Boral Resources (NSW) Pty Ltd v Gangi

BASTEN JA: I agree with Leeming JA.

GLEESON JA: I agree with Leeming JA.

LEEMING JA: At premises owned by the appellant and cross-respondent (Boral) in Artarmon, Sydney, which was formerly a quarry, a concrete batching plant has been in operation for more than 50 years. The quarry excavations had left a cliff approximately 15 metres high adjoining an entrance to the site from Carlotta Street. Many years ago, a steel structure was erected on the surface of the former quarry floor, with a platform at approximately street level and connected to the entrance by a ramp. On the platform, and supported by the steel structure, were three trays, into which sand, aggregate and gravel could be unloaded by trucks. Beneath each of those trays were conical steel bins, each capable of containing some 150 tonnes of materials. The plant was said to be "gravity fed" - by which was meant that sand, aggregate and gravel were unloaded directly into the trays from a truck, or else physically swept or otherwise pushed into the bins from the trays on the platform.

As originally designed, the contents of those bins were weighed and fed into an internal mixer beneath the bins before being transferred to trucks. At a time not identified in the evidence (but many decades ago), the mixer was removed, and instead the contents of the bins, together with cement and water, were weighed and transferred directly into cement agitator trucks which could drive underneath the structure.

There were no precise plans of the structure in evidence. One matter for resolution by the primary judge was how the three bins had been supported, and a great deal of evidence, including competing expert opinion evidence, was adduced on this issue.

On 14 December 2007, the respondent and cross-appellant, Mr Salvatore Gangi, had parked his truck under the structure for the purpose of its being loaded. The bins and the structures below them collapsed, without warning. Fortunately, Mr Gangi was in the cabin of his truck, which was propelled from the collapsing structure, and he sustained relatively minor physical injuries. However, it was common ground that he suffered psychological injury as a result of the collapse. Boral was found liable for negligence and a judgment of some $396,000 was entered against it: Gangi v Boral Resources (NSW) Pty Ltd [2013] NSWSC 569 (Schmidt J).

Boral appeals from the findings on liability. It has no fewer than 14 grounds. Speaking generally, Boral maintains that her Honour erred in finding that it was in breach, given that it had a regular system of maintenance, which proved to be ineffective by reason of the unusual design of the plant, something which was not known to it. Boral's appeal on liability extends, by ground 14, to a challenge of her Honour's finding of causation. By ground 13, Boral also submits that there is error in the process by which the primary judge drew inferences from its failure to tender a forensic engineer's report it had obtained after the accident.

In oral address, senior counsel for Boral confirmed that he had two "substantial complaints" about the finding of liability. The first was that it was said that there had been a deal of impermissible reasoning by way of hindsight. The second was said to be a misunderstanding or misapplication of the rule in Jones v Dunkel. The first was said to infect many of the grounds of appeal; the second was ground 13.

Further, Boral challenges, by ground 15 of its notice of appeal, one aspect of her Honour's findings as to damages.

By his cross-appeal, Mr Gangi challenges (grounds 1-6) other aspects of her Honour's findings of damages. He also challenges (ground 7) the costs order, which, although favourable to him, required Boral to pay only 82% of his costs.

For the reasons which follow, I propose that Boral's appeal be dismissed, as should Mr Gangi's cross-appeal on damages. However, although I would reject the majority of Mr Gangi's submissions on costs, I conclude that error has been demonstrated and the costs discretion must be re-exercised.

The trial was run on the basis that the appellant Boral acquired the plant in late 1982, following a takeover by Boral Limited of BMI Limited, of which Boral Limited was a 30% shareholder. The plaintiff at all times made it clear that he did not assert that Boral Resources (NSW) Pty Ltd had actual knowledge of how the plant had been designed. This was expressed in submissions as being a case of "constructive knowledge", although I prefer to use the statutory language of s 5B(1)(a) of the Civil Liability Act 2002 (NSW) and refer to a risk of harm of which Boral did not in fact know, but ought to have known, by reason of what its directors, officers, employees and agents ought to have known.

The trial was conducted by both sides with a focus on practical considerations as opposed to the precise legal relations which applied. On the one hand, the plaintiff was content to proceed on the basis that Boral had no knowledge prior to 1982. Consistently with this, the primary judge said that: "The plant was a steel structure built in a old quarry. It was a gravity fed plant, some 50 years old in December 2007. Boral acquired it in late 1982 when it purchased the remaining 70% of the shares from its former owner. It then took over operation of the plant as a going concern."

Those sentences are worded loosely. What seems to have happened is that Boral Ltd acquired the remaining shares of BMI Ltd, which became its wholly owned subsidiary, and which may well have changed its named to Boral Resources (NSW) Pty Ltd. However, nothing turns on this. There was no suggestion in the evidence that the appellant (however named) should be treated as having any knowledge of the design by reason of its directors, officers or agents being involved in the original erection of the structure.

On the other hand, as far as the plaintiff's position was concerned, the primary judge recorded that it emerged at the hearing that Mr Gangi was a director and shareholder of Gangi Holdings Pty Ltd: at [3]. This was capable of giving rise to a partial defence on the part of Boral pursuant to s 151Z of the Workers Compensation Act 1987 (NSW). However, although pleaded, this was abandoned during the course of the hearing.

There can be considerable difficulties in applying ss 5B-5E of the Civil Liability Act when there is a high degree of uncertainty as to why and how the plaintiff suffered harm. This proceeding is an example. However, and as required by those sections, her Honour resolved a large factual dispute as to how the plant had been constructed in the 1950s and the physical cause of its collapse in 2007. That enabled those sections to be applied.

On the evidence, Boral had no plans in its possession disclosing how the structure had been constructed. Indeed, even at trial, after subpoenas had been issued, there was no plan showing how the plant had been constructed. This led to there being a large issue at trial, and competing hypotheses, as to how had been constructed and how it collapsed.

Boral had employed, as its maintenance manager until he retired in 2010, Mr Russell, who had looked after some 14 Boral sites in Sydney and had been involved in designing, constructing and demolishing old concrete plants. He had trade qualifications in fitting and machining, heat treatment and metallurgy, but no tertiary qualifications. His evidence was that prior to 2007, he was of the view that every concrete plant in Sydney had its bins supported by what he called a "ring beam" which securely supported the bins. He said: "You would put up stanchions, posts if you like, in a vertical position, and then there would be a beam run around the top of those six or eight stanchions, depending on the size of the bins, and the bins would then be lowered in one or two sections and they would also have a beam which sat on top of that bin so that it was physically impossible for the bin to fall down through the centre. In other words, the top of the bin structurally was wider than what the stanchions were."

It was not disputed that Mr Russell believed that the bins at Artarmon were supported in that manner at all times prior to 2007. His evidence was that "in any bin that I've been involved in construction, that has a ring beam welded to it continuously around the outside and would sit on top of the vertical stanchions".

