is important to recall that because this issue was not at the forefront of BHE's case at the trial, there was no evidence that these documents should be read in one way or the other; though, implicitly, Mr Boyd's evidence was to the effect that they did not mandate any layer heights of uncompacted backfill. The ascertainment of those heights of uncompacted backfill for compaction was for BHE to determine in the hand compaction zone by good construction techniques, including experimentation and testing. The absence of any evidence contradicting what is implicit in Mr Boyd's evidence is telling against BHE's argument as to the mandatory character of the documents as to heights of layers of uncompacted backfill. 203 I do not accept the submissions of BHE that the introductory parts of the construction manual lay out a clear sequential order instructing the laying of all uncompacted backfill in the hand compaction zones before compacting with hand held equipment. Indeed, there is discussion of not compacting close to the wall until connecting the first layer of strips. It is not entirely clear what the importance of this is, but it is not consistent with the notion at the heart of BHE's submission that the documents tell the constructor that it must place 375mm or 660mm of backfill against the panel before commencing any hand compaction. 204 Also, in places, the term "backfill" is used in a way consistent with compacted backfill. 205 I reject the foundational proposition in BHE's argument on appeal that the RE P/L Construction Manual tells one to place a full layer of (uncompacted) backfill before any hand compaction commences.

RE P/L's Design Proposals of March 1992 206 These documents were provided to BHE for both the Seawall and the Millstream walls. In the proposal concerned with the Seawall, the following appeared: Place the sand backfill behind the panel to the level of the first reinforcement connection. Compact to achieve minimum 65% Density Index. … Place and compact sand backfill to the next reinforcing strip layer. Compact to achieve minimum 65% Density Index. … Place and compact the sand backfill to the next reinforcing strip layer. Compact to achieve minimum 65% Density Index. 207 BHE sought to argue that these were an instruction to place uncompacted backfill to the relevant full layer height before beginning hand compaction. I reject that. I do not understand these general instructions, in the form of a composite instruction, as providing any such mandatory direction. 208 Mr Caretti gave evidence that they adopted the construction techniques in the RE P/L Construction Manual. He did not, however, give evidence (nor did anyone else) that he thought that he was required to place uncompacted backfill in the hand compaction zone to a certain height before compaction could begin. Mr Finch referred to evidence which indicated that compaction in the hand compaction zone did in fact take place thus, and so asked it to be inferred that those responsible thought they had to do so. He referred to the evidence of Mr Caretti, Mr Ashton and Mr Scott. None of these witnesses said that he thought the documents told of a compulsory thickness of uncompacted backfill material. The witnesses who could say so were in BHE's camp. No evidence of that kind was called. No inference should be drawn: Commercial Union Assurance Co of Australia v Ferrcom Pty Ltd (1991) 22 NSWLR 389 at 418-419. The fact that layers of uncompacted backfill of those heights were placed for compaction was consistent with BHE thinking they could (as opposed to must) do so, with BHE not testing and thus not becoming aware of an absence of compaction and with BHE failing to attend to proper construction technique. This latter explanation was in accordance, not only with the evidence, but also with the clear findings of the primary judge. 209 One other feature of the evidence making it very difficult to infer that the BHE workmen used 375 and 660 mm layers of uncompacted backfill for the compaction process because they thought that the documents required them to do so is the evidence of Mr Moore. Mr Moore was a BHE employee who carried out an internal investigation. He interviewed many people on the site who had been engaged in compaction work. Not only were those people not called, but also Mr Moore gave no explanation to those to whom he was giving advice in the investigation that those on the job thought that they had to use these layer heights of uncompacted backfill.

The Quality Plans 210 Quality Plans were required by cl D4.1.5 of the Specification to be prepared by BHE and RE P/L. In November 1992, RE P/L produced a "Quality Plan". This set out the respective design responsibilities of BHE and RE P/L and stated RE P/L's responsibility for drawings as follows: The Reinforced Earth Company shall be responsible for the production of working drawings, shop drawings and specifications as required to fully define the manufacture, supply and construction of all elements of the REINFORCED EARTH structure.

The Quality Plan stated that RE P/L would produce a Technical Specification, the purpose of which was described as: [t]o describe the manufacture and supply of the elements of the Reinforced Earth structure and the construction process.

Finally, the Quality Plan gave RE P/L responsibility for the "constructability" of the walls: RE shall verify that the Reinforced Earth structure is constructible in accordance with the construction procedures nominated for this project. The RE P/L Technical Specifications 211 In November 1992, RE P/L produced, in accordance with its obligations under the Quality Plan, two technical specifications - one for the Seawall and one for the Millstream walls. In the Millstream walls technical specification, the contractor was told to follow the erection procedure closely. Under the heading "Select Fill Placement" the following appeared: It is the intention of this specification that select fill of the same type and from the same source shall be used in the REINFORCED EARTH BLOCK and that the same degree of compaction is achieved throughout the REINFORCED EARTH BLOCK including the 1.5 metre wide zone immediately to the rear of the facing panels. Select fill placement shall closely follow the erection of each course of facing panels. The select fill is not to be placed in layers exceeding 375 millimetre thickness without the written approval of the Superintendent. Select Fill is to be placed and spread in a direction parallel to the exposed face of the structure. The path of plant including compaction plant shall also generally be parallel to the exposed face of the structure. At each level of reinforcement the fill shall be compacted and levelled prior to the placing and fixing of the REhas Reinforcing Strips. The level of the surface of the compacted select fill layer shall be equal to or higher than that of the tie points case into the facing panels. Unless otherwise shown on the drawings, or directed by the Superintendent, the RE has [sic] Reinforcing Strips shall be placed normal to the face of the structure. Plant with an equivalent static load in excess of one tonne shall be excluded from a zone extending for 1.5 metres from the rear of the facing panels at all stages of the construction of the REINFORCED EARTH Structures. Compaction of the Select Fill in this zone shall be achieved using hand operated plant with an equivalent static weight of less than one tonne. Care shall be taken to ensure that the density of compacted select fill in this zone is equivalent to that achieved throughout the select block. The select fill shall be compacted to achieve a minimum Density index of 80% as required by the project specification. A uniform density shall be achieved throughout the select block. (emphasis added)

("REhas Reinforcing Strips" was the proprietary name of the reinforcing strips used by RE P/L.)

Five things are to be noted about these instructions. First, the hand compaction zone is said to be 1.5 metres. The primary judge found that it was 1 metre. No challenge is made to this finding. Secondly, hand compaction is clearly to be dealt with differently. Thirdly, the 375mm is not required, but is not to be exceeded. Fourthly, it is not clear whether this is a 375mm compacted or uncompacted layer. Fifthly, there is no advice or instruction not to attempt to hand compact with layers of uncompacted backfill of greater than 150mm. 212 The Seawall technical specification was to similar effect.

CW Technical Specifications 213 Two months after the production of RE P/L's technical specification, in January 1993, CW produced its "Detailed Design Report" for the Millstream walls. It contained a technical specification for the reinforced earth walls. It contained a section entitled "Select Fill Placement" in very similar terms to the sections in the RE P/L technical specifications under the same heading set out above. One of those paragraphs is worth repeating because there is the insertion of a word which clarifies the meaning of the paragraph in an important respect: Select fill placement shall closely follow the erection of each course of facing panels. The select fill is not to be placed in layers exceeding 375 millimetre compacted thickness. Select Fill is to be placed and spread in a direction parallel to the exposed face of the structure. The path of plant including compaction plant shall also generally be parallel to the exposed face of the structure. (emphasis added)

The word "compacted" has been inserted. This makes clear what had previously been unclear. The paragraph is referring to layers in the manner understood by Professor Leshchinsky. The paragraph is not stating (let alone requiring) 375 mm of backfill to be placed before commencement of compaction. It is stating that compacted layers should be no more than 375mm. The document is silent as to how one compacts in the hand compaction zone to reach that compacted layer height. 214 The CW technical specification for the Seawall prepared in March 1993 was to the same effect. It referred to "375mm compacted thickness". (It was common ground that this reference to 375mm in the Seawall technical specifications was a transcription error and should have been 330mm.) 215 Cross-examination of Professor Leshchinsky and Mr Boyd took place on these technical specifications. The question of the cross-examiner often tried to import the proposition that the documents were referring to directions about thickness of backfill to be compacted. A reading of the whole of the evidence of these witnesses does not lead to an acceptance of that proposition. Mr Boyd on a number of occasions sought to make clear his view that these documents were referring to compacted thickness. An example is as follows: Well, the 375 layer is just defining the maximum amount of earthworks in a single operation, whether it's made up of sublayers for compaction is another issue. The change from 330mm to 660mm 216 On 11 May 1993, Mr Scott sent a facsimile to Mr Boyd at RE P/L which enclosed an engineering change proposal for the Seawall. In the context of the documents to which I have made reference, the engineering proposal was expressed as follows: SEAWALL CONSTRUCTION: PROPOSE TO ALLOW BACKFILL TO BE PLACED IN 660MM LAYERS, STRAP TO STRAP

BHE pressed strongly for a construction that these words must only mean that uncompacted backfill would be placed to the inside panels before compaction. In my view, the document states and requires no such thing. First, it is to be understood as a change to be undertaken in the context of the current CW technical specification for the Seawalls dated March 1993, which stated (see [213] above) that "select fill is not to be placed in layers exceeding 375 millimetre compacted thickness …" (emphasis added). (The same wording referring to "compacted thickness" appeared in the January 1993 CW technical specification for the Millstream walls.) Secondly, it is to be recalled that "backfill" was a word used as capable of meaning compacted or uncompacted backfill. Thirdly, no reference was made to techniques of compaction in the two zones - mechanical and hand compaction. Fourthly, it is difficult to see this as other than a statement of compacted thickness since the compacted thickness was the subject of the relevant technical specification (CW of March 1993) and to achieve that a higher layer of uncompacted backfill (by about 10%) would be required. Thus for 660mm of uncompacted backfill to be the subject of compaction would lead to somewhat under 600mm of compacted select fill. To achieve what was required by the CW technical specification (issued after the change, see below) of "660mm of compacted thickness" a further layer (albeit a very shallow one) would need to be placed and compacted. This would have negated the advantage sought by the change in the reduction in layers and time for compaction. 217 The engineering proposal contained a record of a "double layer trial" which did refer to: Single layer (660mm) pushed out with dozer. Compacted with: vertical rammers between buttresses. walk behind plate in 1 metre zone. smooth drum 10 tonne roller in remainder of reinforced earth block.

