Practice with mock tests, learn from structured notes, and get your questions answered by a global forensic community, all in one place.
Forensic engineers occupy a carefully bounded role: they offer independent technical opinions to the court, not advocacy for the party that hired them. This topic covers the duties, ethics, and practical mechanics of that role.
Last updated:
Most engineers solve problems for clients. A forensic engineer is hired to do the same thing, but then something shifts: the work ends up before a judge, a jury, or an arbitration panel, and the engineer is asked to explain what went wrong to people who have no engineering background and who will make a binding legal decision based partly on what they hear. That shift changes almost everything about how the role works.
The central tension is this: the engineer is paid by one party, but they owe their duty to the court. Courts take this seriously. In England and Wales, the Civil Procedure Rules Part 35 states it plainly, the expert's overriding duty is to the court, not the retaining party. Similar obligations exist under the Federal Rules of Civil Procedure in the United States, under Australian procedural rules, and under the rules of most international arbitral tribunals. A forensic engineer who crosses the line from independent witness to partisan advocate risks having their evidence excluded, and risks professional consequences besides.
This topic works through what that dual position means in practice: how an engagement is structured, what a compliant expert report looks like, how courts have developed procedures like single joint experts and hot-tubbing to manage the expert-evidence problem, and where the conflict-of-interest landmines sit. Get these mechanics right and a forensic engineer's technical opinion reaches the decision-maker intact. Get them wrong and the opinion may never be heard at all.
Being hired to investigate is not the same as being retained to testify.
Many forensic engineering assignments start as consulting work: a manufacturer calls after a product failure, an insurer wants to know the cause of a building collapse, a law firm needs someone to look at whether a structure was built to code. At that stage the engineer is a consultant, and their communications are often protected by attorney-client or legal professional privilege. They can explore bad news, draft speculative theories, and give the client candid assessments of risk, all without those drafts ending up in the hands of the opposing side.
The situation changes when the retaining party decides to designate the engineer as a testifying expert. At that moment, and usually at the moment of disclosure, which procedural rules require by a set deadline, the work product transitions from privileged to discoverable. In US federal litigation, Rule 26 requires disclosure of expert reports and, in many cases, draft reports and communications with counsel. UK CPR Part 35.10 requires the report to state the substance of all instructions, written or oral. The expert is now a witness, not just an adviser.
Once designated as a testifying expert, the engineer cannot take instructions from the retaining party about what conclusion to reach. They can receive factual information, be shown relevant documents, and have inaccuracies in a draft corrected. They cannot be coached toward a conclusion. The distinction is one courts police through cross-examination and, occasionally, through sanction when the line is crossed.
A 1993 UK judgment distilled expert witness obligations into a set of duties that spread around the common-law world.
The case that crystallised expert witness duties in English law was The Ikarian Reefer [1993] 2 Lloyd's Rep 68, a shipping insurance dispute in which Cresswell J criticised the conduct of expert witnesses on both sides. His statement of duties became the reference point for generations of expert witness guidance:
These duties were codified in CPR Part 35 and its Practice Directions, which require the expert to include a declaration that they understand their duty to the court and have complied with it. Similar formulations appear in Australian Uniform Civil Procedure Rules, the Federal Court of Australia's Practice Note CM 7, and in the rules of major international arbitration institutions (ICC, LCIA, ICSID).
Good forensic engineering practice is partly a documentation discipline.
An expert witness engagement should be accepted with care. Before taking instructions, the engineer should confirm their expertise covers the technical issues, check for conflicts of interest, and ensure the scope is clear enough that the engagement can be completed and reported without crossing into advocacy. A written retainer letter or engagement letter that captures the scope, the questions to be answered, and the basis on which instructions can be communicated is good practice in all jurisdictions.