However, it was common ground at the trial that there had not been a ring beam supporting the bins at Artarmon. There was photographic and testimonial evidence obtained shortly after collapse, of bolt holes in metal beams and sheared bolts. It is not necessary to address the detail of that evidence. The photographic evidence is difficult for a layperson to interpret, but the testimonial evidence was clear: there were structural bolts supporting the bins.

The primary judge found that the bins were connected to the support structure by a combination of welding and bolts: at [145]. That finding was not challenged, and was supported by the evidence from two Boral witnesses about seeing bolts cleanly sawn off when they attended after the accident. In reaching that finding, her Honour had regard to the expert evidence that it was unconventional, but possible, to construct a bin support structure in that way. Her Honour rejected the competing hypothesis that the bins were held in place solely by bolts, in light of the fact that the structure had survived for 50 years, and the evidence about the maintenance required by such bolts.

It was common ground that the bins were exposed to load shocks when they were filled with sand, aggregate and gravel, and that the plant operated in a caustic environment.

Her Honour's findings about how the bins collapsed followed from her findings about how the bins were supported: at [146] (emphasis added): "On the evidence the failure of this support structure was caused by the connections between the bins and the support structure becoming too weak to continue carrying the weight of the bins and the materials they held. That was the result of none of these connections ever having been inspected or maintained, as they needed to be. While they were either not visible, or difficult to see, they were, however, constantly exposed to the caustic environment in which the plant operated, the result of the interaction of sand, air and water. They were also subjected to considerable ongoing load shocks. This environment was the reason for the otherwise thorough system of inspection and maintenance which Boral operated at the plant. It was a system which should have extended to these connections. Had it been, the bins would not have collapsed as they did."

There was no challenge to the finding as to how the structure collapsed. No submissions in terms were addressed in terms to the concluding two sentences, although consistently with the general attack on impermissible reasoning based on hindsight, I proceed on the basis that, to the extent that they amount to conclusions of breach of duty of care and causation, they fall within those challenged by Boral's appeal.

Boral sought to establish that it had a regular plan of inspection and maintenance at the site. Mr Russell was directly involved. He gave evidence that, in the course of his regular inspections, he would go inside the bins. What could be seen from within the bin (namely a flat plate) was precisely what Mr Russell expected: nothing revealing how the bins were supported. Accordingly (and appropriately), his evidence was that he "assumed that the bin was attached in the same method as any other bin I had been involved in". The language of "assumption" was understandable, for his evidence was prepared and adduced at a time when he was acutely aware of the fact that his assumption had been false.

Mr Gangi was critical, at trial and on appeal, of Boral's system of inspection and maintenance. It is not necessary to say anything of those matters (which were disputed). It is sufficient for present purposes to observe that because of the assumption that the bins were supported by a ring beam, Boral's process of inspection did not extend to the bolts (of which it was unaware) which contributed to supporting the bins, and for the same reason, they were not maintained.

On one occasion there was an opportunity for Mr Russell's assumption to have been tested. In 1991 or 1992 (which year was not reliably established), there was a major refurbishment of the Artarmon plant. Asbestos had been found in the cladding walls around the structure. Experts were brought in to remove the asbestos, taking around a week to do so. Mr Russell said: "the decision was made at that stage that we would sandblast the entire plant and repaint it in a manner which would last for innumerable years."

Mr Russell was personally involved in that process. However, even so, with the cladding off, Mr Russell did not appreciate that the bins were not supported by a ring beam.

The primary judge summarised Mr Russell's evidence on this point at [110] as follows: "Mr Russell explained that in a traditionally constructed plant, it is a ring beam which supports the weight of the bins, which is visible from several angles at the top of the bins. Mr Russell did not see such a beam when the cladding was stripped off this plant in 1990. What he then saw was a beam half a metre from the top of the bin wall. He agreed that he knew from his visual inspection at that time, that this beam was not supporting the bins. He then assumed that there was another horizontal beam welded to the bin, inside the beam, which could not be seen. After the collapse he learned that there were in fact three beams there."

I confess not fully to understand the three beam structure which Mr Russell was describing. There was difficulty at the trial on this point, so much so that it led to two things happening: a call for the photographs taken of the wreckage of the structure after the collapse, and an offer from Boral's counsel that "I intend to re-examine the witness and to ask him to draw this configuration". Boral claimed privilege in response to the call, and did not re-examine Mr Russell as had been indicated (to be fair, Mr Gangi's counsel did not remind him to do so; I am not to be taken to be suggesting that the omission was other than inadvertent).

However, a full understanding of how Mr Russell believed (after the event) the bins were supported is not necessary in order to resolve this appeal. What matters for present purposes is what Mr Russell said he did, and could have done, in 1991 or 1992 when the cladding had been removed.

Mr Russell gave evidence that the absence of a ring-beam, and the fact that the bins were supported by bolts, could have been detected by using mirrors: "We suggested mirrors to look underneath and look up. They didn't use that method. That would have been the only way to see it. Everything that had an external face to it, and a horizontal face was sandblasted and painted, including the bins."

However, the cross-examiner later caused him to qualify his statement that mirrors would have been the only way to detect the bolts. Mr Russell's evidence included the following: "Q. You didn't think that [the top beam] was the ring beam that was the support for the bin, did you? A. No, I believed that was support for the bin and I believed that was welded to the inside of the bin. Q. You didn't inspect that and see that? A. Didn't inspect it closely, no. Q. So contractors got up there and sandblasted that particular area? A. Yes. Q. But you never got up there and had a look? A. No. Q. I'm not being critical; that is because you assumed everything was in order? A. Correct. Q. A matter I want to raise with you is this: Had you got up there and had a look at the time, you would have been able to see whether or not the bin was welded to the ring beam, correct? A. I can't answer that because I don't know what was there. Q. But the opportunity was there to get up and have a look? A. The opportunity was there, yes."

Mr Russell also confirmed that it would have been possible after the new (non-asbestos) cladding had been put on, to remove a few sheets near where the bins were supported and examine the method of support.

The most concise way of resolving Boral's appeal on liability is first to identify the essential evidentiary matters which supported her Honour's dispositive findings. What follows is by no means exhaustive of the evidence adduced throughout a 12 day trial, but will be sufficient.

Mr Russell readily (and understandably) accepted that, by 2007, he knew the following five matters. First, that the plant at Artarmon was the second oldest Boral concrete plant in Sydney. Secondly, that it had been up and running for 50 years. Thirdly, that it was unique, in the sense that it was the only one whose design included asbestos cladding walls around the steel structure. Fourthly, that the steel structure of a batch plant needed maintenance to protect it from the adverse effects of corrosion. Fifthly, that the operator ought to have annual visual inspection of the support structure for the bins and the means by which the bins were attached to the support structure.

Moreover, there was undisputed expert evidence that a reasonably prudent concrete batch plant operator ought have a good knowledge of all structural aspects of the plant. It is sufficient to reproduce the following evidence given concurrently by the parties' experts: "WITNESS APPLEYARD: It is a difficult question in the way it is put, your Honour. ... My view is that a reasonably prudent concrete batch plant operator ought to have had at least a good practical working knowledge of the condition of all aspects of the plant. HER HONOUR: Yes. What do you say, Mr Taylor? WITNESS TAYLOR: I wouldn't disagree with that. I would agree with that."