Test results were given, none of which specifically related to the hand compaction zone. 218 Mr Boyd of RE P/L responded on 11 May 1993 in the following terms: Your proposal to allow backfill placement and compaction in 660mm layers as described in your ECP/DCN No. BHE 78 is acceptable to us. Our criteria for acceptance are that the specified Density Index of 80% is achieved over the full layer thickness and that no increase in panel deformation occurs during the earth works operation.

This response was not in terms an acceptance of any proposition that BHE should or must undertake compaction in the hand compaction zone in any particular way, or that it should or must use layers of uncompacted backfill of any particular height. What was plain was that BHE was required to test in accordance with its obligations under the head contract (which it failed comprehensively to do). 219 CW agreed with the proposal and amended its technical specification for the Seawall on 24 June 1993 which replaced the March 1993 version set out above. The fact that the change was concerned with compacted, rather than uncompacted, backfill is made clear and reinforced by the terms of the changes to cl 4.4 of the CW technical specification under the heading "Select Fill Placement". The amendments, including scoring through, were as follows: Select fill placement shall closely follow the erection of each course of facing panels. The select fill is not to be placed in layers exceeding 375 660 millimetre compacted thickness except for the two layers adjacent to the horizontal steps in the panels which shall be placed in 455 millimetre layers . Select Fill is to be placed and spread in a direction parallel to the exposed face of the structure. The path of plant including compaction plant shall also generally be parallel to the exposed face of the structure. 220 As I have already discussed, there was no evidence, lay or expert, to the effect that persons in the position of BHE's employees read and understood these documents, especially the engineering change proposal and CW's technical specification, as requiring BHE's workers in the conduct of their construction work in the hand compaction area to compact only once a layer of 660mm of uncompacted backfill that had been placed. To the contrary of such an interpretation are the terms of the documents, the use of "backfill" as a phrase dealing with compacted material (especially when linked with the phrase "compacted thickness"), the evidence of Mr Boyd that compaction techniques are for the constructor and depended on a number of factors and that experimentation with layer heights of backfill using hand-held equipment was part of proper construction practice, and the evidence of Professor Ingold in cross-examination as to normal practice. Mr Boyd's evidence was that he understood that BHE would compact in single full layers outside the hand compaction area using a motorised roller, but did not assume that BHE would attempt to compact using one full layer of uncompacted backfill in this area. He thought this was self-evident as a matter of construction technique to an experienced constructor. Mr Boyd's evidence was reinforced by Professor Leshchinsky to like effect. The primary judge accepted this evidence. It is not now in contest, subject to its proper characterisation. It was not squarely put to Mr Boyd that the problem with his evidence was that BHE was told to compact the backfill in these thick layers in the hand compaction zone. There was cross-examination of Mr Boyd about the absence of any warning or direction that BHE should only attempt to compact thinner layers in the order of 150mm so as to avoid wrinkles or folds. But it was not put to him that experimentation and testing was not permitted because BHE was required by the text of the design documents to hand compact using one layer of uncompacted backfill of 375 or 660mm. Nor was it put to Mr Boyd that RE P/L was responsible for the failure to achieve compaction of 80% DI. 221 All these matters, but in particular the terms of the very documents, lead me to conclude with confidence that the relevant design documents do not tell the reader that BHE was required to place 375mm or 660mm of uncompacted backfill in the hand compaction zone, in the Millstream walls and Seawall, respectively before it compacted with hand-held equipment.

The "Method Statements" 222 One further group of documents was relied upon as supporting BHE's argument. As stated earlier cl E3.2.2 of the Specification required BHE to provide "Method Statements" being "a description of the method [it] proposes to use to execute the work". RE P/L and CW reviewed these documents. Method Statements were prepared by BHE for both the Millstream walls and the Seawall. Even if these can be seen as design documents (which is doubtful), they did not require any layer of uncompacted backfill to be placed before compaction could commence in the hand compaction zone. For instance, in the Millstream walls Method Statement of March 1993, that can be accepted as having been approved by RE P/L and CW, the following appears: 375mm thick backfill layers will be placed by the excavator and spread using a D6 Dozer or equivalent to strap level of the precast unit. Compaction of the backfill layers will be achieved by smooth drum roller and/or flooding. Trimming of the layers will be achieved by a combination of backblading with 4 in 1 traxcavator or laser controlled dozer blade and the effects of flooding. The layer will then be tested for compaction and level tolerances (refer ITP No.17) and when passed the straps will be layed and connected to the wall panels.

(The initials "ITP" stand for "Inspection and Test Plan" and comprised, along with the Method Statements, an element of BHE's methodology in performing its testing obligations under clause E4.9 of the Specification.) In the Seawall method statement of January 1993 the following appears: Backfill material to be spread by D6 Dozer. It is anticipated that the layer depths will be as shown on Figure MS07/04. The layers will be compacted by a combination of soaking with seawater and compaction with 10 tonne vibrating smooth drum roller. We intend to trail a vibrating plate fitted to a hydraulic excavator for compaction of the one metre wide zone immediately behind the wall. The system for backfill control is detailed in the ITP.

223 There was a further Seawall method statement in July 1993 which stated: Backfill material to be spread by D6 Dozer. Care will be taken to ensure that the strips are prevented from moving during the spreading operation. The general direction of spreading will be parallel to or away from the wall. When the strips have been effectively anchored by the backfill material, the temporary wedges will be removed. It is anticipated that the layer depths will be as shown on Figure MS-7/04. The layers will be compacted by a 10 tonne vibrating smooth drum roller plus smaller compaction equipment in the 1 metre zone adjacent to the wall. Water shall be added from watercart if required.

Figure MS07/04 was a drawing which picked up the layer configuration after the engineering change to 660mm. 224 None of these documents state what BHE would seek to take from them - that uncompacted backfill is to be placed at any required depth in the hand compaction zone before compaction is undertaken. Indeed, the July 1993 Seawall Method Statement emphasises testing, primarily directed to the compaction by roller: "to the establishment of a compaction method (ie watering and number roller passes)…". The following was also stated: Sample locations within each lot on each layer shall be randomly located however ensuring that at least one test in each layer is within the 1 metre zone adjacent to the panels which is compacted by Plate Compactor and/or Wacker Rammer. This ensures a control over both compaction techniques.

The findings of the primary judge reveal this testing was not undertaken.

Conclusion as to the argument of mandated compaction technique 225 In my view, none of the documents relied upon by BHE told an experienced contractor, such as BHE, how to compact. The evidence and the findings of the primary judge were that proper compaction technique using hand-held equipment was a matter of experimentation and testing for the constructor. The testing was required by the contract. It was not complied with. BHE was not obliged to take any step as to the putting out of layers of uncompacted backfill which would affect or constrict it in any way in undertaking its obligation as constructor. 226 BHE sought to rely on how the work was actually performed to support an inference that the BHE workers and engineers thought that they were required to hand compact 660mm of uncompacted backfill. For the reasons that I have already given, such an inference is not open. That BHE undertook the hand compaction work by placing layers of up to 375mm or 660mm of uncompacted backfill adjacent to the panel facings in the Millstream walls and Seawall, respectively, (as, on the evidence, it appears to have done) is equally consistent with it being considered that this course was permitted and available and the employment of poor construction technique, as it is with it being considered that this course was required. It lay within the responsibility of BHE to prove that its workers adopted a backfill and compaction method that was dictated by the design documents. It did not do so. 227 Thus, in my view the whole argument on appeal fails to the extent that it was based on the asserted meaning of, and consequential reliance upon that meaning of, the design documents.

