//
© 2026 Zoe. All rights reserved.
Zoe is a legal information platform. Always consult the Federal Register of Legislation for authoritative text.
[2026] NSWCA 184
respondent. Appeal dismissed. Cross-appeal dismissed. The appellant is to pay the respondents’ costs.
The Court of Appeal left in place the finding that Hunter Water was liable in nuisance for flooding caused by a ruptured water main. Hunter Water did not prove that a latent manufacturing defect explained the rupture or that its repair methods reasonably minimised the risk to nearby homes. The Court also held that s 43A did not apply because the liability was based on Hunter Water’s conduct, not on exercising a special statutory power. Both the appeal and cross-appeal were dismissed.
AI-generated legal information, not legal advice. Zoe can make mistakes. Check the cited source, and for advice about your situation consult a qualified Australian lawyer.
9 cited instruments linked from this judgment.
2,096 words · generated 05/10/2026
Hunter Water Corporation owned and operated the DN 500 water main involved in this case. It ran through a road reserve near a townhouse complex in Elermore Vale, downhill from the pipe. On 22 September 2017, the pipe ruptured and a large volume of water flooded the complex. The respondents were the owners corporation and individual lot owners.
Hunter Water had been investigating a possible leak from August 2017. On 21 September, its employees excavated down to the pipe and found a small “weep” at a lead collar. Further equipment was expected the next day, so the pipe was left partly exposed overnight. Hunter Water did not reduce the flow or pressure, despite knowing that pressure would rise as reservoirs filled in the early hours of the morning ([34]-[35]).
The pipe was old, having been laid in 1953. In 2010, a nearby section of the same water main had failed because of excessive corrosion ([32]). The expert evidence about the 2017 rupture was disputed. The possible mechanisms included a manufacturing defect, a crack, corrosion, deterioration, or some combination. The Court did not need to identify the precise mechanism to resolve the appeal: all three experts agreed that excavation and leaving the pipe uncovered contributed to the rupture ([82]-[84], [131]-[132]).
At trial, the owners succeeded in private nuisance but not in negligence or their claim under s 22 of the Hunter Water Act 1991 (NSW). The primary judge considered that s 43A of the Civil Liability Act 2002 (NSW) applied to negligence and imposed a higher standard. Hunter Water appealed the nuisance finding and argued that s 43A also applied to nuisance. The owners cross-appealed on other issues, including the s 43A ruling and the statutory compensation claim. The Court of Appeal dismissed both the appeal and cross-appeal, with Hunter Water ordered to pay the respondents’ costs ([36], [41]-[46], [206]).
The Court applied the High Court’s reformulation of private nuisance in Hunt Leather. McHugh JA described three stages: the plaintiff establishes substantial interference with ordinary use of land; the defendant may then establish that the purpose of its use was common and ordinary; and the defendant must establish that its means of use were conveniently done, meaning that they reasonably minimised the interference ([49]-[52]).
The substantial interference and physical damage were conceded. The decisive question was therefore whether Hunter Water established a justification or excuse. The defendant carried the legal onus on that issue ([52]).
The purpose of the repair work was characterised as maintaining and repairing water mains. The Court rejected the respondents’ submission that the purpose should be framed at the level of the specific repair method, including leaving the pipe exposed. That approach would confuse the purpose of using the land with the means by which it was used. The respondents accepted that maintaining and repairing water mains was common and ordinary locally, so Hunter Water would have succeeded at this stage if it had been necessary to decide the issue ([58]-[66]).
It failed at the third stage. “Conveniently done” required assessment of the parties’ relative interests in all the circumstances. The question was not simply whether the work was routine for Hunter Water, but whether the means adopted reasonably minimised the risk of substantial interference with the homes ([67], [123]). Here, leaving the pipe unsupported overnight while water pressure increased created a real, non-remote risk of catastrophic flooding. A previous nearby failure, known to Hunter Water, and the uncertainty about the condition of the buried pipe were relevant circumstances ([106]-[110], [126]-[128]).
The Court also considered alternatives. Other mains provided some redundancy, and the evidence did not establish that reducing pressure was impractical or involved great cost. Evidence about the general inconvenience of shutting down water mains was not sufficiently specific to show that reducing pressure was unavailable or unreasonable in this situation ([113]-[122]). Hunter Water gave no sufficient reason for exposing the respondents’ land to the identified risk ([127]-[131]).
The latent-defect argument did not change the result. The competing experts’ explanations were uncertain, and the Court found no compelling basis to prefer one theory. Hunter Water therefore did not establish that a latent defect caused the rupture ([84]-[93]). Even if a defect contributed, it would not be decisive where other causes included Hunter Water’s own conduct and it knew of a real risk of failure ([94]-[97]).
Finally, s 43A did not apply. The claim was based on Hunter Water’s conduct in excavating and leaving the pipe exposed while water pressure increased. The fact that a statutory power authorised opening the road reserve did not make the liability one based on exercising that power. Nor was the ordinary operation of Hunter Water’s own water main dependent on a special statutory power ([149]-[181]).
The judgment applies the High Court’s decision in Hunt Leather, which it describes as restating private nuisance ([49]). The relevant formulation requires substantial interference with the plaintiff’s ordinary use of land, then places on the defendant the burden of establishing the relevant justification or excuse, including a common and ordinary purpose and convenient use of means ([50]-[52]).
Leeming JA’s additional observations address the difficulty of identifying what propositions in Hunt Leather bind lower courts. The High Court was unanimous in dismissing the appeal there, but divided in its reasons. Leeming JA identified propositions common to the three judges who adopted the reformulated test, including that the plaintiff need not establish a substantial and unreasonable interference and that the defendant bears the legal burden on the two later issues, subject to an evidentiary onus on the plaintiff to identify the respects in which the use was not conveniently done ([13]-[16]).
Leeming JA considered that Hunt Leather changed the law, contrasting it with earlier High Court statements in Gartner v Kidman and Elston v Dore, which formulated the test in terms of reasonableness ([26]-[28]). That is Leeming JA’s analysis, not a separate holding needed to dispose of this appeal. He also cautioned that aspects of the reformulated test may lack a binding ratio where the High Court judges differed about the legal classification of construction or demolition work. In particular, he considered that no majority had decided whether the purpose of such work belongs under the “common and ordinary” stage or the “conveniently done” stage ([17]-[24]).
On s 43A, the judgment does not decide whether the section can ever apply to a nuisance claim. It decides only that it did not apply on these facts. The essential distinction is between liability based on the exercise or non-exercise of a special statutory power and liability based on conduct that merely involves, or occurs alongside, a statutory power ([149]-[154], [181]).
At [61], McHugh JA explains why the particular repair method was not the relevant purpose: “To descend to the level of the particular method of repair would be to go beyond the purpose of the use and to enter upon the third issue, namely, the means of use.” In plain English, the question whether repair is a recognised purpose is different from the question whether the repair was carried out in a way that adequately considered its effects.
At [67], the judgment says convenient use requires considering “not merely the convenience of the defendant, but also the convenience of the plaintiff.” This means that the defendant’s operational needs matter, but they do not exhaust the inquiry. The impact on neighbouring land must also be assessed.
At [126], McHugh JA states that where the defendant’s means create a real risk of catastrophic interference, the defendant needs to show a “sufficient reason” for adopting them. The passage links the seriousness and likelihood of the risk to whether the defendant has established that its methods reasonably minimised the risk.
At [169], the Court distinguishes between having statutory authority and basing liability on its exercise: “The issue is whether HWC’s ‘liability [was] based on … [the] exercise of, or failure to exercise, a special statutory power’.” In practical terms, the statutory permission to excavate did not itself make this a s 43A case. The claim concerned the conduct alleged to have caused the flooding.
Leeming JA’s observation at [26] is also important for understanding the judgment’s treatment of precedent. He said the better view was that the test in Hunt Leather represented “a change in the law”. That observation helps explain his caution about how older nuisance cases should be used, but it is not necessary to the orders in this appeal.
The judgment is most directly relevant where a defendant’s use of land causes substantial interference with a neighbouring occupier’s ordinary use, and the defendant relies on a justification based on the ordinary purpose of its activity. The decision shows that courts may separate the purpose of an activity from the means adopted to carry it out. A routine or socially recognised purpose does not itself answer whether the means were conveniently done ([61]-[67]).
The risk-based reasoning is especially relevant where the alleged interference is not an inevitable or continuous feature of the use, but a serious harm that may occur if a risk materialises. The Court held that the conveniently-done inquiry could be assessed at the time the use created the risk, whether or not that risk ultimately came home ([124]-[126]). Relevant circumstances here included the known pressure increase, the prior nearby failure, uncertainty about the pipe’s condition, the potential scale of damage, and the availability of other operational measures ([106]-[122], [127]-[131]).
The s 43A analysis may be relevant where an authority argues that a tort claim is based on a statutory power. This judgment directs attention to the pleaded basis of liability: whether the claim depends on the exercise or non-exercise of the power, rather than merely involving conduct for which some statutory authority exists ([149]-[154]). The judgment does not establish that s 43A can never apply to nuisance; it expressly leaves that general question open ([139]-[141]).
The supplied judgment is dated 4 September 2026 and does not report any later court treatment of this decision. It therefore cannot establish how subsequent courts have applied or distinguished Hunter Water.
It does discuss later treatment of the High Court’s decision in Hunt Leather. Leeming JA refers to Enkelmann v Stewart [2026] QCA 67, where the Queensland Court of Appeal treated a passage from Gartner v Kidman as an example of applying the nuisance test, rather than as a defence permitting the defendants to redirect water with impunity ([29]). That is the judgment’s account of Enkelmann, not subsequent treatment of Hunter Water itself.
Leeming JA also notes that the UK Supreme Court had returned to a substantial-and-unreasonable-interference formulation in Jalla and The Manchester Ship Canal Company. He nevertheless says Australian courts remain bound by the High Court’s ratio in Hunt Leather ([12]). This is a discussion of comparative and binding authority, not evidence that later Australian courts have adopted or rejected the Hunter Water reasoning.
The judgment leaves several matters unresolved. First, it does not decide whether s 43A can apply to any claim in private nuisance. It holds only that Hunter Water’s liability in this proceeding was not based on a special statutory power ([139]-[141], [181]).
Second, Leeming JA identifies uncertainty about the precise precedential reach of Hunt Leather. The judges there differed about the characterisation of construction work and whether the purpose of construction belongs in the common-and-ordinary-purpose inquiry or the conveniently-done inquiry. Leeming JA considered that the point lacked a majority ratio and would need to be addressed from first principles in an appropriate case ([17]-[24]).
Third, this judgment does not provide a comprehensive measure for identifying the “proper level of generality” in every kind of land use. It holds that repair and maintenance of water mains was the relevant purpose here and that the particular overnight method of repair belonged to the means inquiry ([61]-[66]). Different activities and localities may require a different characterisation.
Fourth, the judgment does not decide the cross-appeal issues concerning negligence and compensation under s 22 of the Hunter Water Act. Those issues were unnecessary to resolve once the nuisance judgment stood ([183]-[185]). Leeming JA separately observed that negligence might arguably also sustain the judgment if s 43A did not apply, but said the point had not been argued in terms ([8]).
This is general legal information, not legal advice. Zoe can make mistakes. Before acting on or relying on this analysis, have it checked by a qualified Australian lawyer.
Source: NSW Caselaw (www.caselaw.nsw.gov.au). Converted to plain text by ZoeLaw. Not the official version.
Medium Neutral Citation: Hunter Water Corporation v The Owners of Strata Plan 16460 [2026] NSWCA 184
Hearing dates: 1 June 2026
Date of orders: 04 September 2026
Decision date: 04 September 2026
Before: Ward P at [1]
Leeming JA at [2]
McHugh JA at [30]
Decision: (1) Appeal dismissed.
(2) Cross-appeal dismissed.
(3) The appellant is to pay the respondents’ costs.
Catchwords: TORTS – nuisance – private nuisance – plaintiffs’ properties affected by rupture of defendant’s water main – level of generality at which to identify purpose of use – whether purpose of use common and ordinary – whether rupture of water main was caused by a latent defect – whether defendant established that its uses of lands were conveniently done – whether s 43A of Civil Liability Act 2002 (NSW) applicable
COSTS – Party/Party – whether exercise of costs discretion manifestly miscarried – where primary judge apportioned costs in respect of one issue but not others – where issues and evidence overlapped
Legislation Cited: Civil Liability Act 2002 (NSW), ss 40, 41, 43A
Hunter Water Act 1991 (NSW), ss 3, 4A, 12, 13, 19, 20, 22, 23
Roads Act 1993 (NSW), ss 7, 145
Supreme Court Act 1970 (NSW), s 101(2)(c)
Cases Cited: Babet v Commonwealth [2025] HCA 21; 99 ALJR 883
Bath v Alston Holdings Pty Ltd (1988) 165 CLR 411; [1988] HCA 27
Cartwright v McLaine & Long Proprietary Limited and Another (1997) 143 CLR 549; [1979] HCA 16
Cole v Whitfield (1988) 165 CLR 360; [1988] HCA 18
Conexa Sydney Holdings Pty Ltd v Chief Commissioner of State Revenue (2025) 117 NSWLR 209; [2025] NSWCA 20
Conkey & Sons Ltd v Miller (1977) 16 ALR 479
Doppstadt Australia Pty Ltd v Lovick & Son Developments Pty Ltd (No 2) [2014] NSWCA 219
Elston v Dore (1982) 149 CLR 480; [1982] HCA 71
Enkelmann v Stewart [2026] QCA 67
Fearn v Board of Trustees of the Tate Gallery [2024] AC 1; [2023] UKSC 4
Fennell v Robson Excavations Pty Ltd [1977] 2 NSWLR 486
Fisher v Nonconformist Pty Ltd (2024) 114 NSWLR 1; [2024] NSWCA 32
Gales Holdings Pty Ltd v Tweed Shire Council (2013) 85 NSWLR 514; [2013] NSWCA 382
Gartner v Kidman (1962) 108 CLR 12; [1962] HCA 27
House v The King (1936) 55 CLR 499
Housman v Camuglia (2021) 104 NSWLR 615; [2021] NSWCA 106
Hunt Leather Pty Ltd v Transport for NSW [2025] HCA 53; (2025) 100 ALJR 89
Jalla v Shell International Trading and Shipping Co Ltd [2024] AC 595; [2023] UKSC 16
March v E & MH Stramare Pty Ltd (1991) 171 CLR 506; [1991] HCA 12
Owners of Strata Plan No 2661 v Selkirk [2024] NSWSC 760
Pantalone v Alaouie (1989) 18 NSWLR 119
Sydney Holdings Pty Ltd v Chief Commissioner of State Revenue (2025) 117 NSWLR 209; [2025] NSWCA 20
Teubner v Humble (1963) 108 CLR 491; [1963] HCA 11
The Manchester Ship Canal Company Ltd v United Utilities Water Ltd (No 2) [2025] AC 761; [2024] UKSC 22
Transport for NSW v Hunt Leather Pty Ltd (2024) 115 NSWLR 489; [2024] NSWCA 227
Vanderstock v Victoria (2023) 279 CLR 333; [2023] HCA 30
Wallace v Kam (2013) 250 CLR 375; [2013] HCA 19
Texts Cited: A Mohseni, “What is a Nuisance? Hunt Leather Pty Ltd v Transport for NSW” (2026) 142 LQR 378
C Sappideen et al (eds), Fleming’s The Law of Torts, 11th ed (2024, Lawbook Co)
D Rolph et al, Balkin and Davis Law of Torts, 6th ed (2021, LexisNexis)
J Dietrich and I Field, “The ‘New’ Test of Nuisance in Australia” (2026) 48 Sydney Law Review (forthcoming)
M Davies et al, Focus: Torts, 10th ed (LexisNexis, 2024)
Category: Principal judgment
Parties: Hunter Water Corporation ABN 46 228 513 446 (Appellant/Cross-respondent)
The Owners of Strata Plan 16460 ABN 15 870 078 430 (First respondent/Cross-appellant)
Tony Mercurio Furgiuele & Jungsun Furgiuele (Second respondent/Cross-appellant)