There was uncontroversial evidence that Boral, had it been aware that there were structural elements of its plant that ought to have been regularly inspected and maintained, would have done so.

Finally, there was Mr Russell's evidence of what could be seen in 1991 or 1992 when the asbestos cladding was removed. Her Honour found at [128] that: "when the cladding was removed from the plant in 1990, it appears that what could then be seen of the bin support structure suggested it may not have been the conventional one. The ring beam which Mr Russell explained could ordinarily be seen to support the underneath edge of the top of the bins in such a structure, was either not visible or very difficult to see."

Her Honour first summarised her reasons (at [26]-[29]) and then, elaborately and extensively, addressed the evidence and made findings which underlay that summary (a course which carried with it the advantage that the reader knew, upfront, the relevance of the evidence).

That the evidence extracted above supported the essential reasons of her Honour may very readily be seen. Her Honour found (by way of summary) at [27]-[29]: "What the evidence did establish was that the plant was not constructed in the standard way in which such plants are generally constructed, where the bins which collapsed are supported in such a way that they could not physically collapse as they did at this plant. Over the course of its 25 years of operation of the plant, Boral assumed that they were supported in the standard way. The bins were in fact supported in such a way that a physical collapse was possible. The actual bin supports were vulnerable to corrosion and were never maintained. In the result, while Boral had safety and maintenance systems constantly in operation at the plant, it never investigated how the bins were supported and never maintained the supports. Eventually they corroded to the point where they could no longer support the weight of the bins, with the result that in December 2007, they collapsed to the ground, injuring Mr Gangi. The evidence showed that while this risk was not foreseen, it ought to have been, and that if it had been, Boral would have taken available precautions, which would have prevented the risk from materialising. In the circumstances a reasonable person in Boral's position would have known of the risk, because it would have taken available steps to ensure that it knew of the way in which the bins were supported; and it would have maintained those supports, in accordance with the systems which it operated at the plant. Had it done so, the supports would not have corroded to the point that the bins collapsed."

To be clear, the summary extracted above would not, if it stood alone, satisfy what is required by ss 5B-5E to be found. However, that reasoning did not stand alone. It was, expressly, a summary, which rested on the more detailed findings which followed, as her Honour addressed the evidence in the case.

Her Honour's factual findings occupied [31]-[146]. Her Honour then addressed what reasonable care required (at [147]-[158]) and the requirements of s 5B at [159]-[161], and causation under s 5D at [162]-[174].

In addressing those issues, her Honour made findings at [126], [130] and [178] which it is convenient to reproduce immediately: "Mr Russell's evidence was that he had always approached his work on the basis of an assumption that at this plant, like conventionally constructed plants, it was impossible for the bins to collapse as they did, given their construction. While the experts agreed that this was a reasonable assumption for him to have made, they also considered that a reasonably prudent operator, having operated the plant for some 25 years, ought to have had a good practical working knowledge of the condition of all aspects of the plant, as well as knowing what the structural supports for the bins were. ... Boral had no drawings of the plant. It was then apparent that there was a difficulty in seeing the bin support structure. That meant that reasonable prudence required that an inspection be undertaken, engaging an engineer if necessary, to determine whether the assumption which had been made as to the construction of the bin support structure was sound. Such an inspection was not undertaken. Had it been, there is no question that the unusual support structure would then have been revealed and that afterwards, the bin supports would have received necessary maintenance, which would have prevented the collapse which occurred in 2007. ... On the expert evidence, it is apparent that if an engineer had been engaged by Boral to investigate how the plant had been constructed, prior to the collapse, its design and the risk of collapse to which the plant succumbed, would have been uncovered. On Mr Russell's evidence, no such investigation was ever undertaken, before the collapse. It was, however, undertaken after the collapse."

The foregoing will permit me relatively concisely to explain why I do not accept the submissions advanced by Boral on liability.

One recurring theme in Boral's submissions was the complaint that her Honour had impermissibly used hindsight reasoning in order to identify the risk of harm. As it was put: "It is not permissible to use hindsight and to say, well it's all very well to continue to operate day in day out but you could have done this. There needs to be a reason, as the case prospectively unfolded, which causes someone reasonably to stop and ask the question of themselves."

That submission is uncontroversial. It may be acknowledged that there are some passages in her Honour's reasons which express conclusions which anticipate the legal reasoning to follow; I have mentioned the last two sentences of [146]. There are others. Boral emphasised her Honour's summary, at [27]-[29], but those reasons were, expressly, anticipatory of the findings and reasoning to follow.

However, contrary to Boral's submissions, a fair reading of the reasons of the whole does not disclose impermissible hindsight reasoning. The events of 1991 or 1992 were, quite plainly, critical to her Honour's reasoning process. At that time, Mr Russell knew that the plant was old, and was the only one in his experience that had asbestos cladding. Further, the ring beam supporting structure that he had expected to see was not located where he had expected it to be. It was in those circumstances correct to conclude, in accordance with the expert evidence, that Boral should have taken steps to determine how the bins were supported.

Her Honour's findings at [130] (reproduced above) disclose no impermissible use of hindsight. The only element of that reasoning which might support a complaint of impermissible hindsight is what reasonable prudence required in 1991 or 1992, when it was not clear how the bins were supported. But the undisputed expert evidence referred to above amply supports her Honour's findings. After the asbestos cladding had been removed in 1991 or 1992, and it had become obvious to Mr Russell that the bin support structure was different from what he had hitherto assumed, it was clear that Boral did not have a good practical working knowledge of all aspects of its plant.

That is the short answer to the first "substantial complaint" advanced by Boral on liability. I turn to the second.

After the accident, Boral retained Gardiner Willis & Associates Pty Ltd, trading as GW Engineers, to conduct a "forensic investigation" in conjunction with the demolition of the remains of the structures. GW Engineers was also to provide project management services for the demolition of the structure. When inviting tenders for the demolition work, GW Engineers stated that there would be a process of: "identification of the structure's components, the taking of photos, physical tagging of the structure's components, equipment required, laid down areas and storage. Preservation of evidence is vital. Components deemed by the Forensic Engineer to be of significance will be photographed in place (where possible) prior to removal ... Once identified, components shall be kept on site in a designated laid down area in sequential order for further examination as required. Components of the structure shall only be disposed of upon the Forensic Engineer's written approval and instruction."

The photographs taken during this process were not provided by Boral to its expert Mr Taylor. Boral resisted producing them, claiming they were privileged (the investigation occurred in the context of an investigation by WorkCover, from which no prosecution resulted).

Her Honour described the position as follows (at [185]-[187]): "In the result, the experts who gave evidence in these proceedings were denied access to relevant information in Boral's hands, which would have shed considerable light not only on how the plant was actually constructed, but what in fact caused the collapse."