The questions of design and construction and the operation of the policies - why the appeal should be dismissed 228 It may be accepted from the above analysis that BHE had design responsibility. Once, however, it is accepted that the design documents did not mandate the particular (defective) method of backfilling and compacting in the hand compaction zone that was used in construction, one is left with the findings of the trial judge about the comprehensive failure to comply with testing in the hand compaction zone and the adoption of inadequate construction techniques, including a failure to experiment and test. These findings, as discussed below, lead to the engagement of exclusion 1(p), to which I will come. 229 BHE submitted that the primary judge gave too little weight to the contractual documentation and too much weight to the evidence of Mr Boyd and Professor Leshchinsky in characterising BHE's defaults as of construction, rather than design. There are a number of difficulties with this submission. First, it rested on the proposition (which is wrong, for the reasons that I have given) that the design documents required BHE to compact in the hand compaction area using layers of uncompacted backfill of 375 and 660mm. Secondly, in assessing the acts and omissions which were the reason or reasons for the sand loss, the primary judge was entitled, indeed bound, to have close regard to the evidence of such experienced witnesses as Mr Boyd and Professor Leshchinsky. The contract and design documents assist in understanding the mutual rights and responsibilities of the parties, but they cannot, and do not here, transform evidence of failures of construction techniques or practice into failures only of design. There may be failures which can be understood in terms of the documents in the specification of construction method. The primary judge found this in relation to the absence of precautionary warning about mechanism two. That does not mean, however, that this is sufficient to prove that this was the only act, error or omission out of which the claim could be said to arise, or that any such act, error or omission falls within the field of engineering professional activities and duties for the purposes of the policy. The evidence was clear that BHE had exhibited critical and causally central construction defaults: it had comprehensively failed to test in accordance with the contract (remembering that this is something it was clearly told to do), it had failed to undertake experimentation and testing in accordance with good practice to compact adequately (to 80% DI) and without damage to the geotextile. Once these related failures were demonstrated in the evidence, the engagement of exclusion 1(p) was clear. The only way that this conclusion could be gainsaid was if BHE's argument on appeal was successful that its failures just identified were irrelevant, because it was obliged to undertake the compaction work in this way and complied with that requirement. Then, it could be said, its failures as identified by the primary judge should not be characterised as causally relevant. No testing would have assisted, until a design change was made directing a construction method (or technique) that would lead to adequate compaction without geotextile deformation. 230 Some of the submissions of Mr Finch were somewhat derisory of the notion that BHE should from time to time have dug up compacted fill adjacent to the wall to inspect for folds or wrinkles. There was no reason, however, why such destructive testing should not have been undertaken as part of proper construction practice to ensure that the compaction of the backfill with hand-held equipment was proceeding according to a technique that would not distort the geotextile. 231 As I have already said, critical and overwhelming problems for this argument about the design documents are that, first, the design documents did not require what was done and, secondly, there was no evidence upon which it could be inferred or concluded that BHE acted as it did because it thought it was required to from the terms of the design documents. It is true, as BHE pointed out, that it had design responsibility which was in significant part passed on to RE P/L and CW as consultants. I have set out earlier the terms and effect of such provisions. These did not, however, disentitle the primary judge from drawing on the evidence to reach a conclusion as to what was involved in the process of construction. 232 To state that the construction defaults could and should have been prevented by a specification which descended to the detail of telling BHE how to do the things that the witnesses said BHE, as an experienced constructor, would be taken to know is to seek to have logic replace evidence. There was no evidence that such a direction as to how to compact using hand-held compaction equipment was required. It does not flow logically from the poor construction technique or practices of BHE. Nor can it be concluded that the placement in the documents as to how to compact would have been followed. Testing was compulsory and it was not done. No witness was called to say he or she needed help, or would have followed directions, about how to compact with hand-held equipment. 233 With respect, BHE's submissions lack an air of reality in that they suffer from a lack of foundational evidence. 234 The above conclusions do not deny the possibility that the primary judge should have attributed the deficiencies in instruction requiring the adoption of a construction practice calling attention to the need for caution in dealing with the geotextile not only to RE P/L and CW, but also to BHE. 235 Gordian and CGU submitted that the primary judge's findings that BHE had not proved any design role or undertaking was correct and should not be disturbed. There is no doubt that the evidence from witnesses did not descend into an investigation of the role of BHE engineers in relation to the formulation of the technical specification, drawings, method statement and like documents. For instance, Mr Ashton, who at some point became the construction manager responsible for overseeing wall construction gave evidence relating to the known circumstance issue, but not about the claim under special provision 2 or BHE's possible design responsibility under insuring clause 1. Neither Mr Caretti nor Mr Scott dealt with the design issues about which the work of RE P/L and CW was found wanting by the primary judge. Mr Betbeder (the Project Director with significant experience) was not called; nor was Mr Burns (project manager for the Seawall) or Mr Dujmovic (the project manager). 236 BHE did, however, prove that it had design responsibility to the FAC under the head contract, that it shared that responsibility with RE P/L and CW under the subcontracts and that it was involved in design considerations and discussion. BHE's submissions on appeal and their references to the contractual responsibility of BHE in relation to design persuade me that to the extent that RE P/L and CW can be the subject of criticism in inadequate design by the failure to incorporate instruction as to construction method or practice, that must have also been a failure of BHE to undertake to the FAC the design responsibilities that it clearly had and shared with RE P/L and CW. 237 The absence of evidence from witnesses is, nevertheless, important in this context. Its importance is best understood in the context of assessing the legitimacy of the primary judge's conclusion that the acts, errors and omissions which potentially engaged special provision 2 did not in fact lead to the engagement of that provision because the claim of SACL did not sufficiently arise out of, or was not proved to arise out of, the relevant act, error or omission: see [955] of the primary judge's reasons. It is at this point that the state of the evidence becomes crucial. Not only did the primary judge find that BHE was upon notice of the need to be cautious, but the evidence was such that it would have been wrong to conclude that, even if the specifications or drawings had contained a precautionary warning of the kind found by his Honour to be necessary, BHE would have relied on it. First, the evidence was clear that BHE had made known that it did not need the advice of RE P/L on construction technique. This was discussed by the primary judge at [1030] and [1031] of his reasons. Mr Boyd's evidence was clear about this. Against the background of the size and experience in the field of work of BHE, this is telling. Secondly, BHE was told to test and it failed comprehensively to do so in the hand compaction area. Thirdly, BHE called no witnesses who dealt with either the design process or the development of construction method or technique about compaction of the reinforced earth walls or who could have given evidence as to:

(a) what they would have considered had a precautionary warning of the kind discussed by the primary judge been included;

(b) the techniques used to compact backfill in the hand compaction area;

(c) whether they read the documents created or approved by RE P/L and CW in relation to design, in particular in relation to compaction, and, if they did, whether they understood them to require 375mm and 660mm layers of uncompacted backfill; and further if they did, whether any greater specificity of direction of the kind the primary judge found RE P/L and CW should have given would have been read. 238 This lack of evidence is all the more telling given the experience and reputation of BHE as a major project constructor. Without such evidence it could not be concluded that any failure of design by RE P/L or CW (as found by the primary judge) was in any way causally relevant. Nor could it be concluded that any correlative design fault of BHE would have led to a different approach at the point of construction. No contradictory evidence was given to that of Professor Leshchinsky and Mr Boyd about the role of a competent constructor in experimenting and testing to find and employ, conformable with the equipment it was using, the most appropriate layers of backfill to compact to achieve 80% DI and without causing wrinkles and folds. Mr Boyd's evidence was clear as to BHE's lack of need for consultation with RE P/L as to construction techniques. He was assured by BHE that they were adopting practices to achieve compaction. 239 Thus, even if one concludes, as one should I think, that any deficiencies in instruction, at least by way of omission, for which RE P/L and CW were responsible should in the contractual context and factual circumstances be also recognised as a default of BHE, that does not mean one can conclude that insuring clause 1 was engaged. For the reasons that I have just identified that special provision 2 was not proved to be engaged, so it was not proved that insuring clause 1 was engaged. There was no basis in the evidence to conclude that any precautionary instruction of the kind discussed by his Honour would have had any effect on the conduct of the work - whether or not it was the responsibility of BHE or RE P/L or CW. 240 BHE sought to plug this fundamental evidential gap with a proposition that an evidentiary onus fell on the insurers to rebut a presumed causal relationship between the absence of the precautionary warning found and the claim. Reference was made to Betts v Whittingslowe (1945) 71 CLR 637 at 649; Chappel v Hart (1998) 195 CLR 232; and Schellenberg v Tunnel Holdings Pty Ltd (2000) 200 CLR 121 at [51]. See also Flounders v Millar [2007] NSWCA 238 and Roads and Traffic Authority v Royal [2008] HCA 19; 245 ALR 653. Whilst in some cases a breach of duty coupled with an accident of the kind that might thereby be caused is enough to justify a factual inference about a causal link, that is not a rule of law. It is an aspect of an available process of drawing of conclusions about causation: Royal at 662 [31] and 689 [143]. Here, BHE chose to direct most of its evidence in terms of compaction and design to showing that it had satisfactorily compacted to 80% DI and tested in accordance with the contract and that design flaws as espoused by Professor Ingold explained the sand loss. Virtually no evidence was led from engineers and construction and project managers as to how they would have taken into account instruction or advice or precautionary warnings in relation to a field of activity that was already, on the evidence, within their direct field of competence. This absence of evidence is crucial, in particular in circumstances where BHE was found to have failed to comply with its obligation to test for compaction density in circumstances where it was told by the contract documents to do so. 241 The primary judge was, in my view, in error in failing to conclude that any deficiency or omission by the absence of a precautionary warning found to be RE P/L and CW's responsibility was also BHE's responsibility. The primary judge, nevertheless, was correct to conclude that the claim did not arise out of RE P/L and CW's act, error and omission for the purpose of special provision 2. Further, because of the absence of attention to the issue in the evidence, it is not open to conclude from the fact that BHE was also legally responsible for the act, error or omission that this would have had any effect on the course of events and the cause of the sand loss. So, just as his Honour concluded (correctly) that special provision 2 was not engaged, it can also be concluded that insuring clause 1 was not proved to be engaged. 242 Two further issues related to the BHE appeal now need to be addressed. The first is whether the existing findings by the primary judge of the deficiency in the precautionary instructions said by his Honour to be an act, error or omission of RE P/L and CW encompassed implicitly a finding that RE P/L and CW should have directed a construction technique of specific layer heights of uncompacted backfill in order to achieve compaction of 80% DI, and in order not to disturb the geotextile. Secondly, if the answer to that is no, should a finding be made by this Court that the design instructions were deficient in that respect. 243 As to the first issue, I agree with the submissions of the respondents that his Honour did not go so far. The primary judge expressed himself in a variety of ways. See the expression of his findings in [955], [972]-[973], [975]-[976], [978]-[980] and [1021] of his Honour's reasons. The primary judge identified a number of steps that might have been suggested, but the gravamen of his conclusions was that in giving instructions as to construction method there should have been a warning as to precautions and care to be adopted in ensuring folds, wrinkles and creases did not occur in the geotextile. It was not a conclusion that BHE should be instructed how to compact to a given Density Index. It was an instruction by way of warning for care as to the geotextile, because of the consequences of sand loss through avenues of passage of sand if folds, wrinkles or creases appeared. If the layer height of uncompacted backfill of no more than 150mm was intended to be encompassed within this, it would be difficult to reconcile with his findings (otherwise soundly based in the evidence) that choice of the optimum layer of depth of backfill was not a theoretical question but one for construction experimentation and testing. 244 This being the case, should this Court now make a finding that there was a further design or instruction defect by RE P/L, CW and (on the reasoning that I prefer) necessarily BHE, in failing to tell BHE as constructor how it should conduct hand compaction, specifically that to achieve 80% DI and to minimise risk of interference with the geotextile, it should experiment and test its layer heights of backfill in the hand compaction zone, but in no case have them higher than about 150mm? 245 In my opinion, the Court should not take this step. No witness, lay or expert, addressed the question squarely. No one stated that BHE had to be directed how to compact. Mr Boyd's and Professor Leshchinsky's evidence to which I have referred was to the effect that BHE as a competent constructor should have known how to compact backfill in a hand compaction zone. No BHE witness said that he or she needed such instruction about that matter. BHE's professed experience tended to the contrary, as did the evidence of BHE stating that it did not require RE P/L's advisory assistance. 246 One of the difficulties faced by BHE in the propounding of this as a fall back position was that it had significant tension with its primary case that 80% DI compaction was achieved. Thus, no body of evidence was directed by BHE to the point. To cure this problem BHE tended in its submissions on appeal to deal with the matter on appeal as one of logic: There was an absence of good construction technique, being the absence of experimentation and testing to find the correct layer of uncompacted backfill to achieve 80% DI, without distorting the geotextile and a layer of uncompacted backfill of too great a thickness was used; the construction technique of the correct layer lift was not addressed in the design documents; therefore those responsible for that absence of specification of construction technique caused the problem. But this way of putting the matter fails to address the question as to whether, in a circumstance of this kind of construction, a designer would be expected to tell a constructor about this kind of thing. The fact (as found) that a constructor failed to undertake adequately what is expected of it, does not (logically or probatively) lead to the conclusion that someone has failed in an obligation to tell the constructor how to construct. This is a particularly acute question in circumstances where BHE was told, by the contract to test in the hand compaction area, and did not do so. 247 In any event, even if one were to conclude that the technical specification or drawings should have told BHE how to compact, the primary judge found (in my view correctly) that these matters were also within the field of exclusion 1(p). That might only be avoided if it could be concluded that BHE only compacted in the way it did because it was following a mandatory statement of compaction technique. For the reasons that I have already given, that central plank of the appeal fails. 248 To the extent that, contrary to the above view, a direction as to the height of backfill to be compacted in the hand compaction area should have been specified or some warning should have been given as to construction technique to achieve compaction of 80% DI in the hand compaction zone by warning about the use of layers of backfill of greater than 150mm in order to achieve compacted layer lifts of 375 or 660mm, it is an open question as to whether such failure would fall within the defined "professional activities and duties" in the policy. No expert evidence was directed to that. I would not be prepared so to conclude without evidence. The fact that it may be (on this hypothesis) a failure under the contractual and subcontractual regime would not be determinative. It is a failure to tell a skilled experienced contractor how to undertake construction work. 249 This consideration also arises in relation to the act, error or omission of RE P/L and CW as found by his Honour. I am far from persuaded that this fell within the definition of "engineers" and "professional activities and duties". It can be accepted that, to a degree, the care of the geotextile during the compaction process that generated debate for the purposes of mechanism two can be seen more likely to meet the description of "designing" and "geotechnical" and construction "method" than a direction as to how to compact sand in a confined space by hand-held equipment in order to obtain compaction at 80% DI and in order not to create folds, wrinkles or creases in the geotextile. The primary judge appeared prepared to conclude that the act, error or omission of RE P/L and CW fell within "professional activities or duties" for the purposes of special provision 2. Whilst I am not prepared to conclude that the primary judge was in error about this, if, contrary to my view, it should be concluded that there was an omission in instruction as to how to compact safely in the hand compaction zone by employing thinner layers of backfill to compact, I am not prepared to conclude without evidence that this falls within the policy as "professional activities and duties" of an engineer.

