Court of Appeal Supreme Court New South Wales Medium Neutral Citation: Curtis v Harden Shire Council [2014] NSWCA 314 Hearing dates: 25 September 2013 Decision date: 10 September 2014 Before: Bathurst CJ at [1]; Beazley P at [58]; Basten JA at [230] Decision: In each of the matters CA 2012/241825 and CA 2012/241832: 1. Appeal allowed. 2. Set aside the verdict for the respondent made by Fullerton J on 9 July 2012. 3. Remit the matter to the Common Law Division of the Supreme Court for the assessment of damages. 4. Order the respondent pay the appellant's costs of the liability hearing at first instance and of the appeal. [Note: The Uniform Civil Procedure Rules 2005 provide (Rule 36.11) that unless the Court otherwise orders, a judgment or order is taken to be entered when it is recorded in the Court's computerised court record system. Setting aside and variation of judgments or orders is dealt with by Rules 36.15, 36.16, 36.17 and 36.18. Parties should in particular note the time limit of fourteen days in Rule 36.16.] Catchwords: TORTS - negligence - duty of care - breach of duty - whether Civil Liability Act 2002 (NSW), s 43A applied - whether placing certain traffic control signs and omitting other signage involved the exercise of a "special statutory power" - prohibition on any person installing prescribed traffic control devices combined with requirement for statutory authority to undertake such activity WORDS AND PHRASES - "special statutory power" - Civil Liability Act 2002 (NSW), s 43A TORTS - negligence - duty of care - breach of duty - standard of liability - Civil Liability Act 2002 (NSW), s 43A - motor vehicle accident - whether omission of signs on road was an act no authority could properly consider a reasonable exercise of its power - expert evidence - evidence of council officer - omission of "slippery road" sign despite direction in Traffic Control Plan - common sense TORTS - negligence - causation - onus - balance of probabilities - whether primary judge erred in comparison of probabilities with possibilities TORTS - negligence - causation - motor vehicle accident - whether absence of signage caused driver to lose control on first resurfaced section - balance of probabilities - no direct proof - whether circumstances give rise to a reasonable and definite inference - consideration of state of the road, circumstances of accident, expert evidence and competing hypotheses Legislation Cited: Civil Liability Act 2002 (NSW), ss 5D, 5E, 5T, 43A Compensation to Relatives Act 1897 (NSW) Criminal Procedure Act 1986 (NSW), s 214 Evidence Act 1995 (NSW), s 140 Judiciary Act 1903 (Cth), ss 56, 64 Road Transport (Safety and Traffic Management) (Road Rules) Regulation 1999 (NSW), cl 131 Road Transport (Safety and Traffic Management) Act 1999 (NSW), ss 50, 51, 52; Pt 4, Div 1 Road Transport Act 2013 (NSW), ss 121, 122, 123 Roads Act 1993 (NSW), s 87 Supreme Court Act 1970 (NSW), s 75A; Transport Administration Act 1988 (NSW), ss 45E, 52A; Pt 6 Uniform Civil Procedure Rules 2005 (NSW), Pt 31, r 23 Cases Cited: Allianz Australia Insurance Ltd v Roads and Traffic Authority of New South Wales [2010] NSWCA 328 Amalgamated Television Services Pty Ltd v Marsden [2002] NSWCA 419 Associated Provincial Picture Houses v Wednesbury Corporation [1948] 1 KB 223 Banque Bruxelles Lambert SA v Eagle Star Insurance Co Ltd [1997] AC 191 Bellingen Shire Council v Colavon Pty Ltd [2012] NSWCA 34 Bennett v Minister of Community Welfare [1992] HCA 27; 176 CLR 408 Betts v Whittingslowe [1945] HCA 31; 71 CLR 637 Board of Fire Commissioners (NSW) v Ardouin [1961] HCA 71; 109 CLR 105 Bolam v Friern Hospital Management Committee (1957) 1 WLR 582 Bradshaw v McEwans Pty Ltd [1951] HCA 480; 217 ALR 1 Briginshaw v Briginshaw [1938] HCA 34; 60 CLR 336 Brodie v Singleton Shire Council [2001] HCA 29; 206 CLR 512 Chester v Afshar [2005] 1 AC 134 Commissioner of Main Roads v Jones [2005] HCA 27; 79 ALJR 1104 Crimmins v Stevedoring Industry Finance Committee [1999] HCA 59; 200 CLR 1 Curtis v Harden Shire Council [2012] NSWSC 757 Duma v Mader International Pty Ltd [2013] VSCA 23 Evans v Queanbeyan City Council [2011] NSWCA 230 Flounders v Millar [2007] NSWCA 238; 49 MVR 53 Fox v Percy [2003] HCA 22; 214 CLR 118 Fraser v Burswood Resort (Management) Ltd [2014] WASCA 130 FTZK v Minister for Immigration and Border Protection [2014] HCA 26 Gorringe v Calderdale Metropolitan Borough Council [2004] 1 WLR 1057 Hargrave v Goldman [1963] HCA 56; 110 CLR 40 Helton v Allen [1940] HCA 20; 63 CLR 691 Jamal v Secretary, Department of Health (1988) 14 NSWLR 252 John Curtis v Harden Shire Council [2012] NSWSC 84 Jones v Dunkel [1959] HCA 8, 101 CLR 298 Kable v State of New South Wales [2012] NSWCA 243; 293 ALR 719 Kuhl v Zurich Financial Services Australia Ltd [2011] HCA 11; 243 CLR 361 Latoudis v Casey [1990] HCA 59; 170 CLR 534 Luxton v Vines [1952] HCA 19; 85 CLR 352 Malec v JC Hutton Pty Ltd [1990] HCA 20; 169 CLR 638 March v Stramare (E & M H) Pty Ltd [1991] HCA 12; 171 CLR 506 Minister for Aboriginal Affairs v Peko-Wallsend Ltd [1986] HCA 40; 162 CLR 24 Minister for Immigration and Multicultural Affairs v Eshetu [1999] HCA 21; 197 CLR 611 MM Constructions (Aust) Pty Limited and Anor v Port Stephens Council (No 3) [2010] NSWSC 243 Nader v Urban Transit Authority of NSW (1985) 2 NSWLR 501 Neat Holdings Pty Ltd v Karajan Holdings Pty Ltd [1992] HCA 66; 67 ALJR 170 Nguyen v Cosmopolitan Homes (NSW) Pty Ltd [2008] NSWCA 246 Presland v Hunter Area Health Service [2003] NSWSC 754 Puntoriero v Water Administration Ministerial Corporation [1999] HCA 45; 199 CLR 575 Qantas Airways Ltd v Gama [2008] FCAFC 69 Roads and Traffic Authority of New South Wales v Refrigerated Roadways Pty Ltd [2009] NSWCA 263; 77 NSWLR 360 Roads and Traffic Authority v Royal [2008] HCA 19; 82 ALJR 870 Rogers v Whitaker [1992] HCA 58; 175 CLR 479) Royal v Smurthwaite [2007] NSWCA 76; 47 MVR 401 Secretary of State for Education and Science v Tameside Metropolitan Borough Council [1977] AC 1014 Secretary of State for Education and Science v Tameside Municipal Borough Council (1977) AC 1014 Shoeys Pty Ltd v Allan [1991] Aust Torts Rep ¶81-104 Sneddon v State of New South Wales [2012] NSWCA 351 Stovin v Wise [1996] AC 923 Strong v Woolworths Ltd [2012] HCA 5; 246 CLR 182 Sutherland Shire Council v Heyman [1985] HCA 41; 157 CLR 424 Sydney South West Area Health Service v Stamoulis [2009] NSWCA 153 Vyner and Waldenburg Bros Ltd [1946] KB 50 Wallace v Kam [2013] HCA 19; 250 CLR 375 Warren v Coombes [1979] HCA 9; 142 CLR 531 Woolworths Ltd v Strong [2010] NSWCA 282 Texts Cited: Australian Standard 1742 - 2002 edition, Manual of uniform traffic control devices, Part 3 M Aronson, "Government Liability in Negligence" (2008) 32 Melb U L Rev 44 at 79 Review of the Law of Negligence, Final Report (Commonwealth of Australia 2002) at par 10.26 Category: Principal judgment Parties: John Curtis (Appellant) John Curtis on behalf of himself and Jasmine Patterson, Jacob Patterson, Angus Curtis and Emily Curtis (Appellants) Harden Shire Council (Respondent) Representation: Counsel: H M Marshall SC / R Taylor (Appellants) R Sheldon SC (Respondent) Solicitors: McCabe Partners (Appellants) DLA Piper (Respondent) File Number(s): 2012/241825; 2012/241832 Decision under appeal Citation: [2012] NSWSC 757 Date of Decision: 09 July 2012 Before: Fullerton J File Number(s): 2007/293512; 2007/293513 HEADNOTE [This headnote is not to be read as part of the judgment] On 20 August 2004, the appellant's partner, Ms Paterson, was fatally injured when the car she was driving ran off the road and hit a tree. At the time of the accident, the respondent, Harden Shire Council, was carrying out roadworks on the section of road where the accident occurred. The appellant brought proceedings for damages for psychiatric injury arising from Ms Paterson's death, and, on behalf of himself and her children, for damages under the Compensation to Relatives Act 1897 (NSW). The appellant claimed that the accident was caused by loose gravel on the road surface, and that the respondent was negligent in failing to provide adequate signage to indicate that the road had been resurfaced and that motorists should reduce their speed. The primary judge, Fullerton J, dismissed both claims: [2012] NSWSC 757. Her Honour found that the Council had breached its duty of care to Ms Paterson in failing to erect adequate signage. However, her Honour was not persuaded that this breach was the cause of Ms Paterson's accident. Her Honour also considered that the appellant did not establish a breach of duty to the standard required by the Civil Liability Act 2002 (NSW), s 43A. This section states that a public authority exercising (or failing to exercise) a special statutory power, within the meaning of that section, will not be liable unless the act or omission was in the circumstances so unreasonable that no authority having the special statutory power could properly consider the act or omission to be a reasonable exercise of, or failure to exercise, its power. On appeal to this Court, the appellant claimed that: (1) The primary judge erred in holding that the statutory defence provided by the Civil Liability Act, s 43A, applied to the circumstances of the case, and that the appellant had not established a breach of duty to the standard required by that section; (2) The primary judge erred in requiring the appellant to establish causation on a standard of proof that was higher than the balance of probabilities; (3) The primary judge erred in failing to find that the accident was caused by the Council's negligence: Civil Liability Act, s 5D. The Court allowed the appeal. In relation to (1): (Basten JA, Bathurst CJ and Beazley P agreeing) 1. Section 43A is engaged as the failure to provide adequate signage on the road involved the exercise of a special statutory power - the signs were a "prescribed traffic control device" under the Road Transport (Safety and Traffic Management) Act 1999 (NSW) which can only be installed by a public authority: [245]-[256] (Basten JA) Roads and Traffic Authority of New South Wales v Refrigerated Roadways Pty Ltd [2009] NSWCA 263; 77 NSWLR 360; Bellingen Shire Council v Colavon Pty Ltd [2012] NSWCA 34; 188 LGERA 169 distinguished. (Bathurst CJ, Beazley P and Basten JA) 2. The standard of liability once s 43A is engaged requires determining whether no authority could properly consider the act or omission to be a reasonable exercise of, or failure to exercise, its power. The emphasis is on what a public authority exercising that statutory power could properly consider reasonable - not what the court considers reasonable: [6] (Bathurst CJ); [224] (Beazley P); [277]-[279] (Basten JA) (Bathurst CJ, Beazley P and Basten JA) 3. The expert evidence established that the failure to include signage indicating that the road was slippery and that motorists should reduce their speed was a decision no body with the special statutory powers in question could reasonably have made in these circumstances: [12] (Bathurst CJ); [225]-[228] (Beazley P); [309]-[311] (Basten JA) In relation to (2): (Bathurst CJ, Basten JA) 1. No appealable error is disclosed in how the trial judge approached the issue of causation: [21] (Bathurst CJ), [335] (Basten JA). (Basten JA) 2. The comparison of probabilities with possibilities, the former satisfying the standard of proof while the latter do not, is entirely conventional and does not demonstrate a misunderstanding of the applicable burden of proof: [332]-[335] Nguyen v Cosmopolitan Homes [2008] NSWCA 246; Sydney South West Area Health Service v Stamoulis [2009] NSWCA 153 approved. (Beazley P) 3. The standard of proof of facts in respect of which a party bears the onus of proof requires the fact finder to compare the probability of the event having occurred against the probability that it did not occur: [177]-[178] Malec v JC Hutton Pty Ltd [1990] HCA 20; 169 CLR 638; Briginshaw v Briginshaw [1938] HCA 34; 60 CLR 336 (Beazley P, dissenting) 4. The trial judge erred in holding that, to establish causation, the appellant was required to prove that the probability that the accident was caused by the respondent's breach was greater than the possibility that it was caused by hypothetical alternatives unrelated to the respondent's breach. This posed a wrong comparison for determining whether a fact has been established on the balance of probabilities: [179]-[187] Bradshaw v McEwans Pty Ltd [1951] HCA 480; 217 ALR 1; Nguyen v Cosmopolitan Homes (NSW) Pty Ltd [2008] NSWCA 246 disapproved In relation to (3): (Beazley P) 1. A finding on causation can be made by the drawing of available inferences, provided that the relevant standard of proof is satisfied: [193] Betts v Whittingslowe [1945] HCA 31; 71 CLR 637; Roads and Traffic Authority v Royal [2008] HCA 19; 82 ALJR 870 (Bathurst CJ, Beazley P) 2. Having regard to the evidence, the trial judge erred in finding that causation was not established on the balance of probabilities: [51]-[53] (Bathurst CJ), [216] (Beazley P) (Basten JA, dissenting) 3. The trial judge was entitled on the evidence to find that causation had not been established: [362], [368] Judgment 1BATHURST CJ: I have had the advantage of reading the judgments of Beazley P and Basten JA in draft. Subject to what I have said below, I agree with the analysis of s 43A of the Civil Liability Act 2002 (NSW) (the Act) carried out by Basten JA and with his conclusions concerning its application in the present case. Section 43A of the Act 2There was no issue between the parties that the respondent owed a duty of care to road users in carrying out road maintenance work including the placement of appropriate warning signs. As Basten JA has indicated, in these circumstances s 43A of the Act sets the standard to be applied in fulfilment of that duty, where the issue in question is the exercise or non-exercise of special statutory powers. 3Section 43A of the Act provides as follows: "43A(1) This section applies to proceedings for civil liability to which this Part applies to the extent that the liability is based on a public or other authority's exercise of, or failure to exercise, a special statutory power conferred on the authority. (2) A special statutory power is a power: (a) that is conferred by or under a statute, and (b) that is of a kind that persons generally are not authorised to exercise without specific statutory authority. (3) For the purposes of any such proceedings, any act or omission involving an exercise of, or failure to exercise, a special statutory power does not give rise to civil liability unless the act or omission was in the circumstances so unreasonable that no authority having the special statutory power in question could properly consider the act or omission to be a reasonable exercise of, or failure to exercise, its power. (4) In the case of a special statutory power of a public or other authority to prohibit or regulate an activity, this section applies in addition to section 44." 4I agree with Basten JA that for the reasons given by him the work being carried out involved the exercise of a special statutory power within the meaning of that expression in the Act. 5As Basten JA has pointed out, the standard of care required by the section derives from the judgment of Lord Greene in Associated Provincial Picture Houses v Wednesbury Corporation [1948] 1 KB 223. The principle enunciated by Lord Greene in that case has caused difficulty in the field of administrative law and these difficulties are compounded when it falls for consideration in the context of tort liability. In one sense if any authority, having the powers in question and acting properly, could reasonably consider that an act or omission was appropriate, it is difficult to see how any liability in negligence could arise with respect to that act or omission, even if the relevant standard was that contained in s 5B of the Act. However, it is evident that the legislature was intending to set a different and lower standard. 6It does not, in these circumstances, seem to be necessary to paraphrase the section by reference to cases derived from the administrative law area, although the manner in which courts have approached the question in these cases may inform consideration of the application of s 43A of the Act. It is sufficient to say that what is important is that the standard is set by what a holder of the statutory power could properly consider reasonable. The Court must look at the matter having regard to what the authority in question could properly consider a reasonable exercise of the power. If the authority could properly consider what was done was a reasonable exercise of the power then there will be no liability. This is so even if the Court considering the matter independently of the section would have concluded there was a failure to fulfil the duty. (Cf in an entirely different context Bolam v Friern Hospital Management Committee (1957) 1 WLR 582 and Rogers v Whitaker [1992] HCA 58; (1992) 175 CLR 479). 7In the present case, in my opinion, the appellant has satisfied the onus that the standard laid down was not met. The Traffic Control at Work Sites manual (the manual) prepared by the Roads and Traffic Authority (RTA) from which the Traffic Control Plan (TCP) was derived, stated that the manual must be used on all RTA road sites by all local bodies undertaking work on behalf of the RTA. The manual sets out a number of TCPs. It states that the selection of the appropriate TCP must only be undertaken by qualified persons. Mr Stephenson, who prepared the TCP for the site in question, was so qualified. 8Section 4 of the manual deals with standard TCPs. Clause 4.4 permits minor modifications if a standard TCP does not exist for the work being planned. Clause 4.4 states that the modification power is not to be used to design new TCPs. The modifications suggested in cl 4.4 although not said to be exhaustive, do not include the removal of signs. 9Mr Stephenson selected TCP 56. Subject to one matter, this selection was entirely appropriate. TCP 56 was designed for roads with speed limits of 60 km per hour or less. Kingsvale Road, where the accident occurred, had a speed limit of 100 km per hour. TCP 56 directed use of TCP 353 when the speed limit exceeded 60 km per hour and the average daily traffic exceeded 5,000 vehicles. Kingsvale Road did not carry this volume of traffic. There was, therefore, some discretion left in the designer to determine whether speed warning signs should be erected or a reduced speed limit imposed. 10Critically, TCP 56 contained a "slippery road" sign with a direction it was to remain in place until loose aggregate was removed. The reason for this direction was clear from the evidence of Mr Coffey. Beazley P has summarised his evidence and I will not repeat it. However, the importance of a "slippery road" sign is self-evident from his comment that driving on loose gravel is like walking on marbles. 11Beazley P has also summarised the expert evidence. Ultimately each expert agreed that a "slippery road" sign was necessary with either a "reduce speed" sign or a lower speed limit. Mr Stephenson was not called to give evidence of why he modified TCP 56 to exclude the "slippery road" sign or a "reduce speed" sign, a decision which his superior, Mr Coffey, said made no sense. 12Having regard to this evidence, particularly the mandate in the manual and the statement by Mr Coffey that the decision to omit the signs made no sense, I am of the view that a failure to provide a speed warning sign and a "slippery road" sign was a decision nobody having the special statutory powers in question could reasonably have made in the circumstances. Causation 13Beazley P and Basten JA have set out the evidence concerning this issue in detail and I will only repeat that evidence insofar as is necessary for a proper understanding of this judgment. 14Sections 5D and 5E of the Act, so far as relevant, provide as follows: "5D(1) A determination that negligence caused particular harm comprises the following elements: (a) that the negligence was a necessary condition of the occurrence of the harm (factual causation), and (b) that it is appropriate for the scope of the negligent person's liability to extend to the harm so caused (scope of liability). (2) In determining in an exceptional case, in accordance with established principles, whether negligence that cannot be established as a necessary condition of the occurrence of harm should be accepted as establishing factual causation, the court is to consider (amongst other relevant things) whether or not and why responsibility for the harm should be imposed on the negligent party. ... 5E In proceedings relating to liability for negligence, the plaintiff always bears the onus of proving, on the balance of probabilities, any fact relevant to the issue of causation." 15The determination of factual causation involves nothing more than the application of the "but for test" for causation: Strong v Woolworths Limited [2012] HCA 5; (2012) 246 CLR 182 at [18] and Wallace v Kam [2013] HCA 19; (2013) 250 CLR 375 at [16]. In the present case this involves the determination of whether, but for the failure to erect a "slippery road" sign and either impose a lower speed limit or a "reduce speed" sign, the accident which resulted in the death of Ms Patterson (the deceased) would not have occurred. 16The appellant's case was that the absence of such signage caused the deceased to drive onto the first resurfaced section of the road at or slightly under the speed limit of 100 km per hour, lose control of her vehicle on the loose gravel on the resurfaced section and ultimately collide with a tree resulting in her death. There was no issue that if this was proved on the balance of probabilities, factual causation was established. Nor was it contended, subject to the matters raised under s 43A of the Act to which I have referred earlier, that it was inappropriate for the respondent's liability to extend to the accused in those circumstances. 17It is convenient to deal with the issues raised by reference to three subheadings. (a) Did the primary judge apply the correct standard of proof on the question of causation? 18As Beazley P has pointed out, the appellant was required to prove factual causation on the balance of probabilities. So much is clear from s 5E of the Act. In the passage from Bradshaw v McEwans Pty Ltd [1951] HCA 480; (1951) 217 ALR 1, cited by Beazley P in her judgment (par [180]), the High Court emphasised that where direct proof is not available it is enough that the circumstances give rise to a reasonable and definite inference. However, the evidence must do more than give rise to conflicting inferences of equal degree of probability so that the choice is a matter of conjecture: See also Luxton v Vines [1952] HCA 19; (1952) 85 CLR 352 at 358 and Jones v Dunkel [1959] HCA 8; (1959) 101 CLR 298 at 304-305. 19In the present case it was incumbent on the appellant to prove that the failure to erect the necessary signs was a necessary condition of the death of the deceased: Strong v Woolworths Limited supra at [44]. 20Dixon J in Briginshaw v Briginshaw [1938] HCA 34; (1938) 60 CLR 336 at 362 stated that reasonable satisfaction was not a state of mind attained or established independently of the nature and consequence of the fact to be proved. However, his Honour emphasised at 361-363 that what is enough is that the affirmant of the allegation is made out to the reasonable satisfaction of the tribunal, which may be based upon a preponderance of probability. 21The primary judge stated (Curtis v Harden Shire Council [2012] NSWSC 757 (the primary judgment) at [109]) that what was required was a finding that the probability that control was lost in the first section of the resurfaced roadwork, due to encountering gravel under speed was greater, than the possibility that the deceased veered progressively onto the incorrect side of the road for reasons unrelated with the condition of the road in combination with the speed at which she was travelling. However, the primary judge immediately thereafter made reference to Flounders v Millar [2007] NSWCA 238; (2007) 49 MVR 53 at [35], where Ipp JA set out the standard of proof in the same manner as in Bradshaw v McEwans Pty Ltd supra and the other cases to which I have referred. In that context her ultimate conclusion (primary judgment at [113]) that she was not persuaded that but for the failure to provide for and position appropriate signage the accident would not have occurred, was not based on any misapplication of the appropriate onus. (b) Legal principles on causation 22As the plurality pointed out in Strong v Woolworths Limited supra, a necessary condition is a condition that must be present for the occurrence of the harm, although there may be more than one set of conditions necessary. Further, the onus can be discharged by a consideration of the competing probabilities. 23Unlike Beazley P I do not derive any assistance from what was said by Dixon J in Betts v Whittingslowe [1945] HCA 31; (1945) 71 CLR 637. Dixon J in that case was dealing with a claim for breach of statutory duty. He stated, in the passage cited by Beazley P at par [193], that a breach complained of, together with an accident of a kind that might thereby be occurred, is enough to justify an inference, in the absence of "sufficient reason to the contrary, that in fact the accident did occur owing to the act or omission amounting to the breach of statutory duty" (emphasis added). 24Roads and Traffic Authority v Royal [2008] HCA 19; (2008) 82 ALJR 870 involved a claim against the appellant for the incorrect positioning of the stop sign which it was said led to an accident at a "T" intersection. Gummow, Hayne and Heydon JJ in a joint judgment rejected reliance on Betts v Whittingslowe supra. They referred to the fact that after the passage in the judgment of Dixon J cited by Beazley P, his Honour said that in the case before him "the facts warrant no other inference inconsistent with liability on the part of the defendant". They pointed out at [31] that in the case before them there were other inferences available. Kiefel J at [139] referred to the same passage in the judgment of Dixon J and emphasised at [143] that it remained a requirement of the law that a plaintiff prove a defendant's conduct materially caused the injury. So much is made clear by s 5E of the Act. 25In Kuhl v Zurich Financial Services Australia Ltd [2011] HCA 11; (2011) 243 CLR 361, French CJ and Gummow J at [50] referred with approval to the remarks of Kiefel J in Roads and Traffic Authority v Royal supra. They pointed out that what was contended for by the appellant in the case before them was not the only inference that could be made. Although their Honours were in dissent as to the result, the majority placed no reliance on Betts v Whittingslowe supra. 26The same approach to the dictum of Dixon J has been taken in relatively recent decisions of intermediate appellate courts: Duma v Mader International Pty Ltd [2013] VSCA 23 at [62]-[64] per Tate JA, Davies AJA agreeing and Fraser v Burswood Resort (Management) Ltd [2014] WASCA 130 at [144] per Martin CJ, McLure P and Newnes JA agreeing. 27As is apparent from the evidence summarised by each of Beazley P and Basten JA, there are logical inferences contrary to those contended for by the appellant. It was for the appellant to prove on the balance of probabilities that the inference contended for should be drawn. It cannot be inferred from the nature of the breach. (c) Was causation established in the present case? 28Although, as I have indicated, I do not consider the primary judge applied the wrong test in concluding that the appellant had not discharged the onus of proof to establish causation, it remains necessary to determine whether the onus was in fact discharged. The appeal, by virtue of s 75A of the Supreme Court Act 1970 (NSW) is an appeal by way of rehearing and the Court, subject to the natural limitations imposed on an appellate court, must conduct its own review of the evidence. Further if, making proper allowance for the advantage of the trial judge, the Court concludes an error has been shown, it is authorised and obliged to discharge its appellate duties under the statute: Fox v Percy [2003] HCA 22; (2003) 214 CLR 118 at [22]-[25] and Warren v Coombes [1979] HCA 9; (1979) 142 CLR 531 at 551. 29In considering whether the necessary connection can be inferred from the evidence, there are a number of matters which I consider to be of relevance. First, the state of the road. Second, what facts are known or can be inferred as to the circumstances of the accident. Third, the expert evidence of Mr Johnston and Mr Stuart-Smith. (i) The state of the road 30Senior Constable Hando described the portion of the road where the accident occurred in his statement as follows: "7. The section of the roadway where the collision occurred runs generally in a north/south direction. The Kingsvale Road has a posted speed limit of 100KM/H. The roadway consisted of a sealed bitumen surface with one northbound lane and one south bound lane separated by double unbroken lines. When travelling in the southbound lane approaching the scene there was a slight uphill grade with a moderate right hand bend followed by a slight downhill grade into a straight section. The section of roadway where the collision occurred was under repair, with sections of recently laid road surface." Senior Constable Hando also observed the following: "There was recently laid road surface along two sections of roadway prior to the collision scene. There was loose gravel on the surface, which covered the entire width of the roadway. The first section was on the bend. This section was 86.9 metres in length and 7.1 metres in width. There were no visible marks on the road surface. Following this section was a sealed bitumen section 32 metres in length. The second section was along the straight and was 32 metres in length and 7.3 metres in width. There were faint marks on the road surface in the northbound lane. Following this section was a sealed section of bitumen. This section was 53.2 metres in length. There was a set of tyre skid marks beginning in the southbound lane, which continued into the northbound lane leading towards the vehicle. The road surface was dry." 31A diagram, prepared by Senior Constable Ryan, showed the first resealed portion commencing as the road eased out of a right hand curve. The road was straight from approximately halfway through the first resealed section until the second resealed section. Senior Constable Ryan estimated the distance between the two resealed sections to be 40 metres, although Senior Constable Hando estimated it at 32 metres. The experts retained by the parties in a joint report expressed the view that the first resealed surface extended for 89 metres, the distance between it and the second resealed surface was 38-39 metres, the second resealed surface extended for 31 metres, whilst there was a gap of 54 metres from that area to the point where the deceased's car collided with a tree. 32It is important to note that Senior Constable Hando stated that there was loose gravel on the surface of each of the resealed sections along the entire width of the roadway. It was in that context that Mr Coffey gave the evidence referred to in pars [91] and [92] of the judgment of Beazley P. He described the effect of driving on the road with loose gravel on the top as walking on marbles, this being the reason speed warning signs were necessary. 33Mr Johnston's opinion was the same. In the portion of his evidence set out in par [298] of the judgment of Basten JA, he expressed the view that a speed limit of 60 km per hour should have been imposed and a "slippery road" sign should have been erected, as the road was "recently sealed with reduced traction". 34Mr Stuart-Smith in the passage of his evidence referred to in par [299] of the judgment of Basten JA, stated that it was better to have a "reduce speed" sign because a driver could lose control on loose gravel. He also agreed that speed was an obvious additional risk where there is a risk of loss of traction by reason of surface gravel. (ii) The circumstances of the accident 35Senior Constable Brand who attended the scene of the accident noted the accident occurred at about 11.28 am. His notes reported that the deceased was 42 years of age at the time of the accident and that a blood alcohol test was negative. In a statement subsequently prepared, he said the day in question was "a clear sunny day". He also recorded the following observations: "9. I noticed that there was skid marks apparent on the road surface. These skid marks were evident on the section of road immediately preceding the accident site and showed several things. Firstly that the vehicle had been travelling in a Southerly direction on the incorrect side of the road and the deceased applied the vehicles brakes whilst doing so. Secondly that the vehicle returned to the correct side of the road briefly - still under brakes. Thirdly that the deceased over corrected and swerved back across the incorrect side of the road, the vehicle turning 180 degrees as this occurred. Finally, that the vehicle travelled a short distance before impacting with a tree." 36Mr Mark Ellis, the Director of Works of the respondent, attended the scene of the accident at approximately 12.15 pm on the day in question, some 45 minutes after it had occurred. He made an incident report which contained the following remarks: "At the time of the accident [the deceased] was driving in a southerly direction from Kingsvale and had just been around through a short winding section of hilly terrain known as 'Stone Rises' and was just rounding a slight curve in the road when it appears she has crossed to the wrong side of the road and then overcorrected before losing control of the vehicle and it skidding sideways into the tree. Impact being the passenger side door. At the time of the accident it was a bright day with little cloud cover. It is also apparent that as the vehicle rounded the curve in the road it was at the start of recent Council Roadworks which involved resurfacing of the road pavement (undertaken on Thursday 19th Aug). As the vehicle traveled [sic] around the bend commencing on then [sic] correct side of the road it appears that at the end of the 1st sealed patch the vehicle was on the wrong side of the road. This being a distance of 90.0 metres. The vehicle has then traveled 38 metres on the wrong side before hitting a second sealed patch of 31.7 metres [in] length. Towards the end of this second patch is were [sic] the vehicle appears to have commenced sliding back across the road (to [the] correct side) for 33.3 metres before completely sliding sideways back (another 54.6 m) to the wrong side of the road, across the road shoulder and hitting the tree some 3.0 metres from the edge of the road. On inspecting the roadworks it is not evident that the vehicle lost control after hitting the first patch or even the second patch. There is no discernible skid marks through the sealing aggregate on the first patch, only some light marks were [sic] it appears the vehicle attempted to correct its path." 37Mr Mark Crisp, the Director of Technical Services of the respondent, attended the site with Mr Ellis. He took photographs of the site. Photograph No 19 (P 19) is of particular importance as it shows a tyre mark on the first section of the resealed surface running from the correct to the incorrect side of the road. 38Mr Crisp also drew a diagram based on his observations. As Beazley P has pointed out, it showed that at the point of entry onto the resealed surface the vehicle was on the correct side of the road. It showed the vehicle commencing to veer to the incorrect side. It then showed tyre marks on the second resurfaced section, again on the incorrect side of the road, which then travelled onto the correct side of the road and then almost immediately had veered to the right and off the road into the tree. 39In the joint report the experts expressed the following view as to the path taken by the vehicle: "Messrs Johnston and Stuart-Smith generally agree that: The vehicle appears to have deviated from the left side of the road within the first patch to the right side of the road at the end of the second patch. The evidence for this is a possible/probable (RSS/GJ) tyre mark shown in some diagrams and possibly evident in a Council photograph. On reaching the right side of the road, (towards the end of the second patch about 60 m before the POI) the vehicle deviated to the left. The evidence for this is the observation by Police and others of a faint tyre mark on the extreme right side of the road towards the end of the second patch. The mark is possibly evident in a Police photograph. The vehicle then passed on to an old section of pavement and crossed substantially back onto the left side of the road before deviating sharply to its right. The vehicle commenced to rotate clockwise and passed back onto the right side of the road and collided in a lateral impact (left side leading) with a tree located a metre or two off the right side of the road. The tree was located approximately level with the start of a new resealed surface. The evidence for this is the presence of diverging tyre marks shown in Police photographs, commencing about 45 m before the POI, as well as vehicle damage. Opposing Views Mr Stuart-Smith is of the view that there is no clear evidence of the vehicle's movement from the first patch to the point where it reached the right side of the road. The possible initial tyre mark was not recorded or photographed by Police and the photograph of the possible mark is not clear. Mr Johnston agrees that the photographic evidence is not definitive but does tend to support the view of the Council Officer who drew the contemporaneous diagrams showing what was understood to be his observation of assumed tyre tracks." 