Her Honour formed the view that an adverse inference should be drawn against Boral, not from the fact that it claimed legal professional privilege in the photographs in the report, but by reason of its failure to rely on information in its possession. The gravamen of her Honour's concern was that the photographs and report might assist determining the factual questions as to why the structure had collapsed. Her Honour concluded that "it must be inferred that the information in Boral's hands as to what caused the plant to collapse, would not have assisted its defence of Mr Gangi's claim": at [194]. Boral submitted that that process of reasoning disclosed error.

A threshold issue confronting this ground of appeal is that her Honour, expressly, regarded the adverse inference as merely corroborative and not determinative. At the conclusion of her consideration of this issue, the primary judge said: "Mr Gangi's liability case was supported by these conclusions, but did not depend upon them. I have already concluded that the evidence otherwise establishes the case he advanced as to Boral's liability."

Boral said that notwithstanding that express disclaimer, it was to be inferred from other statements throughout the judgment that the failure to provide full information to its expert was used by her Honour impermissibly. In particular, when her Honour summarised her conclusions as to breach (at [23]-[29]), Boral pointed to the fact that her Honour expressly relied upon the failure to lead evidence, without any qualifications.

For present purposes, the critical paragraph of her Honour's reasons is [26], which immediately precedes the summary to which I have referred above: "I am satisfied that these submissions cannot be accepted. In summary, the evidence as to the configuration of the plant showed that the bins were located high above the ground, carrying up to 450 tonnes of materials, which were gravity fed into the trucks parked below, underneath the plant. The bins were not supported from below. Their means of support were neither visible nor known to Boral before the collapse. It did not lead evidence as to the result of the investigations which it undertook after the collapse, which would have shed considerable light on how they were supported and what caused the bins to collapse."

There is force in Boral's submission that there is a tension between the unqualified reliance upon Boral's failure to adduce evidence of its own investigations after the collapse in this paragraph rejecting Boral's case, and the expressly qualified reliance upon that failure when her Honour dealt in detail with this issue.

However, not lightly ought it to be concluded that her Honour did not mean what she wrote at [195], expressly discounting reliance upon the failure by Boral to adduce evidence of its investigations. It is a familiar thing for judges expressly to make part of their reasons dispositive, or alternatively to make part of their reasons non-dispositive. That is an important part of the judicial function. It has a variety of consequences. One, applicable to reasons on questions of law, is to alter the precedential status of the reasoning, in the manner explained by Devlin J in Behrens v Bertram Mills Circus Ltd [1957] 2 QB 1 at 24: "It is well established that if a judge gives two reasons for his decision, both are binding. ... [T]he practice of making judicial observations obiter is also well established. A judge may often give additional reasons for his decision without wishing to make them part of the ratio decidendi; he may not be sufficiently convinced of their cogency as to want them to have the full authority of precedent, and yet may wish to state them so that those who later may have the duty of investigating the same point will start with some guidance." Sir Rupert Cross stated that those remarks represented orthodox judicial theory: see now R Cross and J Harris, Precedent in English Law, Clarendon Press Oxford, 4th ed 1991, p 42 (the passage occurs in earlier editions).

Another consequence is that a judge may choose narrowly to base his or her decision upon particular findings, whose effect may be to alter the materiality of other findings or the way in which appellate review is to be conducted. One example of an immensely experienced trial judge (Sheppard J) finding a witness' evidence to be neither honest nor reliable, but expressly not basing that finding upon his demeanour, may be seen in Masterton Homes Pty Ltd v LED Builders Pty Ltd [1996] FCA 171; 33 IPR 417.

I acknowledge that there would be much greater force in Boral's complaint if her Honour's reasons had been delivered ex tempore. If, say, there were repeated and unqualified references in an ex tempore judgment to Boral's failure to tender the report, but the judgment concluded with a statement that that was not material to her Honour's finding, then it would at the least be necessary to reconcile the seeming inconsistency.

But the fact of the matter is that her Honour reserved, and prepared a written judgment, of considerable length (appropriately so, having regard to the length of the trial and the importance and complexity of the issues). The judgment includes a table of contents, and, as I have noted above, a (helpful) summary which anticipates the findings of fact and legal reasoning to follow. It is plain that the reasons (which occupy 101 pages) were drafted over some time, and were subsequently reviewed and revised and proof-read within her Honour's chambers. In those circumstances, powerful considerations would be required to discount her Honour's express finding that she did not need to, and did not, rely on any inference drawn from Boral's maintenance of a claim of privilege.

I do not consider that Boral has demonstrated that her Honour's statement at [195] should not be taken at face value. Boral's submission to the contrary focussed on the summary at the outset of the reasons. Paragraph [26] of which Boral complains within the summary is descriptive of the evidentiary position, and explicative of the fact that a great deal of the judgment to follow would deal with the competing views as to how the bins had been supported. However, it is the three following paragraphs which identify, by way of summary, the essential themes of her Honour's reasons.

In light of the prominence accorded to the report at trial, it was appropriate that her Honour address the submissions based on it. It is not surprising that she did so over 21 paragraphs (from [175]-[195]). However, her Honour's dispositive findings were based on the actual evidence before the Court, rather than inferences based on what was not before the Court.

In short, I see no reason to disregard the clear language of her Honour at [195], reproduced above, to the effect that the inferences drawn from the failure by Boral to provide the primary materials on which the report was based, or the forensic engineer's report itself, were not dispositive. There is no sound basis to proceed on the basis that the dispositive and non-dispositive portions of her Honour's reasons were other than as she had indicated.

Grounds 1-12 of Boral's notice of appeal, which were directed to breach, ranged beyond the two "substantial complaints" which I have addressed above. These grounds were addressed collectively in written and oral submissions, and it is convenient to resolve them in the same manner, and by subject matter, rather than by reference to the particular ground.

One of Boral's submissions (ground 3) was that this was a latent defect, such that no one within Boral ought to have known of its existence. It was said "One cannot have constructive knowledge of something unless it is reasonable to suppose that steps should have been taken to determine the presence of the defect". However, as noted above, it was common ground between the experts that Boral ought to have known the structural elements of the plant it operated. That is a complete answer to this submission.

Boral advanced an elaborate submission directed to whether the risk of harm was foreseeable (written submissions, paragraphs 69-82). But once again, it is sufficient to observe that Boral knew, no later than 1991 and 1992, that this plant had been differently constructed from any others of which Mr Russell was aware. There was no error in concluding that Boral's duty of care required that Mr Russell's assumption should then have been examined afresh.

Boral submitted there was error in her Honour's reasoning as to s 5B(1)(b) and (c). In relation to whether the risk of harm was not insignificant, Boral submitted that it was not established that it ought to have known as at 14 December 2007 that there was a not insignificant risk of the plant suddenly collapsing. There is nothing in this submission. In 2007, Boral knew that the structural elements of its plant needed regular inspection and maintenance, and it ought to have known what those structural elements were. In relation to the reasonable precautions to be taken required by s 5B(1)(c), it is sufficient to observe that a program of inspection and maintenance of all structural elements, including the bolts in shear which supported the bins, was a reasonable precaution. As her Honour said, given the extensive maintenance otherwise carried out on the plant, it is clear that its extension to the bolts was a reasonable precaution.