Popov, Frank & Danka (Third respondent/Cross-appellant)
Mence, Gregory & Renay (Fourth respondent/Cross-appellant)
Brown, Susan Lynette (Fifth respondent/Cross-appellant)
Martin, Christian (Sixth respondent/Cross-appellant)
Pleash, Lorna (Seventh respondent/Cross-appellant)
Smith, Allison Jane (Eighth respondent/Cross-appellant)
Badior, Roy & Patricia (Ninth respondent/Cross-appellant)
Hiep Thi Hoang & Hong Minh Hoang (Tenth respondent/Cross-appellant)
Ewa Krystyna & Robert Miroslaw Swirad (Eleventh respondent/Cross-appellant)
Bridger, Lorraine May (Twelfth respondent/Cross-appellant)
Carter, Melissa Anne (Thirteenth respondent/Cross-appellant)
Parker, Sharna (Fourteenth respondent/Cross-appellant)
Alana Jane Magee & Kathryn Unger (Fifteenth respondent/Cross-appellant)
Filippou, Nicholas (Sixteenth respondent/Cross-appellant)
Bentkey, Jacquelyn Lorraine (Seventeenth respondent/Cross-appellant)
Spicer, Caryne Louise (Eighteenth respondent/Cross-appellant)
Anderson, Ariana Gabrielle (Nineteenth respondent/Cross-appellant)
Christopher James Bright & Nickey Lee Bright (Twentieth respondent/Cross-appellant)
Fitzgerald, Bruce Alan (Twenty-first respondent/Cross-appellant)
Berry, Shannon James (Twenty-second respondent/Cross-appellant)
Sergeant, Timothy Charles (Twenty-third respondent/Cross-appellant)
Richard Nicholson & Dilian He (Twenty-fourth respondent/Cross-appellant)
Beckhous, Jeremy William (Twenty-fifth respondent/Cross-appellant)
Allan Charles Hamilton & Katherine Margaret Hamilton (Twenty-sixth respondent/Cross-appellant)
Element, Janice Dorothy (Twenty-seventh respondent/Cross-appellant)
Sin, Kallyann (Twenty-eighth respondent/Cross-appellant)
Elena Ana Bennett (Twenty-ninth respondent/Cross-appellant)
Collins, Timothy Jerard (Thirtieth respondent/Cross-appellant)
Fauchon, Richard (Thirty-first respondent/Cross-appellant)
Woseen, Kerry Ann (Thirty-second respondent/Cross-appellant)
O'Reilly, Gerald David (Thirty-third respondent/Cross-appellant)
Brown, Peter Conrad (Thirty-fourth respondent/Cross-appellant)
Marynowski, Andrzej (Thirty-fifth respondent/Cross-appellant)
Peter Burges and Taryn Jade Burges (Thirty-sixth respondent/Cross-appellant)
Gulliver, William (Thirty-seventh respondent/Cross-appellant)
West, Mark John (Thirty-eighth respondent/Cross-appellant)
Newton, Brett (Thirty-ninth respondent/Cross-appellant)
Hayes, Deanne Maree (Fortieth respondent/Cross-appellant)
Stuckings, Beverley (Forty-first respondent/Cross-appellant)
Goodwin, Rodney (Forty-second respondent/Cross-appellant)
Alfred John Williams & Tracey Anne Lewis (Forty-third respondent/Cross-appellant)
Storm, Imogen Kathleen (Forty-fourth respondent/Cross-appellant)
Azhar, Asiya (Forty-fifth respondent/Cross-appellant)
Sidney, Glen Keith (Forty-sixth respondent/Cross-appellant)
Stretton, Kirsty Mary (Forty-seventh respondent/Cross-appellant)
Michael David Wilson & Kim Michelle Wilson (Forty-eighth respondent/Cross-appellant)
Zajac, Sara Lina (Forty-nineth respondent/Cross-appellant)
Rebecca Joy, Van Spreeuwel & Luke James McGuiness (Fiftieth respondent/Cross-appellant)
Morris, Christopher Rodney (Fifty-first respondent/Cross-appellant)
Naddaf, Ashraf Ashmed Musfta (Fifty-second respondent/Cross-appellant)
Cassie Martin & Bikramjeet Singh (Fifty-third respondent/Cross-appellant)
Grant Lawrence Duncan, Elsley & Mary Susan Murphy (Fifty-fourth respondent/Cross-appellant)
Jackman, Thomas Matthew (Fifty-fifth respondent/Cross-appellant)
Zdraveski, James Peter (Fifty-sixth respondent/Cross-appellant)
Weik, Roger Grant (Fifty-seventh respondent/Cross-appellant)
Flaherty, Patrick Mateer (Fifty-eighth respondent/Cross-appellant)
Lane, Jordan (Fifty-nineth respondent/Cross-appellant)
Yefremov, Maxim & Iryna (Sixtieth respondent/Cross-appellant)
Lobel, Keren (Sixty-first respondent/Cross-appellant)
Rock, Shane Gunter (Sixty-second respondent/Cross-appellant)
Morgan, Janice (Sixty-third respondent/Cross-appellant)
Geoffrey Robert Munn & Glennis Joy Munn (Sixty-fourth respondent/Cross-appellant)
Aaron Leonid Sverdlov & Doan Thi Minh Ngo (Sixty-fifth respondent/Cross-appellant)
Phillips, Peter John (Sixty-sixth respondent/Cross-appellant)
Bennett, Mark (Sixty-seventh respondent/Cross-appellant)
Bennett, Mark (Sixty-eighth respondent/Cross-appellant)
Lozada, Ralph Jude Neo (Sixty-ninth respondent/Cross-appellant)
Allen, Paul Robert (Seventieth respondent/Cross-appellant)
Jenny Phuoc Duyen Nguyen & John Phuoc-Thie Nguyen (Seventy-first respondent/Cross-appellant)
Cheah, Su Lynn (Seventy-second respondent/Cross-appellant)
Flaherty, Patrick (Seventy-third respondent/Cross-appellant)
Pike, Julie Maree (Seventy-fourth respondent/Cross-appellant)
Potts, Lauren Kate (Seventy-fifth respondent/Cross-appellant)
Miller, Courtney (Seventy-sixth respondent/Cross-appellant)
Steven Allan Bone & Sharyn Bone (Seventy-seventh respondent/Cross-appellant)
Tracy, Julieann (Seventy-eighth respondent/Cross-appellant)
Galloway, Judith Ann (Seventy-nineth respondent/Cross-appellant)
Ameur, Linda (Eightieth respondent/Cross-appellant)
Bruce Laurence Cable & Jennifer Rose Graham (Eighty-first respondent/Cross-appellant)
Huish, Danielle Sarah (Eighty-second respondent/Cross-appellant)
Hajime Furukawa & Kathryn Anne Sullivan (Eighty-third respondent/Cross-appellant)
Mooney, David Paul (Eighty-fourth respondent/Cross-appellant)
Bull, Margaret (Eighty-fifth respondent/Cross-appellant)
Taylor, Elaine (Eighty-sixth respondent/Cross-appellant)
Fletcher, Robyn (Eighty-seventh respondent/Cross-appellant)
King, Gordon Glen (Eighty-eighth respondent/Cross-appellant)
Mason, Louise Anne (Eighty-ninth respondent/Cross-appellant)
Vanderveen, Amy (Ninetieth respondent/Cross-appellant)
Shane Malcolm Curtis & Catherine Jane Curtis (Ninety-first respondent/Cross-appellant)
Kai Zheng & Ran Ding (Ninety-second respondent/Cross-appellant)
Maughan Sophie Roseanne (Ninety-third respondent/Cross-appellant)
Millward, Karen May (Ninety-fourth respondent/Cross-appellant)
James Michael Whatman & Jessica Anne-Marie Howard (Ninety-fifth respondent/Cross-appellant)
Asadi, Bahador (Ninety-sixth respondent/Cross-appellant)
Gary David Rickard & Margarita Rickard (Ninety-seventh respondent/Cross-appellant)
Farrelly, Martin Benjamin (Ninety-eighth respondent/Cross-appellant)
Lewis, Brock Jordan (Ninety-ninth respondent/Cross-appellant)
Lopez-Andrade, Anthony (Hundredth respondent/Cross-appellant)
Mitchell, Carol Elizabeth (Hundred-and-first respondent/Cross-appellant)
March, Rosslyn Jane (Hundred-and-second respondent/Cross-appellant)
Bisby, Maree Louise (Hundred-and-third respondent/Cross-appellant)
Luc, Michael John (Hundred-and-fourth respondent/Cross-appellant)
Boyd, Ainsley (Hundred-and-fifth respondent/Cross-appellant)
Sarkis, Nicholas (Hundred-and-sixth respondent/Cross-appellant)
Sessions, Brett John (Hundred-and-seventh respondent/Cross-appellant)
MD Rashidul Alam & Arifa Alam (Hundred-and-eighth respondent/Cross-appellant)
Grogan Helen Anne (Hundred-and-nineth respondent/Cross-appellant)
Cooper Kane Allan (Hundred-and-tenth respondent)
Mclachlan, Errol & Wilma (Hundred-and-eleventh respondent/Cross-appellant)
Harrison, James Alexander (Hundred-and-twelfth respondent/Cross-appellant)
Galloway, Judith Ann (Hundred-and-thirteenth respondent/Cross-appellant)
Halligan, Brian (Hundred-and-fourteenth respondent/Cross-appellant)
Wilson, Melissa Maria (Hundred-and-fifteenth respondent/Cross-appellant)
Stuart Hildsworth Lowndes & Megan Therese Lowndes (Hundred-and-sixteenth respondent/Cross-appellant)
Saunders, Rachel Elizabeth (Hundred-and-seventeenth respondent/Cross-appellant)
Hafey, Matthew John (Hundred-and-eighteenth respondent/Cross-appellant)
Representation: Counsel:
M T McCulloch SC and R Gambi (Appellant/Cross-respondent)
M Dawson and A Cameron (Respondents/Cross-appellants)
Solicitors:
Makinson d’Apice Lawyers (Appellant/Cross-respondent)
Mills Oakley (Respondents/Cross-appellants)
File Number(s): 2025/385673
Publication restriction: Nil
[This headnote is not to be read as part of the judgment]
Early in the morning of 22 September 2017, a water main operated by the appellant (HWC) ruptured, flooding a strata scheme townhouse complex downhill. At the time of the rupture, the soil above that part of the water main had been excavated and the water main had been left uncovered overnight, as HWC was in the process of investigating and repairing an apparent leak in the water main. The excavation of the soil removed a source of support which ordinarily counteracted the pressure exerted by water within the pipe. Further, there was a structural weakness of some kind in the length of pipe which burst. When the water pressure rose in the early morning hours as reservoirs filled to supply the day’s water, the rupture occurred. HWC had relevantly been using two parcels of land. First, HWC used land it owned in the separate strata containing the water main to carry water. Secondly, it used the nature reserve surrounding the water main in conducting the excavation.
The respondents, who are the owners corporation for the strata scheme and owners of lots in the complex, claimed against HWC in private nuisance, in negligence and under the Hunter Water Act 1991 (NSW). The primary judge rejected the claim under the Hunter Water Act, held that the claim in negligence was not made out by reason of the application of the higher standard under s 43A of the Civil Liability Act 2002 (NSW) (CLA), and upheld the claim in private nuisance. HWC appealed against that decision and the cross-respondents cross-appealed and sought to uphold the judgment by a notice of contention.
The primary issues on appeal which required resolution were:
(i) whether the purpose(s) of the relevant uses of land were “common and ordinary”;
(ii) whether the rupture was caused by a latent defect;
(iii) whether HWC uses of land were ‘conveniently done’;
(iv) whether CLA s 43A applied to the claim in private nuisance in this case; and,
(v) whether the trial judge’s exercise of the discretion as to costs manifestly miscarried.
The Court held (McHugh JA, Ward P agreeing and Leeming JA agreeing with additional observations), dismissing the appeal with costs and dismissing the cross-appeal:
As to issue (i)
(1) There are three stages to be considered in the tort of private nuisance as re-stated in Hunt Leather Pty Ltd v Transport for NSW [2025] HCA 53; (2025) 100 ALJR 89. The plaintiff must (1) establish a substantial interference with their ordinary use of land. In order to be excused from liability, the defendant must then establish (2) that it was using land “for a common and ordinary purpose”; and (3) that it was using “means that involved it being ‘conveniently done’ in the sense of using means that reasonably minimised the extent of the interference”: [49]-[51].
Hunt Leather Pty Ltd v Transport for NSW [2025] HCA 53; (2025) 100 ALJR 89, applied.
(2) At the second stage, it is necessary to identify the purpose of the use of land at the appropriate level of generality. The purpose should not be identified with such specificity as to defeat the object of the inquiry, namely, to identify whether the defendant’s use serves an interest that warrants recognition in the balancing exercise to be undertaken at the third stage: [62]-[65].
Hunt Leather Pty Ltd v Transport for NSW [2025] HCA 53; (2025) 100 ALJR 89, considered.
(3) HWC’s use of its own land for the purpose of carrying water was conceded to be common and ordinary. In identifying the purpose of HWC’s use of the nature reserve, descending to the particular method of repair employed would go beyond purpose and enter upon the means of use. The purpose was to maintain and repair water mains, which was common and ordinary: [56], [66].
Hunt Leather Pty Ltd v Transport for NSW [2025] HCA 53; (2025) 100 ALJR 89, applied.
As to issue (ii)
(4) The primary judge did not err in failing to find that there existed a latent manufacturing defect in the water main which caused the rupture. Three experts gave conflicting evidence. There was no compelling basis on which to prefer the evidence of one over the others. HWC, which carried the onus, failed to prove that a latent defect caused the pipe to rupture: [79]-[93].
(5) Even if a latent defect had caused the rupture, no ‘principle’ emerging from Cartwright v McLaine & Long Proprietary Limited (1979) 143 CLR 549 at 555-6; [1979] HCA 16 would be engaged where other causes of the rupture involved HWC’s own conduct, and HWC knew of a real risk of a failure: [94]-[97].
Hunt Leather Pty Ltd v Transport for NSW [2025] HCA 53; (2025) 100 ALJR 89; Cartwright v McLaine & Long Proprietary Limited (1979) 143 CLR 549 at 555-6; [1979] HCA 16, considered.
As to issue (iii)
(6) The assessment of whether the use of land was “conveniently done” is directed to the concern that the defendant adopted means for their own ordinary enjoyment of land that reasonably minimised the extent of the substantial interference with the plaintiff's ordinary enjoyment of their land. In a case where there is a risk that the defendant’s use of land will cause a substantial interference, the assessment involves considering whether the means the defendant adopted reasonably minimised that risk, having regard to the relative interests of each of the parties in all the circumstances. HWC failed to demonstrate that it adopted means which reasonably minimised the real risk of catastrophic interference with the respondents’ land: [67]-[78], [98]-[134].
Hunt Leather Pty Ltd v Transport for NSW [2025] HCA 53; (2025) 100 ALJR 89, applied.
As to issue (iv)
(7) In order to engage s 43A the defendant must establish that its alleged civil liability in tort is based on its exercise of, or failure to exercise, a special statutory power. The focus is upon the plaintiff’s claim. There is a distinction between liability that is “based on” the exercise of, or a failure to exercise, a special statutory power, and an act or omission merely “involving” the exercise or failure to exercise such a power: [142]-[154].
Hunt Leather Pty Ltd v Transport for NSW [2025] HCA 53; (2025) 100 ALJR 89; Gales Holdings Pty Ltd v Tweed Shire Council (2013) 85 NSWLR 514; [2013] NSWCA 382, considered.
(8) HWC’s liability was based on its conduct in leaving the pipe uncovered overnight while the water pressure increased. The respondents were not required to prove that HWC exercised (or failed to exercise) any statutory power, or did so in any particular way. Moreover, HWC did not require any statutory power to run water through its water main. Section 43A was not engaged: [155]-[181].
Hunt Leather Pty Ltd v Transport for NSW [2025] HCA 53; (2025) 100 ALJR 89; Gales Holdings Pty Ltd v Tweed Shire Council (2013) 85 NSWLR 514; [2013] NSWCA 382, considered.
As to issue (v)
(10) Leave to appeal was not required to raise a ground as to costs where the appeal raised other bona fide grounds not requiring a grant of leave: [187].
Housman v Camuglia (2021) 104 NSWLR 615; [2021] NSWCA 106, applied.
(11) The primary judge’s exercise of the discretion as to costs did not manifestly miscarry by declining to apportion costs as sought by HWC where the aspects of the claim which otherwise failed were related and involved overlapping evidence with those aspects of the claim which succeeded: [188]-[194].
House v The King (1936) 55 CLR 499; Doppstadt Australia Pty Ltd v Lovick & Son Developments Pty Ltd (No 2) [2014] NSWCA 219, applied.
Per Leeming JA
(12) Consideration of the nature and precedential effect of Hunt Leather Pty Ltd v Transport for NSW [2025] HCA 53; (2025) 100 ALJR 89.
WARD P: I have had the considerable advantage of reading McHugh JA’s detailed reasons and Leeming JA’s additional observations. I agree with McHugh JA, for the reasons his Honour gives, that the orders his Honour proposes should be made. I accept the force of Leeming JA’s observations as to the desirability of elucidating the propositions for which Hunt Leather stands. Nevertheless, I do not consider it necessary for the resolution of this particular matter to engage in that debate.
LEEMING JA: I agree with the orders proposed by McHugh JA, and subject to what follows, with his reasons.
As McHugh JA explains, the plaintiffs advanced parallel claims in nuisance and negligence. The primary judge upheld the former but rejected the latter, with the different outcome turning upon the approach taken to s 43A of the Civil Liability Act 2002 (NSW). The primary judge proceeded on the basis that following this Court’s decision in Transport for NSW v Hunt Leather Pty Ltd (2024) 115 NSWLR 489; [2024] NSWCA 227, he was bound to hold that s 43A did not apply to a claim for nuisance: reasons of the primary judge at [10]. That error reflected a submission repeatedly made by HWC in writing (submissions dated 14 August 2025 at [37] and dated 25 August 2025 at [49]) and it would appear that that concession explains the absence of reasons on this point by the primary judge. The plaintiffs, to be fair, did not go so far. They submitted that s 43A did not apply, but only because there had been no exercise of a special statutory power.
It may be doubted that the reasons for this Court’s decision in Transport for NSW v Hunt Leather Pty Ltd at [174]-[181], which are on any view obiter, preclude reliance on s 43A to meet all claims in nuisance. It is unnecessary to pursue that point, because an appeal has been allowed from that decision, and all members of the High Court explicitly acknowledged that s 43A might apply to such a claim: Hunt Leather Pty Ltd v Transport for NSW [2025] HCA 53; 100 ALJR 89 at [55] (Gageler CJ), [142] and [144] (Gordon and Edelman JJ, with whom Jagot J at [193] and [256] and Beech-Jones J at [300] agreed on this issue). The concession made by HWC at trial concerning s 43A was incorrect. It being a pure question of law, there was no opposition to arguments based on s 43A being run on appeal.