40In the joint report Mr Stuart-Smith was prepared to accept only that it was possible that P 19 showed a tyre mark on the first resealed section veering to the wrong side of the road. However, in cross-examination he conceded that the marks in P 19 were consistent with the observations of Mr Ellis. (iii) The evidence as to when and for what reason the deceased lost control of her vehicle 41The experts were also asked to give their opinion of the earliest evidence of the deceased losing control of her vehicle. The following views were expressed in the joint report: "Mr Johnston's View Mr Johnston is of the view that the driver has commenced to lose control around the start of the first section of reseal within the right horizontal curve. The evidence of this is the contemporaneous diagrams by a Council officer, the possible indication of a tyre mark within the Council photographs and the known passage of the vehicle from the left side of the road to the right side of the road commencing within a right horizontal curve consistent with the Council officers drawing of the initial tyre marks. Mr Stuart-Smith's View The earliest clear evidence of loss of control is at the start of the tyre marks located about 45 m before the POI. At this point the vehicle was commencing a clockwise yaw. However, it is probable that the vehicle was out of control about 10 to 20 m earlier at the location of a possible tyre mark at the end of the second patch. It was likely that the vehicle was over-steered to the left at this point. There is no evidence of loss of control at the curve overlapping the first patch, as suggested by Mr Johnston. The factors identified by Mr Johnston are not evidence. Had the vehicle been out of control giving rise to an oversteer to the right (as would be necessary for it to continue towards the POI), evidence, such as tyre marks and windrows on the loose material of the first patch would have been present. No such evidence was recorded by Police or photographed. The long distance of over 100 m from the curve to the point on the right side of the road where the vehicle deviated left is generally inconsistent with it having been out of control. Mr Johnston's suggestion that the deceased lost control at the curve on the first patch is not supported by evidence where evidence would have been expected." 42A further question was asked, "[w]hat does the evidence say was the most likely cause of the loss of control". The experts gave the following report: "Mr Johnston's View Mr Johnston is of the view that initial loss of control was probably due to a partial loss of traction upon transitioning into the first section of reseal. This has caused increased clockwise slip angle and a resulting deviation of the vehicle path to the right. The evidence for this loss of control has been discussed above. Mr Stuart-Smith's Views There is no evidence of loss of control at the curve overlapping the first patch. Had the vehicle lost control in loose gravel, evidence, such as tyre marks and windrows on the loose material of the first patch would have been present. No such evidence was recorded by Police or can be seen in photographs. Mr Johnston's suggestion that the vehicle lost control as a result of rounding the curve in loose gravel is not supported by evidence where evidence would have been expected. An oversteer to the left at the faint tyre mark was the likely cause of loss of control when the vehicle approached the right side of the road about 60 m before the POI. Alternatively, an oversteer to the right less than 45 m before the POI, as evidenced by the diverging tyre marks leading to the POI was the cause of loss of control." 43Mr Stuart-Smith elaborated on his views in his evidence at the trial. The experts gave their evidence jointly and the following exchange sets out his conclusion. After agreeing that what appeared in P 19 was consistent with the observations of Mr Ellis, the following exchange occurred: "MARSHALL: - would you agree that the marks that are shown, the tyre marks that are shown of a vehicle, crossing to its incorrect side of the road, if it be Ms Paterson's vehicle, cause lateral displacement of stones? WITNESS STUART-SMITH: No, no, and - or the marks could be a vehicle going the other way, but assuming it's Ms Paterson's vehicle coming this way, they certainly don't show any evidence of - I'm sorry, they don't show any evidence of lateral displacement of stones in - to one particular side. There could be a lateral displacement of stones either side. MARSHALL: Either side? WITNESS STUART-SMITH: Yes, I'm sorry, yes. I would agree that's possible, yes. MARSHALL: And that would be consistent with a vehicle attempting to get on its correct side of the road, having, for some reason, gone on to its incorrect side? WITNESS STUART-SMITH: No, no, the reason is that a vehicle in Ms Paterson's - assuming Ms Paterson had lost control and was making that mark, she would have been in what we call a clockwise yaw. The rear would have started to move, swing to the left, and - and there would be a displacement of material towards the outside. So looking at the photograph, that's on the right-hand side of the photograph, in other words, Ms Paterson's left as she's coming towards us, so one would have expected some windrow or some build up of material, or some - some shading of the tyre mark to indicate some lateral displacement. But not only that. A vehicle doesn't just sideslip over - just sideslip and then continue just travelling in that same sort of general direction. Once it starts to sideslip, it continues to rotate until the steering has twisted in the other direction, in which case, you either correct - control is regained or else you end up with an oversteer, and the mark is too long. It's - the photograph is like 80 metres from - it's 80 metres from the police car to the end of that tyre, the end of that patch. There's 150 metres from the police car to where the plaintiff's vehicle most likely was on the incorrect side of the road at the end of the second patch. So what we would have to assume for that to have been made by her car out of control, is that she sideslipped - been sideslipping for about 150 metres without her car seriously rotating. It's not really feasible for that to occur over such a long distance without other marks having started to occur, without the car sort of turning around to a greater extent. 150 metres is just too long for some - some little bit of sideslip to have been caused at an originating point, and then to have continued at the speed that she was travelling. MARSHALL: Do you agree with that, Mr Johnston? WITNESS JOHNSTON: No, I don't agree with the last point. I agree that she wasn't in complete loss of control when the car started rotating at the start and continued to rotate at 150 metres. There was a loss of traction that has caused her vehicle to orient and travel the incorrect side of the roadway. She has then made a series of manoeuvres that she has presumably tried to correct the vehicle, and stops leaving marks because it's on the wrong side of the road. And then there's a second patch, and then has a catastrophic loss at the end of that." 44Subsequently the experts were asked the following questions: "MARSHALL: Is, can I then say on the balance of probabilities, what is available to this Court to which I have just referred you, consistent with a loss of some control, at or near the beginning of the first patch, a conservative correction through the first patch into the second, and a subsequent catastrophic loss of control? WITNESS JOHNSTON: Yes MARSHALL: In fairness, I should ask the same question to Mr Stuart-Smith. Do you accept on the balance of probabilities that scenario? WITNESS STUART-SMITH: No, I don't accept that on the balance of probabilities at all. HER HONOUR: It's a possible scenario. WITNESS STUART-SMITH: A possible scenario. HER HONOUR: Is there a more probable scenario in your view? WITNESS STUART-SMITH: I believe so, yes, your Honour. HER HONOUR: What's that? WITNESS STUART-SMITH: A more probable scenario is that the driver deviated for an unknown reason. The reasons for that, are that it's only a balance of probabilities in my view, in that we have the statement by Mr Ellis where he specifically said he couldn't see any, what he called skid marks in there, and had - I think loss of control, in this case, we should only really be concerned with loss of control as a result of traction, and that's all I - what we're dealing with here. But that's all that matters. If she lost control for some other reason, then it's not the loose material that caused her to lose control, but there's no evidence of loss of control from traction where evidence would be expected in that material either through Mr Ellis's statement or from observations by police officers. There are no photographs of that area there, and so that's the actual lack of evidence to support that theory. We then come and look at some objective sort of factors, and that is, that even if you use the minimum likely friction, reduced friction, as proposed by Mr Johnston, it was traversable at a very high speed, admittedly, with constant steering and without braking at those high speeds, and Mr Johnston made that point, and I accept those qualifications. But nonetheless, even if you then apply various other actions which may well have reduced - had the effect of reducing the friction such as braking, such as transitioning from one side to another, it's still traversable, should be comfortably traversable at 100 kilometres per hour without loss of traction, and so given that on one hand, the absence of objective evidence, of loss of traction, and on the other hand, the theoretical calculations that suggest that it was traversable, comfortably traversable at high speed - when I say that high speed, I'm talking 100 kilometres per hour - I say those two elements add up to a balance of probabilities that she lost, she deviated for a reason other than loss of traction. HER HONOUR: What if I added this into your assessment, that despite the theoretical possibility of safe passage through the curve at 100 kilometres an hour, the loss of traction under that speed on this gravel, was the extent of this gravel, was such as to render that theoretical possibility far less likely, such that there was in fact a loss of traction or it was not safely - the curve was not safely traversed at that speed - would that alter your view as to the probabilities of control having been compromised, leading to the driver not maintaining her orientation on the correct side of the road? WITNESS STUART-SMITH: Well, I that that would - that obviously, that reduces the degree to which one could say that what my previous opinion - there is another factor too, and I accept that once a driver has lost control, they can do a number of things, but the fact that if we assume this tyre mark to be hers, that she did travel in this relatively straight line for 150 metres, is, on the far side of likelihood for a driver who has just lost control, it's reasonably uncommon for a driver to freeze or just to do that for that length of time, so whilst I couldn't say that a driver wouldn't do that, it adds weight to the likelihood that it wasn't a loss of traction that caused her to deviate." 45Mr Johnston pointed to the fact that the deceased's vehicle veered onto the incorrect side of the road contemporaneously with entering onto the unsealed surface. He then gave the following evidence: "SHELDON: But if that is the path the vehicle took later on, wouldn't you have to assume the regaining by the driver of complete control of the vehicle in the absence of marks, indicating that the vehicle was in a yaw, as to came [sic] down the incorrect side of the road? WITNESS JOHNSTON: Certainly not complete control, but I would expect the vehicle was under control and largely corrected. So, it was now heading in the wrong direction, but was no longer continuing to rotate. SHELDON: Would you expect there to be marks on the road showing the loss of control that precipitated that manoeuvre? WITNESS JOHNSTON: Provided the area was not completely contaminated by material, yes. I would expect somewhere up near where the police car is, a trained eye would see, or find, that evidence, potentially. HER HONOUR: A trained eye would find what? WITNESS JOHNSTON: A trained eye would find some evidence of that type of loss of control. By that I don't mean general duties police officers or crime scene officers. HER HONOUR: When you say that loss of control - - WITNESS JOHNSTON: A clock-wise rotation of the vehicles due to traction. I suggest if Mr Stuart-Smith or I, if we investigated, might find it. SHELDON: When you say might find it, that is in the realms of speculation - - HER HONOUR: No, would find it; find evidence of that loss of control, that is, a clock-wise rotation of the vehicle due to traction on the road surface? SHELDON: When you say that Mr Stuart-Smith and you might pick it up - you did say 'might' didn't you? WITNESS JOHNSTON: I said that subject to a lot of contamination being up there. So, if we got there fresh, and no one else had been on the roadway and it was clear, I would be fairly confident we would pick it up, when I have been at scenes of crashes." 46It was this evidence that the primary judge criticised as circular. Her Honour made the following remarks: "[107] It is clear from this evidence that Mr Johnston's opinion was also informed by his belief (or expectation) that physical evidence on the road surface in the first section where he believed traction was lost would likely have been discernable to 'a trained eye', and that neither officers Brand and Hando nor Messrs Ellis and Curtis were sufficiently qualified to discern what Mr Johnston claimed he and Mr Stuart-Smith would have been able to discern and indentify as a tyre mark of a vehicle in a clockwise yaw. [108] Mr Johnston's assumption that the tyre marks were present but not seen could only support a finding that they were probably there if there were some evidence that the marks were there to be seen but overlooked. The obvious flaw in his reasoning is that if there were evidence of a yaw mark or sideslip proximate to where Ms Paterson entered the first resurfaced section (whether or not she regained control of the vehicle thereafter) there would be no need to proceed on the assumption that it was overlooked. To the extent that Mr Johnston's opinion that the probable cause of the accident was a loss of control in the first resurfaced section due to a loss of traction is based on the assumption that there either was physical evidence which was missed by all observers, or there might have been evidence that was missed by those people, the reasoning suffers from circularity in the sense that it is assumed that there were marks on the road because one would expect them to be present if there was a loss of control due to lack of traction. I am not persuaded that the theoretical possibility that the tyre marks were missed by those at the scene (and, in the case of Sergeant Brand, missed despite the fact that he was specifically looking for some evidence of the loss of control) permits an inference to be drawn that the tyre marks were there but missed." Consideration 47The primary judge concluded on that evidence that she was not satisfied on the balance of probabilities that the cause of the accident was the deceased losing control as a result of encountering loose gravel on the first resealed section of the road. Her Honour's conclusions were as follows: "[110] I accept that the competing hypotheses for Ms Paterson's loss of control, being a lack of concentration, inattention, drowsiness or fatigue, would logically need to subsist at the very point on the road where there was increased risk of a loss of control by reason of excess speed on gravel. While this state of affairs might undermine the likelihood of either of them (or any combination of them) being the explanation for the loss of control, it is necessary for me to be actually persuaded of one scenario over another as the more probable explanation and I am unable to reach that level of satisfaction. [111] In the absence of any physical evidence of sideslipping due to a clockwise rotation of the vehicle in the first resurfaced section in circumstances where it would be expected to be obvious, and where the road surface was examined and none noted; where the photographic evidence favours a finding of a gradual veering from the correct to the incorrect side of the road before a faint but definite mark 150 metres after the gravel is first encountered and after an intervening section of unworked road without tyre marks; and where Mr Curtis' sketch and Mr Ellis' report are open to interpretation, the balance of the evidence, which includes the evidence of another accident on a stretch of road under the same road repair works 24 hours earlier, simply does not induce in my mind an actual persuasion that the precipitating cause of Ms Paterson's accident was her losing control of her vehicle in the first resurfaced section due to the presence of gravel and her unchecked speed. ... [113] Given the way the plaintiffs have put their case, the failure to persuade me that control was probably lost because of a loss of traction at the beginning of the first resurfaced section, a primary fact upon which the further and critical finding of factual causation depends, carries with it the consequence that I am not persuaded that but for the failure to provide for and position appropriate signage the accident would not have occurred." 48The first question which needs to be considered is whether on the balance of probabilities the deceased lost control of her vehicle on the first section of the unsealed road. In my opinion the photographic evidence of P 19, the statement by Mr Ellis and the drawing by Mr Crisp based on their observations, lead to an inference that on the balance of probabilities there was a loss of control at that point of time. As Constable Hando observed the northbound and southbound lanes were separated by double unbroken lines. A competent driver in control of her vehicle would take care not to cross those lines. I appreciate as her Honour pointed out (primary judgment at [110]) that the report of Mr Ellis is open to interpretation, in that it does not provide certainty that the car veered to the right on encountering the first resealed surface. However, the question is to be decided by reference to the probabilities. The evidence available, particularly P 19, satisfies me to that standard that that was what occurred. 49Even if it is inferred that the vehicle travelled in a straight line following its veering to the right, that does not lead to the conclusion that there was no loss of control or that the deceased regained control. It appears from the evidence of Senior Constable Brand and Mr Ellis that the vehicle was on the wrong side of the road when it reached the second resealed surface. The fact that the car travelled on the wrong side of the road, albeit in a straight line from the end of the first resealed surface to the second, does not indicate to me that the driver was in control or had regained control of the vehicle. Nor do I consider that that fact renders it more probable that the reason the deceased was on the wrong side of the road was due to a loss of concentration, inattention, drowsiness or fatigue as opposed to a loss of control as a result of encountering loose gravel. This is particularly the case given the distance covered by the deceased on the wrong side of the road (approximately 100 to 150 metres) would have been covered within a matter of seconds. In that regard I would prefer the conclusion of Mr Johnston in the passage to which I have referred in par [43] above. 50As Mr Stuart-Smith pointed out in the evidence I have summarised in par [44] above, the key question is whether the loss of control was as a result of a loss of traction caused by the loose gravel. The experts disagreed on this point. Mr Johnston stated he regarded the loss of control as a result of the loose gravel the probable cause. Mr Stuart-Smith, whilst conceding that that was a possibility, disagreed. 51In considering this issue it is appropriate to look at the competing hypotheses. There was no suggestion of mechanical failure and no suggestion that the deceased was affected by drugs or alcohol. The deceased was driving at an estimated speed at or slightly under the speed limit (100 km per hour) on a bright sunny day. The evidence of the experts was that absent any loose gravel, the bend could be safely negotiated at least at 145 km per hour. 52In these circumstances the only alternatives suggested were a loss of concentration, inattention, drowsiness or fatigue, coincidentally subsisting at the very point where there was an increased risk of loss of control by reason of excess speed on gravel (primary judgment at [110]). However, it must be remembered that the deceased had negotiated a significant part of the curve successfully, a fact tending against these explanations. It seems to be more probable than not that in those circumstances, the appellant lost control as a result of loss of traction on encountering loose gravel on the first resealed surface and was unable to regain full control before encountering the second resealed section and ultimately the tree which caused her death. 53In reaching this conclusion I appreciate the force of the fact that there was no physical evidence of sideslipping due to a clockwise rotation of the vehicle in the first resurfaced section and the somewhat circular nature of Mr Johnston's explanation for its absence. Notwithstanding I remain of the view that it was more probable than not that the deceased lost control on the first resealed section due to the lack of traction caused by loose gravel. It must be remembered that Mr Stuart-Smith conceded that that remained a possibility. Taking this and the matters to which I have referred in pars [51] and [52] into account and bearing in mind the hazardous nature of the road as described by Mr Coffey and Mr Johnston (and ultimately accepted by Mr Stuart-Smith), the absence of evidence of sideslipping does not cause me to alter my conclusion. 54In reaching this conclusion I do not draw any support from the fact that there was evidence that a Ms Laura Skorulis apparently had an accident as a result of slipping on loose gravel whilst travelling on the road in the opposite direction to the deceased. I do not think that this demonstrates anything more than it was hazardous to drive at speed on a resurfaced road which had loose gravel on it. That was not ultimately in issue. 55The respondent accepted that if it was established that one of the causes of the accident was loss of control on the first resurfaced section as a result of a loss of traction due to loose gravel, the necessary connection between the breach of duty and the injuries leading to the death of the deceased was established. In these circumstances the appeal should be allowed. Conclusion 56In the result, the appeal should be allowed. There does not appear to be an order for separate questions, so it is not clear whether contributory negligence, which was pleaded in the defence of the respondent remains in issue. 57In those circumstances I agree with the orders proposed by Beazley P. 58BEAZLEY P: On 20 August 2004, Debbie Paterson was fatally injured when, whilst travelling along Kingsvale Road between Young and Harden in country New South Wales, the car she was driving ran off the road and hit a tree. Ms Paterson was the sole occupant of the vehicle and there were no witnesses to the accident. Ms Paterson was 42 years old at the time of the accident. Her occupation was that of home duties. 59The appellant, who was Ms Paterson's partner, brought proceedings for psychiatric injury arising from Ms Paterson's death. He also brought proceedings on his own behalf and on behalf of her children under the Compensation to Relatives Act 1897. 60At the time of the accident, roadworks were being carried out on Kingsvale Road. The appellant claimed that the accident was caused by the presence of loose gravel on the road surface which was present due to the negligence of the respondent, Harden Shire Council (the Council). 61The Council denied negligence and also pleaded contributory negligence pursuant to the Civil Liability Act 2002, s 5T. The Council also pleaded that to the extent the claim relied upon a failure to warn, the risk was obvious: see the Civil Liability Act, Pt 1A, Div 4, s 5G(1). The Council also relied upon the provisions of the Civil Liability Act, Pt 5, ss 42, 43, 43A and 44 relating to limitations on the tortious liability of public authorities. 62The trial judge, in her reasons for judgment given on 9 July 2012, gave judgment for the Council. Her Honour found that the Council had breached its duty of care to Ms Paterson in failing to provide pictorial signage indicating that the road had been resurfaced and speed reduction signage in advance of the commencement of the roadworks. However, her Honour concluded, at [111], that the evidence did not induce in her an actual persuasion that the cause of Ms Paterson's accident was her loss of control of her vehicle in the first resurfaced section of the roadwork due to the presence of gravel on that surface and her unchecked speed. 63Her Honour also held, at [70], that although the Council had been negligent, the appellant had not established that the Council's conduct met "the necessary test for Wednesbury unreasonableness within the meaning of s 43A", applicable to a public authority when acting under a special statutory power within the meaning of that section. The Council was accordingly protected under that section from liability for negligence. Issues on the appeal 64The appellant submitted that her Honour, in giving judgment for the Council, erred in the following respects: first, in the manner in which she applied the Civil Liability Act, s 5D (the causation issue); secondly, in requiring the appellant to establish his claim on a standard of proof that was higher than the balance of probabilities (the standard of proof issue); thirdly, in confining her finding of breach of duty to the Council's failure to erect "Slippery Road" and "Reduce Speed" signs, when her Honour ought to have found that the Council had been negligent in the preparation of the Traffic Control Plan (TCP) (the breach of duty issue); and fourthly, in holding that the statutory defence provided by the Civil Liability Act, s 43A applied to the circumstances of the case (the s 43A issue). 65The first two issues are interrelated and will be addressed in this judgment together as the causation issue. The third and fourth issues are also interrelated and will be addressed together as the s 43 issue. Factual background 66The circumstances relating to the accident, other than how the accident occurred, were uncontroversial and the following account is derived principally from her Honour's judgment. 67As already noted, the accident occurred on Kingsvale Road. Kingsvale Road is a rural highway running generally in a north/south direction between Harden to the south and Young to the north. It has a designated speed limit of 100 km/h. 68The day before the accident, the Council had completed laying bituminous material on the stretch of Kingsvale Road where the accident occurred as part of its road maintenance works program. The work was carried out in five discontinuous sections, of differing lengths, over a distance of about 500 m. Unworked sections of the road, also of differing lengths, separated the resurfaced sections. The work involved spraying bituminous material onto the prepared road surface which was then dressed with loose blue metal gravel (or aggregate). It is an expected incident of that method of resurfacing that an excess of gravel will remain on the road surface. The gravel is progressively embedded in the new surface by passing traffic. The road surface is swept as the final phase in the resurfacing process. 69Just before midday on 20 August 2004, Ms Paterson was travelling south towards Harden. Ms Paterson was travelling at, or perhaps slightly under, the speed limit when she entered the right hand curve in the road at the commencement of the new work. The point of impact was 210 m south of the commencement of the work, adjacent to the start of the third resurfaced section. Ms Paterson's vehicle collided with a tree on the verge of the right hand side (the incorrect side, given Ms Paterson's direction of travel) of the roadway. Her vehicle had rotated approximately 180 degrees whilst still on the roadway and was facing north (that is, in the opposite direction to her direction of travel) at the point of impact with the tree. Neither the condition of the vehicle nor the weather had any relevant bearing on the cause of the accident. It was common ground that Ms Paterson was a careful and experienced driver who was familiar with the road. There were no witnesses to the accident or to any part of the driving sequence that preceded it. 70The respective lengths of the worked and unworked sections were measured by Senior Constable Hando from the Goulburn Crime Scene Section of the New South Wales Police. In his report of the accident, he recorded the following: "● There was recently laid road surface along two sections of roadway prior to the collision scene. There was loose gravel on the surface, which covered the entire width of the roadway. ● The first section was on the bend. This section was 86.9 metres in length and 7.1 metres in width ... ● Following this section was a sealed bitumen section 32 metres in length. ● The second section was along the straight and was 32 metres in length and 7.3 metres in width ... ● Following this section was a sealed section of bitumen. This section was 53.2 metres in length." 71The roadworks were governed by a Traffic Control Plan (TCP) which had been adapted by the Works Manager employed by the Council from a generic plan (TCP 56) in the applicable Roads and Traffic Authority's (RTA) Traffic Control at Worksites Manual (the Manual). It was common ground that the TCP (as adapted) was designed to address the risks associated with or likely to arise from the roadworks on Kingsvale Road, including the risks to road users presented by the presence of loose gravel on the road surface. 72A sign advising of the presence of (unspecified) roadworks was erected approximately one kilometre in advance of the first section of resurfaced roadway. In addition, within 60 m of the commencement of the first resealed section and at various intervals thereafter, signage was placed which advised that there were no centre lines marked and that vehicles should not overtake. There were also pictorial signs indicating the potential for windscreen damage from the presence of stones or gravel. These signs were identified in the evidence as "Chip Hazard" signs. There was no provision made in the adapted TCP for signage advising that the road had been resurfaced or imposing a reduction in the speed limit. Nor did the adapted TCP provide for any pictorial signage warning specifically of the risk of slippage (or loss of traction) due to the presence of loose gravel on the new surface and no such signs were in place. 73It was agreed by the parties that the evidence supported a finding that Ms Paterson entered the first resurfaced section with her vehicle under control. Mr Johnston, the expert retained by the appellant, considered that Ms Paterson lost control in the first section of the new work, whereas Mr Stuart-Smith, the expert retained by the respondent, considered that there was no evidence that Ms Paterson lost control until she underwent the final movements in the last unworked section of the roadway before the collision where skid marks revealed that Ms Paterson's vehicle had spun 180 degrees before leaving the roadway. The differing views of the experts are considered below. Technical requirements for road sealing 74There were various technical requirements and guidelines for signposting and for the implementation of other safety measures issued by the RTA when roadworks were being undertaken. Aspects of these requirements are discussed below in relation to the Council's "public authority defences". Presently relevant, however, are sections 12, 13 and 14 of the National Association of Australian State Road Authorities (now renamed Austroads) (NAASRA) Technical Report, Bituminous Surfacing Sprayed Work, which include the following requirements: "Immediately the binder has hardened to the stage at which no more aggregate can be pressed into it by rolling or by traffic moving at slow speeds all loose aggregate should be removed by sweeping it clear of the edge of the seal ... The road should not be opened to unrestricted traffic until the loose aggregate has been removed." 75In addition, the Council's Safe Work Method Statement for Sweeping Pavement provided that speed restrictions were to be in place until after the road surface was "completely broomed". Inspection of the accident site 76Two Council employees, Mr Crisp and Mr Ellis, attended the scene shortly after the accident. 77Mr Ellis, who was the Council's Director of Works, prepared a report of the accident. Mr Crisp, who was an engineering assistant, prepared a sketch of the scene of the accident. Both the sketch and the report were prepared on the day of the accident. Neither Mr Crisp nor Mr Ellis gave evidence at trial. No Jones v Dunkel inference was drawn from the Council's failure to call these two witnesses, as her Honour considered that there was nothing to suggest that their evidence would be different from their contemporaneous notes: Jones v Dunkel [1959] HCA 8; 101 CLR 298. 