Boral submitted that the primary judge "did not address, or did not adequately address" the relevant issues arising under s 5D (written submissions, para 101). The first limb of this submission indicates much about the way in which an argument ran at trial was sought to be converted into a submission on appeal without regard to the need to identify error in her Honour's reasons. The fact of the matter is that her Honour addressed causation at length at [162]-[174].

In its written submissions, Boral said that her Honour failed to identify how the failure to inspect, or carry out maintenance, would have prevented the collapse on 14 December 2007. As elaborated orally, Boral submitted: "But on close examination ... her Honour didn't address the critical portion of this case which was what were the material facts and circumstances which should have engaged Boral in an inquiry about how the bins were supported."

The submission is not made out. The essence of her Honour's reasoning was that Boral was ignorant of the way in which its plant had been constructed, that it should have taken steps after 1991 or 1992 to determine how it was supported when it became clear that Mr Russell's assumptions could not confidently be relied upon, and that had it done so, it would have identified additional structural elements which had not hitherto been the subject of regular inspection or maintenance.

That reasoning was correct. It was at all times accepted that Boral would undertake any necessary maintenance identified following an inspection, and her Honour so found at [173]. The only questions therefore were whether there was an occasion to identify the unusual structural elements of the plant, and whether they would have been identified on that occasion. Her Honour found at [169] that: "What had been seen when the cladding walls were removed in the early 1990s, called into question that the bins were constructed and supported in the conventional way that Boral had assumed that they were. Its assumptions as to the bin support structure should then have been investigated."

Her Honour also found that the nature of the structure could have been identified by a structural engineer, using mirrors, ultrasound or x-rays: at [171].

In oral submissions, Boral emphasised that when Mr Russell examined the structure in 1991, he did not see anything remarkable. "The effect of his evidence was that he did not at that point notice anything which created, relevantly a red flag." Boral accepted that it was necessary to challenge the finding in the first sentence of [169], and said in substance that that did not reflect Mr Russell's evidence. I acknowledge that there are difficulties in understanding aspects of Mr Russell's verbal description of what seems to be a complex three-beam structure. However, the portions reproduced above sustain her Honour's findings, which are further supported by the consideration that Mr Russell knew after 1991 or 1992 that the structure was unique in that it was constructed with asbestos.

Boral knew that all structural elements needed to be maintained. It had no knowledge that the bins were not supported by bolts. It made an assumption (through Mr Russell) that they were not supported by bolts. However, that assumption was called into question by a combination of facts: (a) the plant was old, (b) it was the only plant built with asbestos cladding, (c) when the cladding was removed, it was not possible to see the assumed ring-beam, and indeed as her Honour put it, "what could then be seen appears to have departed from what could be seen in a conventionally constructed plant": at [129].

Boral advanced one other submission, which was capable of engaging either or both of ss 5B(2) and 5D. It was said, upon the assumption that the bolts had been detected in 1991, that: "we know as a fact that it lasted for a further 17 years. So the question is, what maintenance would have been required and would that maintenance necessarily have prevented this incident?"

There is nothing in that submission, once it is appreciated that Boral would have maintained any structural elements in the plant had it been aware of their existence and the need to do so. There is no reason to consider that there is any error in her Honour's finding that a regular system of maintenance would have prevented the collapse in 2007.

It is perhaps convenient to step back from the detail of the precise grounds of appeal, and return to the summary given by her Honour and the essentially uncontroverted evidence from Mr Russell and the parties' experts.

An operator of a concrete batching plant supporting hundreds of tonnes of raw materials above areas where contractors would drive, whose structural elements were steel, and which functioned in a caustic environment subject to sudden loadings, is required by the law of negligence as modified by the Civil Liability Act to regularly inspect and maintain all structural elements. In order to do that, it was necessary for the operator first to identify what the structural elements were. When once the operator knew that this plant, which it had acquired rather than building itself, was old and had features which it had never before encountered, then it became insufficient to discharge its obligations merely to rely upon untested assumptions as to what was structural and what was not, even though those assumptions may have been sufficient to amount to a reasonably prudent course of conduct at other plants. Finally, where ultimately the plant collapsed by reason of the absence of inspection and maintenance, factual causation is obvious, and there can be no real question that the scope of liability should extend to the harm directly and immediately caused by the collapse.

Ground 15 of Boral's appeal challenged one aspect of the damages awarded for economic loss. It was to the effect that having found that Mr Gangi's psychological injury, although chronic, did not prevent him from working and required only limited further treatment, there was appellable error in awarding him damages for economic loss beyond a closed period. The parties were agreed as to Mr Gangi's entitlement to damages for economic loss from the date of injury until he employed a relief driver (around the beginning of June 2009).

The primary judge addressed the issues of past loss of earning capacity and future economic loss at [309]-[331]. Her Honour found that in the period in contest (from June 2009) Mr Gangi's earning capacity was initially reduced by two days a week, declining to one day per week, and that, for the future, there would continue to be "some limited ongoing impact on his capacity to work, which I assess at one day per week": at [331]. (The materials available on appeal do not readily permit an analysis of the amount of this aspect of the claim; the judgment in Mr Gangi's favour was in an agreed undissected amount of $396,290.)

Boral made no separate oral submissions on this ground of appeal, but relied upon [107]-[108] of its written submissions, which in turn emphasised the evidence of its witness Dr Julian Parmegiani, whose most recent reports were dated 20 March 2012. Dr Parmegiani had diagnosed a post-traumatic stress disorder when he had assessed Mr Gangi in March 2009, and had expressed the hope that Mr Gangi's symptoms should improve significantly over the next 18 months to 2 years. However, by March 2012, Dr Parmegiani was of the view that this had not occurred and that his chronic post-traumatic stress disorder was perpetuated. He expressed the view that Mr Gangi's psychiatric injuries had stabilised, and that "no significant improvement is expected in the foreseeable future". However, he expressed the view that "[Mr Gangi's] psychiatric symptoms would not preclude a return to full-time gainful employment on suitable duties". In a report evaluating Mr Gangi's psychiatric impairment in accordance with the WorkCover guides for the evaluation of permanent impairment, also dated 20 March 2012, Dr Parmegiani said "Mr Gangi could work full-time, if pain allowed him to do so. He would however find it difficult to work in situations which remind him of the accident of 14 December 2007".

The primary judge was faced with the difficulty that there were important aspects of Mr Gangi's claim which she did not accept, partly by reference to the competing medical evidence, partly by reference to her adverse demeanour-based findings of Mr Gangi, and partly because of surveillance evidence showing Mr Gangi performing his ordinary work and engaging in social activities seemingly without discomfort. That said, her Honour adopted a measured approach. At [329] her Honour framed the question as whether Mr Gangi had established that his future capacity to work was less than full-time. Her Honour noted that the evidence supported his ability to work, extending even to driving work, as well as work utilising other skills such as those attained by him from his TAFE studies and his trade as an electrician. Her Honour then said (at [330]-[331]): "Weighing up the risks and chances which the evidence reveals in this case, necessarily means, however, that the ongoing impact of his chronic psychological illness and its impact on his physical condition, may not be ignored. In the result I have concluded, on balance, that in future there will be some limited ongoing impact on his capacity to work, which I assess at one day per week."