However, for the reasons given by McHugh JA, the liability of HWC was not “based on” the exercise or failure to exercise a special statutory power. As McHugh JA explains, the liability was based upon excavating and leaving uncovered overnight the water main known as DN 500 while at the same time maintaining ordinary water pressure in the pipe overnight. It is tolerably clear that the catastrophic failure of the water main was the consequence of either or both of leaving the excavated pipe exposed overnight (without the earth around the pipe to counteract the pressure exerted by the water) and the overnight refilling of the reservoirs (which increased the water pressure). But if s 43A is to apply, HWC’s liability must be based on the exercise of a special statutory power.
We characterise the purpose of the use of land by TfNSW as construction work for a light rail, which we consider not to be common and ordinary in the relevant localities. Beech-Jones J characterises the purpose more generally as construction work, which is common and ordinary in those localities, but finds that the work was not "conveniently done" (as we also find). Beech-Jones J would have found the purpose common and ordinary in the relevant localities even if characterised as construction work for a light rail.
But that is not the only division within those reasons. There is a further divergence, which applies at least in cases where the alleged nuisance is construction work or demolition.
Construction work and demolition work is ordinarily carried on for a purpose. How is that purpose to be taken into account in the test for private nuisance? Does it fall within the second limb or the third limb?
On the one hand, Gordon and Edelman JJ treated the purpose of construction work as going to whether the purpose was common and ordinary, at [134]:
Construction works on a road for the purposes of repair or upgrading the road are works for purposes that are common and ordinary. Construction works on a road for the purposes of expansion or diversion of the road, or to insert speed bumps or roundabouts, are works for purposes that are common and ordinary. But construction works for the purpose of installing a light rail in place of a road are not consistent with the reasonable expectations of the locality about the purposes for which the land, comprising the road, would be used. The purpose of installing a light rail is not a common and ordinary purpose.
where the substantial interference arises from the activity of demolition and building, the relevance of what is being constructed and the time taken to undertake the construction are matters that form part of the assessment of whether the construction activity is being “conveniently done” rather than whether or not the use of the land for construction of that thing is common and ordinary.
Thus there is no majority of the Court which supports the proposition that the purpose of construction or demolition works should be considered under either the second or the third limb of the reformulated test. I think it follows that, at least in that respect, Hunt Leather is a decision without a ratio. “[T]he ratio decidendi is any rule of law ‘expressed in or necessarily implied by reasons for judgment to which a majority of the participating judges assent’ as a necessary step in reaching their conclusion”: Vanderstock v Victoria (2023) 279 CLR 333; [2023] HCA 30 at [430] (citations omitted); Babet v Commonwealth [2025] HCA 21; 99 ALJR 883 at [97] (citations omitted). The different views about how the purpose of construction or demolition work fits into the test, at least as it seems to me, is not the expression of different views as to the application of a legal test; rather it is a difference about the test itself. The result is that lower courts will have to address this from first principle. Of course, this difference may be more academic than real on the facts of a particular case, and that may be reflected in the observation at [65] that the difference in the passage reproduced above is the only “substantive point of difference”.
Although what is binding in Hunt Leather is complicated, what is not binding is more straightforward. This Court and other Australian courts are bound by propositions of law which are necessary to the determination of the appeal by a majority of the five members of the High Court constituting that Court in that appeal: Conkey & Sons Ltd v Miller (1977) 16 ALR 479 at 485. This Court and other Australian courts are not bound by opinions expressed as to historical fact. That includes whether the test in Hunt Leather is unaltered or innovative.
As flagged above, Gordon and Edelman JJ (with whom Beech-Jones agreed) stated at [115] that “the approach of Bramwell B has been the dominant approach for 150 years in England and Australia”. Gageler CJ and Jagot J denied this (at [15]-[17] and [192]). Both positions cannot be right, and what is the correct position does not turn on whether they have participated in or disagreed with the legal test enunciated by Gordon, Edelman and Beech-Jones JJ. I mention this because the correct position may matter in subsequent cases, including in lower courts. That is because the correct position will inform the precedential status of existing authority. For if there has been a signal change in the law of nuisance when Hunt Leather was delivered, then it will be more likely that earlier authority will have been impliedly overruled. If on the other hand Hunt Leather is merely a confirmation of what has long been settled law, then it is unlikely and perhaps impossible for there to have been an implied overruling of earlier authority.
“Based on my finding that the defendant should have turned off, diverted or reduced the pressure in the pipe and that the effect of the pressure in the uncovered pipe was to cause the burst, then I think it follows that the defendant’s failure caused the damage. Put another way, but for the defendant failing to take precautions when leaving the pipe exposed and the water pressure unaltered, the pipe would not have burst and the damage to the complex would not have occurred.”
His Honour awarded damages for physical damage to property (agreed in the sum of $1,161,841.92) and interest (agreed in the sum of $541,914.95). On the assumption that liability is established, the quantum of this head of damage is not in contest on the appeal.
As to the respondents’ claim for diminution in the value of the properties (claimed in the amount of $20,000 per townhouse), which had been based on the alleged stigma of the properties having been flooded, his Honour rejected that claim, which rejection is not in contest on the appeal.
As to the claim under s 22 of the Hunter Water Act, his Honour rejected that claim on the basis that a common law remedy was available, and that HWC had not entered onto the respondents’ land. This is challenged on the cross-appeal by the respondent on a contingent basis if its claim in private nuisance fails.
Two further points should be made at the outset. First, the ownership of the land HWC used for the purpose of investigating and repairing the water main was not a question taken up in any detail in the argument on the appeal. But it does not appear to be controversial that the road reserve in which the water mains were laid was a public road and not a freeway or a Crown road. In accordance with s 145(3) read with s 7(4) of the Roads Act 1993 (NSW), title to the land was vested in fee simple in the council of the local government area (Council). In the ordinary case, the water main, being a fixture, would therefore become part of the Council’s land: see Conexa Sydney Holdings Pty Ltd v Chief Commissioner of State Revenue (2025) 117 NSWLR 209; [2025] NSWCA 20 at [69] (Payne JA, Ward P, Stern and McHugh JJA agreeing) and at [178] (Basten AJA). But in this case, s 19(1) of the Hunter Water Act vested ownership of the water mains in HWC. This created, in effect, separate strata of land owned by HWC, representing the physical space containing the pipes: see Conexa at [70].
Secondly, in circumstances in which there was no doubt that HWC had lawful authority to excavate the Council’s land, no party attempted to take a point that HWC was not capable of being sued in nuisance for its use of land that belonged to the Council (cf. Fennell v Robson Excavations Pty Ltd [1977] 2 NSWLR 486 at 491-2 (Giles JA) and at 499 (Samuels JA); Pantalone v Alaouie (1989) 18 NSWLR 119 at 129 (Giles J)).
The High Court’s decision in Hunt Leather Pty Ltd v Transport for NSW [2025] HCA 53; (2025) 100 ALJR 89 re-stated the law of private nuisance. Although the primary judge did not have the benefit of that decision when he gave judgment, the correctness of his Honour’s reasons is to be determined in light of the law as stated in that case.
For present purposes the law is as summarised in the joint judgment of Gordon and Edelman JJ, with whom Beech-Jones J relevantly agreed at [260] and [274], as follows (citations omitted):
“[96] From the inception of the modern tort of private nuisance, therefore, the starting point has been that the action requires a substantial interference with the plaintiff's ordinary use of land. Once this has been proved, the focus shifts to whether the defendant has acted with ‘any lawful ground of justification or excuse’.
The defendant’s substantial interference must be one that is not ordinary o**r not convenient
[97] If a defendant causes a substantial interference with a plaintiff's enjoyment of land for ordinary purposes, the defendant may have a lawful ground of justification or excuse in an action for private nuisance if they establish that they were using their land for a common and ordinary purpose, by means that involved it being ‘conveniently done’ in the sense of using means that reasonably minimised the extent of the interference.”
(Italicised heading in original.)
There are thus three stages to be considered:
The respondents as plaintiffs must first establish “a substantial interference with [their] ordinary use of land”.
In order to be excused from liability, HWC as defendant must then establish (see Hunt Leather at [111] per Gordon and Edelman JJ and [274] per Beech-Jones J):
HWC conceded that the respondents had discharged their onus at the first stage (of proving that HWC’s use of land had caused a substantial interference with the respondents’ ordinary use of land). HWC also conceded that the substantial interference had caused physical damage (quantified in the agreed sum of $1,161,841.92). That being so, liability in that amount was prima facie made out, subject to HWC’s establishing any lawful ground of justification or excuse. The focus thus shifts to the two remaining issues of justification or excuse, on which it is to be remembered HWC bore the legal onus of proof.
It might be noted that, in view of these concessions, little attention was paid at this first stage to questions of causation. As will be seen, there appear to have been three matters which together caused, in the sense of being necessary conditions of, the rupture in DN 500. First, that the soil above the water main was excavated such that the water main was left uncovered overnight, depriving it of the support hitherto provided by the surrounding earth. Second, an increase in the water pressure in the early hours of the morning as the reservoirs filled to supply the following day’s water. Third, a structural weakness of some kind in the particular length of pipe (whether as a result of a manufacturing defect, corrosion, decades of service life or otherwise). While HWC might not be responsible for the third matter, the first and second matters were conduct, that is, acts and omissions, of HWC itself.
“The identification of whether a purpose is common and ordinary should be made at the proper level of generality by reference to reasonable expectations based upon what is common and ordinary in a locality. The exercise of identifying the purpose at the proper level of generality can sometimes be finely balanced but it will be guided by the conduct that is said to be the infringement.”
Here, HWC was using more than one parcel of land, and did so for more than one purpose.
First, HWC was using the land it owned in the separate strata containing the water mains for the purpose of carrying water. The respondents did not dispute that the use of land for that purpose was common and ordinary in Elermore Vale.
Second, HWC was also using the nature reserve surrounding the water mains. There is no doubt that the nature reserve was not HWC’s own land. As explained above, it does not appear to be controversial that the Council owned the nature reserve. HWC was using that land for a second purpose: in broad terms, to investigate the water leak and to repair DN 500. In particular, HWC was using the Council’s land when it excavated through some 5 metres of the nature reserve in order to reach the water mains.
Whether the purpose of investigating and repairing the water main was a “common and ordinary purpose” was in contest on the appeal. This issue gave rise to a question as to the appropriate level of generality at which to identify HWC’s purpose. The respondents submitted:
“In the ordinary course we'd accept repair of the water main would be common and ordinary, but here we submit it's not.”
Tcpt, 1 June 2026, 42.21-2.
It might be noted that at various points the respondents expressed the second issue in Hunt Leather in terms of the use of land, rather than the purpose of the use. The respondents’ concession reflected in the oral submissions just excerpted was also not expressed in terms of purpose. However, at other points, the respondent quoted passages from Hunt Leather which were expressed in terms of purpose. The distinction between use and purpose may have practical significance in other cases. But not here, where the respondents’ submissions collapsed the distinction. The respondents must accordingly be understood to have conceded that the use of land for the purpose of maintaining and repairing water mains was common and ordinary in the locality.
The respondents’ point was that that was the wrong level of generality at which to identify the purpose of HWC’s use of land. They submitted that the appropriate level of generality was more specific. They referred to what was said in Hunt Leather at [107], that “identifying the purpose at the proper level of generality … will be guided by the conduct that is said to be the infringement”. They submitted that the use (i.e., the purpose) should accordingly be identified in a way which included the specific method of repair and, in particular, that it involved leaving a partially excavated pipe exposed overnight, in circumstances where the same water main had ruptured in 2010. The respondents then disputed that that purpose was common and ordinary.
It is not strictly necessary to decide this issue in light of the conclusion, adverse to HWC, reached below in relation to the third issue (“conveniently done”). However, the respondents’ argument as to the appropriate level of generality at which to identify HWC’s purpose should not in any event be accepted. To descend to the level of the particular method of repair would be to go beyond the purpose of the use and to enter upon the third issue, namely, the means of use.
The rationale for the test of nuisance explains why care should be taken not to identify the purpose of the use too narrowly at this second stage of the test. As Gordon and Edelman JJ said at [94], “[t]he development of the test for the tort of private nuisance … has been concerned with balancing the competing interests of occupiers of proximate land, usually neighbours, in their enjoyment of their land”. An implicit premise of that statement is that the “competing interests” warrant recognition in the balance.
The second stage of the Hunt Leather test effectively determines, in relation to a defendant, whether the relevant interest warrants recognition. Although the second stage is expressed in terms of purpose (“whether a purpose is common and ordinary”), identifying the purpose of the use of land directs attention to the interest the use serves. The inquiry as to whether a defendant’s purpose was common and ordinary “by reference to reasonable expectations based upon what is common and ordinary in a locality” thus effectively answers the question whether the defendant’s interest warrants recognition in the balance in a given case. The same interest (for example, a defendant’s interest in conducting a particular commercial activity involving a particular use of land) may thus be treated as warranting recognition in one locality but not in another: see at [108]-[109]. The “balancing” exercise then effectively occurs at the third stage: was the defendant’s use of the land “by means that involved it being ‘conveniently done’”?
“[111] Even if a defendant who causes a substantial interference with a plaintiff’s enjoyment of land for ordinary purposes does so by using their land in a common and ordinary way, the defendant will still commit a nuisance unless the defendant establishes that they have used their land by actions that Bramwell B described as ‘conveniently done’. Consistently with the treatment by Bramwell B of this aspect of the tort of private nuisance as a lawful ground of ‘justif[cation] or excuse’, it has been held that the defendant bears the onus of proving that their use of the land which causes such interference is ‘conveniently done’. Of course, a defendant does not bear an onus of negating every imaginable allegation that the means adopted for the use of the land were not conveniently done. A plaintiff bears an evidentiary onus to identify the respects in which the defendant’s use of the land is not conveniently done.
[112] The requirement that a defendant establish that their use of land, for common and ordinary purposes, was ‘conveniently done’ has also been described as a requirement that the defendant’s use of land be ‘necessary for the common and ordinary use’ or, perhaps, more accurately, as a requirement that the defendant’s use of land be ‘reasonable according to the ordinary usages’ or ‘reasonable in the circumstances’. But the use of the land need not be ‘necessary’ in the sense of ‘essential’. And the concern with the reasonableness of a defendant’s use of land is not a test for negligence but a concern that the defendant adopted means for their own ordinary enjoyment of the land that reasonably minimised the extent of the substantial interference with the plaintiff’s ordinary use of their land.
[113] The assessment of whether a defendant’s activity on their land for common and ordinary purposes is ‘conveniently done’ thus requires consideration not merely of the convenience of the defendant, but also the convenience of the plaintiff. In other words, the assessment involves consideration of the relative interests of each of the parties in all the circumstances. …”
(Citations omitted.)
The primary judge, who did not have the benefit of the High Court’s judgment in Hunt Leather, did not use the language of “conveniently done” to determine whether HWC adopted means that “reasonably minimised the extent of the substantial interference with the respondents’ land”. Nevertheless, his Honour made a number of primary findings of fact, as well as evaluative findings as to the reasonableness of HWC’s conduct in the circumstances, that are relevant to the “conveniently done” issue.
In this regard, the respondents submitted that the following findings relevant to the reasonableness of HWC’s use of the land (which findings are not challenged on appeal) support the conclusion that HWC’s repair of the water main was not conveniently done.
Two pipes (DN 600 and DN 1200) ran parallel with DN 500 at the site where DN 500 burst: J[47].
DN 600 and DN 1200 provided redundancy in case DN 500 could not be used. It “was an option available to the defendant to shut off DN 500 while repairs were carried out”: J[107]-[108].
Shutting off, reducing the pressure or diversion of water were available options and, bearing in mind the redundancy provided by the two other pipes, would not have had an effect upon any user: J[99]-[101].
The three experts who gave concurrent evidence (Mr Fennell, Prof Melchers and Dr Clegg) “agreed that a major factor in the bursting of the pipe was the excavation of the ground above it and it being left uncovered by the defendant’s employees”: J[81].
The pressure in the pipe was greater during the early morning hours and that increase was known to HWC: J[97]-[98].
At the time the excavation was occurring and when the area was being left overnight, any defect in the pipe was unknown and there was a possibility that the pipe was heavily corroded (as had been the case in 2010): J[102].
(A finding made with respect to the negligence claim:) the risk that a burst water main would cause significant damage to the respondent’s property was “known” to HWC (in fact his Honour found that the risk was reasonably foreseeable and not insignificant in view of modelling available to HWC and the 2010 failure): J[116]-[117].
A key factual premise of HWC’s argument that its repairs to the DN 500 water main were conveniently done was that, as its expert witness Prof Robert Melchers opined, the cause of the catastrophic rupture was a latent manufacturing defect in the relevant section of the pipe, in that the pipe wall had less than the nominal expected thickness (i.e., 15 mm, as opposed to 18 mm). The bulk of the oral argument advanced on behalf of HWC was directed to establishing the latent defect, about which the primary judge had made no express finding. At one stage, Senior Counsel for HWC described that as “the issue in the case”.
HWC complains that the primary judge erred in not making a finding about the specific causal mechanism of the failure of the water main. Although, for reasons developed below, that issue is not ultimately determinative of the appeal, given the focus HWC gave to the issue it is appropriate to consider the expert evidence about it closely.
HWC submitted that the primary judge did not expressly reject Prof Melchers’ evidence that the cause of the failure was a latent manufacturing defect and that, as Prof Melchers was the only expert with both appropriate qualifications and the benefit of having inspected the relevant section of the pipe, his views should have prevailed.