Exclusion 1(p) 250 BHE sought to use some communications before entry into the policy to assist in the construction of exclusion 1(p). The insurance programme was put in place pursuant to clause 20.1 of the head contract between FAC and BHE under which FAC was required to maintain "Contracts Works Insurance", "Public Liability Insurance" and "Professional Indemnity Insurance". The description of the professional indemnity insurance required was as follows: [T]o the limit of cover and for a period of two years after the expiration of the Defects Liability Period for the Works in the joint names of the Principal, the Principal's Representative, the Contractor and all consultants and sub-consultants in respect of liability at law for breach of professional duty by reason of any negligent act, error or omission occurring in the performance of this Contract to the limit of cover and subject to the deductibles set out in the Annexure. 251 In December 1991, a proposal filled out by FAC to the broker, Sedgwick James, and one assumes provided to the underwriter, described the insurance as for "Professional Indemnity for Single Projects". FAC described the activities as follows: • Preliminary design has been completed by FACs principal Consultant, Airplan-GHD • The Managing Contractor is to complete the design with his own consultants. 252 By September 1992, BHE was known to be the contractor. In early September 1992, a meeting took place between representatives of the broker, FAC and BHE. A facsimile sent by BHE to the broker on 3 September records the following: On your comment on Exclusion 1(p) you agree to modification as follows: The intent of the exclusion is to exclude defective workmanship, that is the defective or incorrect application of otherwise sound methods from the viewpoint of design or the exercise of the other professional skills described in the Policy. 253 BHE relied on this in relation to the construction of exclusion 1(p). One immediate difficulty with that is that there was no evidence of authority of the broker (in this non-marine context) from the underwriters. If this difficulty could somehow be overcome it was prayed in aid of a construction to limit the operation of clause 1(p) to circumstances of defective workmanship of otherwise sound design or professional skills. Thus, it was submitted, it helped the Court to reach a construction that exclusion 1(p) did not apply where there was defective workmanship and defective design or professional skills. The difficulty with this proposition is that the document is being used not to establish the factual matrix in which the policy came into existence, but to control the proper interpretation of the words of the policy. This is not permissible: Codelfa Construction Pty Ltd v State Rail Authority (NSW) (1982) 149 CLR 337 at 347 and 352. 254 In my view, these pre-contract communications should not be taken into account in construing exclusion 1(p). 255 The first complaint about the primary judge's invocation of exclusion 1(p) was based on the primary argument on appeal - that BHE was required to effect compaction following an instruction to compact (only) layers of uncompacted backfill of 375 mm and 660 mm. That argument fails with the failure of this essential argument. 256 BHE also submitted that exclusion 1(p) should be narrowly construed so as not to cover construction in accordance with a defective design. 257 It is clear that "construction work" in the first line of exclusion 1(p) is the performance of work, not the finished product or the notion of the contractual task. It is something that is to be "performed" and involves what follows: "means, methods, techniques, sequences, procedures and use of equipment". 258 The exclusion is to be construed in accordance with the principles laid down by the judgment of the High Court in Darlington Futures Ltd v Delco Australia Pty Ltd (1986) 161 CLR 500 at 510 reiterated by the judgment of the High Court in Nissho Iwai Australia Ltd v Malaysian International Shipping Corp Berhad (1989) 167 CLR 219 at 227: … the interpretation of an exclusion clause is to be determined by construing the clause according to its natural and ordinary meaning, read in the light of the contract as a whole, thereby giving due weight to the context in which the clause appears including the nature and object of the contract, and, where appropriate, construing the clause contra proferentem in case of ambiguity. 259 The evident commercial purpose in exclusion 1(p) is to remove from the scope of indemnity claims that arise out of the performance of construction work. The prepositional phrase "arising out of" can be seen to be somewhat wider than the prepositional phrase "by reason of" used in insuring clause 1 (though cf "arising out of" in special provision 2). That difference in wording tends against the submission of BHE that exclusion 1(p) should be seen as limited to circumstances where the claim arises only out of the performance of construction work and not out of design work. So to restrict the clause would tend to make exclusion 1(p) devoid of real operation; for if it only operated when claims arose only out of construction work, it is difficult to see how the insuring clause could ever be engaged based as it was on professional activities and duties of engineers or specialist designers or consultants. In my view, the commercial purpose of exclusion 1(p) was to exclude the risk of defective construction performance. The notion of "defective" is introduced by the relationship of the acts with the claim. There will be a claim arising out of the construction work performed if it is defective. That was certainly the case here, though it is unnecessary to explore the limits of operation of exclusion 1(p) in this respect. 260 On the evidence here, and given the absence of proof as to reliance on the act, error or omission of RE P/L and CW, the proximate cause of the claim by FAC and SACL was the defective construction. As Mr Finch agreed in argument, if the specification and documentation was not mandatory as to thickness of uncompacted backfill layer, and if the claim can be seen to arise also out of that act, error or omission the finding of defective construction takes one into exclusion 1(p) with the claim arising out of both construction work performed by BHE and acts, errors and omissions of BHE, RE P/L and CW. 261 The difficulty inherent within all BHE's submissions on exclusion 1(p) is that once one comes to the conclusion that its primary argument as to the content of the instructions in the design and construction documents is wrong, one is left with the primary judge's views that mechanism one has been caused by poor construction technique and mechanism two has been brought about by the same poor construction technique together with the errors and omissions of the designers RE P/L and CW. Even if one were to conclude that the design error and omission, for which RE P/L and CW were held responsible by the primary judge, should also be held to be the responsibility of BHE, as designer, and that the claim can be seen (for the purposes of insuring clause 1 and special provision 2) to arise out of the error and omission of the designers (BHE, RE P/L and CW) one is still left with his Honour's conclusion that the claim arises out of construction work by BHE for the purposes of exclusion 1(p). Further, even if it be the case that BHE, RE P/L and CW are to be further criticised as designers for failing to include in the technical specifications or drawings more detailed instruction to BHE (as constructor) as to how to execute compaction in the hand compaction zone by prescribing or suggesting layers of uncompacted backfill of less than 150mm, BHE still faces the problem of the application of exclusion 1(p). 262 There may be factual circumstances that lead to the operation of exclusion 1(p) where there is an absence of fault by the constructor. That hypothetical proposition need not be dealt with. It is difficult, however, to envisage circumstances where a design which contains a flaw is incorporated into construction which is undertaken with competence and skill producing an inadequate and unsatisfactory piece of work which would lead to the engagement of exclusion 1(p). If Professor Ingold had been correct and the geotextile was an inappropriate element, the fact that the walls had been constructed using it did not mean that the claim "arose out of construction work". Here, with the failure of BHE's characterisation argument based on the content of the design and construction documents, one is left with the conclusions of the primary judge that BHE's comprehensive failure to test for compaction density and its failure to experiment and use an appropriate construction technique to compact in the hand compaction area were aspects of executing construction responsibilities. Such conclusions were amply supported by the evidence and the failures were found to be an operative cause of the sand loss through both mechanism one and mechanism two. It can be concluded that there was a design act, error or omission of RE P/L and CW, as found by the primary judge and assumed of BHE and that the claim arose out of that also, insofar as this contributed to mechanism two. It can even be assumed for present purposes that there was a failure by BHE, RE P/L and CW to prescribe compaction techniques and layer lifts of uncompacted backfill of 150mm in the hand compaction zone. On these hypotheses one has both the insuring clauses (insuring clause 1 and special provision 2) and exclusion 1(p) engaged because of the operation of two causes - the construction failures of BHE and the design failures of BHE, RE P/L and CW. 263 In these circumstances, where there are two or more causes and one falling within an insuring clause and another falls within the exclusion, the policy will not respond: Wayne Tank and Pump Co Ltd v The Employers' Liability Assurance Corp Ltd [1974] QB 57. In McCarthy v St Paul International Insurance Co Limited (2007) 157 FCR 402 at 429-438 [88]-[116] I examined in detail the content and operation of the so-called "Wayne Tank rule" and the English and Australian cases dealing with it. Without setting it out, I repeat what I there said. Ultimately it is a matter of construction of the policy. Did the parties intend the policy to respond in circumstances where a claim is brought about in circumstances that are encompassed by the contractual exclusion? 264 Here, exclusion 1(p) is tolerably clear: the policy "shall not indemnify" BHE in respect of a claim made against it "[a]rising out of construction work performed involving the means, methods, techniques, sequences, procedures and use of equipment of any nature whatever … in executing any phase of" the works. With the failure of BHE's primary argument on appeal, in my view, it is not open to doubt that BHE's failures to test and to employ proper compaction and construction techniques in the hand compaction area fall squarely within exclusion 1(p). In my view, there could be no doubt that a cause, being a relevant proximate cause, of the claim was this construction inadequacy of BHE in its means, methods, techniques and procedures of construction such that the claim arose out of that. To construe the clause more narrowly such that in the circumstances of this case the policies responded would be to ignore to the words of exclusion 1(p). 265 No point was taken in argument contrary to the discussion and conclusion in McCarthy v St Paul Insurance that there could be more than one operative cause in the answer to a question about the response of an insurance policy in this context. Here, the flaws in BHE's construction means and techniques were plainly an operative cause of the claim such that it can be said to arise out of them. That is sufficient. In any event, for the reasons given above, the primary judge was entitled to conclude that the liability of BHE did not arise out of RE P/L's or CW's act, error or omission or out of BHE's co-ordinate responsibility for such. 266 The above reasons deal with the Notice of Appeal, as follows:

(a) As to grounds 1 and 3, there was no error in the primary judge's conclusion that special provision 2 was not engaged.

(b) As to grounds 4 and 6, there was no error in the primary judge's conclusion that insuring clause 1 was not engaged.

(c) As to grounds 7 and 9, there was no error in the primary judge's conclusion in finding that exclusion 1(p) was engaged.

(d) As to grounds 2, 5 and 8 and the identified subsidiary findings: (i) The findings at [833], [911], [925], [926], [969] and [971] to the effect that there was no relevant design involvement of BHE are incorrect. (ii) The finding in [955] that the claim did not arise from RE P/L's and CW's acts, errors and omissions is correct. (iii) The findings that the claim arose from BHE's construction errors in [980]-[986], [1013], [1020], [1021], [1022], [1028], [1031], [1033] and [1042] are correct and amply supported by the evidence. (iv) The findings as to the construction of exclusion 1(p) at [1004], [1005], [1012], [1013], [1019], [1024], [1025] are correct. 267 For these reasons, the policies did not respond and the appeal should be dismissed.

Other matters argued 268 A number of other matters were argued, but which, given my views on the appeal, do not strictly arise. Nevertheless, for completeness, it is appropriate to address them. These matters are as follows:

(a) Whether the primary judge's conclusion about the default of RE P/L and CW as to the precautionary instructions should be set aside.

(b) The proper construction of special provision 2 as a so-called stand alone insuring clause.

(d) The so-called "excess point" and the operation of exclusions 1(e) and 1(o) and cl 7.

(e) The lack of separate treatment of RE P/L and CW. (a) Whether the primary judge's conclusion about the default of RE P/L and CW as to the precautionary instructions should be set aside 269 It was asserted that the findings about the act, error or omission of RE P/L and CW discussed earlier were not supported by the evidence and were inconsistent with the primary judge's conclusion that BHE as constructor, whether warned or not, knew of the need to proceed with caution and not to compact in a way that would lead to the geotextile being folded, creased or wrinkled. 270 It is true that there was no expert evidence specifically directed to this question. That is not fatal to the finding. The primary judge absorbed over a long trial the evidence of both professors and Mr Boyd. His Honour concluded, in the end, that the instructions for construction method and sequence should have contained some precautionary warning. In the end the respondents were not, in my view, able to demonstrate how his Honour fell into error in this respect. The fact that he thought it appropriate for there to be a warning about a matter that the constructor should be taken to be aware of is not an inconsistency. The limits of the knowledge of the constructor does not necessarily mark the commencement of the obligation of the designer. 271 I would not be prepared to interfere in this respect with the views of the primary judge. (b) The proper construction of special provision 2 as a so-called stand alone insuring clause 272 Some of the difficulties and arguments about the construction of special provision 2 stem, in my view, from the language of the debate. To ask whether special provision 2 is a stand-alone insuring clause is to misdirect attention away from the words used and the place of the provision in the whole policy. 273 Plainly the clause is an extension of the indemnity. Plainly it is "subject to the Policy's terms and conditions". The extension is to cover acts, errors and omissions by specialist designers or consultants, not by the Insured. Thus, I do not accept the respondents' submissions that there was required still to be an act error or omission of BHE. This is not to relieve the operation of special provision 2 of one of the terms and conditions of the policy; rather, it is merely to recognise the point or interface of the extension of indemnity. The relationship with the Insured's "Professional Activities and Duties" arises from the fact that the act, error or omission of the specialist designer or consultant must likewise be in the conduct of "professional activities or duties", acting on the Insured's behalf and for whom the Insured was responsible. 274 Also, I reject the argument of the respondents that the phrase "on the Insured's behalf" restricts or limits the application of the provision to circumstances where the designer or consultant acted for the third party as agent, for or representative, of the insured. The words of the clause do not warrant such a narrow focus of indemnity. Looking at the contractual documentation here, BHE's responsibility to engage and responsibility for, RE P/L and CW and the necessary co-operation in the design function, it is a perfectly natural use of language to say that RE P/L and CW were acting on BHE's behalf and were consultants for whom BHE was responsible. 275 Otherwise, special provision 2 fits into the policy fully subject to the terms and conditions without the need for implication of terms as found by the primary judge. (c) The so-called Gordian attachment points 276 The primary judge examined clause C of the Gordian policy. He identified, correctly, that Gordian's obligation to provide indemnity only arose upon the occurrence of one or more of the preconditions in that provision. These preconditions were expressed by the statement that Gordian would "only be liable … after [relevantly, HIH]":