78In his report, Mr Ellis recorded the following: "At the time of the accident [Ms Paterson] was driving in a southerly direction from Kingsvale and had just been around through a short winding section of hilly terrain known as 'Stony Rises' and was just rounding a slight curve in the road when it appears she has crossed to the wrong side of the road and then overcorrected before losing control of the vehicle and it skidding sideways into the tree. Impact being the passenger side door. At the time of the accident it was a bright day with little cloud cover. It is also apparent that as the vehicle rounded the curve in the road it was at the start of recent Council Roadworks which involved resurfacing of the road pavement (undertaken on Thursday 19th Aug). As the vehicle traveled around the bend commencing on then correct side of the road it appears that at the end of the 1st sealed patch the vehicle was on the wrong side of the road. This being a distance of 90.0 metres. The vehicle has then traveled 38 metres on the wrong side before hitting a second sealed patch of 31.7 metres length. Towards the end of this second patch is were [sic] the vehicle appears to have commenced sliding back across the road (to correct side) for 33.3 metres before completely sliding sideways back (another 54.6 m) to the wrong side of the road, across the road shoulder and hitting the tree some 3.0 metres from the edge of the road. On inspecting the roadworks it is not evident that the vehicle lost control after hitting the first patch or even the second patch. There is no discernable skid marks through the sealing aggregate on the first patch, only some light marks where it appears the vehicle attempted to correct its path." (emphasis added). 79Mr Crisp also took photographs of the road, the roadworks, the loose gravel and the crashed vehicle. The sketch drawn by Mr Crisp depicted the path of Mr Paterson's vehicle from its entry onto the first section of new work to the point of impact with the tree. At the point of entry onto the first section of new work, the vehicle was depicted to be on its correct side of the road. The sketch then showed the vehicle commencing to veer to its incorrect side of the road. No tyre marks were visible on the next section of the road, which was an old unworked section. Tyre marks were next seen on the second new section of roadway on the incorrect side of the road. The tyre marks were shown to have then travelled onto the correct side of the road on the second unworked section and to then almost immediately have veered to the right and off the road into the tree adjacent to the third resealed section. 80Of the photographs taken at the accident site, the photograph numbered 19 was of particular importance, as it was found to show a tyre mark on the first section of new work running from the correct to the incorrect side of the road. 81A number of police officers also attended the scene of the accident. The notes in the police notebook recorded at the scene of the accident indicated that the skid marks leading directly to the vehicle were 69 feet (as paced out). The notes also stated that it was "[n]ot known if the resealed road surface was contributing factor in accident". 82Senior Constable Brand, of Harden Police Station, attended at the scene of the accident, arriving at about 12 noon. In a statement made on 13 September 2004, he recorded that he had observed skid marks on the road. He said that: "These skid marks were evident on the section of the road immediately preceding the accident site and showed several things. Firstly that the vehicle had been travelling in a Southerly direction on the incorrect side of the road and the deceased applied the vehicles brakes whilst doing so. Secondly that the vehicle returned to the correct side of the road briefly - still under brakes. Thirdly that the deceased over corrected and swerved back across the incorrect side of the road, the vehicle turning 180 degrees as this occurred. Finally, that the vehicle travelled a short distance before impacting with a tree." 83This was a description of the vehicle's path of travel in the second unworked section of the road until the point of collision. Senior Constable Brand stated that he did not have the expertise to determine the speed of the vehicle prior to the collision based upon his observation of the skid marks on the road surface and, in the absence of any physical evidence to indicate the cause of the accident, advanced the possibility that Ms Paterson may have fallen asleep or have been distracted. 84Senior Constable Brand inspected the road until approximately 100 m north of the point from where the vehicle left the road. It will be recalled that the accident occurred 210 m from the commencement of the roadworks. Senior Constable Brand stated he was unable to find any physical evidence to indicate a cause for the vehicle to be on the incorrect side of the road. It is apparent, both from Senior Constable Brand's statement and the statements of the other police officers who attended the scene of the accident, that in looking for a cause for the accident, they were seeking to ascertain if there were potholes that Ms Paterson may have tried to avoid, or the presence of broken glass or an animal carcass that would suggest a collision with another vehicle or an animal. There was no evidence of that type. 85Constable Ryan, who was on highway patrol duties, attended the scene of the accident at 12.10 pm. He was also unable to express a reason for Ms Paterson's vehicle having travelled onto the wrong side of the road. However, his observations of the roadway support the observations of Mr Ellis and Mr Crisp of a mark on the first section of new work (see bolded section below). As Constable Ryan also provided a clear explanation of the marks on the road in the section immediately preceding the accident, I will set out this part of his statement in full: "... I continued south and saw a dark set of skid marks. The skid marks consisted of two lines, which began virtually from the centre of the roadway. They then arched in a clockwise direction towards the eastern side of the road before crossing the incorrect side of the road and arching towards the western side of the road ... I looked at the skid marks and determined that the mark closest to the eastern side of the road would have been the rear nearside tyre and the inner mark would have been the rear offside tyre. I walked back to the skid marks and stood where they had begun. I walked in the direction of the marks, which took me onto the incorrect side of the road. I continued on this course until I met the resurfaced section ... where there were two marks, which I thought were marks caused by vehicle tyres, on the point where the new and old surface joined. These marks were some sixty metres north of the impact point. The distance between these marks was the similar size to that of a car and they appeared to be slightly skewed, facing south-southeast, leading me to believe that the marks were made by a car not travelling in a straight line. It appeared to me as though these marks lined up with the beginning of the skid marks. I continued north and I saw a very faint tyre path on the very western side of the road, which when I looked back south, thought possible they may have ran into the path of the previous marks mentioned above. This path was very faint however and I am unable to say with any certainty that the path was caused by the deceased's vehicle." (emphasis added) 86Senior Constable Hando attended the scene at 2.40 pm on the day of the accident and measured the length and breadth of the sections of new work and the old roadway (see [70] above). He observed "faint marks on the road surface in the northbound lane" of the second section of new work. He also observed "a set of tyre skid marks beginning in the southbound lane, which continued into the northbound land leading towards the vehicle [at the point where it collided with the tree]". According to Senior Constable Hando, there were no visible marks on the road surface in the first section of the new work. 87Senior Constable Hando, who had completed the Introduction into Crash Investigation Course conducted by the New South Wales Police Crash Investigation Unit, as well as the Certificate IV in Forensic Investigation accredited by the Canberra Institute of Technology, concluded: "From my examination of the scene and the vehicle it would appear that the yellow 1980 Holden Commodore ... travelled at an unknown speed in a southerly direction along the Kingsvale Road, Kingsvale. Upon entering the right hand bend in the roadway the deceased has lost control of the vehicle, which has slid out of control along the roadway in a southerly direction. The vehicle has rotated in a clockwise direction almost 180 degrees travelling in a westerly direction with middle nearside portion of the vehicle impacting heavily with a gum tree on the western side of the roadway." Other evidence 88The Council's Director of Technical Services at the time of the accident, Mr Coffey, gave evidence relating to the sufficiency of the TCP and, in particular, as to the adequacy of the signage that had been erected. As discussed below, Mr Coffey considered that the TCP was seriously deficient and that there should have been both a "Slippery Road" sign and a speed reduction sign in place. 89The TCP was not reviewed by Mr Coffey prior to its implementation. However, Mr Coffey reviewed it after the accident to determine its suitability, having regard to the roadworks that were being undertaken. During the course of this review, Mr Coffey had sought written explanations from the Works Manager, inter alia, as to whether the absence of a "Slippery Road" sign had been noted and, if so, why no action had been taken to remedy that. The Works Manager failed to provide any explanation, despite Mr Coffey making follow-up enquiries. 90Mr Coffey was of the opinion that the Work Manager's modifications to the generic RTA traffic control plan were "seriously deficient" in two critical respects. The Works Manager had deleted from the generic plan the pictorial Slippery Road sign, which shows a car sideslipping in an uncontrolled movement, leaving the "Chip Hazard" sign as the only pictorial warning of the presence of loose gravel. In Mr Coffey's view, the Slippery Road sign was a standard sign used during road resurfacing operations. In particular, Mr Coffey said that the Slippery Road sign "is one of the most important signs" to be used when bitumen resealing works are being undertaken. He said a speed reduction sign should also have been included because the designated speed limit on Kingsvale Road was 100 km/h. 91During the course of his cross-examination relating to appropriate signage and, in particular, the necessity of having both a "Slippery Road" sign and a "Reduce Speed" sign in place (neither of which were present), Mr Coffey explained that in the course of roadworks where a bitumen surface has been laid, the road would be "slippery", because there would be loose aggregate lying on top of the road. Her Honour sought clarification of this evidence as follows: "[HER HONOUR]: It means slippery, does it, in the sense that there is reduced traction where otherwise there would be good traction on a sealed and set road?" Mr Coffey answered: "That's correct, your Honour; similar to walking on marbles." 92Senior counsel for the Council continued the cross-examination as follows: "Q. But you accept, don't you, that the extent to which a given location might be, as you described it, like walking on marbles, in the context of the car driving on a recently resealed road, it is determined by the sharpness of the curve and the amount of aggregate? A. Yes. Q. Can I suggest to you that a reason why a reduced speed sign was not, or may not, have been necessary, was that the corner with the loose aggregate on it could be safely negotiated at a speed of at least 145km per hour? A. Under normal circumstances? Q. No, with the aggregate on the road? A. Loose aggregate on the road? Q. Yes? A. Definitely not. Q. Are you disagreeing with the speed calculation? A. Would you like to repeat the question? Q. Yes. Firstly I should preface it with this; do you understand that the prevailing speed limit at this location was 100km per hour? A. Correct. Q. If the fact is that even with aggregate on the road, following a resealing operation and before sweeping, the corner could be negotiated safely at a speed of 145km an hour, there would be no need for a reduced speed sign? A. Two parts; I disagree that that could occur, that the corner could be negotiated at that speed with loose aggregate on it. But if it can, as you have assumed for whatever reason a reduced speed sign would not be necessary, but it is a circumstance I don't fathom. The premise of the question is just completely wrong." 93Mr Coffey further explained the purpose of erecting warning signs. He said: "The Reduce Speed sign and all warning signs give the motorist the perception that something is happening here; the road is different from its normal circumstance; be aware, reduce speed, be aware." 94In cross-examination, Mr Coffey agreed that it was a matter for the Works Manager to know which signs to use for the particular roadworks being undertaken. He was asked: "Q. From your perspective as his supervisor, can you see any explanation for their non-use? A. No, it makes no sense to me at all that one wasn't used." (emphasis added) 95The trial judge sought clarification as to whether Mr Coffey was referring to one or both of the signs. He replied: "Both of them, your Honour, but specifically, the 'Slippery Road' sign. It's a standard sign that's used in this situation." 96Her Honour at [26], correctly summarised the effect of Mr Coffey's evidence as follows: "... because aggregate is spread on the road surface as an integral part of a road resurfacing operation, the 'Slippery Road' sign is essential to warn of the risk of a loss of traction, irrespective of a driver being alerted to the presence of roadworks that suggested resurfacing because of the sign advising that there were no centre lane markings. He said the 'Slippery Road' sign is designed to alert a driver to a reduction in traction where a driver would otherwise be entitled to expect good traction on a sealed road. He likened the unswept and dressed road surface to 'walking on marbles'." (emphases added) The expert evidence 97Both the appellant and the Council retained expert engineers to investigate the circumstances of the accident. Their reports were prepared six and seven years after the accident. Each expert had access to the evidence referred to above. 98Mr Johnston, retained by the appellant, provided reports dated 25 July 2010 and 12 May 2011. Mr Stuart-Smith, retained by the Council, provided a report dated 10 November 2010. The experts prepared a joint expert report dated 14 January 2012 in which they provided answers to specific questions posed for their consideration. Both gave oral evidence in accordance with the practice of experts giving evidence concurrently. 99The experts agreed that the evidence indicated that the path of the vehicle from the commencement of the first patch of new work was as follows: "The vehicle appears to have deviated from the left [correct] side of the road within the first patch to the right [incorrect] side of the road at the end of the second patch. The evidence for this is a possible/probable (RSS/GJ) tyre mark [on the first resurfaced section] shown in some diagrams and possibly evident in a Council photograph. On reaching the right hand side of the road, (towards the end of the second patch about 60 m before the POI) the vehicle deviated to the left. The evidence for this is the observation by Police and others of a faint tyre mark on the extreme right side of the road towards the end of the second patch. The mark is possibly evident in a Police photograph. The vehicle then passed on to an old section of pavement and crossed substantially back onto the left side of the road before deviating sharply to its right. The vehicle commenced to rotate clockwise and passed back onto the right side of the road and collided in a lateral impact (left side leading) with a tree located a metre or two off the right side of the road. The tree was located approximately level with the start of a new resealed surface. The evidence for this is the presence of diverging tyre marks shown in Police photographs, commencing about 45 m before the POI, as well as vehicle damage." 100At [2], her Honour stated that it was common ground Ms Paterson was travelling at, or slightly under, the speed limit. On the evidence, that speed was 88 km/h or higher. Mr Johnston was of the view that the speed was more likely to have been closer to 100 km/h. Mr Stuart-Smith stated that the speed could have been in excess of 100 km/h. He accepted that the speed could have been around 100 km/h and that "higher as well as lower speeds cannot be ruled out". 101Mr Johnston and Mr Stuart-Smith agreed that the mark shown on photograph 19 in the first section of new work was probably the mark observed by Mr Ellis as recorded in his report of the accident. Mr Stuart-Smith was not able to say whether that mark was made by Ms Paterson's vehicle. He said: "I really couldn't say that. We'd have to know, were they at the scene before or after the police, or before or after the rescue vehicles have been going all sorts of different directions, and that's not information I have. Your Honour may be presented with that at some time?" 102Mr Stuart-Smith later agreed with the proposition put to him by the trial judge that, although he did not consider the material in the sketch drawn by Mr Crisp and the observations made by Mr Ellis to be entirely irrelevant, and would not discount them, they did not, in the words of the trial judge, "compel that level of comfort that [enabled him] to safely conclude that the marks were those left by Ms Paterson's car". 103Mr Johnston, however, considered (by reference to Mr Crisp's sketch) that the tyre marks on the first section of new work were aligned with the marks seen on the second section of new work and through the unworked section leading up to the point of impact. Mr Johnston noted that this is what Mr Ellis had described in his report and that Mr Crisp's sketch was "pretty close to the police diagram" as well. 104Apart from the above, there was no or limited agreement between the experts relating to how or why Ms Paterson's vehicle left the road. Their respective opinions as stated in the joint report were as follows: "Opposing Views Mr Stuart-Smith is of the view that there is no clear evidence of the vehicle's movement from the first patch to the point where it reached the [incorrect] side of the road. The possible initial tyre mark was not recorded or photographed by Police and the photograph of the possible mark is not clear. Mr Johnston agrees that the photographic evidence is not definitive but does tend to support the view of the Council Officer who drew the contemporaneous diagrams showing what was understood to be his observation of assumed tyre tracks." 105The experts also disagreed on what the evidence demonstrated as to where Ms Paterson lost control of the vehicle. 106Mr Johnston's view, as expressed in the joint report, was as follows: "Mr Johnston is of the view that the driver has commenced to lose control around the start of the first section of reseal within the right horizontal curve. The evidence of this is the contemporaneous diagrams by a Council officer, the possible indication of a tyre mark within the Council photographs and the known passage of the vehicle from the left side of the road to the right side of the road commencing within a right horizontal curve consistent with the Council officers drawing of the initial tyre marks." 107It should be noted that Mr Johnston explained in his oral evidence that by "loss of control" he meant that the vehicle was "not going in the direction or the manner intended by the driver". 108Mr Stuart-Smith's view on this question in the expert report was: "The earliest clear evidence of loss of control is at the start of the tyre marks located about 45 m before the POI. At this point the vehicle was commencing a clockwise yaw. However, it is probable that the vehicle was out of control about 10 to 20 m earlier at the location of a possible tyre mark at the end of the second patch. It was likely that the vehicle was over-steered to the left at this point. There is no evidence of loss of control at the curve overlapping the first patch, as suggested by Mr Johnston. The factors identified by Mr Johnston are not evidence. Had the vehicle been out of control giving rise to an oversteer to the right (as would be necessary for it to continue towards the POI), evidence, such as tyre marks and windrows on the loose material of the first patch would have been present. No such evidence was recorded by Police or photographed. The long distance of over 100 m from the curve to the point on the right side of the road where the vehicle deviated left is generally inconsistent with it having been out of control. Mr Johnston's suggestion that the deceased lost control at the curve on the first patch is not supported by evidence where evidence would have been expected." 109Before turning to the cross-examination of the experts, there are other aspects of the expert evidence to which reference needs to be made. One relates to the loss of traction. The other relates to the speed at which the curve in the road could be safely negotiated. Mr Johnston's evidence 110From the material available to him, Mr Johnston did not consider that the extent of the loose gravel on the new surface was excessive and thought it was commensurable with a typical spray sealing operation prior to sweeping. The final sweeping was scheduled for the morning following the accident. This was in accordance with industry standards. However, Mr Johnston was of the opinion that "[a]ny loose aggregate chippings ... can ultimately present a low traction risk to road users". He considered that restricting vehicle speed during the period after spraying the bitumen and before sweeping was one of the "fundamental ... principles to be followed" (Mr Johnston's emphasis), as was stated in both historical and then current guidance documents produced by the RTA, NAASRA/Austroads and the Australian Asphalt Pavements Association (AAPA). 111Mr Johnston was firmly of the view that a 60 km/h speed limit should have been incorporated into the design of the Traffic Control Plan. In this regard, he considered that provisions in the manual that specified speed reduction signs for this type of road work were mandatory. He was also of the opinion that the "Slippery Road" sign should have been incorporated into the Traffic Control Plan and implemented. 112Mr Johnston considered from "all descriptions of the tyremarks that this event [was] initiated by a loss of traction". He was also of the opinion that in "normal circumstances", a driver of Ms Paterson's experience should have been able to negotiate the corner at 100 km/h, but that the presence of gravel reduced the available friction. This made it more difficult to negotiate the corners in the roadway where the works were being undertaken. 113Mr Johnston stated that the presence of the gravel on the roadway would have reduced the available friction, making it more difficult to negotiate bends in the road and "far more likely that a vehicle may lose control at 100 km/h". Mr Johnston considered that until there had been a final sweeping of loose aggregate the appropriate speed limit was 60 km/h. In his opinion, the bends in the road could have been negotiated at a speed of 60 km/h and that there was little chance of loss of control at that speed. 114Mr Johnston considered that, at a speed of 100 km/h, the presence of the gravel would have made it much harder for a driver to regain control of the vehicle. Braking hard would have exacerbated the problem, probably causing the vehicle to slide off the roadway. The fact that new and old surfaces were interspersed meant that the vehicle would have been transitioning between low traction on the new work and high traction on the old surface. He reiterated that there was little chance of loss of control at a speed of 60 km/h and if control was lost, there was a far greater chance of regaining control of the vehicle without mishap. 115Mr Johnston considered that the likelihood that a vehicle would lose control on the loose aggregate at a speed of 100 km/h had been demonstrated by the fact that two drivers, including Ms Paterson, had lost control on this section of the roadway in a 12 hour period. The other incident had occurred the previous evening, when a vehicle driven by a Ms Skorulis, whilst travelling in a northerly direction (the opposite direction from Ms Paterson's direction of travel), left the road on a similar resealed section of the roadway, approximately 150 m south of Ms Paterson's accident. Ms Skorulis was reported to also have been travelling at 100 km/h. It will be necessary to return to the evidence relating to that accident. 116Mr Johnston also referred in his report to NAASRA's requirements for 'bituminous surfacing sprayed work': see at [74] above, and commented: "Where it is stated that '... The road should not be opened to unrestricted traffic until the loose aggregate has been removed', it is my professional understanding that this means that the 'normal' speed limited for a location (in this case 100km/h) should not be restored (returned) until the road authority is certain that loose aggregate has been removed and does not present a risk to subsequent road users travelling under such conditions." 117Mr Johnston considered that it was not common practice for resealing to be carried out in non-contiguous sections as occurred here and said it was poor practice. In his opinion : "... continuous lengths [were] more commonly specified to ensure a homogenous road surface where a consistent level of skid resistance and surface texture is provided to the road user, rather than a road environment where differentials in skid resistance and surface texture will be common place with resultant implications for vehicle dynamics." 118Mr Johnston's explanation of friction and the consequences of loss of friction, both generally and in relation to what was likely in respect of this accident, was, relevantly, as follows: "9.48 A vehicle travelling along a roadway has a certain limiting friction available at the tyre/pavement interface. This friction supply can be consumed in either a longitudinal or lateral direction or a combination of both. This is known as the friction circle concept whereby utilising all of the friction laterally is called critical speed or longitudinally is emergency braking. The radius of the friction circle is the available friction at the tyre pavement interface which for a clean sealed pavement would be about 0.7g but on a surface with some contamination by loose gravel could be as low as about 0.4g. 9.49 The slip angle relationship ... of a typical understeering vehicle also dictates that when brake force is applied to a vehicle in a turn it requires an increase in the slip angle of the vehicle causing the vehicle to rotate further in a clockwise direction within a right turn. If the driver then applies the brake more firmly further rotation can occur until the combined friction exceeds the outer radius of the friction circle and loss of control occurs. 9.50 This is far more complicated when the vehicle also enters a section of loose gravel within the curve such that the overall radius of the available friction circle is reduced. Therefore the driver of a vehicle which may have been operating well within comfortable performance limits on the uncontaminated road surface suddenly finds themselves operating uncomfortably close to reduced performance thresholds on the contaminated surface which therefore has reduced frictional capacity. ... 9.51 It is for this reason that a reduced speed limit is required whilst the surface remains contaminated by loose aggregate prior to final sweeping. 9.52 In this instance the effect would have been further complicated by the vehicle passing from one surface to another on a repetitive basis as the driver was potentially trying to regain control of the vehicle. The driver and more importantly the vehicle response from one surface to another would have changed significantly based on the varying amounts of longitudinal and lateral friction capacity and the different slip angle relationships the vehicle would have tried to establish for the different surfaces. 9.53 A driver typically operates a vehicle only within the very small central area of the available friction circle. Operating outside of this area is both unfamiliar and uncomfortable to most drivers. It is likely in this instance that the unexpected change in friction capacity from the uncontaminated to the contaminated surface has caused [Ms Paterson] and probably Ms Skorulis to suddenly find themselves operating in the unfamiliar and uncomfortable middle or outer ring of the friction circle (usually felt by increased drift or slip angle). The instinctive response of most typical drivers would then be to brake within the turn which often only exacerbates the problem by taking away more of the lateral capacity to be used in longitudinal braking. In a car with a front wheel brake bias this causes the vehicle to plough or in this case cross to the incorrect side of the roadway ... 9.54 It was therefore my opinion that the suddenly reduced friction capacity within the horizontal curve without prior speed restrictions has created an uncomfortable driving situation for [Ms Paterson] due to a sudden loss of friction capacity. 9.55 This has probably caused her to make an instinctive braking response which has then exacerbated the problem and sent her vehicle into a ploughing type response towards the incorrect side of the roadway. She has seemingly then attempted to correct the path of the vehicle but probably due to the continually changing road surface and friction regime has been unable to do so before ultimately leaving the roadway and striking the road side tree. 9.56 The reduced friction capacity and the initial presence of the curve have probably influenced her not to adopt the option of simply braking heavily at this point for fear of leaving the roadway and possibly striking other roadside objects. It appears that [Ms Paterson] has attempted to bring her vehicle under control within the road surface but the vehicle has eventually left the roadway and struck a roadside tree." 119Mr Johnston dismissed Senior Constable Brand's opinion that Ms Paterson fell asleep. Mr Johnston pointed out that Ms Paterson had "just actively steered her vehicle through at least all or part of the right horizontal curve" and had also apparently "responded on a number of occasions by means of steering input" during the period that she had lost control of the vehicle. 120Mr Johnston also rejected the likelihood of Ms Paterson having either "momentarily lost attention" or having "lost voluntary control of the vehicle by tiredness", being possibilities suggested by Mr Stuart-Smith. Mr Johnston explained: "... a fatigue crash by definition ... is a vehicle that leaves the outside of a curve, so it is generally a driver fails to steer a corner as opposed to oversteers a corner. In this case already part of the right curve is negotiated before reaching the unsealed surface. That suggests vehicle control to that point, then it suggests additional control as opposed to a lack of control which is the normal criteria for fatigue, a vehicle will travel off the outside of a curve or fail to negotiate a curve or simply travel straight ahead." 121Mr Johnston concluded, therefore, that: "... the road surface condition around the material location and time was a contributory factor in the incidents involving Ms Paterson and Ms Skorulis when traversed at 100 km/h." 122Mr Johnston maintained these conclusions in his supplementary report prepared after receipt of Mr Stuart-Smith's report. In the supplementary report, Mr Johnston also expressed his opinion that: "The risk of loss of traction is very significantly increased at 100 km/h on this section of roadway as compared to 60 km/h if a reduced roadworks speed limit had been in place." 123Mr Johnston expressly disagreed with Mr Stuart-Smith's discussion of general friction supply and demand. This is discussed below at [125]-[126]. Mr Johnston considered that Mr Stuart-Smith's view was based on "static or theoretical average values as opposed to the real world loading in this type of situation". He referred to his own experience in testing vehicles at the BMW test facility in South Carolina, USA and his knowledge of the reactions of other experienced drivers in the same group. He stated in his report: "4.3 ... that even experienced drivers when required to brake in a turn can slightly over apply the brakes and achieve instantaneous peaks at values in excess of 0.3g. 4.4 Indeed an experienced test driver who was in my group did this on a regular basis just by his nature of driving. 4.5 This instantaneous peak loading in the real world creates slippage which resolves itself as discomfort in the driver and induces a greater response which can ultimately lead to a loss of control. 4.6 I was not there when this incident unfolded nor is there specific evidence of the course or series of events undertaken by the driver of the involved vehicle. It is known that she has travelled over the section of recently sealed pavement and contemporaneously lost control of her vehicle. As noted these types of incidents are more typically due to driver responses to the development of an uncomfortable situation as opposed to reaching a theoretical maximum static threshold as calculated in [Mr Stuart-Smith's] report. 4.7 To my knowledge there is no evidence of possible involvement of impact from another vehicle, there is no evidence of a sudden swerve to avoid another vehicle or object. Therefore loss of control can only arise from an ultimate loss of traction with the prevailing road surface. 4.8 The risk of a loss of traction is very significantly increased at 100 km/h on this section of roadway as compared to 60 km/h if a reduced roadworks speed limit had been in place." 124It is to be remembered, of course, that Mr Coffey strongly disagreed that the curve could be negotiated safely at 145 km/h (see [92] above). Mr Stuart-Smith's evidence 125Mr Stuart-Smith agreed that it appeared from the photographs that the loose aggregate on the road was not excessive. He described the photographs as depicting a "minimal amount of loose material, consistent with a new surface prior to it having been swept". Mr Stuart-Smith, in his report, disagreed that the appropriate speed for this section of the roadway whilst the roadworks was being undertaken was 60 km/h. He modified this view, at least to some extent, in the course of cross-examination. However, his initial view remains important, because it formed a significant part of the basis for his opinion that the accident was not caused by the reduction in friction on the road surface due to the road works. It had been his view, never completely resiled from, that the first section of new work could have been driven at a speed of at least 145 km/h without there being any sideways slippage. Importantly, he considered that: "Based on the radius of the curve (360 m) and a likely coefficient of friction of about 0.5 to 0.65, the critical speed of the curve was likely to have been about 160 to 180 km/h ... For the minimum likely friction coefficient suggested by Mr Johnston (0.4), the critical speed of the curve with the resealed surface would have been about 145 km/h. This means that, for speeds less than 160 km/h (or at least 145 km/h), vehicles should have been able to travel around the curve whilst following the road alignment without slideslip occurring. In other words, at the speed limit of 100 km/h, travel on the reduced friction surface of the unswept resealed surface should not have been of concern." 126Mr Stuart-Smith explained that the co-efficient of friction "is the ratio of the retarding force due to friction and the weight of a vehicle on a flat surface". Mr Stuart-Smith posited that the evidence was: "... consistent with [Ms Paterson] having had difficulty with the driving task, with the circumstances being consistent with inattention, fatigue, an adverse medical condition or similar factors." 