It is correct to say that the primary judge did not make an express finding that identified the specific causal mechanism of the catastrophic failure of DN 500. On one reading of the primary judge’s reasons, the lack of an express finding may indicate that, in light of the conflict in the opinions of the experts referred to below, his Honour found himself unable to come to an affirmative conclusion about whether a cause of the rupture was a latent manufacturing defect in the relevant section of the pipe.
But whether or not that is so, it is clear that his Honour did not consider the presence or absence of a latent defect to be determinative. That was because, as his Honour said at J[81], “[w]hatever their views were on causation, the three experts who gave concurrent evidence were agreed that a major factor in the bursting of the pipe was the excavation of the ground above it and it being left uncovered by the defendant’s employees.” (Emphasis supplied.) It will be necessary to return to the excavation below. For present purposes it is sufficient to note that the primary judge explained at J[96] and [102]:
“[96] To some extent, the evidence concerning the effect of excavation changes the primary enquiry, for liability purposes, from the identification of a weakness or defect in the pipe to the nature and consequences of the excavation, including its size and it being left overnight. This question also involves whether or not the pipe should have been shut down (or diverted).
…
[102] This is not to say that the nature of the defect in the pipe becomes irrelevant. If the pipe had been heavily corroded the 2010 failure may have assumed a greater significance. But, at the same time, when the excavation was occurring, and when the area was being left overnight, any defect in the pipe was unknown and the possibility of a heavily corroded pipe was present. That the precise defect was unknown does not matter compared to the reasonable action to be taken in case there was a defect capable of leading to a catastrophic failure when the pipe, situated above a large townhouse complex, was left exposed overnight during periods of high water pressure.”
(Emphasis supplied)
As to HWC’s central complaint that the primary judge ought to have found that the pipe contained a latent defect which caused the rupture, the evidence did not compel that conclusion. The main points of divergence among the experts concerned two closely related matters: the location at which the sinusoidal crack in the pipe started and the nature of the mechanism (or “defect”) that led to the rupture. The expert evidence regarding those matters was to some extent speculative, and there was no clear basis (in the experts’ reasoning, investigations, or qualifications) on which to prefer one expert over another on these issues.
As to the experts’ reasoning:
In his first report dated 23 July 2018, Prof Melchers referred to an expert report by Dr David Nicholas dated December 2017. Dr Nicholas had been retained by HWC to undertake a preliminary examination of DN 500 prior to and after abrasive blast cleaning to remove surface corrosion. That report was in evidence, but Dr Nicholas did not give oral evidence at the hearing before the primary judge. Prof Melchers conducted his own visual examination of the pipe on 15 April 2018. He explained his observations as follows.
“As noted, Nicholas considered corrosion as not being the prime cause of failure. The observations on 15 April 2018 confirmed that the nominal wall thickness of the pipe was around 18mm, certainly at the spigot end of the pipe, but that at the cracks in the cracked zone towards, but not at, the socket end of the wall thickness was only 15mm. At these locations there did not appear to be excessive or obvious signs of corrosion. Rather it is likely that the thinner than nominal wall thickness is the result of the manufacturing process. Irrespective of the precise cause, the important fact is that the wall thickness in the main failure zone was (and is) 15mm. It is considered that the occurrence of this low wall thickness where the pipe was to be excavated is not something that could reasonably have been foreseen by Hunter Water.”
(Emphasis supplied.)
“could arise under high internal pressure and removal of the soil above and partly around the pipe, but only if (a) the local wall thickness was significantly less than the nominal wall thickness and (b) the location and orientation of that lower wall thickness was unfavourable in terms of the stresses generated in the local pipe-wall.”
“... the internal pressure of water alone was insufficient to cause rupture of the Water Main if the Water Main did not already have defects present. If the Water Main was defect free, a pressure transient of almost 20 times the head of 58 m would be required to exceed the tensile strength of the pipe. In my opinion, such a transient is likely to have been detected by HWC and no such transient was reported.
Based on these calculations, it is my opinion that a defect in the Pipe must have been present for the Water Main to rupture under the water pressure observed. Furthermore, as the Nicholas Report indicated that there was little evidence of significant pitting, it is considered most likely that the defect was a crack that was under the collar of the adjacent segment of pipe.”
(Emphasis supplied.)
“The reports from Prof. Melchers and Dr. Clegg both state that the forces caused on the pipe by leaving it partially exhumed contributed to the failure of the pipeline. I agree with this analysis, it is not a co-incidence that the failure occurred where it did. Whether there was a pre-existing crack made worse by corrosion, or a manufacturing [sic] is not able to be determined with certainty.
…
The known risk factors prior to the 2017 failure of this pipeline include that it was buried under up to five metres of soil and that any water and mobilised soil, rock and debris would flow directly into residential domestic properties where people lived and slept. This failure to properly consider the risks of the failure led to the pipeline – known to be leaking and heavily corroded after the previous failure in 2010 – to be left in permanent service without any form of flow restriction, and partially exhumed in a manner that was not typically followed when excavating pipes for examination. This failure to consider the risks led not to the failure itself but allowed the avoidable uncontrolled release of water that destroyed property and placed the safety of the public at risk.
In my opinion, the exact mechanism of failure was the combined effects of corrosion and possibly a pre-existing crack in the pipe. However, the catastrophic consequences of that failure was [sic] to partially exhume the pipe whilst fully operational. … ”
“When I first saw the pieces [of the pipe] in the yard at Hunter Water, I just scratched my head and I thought I can’t work out what is going on here; why does a part fail this way, with this sinusoidal curvy behaviour. … A crack along a pipe doesn’t happen by itself; it has to happen under high pressure. But it also has to instigate somewhere, it has to initiate somewhere, and, typically, that’s at a defect along the pipe.
Either usually it’s in the material, but it could be a corrosion pit or something like that, and then it spreads out from there. … “
(Tcpt, 22 August 2025, 157.37-47; emphasis supplied.)
“… Now, what I think is being confused is that the crack spread out … But once that crack started, it then moved sideways and it can only go to the end of the pipe, and that’s the crack that you see at the left-hand end ….
At the other end, there is the a collar, if you like, or the, the bell shaped area that’s much stronger than the rest of the pipe and, therefore, the crack doesn’t go through there. Does that explain it?”
(Tcpt, 22 August 2025, 158.6-15; emphasis supplied)
“… The difficulty one has when you look at a crack like that is how you explain it, because it doesn’t fit any of the standard approaches that we have for the way pipes fail. Pipes fail under pressure and, as this one did, and they fail by cracking, by cracking along the length, if you like. If they fail that way at all, because of what’s called hoop stresses. The pipe basically tries to burst along the whole length circumferentially and outwards, if I can put it that way, and that’s why you get cracks shown at the bottom, in the bottom picture, part C.
So how then do you get this curvy thing? I must be quite clear that it had me stumped for a while; I just couldn't work it out until I looked more closely at the whole geometry of the pipe, and then I discovered that there was a part of the pipe that was thinner. … And the rest of my first report, or part of my first report then deals with how that could possibly raise localised stresses to cause localised failure. Now, one of the things that you find is that if the crack angles are at 45 degrees to particular ways the, the, loads that are applied, or the stresses are applied, then if they’re in, in two directions you typically get 45 degree angle type things … But then I still had to cook up, or find a way to explain it, and that’s in the first report.”
(Tcpt, 22 August 2025, 162.30-50; emphasis supplied.)
“… I don’t believe … it’s possible to say exactly where the crack started. It, it in my experience, the pipes that have broken that I’ve seen, cracks have started at, at the end of the pipe and, and so I, if I were to offer a view as to where that crack started I would say that it started at the spigot end of the pipe and progressed along the pipe and progressed along the pipe. The forces involved in terms of pressure, I think, may have caused that, that sinusoidal shape, as Professor Melchers suggests, but—”
(Tcpt, 22 August 2025, 163.25-31)
“[I] think that the crack started on the plane end, for want of a better word and it started underneath the bell. The explanation for the sinusoidal crack I think is a little bit more subtle. The – I, I think we need to take it in view of the nature of the material, which is cast iron, and my understanding is that the material is centrifugally cast, which is an old technique for producing cast iron. It’s difficult in this method to produce an even wall thickness, because the way in which the, the pipes are made is that they have a, a, a tube which rotates, and then the molten metal is poured into one end as it continues to rotate and then it solidifies in a, in a pipe form.
So the material is quite brittle, but it’s a, you know, it’s an old material that was relatively cheap to make. What happens with cracks in brittle materials is that the crack paths can become very unpredictable, particularly as cracks get long. So if a crack forms at one end and starts to grow, initially the tip of the crack becomes, is, is quite constrained, and it stays straight, then the crack gets longer the level of constraint on the crack tip becomes less and it will tend to wander. And so when I looked at this particular crack I thought that, that’s an indication of, of this wandering effect on a crack.
The exact mechanics of this I’m not entirely sure about, but I certainly found a number of papers which demonstrated that in brittle tubes that are under high pressure and with cracks at one end the crack would initially run straight and then begin to wobble, it would become sinusoidal as it, as it grew through, and it’s largely as a result of, of the crack tip constraint issues. …”
(Tcpt, 22 August 2025, 164.20-43; emphasis supplied.)
“It might have affected the point at which the wobble started, but I don’t think it explains the reason for the crack. I don’t think it, it provides a satisfactory explanation for the crack forming in the position that Dr Melchers, Professor Melchers says.”
(Emphasis supplied.)
“WITNESS CLEGG: It’s possible that the defect started [as] a manufacturing defect within the pipe. If I can just talk about these pipes, these are cheaply made pipes out of cast iron, which is a relatively cheap material and it was made many, many years before – before the failure. It’s a material that is subject to many defects, many potential defects, casting defects of various forms. So it is no surprise to me that there would be defects within the pipe. It’s hard to call this an early failure. It’s been in the ground for 60-odd years, so it’s done a pretty good service, but the defect – but over time the operation of the pipe could lead to the growth of a defect into the formation of a crack, but, yeah, to call it a manufacturing defect causing a failure 60 years after the pipe was put in the ground seems harsh on a manufacturer.
MCCULLOCH: It’s certainly no defect that you could point to that could have occurred through any activity related to the pipe after its installation?
WITNESS CLEGG: Not really. My view is that it’s most likely to have been a gradual degradation.”
(Tcpt, 22 August 2025, 167.26-42; emphasis supplied.)
The reasoning exposed in the experts’ evidence did not provide a compelling basis on which to prefer Prof Melchers’ theory to Dr Clegg’s or Mr Fennell’s (or vice versa). None of the experts suggested that the causal mechanism of the crack was straightforward. To the contrary, even Prof Melchers, whom the primary judge described as “emphatic” in his evidence, acknowledged the difficulty in explaining the cause of the sinusoidal shape of the crack, which “[didn’t] fit any of the standard approaches that we have for the way pipes fail”. That difficulty led to a degree of speculation in the experts’ opinions about what features of the pipe might have resulted in its failure. As Prof Melchers (perhaps unguardedly) expressed the task he had faced, it was “to cook up, or find a way to explain” the crack.
In the result, the overall impression created by the three experts’ evidence was that each had done his best to try to explain the cause of an unusual and catastrophic crack, but that their theories as to the possible mechanisms of the rupture involved a degree of educated guesswork. Mr Fennell and Dr Clegg were candid about the uncertainty involved. Professor Melchers may have been less willing than Dr Clegg or Mr Fennell to entertain theories other than his favoured one, but that does not make Prof Melchers’ reasoning any more compelling. In particular, Prof Melchers emphasised a 3mm departure from nominal thickness of the pipe wall as the cause of the crack. His reasoning on that issue was tied up with his opinion as to where the crack started. Dr Clegg considered it likely that the crack started at the other end of the pipe.
Dr Clegg said that, while the 3mm difference might have affected the shape of the crack (i.e., “the point at which the wobble started” in the brittle pipe material), it did not explain the reason for the crack. In light of Dr Clegg’s evidence (1) that the tensile strength of the pipe material was almost 20 times the hoop stress created by the internal water pressure, and (2) that the nature of the pipe material and the manufacturing process made it difficult to produce an even wall thickness, it is not clear that a 3mm departure from nominal thickness of the pipe wall would, without more, explain the rupture.
There was thus no compelling aspect of their reasoning to prefer any one of the competing theories over the others.
As to the experts’ investigations, although both Prof Melchers and Dr Clegg were called by HWC, HWC submitted that the primary judge should have preferred Prof Melchers’ evidence in light of the fact that, of the three experts, only Prof Melchers had the advantage of inspecting the pipe whereas Dr Clegg relied on photographs. In circumstances where HWC did not elaborate on what specific advantage would be gained by examining the pipe in person and how that examination informed his opinion, that submission should not be accepted. In particular, while it is true that Dr Nicholas’ inspection of the pipe led him to indicate in his report that there was little evidence of significant pitting, Dr Clegg clearly took that into account as one of his assumptions.
In light of that conclusion, it is not strictly necessary to decide whether, if the court had found that the pipe’s failure was caused in part by a latent defect, that finding would have been fatal to the respondents’ claim in nuisance. However, even if the primary judge had made such a finding, it would not follow that the repairs were “conveniently done”.
HWC submits that “there can be no nuisance where the interference with the plaintiff’s land is the direct result of the manifestation of a latent defect”: Tcpt, 6.45-6. It relies on the following statement of Gibbs ACJ in Cartwright v McLaine & Long Proprietary Limited (1979) 143 CLR 549 at 555-6; [1979] HCA 16:
“The reasoning on which [Wringe v Cohen [1940] 1 K.B. 229] was based has been strongly criticized by some academic writers, and the rule which it states has been said to be inconsistent with the general principles of the law of nuisance as laid down in Sedleigh-Denfield v O’Callaghan. However the case has been followed in Heap v Ind Coope and Allsopp Ltd and Mint v Good and it was mentioned without apparent disapproval by Dixon J in Torette House Pty. Ltd. v Berkman. I may assume, without deciding, that the decision should be followed in Australia. The principle for which it is authority applies only when danger has been caused by a want of repair, and is not applicable when the nuisance was caused by the act of a third party. This was made clear by the concluding passage of the judgment of the Court in Wringe v Cohen where it was said:
‘If premises become dangerous as the result of something done by an occupier and they cause damage, the occupier is liable although he did not know of the danger and was not negligent in not knowing. That principle is in harmony with the principle laid down in Tarry v Ashton, that if premises become dangerous as a result of failure to repair and cause damage the occupier is liable, although he did not know of the danger and was not negligent in not knowing. Positive acts and neglect of duty are thus placed on the same footing. On the other hand, if premises become dangerous, not by the occupier’s act or neglect of duty, but as the result of the act of a third party, or of a latent defect, the occupier is not liable without proof of knowledge or means of knowledge and failure to abate it.’
In Cushing v Walker & Son Hallett J. said that if the gravamen of the complaint is the causing of the nuisance, in particular by a breach of a duty to repair, it is irrelevant whether the defendant knew, or ought to have known, that the nuisance had in fact been caused, but that the position is different where the gravamen of the complaint is the continuance of a nuisance caused by another.”
(Citations omitted; emphasis supplied.)
The “principle” to which Gibbs ACJ referred appears to follow as a matter of application of the test of whether the use was ‘conveniently done’, in that, if the sole cause of the interference was a latent defect of which the defendant neither knew nor ought to know, it may be difficult to contend that the defendant had failed to use means which reasonably minimised the extent of the resultant interference.
The passage in Cartwright does not assist HWC. Assuming that the “principle” to which Gibbs ACJ referred is the law, it would be engaged only if the sole cause of the pipe’s failure was a latent defect of which HWC neither knew nor had the means of knowing. The principle is expressed not to apply where the nuisance is caused by the defendant’s “act or neglect of duty”. But here, HWC accepts that there were multiple causes of the rupture, including the excavation, the fact that the pipe was left uncovered overnight, and the increase in the water pressure overnight. Those matters involved HWC’s own conduct of which it was well aware. Moreover, as noted below, the existence of a real risk of a failure in DN500 was known to HWC.
Independently of its argument based on expert opinion as to the presence of a latent defect, HWC also relies on the evidence of several lay witnesses. As I understand the argument, it is that the lay evidence shows that there was nothing out of the ordinary in excavating the land above the pipe or leaving the trench unfilled overnight while equipment was obtained. It follows, so I understand the argument, that those things were conveniently done, in the sense that HWC adopted means which reasonably minimised the extent of the substantial interference with the respondents’ use of their land, bearing in mind, as Gordon and Edelman JJ said, that “the assessment involves consideration of the relative interests of each of the parties in all the circumstances”.
It should be noted that HWC did not press Ground 13 of its Amended Notice of Appeal dated 8 April 2026, which challenged the primary judge’s ruling limiting certain portions of the evidence of Mr Kirby Morrison, Mr Luke Bianchi and Mr Clapham to the respondents’ claim in negligence. As such, this evidence must be disregarded on the appeal against the primary judge’s judgment on the nuisance claim.
The following chronology provides context to HWC’s submissions.
On 3 August 2017, HWC was informed that there was a possible leak at the corner of Aries Way and Apollo Close, Elermore Vale.
HWC employees attended the site on 8 and 9 August to perform a “correlation” to identify the location of the leak. A further correlation was performed by Active Leak Detection on 23 August, who suggested that the leak was likely emanating from DN 600.
On 28 August, two employees of HWC, Mr Mitchell Knight and Mr Les Thornton, attended the site to excavate down to the pipes. Due to the limitations of the excavator they brought with them, they were unable to reach the top of the pipe and they backfilled the excavation.
On 19 September, two different HWC employees, Mr Shane Hayter and Mr Michael Berry, attended the site with a larger excavator. They were unable to reach the pipes. The excavated soil was removed from the site.