Boral's complaint was that the "expert opinion of Dr Julian Parmegiani, consultant psychiatrist, appears to have been largely ignored". Dr Parmegiani's evidence was wholly ignored in Boral's written submissions on economic loss supplied to the primary judge, and his evidence was mentioned briefly (in paragraph 3(ix)) only to be discounted ("it is submitted the doctor's comments at p 76 suggest he did not appreciate the true nature and extent of the plaintiff's alleged complaints at the time his earlier case was proceeding"); that submission was directed to a finding that Mr Gangi was being deliberately non-responsive when giving his evidence. Likewise, in oral address, Boral took her Honour to the evidence of Dr Harvey-Sutton and Dr Rimmer (transcript 19 February 2013 pp 753-756). There was no mention of Dr Parmegiani in relation to economic loss.

Notwithstanding the way in which the case at first instance had been put by Boral, her Honour did, when dealing with the question of economic loss, refer in terms to the necessity of considering Dr Parmegiani's evidence that Mr Gangi could have performed more work: see at [318].

The primary judge concluded that the undoubted ongoing psychiatric injury caused by the incident would have an impact on Mr Gangi's capacity for work. As the primary judge recognised, by reference to State of New South Wales v Moss [2000] NSWCA 133; 54 NSWLR 536 at [71], that inquiry involves "the process of estimation of possibilities" and is "an imprecise and indeterminate one to be carried out within very broad parameters". As a Full Court said in Callaghan v William C Lynch Pty Ltd [1962] NSWR 871 at 877, the trier of fact may have to form conclusions on "slender materials".

In performing that task, there was no error in failing to refer in any more detail than her Honour did to the evidence of Dr Parmegiani.

Grounds 1-6 of Mr Gangi's cross-appeal challenge most of the primary findings of fact which are relevant to the assessment of damages. The grounds are as follows: "1. Her Honour erred in failing to find that the respondent (hereinafter "the plaintiff") had ongoing physical disability arising from the subject accident. 2. Having heard evidence from both doctors, her Honour erred in failing to make a finding preferring the opinion of one of Dr Harvey-Sutton or Dr Rimmer as to the ongoing effect of the plaintiff's physical injuries. 3. Her Honour erred in that her allowance for non-economic loss was manifestly inadequate. 4. Her Honour erred in finding that the plaintiff had not been exercising his full capacity for work from 30 June 2009 to the date of hearing. 5. Her Honour erred in finding that the plaintiff did not have ongoing physical disability precluding him from full-time work. 6. Her Honour erred in finding that the plaintiff had the residual capacity to work five out of six days a week."

This Court faces an obvious difficulty in dealing with those grounds. All save ground 2 were on their face informed by her Honour's unfavourable assessment of Mr Gangi's credibility. For example, although Mr Gangi's case was that his ongoing pain prevented him from working more than three days a week, her Honour rejected that claim: "I am not satisfied that it has been established that in 2013 Mr Gangi continues to experience ongoing pain of such severity, that it prevents him from working at all, on more than three days a week. That conclusion is simply not consistent with much of the medical evidence to which I have referred, or with what the surveillance tapes reveal."

In written and oral submissions, Mr Gangi complained that the primary judge had failed to take account of the fact that the chronic soft-tissue injuries sustained by him had resulted in the development of chronic pain disorder with the need for a pain management program and cognitive behavioural therapy to deal with ongoing chronic psychological disability. It was said, for example, that there was objective evidence (including mild wasting in the right shoulder joint and thickened subdeltoid bursa with impingement of the bursa: Blue 908E, 806N, 1388R) sustaining a finding of ongoing physical injury. As much may be accepted, but (what appear to be relatively minor) ongoing physical manifestations of the injury do not translate into appellable error in the assessment of damages. Moreover, even if there were error, it is impossible for this Court to make appropriate findings without having seen Mr Gangi give evidence.

Ground 2 is in a different category, but presents a similar difficulty. Doctors Harvey-Sutton and Rimmer gave evidence before her Honour. Dr Harvey-Sutton gave concurrent evidence with Dr Nakhle, while Dr Rimmer (an orthopaedic surgeon) gave individual evidence and was cross-examined on the seventh day of the trial. This Court cannot resolve which of those doctors' opinions is to be preferred.

In short, each of grounds 1-6 cannot be determined by this Court.

The submissions advanced by Mr Gangi (and those in opposition from Boral) were (considerably) more elaborate than is indicated by the foregoing. However, there is no utility in reproducing them or expressing views about them. The essential and insuperable problem facing all of these grounds was squarely raised with counsel for Mr Gangi during the hearing: "Basten JA: Doesn't that highlight the problem for us? All of these medical experts that you're relying on gave evidence, the plaintiff gave evidence, the trial judge assessed all these witnesses, what do we do on the paper with this?"

Counsel for Mr Gangi, after taking instructions, advised the Court as follows: "We have reflected on your Honour Basten JA's conundrum that you put to me, which is, does this mean there should be a new trial? The answer is, we don't seek that. And if that means that this is a good point that fails, so be it, because I can't put the plaintiff through another 12 days of hearing."

In those circumstances, it is neither necessary nor appropriate to express a view on whether error has been made out on grounds 1-6 of the cross-appeal. It is sufficient to observe that any error identified could not be resolved by this Court. That, coupled with the (entirely understandable) concession made by Mr Gangi, means that these grounds of the cross-appeal must be dismissed.

By a separate reserved judgment: [2013] NSWSC 1388, her Honour resolved a suite of arguments concerning costs. First, her Honour exercised a discretion under Uniform Civil Procedure Rules 2005 (NSW) r 42.34 to make a favourable costs order, notwithstanding that, as it turned out, the litigation could have proceeded in the District Court. No challenge is made to that exercise of discretion. Secondly, her Honour formed the view that the plaintiff should have an order for the costs thrown away by the adjournment of the 2012 hearing: at [26]-[29]. Again, no challenge was made to that conclusion. Thirdly, her Honour rejected Mr Gangi's application for an order for indemnity costs in his favour: at [30]-[37]. Mr Gangi challenges this, on two bases: by reference to a Calderbank letter preceding the trial, and by reason of claimed error by her Honour rejecting the submissions that Boral should have known that it had no chance of success on liability, and had conducted its defence for some ulterior motive or because of some wilful disregard of the known facts. Fourthly, her Honour made a special costs order, depriving Mr Gangi of some (18%) of his costs, relying on the fact that he had failed to make out a significant part of his case on damages as a result of the adverse credit findings she had made: at [38]-[43].