127Mr Stuart-Smith also dealt with Mr Johnston's 'hypothesis' as to how the accident occurred, namely, that the reduced friction created an uncomfortable driving situation, probably causing Ms Paterson to brake instinctively (see [118], paras 9.54 and 9.55). He stated that a driver was not "usually aware of reduced friction unless the demand for the friction exceeds the available friction". Indeed, Mr Stuart-Smith considered that if Mr Johnston's postulations were correct, it was indicative of imprudent driving on the part of Ms Paterson. However, based on Mr Johnston's hypothesis, Mr Stuart-Smith accepted that Ms Paterson: "... must have applied excessive braking whilst travelling at a relatively high speed around the curve, to have lost control as a result of inadequate friction." He added, however, that there was: "... no evidence (where evidence would have been expected) that this was the cause of the crash. ... Whilst a precise cause for the crash cannot be determined, driver related factors were likely to have been the case. There is no evidence that the resealing operations even played a role in the crash." 128The 'evidence' that Mr Stuart-Smith suggested would have been present, had inadequate friction played a role in the accident, was evidence of sideslip marks in the loose material in either of the two resealed surfaces that Ms Paterson travelled over up to the point where she approached the incorrect side of the road, where he considered that "the preliminary tyre mark was reported to have been located". 129Mr Stuart-Smith reiterated in his report that: "... there was ample friction supply on the resealed surface to cater for drivers travelling around the curve at 100 km/h with an adequate safety margin." 130He also considered that it would have been a potentially dangerous manoeuvre had Ms Paterson sharply applied her brakes as she entered onto the new surface. He stated that it was "common knowledge that, in a car without ABS brakes, this is a potentially dangerous combination ... on any surface". 131Mr Stuart-Smith, in section 12 of his report, stated: "In summary, there is an absence of evidence of [Ms Paterson] having lost control on a resealed surface as a result of inadequate friction before she approached the right side of the road. Had inadequate friction played a role, evidence of sideslip tyre marks in loose material would have been expected to have been observed by Police and to have been present in the photographs. As a consequence, and given the distance from the curve of the location where [Ms Paterson] approached the right side of the road, on the balance of probabilities, [Ms Paterson] drifted or steered to her right for a driver related reason (such as fatigue, inattention or medical reasons), possibly on the path indicated by Mr Ellis." It is apparent from his preceding analysis that the reference to "the right side" in this summary is a reference to the right side of the road immediately prior to the point of impact. 132In the conclusion to his report, Mr Stuart-Smith repeated that there was no evidence as to why Ms Paterson lost control. He stated: "Road surface condition may have been a factor. However, if this was the case, it would be indicative of a driver related factor predominating or of imprudent driver actions in the circumstances." 133Mr Stuart-Smith's conclusion was based upon his opinion that there was no need for a reduced speed sign and that a vehicle could negotiate the curve at the commencement of the new work at a speed of at least 145 km/h. 134Mr Stuart-Smith also commented upon the manner of resurfacing by resealing the roadway in non-contiguous sections. In his opinion, this was not normal practice but was nonetheless acceptable. Evidence of Ms Skorulis' accident 135I have referred above, at [115], to Ms Skorulis' accident. It is convenient to refer to it further at this point. There was no question that Ms Skorulis' accident, which occurred at about midnight the day before Ms Paterson's accident, was caused by the vehicle she was driving losing traction on the loose gravel such that she lost control of her vehicle and ran off the road. However, there was an issue as to the admissibility of evidence relating to Ms Skorulis' accident. 136Ms Skorulis was 19 years of age at the time of her accident, with approximately a year's driving experience. She had been travelling in a northerly direction at 100 km/h. She remembered hitting gravel and ending up in a ditch. Immediately before leaving the road, she was negotiating a curve to the right. Her vehicle went off the road to the left, approximately 30 to 40 m after she entered onto the section of the roadway with loose gravel. In her evidence, Ms Skorulis said she could not recall whether she attempted to correct her direction of travel but from memory she said she did not brake. However, in her statement to the police given at the scene of the accident, she said that she: "... hit a patch of road that had loose gravel on the road, and the steering wheel started to move from side to side. I tried to correct it but I couldn't and the ute went off to the left ..." 137Both experts gave evidence on the voir dire in respect of Ms Skorulis' accident, particularly directed to the question whether there were similarities between the two accidents such that a conclusion might be drawn that the cause in each case was the same. 138In Mr Stuart-Smith's opinion, the similarities between the two accidents were limited to the fact that both vehicles had entered onto a resealed section of road and both had left the road. The dissimilarities, however, were that: "... Miss Skorulis clearly left the road within a curve and she left it on the outside of the curve so that is a typical path for a vehicle that has failed to take the curve ... In Miss Paterson's case, she clearly succeeded in rounding the curve because ... she left the road on the opposite side to the outside of the curve, she left the road on the same side as the inside of the curve but in fact she left the road some one hundred and fifty metres ... after the curve on the right-hand-side of the road which is the same side as the inside of the curve so quite different ... vehicle dynamics involved." 139He added that a slide to the inside of the curve could be consistent with lack of friction in the road surface: "... because if a driver loses friction in the road surface and slight sliding towards the outside of the curve and the vehicle slows sufficiently for friction to be regained, you can then be oriented towards the inside of the curve and then the driver might well get into difficulties heading towards the inside." 140Mr Stuart-Smith agreed that inadequate friction within the curve could not be excluded as a factor in causing a vehicle to lose control. It is apparent from his evidence on the voir dire, as elsewhere, that he considered that there was an absence of evidence of Ms Paterson having lost control in the first section of new work, as there were no close up photographs of tyre marks in the gravel. To the extent that the photographs may have shown a tyre mark, he did not consider that there was evidence of lateral displacement by way of a build up on the outside of the tyre marks in the loose gravel. 141Mr Stuart-Smith did not necessarily accept that what the Council officers had recorded as to the presence of a tyre mark indicated loss of control. He reiterated that the tyre marks would have to be associated with the build up of material on the outside to indicate some degree of side slipping. He said if that was the case, it could be evidence of loss of control. He accepted, in cross-examination, that there could be varying degrees of loss of control, but insisted that there was no evidence, in the case of Ms Paterson's accident, of any loss of control. 142Mr Stuart-Smith's conclusion was that even if there were some similarities, there was sufficient dissimilarity, as Ms Paterson had succeeded in passing through the curve before the loss of control occurred, to make any reference to that accident irrelevant. In the context of the whole of Mr Stuart-Smith's evidence, his reference to the place where the loss of control occurred was on the second section of unsealed work immediately before leaving the roadway. 143Mr Johnston was critical of the manner in which "loss of control" was being discussed. He stated: "Everyone is defining the negative. A vehicle in control is travelling on its intended course and its intended manner so it is rounding a curve at its intended speed maintaining its intended path presumably on its correct side of the roadway. In this instance the evidence suggests that after entering the section of loose material in rounding this curve [Ms Paterson's] vehicle has commenced to travel towards its incorrect side of the roadway where it ultimately left different, various pieces of evidence along its path and then ultimately left the road side of the roadway and impacted with a tree." 144His evidence continued: "Q. ... what you see in photograph 19 and 20, does that show lateral displacement of loose material, can you say or not? A. Yes, because lateral placement of loose material means that the material is deposited out of the wheel path. The reason you can see that if there is less aggregate in that path. It has been displaced out. Q. Does one need skidding for that to occur? A. No, I heard that evidence earlier. Material is displaced out by tyre forces pushing the material sideways. I don't agree that in a high speed rotating tyre that you get this row of two or three stones right on the lateral edge. The stone is moved with some velocity from under the tyre and is thrown out. That's why you have a warning about windscreens damage, stones are thrown reasonable distances." 145Mr Johnston confirmed that he had no doubt that there was "some clear evidence [in photographs 19 and 20] of lateral displacement". 146The trial judge asked questions as to what was shown on the photographs as being equally consistent with Ms Paterson having "momentarily lost attention" or having "lost voluntary control of the vehicle by tiredness", as from a loss of control through a loss of traction with the road surface. Mr Johnston responded with the explanation set out at [120] above. 147The trial judge admitted the expert evidence given on the voir dire relating to Ms Skoluris' accident as evidence in the proceeding being relevant to a fact in issue, namely, how Ms Paterson's accident occurred. See also Amalgamated Television Services Pty Ltd v Marsden [2002] NSWCA 419 at [202]-[204]; MM Constructions (Aust) Pty Limited and Anor v Port Stephens Council (No 3) [2010] NSWSC 243 at [49]. Cross-examination of the experts 148During the course of cross-examination of the experts, Mr Stuart-Smith altered his opinion in relation to the signage that was appropriate having regard to the roadworks that were being undertaken. In particular, he stated that it would have been better to have a reduced speed sign. He also gave evidence as to the risks of travelling on the new work at a time when there was still loose gravel on the surface. This evidence, although directed to breach of duty, is also relevant to the question of causation and needs to be considered in full. 149The questioning commenced with seeking Mr Stuart-Smith's view as to what the "Chip Damage" sign, which was in place, would convey to a driver. He said that that sign would warn that there was "gravel, loose stones". The questioning, including by the trial judge, and his answers continued as follows: "HER HONOUR: What is the driver to draw from that, that there is risk of windscreen damage? WITNESS STUART-SMITH: Whatever might happen when you drive on loose stones. ... HER HONOUR: Yes, but the risk that there being damage occasioned either to the driver or another driver? WITNESS STUART-SMITH: Well I can't obviously agree that that would be the case. HER HONOUR: Why can't you agree with it? WITNESS STUART-SMITH: I do agree, I obviously agree that would be the case. HER HONOUR: It doesn't do more than that, though, does it? WITNESS STUART-SMITH: I believe it does. HER HONOUR: What do you say it does more than that? WITNESS STUART-SMITH: I believe it would convey that a driver would be aware, the fact that there are loose stones and would be conscious of whatever other risks are associated with loose stones. HER HONOUR: What are those? ... WITNESS STUART-SMITH: If you were to undertake an unconventional manoeuvre there may be the possibility of reduced friction. HER HONOUR: Only by reason of an unconventional manoeuvre? Or by travelling fast? WITNESS STUART-SMITH: Yes. HER HONOUR: Yes what? WITNESS STUART-SMITH: Yes, yes. By travelling fast. HER HONOUR: Where is there in the ['Chip Hazard' sign], a risk of harm if speed is not decreased? ... WITNESS STUART-SMITH: Quite clearly the picture doesn't show that, it is the implications a driver would have to take from that. HER HONOUR: It would be an implication a driver would have to take? ... WITNESS STUART-SMITH: ... I would say that a driver would take from [the 'Chip Hazard'] sign that there is a risk of windscreen damage quite clearly because that is what the sign is showing but there are possible other risks. ... what I am saying is a driver should take on board the fact that there could be other risks which may be unknown. ... HER HONOUR: There was no necessity in your view Mr Stuart-Smith for any warning in the most direct and glaring terms to reduce speed either by reduce speed, those words, or a reduced speed designation, namely 60 km an hour or 40 km an hour or something way less than 100 km an hour. Do you stand by that view? WITNESS STUART-SMITH: Your Honour used the word necessity, and necessity is, I would have come to this point, the necessity is based on what is specified, and in those terms, there was no necessity, whether or not-" 150As is apparent from the totality of his evidence, Mr Stuart-Smith's reference to the "necessity ... based on what is specified" was based on his opinion that the Council had used the correct TCP, which did not specify that a "Reduce Speed" sign should have been placed in position prior to the commencement of new work, but did specify that a sign indicating there was loose gravel on the road be erected. The specification of this latter sign had been crossed out by the Works Manager, as discussed below. The questioning continued: "HER HONOUR: I am talking about what your opinion is? WITNESS STUART-SMITH: My opinion is it would have been better to have a reduced speed sign. HER HONOUR: Why is it better to have a reduced speed sign? WITNESS STUART-SMITH: Well, a driver getting into trouble. HER HONOUR: What sort of trouble Mr Stuart-Smith? WITNESS STUART-SMITH: A driver that could lose control in loose gravel. HER HONOUR: And speed is an obvious additional risk where there is a risk of loss of traction by reason of surface gravel. Is that not the case? WITNESS STUART-SMITH: I have said so already your Honour in previous evidence, yes, of course it is. ... [Counsel for the appellant]: The reason, Mr Stuart-Smith, for reduced speed on these roads that have been subject to resealing is similarly the danger to road users by not being aware of the change in the frictional control available to them, isn't that right? WITNESS STUART-SMITH: There are two reasons, that is one potential reason, and the other is possibly road workers which is not applicable in this circumstance. [Counsel]: There aren't road workers on there in the period between the resealing and the brushing? WITNESS STUART-SMITH: As I said, yes. MARSHALL: Primarily it is for the protection and safety of motorists? WITNESS STUART-SMITH: In - yes that's correct." (emphases added) 151Mr Stuart-Smith was further cross-examined as to the reason for the incorporation into the NAASRA technical report of the guideline that, following the spraying, the bitumen aggregate could be pressed into the surface by vehicles travelling at slow speeds, and a similar guideline in Austroads 2004 edition that "traffic speeds should be controlled whilst there is loose aggregate on the surface to less than the speed which will cause stones to fly high enough to strike windscreens". The cross-examination produced the following evidence: "[Counsel]: The very reason for that proscription [sic] is both to protect cars from broken windscreens, correct? WITNESS STUART-SMITH: Yes. [Counsel]: And to reduce speed so that cars do not lose control from passing on to the resealed sections where there are loose stones? WITNESS STUART-SMITH: Yes, to minimise that." (emphasis added) 152Mr Stuart-Smith said that the chip hazard sign did "[n]ot directly" say anything about the possibility of slipping or sliding, or the loss of traction as a vehicle went from one section of the new work to another. 153There was further cross-examination as follows: [Counsel]: But you, of course, agree there is nothing that warns the motorists of the impending nature of what Miss Paterson was about to engage with as far as new work was concerned? WITNESS STUART-SMITH: I wouldn't be that strong, that there was nothing. I would agree that the adequacy was not sufficient. [Counsel]: That said, if speed had been reduced to sixty kilometres an hour, that would have significantly reduced the risk to Miss Paterson, wouldn't it? WITNESS STUART-SMITH: Sixty, it would reduce the risk compared to the risk she would have had at a hundred kilometres an hour. ... WITNESS STUART-SMITH: Of any untoward event happening in loose gravel. ... [Counsel]: If her speed was reduced to sixty kilometres an hour she would have been able much better to respond to loss of control if there were any? WITNESS STUART-SMITH: Correct. ... HER HONOUR: Do you agree or disagree with this proposition, there are road safety consequences of discontinuous pavement textures and friction supply characteristics both during construction where a driver is transitioning from good traction to poor traction and back again ... what do you say to that proposition? WITNESS STUART-SMITH: ... I think I would agree to some degree with that. HER HONOUR: You would agree on this particular stretch of roadway on the Kingsvale Road, there were by reason of the discontinuity between the sealed and or the worked and un worked sections, road safety consequences because of the differing friction supply characteristics of the road surface? WITNESS STUART-SMITH: Yes, to a minor degree. ... WITNESS STUART-SMITH: ... in that, at subsequent curves, at the subsequent curve, it would be advisable to evaluate whether there was a likelihood drivers had sped up in-between on a good section of roadway. So that they were going at a faster speed when they encountered loose gravel at a subsequent curve. ... HER HONOUR: In this instance then, proceeding on the assumption which I think is the most concrete of all working assumptions that there was nothing in and of itself to indicate to a driver that something very much less than the speed limit was called for, for a driver to then travel at speed, at or near the speed limit over discontinuous segments of worked and un worked road, rendered or produced, was productive, was it not, of adverse road safety consequences by reason of the friction supply, do you agree with that? WITNESS STUART-SMITH: I partially agree your Honour, because as I said, the friction supply is only really necessary on curves ... ... HER HONOUR: ... Any layer of gravel on a road surface reduces the coefficient of friction does it not? WITNESS STUART-SMITH: Yes, your Honour. HER HONOUR: It makes curves more dangerous, you agree with that? WITNESS STUART-SMITH: Yes. ... [Counsel]: Then the fact that [Ms Paterson is] travelling at or about 100 kilometres an hour when she encounters the second patch, trying to correct, poses increased dangers for her, does it not? WITNESS STUART-SMITH: Yes, well, in those circumstances, she would be trying to manoeuvre in, in a hard steer, and yes, then in those circumstances, it could be a factor. HER HONOUR: The discontinuity between-- WITNESS STUART-SMITH: Yes it would be. HER HONOUR: ...The faint so called, or I think you referred to it as a possible tyre mark in the second worked section, is a, I think you have proceeded on the assumption, was left by Ms Paterson's vehicle, and she was, on any view on that stage, on the incorrect side of the road? WITNESS STUART-SMITH: Correct. HER HONOUR: She then, as we know, travelled on to an un worked section where she made valiant but unsuccessful attempts to reorient the vehicle in a forward going motion? WITNESS STUART-SMITH: Yes, your Honour. HER HONOUR: The transit from the incorrect side of the road under a steering manoeuvre, back on to the correct side of the road between the discontinuous parts of worked and un worked roadway, were themselves, or was itself, a factor potentially adding to her incapacity to regain control or at least correct a steering manoeuvre. Is that agreed? WITNESS STUART-SMITH: I see. So, well it could have been." 154Mr Stuart-Smith agreed in cross-examination that photographs 19 and 20 showed what Mr Ellis had described in his report, namely, that on the first patch of new work there were some "light marks where it appears the vehicle attempted to correct its path". 155The final position of the experts at the conclusion of their cross-examination was as follows: "[Counsel]: Is, can I then say on the balance of probabilities, what is available to this Court to which I have just referred you, consistent with a loss of some control, at or near the beginning of the first patch, a conservative correction through the first patch into the second, and a subsequent catastrophic loss of control? WITNESS JOHNSTON: Yes. [Counsel]: In fairness, I should ask the same question to Mr Stuart-Smith. Do you accept on the balance of probabilities that scenario? WITNESS STUART-SMITH: No, I don't accept that on the balance of probabilities at all. HER HONOUR: It's a possible scenario. WITNESS STUART-SMITH: A possible scenario. HER HONOUR: Is there a more probable scenario in your view? ... WITNESS STUART-SMITH: A more probable scenario is that the driver deviated for an unknown reason. The reasons for that, are that it's only a balance of probabilities in my view, in that we have the statement by Mr Ellis where he specifically said he couldn't see any, what he called skid marks ... We then come and look at some objective sort of factors, and that is, that even if you use the minimum likely friction, reduced friction, as proposed by Mr Johnston, it was traversable at a very high speed, admittedly, with constant steering and without braking at those high speeds, and Mr Johnston made that point, and I accept those qualifications. But nonetheless, even if you then apply various other actions which may well have reduced - had the effect of reducing the friction such as braking, such as transitioning from one side to another, it's still traversable, should be comfortably traversable at 100 kilometres per hour without loss of traction, and so given that on one hand, the absence of objective evidence, of loss of traction, and on the other hand, the theoretical calculations that suggest that it was traversable, comfortably traversable at high speed - when I say that high speed, I'm talking 100 kilometres per hour - say those two elements add up to a balance of probabilities that she lost, she deviated for a reason other than loss of traction. HER HONOUR: What if I added this into your assessment, that despite the theoretical possibility of safe passage through the curve at 100 kilometres an hour, the loss of traction under that speed on this gravel, was the extent of this gravel, was such as to render that theoretical possibility far less likely, such that there was in fact a loss of traction or it was not safely - the curve was not safely traversed at that speed - would that alter your view as to the probabilities of control having been compromised, leading to the driver not maintaining her orientation on the correct side of the road? WITNESS STUART-SMITH: Well, I think that would - that obviously, that reduces the degree to which one could say that what my previous opinion - there is another factor too, and I accept that once a driver has lost control, they can do a number of things, but the fact that if we assume this tyre mark to be hers, that she did travel in this relatively straight line for 150 metres, is, on the far side of likelihood for a driver who has just lost control, it's reasonably uncommon for a driver to freeze or just to do that for that length of time, so whilst I couldn't say that a driver wouldn't do that, it adds weight to the likelihood that it wasn't a loss of traction that caused her to deviate. HER HONOUR: But that her right orientated incorrect leaning was due to some other driver impacted environmental factor of some kind? WITNESS STUART-SMITH: Yes, your Honour. Correct." 156The essential difference between Mr Johnston and Mr Stuart-Smith in respect of what happened in the first section of new work was that Mr Johnston considered that there had been a deviation to the right contemporaneously with entry onto the new work and that was consistent with some loss of traction. Mr Stuart-Smith considered that for there to have been loss of traction at that point, a more dramatic change would have been observable in the tyre marks. Reasons of the trial judge on causation 157The trial judge, at [76], described the roadway from the commencement of the new work to where Ms Paterson left the roadway. Her Honour identified the point of impact to be about 210 m from the commencement of the first resurfaced section. 158Her Honour next stated, at [77], that the experts agreed that there was no physical evidence capable of fixing where Ms Paterson was likely to have first lost control of her vehicle over that 210 m. The appellant challenges this finding. 159Her Honour, at [81], observed that in order for the appellants to prove that the failure of adequate signage was causally connected to the collision which occurred it was necessary to establish that Ms Paterson probably lost control in the first resurfaced section of the roadway whilst travelling at, or near, a speed of 100 km/h. 160At [86], her Honour cited with approval and applied the following observations of McDougall J in Nguyen v Cosmopolitan Homes (NSW) Pty Ltd [2008] NSWCA 246 at [55]: "(1) A finding that a fact exists (or existed) requires that the evidence induce, in the mind of the fact-finder, an actual persuasion that the fact does (or at the relevant time did) exist; (2) Where on the whole of the evidence such a feeling of actual persuasion is induced, so that the fact-finder finds that the probabilities of the fact's existence are greater than the possibilities of its non-existence, the burden of proof on the balance of probabilities may be satisfied; (3) Where circumstantial evidence is relied upon, it is not in general necessary that all reasonable hypotheses consistent with the non-existence of a fact, or inconsistent with its existence, be excluded before the fact can be found; and (4) A rational choice between competing hypotheses, informed by a sense of actual persuasion in favour of the choice made, will support a finding, on the balance of probabilities, as to the existence of the fact in issue." 161Her Honour remarked that the application of those principles were of considerable significance in resolving the difficult question of causation in the case. The appellant challenges her Honour's reliance on a test of actual persuasion. 162At [91], her Honour referred to Senior Constable Hando's opinion stated in his report that "the deceased [had] lost control" upon entering the right hand bend at the commencement of the work. However, her Honour considered that his opinion was undermined by his statement that there were "no visible marks on the road surface". This finding is also challenged, as her Honour failed to note that the vehicle had crossed from the correct to the incorrect side of the road. Senior Constable Hando was not called to give oral evidence. No Jones v Dunkel inference was drawn. This is important because, although Senior Constable Hando did not record a tyre mark on the first section of the roadway, Sergeant Brand, in cross-examination, when shown photographs 19 and 20, conceded that he may have missed seeing that mark. Her Honour recorded this concession at [93]. 163The trial judge referred to the sketch drawn by Mr Crisp which her Honour, at [94], recorded as showing: "... a discontinuous set of parallel tyre tracks indicating a path of travel from the left-hand side of the road as the vehicle entered the first resurfaced section towards, and then across the centre line at the end of that section. The path of travel is then picked upon the far right-hand side of the road towards the end of the second resealed section (that is, on the incorrect side of the road) before the vehicle moves back towards the left (that is, towards the correct side of the road) before again crossing the centre line so as to be in alignment with the yaw marks leading to the point of impact. The sketch indicates nothing of the passage of the vehicle over the first unworked section (it being common ground there were no marks of any kind on that section) and nothing of its passage on the second resurfaced section, such as might suggest how or why she was on the incorrect side of the road at the end of that section, at a point 150 metres from where she entered the curve and within 60 metres (or thereabouts) of the point of impact." 164Her Honour found, at [95], that the photographs showed the tyre marks that had been described by Mr Ellis in his report, as Mr Stuart-Smith had conceded, and that those tyre marks were consistent with the path of travel of the vehicle that Mr Crisp had drawn on his diagram. Her Honour concluded, at [95], that these were probably the tyre marks of Ms Paterson's vehicle as she veered across the road from left to right before tracking into the shoulder on the right-hand side of the road. Her Honour then noted, at [98], that even if the light mark on the first section of the new work indicated that Ms Paterson's control of the vehicle was compromised, the experts disagreed whether a precipitating cause of that loss of control was a loss of traction due to the presence of loose gravel or was due to some other cause or a combination of possible causes. 165Her Honour, at [99]-[100], reviewed Mr Stuart-Smith's opinion that the mark in the first section showed a deviation of Ms Paterson's vehicle across to the wrong side of the road in a reasonably straight line without any obvious attempt to correct that course. Her Honour noted that he was of the opinion that the probabilities favoured the cause of the deviation as being unconnected with gravel on the road at an unsafe speed. Mr Johnston, however, did address those possibilities. Her Honour recorded his evidence at [103]. It should be noted that this evidence was not refuted by Mr Stuart-Smith. 166Her Honour commented upon other hypotheses advanced by Mr Stuart-Smith, such as drowsiness or inattention, but noted that he was not invited to comment upon whether the physical evidence supported either hypothesis. 167Her Honour's critical reasoning then appears at [108], as follows: "Mr Johnston's assumption that the tyre marks were present but not seen could only support a finding that they were probably there if there were some evidence that the marks were there to be seen but overlooked. The obvious flaw in his reasoning is that if there were evidence of a yaw mark or sideslip proximate to where Ms Paterson entered the first resurfaced section (whether or not she regained control of the vehicle thereafter) there would be no need to proceed on the assumption that it was overlooked. To the extent that Mr Johnston's opinion that the probable cause of the accident was a loss of control in the first resurfaced section due to a loss of traction is based on the assumption that there either was physical evidence which was missed by all observers, or there might have been evidence that was missed by those people, the reasoning suffers from circularity in the sense that it is assumed that there were marks on the road because one would expect them to be present if there was a loss of control due to lack of traction. I am not persuaded that the theoretical possibility that the tyre marks were missed by those at the scene (and, in the case of Sergeant Brand, missed despite the fact that he was specifically looking for some evidence of the loss of control) permits an inference to be drawn that the tyre marks were there but missed." The appellants also point to error in this part of her Honour's reasoning, contending that she conflated loss of traction with loss of control. 168Her Honour, at [110], dealt with the competing hypotheses advanced by the respective experts, as follows: "I accept that the competing hypotheses for Ms Paterson's loss of control, being a lack of concentration, inattention, drowsiness or fatigue, would logically need to subsist at the very point on the road where there was increased risk of a loss of control by reason of excess speed on gravel. While this state of affairs might undermine the likelihood of either of them (or any combination of them) being the explanation for the loss of control, it is necessary for me to be actually persuaded of one scenario over another as the more probable explanation and I am unable to reach that level of satisfaction." 169Her Honour then concluded, at [111]: "In the absence of any physical evidence of sideslipping due to a clockwise rotation of the vehicle in the first resurfaced section in circumstances where it would be expected to be obvious, and where the road surface was examined and none noted; where the photographic evidence favours a finding of a gradual veering from the correct to the incorrect side of the road before a faint but definite mark 150 metres after the gravel is first encountered and after an intervening section of unworked road without tyre marks; and where Mr Curtis' sketch and Mr Ellis' report are open to interpretation, the balance of the evidence, which includes the evidence of another accident on a stretch of road under the same road repair works 24 hours earlier, simply does not induce in my mind an actual persuasion that the precipitating cause of Ms Paterson's accident was her losing control of her vehicle in the first resurfaced section due to the presence of gravel and her unchecked speed." 170The appellant alleges that this conclusion is also in error. Consideration: causation (a) Did her Honour apply the correct standard of proof on the question of causation? 171The appellant contended that her Honour applied a wrong standard of proof in determining whether the appellant had proved that the accident was caused by the Council's negligence. 172Her Honour relied on the first proposition noted by McDougall J in Nguyen v Cosmopolitan Homes to hold, at [86] and [111], that she was required to have a state of "actual persuasion" before being able to find the existence of a fact, which, relevantly to the issue on appeal, was the cause of Ms Paterson's vehicle running off the road. In Nguyen, McDougall J used the language of 'actual persuasion', by reference, inter alia, to the statement of Dixon J in Briginshaw v Briginshaw [1938] HCA 34; 60 CLR 336. In that case, Dixon J, at 361-362, stated the principle in the following terms: "... when the law requires the proof of any fact, the tribunal must feel an actual persuasion of its occurrence or existence before it can be found. It cannot be found as a result of a mere mechanical comparison of probabilities independently of any belief in its reality. No doubt an opinion that a state of facts exists may be held according to indefinite gradations of certainty; and this has led to attempts to define exactly the certainty required by the law for various purposes. Fortunately, however, at common law no third standard of persuasion was definitely developed. Except upon criminal issues to be proved by the prosecution, it is enough that the affirmative of an allegation is made out to the reasonable satisfaction of the tribunal. But reasonable satisfaction is not a state of mind that is attained or established independently of the nature and consequence of the fact or facts to be proved. The seriousness of an allegation made, the inherent unlikelihood of an occurrence of a given description, or the gravity of the consequences flowing from a particular finding are considerations which must affect the answer to the question whether the issue has been proved to the reasonable satisfaction of the tribunal." 