On 20 September, Mr Berry and Mr Sean Manning attended the site. They excavated further and requested a shoring box to stabilise the excavation. The excavated soil was again removed.
On 21 September (the day before the pipe ruptured), Mr Berry and Mr Hayter returned to the site to continue the excavation. They managed to reach the water main, at a depth of about 4 to 5 metres underground, using an excavator and hand-digging the last 300mm. The collar of the water main was visible after removing dirt. Mr Berry observed that the lead jointing of the collar had been pushed out slightly (less than 10mm) at approximately 12 o’clock, that the pipe was damp but the soil still firm, and that a “weep” was visible on the side of the pipe at approximately 9 o’clock. Mr Berry and Mr Hayter removed more soil so that approximately 0.5 metres of the length of the pipe was visible**.**
“Inspection on site at Elermore Vale on the day of failure identified significant corrosion along the piece that had already been removed on arrival at site … . The corrosion was locat[ed] on the underside of the pipe. This pipe is buried quite deep in a clay layer, our normal soil sampling program only extends to depths of about 2 metres … .The soil sample would have missed the clay layer the pipe is exposed to and would have picked up the topsoil instead. This, in all probability, underestimated the corrosiveness at this site. This limitation of the augering is quite unusual, as is pipes in excess of 2 metres in depth. The pipe depth was almost certainly a result of fill being placed over the pipe during the construction of Aries way. Along this 200 metre section the pipe is up to 7 metres deep. It is very difficult to obtain samples at this depth.”
Mr Dafter’s report suggests that, in the days leading up to the 2017 incident, there was a real risk that the pipe was affected by corrosion and that the pipe could fail in a manner that would have catastrophic consequences, as had been the case in 2010. Mr Dafter’s observation that the corrosion that caused the 2010 failure was located on the underside of the pipe has particular significance given that on 21 September 2017, HWC’s workers had been able to expose the top of the pipe, but were unable to expose the circumference of the pipe’s collar. It was not possible for them to assess the condition of the pipe’s underside and further excavation would have been necessary in order for the proposed repairs to be undertaken.
Secondly, as the primary judge recognised, in light of Mr Dafter’s report HWC should be taken to have been aware of the possibility that DN 500 was buried unusually deep (on the evidence, it was approximately five metres deep). As his Honour explained, this was overlooked by HWC’s staff when they twice attended the site in 2017 with excavators that were not capable of digging to the relevant depth (a factor that led to the pipe’s being left exposed overnight).
Thirdly, the depth of the pipe combined with the incorrect identification of which pipe was leaking meant that the excavation was also far more extensive than originally predicted.
Another limb of HWC’s argument is directed to the fact that it did not shut off DN 500 or divert or reduce the water flow. Instead, it continued to carry water through DN 500 in the ordinary way, including the usual increase in the water pressure in the early hours of the morning. As noted above, the increase in the water pressure was one of the causes of the pipe’s failure.
It was not ultimately disputed at trial that DN 1200 and DN 600 provided a level of redundancy if DN 500 was shut off: see J[107], referring to Mr Wynn’s eventual concession of this point in cross-examination. Logically, the same redundancy would be available if required because the water pressure in DN 500 was reduced. However, HWC submits that the primary judge’s finding at J[112] that HWC “acted unreasonably in leaving the pipe exposed overnight without shutting off, diverting the water flow or reducing the pressure in the pipe” — which goes to the “conveniently done” question — involved “the error of using hindsight to reason towards a conclusion”. HWC submits that there “was absolutely no warrant for turning off, diverting or reducing the water flow in DN 500”.
With respect to the primary judge’s finding at J[99] that “[s]hutting off, reducing the pressure or diversion of water were available options and, bearing in mind the redundancy available through the other two pipes, would have had no effect upon any user”, HWC relies on the evidence of Mr Main and Mr Wynn which it submits “demonstrated why intentional interference with the flow of water required a major event such as the major loss of water which only occurred once the latent defect manifested itself.”
Mr Main’s affidavit of 4 November 2024 concerned HWC’s licence obligations. He referred to several clauses of HWC’s Operating Licence 2017-2022, issued under s 12 of the Hunter Water Act:
“(a) Clause 3.1, which required that drinking water supplied to customers had to meet the performance requirements specified in the Australian Drinking Water Guidelines or such other standard as directed by NSW Health;
(b) Clause 3.3.1, requiring Hunter Water to ensure that no more than 4,800 properties in a financial year experienced a water pressure failure (being a pressure of less than 20 metres pressure for a continuous period of 30 minutes or more) … ;
(c) Clause 3.3.2, requiring Hunter Water ensure that no more than 10,000 properties in a financial year experience unplanned water interruption that lasts for more than 5 hours … .”
HWC submitted that the “statutory requirements under its operating licence” to provide water on a regular basis were relevant to the question whether HWC should have shut off, diverted or reduced the pressure in DN 500. However, Mr Main did not elaborate further on HWC’s “statutory” obligations, nor on how shutting off, diverting or reducing pressure in DN 500 would have caused HWC to breach either the statute or its licence obligations. Mr Wynn said in his affidavit of 23 May 2025 that “[a] failure to ensure the continued distribution of water throughout the Region would amount to a breach of the Operating Licence”, and that the “need for actioning rectification requires somewhat of a balancing act between the risk of a failure and the cost associated with damage and rectifying a failed asset.”
Statements at that high level of generality are of little assistance in determining how, if at all, HWC’s licence conditions bear on an assessment of whether what occurred overnight on 21 and 22 September 2017 was conveniently done, particularly given the redundancy in DN 600 and DN 1200. That being so, little weight can be given to HWC’s argument that its statutory or licence obligations assist it to establish that its continued use of DN 500 to carry water in the usual course (rather than shutting it off, or diverting or reducing the pressure) was conveniently done.
Mr Wynn’s affidavit of 23 May 2025 also contained the following statement concerning HWC’s general approach to shutting water mains, as applied to DN 500.
“In respect of the subject watermain, even if Hunter Water concluded based on sufficient relevant information, that a failure of some part of the watermain might be imminent, it would not have shut that watermain because, inter alia:
(a) it was required, as best it can, to provide a base-line level of water supply;
(b) shutting down that watermain would not be temporary, and its capacity was necessary particularly with the peak water supply season approaching;
(c) it still needed to supply water, which could not be maintained with increased demand;
(d) the system required a back-up supply in case of difficulties being encountered with the other watermains;
(e) it would need to replace the entire pipeline, which would create a significant delay during design so as to minimise interruption to its customers, request for funding, and construction.”
This evidence was directed to shutting off the DN 500 water main altogether. To that extent, the evidence was not irrelevant to the “consideration of the relative interests of each of the parties in all the circumstances” for the purposes of assessing whether HWC’s use of its land to continue to carry water through the mains in the usual way on 21 and 22 September 2017 was conveniently done. But importantly, that evidence did not address the possibility of reducing the pressure in DN 500 overnight. Nor did it address the available redundancy in DN 600 and DN 1200.
As to the practical steps involved, HWC submits that it was an “oversimplification to say that [DN 500] could just be turned off” and that this was “a very good example of an error in hindsight reasoning”. HWC relies on an affidavit of Mr Morrison, Manager of Water Network Operations at HWC, dated 31 October 2024, who outlined the process to shut down a water main as follows.
“The time which is required to shut down a water main would depend upon the water main in question, including the condition of its valves and whether any damage occurred (such as a valve breaking) during the process of shutting down each main. Decisions also need to be made as to the number of valves which need to be shut down for the relevant repair.
The process of returning the water main to service following a shut down for repair is also taken into consideration. Depending on the type of repair undertaken, there are a number of issues which need to be managed when returning the water main to service. The empty section of the water main may need to be flushed, bled for removal of entrained air, and sometimes disinfected. The section would need to be refilled and continuously flushed until the water quality passing through the pipe is of acceptable quality, with the flushed water being appropriately treated to remove chloring prior to discharge to avoid environmental harm. This can be a complex and time consuming task and could take a number of hours, and potentially days depending on the number of sections shut down and the length of those sections, as well as the build up of sediment or other contamination that may have occurred during the repair work within those sections.
In my experience, Hunter Water’s field workers would normally make a judgment as to whether a water main needed to be shut down.”
That was, again, evidence at a high level of generality. Despite noting that the time required would depend “upon the water main in question”, the evidence did not descend to analysis of the relevant water main. Again, it had some relevance for assessing, in light of the parties’ relative interests in all the circumstances, HWC’s conduct in not shutting off the water to DN 500. But that evidence fell well short of showing that shutting off the water was not an available option, particularly when it was not directed to a situation where other water mains were available which provided redundancy for the one that was to be out of operation. Still less did Mr Morrison’s evidence explain why reducing the water pressure in DN 500 (as opposed to shutting it off) was not an available and practical option. It follows that there was no error in his Honour’s findings at J[99] and J[108] that shutting off DN 500 or reducing the water pressure were available options.
However, as Gordon and Edelman JJ made clear, assessing whether a defendant’s use of land is conveniently done involves considering not only the plaintiff’s interests but also the defendant’s. Although the evidence led by HWC was largely at a high level, it did suggest that shutting off the water to DN 500 would be a significant matter. On the other hand, there was little if any evidence that reducing the water pressure in DN 500 would cause HWC any great difficulty or cost. Those matters may be relevant to assessing whether HWC’s conduct in using its own land to continue to carry water in DN 500 as usual overnight (that is, on the footing that the water pressure in DN 500 would rise in the early hours of the morning as usual) was conveniently done.
As noted above, at the third stage of the Hunt Leather test, “the assessment involves consideration of the relative interests of each of the parties in all the circumstances”: at [113]. The assessment of whether the use was “conveniently done” is directed to what Gordon and Edelman JJ described as the “concern that the defendant adopted means for their own ordinary enjoyment of land that reasonably minimised the extent of the substantial interference with the plaintiff's ordinary enjoyment of their land”: at [112].
Here, HWC’s uses of both its own land (to carry water in the mains) and the Council’s land (to conduct the repairs) would not necessarily entail any substantial interference with the plaintiffs’ ordinary use of their land. The case is to that extent different from the common cases of excessive noise, dust, or foul odours. Instead, HWC’s uses of its own land and the Council’s land created a risk of the manifestation of a substantial interference with the respondents’ ordinary use of their land.
In such a case, consideration of the parties’ relative interests in all the circumstances involves considering whether the means the defendant adopted reasonably minimised that risk. That in turn requires an assessment of the means adopted by the defendant in light of the likelihood that the risk will come home and the nature and extent of the interference to a plaintiff’s land if it does. Further, in a case involving a risk that a defendant’s use of land will cause substantial interference with a plaintiff’s ordinary use of land, the question whether the defendant’s use is conveniently done must be capable of determination at the time the use occurs and creates the risk, whether or not the risk ultimately comes home.
Here, the risk was that the means adopted by the defendant would cause the water main to burst and flood the respondents’ land. As to the likelihood that the risk would come home, in light of the matters discussed above, that risk was plainly real (as opposed to remote). As to the nature and extent of the interference to the respondents’ ordinary use of their land if the risk came home, the effect of a burst water main on the properties downhill would be — as it in fact was — catastrophic. In such a case — where the means the defendant adopted created a real risk of catastrophic interference — in order for the defendant to establish that its use of land was conveniently done, the defendant will need to show a sufficient reason for adopting those means to justify exposing the plaintiff’s land to that risk. Otherwise, the defendant will not have shown that the means adopted reasonably minimised the risk.
In the present case, no sufficient reason has been shown. The means HWC adopted involved:
(a) leaving a leaking water main that had been excavated uncovered (and hence unsupported by the additional strength provided by metres of covering soil) overnight; and (b) permitting the pressure to build inside the water main;
in circumstances where:
(c) the leak was in a section of the water main that was within 100m of another section of the same main which had catastrophically failed only 7 years earlier because of corrosion; and (d) it was unknown (and could not be determined without further excavation) whether the pipe was corroded or, indeed, subject to some other kind of defect.
Nothing about the means HWC adopted in those circumstances was particularly directed to “minimising” the risk of substantial interference with the respondents’ land. To the contrary, those means created a real and not remote risk of catastrophic interference.
As to whether HWC has shown a sufficient reason for adopting those means to justify exposing the respondents to that risk:
HWC does not point to any particular advantage to it in excavating the pipe and then leaving it uncovered overnight. Doing so no doubt saved HWC the costs of having to re-cover the water main and then re-excavate it the following day. However, there was no evidence of the quantum of the cost to HWC of adopting such measures in this case, still less of the cost of doing so in like cases, nor of how many like cases there might be.
Nor has HWC established that there was any particular advantage to it in not reducing the water pressure in DN 500 (and/or making use of the available redundancy in DN 600 and DN 1200). HWC’s evidence did suggest that there might be real disadvantages to it in shutting down the DN 500 water main altogether. But in light of the alternative available course of reducing the water pressure, it is unnecessary to decide whether any disadvantage to HWC in shutting DN 500 off would have provided a sufficient reason for HWC to expose the respondents to the risk of catastrophic flooding by allowing the water pressure to rise in the usual way.
In any event, in order to discharge its onus at the third stage of the Hunt Leather analysis, HWC must establish that both uses of land (the Council’s land to repair the water main and its own land to carry water) were conveniently done. It is sufficient for present purposes to focus on HWC’s use of the Council’s land to repair the water main, regardless of whether its use of its own land was conveniently done. That is, it is not necessary to determine whether HWC’s use of its own land (which permitted the water pressure to rise in the early hours of the morning) reasonably minimised the risk of interference to the respondents’ land so as to be conveniently done. It is enough to determine whether the means by which HWC used the Council’s land (which involved leaving the excavated pipe uncovered overnight in circumstances where the water pressure was known to rise, and did rise, in the early hours) was conveniently done.
Here, “consideration of the relative interests of each of the parties in all the circumstances” can result in only one conclusion: that HWC’s use of the Council’s land to repair the water main was not conveniently done. All three experts agreed that leaving the pipe uncovered overnight as pressure built in it was a cause of the pipe’s catastrophic failure. The pipe had, after all, apparently survived many years of pressure building in it in the early hours of the morning without bursting while supported by covering earth. Having regard to the relative interests of the parties, HWC has not shown, on the limited evidence before the Court, that any costs saved justified the risk of the damage the respondents would suffer if their land was flooded following a rupture of the water main.
it makes no difference whether the mechanism of the water main’s ultimate failure involved corrosion (as had been the case in 2010), a latent manufacturing defect (as Prof Melchers opined), deterioration with the passage of time (as Dr Clegg opined) or some other contributing cause.
Contrary to HWC’s complaint, those conclusions do not depend on hindsight reasoning. Instead, they flow from a recognition that, in light of the information available to HWC in 2017, HWC’s conduct in excavating the pipe and leaving it uncovered overnight created the real risk of a catastrophic failure in DN 500, which risk was not shown to have been reasonably minimised.
At common law, the elements of the respondents’ cause of action in private nuisance were thus made out, and HWC’s attempt to establish a “lawful ground of justification or excuse” failed.
HWC contends that the primary judge erred in failing to find that s 43A of the CLA applied to the respondents’ claim in private nuisance, such that HWC was not liable: Ground 11. By Ground 1 of their notice of contention, the respondents contend that s 43A was inapplicable because HWC’s “acts and omissions … were not based on the exercise of a special statutory power” (emphasis in original).
Section 43A provides:
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.
40 Application of Part
(1) This Part applies to civil liability in tort.
(2) This Part extends to any such liability even if the damages are sought in an action for breach of contract or any other action.
(3) This Part does not apply to civil liability that is excluded from the operation of this Part by section 3B.
On its face, s 40(1) would appear to make s 43A applicable to claims in private nuisance (i.e., a “tort”), to the extent that the terms of s 43A itself permit.
However, as Gordon and Edelman JJ pointed out in Hunt Leather at [142] (Beech-Jones J agreeing generally as to s 43A at [300]):
“There may be doubt about whether s 43A applies to claims for private nuisance. … [T]he extrinsic materials suggest that s 43A was enacted in response to a claim for negligence in relation to the provision of health services, and the focus of s 43A is on an exercise of, or failure to exercise, a particular statutory power rather than the relative uses of land which are the focus of the law of private nuisance.
(Citations omitted.)
It is not necessary in the present appeal to answer the general question whether s 43A can ever apply to claims in private nuisance.
It does not appear to have been disputed that HWC is a “public or other authority” as defined in s 41 of the CLA as including, relevantly, “any public or local authority constituted by or under an Act” (a matter on which, in light of the conclusion reached below, it not necessary to express a view).
It is convenient to deal with one issue at the outset. The primary judge proceeded on the basis of a concession by HWC to the effect that the decision of this Court in Transport for NSW v Hunt Leather Pty Ltd (2024) 115 NSWLR 489; [2024] NSWCA 227 (Hunt Leather CA) is authority for the proposition that, as a matter of principle, s 43A can never apply to claims in private nuisance. His Honour thus said at J[10] that HWC’s reliance on s 43A “does not extend to the nuisance claim.”
The Court of Appeal’s decision did not go so far as the concession suggests. The relevant reasoning in Hunt Leather CA appears at [175]-[181]. Although the Court identified some of the difficulties that may be involved when applying s 43A in the context of a claim in private nuisance (at [176]-[179]), the ratio of the case was not that s 43A can never apply to that tort. Instead, the Court of Appeal found that in the circumstances of the case Transport for NSW (TfNSW) was not relevantly exercising a special statutory power and, consequently, that the section was not engaged (at [180]). Moreover, because liability in nuisance had not been made out, the grounds did not arise, and it was “neither necessary nor desirable to address the operation of s 43A comprehensively” (at [174], [181]). It was therefore not necessary for the Court to decide any broader question of principle.