The order ultimately made was "Boral is to bear 82% of Mr Gangi's costs of the proceedings as agreed or assessed". No point was taken in the appeal as to the differential treatment accorded to two aspects of the costs of the proceedings: the costs thrown away by reason of the adjournment, and the balance of the costs. It is not necessary to determine whether any process of construction permits the order to be construed as an order that the balance of Mr Gangi's costs (that is, excluding the costs thrown away by reason of the adjournment) were to be reduced by 18%, while he was entitled to an unreduced order in relation to costs thrown away by reason of the adjournment (which is what is consistent with the reasons at [29] and [43]). For the reasons to be given below, the costs discretion should be re-exercised.

By letter dated 28 November 2011, expressed to be "Without Prejudice Save as to Costs", the plaintiff's then solicitor noted that Boral denied the existence of a duty of care and denied that such a duty was breached, and said that the factual circumstance of this claim gave rise to a prima facie inference of negligence on the part of the defendant. She said that "it cannot be seriously be in issue that the plaintiff is entitled to the verdict as against the defendant". She sought an admission of breach of duty of care, and said that if such an admission were not given, and the plaintiff succeeds in establishing duty and breach at trial, he would seek his indemnity costs of proving those issues. The letter also referred to the expert's report which had not been served and which was the subject of a claim for privilege.

Her Honour did not expressly deal with the submission made on appeal that the letter warranted a special costs order. (The submissions made to the primary judge on costs were not included in the appeal materials, and so it is not clear whether this submission was made to her Honour in terms.) However, contrary to the submission made by Mr Gangi on appeal, the letter does not engage the principles in Calderbank v Calderbank [1975] 3 All ER 333.

The offer was made at a time when the plaintiff advanced a "rolled-up" statement of claim, which did not even make a separate allegation of duty, nor mention any provision of the Civil Liability Act 2002. Moreover, it was made at a time, as was reflected in its terms, when the plaintiff was advancing a claim of "Res Ipsa Loquitor [sic]", which was abandoned by his amended statement of claim. Boral's amended defence did admit that it owed the plaintiff a duty of care, although different from that pleaded by the plaintiff; that defence was understandable since the ultimate pleading that went to trial was the duty of an occupier to an entrant to "avoid foreseeable risk of injury", which is not the law: cf Bunnings Group Ltd v Borg [2014] NSWCA 240 at [50].

In those circumstances, it was not unreasonable for Boral to reject the offer. It follows that although her Honour failed separately to deal with the letter, nothing turned on that fact, even if the submission was in fact made.

The reasons given for rejecting the other basis for an order for indemnity costs were as follows: "Given the dispute between the experts, discussed at length in the principal judgment, it cannot be concluded that this is a case where Boral, properly advised, should have known that it had no chance of success on liability, notwithstanding the adverse inference that I drew in relation to its failure to rely on the expert advice it obtained after the collapse of the plant. Nor has it been established that it conducted its defence, for some ulterior motive, or because of some wilful disregard of the known facts. Mr Gangi accepted that privilege was available to be claimed over the expert advice Boral did not tender. What it contained is not known. While I concluded that an inference could be drawn that the information which so came to Boral's hands after the collapse, over which it claimed privilege, would not have assisted its case, that its defence involved 'wilful disregard of the known facts' has not been established."

Her Honour was exercising discretionary power conferred by s 98 of the Civil Procedure Act 2005 (NSW). It was for Mr Gangi to identify House v The King error in the exercise of that discretion. It was put that her Honour erred by mistaking the facts, and that her Honour should have concluded either that Boral, properly advised, should have known that it had no chance of success on liability, or that it had acted for an ulterior motive or because of some wilful disregard of the known facts.

I would reject that submission. First, its acceptance would carry with it the unpalatable consequence that Boral's appeal as to liability must also be one which has been improperly maintained (with the consequence that it must be dismissed, and with an indemnity costs order).

Secondly, Boral's amended defence was certified by its solicitor to the effect that there were "reasonable grounds for believing on the basis of provable facts and a reasonably arguable view of the law" that its defence had reasonable prospects of success. Although it was said that that certification could remain accurate and yet the factual matters as to Boral's knowledge and purpose for which Mr Gangi contended made out, I respectfully disagree.

Thirdly, Boral had retained an expert who, although not briefed with all the material available to Boral, nevertheless expressed opinions consistent with the absence of breach of duty.

Finally, it may be added that the principles in Briginshaw v Briginshaw and s 140 of the Evidence Act applied to the findings her Honour was asked to make.

For similar reasons, it was not established that Boral defended the proceedings for some ulterior motive or in wilful disregard of the known facts. That conclusion could not be reached by reason merely of the concededly proper claim of privilege maintained by Boral over the report obtained by it after the accident. No error has been shown by her Honour's declining to find wilful disregard of the known facts.

Mr Gangi had originally sought damages of some $188,000 for past economic loss and almost $1.2 million for future economic loss, together with other substantial heads of damage. By the close of the trial, the claim had reduced to just less than $1.1 million. Parts of that claim were rejected by the primary judge. In particular, her Honour was not satisfied that a claim for future domestic assistance or care had been made out. He ultimately obtained a judgment of $396,290.27, slightly more than 1/3 of the amount pressed by the conclusion of the trial. In that context, her Honour reasoned as follows: "In assessing that percentage, consideration has to be given to the claims advanced, the measure of success which Boral had in resisting that aspect of Mr Gangi's case and what I have said as to its own approach to the proceedings and how that unnecessarily increased costs. Having taken the evidence led on costs into account, in undertaking the exercise discussed in James v Surf Road Nominees Pty Limited [No 2], including Mr Davidson's unchallenged estimation that the hearing on damages occupied some 24% of the hearing, I have concluded that Mr Gangi's costs should be reduced by 18%. That reflects that there was no issue as to the psychological injury which he had suffered or that he had suffered various physical injuries. It was the ongoing physical injury after a certain time and into the future, on which Mr Gangi's case was not entirely accepted. This reduction in my assessment fairly reflects the impact of that failure on the costs which he should, in justice, receive in these proceedings."

Mr Gangi's primary submission was that it was not open, in a case such as this where the "issue" was damages and where a substantial verdict had been ordered, to reduce the percentage of costs awarded. That submission is contrary to principle, statute and authority.

First, Mr Gangi's submission ultimately proceeds on the basis that the undoubted discretion to make costs orders in relation to particular issues does not extend to either components of heads of damages, or the factual issues underlying those components (such as whether in truth there were ongoing physical impairments). There is no reason to circumscribe the discretion in the way Mr Gangi contends.