173The standard of proof of facts in respect of which a party bears the onus was again considered by the High Court in Neat Holdings Pty Ltd v Karajan Holdings Pty Ltd [1992] HCA 66; 67 ALJR 170 at 170-171, where Mason CJ, Brennan, Deane and Gaudron JJ stated: "The ordinary standard of proof required of a party who bears the onus in civil litigation in this country is proof on the balance of probabilities. That remains so even where the matter to be proved involves criminal conduct or fraud. On the other hand, the strength of the evidence necessary to establish a fact or facts on the balance of probabilities may vary according to the nature of what it is sought to prove. Thus, authoritative statements have often been made to the effect that clear or cogent or strict proof is necessary 'where so serious a matter as fraud is to be found'. Statements to that effect should not, however, be understood as directed to the standard of proof. Rather, they should be understood as merely reflecting a conventional perception that members of our society do not ordinarily engage in fraudulent or criminal conduct and a judicial approach that a court should not lightly make a finding that, on the balance of probabilities, a party to civil litigation has been guilty of such conduct." (citations omitted) 174The Evidence Act 1995, s 140, which now specifies the standard of proof that applies in civil cases, restates the common law position that the civil standard of proof is on the balance of probabilities: s 140(1). Section 140(2) provides that the matters the court may (non-exclusively) take into account in deciding whether it is satisfied on the balance of probabilities include: the nature of the cause of action or defence (para (a)); the nature of the subject matter of the proceeding (para (b)); and the gravity of the matters alleged (para (c)). It has been accepted that para (c) reflects the common law principles stated in Briginshaw: see Pedler v Richardson (unreported, Supreme Court, Young J, 16 October 1997); Wily v Fitz-Gibbon (unreported, Federal Court, Hill J, 2 March 1998); Amalgamated Television Services v Marsden [2002] NSWCA 419 at [60]. In Qantas Airways Ltd v Gama [2008] FCAFC 69 at [139], Branson J adopted the language of the High Court in Neat Holdings v Karajan Holdings, in finding that under s 140: "... the strength of the evidence necessary to establish a fact in issue on the balance of probabilities will vary according to the nature of what it is sought to prove." 175Recently, in FTZK v Minister for Immigration and Border Protection [2014] HCA 26, French CJ and Gageler J, at [12], in response to the appellant's reliance on Briginshaw v Briginshaw, observed that Briginshaw was concerned with the degree of satisfaction necessary to discharge the civil standard of proof in relation to an allegation of criminal conduct in civil proceedings. Their Honours added, "[t]he requisite degree of satisfaction is informed by the seriousness of the allegation". 176Whilst it is correct to say that a judge must have an "actual persuasion" of the existence of an event or occurrence (or, as in this case, the cause of the relevant event), the relevant standard required her Honour to be satisfied on the balance of probabilities, including by the drawing of such available inferences as she considered appropriate, whether the accident was caused by the negligence of the Council. Although it is often difficult to determine the cause of an accident where only one vehicle is involved, there are no witnesses and the driver is not able to give evidence, there was no feature of the accident that engaged that aspect of the Briginshaw principle enacted in s 140(2)(c). Nor did her Honour purport to proceed under that provision. It is necessary, therefore, to consider her Honour's reasoning process in concluding that factual causation had not been established. 177Her Honour's reasoning in this regard commenced with the assistance she obtained from the second proposition stated by McDougall J in Nguyen, where his Honour explained that a decision maker reached "that feeling of actual persuasion" by a determination that "the probabilities of the fact's existence are greater than the possibilities of its non-existence". 178This proposition appears to have its source in the observations of McHugh and Gaudron JJ in Malec v JC Hutton Pty Ltd [1990] HCA 20; 169 CLR 638. In that case, Deane, Gaudron and McHugh JJ said at 642-643: "A common law court determines on the balance of probabilities whether an event has occurred. If the probability of the event having occurred is greater than it not having occurred, the occurrence of the event is treated as certain; if the probability of it having occurred is less than it not having occurred, it is treated as not having occurred." 179McDougall J, whilst correctly restating the reference in Malec v JC Hutton to the probabilities of an event having happened, incorrectly stated the balance of the principle in referring to the possibilities of an event not occurring. The test in Malec v JC Hutton is a comparison of probabilities of an event occurring or not occurring. The test set out in the second proposition thus posed a wrong comparison for determining whether a fact has been established on the balance of probabilities. It, in effect, invited the fact-finder to weigh the possibilities of a fact not existing against the probabilities that a fact exists. 180That error is significant and contrary to a long line of authority explaining that, in regard to the matter in issue here, the court is concerned with probabilities, not possibilities. In Bradshaw v McEwans Pty Ltd [1951] HCA 480; 217 ALR 1, the plaintiff brought a claim under the Compensation to Relatives Act relating to a fatal accident involving a motor vehicle and the plaintiff's husband who was riding a bicycle. At trial, the defendant submitted that no case had been made out as there was no evidence as to how the accident occurred; the plaintiff's case relied on circumstantial evidence and upon certain admissions made by the defendant. The defendant submitted that he was entitled to a non-suit or a direction that the jury find in his favour. The defendant called no evidence, but suggested possible alternatives to the appellant's case as to how the accident occurred. This submission was rejected at trial and the jury awarded compensation to the complainant. Although the defendant's appeal to the Court of Appeal was allowed, the original verdict was restored by the High Court. After noting the various alternate hypotheses as to how the accident could have occurred, Dixon, Williams, Webb, Fullagar and Kitto JJ observed, at 5: "Of course as far as logical consistency goes many hypotheses may be put which the evidence does not exclude positively. But this is a civil and not a criminal case. We are concerned with probabilities, not with possibilities. The difference between the criminal standard of proof in its application to circumstantial evidence and the civil is that in the former the facts must be such as to exclude reasonable hypotheses consistent with innocence while the latter you need only circumstances raising a more probable inference in favour of what is alleged. In questions of this sort where direct proof is not available it is enough in the circumstances appearing in evidence give rise to a reasonable and definite inference: they must do more than give rise to conflicting inferences of equal degrees of probability so that the choice between them is mere matter of conjecture ... But if circumstances are proved in which it is reasonable to find a balance of probabilities in favour of the conclusion sought then though the conclusion may fall short of certainty it is not to be regarded as mere conjecture or surmise ..." (citations omitted) 181The Court then went on to hold, at 6: "All that is necessary is that according to the course of common experience the more probable inference from the circumstances that sufficiently appear by evidence or admission, left unexplained, should be that the injury arose from the defendant's negligence. By more probable is meant no more than that upon a balance of probabilities such an inference might reasonably be considered to have some greater degree of likelihood." See also Luxton v Vines [1952] HCA 19; 85 CLR 352 at 358; Jones v Dunkel [1959] HCA 8, 101 CLR 298 at 304-305. 182In Shoeys Pty Ltd v Allan (1991) Aust Torts Reports 81-104, Mahoney JA, in response to a submission in an occupier's liability slip and fall case that even if there had been a proper monitoring system, the plaintiff may still have been injured, Mahoney JA commented: "I do not think that such an argument should prevail in this case. The plaintiff's onus requires, not that she prove that the defendant's failure to monitor certainly caused her to fall; she is required to prove merely that it is probable that it did. The issue is therefore whether it is possible to infer that the absence of a system of monitoring probably caused the plaintiff to fall, that is, that a proper system of monitoring would probably have detected the vegetable matter on the floor before the plaintiff slipped on it." 183In Strong v Woolworths Ltd [2012] HCA 5; 246 CLR 182, the appellant slipped when the tip of one of the crutches she was required to use came into contact with a greasy chip on the floor of a shopping centre in an area under the control of the respondent. As the plurality made plain, at [84]: "Woolworths' submission that it was necessary for the appellant to point to some evidence permitting an inference to be drawn concerning when the chip was deposited must be rejected. It was incumbent on the appellant to prove that it was more probable than not that Woolworths' negligence was a necessary condition of her fall, but this onus could be discharged by consideration of the probabilities in circumstances in which the evidence did not establish when the chip was deposited." 184At [109], her Honour stated that the question whether the appellant had proved causation required: "... a finding that the probability that control was lost in the first section due to Ms Paterson encountering gravel under speed (and that she did not thereafter regain control sufficiently for her to avoid the accident), is greater than the possibility that she veered progressively onto the incorrect side of the road (as shown in photographs 19 and 20) for reasons unrelated to the condition of the road in combination with the speed at which she was travelling." 185For the reasons I have given, I do not consider that is a correct approach. However, it should be noted that her Honour also referred to Flounders v Millar [2007] NSWCA 238; 49 MVR 53 where, at [35], Ipp JA stated the test correctly, namely, that in a case based on circumstantial evidence, a plaintiff to succeed must "prove that the circumstances raise the more probable inference in favour of what is alleged". 186At [110], her Honour reasoned that: "... the competing hypotheses for Ms Paterson's loss of control, being a lack of concentration, inattention, drowsiness or fatigue, would logically need to subsist at the very point on the road where there was increased risk of a loss of control by reason of excess speed on gravel." 187Her Honour added, at [110], that it was necessary for her to be actually persuaded of one scenario over another as the more probable explanation for the loss of control and stated that she was not able to reach that level of satisfaction. That was a correct approach. However, read in the context of her reference at [109] to the "probability" of losing control on gravel and the "possibility" of having veered onto the incorrect side of the road for reasons unrelated to the condition of the road and the immediately preceding remarks at [110], it is possible that her Honour considered she had to be persuaded of a probability over a possibility. If that is correct, that reasoning contains a repetition of the error I consider is to be found in the second proposition in Nguyen. 188However, an error will not be sufficient to call for appellate interference in the orders made by the lower court, unless it can be established that the error had a material effect on the outcome. Thus, in order for the appeal to succeed, it will be necessary for the Court to be satisfied, on the balance of probabilities, that a cause of Ms Paterson's vehicle running off the road was the Council's negligence in failing to place speed reduction and slippery surface signs prior to the commencement of the roadworks. That requires a consideration of her Honour's reasoning in respect of that question. (b) Was the Council's negligence a cause of the accident? 189The question of the Council's liability for the accident was governed by the Civil Liability Act. Section 5D(1)(a) and (b), specify the requirements for "factual causation" and "scope of liability" respectively. It is factual causation that is relevantly at issue in this case. 190In Strong v Woolworths, the plurality, at 191, accepted, as had earlier been explained by McHugh J in March v Stramare (E & M H) Pty Ltd [1991] HCA 12; 171 CLR 506, that "the concept of a condition that is necessary to an occurrence" is an adaptation of John Stuart Mill's theory that the cause of an event is the sum of the conditions which were jointly sufficient to produce the event. As McHugh J explained, at 529-530, in its adapted form, the theory holds that: "... every necessary member of the set of conditions or relations which is sufficient to produce the relevant damage is a cause of that damage: International Encyclopedia of Comparative Law, vol XI, (1983), Ch 7, at p 27; Prosser, Law of Torts, 4th ed (1971), p 237; Fleming, Law of Torts, 7th ed. (1987), p 173 ..." His Honour had expressed this view in his earlier decision in this Court: Nader v Urban Transit Authority of NSW (1985) 2 NSWLR 501 at 531. 191Having referred to McHugh J's statement above, the plurality continued, at [20]: "A necessary condition is a condition that must be present for the occurrence of the harm. However, there may be more than one set of conditions necessary for the occurrence of particular harm and it follows that a defendant's negligent act or omission which is necessary to complete a set of conditions that are jointly sufficient to account for the occurrence of the harm will meet the test of factual causation within s 5D(1)(a). In such a case, the defendant's conduct may be described as contributing to the occurrence of the harm." (citation omitted) 192In Strong v Woolworths, the Court was concerned with an omission to regularly clean the floor of a particular area in a shopping centre. As noted above, the Court observed, at [34], it was not necessary for the appellant to point to some evidence permitting an inference to be drawn concerning when the chip upon which she slipped was deposited on the floor. Rather, s 5D(1)(a) required the appellant: "... to prove that it was more probable than not that Woolworths' negligence was a necessary condition of her fall, but this onus could be discharged by consideration of the probabilities in circumstances in which the evidence did not establish when the chip was deposited." 193In other words, a finding on causation could be made by the drawing of available inferences, provided the relevant standard of proof was satisfied. Inferences have always been an available form of fact finding on the question of causation. In Betts v Whittingslowe [1945] HCA 31; 71 CLR 637 Dixon J referred to Vyner and Waldenburg Bros Ltd [1946] KB 50 at 51 and then stated, at 649: "It is not necessary to inquire whether their Lordships [in Vyner] meant more than that the breach of duty coupled with an accident of the kind that might thereby be caused is enough to justify an inference, in the absence of any sufficient reason to the contrary, that in fact the accident did occur owing to the act or omission amounting to the breach of statutory duty. In the circumstances of this case that proposition is enough. For, in my opinion, the facts warrant no other inference inconsistent with liability on the part of the defendant." 194In Commissioner of Main Roads v Jones [2005] HCA 27; 79 ALJR 1104, Callinan J emphasised, at [80], that contained within Dixon J's statement in Betts v Whittingslowe was: "... the important qualification of an 'absence of any sufficient reason to the contrary'." 195Betts v Whittingslowe was commented upon more recently in Roads and Traffic Authority v Royal [2008] HCA 19; 82 ALJR 870, where Kiefel J, at [140], observed that Betts v Whittingslowe does not provide support for a conclusion of liability to be drawn from a failure to address or reduce a risk, an approach her Honour considered Santow JA in this Court had applied in that case: see Royal v Smurthwaite [2007] NSWCA 76; 47 MVR 401. As her Honour observed, at [139]: "Betts concerned a statutory duty, on the part of the employer, to securely fence and safeguard all dangerous parts of machinery. The observations of Dixon J were referrable to the circumstances of that case. The fact that a young worker's hand came into contact with part of the machinery which needed to be guarded permitted an inference of breach of that duty. Indeed, as his Honour went on to say immediately after the passage relied upon, 'the facts warrant no other inference inconsistent with liability on the part of the defendant'. His Honour's reasons do not suggest any presumption to operate or any alteration to the requirement of proof of causation. They have not been understood to suggest any lessening of it. As Dixon CJ later confirmed in his judgment in Jones v Dunkel, the facts proved must form a reasonable basis for a definite conclusion, affirmatively drawn." 196In Evans v Queanbeyan City Council [2011] NSWCA 230, Allsop P characterised Dixon J's approach in Betts v Whittingslowe as using the breach and the occurrence of the harm "as part of the fact finding process of drawing the relevant inference". 197In the Review of the Law of Negligence Final Report (the Ipp Report), the authors commented unfavourably on the approach taken in Betts v Whittingslowe and also in Bennett v Minister of Community Welfare [1992] HCA 27; 176 CLR 408, on the basis that the onus of proof was thereby cast on the defendant. I do not agree that that is the effect of the proposition stated in those cases. The proposition stated in those cases is about the drawing of available inferences, as Kiefel J made clear in Roads and Traffic Authority v Royal at [139]. The basal proposition remains that the court must be satisfied on the balance of probabilities that the defendant's breach caused the harm suffered. This is not to place any burden of proof on a defendant. Nor does it involve the proposition that liability may be inferred or proved merely because of a failure to address or reduce a risk. Both at common law and now pursuant to the Civil Liability Act, s 5E, the plaintiff bears the onus of proving any fact relevant to causation: see Flounders v Millar; Evans v Queanbeyan City Council. 198With the principles and considerations in mind to which I have referred and noting that the appeal to this Court is by way of rehearing: the Supreme Court Act 1970, s 75A; Fox v Percy [2003] HCA 22; 214 CLR 118 at [22], I consider that the evidence established the following. It was accepted that Ms Paterson entered the curve with her vehicle under control (see above at [73]). According to the two Council officers who attended the scene of the accident, there was a light tyre mark on the first section of new work and this mark showed that a vehicle had veered from its correct side of the road to the incorrect side of the road. Photographs taken at that time also indicated a mark on the road. Constable Ryan's observations of a mark on the first section of new work was consistent with Mr Ellis' statement and Mr Crisp's sketch. 199Her Honour accepted that the mark on the first section of new work had been made by Ms Paterson's vehicle and showed the vehicle veering from the correct to incorrect side of the roadway. This was a critical finding. It meant that something occurred on the first resealed section of the roadway. Her Honour did not accept, however, that the mark on the first resealed section was due to loss of control. In not being satisfied that there was that evidence, she focussed on the fact that Mr Johnston had sought to explain away why the police officers had not seen evidence indicative of a loss of control. 200However, there was much more in the evidence than that. The starting point for an analysis of that evidence is her Honour's acceptance that the mark was made by Ms Paterson's vehicle. The evidence then establishes that the vehicle was on the incorrect side of the road until the second old section of the road. It was on that section that there was a loss of control that resulted in the vehicle rotating and leaving the road. 201The differing views of the experts revolved essentially around the question whether there was evidence of a loss of control on the first resealed section and what was a safe speed for travelling around the curve with its resealed surface commencing part way through the curve. Mr Coffey and Mr Johnston considered a safe speed was 60 km/h. Mr Stuart-Smith considered it was 145-160 km/h. 202On the basis that the mark on the first section of new roadwork was made by Ms Paterson's vehicle, Mr Johnston considered that photograph 19 showed evidence of a side slip, precipitated by some braking. The dynamics at that point were not sufficient to indicate that there was a significant loss of control. 203Mr Stuart-Smith did not explain what action of the vehicle caused the mark because he did not accept that there was evidence of loss of control, if indeed there was, in fact, any evidence of a mark on the road at that point. However, by the conclusion of the evidence, Mr Stuart-Smith had agreed that there were risks associated with the loose aggregate on the roadway additional to the risk of windscreen damage, including the possibility of reduced friction or loss of traction if a vehicle was travelling fast. Mr Stuart-Smith accepted that it "would have been better" had a reduced speed sign been in place because a driver could lose control on loose gravel. The reference to "travelling fast" in Mr Stuart-Smith's evidence was a reference to a speed at or just under 100 km/h. 204Mr Stuart-Smith also accepted that a reason for reduced speed on roads subject to resealing, which he agreed was primarily for the protection and safety of drivers, was the danger to road users in not being aware of the change in functional control. In this regard, he conceded that reduced speed signs were recommended by national road authorities so as to minimise vehicles losing control when passing onto newly resealed sections of roadway. 205Mr Stuart-Smith also agreed that if the speed limit for the section of the roadway where the roadworks were being undertaken was reduced to 60 km/h, the risk to Ms Paterson would have been significantly reduced compared to the risk she would have faced if travelling at 100 km/h. He accepted that at 60 km/h, Ms Paterson would have been able much better to respond to any loss of control. He likewise agreed that the discontinuity between sections of new and old work would be a factor increasing the danger to Ms Paterson, travelling at about 100 km/h, if she were trying to correct her line of travel. 206These various concessions by Mr Stuart-Smith brought his evidence into substantial concurrence with the evidence of Mr Johnston and Mr Coffey. In particular, he accepted that there was a risk of loss of traction and a danger that a driver, on entering upon the newly surfaced section of the roadway, may not be aware of the change in functional control of the vehicle. This had been a criticism that Mr Johnston made of Mr Stuart-Smith's view that a safe speed for travelling around the curve was 145-160 km/h. Mr Stuart-Smith's view had been based on a static calculation without any allowance for human reaction. As I understand it, Mr Stuart-Smith's concession that a driver would not be aware of the change in functional control was consistent with Mr Johnston's evidence that a driver, in this case Ms Paterson, entering upon the new work at speed, would find herself in an uncomfortable situation and was likely to instinctively react by braking. Braking, of course, was a dangerous reaction when there was gravel on the road, which, as Mr Coffey described it in this case, was like "walking on marbles". 207In my opinion, the evidence was overwhelming that 100 km/h was not a safe speed for the negotiation of the curve with the new work. Her Honour's finding of breach accepted that to be the case. 208There were then two competing hypotheses as to the cause of the accident. Mr Stuart-Smith postulated that the likely cause of the accident, on the balance of probabilities, was that Ms Paterson deviated from the correct side of the road for an unknown reason, that is, a reason other than a loss of traction. Although he postulated that the reason for the deviation was unknown, he advanced as possible reasons that the deviation was due to driver fatigue, inadvertence or illness. Again, his view in this regard was based upon his opinion that the curve could be safely negotiated at speed. 209There was no evidence of illness and it appears that little attention was paid to it after initial mention of it in Mr Stuart-Smith's report. The accident occurred prior to midday. Ms Paterson lived in the area and her occupation was home duties. It was not suggested that Ms Paterson had driven for a long distance or that she was travelling home after working a shift that had started earlier that morning. Accordingly, there was no evidence to indicate the likelihood of fatigue. Nonetheless, there can be other reasons for fatigue, so that fatigue cannot be eliminated. Nor can inadvertence be eliminated. 210However, the "unknown" reasons advanced by Mr Stuart-Smith for Ms Paterson having veered off course on the first resealed section remained possibilities. There was no evidence to support these factors as possible, let alone probable causes for the accident. In addition, Mr Johnston explained why the factors suggested by Mr Stuart-Smith were unlikely, having regard to the direction of travel of the vehicle on the first resealed section. Had any of those factors been present, the vehicle's path of travel would have been straight ahead, rather than a veering to the right. Mr Stuart-Smith did not respond to that analysis. 211There is another important consideration on this issue. Both experts agreed that a driver would have been able better to respond to a loss of control, for whatever reason, if travelling at 60 km/h. The evidence was that Ms Paterson was an experienced and competent driver, and it was not suggested that had the speed limit been 60 km/h she would not have observed it. Mr Johnston considered the accident was unlikely to have occurred at all if Ms Paterson had been travelling at 60 km/h. Mr Stuart-Smith accepted that the risk of losing control at that speed was significantly reduced. Her Honour did not consider this aspect of the evidence. Critically, that was the very reason why there should have been a reduced speed sign and why it was negligent for the Council not to have had one in place. 212Further, contrary to her Honour's finding, there was evidence that Ms Paterson lost control on the first section of the new work. Once it was accepted, as her Honour did, that the tyre mark on the first section of the roadway was that of Ms Paterson's car, it was clearly a mark showing a loss of control at that point in the sense that the vehicle was no longer travelling along the path it should have been travelling. A loss of control does not have to involve a dramatic loss of direction or to be such as to cause the vehicle to rotate. This was explained by Mr Johnston in the evidence set out at [143] above. 213The evidence was clear that a loss of friction was likely to occur on a curve on a surface with loose stones if entered with speed. All of those conditions were satisfied in this case. The evidence was also clear that even had there been factors involved such as fatigue or inattention, those factors would have been accommodated had the new work been entered at a significantly lower speed, such as 60 km/h, as her Honour was required. The evidence also established that the likelihood of a person over-braking in reaction to the change in friction would have been significantly lower had the curve been entered at a lower speed. 214There are two final factors that should be mentioned relating to her Honour's reasons for not accepting Mr Johnston's evidence. At [108], her Honour stated that she was not prepared to draw an inference that the tyre marks were present but missed at the scene by the police officers. She made particular reference to Sergeant Brand, who was specifically looking for some evidence of loss of control. However, Sergeant Brand only looked for tyre marks for a distance of 100 m from the point where Ms Paterson's vehicle went off the road. The marks observed by a Mr Ellis and Mr Crisp were at least 150 m from that point. Further, Constable Ryan saw a mark that was consistent with the mark seen by Mr Ellis and Mr Crisp. 215The second matter to which mention should be made is the discounting of Mr Johnston's reliance upon the accident involving Ms Skorulis as supporting his opinion that it was probable that Ms Paterson had lost control on the loose gravel. As is explained above, there was an attempt made to compare the two accidents by reference to the similarities and dissimilarities between them. Mr Stuart-Smith considered that the only similarities were limited to the fact that both vehicles had entered a resealed section of the road and had left the road. Ms Skorulis' accident may provide some support for the proposition that the curve could not be driven safely at 100 km/h. But even if that is so, it does not matter in this case, when the evidence established that a 60 km/h sign should have been erected. Perhaps more relevant, Mr Johnston's evidence was not centrally reliant upon it and his reference to Ms Skorulis' accident did not otherwise undermine his expert opinion. 216In my opinion, on the balance of probabilities, the only reasonable inference to draw from the whole of the evidence was that when Ms Paterson entered into the curve and onto the new road work, there some loss of friction at the speed she was travelling. This was likely to have caused her to react and eventually lose control of the vehicle. Had she been travelling at the lower speed of 60 km/h, the evidence supported the inference that she was most likely to have been able to take the curve without mishap. As the Council was negligent in failing to erect a reduced speed sign specifying a speed limit of 60 km/h, I am of the opinion that factual causation has been established. Her Honour's failure to be satisfied is an indication that she erred in failing to determine the matter on the probabilities of events occurring or not occurring in accordance with the appropriate standard of proof. But even if that was not so, I have reached the conclusion that causation has been established. Reasons of the trial judge: s 43A 217Her Honour concluded, at [48], that a speed reduction sign, in combination with the pictorial "Slippery Road" sign, should have been put in place in advance of the commencement of the roadworks and at appropriate intervals thereafter to alert road users to the potentially hazardous condition of the road. Her Honour concluded that the respondent was negligent in failing to do so. 218However, at [70], her Honour concluded that the test stated in the Civil Liability Act, s 43A had not been met. Her Honour's reasons for this were as follows: "I accept that although other Council officers in [the Works Manager's] position might legitimately have made no provision for the placement of a 'Slippery Road' sign or a sign warning or directing of the need for a reduction in speed from 100 km/h, the fact that minds might differ as to the need for that additional signage and that the Manual allows the exercise of a considered judgment as to the number and type of road condition signs where there is gravel on the road as an incident of road sealing works does not, in my assessment, meet the necessary test for Wednesbury unreasonableness within the meaning of s 43A." Consideration: s 43A 219Section 43A provides as follows: "43A Proceedings against public or other authorities for the exercise of special statutory powers (1) This section applies to proceedings for civil liability to which this Part applies to the extent that the liability is based on a public or other authority's exercise of, or failure to exercise, a special statutory power conferred on the authority. (2) A special statutory power is a power: (a) that is conferred by or under a statute, and (b) that is of a kind that persons generally are not authorised to exercise without specific statutory authority. (3) For the purposes of any such proceedings, any act or omission involving an exercise of, or failure to exercise, a special statutory power does not give rise to civil liability unless the act or omission was in the circumstances so unreasonable that no authority having the special statutory power in question could properly consider the act or omission to be a reasonable exercise of, or failure to exercise, its power. (4) In the case of a special statutory power of a public or other authority to prohibit or regulate an activity, this section applies in addition to section 44." (emphasis added) 220In my opinion, the respondent was exercising a "special statutory power", within the meaning of s 43A: see Basten JA at [245] ff. 221Basten JA has approached the meaning of the standard of care prescribed by s 43A by reference to English and Australian authorities in the administrative law sphere and, in particular, by reference to Minister for Immigration and Multicultural Affairs v Eshetu [1999] HCA 21; 197 CLR 611. The trial judge also referred to the authorities in this Court which had dealt with this aspect of s 43A. Her Honour summarised the explanation of the test in s 43A by reference to the authorities reviewed by Giles JA in Allianz Australia Insurance Ltd v Roads and Traffic Authority of New South Wales [2010] NSWCA 328. 222In my opinion, the statement of Lord Diplock in Secretary of State for Education and Science v Tameside Municipal Borough Council (1977) AC 1014, at 1064, is consonant with the language of s 43A and appropriately characterises the nature of a special statutory power: "... in public law 'unreasonable' as descriptive of the way in which a public authority has purported to exercise a discretion vested in it by statute has become a term of legal art. To fall within this expression it must be conduct which no sensible authority acting with due appreciation of its responsibilities would have decided to adopt." 223As Lord Diplock recognised, at 1064: "The very concept of administrative discretion involves a right to choose between more than one possible course of action upon which there is room for reasonable people to hold differing opinions as to which is to be preferred." 