The High Court in Hunt Leather did not resolve the question whether s 43A can ever apply to claims in private nuisance. Instead, all five judges in that case held that TfNSW’s liability in nuisance was not “based on” the exercise of a special statutory power and, therefore, s 43A was not engaged (see Hunt Leather at [56] (Gageler CJ), [146] (Gordon and Edelman JJ), [256] (Jagot J), [300] (Beech-Jones J)).
TfNSW had submitted that the plaintiffs’ claims against it in nuisance were based on two “elements”. Those elements, as summarised in the judgment of Gordon and Edelman JJ at [145], were said to be “(i) the relevant interference with the interest of Hunt Leather and Ancio Investments in their land; and (ii) TfNSW’s responsibility for that interference because it developed, planned, procured and organised the works that caused the interference.” (Emphasis supplied.) TfNSW contended that the second of those elements was the exercise of a “special statutory power”.
TfNSW specifically relied on s 104O of the Transport Administration Act 1988 (NSW) which provides that it may develop and operate light rail systems (or facilitate their development or operation by other persons). TfNSW also invoked the “general powers conferred” by s 3E of that Act, read together with cll 3 and 9(1) of Sch 1. Those provisions conferred TfNSW’s powers to “develop, establish, hold, manage and maintain transport infrastructure on behalf of the State”, to “hold, manage, maintain and establish assets associated with transport infrastructure developed or proposed to be developed by TfNSW”, to “make and enter into leases or licences, or other arrangements, with persons for developing transport infrastructure” and to “make or enter into contracts or arrangements with any person in connection with the exercise of TfNSW’s functions”.
Justices Gordon and Edelman (Beech-Jones J agreeing at [300]) held at [146] that the relevant acts upon which TfNSW’s liability was “based” were not the acts of TfNSW in procuring the construction of Sydney’s light rail system, but the “acts of doing the construction work”. As their Honours explained, “[t]he liability in private nuisance was based on the physical acts not on the reason (i.e., procurement) that those acts were attributed to TfNSW.”
Gageler CJ and Jagot J each implicitly accepted (at least for the purposes of determining the s 43A issue) TfNSW’s submission that the relevant acts for the purposes of the plaintiffs’ claim in nuisance were those by which it procured the development of the light rail system. However, Gageler CJ held at [56] that the power granted by s 104O to “develop light rail systems, or to facilitate their development by other persons” was not engaged because at the time, there were no regulations bringing the light rail in Sydney within the statutory definition of a “light rail system”. His Honour further held that TfNSW’s power to contract and to plan and develop transport infrastructure did not require any specific authority. Justice Jagot similarly resolved the issue on the basis that “[p]ersons generally have capacity to apply for approvals to construct any development … and persons generally have the capacity to agree a design and construction contract” (at [256]. TfNSW’s acts thus fell outside the definition of “special statutory power” in s 43A.
“In any event, it was properly common ground in these appeals that an assessment of whether liability is based on a breach of statutory duty (s 43) or based on the ‘exercise of, or failure to exercise, a special statutory power’ (s 43A) requires characterisation of the plaintiff's claim.”
(Citations omitted.)
That the focus of the section is the “plaintiff's claim” follows from the language of the provision. Section 43A refers to “proceedings for civil liability”; that is, proceedings in which such a liability is claimed. That claimed liability is the “liability” to which the words “to the extent that the liability is based on …” refer later in the same sentence.
The starting point for analysing HWC’s statutory “defence” is thus the identification of the matters on which HWC’s liability is “based”. Little guidance is given as to the appropriate level of generality or specificity to be applied when identifying the matters on which the liability of a public authority is “based” for the purposes of s 43A. However, as was pointed out in Hunt Leather CA at [177], quoting Gales Holdings Pty Ltd v Tweed Shire Council (2013) 85 NSWLR 514; [2013] NSWCA 382 at [196] (Emmett JA), there is a distinction between liability that is “based on” the exercise of, or a failure to exercise, a special statutory power, and an act or omission merely “involving” the exercise or failure to exercise such a power.
That distinction may be significant in private nuisance, where the essence of the alleged liability is the defendant’s conduct in using land, rather than the exercise or non-exercise of any statutory power when doing so. In light of the incidence of the burden of proof as explained in Hunt Leather, the cause of action in private nuisance is complete when the plaintiff establishes that the defendant’s use of land has “cause[d] a substantial interference with [the] plaintiff’s enjoyment of land for ordinary purposes”: at [97]. The test of causation in private nuisance is the general test of causation in tort at common law, that is, it is a matter of common sense: Fennell at 492 (Glass JA), 493 (Samuels JA); March v E & MH Stramare Pty Ltd (1991) 171 CLR 506 at 515-6; [1991] HCA 12. The reference to common sense should not obscure that two overlapping questions are involved: “a question of historical fact as to how particular harm occurred; and a normative question as to whether legal responsibility for that particular harm occurring in that way should be attributed to a particular person”: Wallace v Kam (2013) 250 CLR 375; [2013] HCA 19 at [11] (French CJ, Crennan, Kiefel, Gageler and Keane JJ); see Fisher v Nonconformist Pty Ltd (2024) 114 NSWLR 1; [2024] NSWCA 32 at [105]-[114]; Owners of Strata Plan No 2661 v Selkirk [2024] NSWSC 760; (2024) 21 BPR 44757 at [130]-[137] (Leeming JA). Once those matters are established, a plaintiff is not required to disprove that the defendant’s use of land was for a “common and ordinary purpose”, or to prove that it was not “conveniently done”, in order to complete the cause of action.
It is true that in this case HWC failed in its attempt to establish that it had a “lawful ground of justification or excuse” for its admitted substantial interference with the respondents’ enjoyment of their land. In particular, HWC failed to discharge its onus of proving at the third stage of the analysis that its use of the Council’s land was “conveniently done”. However, it cannot sensibly be said that HWC’s liability was “based on” any facts which HWC failed to establish, or on any established facts which meant that HWC’s use of land was not “conveniently done”. The respondents’ cause of action was already complete at the first stage of the analysis.
As explained above, it was common ground that DN 500 had been laid within the road reserve to the east of Aries Way. Title to the land was vested in fee simple in the Council.
Since the pipe was buried within the road reserve, HWC’s employees were free to walk on the land without any special authority (such as that conferred by s 20(a) of the Hunter Water Act to enter onto land to “operate, repair, replace, maintain, remove, extend, expand, connect, disconnect, improve or do any other things that [HWC] considers are necessary or appropriate to any of its works …”.). But HWC’s employees excavated through some 5 metres of the road reserve to reach the water mains. In the apparent absence of permission from the Council (as owner of the road reserve), it was not suggested that HWC had any right under the general law to excavate the Council’s land. HWC needed statutory authorisation to dig up the road reserve.
That statutory authorisation appears to have derived from, at least, s 23 of the Hunter Water Act, which relevantly provides as follows:
23 Power to open roads
(1) The Corporation may, for the purpose of complying with the operating licence and after giving reasonable notice to persons likely to be affected, open and break up:
(a) the soil and pavement of a public road or public reserve, and
(b) any sewer, drain or tunnel in or under or any ground under a public road or public reserve.
There was no suggestion that HWC had not given reasonable notice in accordance with that section.
It is also relevant that HWC was conducting the excavation for the purpose of investigating the reported leak and, if necessary, repairing the pipe. As owner of the pipe, HWC had the rights of an owner to deal with its property, including to make repairs to the pipe itself (assuming HWC had physical access to the pipe). HWC was also given statutory authority to undertake repairs by s 19(2) of the Hunter Water Act, which provides as follows:
19 Ownership of works
…
(2) The Corporation may, subject to this Act, operate, repair, replace, maintain, remove, extend, expand, connect, disconnect, improve or do any other things that are necessary or appropriate to any of its works to ensure that, in the opinion of the Corporation, the works are used in an efficient manner for the purposes for which they were installed.
“Works” is defined in s 3(2) to mean “water mains, sewer mains, drainage channels and any works ancillary to those works.” Plainly that included DN 500. The conferral of power in s 19(2) must be read in light of the nature of water infrastructure, which is itself recognised in the words of s 19(1): “[HWC] is the owner of all works installed in or on land by [HWC] …”. However, the damage here was not caused by anything HWC did by way of repairs to its works on its land. This was not a case where in the course of excavation of repair work, a machine ruptured a water pipe and caused loss and damage. Instead, the damage here was relevantly caused by HWC’s excavation of the Council’s land. The s 19(2) power may therefore be put to one side.
It may be accepted that (at least) HWC’s power in s 23(1) to open and break up the Council’s reserve without the consent of the Council was “of a kind that persons generally are not authorised to exercise without specific statutory authority” so as to meet the definition of a “special statutory power” in s 43A(2). HWC was apparently exercising that power when it excavated the DN 500 water main.
But that is not the issue. The issue is whether HWC’s “liability [was] based on … [the] exercise of, or failure to exercise, a special statutory power”. The important point for purposes of the s 43A analysis is that HWC’s liability in nuisance is “based on” its conduct: relevantly, leaving the pipe uncovered overnight while the water pressure increased. No part of what the respondents were required to prove in order to establish that liability was that HWC was exercising (or failing to exercise) any statutory power, or that HWC was exercising such a power (or failing to do so) in any particular way.
In Gales Holdings, the gravamen of the plaintiff’s complaint was that the Tweed Shire Council had caused or allowed stormwater runoff to flow onto its land and prevented and obstructed stormwater runoff from passing and flowing away from its land. Emmett JA rejected the Council’s reliance on s 43A at [197]:
“Gales’ claims are not based on any allegation that the Council failed to exercise a particular statutory power or that it exercised a particular power negligently or unreasonably. Rather, Gales sues in nuisance, not for breach of statutory duty, in relation to the continuing operation of the Council’s stormwater drainage network. It is a misconception to suggest that Gales’ claim is based on the Council's exercise of any particular statutory power. Accordingly, s 43A has no application to the proceedings.”
The position is a fortiori to the extent that HWC’s liability was alleged to arise out of its use of its own land (i.e., the separate strata of land containing its water infrastructure). The relevant conduct on which liability is “based” here is carrying water through the DN 500 water main.
HWC did not need a statutory power to engage in that conduct. HWC owned the water mains. The ability to run water through DN 500 was an ordinary incident of ownership. The ability to shut off or divert the water or to reduce the water pressure was equally an incident of ownership.
Moreover, HWC did not identify any special statutory power necessary in order for it to run water through DN 500. The furthest the argument went was by reference to general provisions in the Hunter Water Act addressing HWC’s functions and HWC’s operating licence. As will be seen, none of those provisions amounted to a statutory power, either to run water through the water mains, or to shut off or to divert the water or to reduce the pressure.
HWC pointed to ss 4A, 12 and 13. Section 4A addresses the Corporation’s principal functions by reference to s 12. Section 12 provides:
12 Grant of operating licence
(1) The Governor may grant an operating licence to [HWC] to enable [HWC] in accordance with this Act, in the area of operations specified in the licence, to provide, construct, operate, manage and maintain systems and services for—
(a) supplying water, and
(b) providing sewerage and drainage services, and
(c) disposing of waste water.
(2) Except to the extent to which this Act expressly provides, nothing in an operating licence limits the requirements imposed by or under any other Act or law with respect to the provision, construction, operation, management or maintenance of any system or service referred to in subsection (1).
13 Terms and conditions of operating licence
(1) The operating licence is subject to the terms and conditions determined by the Governor but must include terms or conditions under which the Corporation is required—
(a) to provide, construct, operate, manage and maintain efficient, co-ordinated and commercially viable systems and services for supplying water, providing sewerage services and disposing of waste water, and
(b) to provide, operate, manage and maintain a drainage service within the capacity of the drainage service included in the business undertaking transferred under Part 3 by the Hunter Water Board to the Corporation as at the date of the transfer of the business undertaking, and
(c) to ensure that the systems and services meet the quality and performance standards specified in the operating licence in relation to water quality, service interruptions, price levels and other matters determined by the Governor and set out in the operating licence.
(2) The operating licence must also include terms or conditions that require the Corporation to maintain procedures under which the Corporation is to consult with its customers at regular intervals in relation to the provision of the systems and services referred to in this section.
Sections 12 and 13 are concerned with licence conditions, not statutory powers. Neither section confers a statutory power to run water through the water mains. Nor do those sections confer a statutory power to shut off or to divert the water from a water main or to reduce the water pressure. That is as expected, given that HWC’s ability to run water through DN 500 was, as noted above, an incident of its ownership of the water main. To the extent that s 13(1)(a) creates a licence requirement, it is not a requirement to supply water. It is instead a requirement, “subject to the terms and conditions” of the licence, to operate and maintain efficient, co-ordinated and commercially viable systems and services for supplying water.
The specific conditions under HWC’s Operating Licence 2017-2022 were referred to above. They included performance standards with respect to rates of water pressure failure and unplanned water interruption for each financial year. That HWC had performance standards to meet would be, at most, a constraint on HWC’s right, as the owner of the water mains, to shut off or divert the water or to reduce the water pressure. Such a constraint is the opposite of a conferral of power. Section 43A is concerned only with power.
It follows that there was no “special statutory power” engaged in HWC’s use of its land to run water through the DN 500 main.
But even if it were assumed that HWC had a special statutory power to run water through DN 500, it would not follow that HWC’s liability was “based on” the exercise of that power so as to engage s 43A. To the extent that HWC’s alleged liability arose out of its use of its own land to run water through DN 500, the liability was “based on” alleged conduct. Again, as in Gales Holdings, the respondents’ claims were “not based on any allegation that [HWC] failed to exercise a particular statutory power or that it exercised a particular power negligently or unreasonably” (at [197]). Again, HWC’s liability in nuisance would be no different, whether or not it had a statutory power to run water through DN 500.
It follows that s 43A has no application to the claim in private nuisance in this proceeding.
Since the appeal with respect to the cause of action in private nuisance fails, it is unnecessary to address Ground 1 or Ground 2 of the notice of cross-appeal. Those grounds challenged the primary judge’s findings with respect to the engagement and operation of s 43A in connection with the respondents’ claim in negligence.
It is also common ground that since the appeal with respect to the nuisance claim fails, it is unnecessary to address the respondents’ claim for compensation under s 22 of the Hunter Water Act, which is the subject of Ground 3 in the notice of cross-appeal*.* It would be inappropriate to address the questions of statutory interpretation raised by s 22 of the Hunter Water Act in light of the focus of the parties’ arguments on other issues on the appeal.
In those circumstances, the cross-appeal should be dismissed with no separate order as to costs.
What remains is Ground 12 in the amended notice of appeal: “The trial judge erred in making the order for costs on 19 November 2025.”
This ground was advanced as an independent error, regardless of the failure of HWC’s other grounds of appeal. HWC assumed that in those circumstances leave to appeal was required, apparently on the basis of Supreme Court Act 1970 (NSW) s 101(2)(c). That assumption was erroneous: see Housman v Camuglia (2021) 104 NSWLR 615; [2021] NSWCA 106 at [43]-[84] (Leeming JA, Bell P and White JA agreeing). Leave is not required, given that the appeal raised other bona fide grounds not requiring a grant of leave as well as challenging an order as to costs.
The primary judge gave a separate judgment on the issue of costs: The Owners - Strata Plan No 16460 v Hunter Water Corporation (No 2) [2025] NSWSC 1361 (PJC). HWC had submitted at first instance that the respondents should pay HWC’s costs in respect of the claim under s 22 of the Hunter Water Act, and that it should pay only one third of the remainder of the respondents’ costs, in circumstances where “the plaintiffs [lost] one of the causes of action they had pursued”: PJC [4]. The primary judge observed that the respondents’ damages claim effectively had two parts: a claim for physical damage and a claim for loss of value (referred to as the stigma claim): PJC [12]-[13]. The former had succeeded; the latter had failed.
The primary judge accepted HWC’s submission concerning the s 22 claim, and apportioned “a half day of hearing” to that issue: PJC [6], [9], [11]. But his Honour rejected HWC’s submission seeking apportionment of the balance of the costs. His Honour considered that it was not appropriate to separate the costs of the physical damage and stigma components of the claim in circumstances where: (unlike the s 22 claim, which was introduced by amendment at the hearing: PJC [7]) the physical damage and stigma components had formed the original basis for the damages sought by the respondents; the whole of the respondents’ claim had been disputed by HWC (that is, as to both nuisance and negligence); and HWC had not disputed the quantum of the physical damages claim but failed to make any pre-hearing offer to settle that claim: PJC [12]-[15]. The respondents having succeeded in the proceedings, costs followed the event: PJC [15].
It is common ground that HWC must demonstrate error in the costs order in accordance with House v The King (1936) 55 CLR 499. HWC does not point to any specific error, but argues that the discretion “manifestly miscarried” and the decision was “clearly wrong”. HWC submits that the s 22 claim was without a proper foundation, the negligence claim was bound to fail by reason of s 43A of the CLA, and the claim for diminution in the value of the land involved expert evidence, and was misconceived.
1. The Trial Judge erred in finding that the Appellant had committed a private nuisance because the cause of the escape of the water from the DN500 mains pipe on 21 September 2017 was not caused by any unreasonable use by the Appellant of the mains pipe but, rather, was caused by the materialisation of an inherent manufacturing defect in the pipe which was neither known to or reasonably foreseeable by the Appellant.
2. The Trial Judge erred in finding that it was reasonably foreseeable that there was a defect in the pipe that required shutting off or reducing the water pressure in the DN500 from 21 to 22 September 2017 in circumstances where the pipe had been excavated and inspected and there was no detectable defect that warranted such action.