(iii) has "been held liable to pay the full amount of its indemnity" . (i) The question whether HIH had "paid" 277 In its written submissions, BHE accepted that this first precondition had not been satisfied. In argument, the point was taken and reliance was placed on the discussion of the word "paid" in the context of reinsurance treaties in Charter Reinsurance Co Limited v Fagan [1997] AC 313, a case not cited to his Honour. 278 In my view, Charter Reinsurance v Fagan is not determinative of the meaning of this policy. Here, as the primary judge noted, the policy distinguishes between admission, being found liable and paying. These are the three (alternative) acts or circumstances which will crystallise the obligations to indemnify of the following excess insurer. It is a form of linkage of the policy to have Gordian as an excess insurer follow the primary insurer, but only where full value of the underlying policy is or will be taken up indemnifying the insured. Gordian has taken its share of the premium on the basis of a response by the underlying cover. The way that response may attach is set out in clause C. It is plain that there is no requirement for HIH to have actually disbursed funds for the indemnity of Gordian to attach: HIH can admit liability to pay or be held liable to pay. 279 In Charter Reinsurance v Fagan, three reinsurance policies were in issue: two whole of account excess of loss reinsurance policies and an aviation excess of loss reinsurance policy. All three contracts provided that the reinsurer was to be liable for the losses of the reinsureds in excess of an ultimate net loss of a specified amount. The relevant terms "net loss" and "ultimate net loss" were defined as "the sum actually paid by the reinsured in settlement of losses or liability after making deductions for all recoveries, all salvages and all claims on other reinsurances whether collected or not and shall include all adjustment expenses arising from the settlement of claims". The issue was what the phrase "sum actually paid" in this context meant. 280 The reinsureds were insolvent. The words were in common usage in the market and the consequences of success of the reinsurers' arguments were considerable in the London reinsurance market: [1997] AC at 318 in the judgment at first instance of Mance J (as his Lordship then was). The judgment of Mance J contains, if I may respectfully say so, an illuminating exposition, not only of the principles governing the construction of insurance policies (see [1997] AC at 325-328), but also of the historical development of following clauses. Though the case was dealt with by the Court of Appeal and House of Lords, the reasons of Mance J can be seen as surviving the appeal process. Indeed Lord Mustill (with whose speech Lord Goff of Chieveley, Lord Griffiths and Lord Browne-Wilkinson agreed) agreed with Mance J's interpretation and Lord Hoffmann expressed his full agreement with Mance J's reasons. The reasons of Mance J illuminate why the case is not determinative of the present problem. In the history of following clauses discussed by Mance J, one finds reference to the proposition that "to pay as may be paid" did not affect the prima facie rule at common law that recovery under a reinsurance policy depended upon proof of legal liability as well as measurement of the reinsured's financial commitment: Chippendale v Holt (1895) 1 Com Cas 197; Versicherungs und Transport A/G Daugava v Henderson (1934) 49 Ll L Rep 252 at 253 and 254. The distributive wording used by Gordian here can be seen to be wider than a clause such as "pay as may be paid". After a comprehensive review of relevant authorities and a careful consideration and discussion of the objects of ultimate net loss clauses, Mance J concluded that the words did not signify a precondition of liability or indemnity, but were part of the measure of loss. The Court of Appeal (Simon Brown and Nourse LJJ, Staughton LJ dissenting) dismissed the appeal. The House of Lords unanimously dismissed the appeal. 281 It is unnecessary to discuss further the reasoning in Charter Reinsurance v Fagan or the meaning of similar words and phrases in following clauses or net loss clauses: cf Firma C-Trade SA v Newcastle Protection and Indemnity Association (The Fanti); Socony Mobil Oil Co Inc v West of England Shipowners Mutual Insurance Association (London) Limited (No 2) (The Padre Island) [1991] 2 AC 1; Pine Top Insurance Co Ltd v Unione Italiana Anglo Saxon Reinsurance Co Ltd [1987] 1 Lloyd's Rep 476. Unlike the relevant phrase in Charter Reinsurance v Fagan, the words of clause C express a clear commercial purpose of expressing a precondition to the attachment of liability. A composite phrase which might need to be disaggregated to discover its full meaning has not been used. The disaggregation is express: paid or admitted liability or held liable. One, at least, must be present for Gordian to be liable. 282 Here, in my view, "paid" has its usual or ordinary meaning and means funds disbursed.

(ii) The question whether HIH had admitted liability 283 The primary judge concluded that HIH had made no admission of liability. The debate about this was in short compass. It concerned the effect of correspondence between HIH and SACL concerning the claim. There was no evidence of the lodgement of a proof of debt in HIH's insolvency or as to whether the liquidator of HIH had admitted or rejected any such proof. 284 The letters in question were set out by the primary judge at [1248] and [1249] of his reasons. On 6 November 2002, the assistant claims manager of HIH wrote to Mr Finnigan of Phillips Fox (which firm acted for BHE at the time). The letter stated: We refer to the meeting on the 31st of October 2002 at your office and confirm that indemnity is granted to Baulderstone Hornibrook Engineering pursuant to policy no: 9541NK18365 in relation to the claim against it by SACL resulting from the subsidence at the millstream and seawalls of the third runway at Sydney Airport. The grant of indemnity is subject to the policy terms, conditions and exclusions and is based on the facts presently known to HIH. Also confirmed that HIH no longer asserts that policy 9541NK18365 on its proper construction sits in excess of AMPG policy no: D0020608W. Please keep our office fully informed of the claim position and progress.