224I agree with Basten JA, at [277]-[279], that for the purposes of s 43A, the relevant test is not whether the court considers that the decision had the requisite character of unreasonableness specified in the section, but whether "no authority" could "properly consider the act or omission" to be a reasonable exercise of, or failure to exercise, its power. In this regard, I consider that her Honour applied the correct test. 225On the evidence, however, I do not accept that it could be said that the failure to install the "Slippery Road" signs and either reduction of speed signs or signs imposing a lower speed limit, were matters about which reasonable minds could differ. Mr Coffey was not only the Director of Technical Services and the person to whom the RTA certified Works Manager reported to, he was also the person designated by the respondent after the accident to investigate the suitability of the use of the Traffic Control Plan. 226His evidence was that the "Slippery Road" sign was the most important sign to use. It was essential to warn of the risk of a loss of traction, irrespective of a driver being alerted to the presence of roadworks. He said it made "no sense" to him that it was not used. This was because driving on the newly sealed portions of the roadway was like "walking on marbles". His evidence was totally supported by the appellant's expert, Mr Johnston and was ultimately supported by the respondent's expert, Mr Stewart-Smith. 227To the extent that there may have been any reticence in Mr Stewart-Smith's agreement, it is to be noted that her Honour had cause to remark, at [47], upon his "seemingly limited appreciation of the necessity to comply strictly with the expert witness' obligation of impartiality": see the Uniform Civil Procedure Rules 2005 (UCPR), Pt 31, r 23. Nonetheless, her Honour assessed Mr Stewart-Smith's evidence as accepting the necessity for the respondent to provide both the pictorial and speed reduction signage that Mr Coffey stated was required. 228In my opinion, the omission to include that signage was, to quote Lord Diplock, conduct which no sensible authority acting with due appreciation of its responsibilities would have decided to adopt. Conclusion 229It follows, in my opinion, that the appeal should be allowed. I would therefore propose the following orders: In each of the matters CA 2012/241825 and CA 2012/241832: 1. Appeal allowed. 2. Set aside the verdict for the respondent made by Fullerton J on 9 July 2012. 3. Remit the matter to the Common Law Division of the Supreme Court for the assessment of damages. 4. Order the respondent pay the appellant's costs of the liability hearing at first instance and of the appeal. 230BASTEN JA: Kingsvale Road is a country road in central New South Wales, running south from Young to the township of Harden. In 2004 it was sealed and carried some 300 cars per day. In August 2004 Harden Shire Council ("the Council") was undertaking resurfacing of sections of the road. At about 11.28am on 20 August 2004 Ms Debbie Ann Paterson was travelling south along Kingsvale Road when she lost control of her car on a section of the road which was subject to roadworks, left the road and crashed into a tree. She died of her injuries. 231The appellant, her de facto partner, brought proceedings for damages, both for psychiatric injury resulting from nervous shock and, on behalf of himself and their three children, for damages under the Compensation to Relatives Act 1897 (NSW). 232The proceedings were heard together by Fullerton J in the Common Law Division. By judgment delivered on 9 July 2012, both claims were dismissed: Curtis v Harden Shire Council [2012] NSWSC 757. Appeals were brought in both proceedings by Mr Curtis. For the reasons given below, the appeals in each matter, which turned on establishing negligence on the part of the Council, should be dismissed. Engagement of s 43A (a) background and issues 233The primary basis upon which Fullerton J dismissed the proceedings was that the plaintiff did not establish a breach of duty to the standard required by s 43A of the Civil Liability Act 2002 (NSW) (see at [3] and [219] above). That section is engaged when liability is asserted on the part of a public authority exercising (or failing to exercise) a special statutory power. (The trial judge also found that the plaintiff had not established that the conduct complained of caused the accident.) 234Where s 43A applies, an act or omission involving the exercise of a special statutory power "does not give rise to civil liability" unless it was so unreasonable an exercise of the relevant power that no authority could properly consider it to be a reasonable exercise of the power: s 43A(3). The section thus assumes the existence of a duty of care and identifies the standard to be applied in determining whether there has been a breach. It was common ground that, in carrying out road maintenance works, the Council owed a duty of care to road users, including the deceased. That duty included placing appropriate warning signs near the works. It was necessary to consider the application of s 43A, prior to determining breach, because, if applicable, the section prescribed the relevant standard of care. 235The question whether the section applies must be approached in two stages. The first is to identify whether and to what extent the liability upon which the defendant is sued "is based on" the exercise of a statutory power conferred on the defendant: s 43A(1). The second stage requires reference to the power so identified to determine whether it is a "special statutory power" as defined in sub-s (2). To satisfy the definition, it must be a power "conferred by or under a statute" and "of a kind that persons generally are not authorised to exercise without specific statutory authority": s 43A(2)(a) and (b) respectively. 236The maintenance work being undertaken by the Council widened the sealed surface, removed the uneven border between the tarmac and the dirt shoulder of the road and no doubt removed potholes or other defects in the sealed area. In the short term, it created two potential hazards for motorists. First, there was no marked centreline; secondly, there was loose gravel on the resealed surface. 237The appellants did not contend that there was any negligence involved in the actual roadworks: rather, they submitted that the signage was inadequate, due to the preparation of an inadequate "Traffic Control Plan", referred to as a "TCP". 238No work was being undertaken at the time when the deceased was travelling on the road. There were, however, signs at each end of the section which had been the subject of roadwork on the previous day, one stating "Roadwork Ahead", a second stating "New Work No Lines Marked" and a third, described as a "Chip Hazard" sign, showing stones being thrown up by the wheels of one car and damaging the windscreen of another car. 239Although the pleadings raised a number of particulars of negligence, the trial judge held that "at the hearing it was the failure to provide adequate signage during the inspection stage that was relied upon as constituting the breach of duty": at [7]. (That assessment was challenged on the appeal, in a way which will be considered shortly.) In respect of those matters, the trial judge expressed her satisfaction that "a speed reduction sign in combination with ... a pictorial 'Slippery Road' sign should have been put in place in advance of the commencement of the roadworks (and at appropriate intervals thereafter) to alert road users to the potentially hazardous condition of the road": at [48]. The Council had not taken those further steps. The speed limit on that section of the road was 100km/h: the judgment envisaged a reduction to 80km/h. 240The appellants alleged that, in addition to the absence of the identified signage, there was negligence in failing to design and implement an appropriate TCP. In a practical sense, this complaint appeared to have little relevance. Assuming that the inadequate signage was the cause of the accident, and that the inadequacy of the signage involved a breach of duty, it did not matter whether there was a TCP prepared by the Council which directed the workers as to the signage to be used, or no plan at all. If there were a TCP it also did not matter, in practical terms, whether the plan was adequate or inadequate. The purpose of the complaint lay elsewhere: the appellants submitted that even if the placing of the signage was the exercise of a special statutory power, the design and implementation of a TCP was not and, if negligence which indirectly caused the accident could be traced to the inadequate TCP, s 43A would not be engaged. That submission, it was contended, was not dealt with by the trial judge, although it had been articulated by the plaintiff at the trial. (b) meaning of "based on" 241The appellants' first submission was that s 43A was not engaged because the claim as to liability was not "based on" any exercise of a special statutory power by the Council. Rather, it was based on a breach of a common law duty to warn of dangers created by the carrying out of the roadworks. In support of that approach, the appellants referred to statements by Campbell JA (with the agreement of McColl JA and, "in general", Sackville AJA) in Roads and Traffic Authority of New South Wales v Refrigerated Roadways Pty Ltd [2009] NSWCA 263; 77 NSWLR 360. The question in that case was whether the Roads and Traffic Authority ("the RTA") was negligent in failing to erect screens along the sides of a bridge across the F5 freeway (near Menangle) which might have prevented a person dropping a large piece of concrete from the bridge, which struck the windscreen of a truck passing underneath, causing the death of the driver. The Court held that the RTA could not rely upon s 43A on the appeal, when it had not been raised at trial: at [361]. Views as to how it might have operated if properly invoked were therefore obiter. The passage relied upon by the appellants was in the following terms, at [375]: "Even if it had been the case that construction of the screens had required temporary closure of the bridge, it would not have been negligence in exercising any power to close the bridge, or even delay in exercising a power to close the bridge, that is the basis of the plaintiff's claim. Thus, any civil liability of the RTA in the present case would not be 'based on ... failure to exercise a special statutory power' and so the condition for the operation of section 43A would not arise." 242This conclusion followed from the following finding, at [368]: "Further,... the ownership of the bridge is quite sufficient to empower the RTA to erect screens on it. Its ownership of the bridge is not a 'power ... conferred by or under a statute'. Rather, it is a property right." 243If the omission pleaded, namely the failure of the RTA to erect screens along the bridge over the freeway, did not constitute a failure to exercise a special statutory power, it was correct to say that the claim was not "based on" the failure to exercise a special statutory power. Whether a similar conclusion would arise in the present case depends upon an analysis of the acts and omissions upon which the appellants relied at trial. 244The expression "based on" is not to be extracted from its statutory context and treated as a reference to the terminology used in the pleadings, as the appellants' submissions tended to suggest. What must be pleaded are the acts or omissions which give rise to a cause of action. Adopting the language of sub-s (3), the question for the Court will be whether "any act or omission ... [gives] rise to civil liability". Once the act or omission has been identified, it will then be necessary to inquire whether it is one "involving an exercise of, or failure to exercise, a special statutory power". A statement of claim need not identify the statutory source of any power which might be involved. While s 43A does not merely identify a defence, there is no doubt that a defendant must plead, not the terms of the provision as such, but the facts giving rise to its engagement. The plaintiff will then have to establish negligence to the statutory standard to succeed. (The pleadings on both sides left much to be desired in this case.) (c) special statutory powers 245The definition of the phrase "special statutory power" in sub-s (2) has already been noted. It is neither helpful nor desirable to try to paraphrase or explain that language: the proper course is to seek to apply it to the acts or omissions relied on at trial. Those were the omission to install certain traffic control signs. (This requires reference to the relevant statutory provisions as they applied at the date of the accident; most have since been replaced: see Road Transport Legislation (Repeal and Amendment) Act 2013 (NSW), effective from 1 July 2013.) The missing signs fell within the definition of "prescribed traffic control device" as defined in the Road Transport (Safety and Traffic Management) Act 1999 (NSW) ("the Traffic Management Act"), s 50 (since repealed and now in the Road Transport Act 2013 (NSW), s 121). That provision came in Div 1 of Pt 4, and provided the following definition: 50 Interpretation In this Division: ... prescribed traffic control device means a sign, signal, marking, structure or other device to direct or warn traffic on a road ... that is prescribed by the regulations for the purposes of this definition. 246The Road Transport (Safety and Traffic Management) (Road Rules) Regulation 1999 (NSW) prescribed, at the relevant time and for the purposes of s 50 of the Traffic Management Act, "any traffic control device of a kind mentioned in the Australian Road Rules that has effect for the Rules under rule 315 of the Rules": cl 131. (On 13 June 2008 the Regulation was renamed the Road Transport (Safety and Traffic Management) Regulation 1999 and was repealed on 1 July 2013.) It was not in dispute that the "Slippery Road" sign and a restricted speed sign fell within this definition. 247The Traffic Management Act prohibited anyone without appropriate authority installing a traffic control device. Thus s 52 (now the Road Transport Act, s 123) stated: 52 Unauthorised prescribed traffic control devices (1) A person must not, without appropriate authority: (a install or display a prescribed traffic control device on, above or near a road or road related area, or (b) interfere with, alter or remove any prescribed traffic control device installed or displayed on, above or near a road or road related area. (2) A person must not install or display on, above or near a road or road related area any sign, signal, marking, structure or other device that might reasonably be mistaken to be a prescribed traffic control device. 248This section is clear beyond doubt and reflects the commonsense proposition that only someone with authority could place a regulatory sign beside a road limiting the speed at which vehicles might travel, or otherwise warning of hazards. Questions of judgment are involved: unnecessary signs can cause confusion. Accordingly, the placement of such signs involves a power of a kind that persons generally are not authorised to exercise without specific statutory authority. 249Appropriate authority is identified by s 51 of the Traffic Management Act (now the Road Transport Act, s 122) in the following terms: 51 Appropriate authority for the purposes of this Division For the purposes of this Division, a person has appropriate authority to install or display (or to interfere with, alter or remove) a prescribed traffic control device if: (a) the person is a public authority that has been directed by the Authority under Division 1C of Part 6 of the Transport Administration Act 1988 to install or display (or to interfere with, alter or remove) the device, or (b) the person is otherwise authorised in writing by the Authority to install or display (or to interfere with, alter or remove) the device. Note. Division 1C of Part 6 of the Transport Administration Act 1988 enables the Authority to give certain public authorities directions in respect of safety and traffic management. 250Part 6 of the Transport Administration Act 1988 (NSW) deals with what was then known as the RTA: it was, relevantly, an "authority" for the purposes of that Act. The term "public authority" appears not to be defined in the Traffic Management Act, but the intention of s 51(a) was to adopt that concept, as used in Pt 6 of the Transport Administration Act, where it was defined to mean a "public or local authority constituted by or under an Act": s 45E(1). That definition would include the Council. Further, the RTA was authorised by the Transport Administration Act to direct public authorities to implement plans or proposals formulated or adopted, general standards or general principles established, or other decisions made, by the Authority in exercise of its functions under Pt 6. Those functions are stated in broad terms and include promoting traffic safety measures or activities, including those for the prevention of accidents on roads: see generally, s 52A. 251It appears to have been in exercise of these functions that the RTA promulgated a manual entitled "Traffic Control at Work Sites". The manual noted in the foreword that it was released "following the publication of the 2002 edition of the Australian Standard 1742.3, Manual of uniform traffic control devices, Part 3, Traffic control devices for works on roads and to bring it up to date with developments made in traffic control practice since the release of the second version in 1998". The foreword continued: "Whilst this manual must be used on all RTA road and bridge sites for work being undertaken by the RTA's own workforce, RTA contractors or local government and public utility bodies undertaking work on behalf of the RTA, its use is commended to all practitioners who are responsible for the control of traffic at work sites, on non-RTA sites." 252Appendix D to the manual included a set of standard traffic control plans, indicating the signage, with instructions as to when and how they were to be used. Each sign was identified by a number, which was a cross-reference to Australian Standard 1742 (now 1743). 253The evidence also indicated that the Council officer responsible for designing and implementing the traffic control plan with respect to the work undertaken on Kingsvale Road, was trained and certified by the RTA. The basis on which such training and certification was provided was not identified: however, it is sufficient for present purposes to note that the power to place prescribed traffic control devices beside the road was one which could not be undertaken by persons generally, but could be undertaken by a public authority, including the Council, in accordance with the statutory regime identified above. It is therefore clear that the Council was exercising a "special statutory power" in placing the signage used and in omitting other signage. It was not a case where the Council "failed to exercise" the relevant power: to the extent that it did not adopt a particular sign, it made a decision in exercise of a discretionary power, after considering the available options. 254The appellants drew attention to the comments of Campbell JA in Refrigerated Roadways as to the manner in which the courts have read down protective provisions granting statutory authorities immunity from suits in negligence when carrying out their functions in good faith. The immunity is generally restricted to activities which are reliant for their lawfulness on a statutory power and does not extend to activities which can be undertaken in accordance with the general law. The appellants submitted that conduct which involves the exercise of a liberty available to any person under the general law, such as driving on a road or the carrying out of a contractual agreement with a particular party, may not be covered by s 43A: Refrigerated Roadways at [376]; and see Board of Fire Commissioners (NSW) v Ardouin [1961] HCA 71; 109 CLR 105; Puntoriero v Water Administration Ministerial Corporation [1999] HCA 45; 199 CLR 575 at [14]-[15] (Gleeson CJ and Gummow J), [35] (McHugh J), [114] (Callinan J), see also [61]-[63] (Kirby J, in dissent); M Aronson, "Government Liability in Negligence" (2008) 32 Melb U L Rev 44 at 79. It was submitted that the Council was exercising powers as a landowner. Attention was also drawn to the possibility that the use in s 43A(2)(b) of the term "specific statutory authority" might invite a comparison with powers exercised under some more general statutory authority. However, these factors have no direct application in the present case: the prohibition on any person installing prescribed traffic control devices combined with the requirement for statutory authority to undertake such an activity, placed the grant of authority for such activities squarely within the concept of a "specific statutory authority", as used in s 43A(2)(b). 255Finally, the appellants suggested a different conclusion might be reached on the basis of statements in the judgment of Beazley JA (with whom Whealy JA and Sackville AJA agreed) in Bellingen Shire Council v Colavon Pty Ltd [2012] NSWCA 34; 188 LGERA 169 at [34]-[38]. The complaint in that case involved a failure on the part of the Bellingen Shire Council to "install guide posts along the road so as to delineate the edge of the formed surface of the roadway from the soft edge", in circumstances where a tanker had strayed onto the side of the road and rolled down the embankment. Again, the question involved not compliance with s 43A, but whether s 43A had been pleaded and whether it could be relied upon by the appellant, if not pleaded. The Court held that the appellant should not have been allowed to raise s 43A as a defence after the evidence had concluded: at [33]. Beazley JA identified an issue, involving both fact and law, as to whether the authority to install guideposts involved the exercise of a "special statutory power" at [38]. The statutory power relied upon was s 87 of the Roads Act 1993 (NSW), pursuant to which an appropriate roads authority was empowered to carry out "traffic control work" on any unclassified road: s 87(3). 256In the present case, it was common ground that Kingsvale Road was not an unclassified road and, accordingly, the nature of that provision was not in issue. This case must be resolved by reference to its own facts and the relevant statutory provisions and not by reference to dicta in other cases involving different facts and different statutory provisions. For the reasons given above, s 43A was engaged: neither Refrigerated Roadways nor Bellingen Shire Council supported a different approach or conclusion. The statutory standard of care (a) source of the language 257The concept of an exercise of power being so unreasonable that no reasonable person could so exercise it derived from the judgment of Lord Greene in Associated Provincial Picture Houses v Wednesbury Corporation [1948] 1 KB 223, where it was used to was used to identify a misuse of a discretionary power so serious as to be set aside. Its application as a basis of liability in tort suffers from the many difficulties which attend its application in administrative law, where it does not give rise to any right to claim damages. In the sphere of administrative law it has been characterised and reformulated in a variety of different ways, which do not necessarily assist in understanding s 43A: see discussion in Allianz Australia Insurance Ltd v Roads and Traffic Authority of New South Wales [2010] NSWCA 328; 57 MVR 80 at [81]-[89] (Giles JA, McColl JA and Sackville AJA agreeing). 258The classic language of administrative law is to ask whether the decision was "so unreasonable that no reasonable [t]ribunal, acting within jurisdiction and according to law, would have come to such a conclusion": Minister for Immigration and Multicultural Affairs v Eshetu [1999] HCA 21; 197 CLR 611 at [39] (Gleeson CJ and McHugh J) and [159] (Hayne J). On this approach, the task is to characterise the decision in a particular way, rather than to analyse the process of reasoning. Eshetu involved a challenge by way of judicial review to a decision of the Refugee Review Tribunal: in contrast to many forms of administrative decision-making, the Court was assisted by extensive reasons given by the Tribunal. Further, the language of unreasonableness must be understood in its legal context. Thus, Gleeson CJ and McHugh J stated at [40]: "Someone who disagrees strongly with someone else's process of reasoning on an issue of fact may express such disagreement by describing the reasoning as 'illogical' or 'unreasonable', or even 'so unreasonable that no reasonable person could adopt it'. If these are merely emphatic ways of saying that the reasoning is wrong, then they may have no particular legal consequence." 259Gummow J in Eshetu, referring to the earlier judgment of Mason J in Minister for Aboriginal Affairs v Peko-Wallsend Ltd [1986] HCA 40; 162 CLR 24 at 39-46, noted that the so-called Wednesbury unreasonableness principle was concerned with identifying the limited role of a court in reviewing the exercise of an administrative discretion: at [124]. The principles, Gummow J explained, had been developed by analogy from principles governing the judicial control in private law of the exercise of powers and discretions vested in trustees. The idea of a "discretion" in this context involves a choice to be made by the decision-maker between two or more outcomes. As explained by Lord Diplock in Secretary of State for Education and Science v Tameside Metropolitan Borough Council [1977] AC 1014 at 1064: "The very concept of administrative discretion involves a right to choose between more than one possible course of action upon which there is room for reasonable people to hold differing opinions as to which is to be preferred." 260Thus, if the law recognises no room for reasonable people to differ in a particular case, there is, in effect, no discretion; the decision-maker is under a legal duty to act in a particular way. When that point is reached, but only then, the court may intervene where a different and unavailable choice is made. 261Bearing these principles in mind, it is convenient to turn to the terms of s 43A, which do not simply reflect the administrative law principles. (b) a linguistic analysis 262Because s 43A was engaged, liability could only be established if the appellants proved on the balance of probabilities that the omission by the Council was in the circumstances so unreasonable that no authority having the special statutory power in question could properly consider that omission to be a "reasonable exercise of, or failure to exercise, its power": s 43A(3). (The grammatical construction is awkward with respect to a failure to exercise a power, but that issue does not arise in this case.) Whilst the terms of s 43A are said to find inspiration in the administrative law principle noted above, it is necessary, in construing the section, to bear in mind its quite different purpose. Thus, whereas Wednesbury unreasonableness constitutes a basis for the intervention of a court in judicial review proceedings to declare an exercise of power invalid, tort law assumes that the power had in fact been exercised and seeks to impose liability to pay compensation. In tort, damages are available if a public authority negligently causes harm to an individual, regardless of whether its action is valid or invalid. (Whether damages should be available for an improper or invalid exercise of power by a public authority is a different question.) 263Historically, public authorities enjoyed various forms of immunity from liability in tort. The most dramatic change in derogating from the immunity came (in New South Wales) with the Claims against the Colonial Government Act 1876 (39 Vic No 38), replaced by the Claims against the Government and Crown Suits Act 1897 and, in due course, the Claims against the Government and Crown Suits Act 1912 (NSW). Its most recent iteration is the Crown Proceedings Act 1988 (NSW). In the Commonwealth sphere, the derogation from the immunity is found in the Judiciary Act 1903 (Cth), ss 56 and 64. Those statutes removed the general immunity of the executive government from action in tort by its citizens. Nevertheless the immunity remained for acts undertaken in the exercise of independent (discretionary) functions, for example by police constables: see Kable v State of New South Wales [2012] NSWCA 243; 293 ALR 719 at [169]-[170]; Sneddon v State of New South Wales [2012] NSWCA 351 at [57]-[64]. That immunity was removed by the Law Reform (Vicarious Liability) Act 1983 (NSW). 264The inroads on the immunity of the government and its public officers have not removed a concern that public authorities acting in the public interest should not be treated identically with ordinary citizens. For example, the courts were long reluctant to award costs against the police in the case of failed prosecutions for minor criminal matters: cf Latoudis v Casey [1990] HCA 59; 170 CLR 534. In serious criminal matters, the principle that the State neither pays nor receives costs remains; in minor matters it is regulated by statute: Criminal Procedure Act 1986 (NSW), s 214. Further, road authorities which were liable for misfeasance in exercise of their statutory powers were long held not to be liable for mere nonfeasance: cf Brodie v Singleton Shire Council [2001] HCA 29; 206 CLR 512 at [102]-[106] (Gaudron, McHugh and Gummow JJ). However, many years before that decision, it had been accepted that public authorities may be liable in tort for negligence in carrying out statutory functions: Sutherland Shire Council v Heyman [1985] HCA 41; 157 CLR 424 and Crimmins v Stevedoring Industry Finance Committee [1999] HCA 59; 200 CLR 1. 265In Stovin v Wise [1996] AC 923 the House of Lords considered the liability of a road authority for an accident which resulted from reduced vision at an intersection, where a mound of earth on adjoining private land obstructed the view of road users. The road authority was aware of the problem and had power to require the landowner to remove the mound, but had not done so. Lord Hoffmann, speaking for the majority, made two points which are relevant in the present case. He noted, first, that the case involved an omission, rather than positive conduct and, secondly, that the common law did not treat omissions in the same way as positive acts, referring to Windeyer J in Hargrave v Goldman [1963] HCA 56; 110 CLR 40 at 66. Lord Hoffmann continued at 943G-H: "There are sound reasons why omissions require different treatment from positive conduct. It is one thing for the law to say that a person who undertakes some activity shall take reasonable care not to cause damage to others. It is another thing for the law to require that a person who is doing nothing in particular shall take steps to prevent another from suffering harm from the acts of third parties ... or natural causes." 266The distinction is apposite in the present context because s 43A picks up both acts and omissions in prescribing a low standard of care. On the other hand, as will be noted below, the point of distinction is not always clear. 267Turning to the exercise by a public authority of a statutory power, Lord Hoffmann said at 953B: "In the case of a mere statutory power, there is the further point that the legislature has chosen to confer a discretion rather than to create a duty. Of course there may be cases in which Parliament has chosen to confer a power because the subject matter did not permit a duty to be stated with sufficient precision. It may nevertheless have contemplated that in circumstances in which it would be irrational not to exercise the power, a person who suffered loss because it had not been exercised, or not properly exercised, would be entitled to compensation. I therefore do not say that a statutory 'may' can never give rise to a common law duty of care." 268When Lord Hoffmann returned to consider whether the council owed a duty of care to take steps to improve the intersection (at 956B-C) he stated: "Since the only basis for such a duty is the authority's statutory powers..., I will start by asking whether in the light of what the council knew or ought to have known about the junction, it would have had a duty in public law to undertake the work. This requires that it would have been irrational not to exercise its discretion to do so." 269This was not a case in which the road authority had negligently undertaken some work on the road: rather, it was a case in which, knowing of a danger and having power to have it removed, it took some steps to exercise the power but did not take the final step of requiring the landowner to remove the mound. 270The House of Lords has since backed away from this approach. In Gorringe v Calderdale Metropolitan Borough Council [2004] 1 WLR 1057, Lord Hoffmann stated at [26]: "This was the reasoning by which the majority in Stovin v Wise came to the conclusion that the council owed no duty to road users.... But misunderstanding seems to have arisen because the majority judgment goes on to discuss, in the alternative, what the nature of such a duty might have been if there had been one. It suggests that it would have given rise to liability only it if would have been irrational in a public law sense not to exercise the statutory power to do the work. ... The suggestion that there might exceptionally be a case in which a breach of a public law duty could found a private law right of action has proved controversial and it may have been ill-advised to speculate upon such matters." 271In Gorringe, the House of Lords was considering a claim against a local authority by a driver who, on approaching the crest of a hill, applied her brakes suddenly with the result that they locked and caused her to skid into a stationary bus. The plaintiff alleged negligence on the part of the road authority in failing to paint a "SLOW" sign on the roadway approaching the crest. Although there had been a sign painted on the road some years earlier, it had apparently been obliterated by later maintenance work. Nevertheless, because it was not correct to say that the Council had made the highway more dangerous by some positive act, there was no liability to the plaintiff: at [13] (Lord Hoffmann). 272The remarks of Lord Hoffmann in Stovin v Wise appear to have been at least part of the inspiration for the recommendation which gave rise to s 43A: Review of the Law of Negligence, Final Report (Commonwealth of Australia 2002) at par 10.26. (The actual trigger for the introduction of the section was the judgment of Adams J in Presland v Hunter Area Health Service [2003] NSWSC 754, delivered on 19 August 2003.) What is curious about the recommendation is that it appears to assume, as noted above, a duty of care and imposes a standard with respect to which liability is to be judged. In Stovin, Lord Hoffmann was concerned with the existence of a duty of care in respect of the exercise of a statutory power, not amounting, on its face, to a duty. That course has subsequently been abandoned in the UK, in favour of the approach adopted in Gorringe of considering the standard of care as an aspect of breach. In that sense, s 43A conformed to the English approach as at the date of its enactment. 273With this understanding of its historical origins, it is appropriate to turn to the rather awkward terminology of s 43A(3). 274The provision may be addressed in three parts: first, there is the pivotal clause "does not give rise to civil liability". Secondly, there is the subject of that clause, namely the words which precede it. Thirdly, there are the words which follow "unless" and prescribe the circumstances in which liability may arise. It is not necessary to say more about the pivotal clause, other than to note that it assumes the existence of a duty of care. (c) subject matter - "act or omission" 275There are three aspects of the subject matter which require attention. First there is the definition of special statutory power, which has already been addressed. Secondly, there is the relationship between the activity and the power, namely the former "involving" the latter. Nothing turns on that in this case. Thirdly, there is the identification of the relevant acts or omissions: this requires further consideration. 