A plaintiff sues a construction company in the County Court of Victoria after scaffolding collapses on a residential build. The plaintiff alleges negligence. The injuries are serious. The damages claim is substantial. The defence has a credible argument that the plaintiff's own foreman modified the scaffolding platform against the site supervisor's verbal instructions, and that this intervention broke the chain of responsibility in a way the defendant could not have foreseen. Causation is the live issue.
Section 51 of the Wrongs Act 1958 (Vic) adopts the modified Civil Liability Act framework for breach of duty and causation. The provision mirrors the structure of section 5B of the Civil Liability Act 2002 (NSW): a determination of whether negligence was a necessary condition of the occurrence of the harm, then an assessment of whether it is appropriate for the liability of the defendant to extend to the harm so caused. The language is familiar. The application is not.
The difficulty for defence counsel in negligence matters is the second limb. The scope-of-liability inquiry asks whether the harm fell within the scope of the defendant's negligent act, and Australian courts have interpreted this inquiry with varying degrees of flexibility depending on the factual matrix. The plaintiff's foreman intervening to modify the platform is an intervening act, but whether it severs the causal chain depends on whether the intervention was reasonably foreseeable and whether the defendant's breach remained a necessary condition of the resulting harm. That question is doctrinally settled at the level of principle and fiercely contested at the level of application.
You need the intermediate appellate decisions. The High Court has spoken on the general framework for causation in the Civil Liability Act era, but the granular application of section 5D in New South Wales and section 51 in Victoria to intervening human conduct happens in the Court of Appeal. Those decisions turn on fact patterns, and the analogies to your scaffolding matter are imperfect. A plaintiff modifying safety equipment is different from a plaintiff ignoring a warning, which is different again from a third party's independent act of negligence. The doctrinal thread runs through those distinctions.
Conventional database search handles the statute and the leading authorities well enough. A Boolean query for "scope of liability" and "intervening act" returns the landmark decisions. What it does not give you is structure. You receive a list of citations, ranked by citation frequency or relevance scoring, and you are left to read each decision and map the factual distinctions yourself. Cross-referencing how the New South Wales Court of Appeal applied section 5D to a third-party intervention in one case against how the Victorian Court of Appeal treated a plaintiff's own conduct under section 51 in another is a morning's work, minimum. The risk is that you miss the one intermediate decision that cuts against your client's position because it sat on page three of the results and the headnote did not signal the relevant factual analogy.
The trap in this matter is specific. Defence counsel might read the scope-of-liability inquiry as a straightforward foreseeability test and argue that the foreman's intervention was not foreseeable, therefore the defendant's breach was not the appropriate scope of liability. That argument works in some contexts, but the Court of Appeal in both states has held that the scope-of-liability inquiry is not coextensive with foreseeability. The foreseeability of the intervening act is one factor in the section 5B or section 51 assessment, but the provision asks a broader question about whether it is appropriate for the defendant to be liable for the harm. Several intermediate decisions have held that even unforeseeable intervening conduct does not necessarily sever the causal chain if the defendant's breach created the material risk of the class of harm that eventuated. The trap is reading the causation provision too narrowly and missing the decisions that expand the scope of liability beyond what a pure foreseeability argument would allow.
This is where the research tool matters. Habeas is built on an Australian-first corpus of over 300,000 cases and pieces of legislation, and its semantic Search Engine lets you ask questions in the language a practitioner uses. You type: "scope of liability under section 5D where a third party's unforeseeable intervention contributed to the harm, and whether the defendant's breach remains a necessary condition." The search returns authorities grounded in Australian primary law, with citations that link directly to the source material. You can check each reference. The results are traceable, never fabricated. The distinction between section 5D of the New South Wales Act and section 51 of the Victorian Act is the substance of the argument, and a closed dataset of legitimate Australian legal sources treats those differences as substance, not noise.
The semantic search maps the doctrinal thread. Habeas surfaces Court of Appeal decisions on intervening acts under both section 5D and section 51, side by side, so you can compare how each court treated the foreseeability of the intervention and the relationship between foreseeability and the broader scope-of-liability inquiry. You see the factual analogies that matter for your scaffolding matter without reading six headnotes to find them. Foundational research processes that used to take a full morning can now be completed in minutes, and you spend the returned time reading the authorities, testing your argument against the closest factual analogies, and identifying the decision the plaintiff will rely on to defeat your intervening-act defence.
That structure is what lets you identify the trap before you file your outline of submissions. A conventional search returns raw hits. Habeas returns authorities organised around the doctrinal relationship you asked about. You see the thread from the statute through the intermediate appellate decisions that apply it to fact patterns resembling your matter.
For the scaffolding matter, the research changes the advice. The search surfaces Court of Appeal authority that a defendant's breach in failing to secure equipment remained the appropriate scope of liability even though the immediate cause of the harm was a third party's modification. The principle is that the defendant's breach created the material risk, and the intervening conduct fell within the scope of that risk even if the specific modification was not foreseeable. That authority sits uneasily with a clean intervening-act defence. The advice needs to acknowledge that exposure and frame the defence around the specific nature of the foreman's modification, arguing that it fell outside the class of risk created by the defendant's breach rather than arguing that the intervention was unforeseeable.
That reframing is the difference between a defence that survives the plaintiff's reply and one that collapses at the first interlocutory skirmish. The research found it because the semantic search connected the doctrinal threads across jurisdictions and surfaced the decisions that a keyword search would have buried.
Habeas does not write the submissions. It does not predict the outcome. The judgment about how to frame the defence, which authority to lead with, and how to distinguish the unfavourable Court of Appeal decision belongs to counsel. What Habeas does is compress the research time between receiving the brief and forming the preliminary view, so that the hours you spend are on the analytical work that requires a litigator. Reading the authorities, testing the argument, advising the client.
The scaffolding matter will settle or proceed to trial. Either way, the causation argument is grounded in the right authorities, the trap has been identified, and the research was done in focused sessions rather than mornings of manual trawling. For defence counsel who handle negligence matters across jurisdictions, that compression is the value. The Civil Liability Acts are not getting simpler, and the intermediate appellate decisions are not getting fewer. The research tool that maps the thread from statute to application, grounded in Australian primary law with citations you can verify, is the one that earns its place in the workflow. You can test that search on your next causation brief at habeas.ai.
If you want to try for yourself or get in contact, book a demo with us here. We also offer the capacity for self-serve individuals to sign up, and subscribe or register a free trial at app.habeas.ai.
The legal research in this article was conducted and every citation verified using Habeas, the Australian legal AI research platform.
Hero image: mehmetakifarts on Pexels
