Part 4 – Cross‑Examination and Expert Witness Conferencing: Sequencing, Structure and Tribunal Utility

Introduction

Cross-examination remains the most familiar way of testing expert witness evidence in international arbitration. It is the traditional model: the expert witness takes the stand, opposing counsel asks the questions, and the tribunal listens, intervening to a greater or lesser extent depending on the tribunal and the case. It is well established, well understood and, done properly, very effective.

Expert witness conferencing, often referred to as concurrent expert witness evidence or “hot tubbing”, is different. It involves two or more expert witnesses giving evidence at the same time, usually on the same issues, and usually with a more direct role for the tribunal. It is now a familiar feature of international arbitration and is expressly contemplated by both the IBA Rules^[1]IBA Rules on the Taking of Evidence (2020) and Ciarb Guideline.^[2]Guidelines for Witness Conferencing in International Arbitration

Importantly, Ciarb approaches witness conferencing as a flexible process rather than a fixed formula. Tribunal-led, witness-led and counsel-led models are all possible, and combinations of them are possible too.

Although witness conferencing has its origins in common law jurisdictions, it is now firmly embedded in international arbitration practice. It is encountered regularly across a wide range of seats, legal traditions and industry sectors, and is no longer characterised as a procedural novelty.

That flexibility is one of the strengths of witness conferencing. It is also one of its risks. A flexible procedure can be adapted intelligently to suit the dispute, but it can also be deployed in a way that looks impressive in a procedural order and proves less impressive at the hearing.

The central question is not whether cross-examination is more traditional, or whether conferencing feels more modern, or whether one sounds more sophisticated than the other. The central question is simpler than that: what best assists the tribunal?

That question is consistent with the themes emerging from the recent Ciarb Report.^[3]Ciarb Report on Party Appointed Expert Witnesses in International Arbitration: Effectiveness and Best Practice The Ciarb Report emphasises that expert witness evidence is most effective where tribunals are proactive, where methodology and issues are addressed clearly, and where procedures are structured in a way that serves the case rather than follows a rigid template. It also makes clear that no single procedural model should be treated as universal.

This article advances a practical sequencing model: joint report, followed by witness conferencing, followed by cross‑examination. That sequence, in our experience, most effectively assists tribunals in identifying, understanding and then testing the real areas of technical disagreement.

The authors of this paper^[4]Michael Tonkin is a Partner and Quantum Expert based in HKA’s Dubai office. Michael has acted as a lead expert on 80+ occasions and has been appointed arbitrator on over 70 matters. Michael is … Continue reading are both seasoned quantum expert witnesses, with over 50 years of combined experience in the international construction industry. Between them, they have been appointed as lead quantum experts on more than 95 matters. The views set out herein are grounded in their collective professional experience and a desire to enhance the quality and efficiency of expert witness evidence in complex construction disputes.

Procedure Should Help the Tribunal

The starting point ought to be obvious, but it is still worth stating. Procedure is there to help the tribunal understand and decide the case. It is not there to become a performance in its own right.

The Ciarb Report deliberately avoids prescribing a single procedural model and instead emphasises flexibility, informed discretion and adaptation to the circumstances of the case. The witness conferencing guideline adopts the same approach, recognising that different factors will influence whether a conference is appropriate at all and, if so, what form it should take.

In practice, that means there is no universally correct format. There is only a format that is more or less useful in helping the tribunal understand and test the expert witness evidence before it.

That may sound uncontroversial, but hearing procedures can sometimes gather a surprising amount of ceremony. One occasionally gets the impression that the process is expected to be admired for its own sake, like an antique clock: intricate, polished and faintly intimidating, even if nobody is quite sure whether it is telling the right time.

If the tribunal finishes the hearing clearer on the mechanics of the procedure than on the actual differences between the expert witnesses, something has gone wrong.

Cross-Examination: What It Still Does Very Well

Cross-examination remains a powerful tool. Its purpose is to test the expert witness’s analyses, assumptions, conclusions and reasoning, to narrow the issues between the expert witnesses, and to help the tribunal understand how the expert witness evidence should be used. That is very much in line with the approach discussed in our earlier articles in this series.