276The dual elements of "act or omission" seem to be reflected in the following pairing of "exercise of, or failure to exercise" the relevant power. However, the apparent structure may be deceptive. In the present case, there was a complaint of omission (with respect to particular signs) in the exercise of the power. That constituted a negligent exercise of the power to install prescribed traffic control devices; it is not apt to say that the power was not exercised. The same analysis applies to the pairing of those expressions in the material following the word "unless". (d) final clause 277The final clause sets a standard. The standard is an act or omission that is "so unreasonable" that no authority could "properly consider the act or omission to be reasonable". This is a curious form of expression: it is not that the act be so unreasonable that no reasonable authority could do the act, but it may perhaps be assumed that the reference to "no authority" is a reference to "no authority acting reasonably". That conclusion is supported by the reference to an authority which "could properly consider" the act to be reasonable. That reading should be accepted. 278The other awkwardness of expression is that the court is required to consider not whether it considers the act (to paraphrase) grossly unreasonable, but rather whether no authority properly considering the matter could consider it to be reasonable. This test has two aspects. First, the identification of the body to be satisfied as to the proper characterisation of the act or omission is not the court but another public authority. That is, it is like the test of apprehended bias in administrative law, which asks not whether the court might think that the decision-maker might not bring a fair and unbiased mind to the task, but whether a fair-minded lay observer might think that the decision-maker might not bring such a mind to the task. With apprehended bias, the identification of the holder of the opinion is understandable; with liability in tort, it is less clear. It means that the court must view the matter through the eyes of a responsible public authority, having particular expertise and functions. (See also the distinction noted in Jamal v Secretary, Department of Health (1988) 14 NSWLR 252 at 262D-F (Kirby P).) 279The second aspect of the statutory language is that the state of mind of the authority is not identified as one which it would or should hold, but rather one which no authority could hold. In other words, it envisages a range of opinions as to what might constitute a reasonable act or reasonable failure to act, but asks if no public authority properly considering the issue could place it within that range. Application of section 280In order to apply s 43A, it is necessary to identify with some care the statutory power being exercised. In this case it was not the power to repair and maintain roads, but the power to place prescribed traffic control devices at the side of a road. Whether that power gave rise to a statutory duty, or whether the common law imposed a duty to exercise the power in a particular manner is not a question which was addressed in the course of the hearing. It was assumed that the Council owed road users a duty of care in determining what signs to place along the road and where they should be placed. 281It must be accepted for the purposes of this case that that conclusion was correct. However, it may be noted that in Gorringe the House of Lords unanimously rejected an argument that the failure by a roads authority to paint a sign on the roadway approaching a crest constituted a breach of duty to a driver who, approaching the crest too fast, applied her brakes suddenly, so that they locked and she slid into a stationary bus. It may be that a different approach should be adopted where the road authority has created the hazard potentially requiring a warning. 282The conclusion of the trial judge in the present case was expressed in the following terms at [70]: "I accept that although other Council officers in Mr Stephenson's position might legitimately have made no provision for the placement of a 'Slippery Road' sign or a sign warning or directing of the need for a reduction in speed from 100 km/h, the fact that minds might differ as to the need for that additional signage and that the Manual allows for the exercise of a considered judgment as to the number and type of road condition signs where there is gravel on the road as an incident of road sealing works does not, in my assessment, meet the necessary test for Wednesbury unreasonableness within the meaning of s 43A." 283In reaching that conclusion, the trial judge set out in detail the various arguments of the parties which established, in her view, a lack of reasonable care - at [48] - but not "Wednesbury unreasonableness". The arguments on appeal reagitated the matters raised before the trial judge. 284The conclusion reached by the trial judge is not, however, determinative. As noted above, the statutory test did not require the judge to form her own view about the matter, but rather required that she determine whether no road authority properly considering the selection of appropriate road signs could have adopted the solution adopted by Mr Stephenson, as the officer of the Council. There was no evidence which, in express terms, supported such a conclusion. 285It is necessary to look in greater detail at the evidence relevant to the issue. Two witnesses were called by the plaintiff. The first, Mr Gerrard Coffey, had been the Director of Technical Services with the Harden Shire Council at the time of the accident. He described his role at that time as technical advisor to the Council on matters relating to engineering, a position formerly described as "the Shire Engineer": Tcpt, 14/02/12, p 96(25). The detail of the traffic control plan used at the site of the roadworks had been prepared by Mr Mark Stephenson, who had been a member of Mr Coffey's staff. 286The plaintiff obtained an expert engineering report from Mr Grant Johnston. The respondent called one witness, Mr Stuart-Smith, from whom it had obtained an engineering report. All three witnesses were asked to comment on certain RTA documents which were in evidence, which included generic traffic control plans and instructions as to the circumstances in which they were to be used. 287Mr Coffey no longer worked for the Council and no statement was obtained from him prior to the trial, nor was notice given that he would be expected to give expert evidence. The tender bundle of documents contained a memorandum from Mr Coffey to Mr Stephenson dated 22 September 2004 and a response from Mr Stephenson, together with a report from Mr Coffey to the General Manager of the Council dated 5 November 2004. 288The thrust of Mr Coffey's concern, as expressed in his evidence, was that the sign for a "Slippery Road", identified as T3-3 on TCP 56, had not been used. His memorandum, which attached a copy of the TCP signed by Mr Stephenson, asked why the slippery road sign was deleted. The answer was said to be "non-responsive", but that was not so. The answer to the previous question given by Mr Stephenson was that the attached TCP was that "only for the spray sealing works", and not for long term warning. Mr Stephenson said that the signage was changed after the sealing works were completed. 289Mr Coffey's memorandum also asked, however, if signs were "checked" at the completion of the sealing and whether the lack of a slippery road sign was noted and "actioned". To that Mr Stephenson had responded that he did not know, as he did not check the site until 8.30pm on the evening before the accident. However, he stated, "at that time I considered the site to be safe as stated in my stat deck [sic]". The latter answer was, in a sense, unresponsive. Mr Coffey gave evidence that he recalled speaking to Mr Stephenson about the answer given and suggesting that it was unresponsive, but said that Mr Stephenson "did not go into any further detail or expand on it at all": Tcpt, p 105(40). Mr Coffey's evidence continued: "Q. Was the 'Slippery' sign part of the RTA specifications adopted by council for use? A. It would have been, yes. It's a standard sign. Q. And what about a 'Reduced Speed' sign? A. Yes, that would have been included. Q. And as part of a TCP plan, was it within Mr Stephenson's power as an accredited or certified RTA - certified by the RTA to do TCPs [-] to put a restricted speed up on these works? A. Yes, he would have used his judgment, being an accredited traffic control plan designer, to know which signs to use for this situation. Q. Dealing with both a 'Restricted' speed sign and the 'Slippery Road' sign, can you see any - firstly, did you receive any explanation for their nonuse in the aftercare of these roadworks? A. No, I never received a proper explanation of that at all. Q. From your perspective as his supervisor, can you see any explanation for their nonuse? A. No, it makes no sense to me at all that one wasn't used. HER HONOUR Q. One being the 'Reduced Speed" sign and/or the 'Slippery Road' symbolic picture? A. Both of them, your Honour, but specifically, the 'Slippery Road' sign. It's a standard sign that's used in this situation." 290Mr Coffey accepted that the same plan appeared to have been used for three sites and agreed that there should have been a site-specific TCP for each. However, as no evidence was given as to the nature of the other sites, it was not possible to infer that the error had occurred because a plan suitable for another site had simply been adopted without proper consideration of the topography of the Kingsvale Road works. However, Mr Coffey did give evidence in relation to the particular site in the following terms: "Q. The Kingsvale Road, as far as roads within the shire council, how does that rate in terms of busyness? A. It's a very busy rural sealed road. It's probably one of the council's more busy roads. Q. And the topography or the country through which it foes, how would you describe that country? A. The alignment of the road or the adjacent vegetation? Q. Well, the alignment? A. Okay, it's quite windy and hilly, so vertical and horizontal alignments are quite windy. Q. I think you told us that Stephenson had the requisite authority to have a "Speed Restriction" sign in place pursuant to a TCP? A. Correct. Q. There's no need for him to seek specific RTA or police permission, is there? A. Not as part of roadworks." 291Mr Coffey gave evidence in cross-examination that he believed the daily traffic on Kingsvale Road would have been 300-400 vehicles: Tcpt, p 167(47). He did not suggest that the generic TCP 56 was other than the most appropriate plan, although it said it applied to roads subject to a 60km/h speed limit, or below. He also agreed that it was appropriate to adapt the TCP to the specific requirements of the site, and that Mr Stephenson was appropriately qualified to prepare the site-specific plan. 292Mr Coffey was cross-examined on the basis that the 'Slippery Road' sign was not required by TCP 56 but was one of a number of signs which were to be used as appropriate (note 4 on the TCP) although it was also said that, "for bituminous surfacing work, [the slippery road and flying stones signs] are to remain in place until loose aggregate is removed": note 9. Although he agreed that it was necessary to have regard to the curvature of the road and the safe speed for negotiating the curve (Tcpt, p 166), he adhered to his view that a slippery road sign was required in the circumstances. 293His evidence concluded with the following exchange both with the trial judge and with Mr Sheldon, senior counsel for the respondent (Tcpt, 170(20)): "HER HONOUR Q. I am not sure that I understand what it means in the context of road works where a bituminous surface has been laid. It does not mean wet to touch, does it? A. No, it does not, your Honour. Q. What does it mean? A. It means that there would be loose aggregate lying on top of the road. When they lay the aggregate, they put down a certain volume per square metre, with the impression that that gets pushed into the bitumen. Q. It means slippery, does it, in the sense that there is a reduced traction where otherwise there would be good traction on a sealed and set road? A. That's correct, your Honour; similar to walking on marbles. SHELDON Q. But you accept, don't you, that the extent to which a given location might be, as you described it, like walking on marbles, in the context of the car driving on a recently resealed road, it is determined by the sharpness of the curve and the amount of aggregate? A. Yes. Q. Can I suggest to you that a reason why a reduced speed sign was not, or may not, have been necessary, was that the corner with the loose aggregate on it could be safely negotiated at a speed of at least 145km per hour? A. Under normal circumstances? Q. No, with the aggregate on the road? A. Loose aggregate on the road? Q. Yes? A. Definitely not. Q. Are you disagreeing with the speed calculation? A. Would you like to repeat the question? Q. Yes. Firstly I should preface it with this; do you understand that the prevailing speed limit at this location was 100km per hour? A. Correct. Q. If the fact is that even with aggregate on the road, following a resealing operation and before sweeping, the corner could be negotiated safely at a speed of 145km an hour, there would be no need for a reduced speed sign? A. Two parts; I disagree that that could occur, that the corner could be negotiated at that speed with loose aggregate on it. But if it can, as you have assumed for whatever reason a reduced speed sign would not be necessary, but it is a circumstance I don't fathom. The premise of the question is just completely wrong." 294Mr Johnston expressed the view in his written report of 25 July 2010 that traffic control signs should have included the imposition of a 60km/h speed zone, whilst there was loose gravel (aggregate) on the roadway. He stated at p 70: "Council has failed to adequately manage the aftercare requirements of this spray sealing project and exposed passing traffic to an unacceptable risk of collision by allowing them to continue passing through the area of this incident at an unsafe speed of 100 km/h." 295Mr Roger Stuart-Smith, the engineer who prepared a report (dated 10 November 2010) for the Council, expressed the conclusion that traffic control at the site "in the form of warning signs was more than adequate and was consistent with TCP 56, which is the relevant traffic control plan for bituminous resurfacing on roads with traffic volumes less than 5,000 vehicles per day". He also expressed the view that neither TCP 56 nor the "over-riding Australian Standard required a reduced speed limit to have been implemented at the site". He also found that a vehicle should have been able to travel around the curve whilst following the road alignment without sideslip occurring at a speed of 145km/h and, accordingly, "at the speed limit of 100km/h, travel on the reduced friction surface of the unswept resealed surface should not have been of concern": p 28. He noted that a loose stones sign had been included in the warnings some 60 metres before the resealed area commenced. 296Prior to trial, the experts were required to prepare a joint report, answering a number of questions. Broadly speaking, both experts agreed that, using Mr Johnston's estimate of available friction, and the estimate of the radius of the curve, "the curve was theoretically passable at a speed of 145km/h". They noted that the calculations assumed "no braking and idealised smooth steering inputs". However, they maintained their disagreement as to whether the traffic control plan used was appropriate. Mr Johnston maintained his view that a 60km/h speed limit was required. 297Neither expert indicated that a "slippery road" sign was required and Mr Johnston expressed no view as to the speed limit, except to find negligence according to the ordinary standard in failing to impose the restriction. A significant focus of the evidence of both experts was the manner in which the accident occurred, which will be dealt with below in discussing causation. 298In oral evidence Mr Johnston adhered to the view that the traffic control plan was inadequate, expressing his opinion in the following terms (Tcpt, p 178(30)): "Because it does not convey the necessary message. It informs the road user that there has been road work. Strictly speaking, the Windscreen Screen Damage sign warns you that loose chips may break the windscreen of an opposing vehicle. It does not guide you, or inform you, of what you should do, where as it does in the case of Reduced Traffic, Slow Down. I personally believe, and from a road safe and technical requirement, it should have been a 60km hour speed limit, and it is almost a 60km speed limit on a recently sealed with loose aggregate road. A loss of reduced speed may have had a similar effect, or better than nothing, and most importantly, the slippery signs were not present, which is a mandatory sign, as I agree with the previous witness, on a recently sealed road with reduced traction." 299The need for a slippery road sign appears to have been raised by Mr Johnston for the first time after hearing the evidence of Mr Coffey, which he effectively adopted. There followed a lengthy passage in which Mr Stuart-Smith (who at that time adhered to the view that the traffic controls were adequate) was questioned by the trial judge with respect to the inference which a driver might derive from the loose stones sign. The cross-examination continued: "HER HONOUR: But you are positing the opinion Mr Stuart-Smith that these signs which consist of the loose stone sign and the road work ahead sign, were adequate to warn Miss Paterson of the risk of harm were she to travel across the forthcoming section of road works where there was loose gravel on the road. You say they were adequate? WITNESS STUART-SMITH: Yes, your Honour. HER HONOUR: There was no necessity in your view Mr Stuart-Smith for any warning in the most direct and glaring terms to reduce speed either by reduce speed, those words, or a reduced speed designation, namely 60 km/h or 40 km/h or something way less than 100 km/h. Do you stand by that view? WITNESS STUART-SMITH: Your Honour used the word necessity, and necessity is, I would have come to this point, the necessity is based on what is specified, an in those terms, there was no necessity, whether or not-- HER HONOUR: I am talking about what your opinion is? WITNESS STUART-SMITH: My opinion is it would have been better to have a reduced speed sign. HER HONOUR: Why is it better to have a reduced speed sign? WITNESS STUART-SMITH: Well, a driver getting into trouble. HER HONOUR: What sort of trouble Mr Stuart-Smith? WITNESS STUART-SMITH: A driver that could lose control in loose gravel. HER HONOUR: And speed is an obvious additional risk where there is a risk of loss of traction by reason of surface gravel. Is that not the case? WITNESS STUART-SMITH: I have said so already your Honour in previous evidence, yes, of course it is. HER HONOUR: I am going to ask you the direct question Mr Stuart-Smith and be aware obviously that the weight of your evidence as tested in this court, has a direct bearing upon my ultimate view as to the weight to which any aspect of your opinion is to be accredited. Do you stand by the view, despite the fact that it would have been better to have a reduced speed sign or a redesignation of the appropriate speed in the absence of anything to indicate loss of speed, the signage on this part of the roadway was adequate, do you stand by that view? WITNESS STUART-SMITH: I am just thinking about the implications. HER HONOUR: As well you might? WITNESS STUART-SMITH: -- of what your [Honour?] is asking. HER HONOUR: Do you want to take the time? I will give you five minutes if you want? WITNESS STUART-SMITH: No. Well I wouldn't mind a minute or two. HER HONOUR: Very good. I will adjourn for five minutes. SHORT ADJOURNMENT HER HONOUR: Yes Mr Stuart-Smith? WITNESS STUART-SMITH: Well the question your Honour, in my opinion, there should have been extra signage to indicate the possibility of reduced friction of some sort or another. I should say though-- HER HONOUR: When you say 'of some sort or another'? WITNESS STUART-SMITH: Possibly the slippery sign would have been an added thing. You could have had a reduced speed sign, you can have had a speed limit obviously. HER HONOUR: And they would have each been advisable in isolation, one of the other, and in combination it would also have been desirable to have such signage in place would it not? WITNESS STUART-SMITH: Primarily in combination, yes. HER HONOUR: But in the absence of the slippery sign, the reduce speed sign or a direction that there is to be a redesignated speed sign, you accept the proposition that the signage was inadequate? WITNESS STUART-SMITH: Yes. I do need to qualify that your Honour, in that I agree with that, the qualification is that it does appear to comply with the recommended TCP." 300Under some pressure from the trial judge, Mr Stuart-Smith accepted the views of Mr Johnston and Mr Coffey that the speed limit on the new section should have been reduced to 60km/h. However, neither Mr Johnston nor Mr Stuart-Smith expressed a view as to the degree of departure from the standard imposed by the duty of care. Even Mr Johnston, who had steadily maintained the view that the 60km/h speed limit should have been applied did not suggest that the failure to do so constituted anything more egregious than a straightforward breach of the duty of care. Mr Coffey's view, on the other hand, could reasonably support an inference that a failure to impose a restricted speed zone, and to provide a slippery road sign, involved conduct which fell outside that which could be expected of a responsible and reasonable Council officer properly considering the design of the traffic control plan for this site whilst there was loose aggregate on the newly sealed section. 301The trial judge was not prepared to draw that inference. In the passage at [70] (set out above at [282]) she concluded that whatever consideration Mr Stephenson did nor did not give to the matter, other council officers in his position might legitimately have failed to take those steps, the need for such steps being a matter as to which "minds might differ". Clearly the trial judge was influenced (and properly so) by the fact that the manual allowed "for the exercise of a considered judgment" in these respects. 302However, there are three factors which support a conclusion that the failure of the Council to impose a speed limit and a slippery road warning sign constituted conduct which was so unreasonable that no road authority properly considering an appropriate traffic control plan for this area of roadworks could have failed to take the suggested steps. 303First, the fact that Mr Coffey took such a view is significant. Not only was Mr Coffey a senior officer with the Council at the time of the accident, who appears to have formed the view at that time that the omissions were inexplicable, but he had prior experience with and training in the preparation of traffic control plans, including whilst an officer at Urana Shire Council: Tcpt, p 164(26). Further, at the time of giving evidence, he was the principal engineer with the ACT Government. His views were therefore the closest evidence at trial to the requisite statutory standard, namely the consideration of how a reasonable road authority (acting through a trained officer) would properly consider the exercise of power. These factors were recognised by the trial judge who stated at [63]: "The only direct evidence bearing on this issue is from Mr Coffey. I am satisfied that he had a level of experience which was sufficient to support what was implicit in his evidence, namely that in his opinion there was an unacceptable departure from what would be expected of other councils exercising similar areas of responsibility under the Roads Act (see Firth v Latham [2007] NSWCA 40 per Hoeben J (with whom Santow JA and McClellan CJ at CL agreed))." 304Conversely, neither of the consulting engineers was invited to address the question in those terms (nor were they qualified on that basis). Indeed, neither was asked to express an opinion otherwise than in conventional terms, that is addressing the usual standard of reasonable care. 305Secondly, although there was general agreement that TCP 56 was the appropriate generic control plan, limited attention was given to the fact that it applied where the speed limit was 60km/h or less. In that circumstance, it is understandable that the plan contained no specific provision for a reduced speed limit and did not require the slippery road sign. By contrast, TCP 353, which did not directly apply because the daily volume of traffic was not reached, but dealt with speed limits exceeding 80km/h, did require a speed restriction to 60km/h and a slippery road sign. Accordingly, whilst TCP 56 in its terms left the question of the slippery road sign to the discretion of the council officer, and did not require a speed limit, both factors required further consideration in circumstances which the TCP did not envisage. 306Thirdly, as Mr Coffey said, his views were based, at least in part, on "common sense": Tcpt, p 169(1). However, he said it was a common sense approach which would have been understood by Mr Stephenson because "that would have been brought out as part of his training". 307While Mr Coffey's simile of "walking on marbles" may have involved a degree of hyperbole, there is no doubt that the potential slipperiness of loose aggregate was the reason for the suggested reduction in speed limit. The characteristic of the section of road under repair which required the reduction in speed limit was not merely the existence of loose aggregate (which might have been adequately indicated by the loose stones sign) but also the fact that there was a curve in the road which had the potential to cause a driver to lose traction. Neither the 'roadworks ahead' sign nor the 'loose stones' sign gave any necessary indication of that risk. 308Although the trial judge was entirely correct in considering whether this was a case where "minds might differ" the answer given did not give full weight to the significance of Mr Coffey's evidence, in the context of the statutory test. Further, there was no suggestion of any countervailing considerations. That is, the additional signage would not have been expensive, time consuming to erect or otherwise inconvenient to the Council. Nor was there any suggestion in the evidence that the additional signs would give rise to potential confusion on the part of a driver. The effect of the combination of the proposed signage would have been internally consistent. 309It was no doubt unfortunate that, as a result of the absence of any notice of Mr Coffey's evidence, the experts were not in a position to comment on it, nor were they asked to. Ultimately Mr Coffey's evidence was admitted without objection. There was no suggestion that Mr Coffey was otherwise than an entirely reliable and credible witness. This material was sufficient to warrant a finding that s 43A of the Civil Liability Act did not preclude civil liability on the part of the Council in the respects identified above. 310I reach this conclusion with some diffidence. First, it is by no means clear that the matter was presented in the manner set out above at trial. Rather, at trial as on appeal, emphasis was placed on the evidence of the expert traffic engineers, which did not support satisfaction of the standard adopted in s 43A. Secondly, although the trial judge did not spell out what she meant in saying that it was a matter about which "minds might differ", that is likely to be a reference to Mr Stuart-Smith's evidence which, even when accepting a breach of duty, did not come close to a concession of (in colloquial terms) gross negligence, or any other expression which might satisfy s 43A. However, neither he nor Mr Johnson was invited to do so. No doubt Mr Stephenson believed he had done all that was required, but he did not give evidence to support such a view. 311Apart from the importance of having regard to the position of a responsible Council, as reflected in the evidence of Mr Coffey, the s 43A test could not be satisfied where expert evidence of value suggested a range of opinion. If the trial judge had properly held that such evidence existed, the appellants would fail. It is arguable that she did; nevertheless the preferable view is that the expert evidence was uncontradicted. Causation 312There remains a question as to whether the failures of the Council, (a) to restrict the speed limit to 60km/h and (b) to include a slippery road sign, caused the accident. The question of causation turned on three issues, namely, (a) the speed at which the deceased was driving when she reached the first section of resurfaced road; (b) the point at which she lost control of the vehicle, and (c) the likely effect of a restricted speed zone and a slippery road sign. 313For reasons which will be explained, the experts were in agreement that the deceased was travelling in excess of 80km/h when she lost control. Mr Johnston undertook a calculation, working back from the point of impact, to suggest an initial speed of about 67km/h-80km/h (first report, par 9.33), which he described as "very approximate and most probably a conservative estimate", expressing a "realistic estimate" as between 80km/h and 100km/h. Mr Stuart-Smith calculated a speed of 88km/h "just before she commenced her oversteer", also expressing the view that the figure may have been higher: report, p 36. 314The section of roadway on which the deceased lost control included three areas of resurfacing. The distances were revealed in the following extract from the trial judge's summary of the facts, taken from her reasons for a ruling on evidence, John Curtis v Harden Shire Council [2012] NSWSC 84: "[8] The tree Ms Paterson struck was in alignment with the commencement of the third section of resealed road (in a southbound direction). This was approximately 210m from where the first section of resealing commenced. There were distinct tyre marks curving across the centre line from left to right on the immediately preceding unworked second section. The tyre marks commenced about 45m from the point of impact. It is common ground that the appearance of the marks and their arrangement are consistent with her vehicle being in a distinct yaw as it 'side slipped' and then rotated in a 180 degree arc towards the point of impact. [9] The first section of reseal travelling north extended over 89m and commenced part of the distance around a 230m curve in the roadway to the right. The second section of reseal over which Ms Paterson must also have passed before losing control ... measured 31m with an unworked 40m section of road intervening between those two segments. There were vehicle track marks of differing density and orientation over both the first and second resealed sections. There were no tyre marks or tracks of any kind on the unworked section in between those two sections." 315Thus, in the southbound direction in which the deceased was travelling, the road commenced a gently curve to the right, followed by a curve back to the left, the latter being beyond the point of impact. The first resealed section (with loose aggregate) commenced as the road eased out of the right hand curve. The resealed section was 87m long, and was followed by a gap of 39m before the second section of resealed road commenced. At least from a point halfway through the first resealed section to the point of impact, the road was straight. 316The tyre marks on the road which were associated with the deceased's vehicle commenced after she had covered approximately 150m from the beginning of the first resealed section. The first tyre mark indicated that as she was coming out of the second resealed area she was on the wrong side of the road, but directing her vehicle back towards the correct side of the road. Immediately she left the second resealed surface, some 54m before the point of impact, there were no tyre marks, however it was thereafter possible to chart her position as she crossed back to the correct side of the road at the gap between the second and third resealed surfaces (an area where there was no loose aggregate) but immediately overcorrected, so that she lost control of the vehicle some 45m before the point of impact with the tree, which was on the right hand (wrong) side of the road, with the rear of the car apparently "fishtailing" so that by the time she left the road on the wrong side and struck the tree, the car had swung through approximately 120 degrees. (By the time it came to rest behind the tree, it had completed a 180 degree turn.) 317The trial judge described the manner in which the question of causation was addressed by the parties in the following passage: "[81] In the way the plaintiffs' case on liability was advanced at trial ... it was conceded that the patent evidence of Ms Paterson's yawing out of control 45 metres from the point of impact (160 metres from the commencement of the roadworks), associated with some evidence of a loss of control in the preceding resurfaced second section by reference to the faint mark in the gravel on the incorrect side of the road, would not support a finding of factual causation. It was common ground that the plaintiffs' case depended upon my being satisfied that Ms Paterson probably lost control in the first resurfaced section on encountering the loose gravel while travelling at or near 100 km/h, and that it was this that caused her to veer progressively off to the right before she travelled (probably on the incorrect side of the road) for some distance, before returning to the correct side of the road after her failed efforts to regain control ...." 318After considering in fine detail the competing hypotheses presented by the experts on the question of causation, the trial judge concluded at [113]: "Given the way the plaintiffs have put their case, the failure to persuade me that control was probably lost because of a loss of traction at the beginning of the first resurfaced section, a primary fact upon which the further and critical finding of factual causation depends, carries with it the consequence that I am not persuaded that but for the failure to provide for and position appropriate signage the accident would not have occurred." 319The appellants' case before this Court implicitly, and somewhat obliquely (there being no ground of appeal specifically addressing this question) took issue with the underlying approach of the trial judge. Secondly, it took issue with the actual conclusion with respect to lack of satisfaction as to any loss of control upon entering the first resurfaced area. Thirdly, the grounds of appeal asserted that the trial judge erred in not finding that causation had been proved pursuant to s 5D(2) of the Civil Liability Act, whilst conceding that the point had not been argued before her. Nothing was said, either in written submissions or in oral argument, in support of this third approach: the ground should be taken to have been abandoned. It remains necessary to consider whether causation was established pursuant to s 5D(1) of the Civil Liability Act. It is convenient to set out the whole section. 5D General principles (1) A determination that negligence caused particular harm comprises the following elements: (a) that the negligence was a necessary condition of the occurrence of the harm (factual causation), and (b) that it is appropriate for the scope of the negligent person's liability to extend to the harm so caused (scope of liability). (2) In determining in an exceptional case, in accordance with established principles, whether negligence that cannot be established as a necessary condition of the occurrence of harm should be accepted as establishing factual causation, the court is to consider (amongst other relevant things) whether or not and why responsibility for the harm should be imposed on the negligent party. (3) If it is relevant to the determination of factual causation to determine what the person who suffered harm would have done if the negligent person had not been negligent: (a) the matter is to be determined subjectively in the light of all relevant circumstances, subject to paragraph (b), and (b) any statement made by the person after suffering the harm about what he or she would have done is inadmissible except to the extent (if any) that the statement is against his or her interest. (4) For the purpose of determining the scope of liability, the court is to consider (amongst other relevant things) whether or not and why responsibility for the harm should be imposed on the negligent party. 