(No reliance was placed on what was said at any meeting on 31 October 2002 referred to in the first paragraph of this letter.) 285 Having set out this letter of 6 November 2002, the primary judge at [1248] and [1249] said the following: The underlined material [the second paragraph in the letter of 6 November quoted above] makes it clear there was no admission of liability by HIH. 286 For the reasons below, I cannot agree with this conclusion of his Honour. He gave no reasons for his view. He appeared to consider the issue too clear for debate. I also view the matter as plain, but, I regret, differently to his Honour. I see the letter as an admission of liability, though in some respects qualified. 287 On the same day, HIH wrote to Mr Finnigan again in his capacity as solicitor for AMP. In this letter the HIH claims manager said: It appears that HIH in regard to policy 9541NK18365 & AMPG in regard to policy D0020608W, have both confirmed indemnity to Baulderstone Hornibrook Engineering Pty Limited. 288 On 16 December 2003, CMGL ("Claims Management Group Limited", the entity administering HIH's claims after its liquidation) sent a letter to Phillips Fox. The letter contained the following: We refer to your letter of 15 December 2003 regarding the facilitation before Mr Terence Cole QC and the imminent settlement with SACL and respond as follows: 1 HIH grants indemnity to BHE under policy number 9541NK18365 for the above claim subject to policy terms, conditions and exclusions and based on the facts presently known to HIH; 2 HIH agrees that the proposed "without prejudice" settlement as detailed in the Estimate of Rectification Costs in the Parallel Runway Rectification Grouted Column Solution Estimate Review dated 26 November, 2003 is reasonable and consents to BHE entering into the settlement agreement subject to the final terms remaining substantially as set out in the Estimate Review dated 26 November, 2003 and subject to appropriate funding being put in place so that the proposed settlement can be effected; 3 HIH agrees that the quantum of the settlement exceeds the excess and level of cover available under the HIH policy; 4 HIH accepts that BHE's reasonable costs and expenses incurred in the defence of the claim by SACL will be covered by HIH on a proportional basis; and 5 HIH reserves its rights in relation to the issue of whether the corrosion of the bolted connections gives rise to a separate claim or cause, but only if the proposed settlement does not proceed. 289 The primary judge stated at [1250] of his reasons: The reservations to the effect that the grant of indemnity was subject to the policy terms, conditions and exclusions and that the grant was based on the facts than [sic] known to HIH, put the matter out of Court. There was no "[admission] of liability" within the meaning of Clause C. (emphasis in original) 290 I cannot agree. Insurers (as they did here) often speak of a "grant of indemnity" as a way of stating an acceptance of the response of their policy. It is of course important to recall that an insurer has entered a policy under which it is either legally liable to indemnify or not. (There may be exceptions to this proposition, in particular areas of mutual or other insurance where the grant of an indemnity may contain an element of choice by the insurer: see, for example the policy in Bailey v New South Wales Medical Defence Union Limited (1995) 184 CLR 399.) In one sense, there is nothing to "grant". The expression "grant indemnity" in these circumstances at least contains the recognition or admission, in perhaps customary language, of liability under the policy. 291 The "grant of indemnity" or admission of liability may also be seen, in certain circumstances, to give rise to another agreement. For instance, if after some dispute as to coverage, an insurer stated that the parties should now proceed "on the basis of an admission of liability", a fresh agreement, supported by fresh consideration may come into existence: cf Lubovsky v Snelling [1944] KB 44. See also Law Society v Sephton & Co [2004] EWHC 544 and Thiess Pty Ltd v ERC Frankonia Reinsurance Limited [2007] QSC 4 where Chesterman J said at [38], [40] and [41]: In my opinion an insurer's acceptance of an insured's claim necessarily connotes an agreement. … Put more briefly, to accept a claim is to agree that the claim is valid and to promise, or impliedly promise, to pay it. … [A]n insurer's acceptance of its insured's claim does away with the potential uncertainties and their capacity to generate costs which the insurer may have to bear, if the claim is litigated, and replaces them with the insurer's intimation (or promise) that it will pay the claim. There is, in accepting a claim, an element of compromise, or of admission that the insured is entitled as a matter of contractual right to the indemnity contained in the policy. The consideration is the benefit to the insurer that it will not be liable to pay the insured's, and its own, costs of an action, and the detriment to the insured of forbearing to sue for the proceeds of the policy. The acceptance of the claim is in form and substance an agreement. Moreover it will normally amount to a contract, legally binding on the parties. 292 Here, the "grant of indemnity" was not expressed to be "without admission" or "without prejudice". Such qualifications would probably, depending on the full context, not amount to an admission of liability, but to a statement of the position that the insurer was prepared to take pending any acceptance of legal liability or for some other reason. Here, the only qualifications were, first, that the grant was in accordance with the terms of the policy. Save for an abundance of caution, that need not have been said. The second was that the grant was on the facts as then known. Again, caution may require this: but, if facts become known that take the circumstances out of the scope of cover, the insurer may well have a basis to withdraw its admission previously made. Again the prudence of using this clear expression gave the insurers the capacity to seek to alter their position should the known facts materially change. These qualifications may limit the scope of any fresh agreement of the kind contemplated by Chesterman J, but they do not detract from the character of the communication - a recognition of liability (that is an admission) under the terms of the policy and on the basis of what was known. 293 In my view, if the policy otherwise would respond, clause C of the Gordian policy was satisfied. (iii) The question whether HIH could be held liable to pay 294 The primary judge stated at [1251] of his reasons that this precondition was not satisfied because of his view of the non-response of the policies. It is necessary, however, if this hypothesis be wrong and if the HIH policy otherwise responds, to assess whether this precondition is necessarily satisfied. 295 The argument put forward by Gordian in answer to this was as follows. BHE sought indemnity from AMPG and accepted $20 million payment from AMPG before the commencement of the proceedings. The AMPG policy covered the same risk as the HIH policy, HIH and AMPG being co-insurers under double insurance. By the acceptance of this payment, BHE lost its right (said to be inconsistent) to claim under the HIH policy. Thereafter, HIH could never be liable to BHE. Thus, under the terms of clause C of the Gordian policy, and with particular reference to the temporal element in the word "after", there would never be a holding of HIH as liable to pay the full amount of their indemnity. 296 Reliance was placed on Sydney Turf Club v Crowley [1971] 1 NSWLR 724 at 730, 734 and 735; Albion Insurance Co Limited v Government Insurance Office of New South Wales (NSW) (1969) 121 CLR 342 at 345 and 349-352; Bonner v Tottenham and Edmonton Permanent Investment Building Society [1899] 1 QB 161 at 176; James Hardie & Co Pty Ltd v Wyong Shire Council (2000) 48 NSWLR 679 at 687-688; Cockburn v GIO Finance Ltd (No 2) (2001) 51 NSWLR 624 at 640 [78]; and Burke v LFOT Pty Ltd (2002) 209 CLR 282 at 299 [38], 300-301 [41] and [42] and 302 [46]. 297 The submission is, in my respectful view, misconceived. 298 The basal principles contained in these cases include the proposition that where several insurers have severally insured a risk, the insured can only receive one indemnity for its loss. Upon the receipt of that full indemnity from one insurer, the others are discharged against the insured. They remain, however, liable to contribute to the insurer that has paid. They are discharged because the insured can have no claim, because it has received full reparation for its loss. 299 In Albion Insurance v GIO, Kitto J used the expression "a single reparation for the loss" to identify the entitlement of the insured and the extent of its rights. One must be careful to appreciate that full payment under one policy (in that sense a complete or full indemnity under that policy) may not see the insured fully indemnified for its whole loss. Leaving aside the existence of a substantial excess, under insurance is the most obvious example. Once an insured's loss has been the subject of reparation, the insured cannot claim indemnity from another insurer in respect of the same loss, because by then there is nothing to indemnify. Subject to the terms of the policies, if there are several insurers, each is liable under its terms until the insured is provided with full reparation or indemnification for its loss. 300 In GRE Insurance Ltd v QBE Insurance Ltd [1985] VR 83, the Full Court of the Victorian Supreme Court was dealing with an argument similar to that (unsuccessfully) put in Sydney Turf Club v Crowley that the paying insurer was entitled to recover through a subrogated action against a second insurer. This was rejected. The applicable principle was contribution, not subrogation. The basis of the absence of a right of subrogation was the full reparation and indemnification of the insured, thereby denying it (and so also the first insurer said to be subrogated) the right to claim an indemnity (against the second insurer) for a fully indemnified loss. At [1985] VR 95, Murray J said: Upon the narrow basis that there was no right left in the insured to support any claim against the appellant the subrogation claim must fail. Even if the respondent had paid only part of the loss to the insured I think the position would have been no different. Any remaining right in the insured against the appellant would have inured only to the benefit of the insured and any claim in excess of that amount would have failed. 301 Thus, if an insured took out four policies each for $20 million and suffered a loss of $60 million, it could sue three insurers and recover $60 million in total. It cannot sue the fourth for an indemnity, its loss having been fully indemnified. The three insurers who paid can seek contribution against the fourth on the basis of equality. Suit against the first, and receipt of a complete contractual indemnity under the first policy, does not discharge the other three insurers. The insured still has a loss to be indemnified. This is what Murray J was referring to in the second and third sentences above. Here AMPG paid $20 million. BHE was still in a position to require reparation and indemnification. It could still sue HIH. If the balance of its loss were minor, say $1 million, a question of contribution would then arise between AMPG and HIH (leaving aside the latter's insolvency). 302 Thus, the acceptance of moneys which gave partial reparation to BHE for its whole loss from a full indemnity under the AMPG policy, did not deny BHE the right to sue HIH for indemnity under its policy because there still remained (on this hypothesis) loss to be the subject of reparation or indemnity. 303 There was no "other insurance clause" in the HIH policy which would disturb this analysis. Exclusion 2 set out earlier dealt with compulsory forms of insurance. In any event, the Insurance Contracts Act 1984 (Cth), s45 would make any applicable clause void. 304 It is unnecessary to discuss the permutations and combinations of contribution rights between AMPG, HIH and the upper layers of the HIH-led programme. It is also unnecessary to deal with how BHE's excess (of $1 million) fits into this analysis. 305 None of the above is directed to the response of policies such as can be characterised as "valued" policies that may not be policies of true indemnity. 306 Similar issues of partial indemnity (by reference to the whole loss suffered) and full indemnity (by reference to the terms of the policy) can arise in relation to the entitlement to take an action concerning which rights of subrogation exist and to the proceeds thereof (where there is no co-insurer). See Page v Scottish Insurance Group (1929) 140 LT 571 at 576. In Canada, there is a long-standing body of authority for the proposition that an enforceable right to sue in the name of the insured does not arise until the insured has been fully indemnified for its loss: see Derham SR Subrogation in Insurance Law (Lawbook Co 1985) at 53ff, and, eg, The National Fire Insurance Co v McLaren (1887) 12 OR 682; Globe & Rutgers Fire Insurance Co v Truedell [1927] 2 DLR 659; Glens Falls Insurance Co v Tom Peters Limited (1957) 10 DLR (2d) 459; Sheridan v Tynes (1971) 19 DLR (3d) 277; and Lawton v Dartmouth Moving & Storage Limited (1975) 64 DLR (3d) 326. See also Halsbury's Laws of England (4th ed) Vol 25 at 185 [333]. These views are consistent with the views of Palles CB in Andrews v The Patriotic Assurance Co of Ireland (1886) 18 LR I 355 at 365 and Brett LJ in Castellain v Preston (1881-83) LR 11 QBD 380 at 386-389. The view that compliance with the terms of the insurance contract was sufficient to give rise to a right to use the insured's name by the insurer was expressed by Sir Boyd Merriman in Morley v Moore [1936] 2 KB 259 at 365. See also Welford's Accident Insurance (2nd ed) at 329. This would now appear to be the preferred view in England: Lord Napier and Ettrick v Hunter [1993] AC 713. 307 The question is one which might arise at two points. The first is whether the insurer can require the name of the insured to be used in an action against a third party. If an under-insured insured refused to allow his name to be used against a third party by an insurer who had paid in full on its policy, on the hypothesis that the first view is correct, a court would not force him to lend his name because the right of subrogation had not become exercisable. In those circumstances, however, if the insured did not itself take action against the third party (and, perhaps, run it as well as the insurer would have) the insured would probably be prejudicing an existing (though not fully crystallised) right of subrogation (which exists from the time of the commencement of the policy and which by the payment of money had accrued into a valuable equitable right) and accordingly he may be liable to compensate the insurer for any loss thereby suffered. The second point at which the question might arise is the destination of recovery funds from any action brought by the insured or insurer against the third party. If an insured had a house worth $100,000, which was insured for $50,000, which latter sum was paid by the insurer and an action was successful against the person responsible for the loss of the house, though less than $100,000 was recovered, what, in those circumstances, would be the appropriate division of the recovery funds? The view preferred by MacGillivray & Parkington on Insurance Law (8th ed) at 490-491 [1180]-[1181] and also Derham op. cit. at 53 is that the right to subrogation (in the sense of receipt of part or all of the proceeds of action against the third party) becomes enforceable upon performance of the contractual terms by the insurer, but that until the insured has been fully indemnified for this total loss he remains entitled to control the litigation as "dominus litis". See Sutton K Insurance Law in Australia (2nd ed) at 882-889 and Kelly and Ball Principles of Insurance Law (LexisNexis Butterworths) [9.0090]. See also SGIO (Qld) v Brisbane Stevedoring Pty Ltd (1969) 123 CLR 228 at 249-241; Santos Ltd v American Home Assurance Co (1987) 4 ANZIC 60-795 at 74,874-75; Morganite Ceramic Fibres Pty Ltd v Sola Basic Australia Ltd (1987) 11 NSWLR 189; and Arthur Barnett Ltd v National Insurance Co of New Zealand [1965] NZLR 874 at 882 and 885-886. 308 It is unnecessary to resolve these issues or to deal with the important judgment of the House of Lords in Lord Napier v Hunter about the rights of a reinsured (and perhaps an insured) to the proceeds of a subrogated action in competition with a reinsurer (and perhaps insurer) which has paid in full under a policy. What is, however, revealed by this discussion (whoever is entitled to start and control litigation and in whatever order the funds should be distributed) is the entitlement or interest of the insured in proceeding with a claim concerning which rights of subrogation exist until its loss is the subject of full indemnification or reparation. It is the same notion of full indemnification or reparation that destroys the right to claim indemnity under another policy of insurance and so discharges the later insurer. The right to indemnity for a loss under a second policy is not lost by election, by proceeding against another (or first) policy, if that other (or first) policy is an inadequate fund to make good the whole loss. The loss of the right to seek indemnity arises from prior full indemnification of the insured's loss. 309 Another approach to the third precondition which would see attachment occur, even if the above be incorrect, is to give the operation of the words "after [HIH] … have been held to be liable to pay the full amount" a timeless construction such that if at some time (that is before payment by AMPG) it could have been found liable that would suffice. Given the view expressed above it is unnecessary to resolve this issue. 310 Thus, HIH can still be held liable, and, on this hypothesis, has been held to be liable.