3. The Trial Judge erred in failing to make a finding as to the nature of the defect which caused the DN500 to catastrophically fail and should have found that it was caused by the materialisation of an inherent manufacturing defect in the pipe which was neither known to or reasonably foreseeable by the Appellant.
4. The Trial Judge erred in failing to find that the mode of failure was through high tensile stresses as a direct result of localized bending and localized attempted bulging of the pipe at a point where, through manufacturing defect, unknown to and not reasonably foreseeable to the Appellant, the nominal wall thickness was 15mm rather than18mm.
5. The Trial Judge erred in finding that the Appellant acted unreasonably in excavating the site of the reported leak and leaving the excavation exposed overnight as it could not be completed.
6. The Trial Judge erred in finding that reducing the pressure of water in the DN500 on 21 September 2017 would have prevented the burst in circumstances when there was no evidence of by how much the water pressure would need to be reduced to have prevented the failure.
7. The Trial Judge erred in finding that the failure to restrict the pressure of water in the DN500 on 21 September 2017 was relevantly causative of a nuisance or damage, in circumstances where there was no evidence how this would have impacted the amount of water that escaped the pipe following failure.
8. The Trial Judge erred in finding that neither Mr Hayter nor Mr Berry turned their minds to the possible consequences of what they found on excavating the DN500 in circumstances where Mr Berry gave evidence that he did not consider that the weep in the lead caulking to constitute a circumstance which could give rise to the risk of failure of the pipe.
9. The trial Judge erred in failing to give any weight, or alternatively sufficient weight, to the considerable body of evidence from the Appellant as to what a weep from the caulking at the collar indicated in terms of the reasonable foreseeability of a catastrophic failure of the pipe which was wholly unrelated to the weep observed, which was a regular and common occurrence which did not necessitate any of the actions found by the Trial Judge.
10. The trial Judge erred in accepting the evidence of Mr Fennell on the issue of turning off or reducing the flow of water in the DN500 main pipe when that evidence was given only with the benefit of hindsight and did not consider at all the cost, expense and inconvenience to the Appellant in doing so and was contrary to the evidence of Mr Main and Mr Wynn adduced by the Appellant, which established that shutting off or reducing flow in its pipes carried with it interruption to supply and costs and expense that was not warranted.
11. The Trial Judge erred in failing to find that section 43A of the Civil Liability Act, 2002 (NSW) applied so that the Appellant was not liable for private nuisance.
12. The Trial Judge erred in making the order for costs on 19 November 2025
The substance of each ground is dealt with above.
Grounds 1, 3, and 4 are dealt with particularly at [79]-[97].
Ground 2 is misconceived insofar as reasonable foreseeability is not an element of private nuisance. In any event, for the reasons stated at [106]-[109] there was ample basis to find that there was a real risk of a defect of some kind and in turn a failure with catastrophic consequences which was reasonably foreseeable.
Grounds 5 and 9 are dealt with at [98]-[111]. Ground 5, like a number of other grounds, inverts the onus of proof. It was a matter for HWC to prove that its use of the land was “conveniently done”, which, for the reasons above, HWC failed to do. Ground 9, as explained above, involves the fallacy that because previously HWC had not taken precautions, the precautions were not reasonable, or the risk was not reasonably foreseeable.
Grounds 6 and 7 invert the onus of proof. The substance of the issues they raise is dealt with at [112]-[122]. To the extent that these grounds assert that there was “no evidence” for certain propositions, HWC bore the onus of proof on those issues.
Ground 8 is dealt with at [101].
Ground 10 inverts the onus of proof and is dealt with at [112]-[122]. For the reasons set out above, HWC failed to establish that turning off or reducing the flow of water were not available options, or that they would have occasioned great difficulty or cost.
Ground 11 is dealt with at [135]-[181].
Ground 12 is dealt with at [186]-[194].
The reasoning above also disposes of Grounds 1 and 2 in the notice of contention which concerned, respectively, s 43A and the issues of whether HWC’s use of the land(s) was “common and ordinary” and “conveniently done”.
The orders I propose are:
Appeal dismissed.
Cross-appeal dismissed.
The appellant is to pay the respondents’ costs.
**********
Hunter Water Corporation
The Owners of Strata Plan 16460
[2026] QCA 67
(1988) 165 CLR 411
(1997) 143 CLR 549
(1988) 165 CLR 360
(2025) 117 NSWLR 209
(1977) 16 ALR 479
(1982) 149 CLR 480
(2024) 114 NSWLR 1
(2013) 85 NSWLR 514
(1962) 108 CLR 12
(1936) 55 CLR 499
(2021) 104 NSWLR 615
(2025) 100 ALJR 89
(1991) 171 CLR 506
(1989) 18 NSWLR 119
(1963) 108 CLR 491
(2024) 115 NSWLR 489
(2023) 279 CLR 333
(2013) 250 CLR 375
(1979) 143 CLR 549
(2017) 261 CLR 328
Appeal dismissed. Cross-appeal dismissed. The appellant is to pay the respondents’ costs.
No special statutory power was involved in HWC failing to reduce the flow of water through DN 500 on that evening, which was at the forefront of the pleaded allegations of breach. It is true that the conditions of HWC’s operating licence, upon which it placed reliance, ultimately had a statutory source. But the conditions were not the source of the power to control the water in its water mains. To the contrary, those conditions were premised upon and qualified the power to control water in its water mains.
For completeness I should mention the excavation which preceded the failure of DN 500. The excavation took place on the road reserve, and no statutory power was required in order for HWC to enter the land. The pipe was owned by HWC. HWC had a statutory entitlement to “open and break up” the pavement and soil of the road reserve. However, this is a not a case where, say, the negligent operation of excavation machinery ruptured a water main. The excavation of the road reserve preceded the failure of DN 500, and no doubt satisfied the undemanding test of but-for causality, but it was not said (nor could it be said) that HWC’s liability was based on the excavation; after all, the excavation was an essential precursor to repairing the water main. Leaving the excavated water main exposed overnight was connected with the failure of DN 500, but it was not alleged that this was a breach of duty, for, once again, the pipe needed to be exposed in order for it to be repaired. Even if such a breach had been pleaded, I do not consider that HWC’s liability would have been based on the exercise of a special statutory power. Instead, the claim against HWC was based on the failure to take precautionary measures in the years prior to 22 September 2017, and in failing to reduce pressure in DN 500.
I agree with the rejection of the challenge to the primary findings of fact analysed by McHugh JA under the heading “Third issue: ‘conveniently done’”, for the reasons given by his Honour. A cause of the failure of DN 500 was leaving the water main exposed overnight while the water pressure rose, and it matters not whether also contributing to that cause were any of the theories espoused by the experts (corrosion, manufacturing defect or deterioration with age). I also agree with what McHugh JA has said concerning the rejection of ground 12, which separately challenged the exercise of the discretion as to costs. It follows that the appeal must be dismissed. It seems probable that the inapplicability of s 43A carries with it the consequence that the judgment against HWC may also be sustained by reason of negligence. (This was not argued in terms but arguably fell within ground 2 of the cross-appeal if that were treated as a notice of contention.)
Although the foregoing disposes of the appeal, it is desirable to say something about the new issues which arise following the High Court’s decision in Hunt Leather. Similar issues will recur in actions based on private nuisance, and it is far from clear that the parties to this appeal fully engaged with those difficulties.
The parties agreed in their written submissions that a private nuisance is established where (1) the defendant’s use of land has caused a substantial interference with the plaintiff’s ordinary use of land, and (2) “the defendant’s use of land was either not a common or ordinary use; or, if it was”, then (3) it was not “conveniently done”, that is to say, undertaken with a proper consideration of the plaintiff’s interests and to minimise the substantial interference with the plaintiff’s enjoyment of or physical damage to the land. The parties agreed that the test was said to derive from what was said in Hunt Leather at [97] and [260] and by Lord Leggatt in Fearn v Board of Trustees of the Tate Gallery [2024] AC 1; [2023] UKSC 4. Very little attention was given in oral submissions to the legal principles; instead the focus was on the challenges to findings of fact. For that reason I would prefer to follow the same course, and do not wish to be understood as expressing views on the content of the meanings of “common and ordinary” and “conveniently done”.
On the legal test which was common ground between the parties for the purposes of this appeal, what has already been said is sufficient to dispose of it. The findings by the primary judge sustain the conclusion that HWC did not discharge its onus of establishing that the repair of HN 500 was “conveniently done” whatever be the precise meaning of that language. But lest the test relied on in this litigation be used unthinkingly in other cases, it is desirable to make the following observations.
First, the parties’ reliance on Fearn is scarcely to the point. One reason for this is the apparent reversion by the Supreme Court of the United Kingdom to the test of a substantial and unreasonable interference in Jalla v Shell International Trading and Shipping Co Ltd [2024] AC 595; [2023] UKSC 16 at [2] (Lord Burrows with the agreement of Lords Reed, Briggs, Kitchin and Sales) and The Manchester Ship Canal Company Ltd v United Utilities Water Ltd (No 2) [2025] AC 761; [2024] UKSC 22 at [6] (Lords Reed and Hodge, with the agreement of Lords Lloyd-Jones, Burrows, Stephens, Lady Rose and Lord Richards). That said, this Court is and continues to be bound by the ratio of the High Court’s decision in Hunt Leather, even if it relied on propositions from which the Supreme Court of the United Kingdom has subsequently departed.
Secondly, Hunt Leather gives rise to somewhat technical issues of precedent. The High Court was unanimous in dismissing the appeal, but for sharply divided reasons. Two members of the High Court (Gageler CJ and Jagot J) maintained that the lower courts had deployed the correct test, but that the Court of Appeal had wrongly interfered with the factual assessment made by the trial judge, and for that reason restored the judgment at first instance. Three members of the High Court (Gordon J, Edelman J and Beech-Jones J) applied a different legal test, on which the plaintiff would succeed by showing a substantial interference with its land unless the defendant thereafter succeeded in establishing that the purpose of its use of land was “common and ordinary”, or that the defendant’s conduct giving rise to the interference was “conveniently done”. Dispositively on the approach their Honours took, it was not necessary for the plaintiff to establish an unreasonable interference with the enjoyment of its land. There was also pointed disagreement as to whether the test applied by the three was a new test for private nuisance (a “retreat from the position established in Gartner and Elston”) or conventional (“the dominant approach for 150 years in England and Australia”); compare [17] and [115].
Plainly there was no common ground on the formulation of legal principle between Gageler CJ and Jagot J on the one hand, and Gordon, Edelman and Beech-Jones JJ on the other. That would not prevent the decision being binding authority for one or more propositions of law, if they were common to the reasons of Gordon, Edelman and Beech-Jones JJ. And it is true that there is a deal of common ground in those reasons.
Common to all three of their Honours’ reasons are the propositions that while the plaintiff must establish a substantial interference with its ordinary enjoyment of land, once that is done, then there are two negative propositions which fall to be addressed. The burdens of proving that (a) the purpose of a use is “common and ordinary” and (b) the use has not been “conveniently done” fall on the defendant, although there remains an evidentiary onus on the plaintiff “to identify the respects in which the defendant’s use of land is not conveniently done” (at [111] and [274]). All three judges hold that there is no obligation for the plaintiff to establish a substantial and unreasonable interference with the enjoyment of its land.
The parties’ agreed formulation of issues for the purposes of this appeal eschewed the nuance of the shifting evidentiary and legal burdens on the second and third issues. That may well have been appropriate after a trial where the parties fully engaged with the mechanism of the rupture and whether it was brought about by a latent defect. But it would not be safe to assume that the three propositions extracted alone will suffice in other litigation.
The parties framed the second element as turning on whether the use was “common and ordinary”. However, the test stated by a majority of the High Court was whether the purpose of the defendant’s use of the land was “common and ordinary”. The difference may matter, not least because on the facts of Hunt Leather itself, there was further disagreement. Gordon and Edelman JJ characterised the purpose of the use as construction of a light rail network, which was not “common and ordinary”, while Beech-Jones J characterised the purpose of the use as construction work, which was “common and ordinary”. How then are litigants and courts to determine what is the content of the second issue? All members of the majority stated that it is necessary to identify the purpose of the use at the “proper level of generality” at [107] or the “appropriate level of abstraction” (at [298]) but it is clear that those words, at least in their application to construction work, may be applied differently.
It is not uncommon for judges to be agreed as to the legal test but to have different views as to its application to the particular facts of the case. Bath v Alston Holdings Pty Ltd (1988) 165 CLR 411; [1988] HCA 27, decided weeks after Cole v Whitfield (1988) 165 CLR 360; [1988] HCA 18, saw the High Court narrowly divided on whether the impugned law was protectionist within the freshly enunciated meaning of s 92 of the Constitution. If the different characterisation of the works giving rise to the claimed nuisance in Hunt Leather were the only divergence between the reasons of Gordon, Edelman and Beech-Jones JJ, the position for lower courts would be clear. The principle as expressed would bind, albeit that its application in facts similar to those which divided Gordon and Edelman JJ and Beech-Jones J might be contestable in any particular case. That is the point helpfully made at [65] of the reasons of Gordon and Edelman JJ:
On this point Gageler CJ and Jagot J are, in my respectful opinion, correct. The question turns on reading the reasons for judgment in a way which distinguishes the formulation of a legal test from its application to the facts of the case, in accordance with Windeyer J’s observations in Teubner v Humble (1963) 108 CLR 491 at 503; [1963] HCA 11 that “Observations made by judges in the course of deciding issues of fact ought not to be treated as laying down rules of law” and “Reports should not be ransacked and sentences apt to the facts of one case extracted from their context and treated as propositions of universal application”. Adopting that approach, the better view is that the test enunciated by Gordon, Edelman and Beech-Jones JJ in Hunt Leather is a change in the law.
One way of seeing this is to identify the propositions of law in earlier High Court decisions. In Gartner v Kidman (1962) 108 CLR 12 at 46; [1962] HCA 27 Windeyer J formulated the “fundamental rule” in terms of reasonableness, while in Elston v Dore (1982) 149 CLR 480 at 488; [1982] HCA 71, Gibbs CJ, Wilson and Brennan JJ likewise formulated the “proper test to apply” in similar terms, as well as endorsing that aspect of Windeyer J’s reasons. The requirement for there to be an interference “beyond what is reasonable in the circumstances” with the plaintiff’s use or enjoyment of the land was stated in Brown v Tasmania (2017) 261 CLR 328; [2017] HCA 43 at [385], consistently with (and citing) the passage in Elston mentioned above.
A second way of determining whether the law has changed is to examine the position stated in leading Australian academic works: see D Rolph et al, Balkin and Davis Law of Torts, 6th ed (2021, LexisNexis) [14.19]-[14.20] under the heading “Unreasonable”; C Sappideen et al (eds), Fleming’s The Law of Torts, 11th ed (2024, Lawbook Co) [19.80] under the heading “unreasonable interference” which commences “to constitute a legal nuisance, the annoyance or discomfort must be substantial and unreasonable”; M Davies et al, Focus*:* Torts, 10th ed (LexisNexis, 2024) [15.4] (“to establish that an inference with their use and enjoyment of property constitutes a nuisance, the plaintiff must show that the interference was unreasonable in all the circumstances”); those references have been collected in an aptly named paper presented by Professors Dietrich and Field, “The ‘New’ Test of Nuisance in Australia” at the University of Queensland arranged by the Australian Centre for Public Law on 6 May 2026 which has assisted me to understand the High Court’s decision: see (2026) 48 Sydney Law Review (forthcoming). The change in test has also been noted by a commentator in the Law Quarterly Review: A Mohseni, “What is a Nuisance? Hunt Leather Pty Ltd v Transport for NSW” (2026) 142 LQR 378 (“cast aside what was once assumed to be the established position – that it is entirely for the plaintiff to prove a ‘substantial’ and ‘unreasonable’ interference with the enjoyment of their land”).
It will be a task for lower courts in applying the new law stated by the High Court in Hunt Leather to determine to what extent particular aspects of earlier authority continue to apply. An example may be seen in the decision of the Queensland Court of Appeal in Enkelmann v Stewart [2026] QCA 67. At a trial which preceded the High Court decision, reliance had been placed on part of the reasons in Gartner v Kidman. However, it was common ground on appeal that what had been said in that case was now “to be understood as an example of the application of the test for nuisance and not as laying down, as it was advanced below, a defence available to the appellants to be able to ‘turn back’ the waters onto the respondents’ land with impunity”: at [43] and see [46].
McHUGH JA: The appellant (HWC or the Corporation) is a corporation established by the Hunter Water Act 1991 (NSW). HWC owns and operates a water main running alongside Aries Way in Elermore Vale, a suburb of Newcastle in New South Wales. The water main is located immediately to the west of, and uphill from, a strata scheme townhouse complex built on sloping land. The respondents are the owners corporation for the strata scheme and owners of lots in the complex.
Early in the morning of 22 September 2017 the water main, known as DN 500, ruptured and discharged a large volume of water which flooded the townhouse complex. The circumstances in which that event occurred will require close attention. It suffices to note the following by way of introduction.
DN 500 was laid in 1953. The part of the water main with which the appeal is concerned is made up of lengths of 500mm diameter cast iron cement lined (CICL) pipe. In March 2010, DN 500 had burst at a location about 90 metres from the location of the 2017 failure. The cause of that failure had been excessive corrosion in that section of the pipe.
The 2017 rupture occurred in a 3.7 metre length of pipe which had been laid within the road reserve on which Aries Way was later constructed. That part of DN 500, together with two other water mains owned by HWC (DN 600 and DN 1200), was buried some 5 metres below a nature reserve to the east of Aries Way. DN 600 and DN 1200 run parallel to DN 500, spaced around 1.5 metres apart, and while DN 600 and DN 1200 were in “excellent condition”, DN 500 was not.