Section 56(5) of the Civil Procedure Act in terms empowers the Court to "take into account any failure to comply with subsection (3) or (4) in exercising a discretion with respect to costs". Those subsections refer to the obligation upon parties, their barristers and solicitors and any other person with a relevant interest in the proceedings to assist the court to further the "overriding purpose", which is to "facilitate the just, quick and cheap resolution of the real issues in the proceedings". A party who, for example, lies about an aspect of his or her claim, and thereby causes expense and time, is not furthering the overriding purpose. Her Honour expressly (at [34]) acknowledged that earlier statements as to the exercise of the discretion as to costs "must now be considered in the context of the provisions made in ss 56-60 of the Civil Procedure Act". The obligation in s 56 is not confined, contrary to what was submitted on appeal, to procedural matters such as compliance with court directions. There is no reason to give a narrow meaning to s 56 and there is every reason to give it a broad meaning; it is after all the overriding purpose.

Secondly, Mr Gangi's submission is inconsistent with authority, both long-standing and recent, which reflects basal principle in this area. It is sufficient to turn to Maestrale v Aspite [2014] NSWCA 182, which was squarely on point, although neither party referred to it. That appeal was from a decision of the primary judge to award a successful plaintiff only 50% of his costs. Beazley P, with the agreement of Macfarlan and Barrett JJA, stated the applicable legal principles as follows. First, the "general rule" in UCPR r 42.1 is subject to it appearing to the court that "some other order should be made as to the whole or any part of the costs": at [78]. Secondly, a successful party may be deprived of costs if guilty of conduct that makes it appear to the court that some other order should be made, including where the successful party "unnecessarily protracts the proceedings": Oshlack v Richmond River Council [1998] HCA 11; 193 CLR 72 at 97. Thirdly, in Arian v Nguyen [2001] NSWCA 5, the Court of Appeal approved the reasoning of a District Court judge who found that the serious dishonesty of the successful party in the conduct of the trial justified the displacement of the usual costs order. Fourthly, her Honour said: "In Apostolidis v Kalenik (No 2) [2011] VSCA 329, the trial judge made findings of serious dishonesty against both the appellant and respondent. The trial judge ordered that the successful respondent be deprived of 25 per cent of the costs of the trial on the basis that she had engaged in improper conduct, told numerous lies, dishonestly exaggerated her case, had failed altogether on some issues; and her own conduct as a witness had wasted court time."

Although the Victorian legislation is in different terms, it is plain that her Honour regarded the same principle as applicable under the generally worded New South Wales Act and Rules.

Her Honour's reasoning in Maestrale as summarised above is, with respect, plainly correct. (The Court proceeded to set aside the costs orders made by the primary judge, but not because of any error in principle.)

In short, it is open to a judge to subtract from the percentage of costs recovered where there has been found to be non-compliance with s 56. It seems clear that a deal of time at trial was occupied by Boral exposing the exaggerated claims made by Mr Gangi. There is no error in this aspect of her Honour's reasons.

However, Mr Gangi also pointed to three further difficulties in her Honour's reasons. He said that it was wrong for her Honour to state that Mr Gangi "should not be rewarded for" his exaggeration of his symptoms in an attempt to mislead medical examiners and the court, because costs orders do not reward a party, but merely provide a partial indemnity for expenses incurred.

There is a threshold question, on a fair reading of her Honour's reasons, as to whether her Honour had fully accepted that submission advanced by Boral. In circumstances where her Honour summarised the position at [8]-[11], essentially by reference to the partial failure of Mr Gangi's case on damages, expressly referring to the fact that costs are not awarded to punish an unsuccessful party, and the proposition that "where a successful party has failed on matters which took up a significant part of the trial, it may be appropriate to deprive that party of a costs order in relation to those matters", her Honour's reasons should not be read as based on an erroneous approach of denying a reward and in effect punishing Mr Gangi for those failures.

However, there is substance to other aspects of Mr Gangi's challenge to the exercise of discretion as to costs.

It was common ground that, by reference to pages of transcript, questions of damages occupied 24% of the hearing time. It by no means follows that 24% of Mr Gangi's costs of the proceedings related to damages and 76% to liability. Pages of transcript are an imperfect proxy for time in court, and, more importantly, time in court is an imperfect proxy for costs incurred. Some aspects of a proceeding are relatively cheap to prepare but take a deal of time in court; others the opposite. Mr Gangi submitted that there was error in using hearing time as a proxy for proportion of a party's costs.

Moreover, the effect of an 18% reduction in Mr Gangi's costs was for Mr Gangi to receive the benefit of an ordinary costs order in respect of the costs of liability (treated under this hypothesis as being 76% of the costs of proceedings), but only a small fraction (25%) of the (much smaller) costs of quantum. (If that is not immediately obvious - it is a consequence of 18% being ¾ of 24%. Let it be supposed that, to take a round number which bears no relation to this case, recoverable costs were $1 million, and therefore recoverable costs of liability were $760,000 and recoverable costs of quantum $240,000. The order that Boral pay 82% of Mr Gangi's costs translates to an order that it pay all of his recoverable costs of liability but only ¼ of his recoverable costs of quantum.)

It is far from clear that her Honour intended such a drastic reduction. If in fact her Honour had appreciated that she was distinguishing issues of liability and quantum, and ordering that the successful plaintiff not obtain 75% of his costs on quantum, then more extreme language than "Mr Gangi failed to make out a significant part of his case on damages" would be expected to be found in [41].

In summary, there is error in relation to the costs of the adjournment. More importantly, there is error in the way in which the percentage of 18% has been calculated, and I would infer on a fair reading of the reasons that her Honour did not intend for Mr Gangi to be deprived of ¾ of what he would otherwise be entitled to recover of his costs insofar as they were attributable to the hearing directed to quantum.

This ground of the cross-appeal should be allowed. I propose that the costs discretion be re-exercised, such that Mr Gangi is to be paid (a) the costs thrown away by reason of the adjournment, and (b) a proportion of his costs of the balance of the proceedings. As for the latter, there should be some deduction for the time and cost wasted by reason of Mr Gangi's exaggerated claims, accepting that most of the trial was occupied by questions of liability, but also that most of Mr Gangi's claims for damages were refused. A broad brush approach should be applied to what is largely a matter of impression and evaluation: see the authorities summarised in Doppstadt Australia Pty Ltd v Lovick & Son Developments Pty Ltd (No 2) [2014] NSWCA 219 at [18]-[19]. On that basis, Boral should be ordered to pay the whole of Mr Gangi's costs thrown away by reason of the adjournment, and 90% of the balance of Mr Gangi's costs at first instance.

Accordingly, I propose that the appeal be dismissed, ground 7 of the cross-appeal be allowed, but the cross-appeal otherwise dismissed, the costs order made on 20 September 2013 be set aside, and in lieu thereof Boral be ordered to pay (a) Mr Gangi's costs thrown away by reason of the adjournment, and (b) 90% of the balance of Mr Gangi's costs of the proceedings.

Prima facie, Boral must pay Mr Gangi's costs of its appeal. Mr Gangi has enjoyed some limited success on his cross-appeal. An order reflecting that limited success is that Boral pay 25% of Mr Gangi's costs of the cross-appeal. 13 March 2019 - [59] - "at 25" changed to "at 24" 13 March 2019 - [13] - "later" changed to "late" [59] - quotation, third sentence, "decisions" changed to "decision"