320The plaintiffs bore the onus of proving on the balance of probabilities "any fact relevant to the issue of causation": s 5E. Although the section does not say so in so many words, it should be accepted that the plaintiffs bore the onus of proving the ultimate factual inference of causation. The appellants contended that this was not inconsistent with "probabilistic reasoning", by which they meant being satisfied of an inference on the balance of probabilities even where there were competing inferences, so long as the one relied upon was more probable than the others. 321One purpose of s 5D(1), dividing causation into two elements, is to encourage a degree of precision in the way in which such issues are presented and determined. In warning cases (of which this is one) particular complexities can arise. For example, as the trial judge noted with deliberate care, the manner in which the case was argued before her depended upon the plaintiffs establishing that there was loss of control upon reaching the first unsealed section. That was not necessarily how a methodical application of s 5D(1) would operate. The necessary connection to be established was between (a) the failure to warn of the slippery surface of the roadway and to impose a speed restriction because of the slippery surface and (b) the injury resulting from loss of control. 322The evidence did not clearly demonstrate that the deceased's loss of control resulted from the slippery surface or excessive speed. However, whatever the cause, it was common ground between the experts that, had she been travelling at 60km/h she would have stood a much better chance of regaining control, assuming that she had lost it, for whatever reason. On that approach, it was not necessary to establish that she lost control because of the slippery surface in negotiating the resealed sections. Arguably, the harm would not have occurred if the speed limit had been imposed and, thus, one element of the negligence (the absence of a speed restriction) was at least a sufficient condition for the harm. 323A similar (though by no means identical) issue arose in Wallace v Kam [2013] HCA 19; 250 CLR 375, as discussed in the joint reasons of French CJ, Crennan, Kiefel, Gageler and Keane JJ. The plaintiff underwent a surgical procedure which was accompanied by two significant inherent risks of which he was not warned. The trial judge (Harrison J) held that if warned of the risk which materialised, he would nevertheless have undergone the procedure. He therefore dismissed the claim. On appeal to this Court, the plaintiff alleged that if warned of the other (more serious) risk, he would have declined the operation. This Court accepted that factual premise, but concluded that he was not entitled to recover. The case turned on the application of s 5D of the Civil Liability Act. The High Court upheld that result. Referring to the analysis earlier undertaken in Strong v Woolworths Ltd [2012] HCA 5; 246 CLR 182 at [18], the joint reasons in Wallace v Kam noted at [16]: "The determination of factual causation in accordance with s 5D(1)(a) involves nothing more or less than the application of a 'but for' test of causation. That is to say, a determination in accordance with s 5D(1)(a) that negligence was a necessary condition of the occurrence of harm is nothing more or less than a determination on the balance of probabilities that the harm that in fact occurred would not have occurred absent the negligence." 324The joint reasons continued at [19]: "Another scenario is where the patient would have chosen not to undergo the treatment at all if warned of all material risks. In that scenario, a determination of factual causation can be made without difficulty. That is because, absent the negligent failure to warn, the treatment would not have gone ahead at any time and the physical injury would not have been sustained." 325It is at this point that the analogy arises in the present case. Although there are factual findings to be identified, it may be assumed for present purposes that, (a) in veering to the wrong side of the road and attempting to correct her position unsuccessfully, the slippery surface caused by the loose aggregate played no part; (b) had there been a sign showing a restricted speed limit of 60km/h, the deceased would have complied with it, and (c) had she veered onto the wrong side of the road, she would nevertheless have been able to correct her position without losing control of the vehicle. The harm would therefore not have occurred. In these circumstances, should the Council be liable? 326The purpose of the restricted speed zone was to prevent loss of control on the resealed sections, not to limit the consequences of veering across the road through sleepiness or inattention. This, as the High Court noted in Wallace v Kam at [22]-[23], requires a finding as to the "scope of liability" in accordance with s 5D(1)(b) and, to the extent that it is a novel case, by application of s 5D(4). As the joint reasons in Wallace v Kam further explained, attribution of responsibility requires consideration of the purpose served by the imposition of the particular duty of care which has been breached. In some circumstances, liability has been excluded where the harm which results may be described as "sheer coincidence": Chester v Afshar [2005] 1 AC 134 at [94] (Lord Walker of Gestingthorpe). Lord Walker gave the example of a taxi driver driving too fast, with the result that the cab was in the way of a falling tree and the passenger was injured. Absent breach of the duty to drive slower, the injury would not have occurred; nevertheless liability should not attach. The example is similar to that referred to in Wallace v Kam, "of a mountaineer who is negligently advised by a doctor that his knee is fit to make a difficult climb and who then makes the climb, which he would not have made if properly advised about his knee, only to be injured in an avalanche": at [24]. In the language of Lord Hoffmann in Banque Bruxelles Lambert SA v Eagle Star Insurance Co Ltd [1997] AC 191 at 213, the injury was "a foreseeable consequence of mountaineering but has nothing to do with his knee". 327The present case is distinguishable from Wallace v Kam in two important respects (amongst others). First, the nature of the harm suffered by the plaintiff in Wallace v Kam was different in kind, resulting from the materialisation of a risk, different from that which he would not have accepted. In the present case, the harm (losing control of the vehicle and sliding off the road) is of the same kind as that which would have resulted had control been lost due to skidding on loose aggregate. On the other hand, in Wallace v Kam both risks were risks of which the plaintiff should have been warned. In the present case, the risk which eventuated (on the assumptions identified above) was the risk of leaving the road through inattention or distraction, being a risk of which the Council had no duty to warn or guard against by reducing the speed limit. 328In the present case, and on the assumptions set out above, the liability of the Council should not extend to the harm suffered, if the harm did not result from the slippery nature of the loose aggregate. On those assumptions, it was in the nature of a coincidence that her inattention occurred at a point where she should have been warned to slow down because of a risk which she (and no doubt many others) had successfully negotiated. 329Although the foregoing analysis was not undertaken by the trial judge, that was because its conclusion was accepted by the plaintiffs without demur. Accordingly, to the extent that the appellants relied in their written submissions in this Court (at paragraph 26) on the fact that, if the deceased had been travelling at 60km/h she would have been able to respond better to her loss of control, as an independent basis of asserting liability, that should not be accepted. 330The remaining question is whether the appellants' submissions that the trial judge should have been satisfied that the loose aggregate was a material contributor to the loss of control, should be accepted. 331The appellants contended that, where the precise cause of the accident was unknown, the driver being deceased and there being no eyewitnesses, the trial judge was required to determine whether the negligence of the respondent was "a necessary condition of the appellants' harm by the process of probabilistic reasoning adopted in Shoeys Pty Ltd v Allan" [1991] Aust Torts Rep ¶81-104, relied on in Woolworths Ltd v Strong [2010] NSWCA 282 at [60] and referred to without dissent by the High Court in Strong v Woolworths Ltd at [30]. That premise may be accepted: the proposition that the trial judge did not in fact adopt such an approach should be rejected. 332The trial judge noted that the question of causation was to be determined in accordance with s 5D of the Civil Liability Act, and that the burden of proof lay on the plaintiffs, pursuant to s 5E: at [85]. She further noted the acceptance by Ipp JA in Sydney South West Area Health Service v Stamoulis [2009] NSWCA 153 at [138] that a finding of causation "may be made even when the expert evidence does not rise above the possible; the question is always whether the evidence as a whole establishes causation on a balance of probabilities": set out at [84]. In addition, the trial judge referred to the principles stated by McDougall J in this Court, (McColl and Bell JJA agreeing) in Nguyen v Cosmopolitan Homes [2008] NSWCA 246 at [44]-[45]. McDougall J noted that "an actual persuasion" was required to satisfy the burden of proof, meaning that the court must be satisfied that "the probabilities of the fact's existence are greater than the possibilities of its non-existence": at [55]. 333The comparison of probabilities with possibilities, the former satisfying the standard of proof while the latter do not, as used by Ipp JA and McDougall J, is entirely conventional. In one sense, any competing finding may be described as a possibility and the likelihood of its occurrence as a probability; in this context, however, it is common to use the term "probability" to refer to a state of satisfaction greater than 50% and the term "possibility" to refer to something less likely. In any event, such linguistic usage could not demonstrate error. 334It is true that McDougall J referred in Nguyen at [44] to Dixon J in Briginshaw v Briginshaw [1938] HCA 34; 60 CLR 336. However, Briginshaw stands for more than a single proposition. The proposition relied on by McDougall J was expressly identified by him as that set out by Dixon, Evatt and McTiernan JJ in Helton v Allen [1940] HCA 20; 63 CLR 691 at 712: "'When the law requires the proof of any fact, the tribunal must feel an actual persuasion of its occurrence or existence before it can be found. It cannot be found as a result of a mere mechanical comparison of probabilities independently of any belief in its reality'." 335This straightforward proposition, applied daily by judges in civil cases, was not misunderstood by this Court in Nguyen, nor is there any reason to suppose that the trial judge misapprehended its meaning in this case. 336The evidence relied on by the parties fell into three categories. First, there were the objective features of the environment in which the accident occurred, including the signs, the resealed sections of the road with loose aggregate, the unaltered sections and the general topography. Secondly, there was the evidence which sought to reconstruct the events immediately preceding the impact of the car with the tree after it left the roadway. This evidence was primarily the analysis by the expert engineers, based on notes and photographs taken by two Council engineers (Messrs Ellis and Crisp) and the police officers (Messrs Hando and Brand) who attended the scene and recorded their observations. The engineers also made their own observations of the scene of the accident, although much later. 337Thirdly, there was an assortment of material from which less direct inferences could be drawn as to the probabilities of various events. This evidence included the circumstances in which another driver, Ms Skorulis, travelling in the opposite direction, had lost control some 12 hours earlier in the same area. It also included the character and driving history of the deceased, namely that she had an "unblemished driving record over 20 years" and had not been involved in previous accidents, together with the absence of any evidence that her capacity to drive on the day of the accident was impaired in any way. 338The only significant evidence given orally in relation to this issue was that of the two engineering experts. The appellants noted that Mr Stuart-Smith, called by the respondent, had been the subject of "pointed questions" by the trial judge, the passage in the transcript being that set out above at [299]. The submissions on appeal stated: "At the very least, the exchanges between her Honour and the Respondent's expert illustrate that his evidence was presented from a partisan perspective. Where it conflicted with that of the Appellants' expert, it ought to have been entirely rejected." 339This submission should not be accepted. First, the passage set out above in which the trial judge pressed Mr Stuart-Smith to consider whether his adherence to his written report was credible related specifically to the need for a restricted speed limit and a warning sign. It had nothing to do with the issue of causation. Secondly, the trial judge made no finding adverse to the reliability of Mr Stuart-Smith's testimony, either in respect of this issue, or more generally. Given that the issue had been raised specifically by the judge, this cannot have been an oversight. Thirdly, even bearing this robust exchange in mind, there is no hint in Mr Stuart-Smith's evidence with respect to causation that his opinions did not comply with the obligations of an expert witness, with which he said he was familiar. 340In oral submissions, the appellants focused on two specific passages in the reasoning of the trial judge with respect to causation which were said to reveal error. The first, at [77], was in the following terms: "It was agreed between the experts that there was no physical evidence which was capable of fixing where Ms Paterson was likely to have first lost control of her vehicle over that 210 metres." 341The distance of 210m was calculated from the commencement of the first resurfaced section to the point of impact. The trial judge also noted (again as a matter of agreement) that, according to the physical evidence, the vehicle was under control when the deceased entered upon the first resurfaced section. Importantly, neither expert viewed the physical evidence in situ: Mr Johnston was not engaged by the plaintiffs until more than three years after the accident. The road surface was swept clear of gravel later on the afternoon of the accident: at [78]. Accordingly, both experts were dependent upon the sketches and photographs taken by police and Council officers on the day of the accident. 342Mr Crisp, the Council's engineering assistant at the time of the accident, prepared a sketch diagram which included tyre marks running the length of the first resealed section (some 87m) which commenced with the vehicle in its correct lane and left at the southern end of the resealed section with the nearside tyre mark approximately at the centre lane of the road (meaning that the vehicle was almost entirely on the wrong side of the road and continuing in that direction). Senior Constable Hando, who attended the scene and prepared the primary police account of the events stated, in relation to the first resealed section (which he measured with precision) that "[t]here were no visible marks on the road surface." He took a number of photographs of the roadway. 343In his initial written report, Mr Johnston stated: "9.46 As there were no witnesses to the actual loss of control event, the fact that Ms Curtis did not survive to provide an account and the limited quality of post incident recording of the forensic evidence it is necessary to hypothesise on the probable loss of control mechanism. 9.47 It is certainly suggested by all descriptions of the tyremarks that this event is initiated by a loss of traction." 344In his written report, Mr Stuart-Smith stated at 12.2: "As was stated by Senior Constable Brand, there is no apparent reason for the Deceased having drifted or steered to the right side of the road. There is no evidence in the photographs of slideslip marks in the loose material in either of the two resealed surfaces traversed by the Deceased up to the point where she approached the right side of the road where the preliminary tyre mark was reported to have been located." 345In the joint expert report, the engineers were asked: "What was the path taken by the deceased's vehicle from the curve at the start of the roadworks to the point at which the vehicle came to rest? Please list the physical and other evidence which supports the conclusion as to the path taken." 346The agreed answer commenced: "The vehicle appears to have deviated from the left side of the road within the first patch to the right side of the road at the end of the second patch. The evidence for this is a possible/probable (RSS/GJ) tyre mark shown in some diagrams and possibly evident in a Council photograph." 347The latter sentence indicates that Mr Stuart-Smith thought that the tyre mark indicated in diagrams was only possibly a tyre mark, whereas Mr Johnston thought it was indicated as being probably a tyre mark, although both agreed that it was only "possibly evident" in a Council photograph. There is no suggestion that it was evident in any of the photographs taken by the police officer. 348The engineers were then asked, "What is the earliest evidence of the deceased losing control? What and where is that evidence?" Mr Stuart-Smith expressed the view that "[t]he earliest clear evidence of loss of control is at the start of the tyre marks located about 45 metres before the [point of impact]." In his view: "There is no evidence of loss of control at the curve overlapping the first patch", as suggested by Mr Johnston. The factors identified by Mr Johnston are not evidence." 349Mr Johnston expressed the view that the driver had "commenced to lose control around the start of the first section of reseal within the first horizontal curve." He described the evidence as being "the contemporaneous diagrams by a Council officer, the possible indication of a tyre mark within the Council photographs and the known passage of the vehicle from the left side of the road to the right side of the road". The trial judge was entitled to take the view adopted by Mr Stuart-Smith that this material was not "physical evidence" of a loss of control in the first section of resealed surface. That view was not shown to be wrong. 350The experts were asked a further question: "What does the evidence say was the most likely cause of the loss of control...?" Mr Johnston's view was recorded as being "that initial loss of control was probably due to a partial loss of traction upon transitioning into the first section of reseal. This has caused increased clockwise slip angle and a resulting deviation of the vehicle path to the right." Mr Stuart-Smith repeated his view that there was no loss of control at the curve overlapping the first patch and continued: "Had the vehicle lost control in loose gravel, evidence, such as tyre marks and windrows on the loose material of the first patch would have been present. No such evidence was recorded by Police or can be seen in photographs. Mr Johnston's suggestion that the vehicle lost control as a result of rounding the curve in loose gravel is not supported by evidence where evidence would have been expected." 351Mr Johnston did not identify any other evidence for the proposed loss of control in the first section, merely referring back to that which had been identified before. If that did not constitute "physical evidence", none was added. 352In the course of oral evidence, Mr Johnston expressed the view that because the "potential effect of the loose gravel" was to reduce traction, that was the "precipitating factor": Tcpt, p 204(35). He then explained (Tcpt, 205(30)): "If she lost traction on that corner, would you expect an initiation of a clock-wise rotation of the vehicle, or a sense of clock-wise rotation of the vehicle, yes." 353The questioning continued by Mr Sheldon (for the Council): "SHELDON: If you consider that Ms [Paterson] lost control with a clock-wise rotation of the vehicle, you would not expect to see a straight line across to the opposite side of the road as a result of that loss of control, would you? WITNESS JOHNSTON: Not where that actually occurred. Later on, if that was the path the vehicle took, you might. But where the initial precipitating event was, no, you I would not expect a line there. SHELDON: But if that is the path the vehicle took later on, wouldn't you have to assume the regaining by the driver of complete control of the vehicle in the absence of marks, indicating that the vehicle was in a yaw, as to [come] down the incorrect side of the road? WITNESS JOHNSTON: Certainly not complete control, but I would expect the vehicle was under control and largely corrected. So, it was now heading in the wrong direction, but was no longer continuing to rotate. SHELDON: Would you expect there to be marks on the road showing the loss of control that precipitated that manoeuvre? WITNESS JOHNSTON: Provided the area was not completely contaminated by material, yes. I would expect somewhere up near where the police car is, a trained eye would see, or find, that evidence, potentially. HER HONOUR: A trained eye would find what? WITNESS JOHNSTON: A trained eye would find some evidence of that type of loss of control. By that I don't mean general duties police officers or crime scene officers. HER HONOUR: When you say that loss of control -- WITNESS JOHNSTON: A clock-wise rotation of the vehicles due to traction. I suggest if Mr Stuart-Smith or I, if we investigated, might find it. SHELDON: When you say might find it, that is in the realms of speculation -- ... WITNESS JOHNSON: I said that subject to a lot of contamination being up there. So, if we got there fresh, and no one else had been on the roadway and it was clear, I would be fairly confident we would pick it up.... SHELDON: As far as that view applies to this case, that is mere speculation, isn't it? WITNESS JOHNSTON: That's right, we can't even look at a photograph of that area and tell you it's there. There's just not even a photograph there of that." 354The first passage in the judgment of the trial judge subject to challenge (that there was no physical evidence which was capable of fixing where Ms Paterson was likely to have first lost control of her vehicle) was intended to refer to the question whether there was evidence of any loss of control in the first resealed section. The statement is clearly correct: the submissions to the contrary must be rejected. 355The second challenge came to a passage at [83]: "Although Mr Stuart-Smith conceded that it was possible that control was lost in the first resurfaced section, the absence of physical evidence showing the vehicle's passage from the correct to the incorrect side of the road under a loss of traction on the curve - evidence which he expected would have been clearly visible on the road surface were that to have occurred - did not allow him to conclude that the loss of control was probably due to a loss of available friction when, in his view, other competing hypotheses were open, including momentary inattention, drowsiness or fatigue." 356The trial judge continued, in a passage central to her conclusion: "[99] Accepting that the light tyre marks observed by Mr Ellis, and shown in photographs 19 and 20, were probably deposited by Ms Paterson's vehicle, Mr Stuart-Smith noted that she deviated from left to right over the first section of resurfaced roadway in a reasonably straight line without any obvious attempt to correct that course. The photographs objectively support that description. In his view, this was inconsistent with a driver experiencing a loss of traction or recovering or attempting to recover from a loss of traction. [100] He also placed considerable reliance on the absence of any physical evidence indicating the tyres were rotating in a clockwise yaw or sideslip, which might have suggested that the precipitating cause of the deviation to the incorrect side of the road was a result of a loss of traction upon encountering the gravel surface as Ms Paterson continued to negotiate the moderate curve to the right. With no photograph of the road surface where the roadworks commenced or within metres of it (and where, on the plaintiffs' case, traction was lost) and no physical evidence on the road surface of any attempt by Ms Paterson to correct any sideslip or undertake any corrective manoeuvres noted either by the police, Mr Ellis or Mr [Crisp], evidence which Mr Stuart-Smith expected would be visible even to an untrained eye, it was his view that it was unlikely that control was lost as a result of a loss of traction upon encountering the gravel at the commencement of the roadworks. He was of the opinion that the probabilities favoured Ms Paterson deviating from her path around the curve in the first resurfaced section for a reason or reasons unconnected with encountering the gravel at an unsafe speed. While he appointed other hypotheses, including drowsiness or inattention, he was not invited to address whether such physical evidence as was apparent at the accident site supported either hypothesis." 357Although the appellants did not challenge the reasoning in [99], they did assert that the trial judge confused "loss of control" and "loss of traction". For this purpose, they relied upon a passage in the reasoning at [108] (italicised below), but in order to understand the particular passage, it is necessary to put it in its context, commencing at [107]. "[107] It is clear from this evidence that Mr Johnston's opinion was also informed by his belief (or expectation) that physical evidence on the road surface in the first section where he believed traction was lost would likely have been discernable to 'a trained eye', and that neither officers Brand and Hando nor Messrs Ellis and [Crisp] were sufficiently qualified to discern what Mr Johnston claimed he and Mr Stuart-Smith would have been able to discern and [identify] as a tyre mark of a vehicle in a clockwise yaw. [108] Mr Johnston's assumption that the tyre marks were present but not seen could only support a finding that they were probably there if there were some evidence that the marks were there to be seen but overlooked. The obvious flaw in his reasoning is that if there were evidence of a yaw mark or sideslip proximate to where Ms Paterson entered the first resurfaced section (whether or not she regained control of the vehicle thereafter) there would be no need to proceed on the assumption that it was overlooked. To the extent that Mr Johnston's opinion that the probable cause of the accident was a loss of control in the first resurfaced section due to a loss of traction is based on the assumption that there either was physical evidence which was missed by all observers, or there might have been evidence that was missed by those people, the reasoning suffers from circularity in the sense that it is assumed that there were marks on the road because one would expect them to be present if there was a loss of control due to lack of traction. I am not persuaded that the theoretical possibility that the tyre marks were missed by those at the scene (and, in the case of Sergeant Brand, missed despite the fact that he was specifically looking for some evidence of the loss of control) permits an inference to be drawn that the tyre marks were there but missed." 358Referring to the italicised passage, as an example of the confusion, counsel's submissions proceeded (CA Tcpt, p 34(38)): "Firstly, that passage ... involves a misconception that there would have to be a yaw or side-slipping mark for the vehicle to have lost control in the first section when it entered. That ignores Mr Johnson's evidence that a loss of control may be something less than a loss of traction. It is not the same as. The second error apparent in that paragraph is that Mr Johnson's opinion was based upon a loss of traction. It was never suggested that there was a loss of traction per se. What was suggested was loss of control occurred in the first section evident by those marks that Crisp and Ellis saw and which accounted for the crossing from the correct to the incorrect side of the road in circumstances where you would not expect it to have occurred." 359The criticism is obscure. As appears from the passage set out above at [343], from his first report, Mr Johnston was at pains to explain that the "mechanism" by which control was lost was a loss of traction on entering the first resurfaced section. From the appellants' perspective, there was a good reason to maintain such a position. If loss of traction did not lead to a loss of control, causing the vehicle to veer across the road, then the veering was to be accounted for by some other mechanism which was independent of the negligence of the respondent. Furthermore, "loss of traction" is a relative concept: it may or may not lead to a loss of control which causes the vehicle to go in an unintended direction. The appellants' case required that the loss of traction had that consequence. For the trial judge to refer to "a loss of control in the first resurfaced section due to a loss of traction" was to identify precisely what Mr Johnston hypothesised: it was the appellants' case. 360The trial judge considered that Mr Johnston's position was, if not circular, at least compromised. Mr Stuart-Smith's evidence, set out by the trial judge at [101], included the following proposition: "So looking at the photograph, that's on the right-hand side of the photograph, in other words, Ms Paterson's left as she's coming towards us, so one would have expected some windrow or some build up of material, or some ... shading of the tyre mark to indicate some lateral displacement. But not only that. A vehicle doesn't just sideslip ... and then continue just travelling in that same sort of general direction. Once it starts to sideslip, it continues to rotate until the steering has twisted in the other direction, in which case, you either correct - control is regained or else you end up with an oversteer, and the mark is too long." 361Although, as the trial judge noted, Mr Johnston disagreed that "definite windrows or furrows would be expected to be created (or visible) under lateral displacement" - at [105] - she also accepted that it was his opinion that physical evidence would have been available "to a trained eye": at [107]. That approach was entirely consistent with the evidence set out above at [351]. In other words, Mr Johnston's hypothesis depended upon there being some physical evidence which had not been detected. Her Honour concluded that it remained just that: an hypothesis, which she could not be satisfied was probable rather than possible. 362There is nothing in that reasoning which gives rise to doubt: indeed, a careful consideration of the evidence and the manner in which the trial judge dealt with it leads me to the same conclusion. 363The thrust of the appellants' remaining submissions was that had proper weight been given to the circumstances in which another driver, Ms Skorulis, also lost control and went off the road, and had proper weight been given to the known characteristics of the deceased as a driver, it was more probable than not that the reduced traction on the first resealed section which led to her losing control of her vehicle. 364The submissions are unpersuasive. Although at times counsel sought to contend that the trial judge had ignored or not taken account of aspects of the evidence, her consideration of the relevant material was comprehensive. The real force of her rejection of the significance of Ms Skorulis' accident was that Mr Johnston had placed significant weight on aspects which he found compelling, as supportive of his conclusions as to how Ms Paterson lost control. However, Ms Skorulis had been travelling in the opposite direction, at midnight, and lost control on a different section of the road. The trial judge noted, at [112], "that Ms Skorulis' evidence and the objective facts bearing upon her accident, were capable of informing in a rational way how and why it was that Ms Paterson was on the incorrect side of the road some 150 metres from the commencement of roadwork". The trial judge continued: "While I accept that there are a constellation of objective features in both accidents that Mr Johnston found compelling, and which prima facie would tend to suggest the accidents shared a common cause, to make the finding that they were both probably the result of encountering the gravel without adequate forewarning and under speed would also require me to be able to comfortably draw the inference (as distinct from simply assuming) that Ms Paterson, being a more experienced driver than Ms Skorulis, had managed to regain some control of her vehicle after a loss of traction (thus explaining the distance of 150 metres or more over which her vehicle travelled before control was ultimately lost) but Ms Skorulis did not. I am unable to draw that inference. Moreover, it was only after the experts gave further evidence that the marked dissimilarities between the known passage of Ms Skorulis' vehicle and the reconstructed passage of Ms Paterson's vehicle were highlighted." 365Ms Skorulis' accident in effect demonstrated that part of the appellants' case which was upheld, namely the need for speed restrictions and warnings as to the slippery surface. When one reaches the issue of causation, the risk is not in doubt: that the evidence establishes that another driver succumbed to the risk is of little significance in establishing how the deceased lost control. The trial judge was correct to focus on the evidence relating to the passage of the deceased's vehicle. 366So far as Ms Paterson's characteristics were concerned, there is nothing suggestive of error in the trial judge's analysis. It took into account and gave weight to the evidence of Ms Paterson's experience and carefulness as a driver. How much weight was given to her "unblemished" record was a matter for the trial judge. Even experienced drivers with good records are capable of momentary inadvertence; further, the evidence that she reached the resurfacing roadworks maintaining a speed of 100km/h, despite the 'roadworks ahead' and 'loose stones' signs required some allowance to be made for error of judgment. The trial judge had referred to the fact that it was "common ground that Ms Paterson was a careful and experienced driver who was familiar with the road": at [2]. It was referred to again in the passage at [112] set out above. Conclusion 367Without diminishing the high standard of unreasonableness, as explained for example in Allianz Australia v Roads and Traffic Authority, the application of the test was satisfied in the present case. Mr Coffey's evidence justified such a conclusion; that of the traffic engineers did not, but they were not asked to address the relevant standard. 368The trial judge dealt lucidly and comprehensively with the issue of causation, without appellable error. As the appellants' claim must fail on this issue, the appeal must be dismissed; the appellants must pay the respondent's costs in this Court. ********** Amendments 16 March 2015 - correction to coversheet 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. Decision last updated: 17 March 2015