From early August 2017, employees of HWC had been investigating a possible water leak at Aries Way. Eventually, on 21 September 2017, HWC’s employees excavated down to the water main and observed a small leak on the lead collar of one of the lengths of pipe forming DN 500. The leak was considered to be a “weep”, which is not uncommon and may readily be repaired by caulking the collar and placing a bell clamp around it.
Pending delivery the following day of a shoring box, a bell clamp and a 23-tonne excavator for further excavation, the partially excavated pipe was left exposed overnight (thereby potentially increasing the stress on the pipe by some 20%). HWC took no steps to reduce the rate of flow within the pipe, knowing that the pressure would increase as the water reservoirs filled in the early hours of the morning. At about 3:27 am on 22 September 2017 DN 500 ruptured and flooded the townhouse complex downhill from the water main.
The respondents sued for damages in nuisance and negligence, as well as claiming (in effect in the alternative) for compensation under s 22 of the Hunter Water Act, in a trial that proceeded over five days before the primary judge, Elkaim AJ: The Owners – Strata Plan No 16460 v Hunter Water Corporation [2025] NSWSC 1029 (J). For reasons explained below, the respondents were successful on the nuisance claim but not on the negligence or s 22 claims. HWC appealed and the respondents cross-appealed and filed a notice of contention.
HWC’s amended notice of appeal contained thirteen grounds of appeal. Many of these were repetitive and one was not pressed. The respondents’ notice of contention and notice of cross-appeal contained a further two and three grounds, respectively. HWC’s written submissions dated 9 March 2026 were, to some extent, organised by reference to its grounds of appeal. However, the respondents’ submissions of 26 March 2026 and HWC’s reply submissions of 15 April 2026 were organised instead by reference to various “issues” under which the grounds in the notice of appeal, notice of cross-appeal and notice of contention were loosely grouped.
The parties did not structure their oral arguments by reference to either the grounds of appeal, cross-appeal and notice of contention, or the “issues” identified in their written submissions. Nor did the parties’ oral arguments particularly reflect the terms in which the grounds were formulated. It is appropriate in the circumstances to organise these reasons in accordance with the issues as they emerged during the hearing.
In summary, the overall conclusions of the primary judge, Elkaim AJ, and the extent to which they are challenged on appeal, are as follows.
As to nuisance, the claim succeeded. The question whether the claim in nuisance was established at common law is in contest on HWC’s amended notice of appeal. The focus of HWC’s submissions in this regard is an argument that his Honour ought to have found that the cause of the rupture was a latent manufacturing defect of which HWC was unaware, and that in those circumstances HWC could have no liability in nuisance. This issue occupied by far the largest part of the appellant’s submissions at the hearing of the appeal.
As to s 43A of the Civil Liability Act 2002 (NSW) (CLA), which requires proof of a heightened standard of unreasonableness for liability “based on a public … authority’s exercise of, or failure to exercise, a special statutory power”, his Honour held that it did not apply to liability in nuisance. The operation of s 43A with respect to liability in nuisance is also in contest on the appeal.
As to negligence, the claim failed, but only because of s 43A. His Honour’s conclusion that the acts and omissions of HWC were based on the exercise of a “special statutory power” within the meaning of s 43A is in contest on the respondents’ notice of cross-appeal. In the alternative, the respondents challenge the primary judge’s conclusion that HWC’s acts and omissions did not breach the standard imposed by s 43A.
His Honour’s reasoning as to negligence is not otherwise in contest on the appeal. His Honour held that his findings on nuisance covered the requirements of s 5B of the CLA with respect to negligence. Those findings included that the “inescapable conclusion is that the defendant acted unreasonably in leaving the pipe exposed overnight”: J[98]. With respect to causation, his Honour held at J[138]:
It follows that, at the second stage of the Hunt Leather test, the purpose of the defendant’s use of the Council’s land should not be identified with such specificity as to defeat the object of the inquiry, namely, to identify whether that use serves an interest that warrants recognition in the balancing exercise.
That is consistent with the Court’s application of the test in Hunt Leather. Gordon and Edelman JJ identified the relevant use in that case as “for the construction of a light rail”: at [132]. Beech-Jones J identified the relevant use at a higher level of generality as construction activity: at [295]. Neither judgment descended in the task of characterisation to the particular manner in which the construction was carried out, which was considered under the third stage.
In this case, the appropriate level of generality at which to identify the purpose of the defendant’s use of land is maintaining and repairing water mains. In light of the respondents’ concession that the use of land for that purpose was common and ordinary in the locality, if it had been necessary to decide this issue, HWC would have succeeded at the second stage of the Hunt Leather analysis.
HWC submitted that the primary judge ought to have preferred the evidence of Prof Melchers over that of Dr Richard Clegg (who was also called by HWC) and Mr Luke Fennell (who was called by the respondents). As will be seen, these experts gave different explanations for the pipe’s failure.
HWC accepted that the rupture was at least partially caused by “the surrounding circumstances which allowed the pipe to fracture at the point where the defect occurred related to the excavation of the area, and leaving the excavation exposed overnight, and there being differences in pressure which gave rise to the sinusoidal wave pattern [in the crack]”: Tcpt, 1 June 2026, 2.32-35. (“Sinusoidal” was used in this context to describe the wave-like or “curvy” shape of the longitudinal crack in the pipe.) But HWC submitted that the excavation and inspection were “relatively routine and expected” (Tcpt, 5.40-1), and that the excavation and leaving the pipe uncovered in and of themselves could not have caused the failure of DN 500 (Tcpt, 3.24). The thrust of the argument was that the latent defect “trumped” the other causes, which HWC submitted merely “created an environment where the defect was able to transform from latent to patent”: Tcpt, 3.14, 3.34.
The effect of the argument was that HWC’s use of the land was “conveniently done” because the true cause of the pipe bursting was the manifestation of a latent manufacturing defect that HWC did not know about and could not have been expected to know about.
As will be seen, the premise of that argument (that there was, in fact, a latent manufacturing defect) was not established on the evidence.
However, even if HWC had succeeded in establishing that premise, the significance of any latent defect cannot be analysed in isolation. The other matters known about DN 500 and the other causes of the failure (which HWC expressly acknowledges) must be considered as part of the inquiry as to whether the means which HWC adopted of using the land “reasonably minimised the extent of the substantial interference” with the respondents’ land. In circumstances where there were multiple causes, two of which HWC was responsible for, that a third cause was a latent defect could not alone negative liability.
HWC also relied on lay evidence for the purpose of establishing that its conduct in relation to DN 500 aligned with how it ordinarily carried out its functions, and that it followed that its use of land was conveniently done. However, as will be seen, that lay evidence established several matters suggesting the works were not, in the circumstances, conveniently done in light of what was known to HWC at the time.
HWC further argued that, to the extent that the flow of water, and the failure to reduce the flow of water, is relied upon as an aspect of the use which was not “conveniently done”, shutting off, diverting the flow or reducing the pressure of water were not options which were available or reasonable. As will be seen, while the evidence does suggest that taking up these options was considered by HWC to be a significant matter, they were available, and that availability, and the absence of evidence demonstrating they would occasion great difficulty or cost, is relevant.
HWC’s argument should not be accepted. HWC failed to discharge its onus in that it failed to establish that the means of its use were conveniently done. Specifically, those means created a real and not remote risk of catastrophic interference with the land of the respondents, and HWC failed to show good reason for adopting the means which were employed.
As to the experts’ qualifications, HWC also sought to distinguish among them on this basis. In particular, both at first instance and on appeal, HWC submitted that the expert called by the respondents, Mr Fennell, did not have the same specialist qualifications as Prof Melchers and Dr Clegg and, as a result, that his evidence should be given “very little weight”. That submission should not be accepted. Although Mr Fennell did not have comparable academic experience to that of Prof Melchers and Dr Clegg, the primary judge rightly accepted that he had “significant experience in operating pipelines”: J[80]. That includes testing, commissioning, and assessing risks of operating pipelines over 25 years, in addition to a degree in mechanical engineering, a science degree in chemistry and membership or registration in relevant professional associations.
In any event, Mr Fennell’s evidence about the nature of any defect in the pipe was, in contrast to the evidence of Prof Melchers, “more circumspect and allowed for more than one possibility”: J[88]. For that reason, Mr Fennell’s evidence was, to some extent, consistent with Prof Melchers’ (but also to some extent with Dr Clegg’s). For example, among the “combination of factors” which he considered caused the pipe to fail, Mr Fennell included both “[t]he fact that the pipe wall was thinner than specified” and his belief that “a crack was probably present prior to the pipe being excavated”.
There being no reason to prefer Prof Melchers’ opinions over those of the other two experts, HWC, which carried the onus on the third stage of the Hunt Leather test and therefore on the question whether a latent defect caused the pipe to rupture, failed to discharge it. The primary judge did not err in failing to find that a latent defect caused the pipe to rupture.
Mr Berry gave evidence that the usual procedure to repair a weep was to use caulk to knock back in the lead collar. A bell clamp would then be fitted around the whole collar to provide a “permanent fix”. In order to do this, it was necessary to excavate a further 500mm to expose the circumference of the collar. This would allow a worker to lie on his back and caulk the lower half of the collar and apply the bell clamp. As the excavator on site on 21 September was unable to reach further depths, a larger 23 tonne excavator, a shoring box and a bell clamp were ordered for the following day, and a team was commissioned to attend to the “weep”.
Importantly, when Mr Berry and Mr Hayter left the site at 2:30pm on 21 September, they left the upper portion of the 0.5 metre length of water main exposed overnight.
Mr Berry said in his affidavit dated 8 August 2025 that he had “left underground water mains uncovered plenty of times” and had “left numbers of excavated watermains uncovered overnight in order to undertake repairs the following day.” With respect to his work on the site in the days leading up to the 2017 incident, Mr Berry said that there was “no indication that there was anything other than a weep on the collar of the pipe”. He also “did not consider it necessary to request the watermain be shut down notwithstanding that [he] was aware that the watermain had failed in 2010” because “[t]he Site of the 2010 failure was a fair distance down the hill from the Site of where this collar was weeping.” (It should be noted on this topic that Ground 8 in the amended notice of appeal misfires. This ground complains that the primary judge erred “in finding that neither Mr Hayter nor Mr Berry turned their minds to the possible consequences of what they found on excavating the DN 500”. But the italicised words are not part of the finding at J[104], which was instead directed to “any possible consequences” of the excavation itself, which was “more extensive than it would have been but for the mistaken identification of DN 600”: J [103].)
Mr Wynn, a Manager, Civil Engineering employed by HWC, gave evidence about HWC’s approach to repairing water mains. In his affidavit dated 23 May 2025, he said that between 2012 to 2017, there had been a total of 470 failures along critical water mains. Mr Wynn resisted the proposition that it would be practical to shut down DN 500, notwithstanding that it was not in dispute that DN 1200 and DN 600 could provide redundancy in the case that DN 500 could not be used. It will be necessary to return to the evidence on this topic in some detail below.
HWC acknowledged at the hearing of the appeal that Mr Hayter’s evidence added nothing significant in this regard.
HWC submits that this lay evidence indicates that what was presented to HWC’s workers when they attended the site was “a minor expected weep leak, which could easily be fixed once excavated and a bell-clamp obtained.”
HWC’s argument that the lay evidence established that in all the circumstances the repairs to DN 500 were conveniently done should not be accepted. To the contrary, the lay evidence raised several matters that lead to the opposite conclusion.
First, HWC and its staff were aware of the catastrophic consequences of the 2010 failure, which had resulted in 4.5 million litres of water being discharged onto generally the same properties as in 2017. The 2010 failure had occurred at a location about 90 metres from the 2017 excavation site. The 2010 failure had been found to be caused by excessive corrosion. As will be seen, there does not appear to have been any basis to exclude similar corrosion as a risk when HWC’s staff left the pipe uncovered at about 2:30pm on 21 September 2017.
From 2008 to 2009, the Materials Engineering group of Hunter Water Australia Pty Ltd (HWA) had conducted a condition assessment of the water mains between Wallsend pumping station and Valentine Reservoir on behalf of HWC. HWA used the linear polarisation resistance technique to conduct this assessment, which involves the assessment of soil samples taken along the pipeline route to predict present and future performance of the pipes. The report from this assessment dated 7 October 2009 indicated that the annual probability of a failure in section 2 of DN 500 (the section in which both the 2010 and the 2017 failures occurred) was around 10% in 2012 and was predicted to rise continually thereafter. The cumulative number of predicted failure(s) in section 2 was one failure by 2012, and around three failures by 2023. Of course, the occurrence of one failure in that section in 2010 would not make further failures any less likely.
In HWC’s Trunkmain Failure Report, dated 23 March 2010 (20 days after the 2010 failure), Mr Matthew Dafter wrote:
Since:
HWC’s use of the Council’s land was, for those reasons, not “conveniently done”, and
that use was a cause of substantial interference with the respondents’ use of their land,
It follows that, for the purposes of s 43A, the matters on which HWC’s liability is “based” are the essential facts alleged to establish the proposition that HWC’s use of land (whether its own land or the Council’s land) caused a substantial interference with the respondents’ enjoyment of their land.
The essential facts alleged (and proved) to establish that proposition were as follows.
HWC used (the C**ouncil’s) land to excavate down to the DN 500 water main and then left it uncovered overnight. To the extent that this involved an omission (leaving the pipe uncovered), the fact that the pipe was uncovered was simply the continuing consequence of HWC’s act (the excavation).
HWC continued to use (its own) land to carry water in the DN 500 water main, such that the pressure within the water main increased in the early hours of the morning in the usual way.
HWC’s conduct in each of (1) and (2) caused the water main’s rupture, which led directly to the inundation of the respondents’ land. (The conclusion that the conduct in each of (1) and (2) caused the pipe to fail is not affected by the fact that the pipe was also subject to a structural defect of some kind.)
The conduct in (1) and (2) encompassed two different uses of two different parcels of land. Those uses are conceptually distinct (i.e., they raise two distinct potential bases of liability) but also factually connected.
As to HWC’s use of the Council’s land, the facts in (1)-(3) establish that HWC’s conduct in (1) caused, as a matter of common sense for the purposes of attributing liability, a substantial interference with the respondents’ enjoyment of their land. That completed the cause of action in nuisance arising out of HWC’s use of the Council’s land. It follows that HWC’s liability in private nuisance for use of that land was “based on” at least HWC’s conduct in (1).
HWC’s liability arising out of its use of the Council’s land was factually connected with the circumstance that the water pressure increased in the early hours of the morning in the usual way. But HWC’s liability arising out of its use of the Council’s land (i.e., the excavation) did not necessarily depend on the fact that it was also HWC itself which had control of the water pressure through its use of its own land. HWC would have had the same liability in private nuisance arising out of its excavation of the Council’s land if the ability to control the pressure in the water main had rested with a third party. That is because it was sufficient to establish HWC’s liability that in circumstances where the water pressure increased overnight, HWC’s conduct in excavating the Council’s land and leaving it uncovered was, as a matter of common sense, a cause of the substantial interference with the respondents’ enjoyment of their land.
As to HWC’s use of its own land, it is perhaps less clear that HWC would have had an independent liability “based on” its conduct in (2) alone. That is because, if HWC had not also engaged in the conduct in (1) on the Council’s land (e.g., if instead the Council had carried out the excavation), it might be argued that HWC’s mere continuing use of its own land to run water through DN 500, as it had for many years, would not, without more, give rise to any liability. Specifically, it might be argued that HWC, in merely continuing to use its own land to run water through DN 500, should not have responsibility attributed to it as a normative matter, applying the overall common sense approach. It might also be argued that HWC’s mere continuing use of its land to run water through DN 500, without more, was “conveniently done” so as to excuse the interference with the respondents’ use of their land.
Nevertheless, in this case, HWC engaged in both the conduct in (1) and the conduct in (2). In circumstances where the fact that the water pressure increased overnight was a necessary condition of the pipe’s rupture, it would be artificial to proceed on the footing that HWC’s liability was not “based on” its conduct in both (1) and (2).
However, in neither case was HWC’s liability “based on [its] exercise of, or failure to exercise, a special statutory power”.
As to the submission based upon s 22, HWC’s submission below was accepted and a special costs order was made. There was no manifest error in his Honour’s approach.
As to the submission that the negligence claim was bound to fail by reason of s 43A, that is by no means clear. In any event, much the same evidence was relevant to the claims in both negligence and private nuisance. There was no manifest error in his Honour’s decision not to depart from the usual rule that costs follow the event.
As to the submission concerning the failure of the stigma claim, HWC’s complaint is in substance that his Honour should have apportioned the costs. The applicable principles were set out in Doppstadt Australia Pty Ltd v Lovick & Son Developments Pty Ltd (No 2) [2014] NSWCA 219 at [18]-[21]. The “question of apportionment is very much a matter of discretion” and the issue or group of issues must generally be clearly dominant or separable: [18]-[19]. The primary judge considered HWC’s submission. HWC does not identify any specific error in his Honour’s reasons for rejecting it, instead complaining that the costs discretion manifestly miscarried. When an appellant makes such a complaint it is important to keep steadily in mind who was the true victor in the litigation. Here it was the respondents. There must be some good reason for departing from the usual rule that costs follow the event. In circumstances where the valuation component of the respondents’ claim was not wholly unrelated to the physical damage component, the primary judge’s decision not to separate the costs, made in the exercise of his discretion, was not manifestly wrong.
Ground 12 is not made out.