(d) The so-called "excess point" and the operation of exclusions 1(e) and 1(o) and clause 7 311 These submissions were put by Gordian, not CGU. 312 The primary judge concluded that no claim had been made against BHE within the policy period, but that a claim was deemed to have been made within the policy period by operation of condition 4 of the HIH policy: the notification and deemed claim provision. His Honour concluded at [870] of his reasons that the notification was of the acts, errors or omissions giving rise to the sand loss then manifesting itself. 313 There was no notification, however, of any act, error or omission in respect of pre-contract representations by BHE to FAC or any act, error or omission on the part of Airplan-GHD, a joint venture design consultant: see [871] of the primary judge's reasons. Also, at the trial, BHE abandoned a case that the acts, errors or omissions giving rise to accelerated corrosion of the metal components used in the walls were the subject of notification: see [1229]-[1235] of the primary judge's reasons. 314 Gordian submitted that the expert evidence established that the sand loss and accelerated corrosion affected the walls independently and in different ways. 315 Gordian submitted that the solution to the sand loss problem was either one based on long term maintenance (filling the holes left by sand loss with coarser sand) or a hybrid solution based partly on a grout curtain wall and maintenance of the kind identified above. Both solutions, it was said, would maintain the stability of the walls. The long term maintenance solution could be effected, it was said the evidence revealed, for under $20 million. The evidence did not reveal a cost of the hybrid solution. 316 Thus, Gordian submitted, since the notified claim could be remedied in these ways, the indemnity was limited to these solutions and did not extend to the significantly more expensive gravity wall solution agreed between BHE and SACL. The precise cost of this solution was found by the primary judge to be over $65 million, a sum sufficient to exhaust the Gordian and CGU policies even taking into account the AMPG payment. The gravity wall solution cured not only the sand loss problem, but also the corrosion problem. It was said that the cost of the solution did not thus "arise out of" the claim notified. It was therefore said that the gravity wall solution was not, and could not be, a reasonable settlement of the notified claim. Also, it was submitted that the gravity wall solution could not be a reasonable settlement at all, unless it was proved that there was an excessive corrosion problem along the entire length of the wall. The primary judge did not decide this issue on the basis that the accelerated corrosion claim was abandoned: [1234] and [1235] of his Honour's reasons. 317 The primary judge:

(a) accepted BHE's submissions as to quantum and that the gravity wall solution would cost over $65 million if used over the entirety of the walls: [1127]-[1147];

(b) found the settlement was reasonable, being for all claims, some identified, some not: [1160]-[1164];

(c) accepted that BHE bore the onus of showing that it was a reasonable settlement of the indemnified claims: [1166] and [1167]; and

(d) found that an agreement to use the gravity wall solution using grouted columns was a reasonable solution to the sand loss claim and thus formed the basis of a reasonable settlement: [1170] and [1171]. 318 Gordian complained that the last finding was without reasons and in the teeth of evidence of the existence of the solution by way of long term maintenance and the hybrid solution. It also asserted that the primary judge had misdirected himself about its submissions and confused them with the issues in the AMPG case. 319 I do not think that it is necessary to unravel the submissions in the two cases (under the HIH and following policies and under the AMPG policy). It is tolerably clear on the material that, if one works on the hypothesis that it was reasonable for BHE to settle with SACL on a solution which would not require maintenance, the solution accepted in the settlement was reasonable. 320 Gordian submitted that the cost of any cure beyond that of the long term maintenance or hybrid solutions was a cost that:

(a) for the purposes of exclusion 1(e) was a "legal liability … imposed … pursuant to the Law of Contract … when such legal liability would not have been imposed … pursuant to the Law of Tort";

(b) for the purposes of exclusion 1(o), could be seen as "arising out of the provision by the Insured of any express warranty or guarantee which increases the Insured's measure of liability …" ; or

(c) for the purposes of clause 7, "relates to any contractual or other duty or obligation assumed by the Insured which is not assumed in the normal conduct of the Insured's profession" . 321 Gordian also submitted that the evidence before the primary judge was inadequate to conclude that the only way of preventing sand loss on a "maintenance free" basis was to construct at least a grout curtain wall. During the trial, there was a debate as to whether it was technically viable to build a grout curtain which used the existing steel straps. BHE said that this would cause a separate corrosion cell. BHE said that the only solution was a gravity wall solution, which encased the metal straps in concrete and did not rely upon the strength of the remaining lengths of straps. This was more expensive. It was submitted by Gordian that the evidence about this was inconclusive and so BHE had not proved the reasonableness of the preferred solution. 322 I do not accept any of these submissions of Gordian. Taking the last matter first, the evidence of Dr Abel and Dr Sussex was not of a "theoretical" problem, using that term, as Gordian did, to connote not real or substantial. Rather, they both expressed the view that the solution should not be one that had the real capacity to introduce another corrosion problem, this time into the solution to rectify the problems. The lack of precise calculation of the problem is not to the point. Dr Abel said in cross-examination when asked about the lack of calculation: I did not and nobody attempted to make such a calculation. The possibility of a corrosion cell developing is highly on the agenda. No one would dare to approach this figuratively how severe would be this corrosion and what would be the long term effect. 323 There was ample basis to conclude that the taking of a risk by a method that could well lead to such a problem was unreasonable and to avoid that risk was reasonable. 324 It is also important to appreciate that what BHE was required to prove was that the settlement with SACL was reasonable upon the information available at the time of settlement, provided that BHE had acted reasonably in collecting information: Unity Insurance Brokers Pty Ltd v Rocco Pezzano Pty Ltd (1998) 193 CLR 603 at 609 [7], 616-618 [36]-[40] and 653 [130]. 325 On the basis of the evidence before BHE at the time of the adoption of the gravity wall solution, I do not think that it is open to challenge the conclusion that it was a reasonable settlement. First Mr Price the solicitor who acted for BHE at the time, who was not cross-examined on his advice, recommended the solution, in part because of a perceived risk of collapse of the walls with continuing sand loss and the necessity for the work, even putting aside the corrosion issue. No cross-examination of him was directed to any asserted unreasonableness of the solution, to the problem of sand loss or that it would not have been adopted but for the accelerated corrosion claim, or to any theoretical corrosion cell issue. 326 Secondly, the sequence of events contained in the evidence of Mr Price included the fact that the grout curtain method actually used was the preferred solution of a conclave of experts. After this expression of view by experts, HIH confirmed that it considered the settlement on the basis of the gravity wall solution to be reasonable. 327 A complete statement of the primary judge's reasons as to why the settlement was reasonable was set out at [1159] and [1160]. No error is there revealed. 328 I also reject the submission that the design requirement of the work that the reinforced earth walls be maintenance free for 50 years was a warranty or guarantee or amounted to assumed contractual liability. This requirement was an inherent part of the design. The requirement was in the Specification, not the conditions of contract dealing with defects or warranties. Clause E1 ("General Requirements") of Section E of the Specification dealing with Marine Works provides: E1.1 General The work in this portion of the project is for the detailed design and construction of the formation of the reclaimed land on which the new runway and its associated features are sited, for the formation of the Mill Stream outlet diversion channel, for the seawall that contains the formations and for the sea wall along the eastern edge of the existing runway. E1.2 Essential Design Parameters The various elements to be constructed under this Contract shall have a maintenance free design life of 50 years, except that the wall panels and reinforced strips in the main reinforced earth sea walls shall have a maintenance free design life of 100 years. 329 The liability of BHE arises by reference to the task it undertook. The task is in part defined by these design parameters. I see no basis for the engagement of exclusions 1(e) or 1(o) or clause 7. (e) The lack of separate treatment of RE P/L and CW 330 CGU submitted that the primary judge erred in failing to separate RE P/L and CW and in relation to CW in failing to find that it had no case to answer. 331 The primary judge explained early in his reasons at [62] how he proposed to deal with them, as follows: The approach generally taken in the judgment as a matter of convenience is not to draw any material distinction between the respective positions of Reinforced Earth and Connell Wagner [often referred to as "RE/CW"]. It is common ground that during most of the hearing almost no attention was given to any material difference in the positions of these two companies each of which had entered into subcontracts with BHE. The matter was raised with leading counsel for BHE [transcript 2893] where it was accepted that it seemed appropriate to refer throughout to "Reinforced Earth/ Connell Wagner". I may not have universally succeeded in adopting that precise practice, but although differentiating between the two companies when it is obviously appropriate to do so [as where documents have been written or received by one or the other], have intended to do so. I now make plain as an internal dictionary, that by and large the sundry references to RE are also to be regarded as shorthand for RE/CW without distinguishing between them. The parties will be given leave at the same time as addressing on short minutes of order and costs, to address on whether there is perceived to be any particular difficulty arising from this internal dictionary exercise, and if so, it may become necessary to expressly differentiate between these companies for some particular special reason. If this ends up requiring some additional findings, so be it. I would doubt that there will be any such problem. 332 Given how the trial was conducted, it is not now appropriate to seek to differentiate the two, especially when even if such differentiation were clearly shown, there would be no effect on the reasoning in the case.

Orders 333 For the above reasons, I would dismiss the appeal with costs. 334 BEAZLEY JA: I agree with Allsop P. 335 CAMPBELL JA: I agree with Allsop P. **********

17/10/2008 - The repeating sentences from computer error removed. - Paragraph(s) 1

DISCLAIMER - Every effort has been made to comply with suppression orders or statutory provisions prohibiting publication that may apply to this judgment or decision. The onus remains on any person using material in the judgment or decision to ensure that the intended use of that material does not breach any such order or provision. Further enquiries may be directed to the Registry of the Court or Tribunal in which it was generated.