One practical tension at the report-writing stage: the retaining party's lawyers may want the expert to soften a conclusion, emphasise a qualification, or add caveats that make the opinion appear less certain. The expert can and should explain their views to counsel and incorporate factual corrections. They cannot change their genuine technical conclusion because it is inconvenient. An expert who does faces cross-examination on the original draft, which in many jurisdictions will be disclosed, and the credibility damage that follows.
When courts tire of duelling experts, they sometimes remove the choice.
The adversarial model produces expert battles: each side appoints an expert, both experts reach predictably opposite conclusions, and the court spends days hearing argument about which qualified engineer to believe. Courts in many jurisdictions have introduced reforms to reduce this. The most direct is the single joint expert (SJE).
Under UK CPR Part 35.7, the court can direct parties to use a single joint expert rather than each appointing their own. The SJE is jointly instructed, answers questions from all parties, and owes the same duty to the court as any other expert. Neither side can coach them or communicate with them without copying the other. The advantage is cost and neutrality. The disadvantage, from a party's perspective, is the loss of confidential expert advice before the SJE is appointed.
SJEs are common in lower-value personal injury and property cases in England and Wales, but rarer in high-value technical disputes where the parties need their own independent investigations and where the cost of a shared expert is less compelling. International arbitration almost never uses SJEs; the parties' own technical evidence is the norm, and the tribunal decides between them.
Putting the experts in the same room at the same time speeds up the real technical argument.
Concurrent evidence (colloquially called hot-tubbing, a term from Australian practice) is a procedure in which opposing experts give evidence simultaneously. They typically sit together in the witness box or at a table facing the judge. The judge leads the discussion, moving through the issues in the joint statement. Each expert can respond to what the other says, rather than waiting for cross-examination to expose the disagreements.
The procedure originated in the Australian Land and Environment Court and the Federal Court of Australia and is now encouraged by the Federal Court of Australia's practice notes and used in the Technology and Construction Court in England. It has also been used in international arbitration and in some US federal proceedings, though it is not yet mainstream in American courts.
| Feature | Sequential expert evidence | Concurrent evidence (hot-tubbing) |
|---|---|---|
| Order of testimony | Each expert in turn, cross-examined separately | Both experts present at once |
| Technical debate | Through counsel's questions, not direct | Direct between experts, judge-moderated |
| Time to narrow issues | Often slow; real disagreements emerge late | Faster; genuine differences surface quickly |
| Best for | High-stakes cases needing full cross-examination | Cases with defined technical disputes |
| Risk | Expert battles with theatrical questioning | Dominant expert may overshadow the other |
For the forensic engineer, hot-tubbing demands a different preparation: rather than preparing for a lawyer's cross-examination, the expert needs to know the opposing report well enough to engage in real-time technical debate. Being corrected on a technical point in front of the judge is more damaging than during cross-examination, because the context makes the error visible to all.
A conflict discovered after a report is served is far worse than one disclosed before instructions are accepted.
Conflict-of-interest problems for forensic engineers take several forms. A prior engagement for the opposing party, even years earlier, can generate sufficient bias concern to require disclosure or recusal. A financial stake in the outcome (owning shares in a party, or depending on a single law firm for most of one's instructions) creates appearance problems. A close professional relationship with a key witness on one side can imply alignment. None of these is automatically disqualifying, but all require disclosure at the outset.
Scope creep is a different category of problem. A forensic engineer retained to examine a steel bridge connection may find, during the investigation, evidence suggesting that design calculations were also deficient. Commenting on the design calculations may be outside their retainer, or outside their competence if they are a structural materials expert rather than a bridge designer. The right response is to flag the issue to the retaining party and, if it is relevant to the litigation, to ensure a suitably qualified expert is engaged for that aspect. Overreaching expertise in a report invites cross-examination challenges and can taint the admissible conclusions.
Under UK CPR Part 35, to whom does a testifying expert's overriding duty run?
Test yourself on Forensic Engineering with free, timed mocks.
Practice Forensic Engineering questionsSpotted an error in this page? Report a correction or read our editorial standards.