It has some obvious strengths. First, it brings focus. Counsel will generally direct their questions to the points their side considers important, which gives the tribunal a clear sense of where each party says the real battleground lies. Second, it provides structure. On complex technical issues, structure is not a vice. Third, it is well suited to probing methodology, assumptions, omissions and inconsistencies in a disciplined way.

Cross-examination is particularly effective where the issues are relatively discrete, where methodological differences are central, or where credibility-related matters – including issues of independence or qualifications – may instead be matters for the tribunal to address outside any concurrent process. The Ciarb Guideline expressly acknowledges that credibility issues — whether relating to independence, qualifications or expert witnesses — may sometimes militate against conferencing or may need to be dealt with outside any concurrent process.

There is, however, a practical limitation to the traditional model. Cross-examination is usually sequential. One expert witness is examined, then the other, often at a different point in the hearing. The tribunal must hold one set of answers in mind while waiting for the competing set later on. That is not impossible, of course. Tribunals are used to that. But it is not always the most efficient way of comparing technical evidence. It is rather like comparing two spreadsheets when one is on the screen and the other is somewhere in a lever arch file on the table. Entirely manageable, but not ideal.

Cross-examination can also, at times, place too much emphasis on the witness rather than the issue. Tribunals are usually trying to decide which analysis is to be preferred, not which expert witness appears most comfortable under fluorescent lighting and mild hostility.

There is also a more practical risk in some cases. The structure of cross-examination can create an opportunity – particularly in less disciplined hands – to focus on the witness rather than the analysis. Time may be spent on tone, presentation or perceived weaknesses of the individual expert witness rather than on the underlying technical issue the tribunal actually needs to decide. Tribunals are generally well alert to that distinction, but the risk remains that the exercise becomes more about advocacy against the person than proper testing of the opinion.

Lastly, the expert witnesses can cause issues. Despite counsel’s precisely targeted questions, ‘skilled’^[5]Or in other words ‘partisan’ expert witnesses may deliberately avoid providing the answer – or worse – convolute matters. Proactive and experienced tribunals may interfere and/or subsequently disregard such expert witness evidence, but if not, such evidence may mislead the tribunal when writing the award distant from cross examination. This risk should not be overstated. Experienced tribunals will usually identify and disregard such evidence. The greater concern arises where less experienced tribunals may give weight to irrelevant or misconceived matters simply because they have been explored at length in cross-examination.

Witness Conferencing: Comparative Utility

Witness conferencing has become increasingly common for good reason. The Ciarb Guideline notes a number of perceived advantages: it allows evidence to be compared side by side, creates the possibility of immediate response and rebuttal, and may improve the efficiency of the hearing. It also notes the rather practical point that expert witnesses may be less inclined to make technically questionable assertions when sitting beside someone who can immediately correct them without needing to wait until tomorrow afternoon.

That is broadly consistent with our experience and in our view the most significant benefit of the witness conferencing is keeping both experts on topic; i.e. it significantly reduces the risk of any expert ‘side-tracking’ from real issues. Done well, witness conferencing can be extremely effective because it forces the real areas of disagreement into the open. The tribunal does not need to reconstruct them from separate transcripts or from memory. It can see them emerge in real time, issue by issue, with both expert witnesses present and able to respond to one another directly.

It also often changes the tone in a helpful way. Expert witnesses tend to be more forthcoming when the tribunal asks questions directly. The exchange becomes less about surviving a sequence of pointed questions and more about assisting the tribunal on the technical matters that actually need to be understood. The process can feel less adversarial without becoming any less rigorous. That is often a good combination.

There is another benefit, which is not always said aloud but is usually there. Expert witnesses sitting beside each other tend to become a little more disciplined. Positions are more difficult to overstate when the competing expert witness is present, listening carefully and fully capable of responding. One does not need to be unduly cynical to think that peer review can be healthy when it takes place live and in public.

That said, witness conferencing should not be oversold. In our experience, it does not usually reduce the amount of work carried out by expert witnesses in any material way. Nor is it a cure for weak preparation. Its benefits are more qualitative than quantitative. It improves the tribunal’s ability to compare, question and understand. In a hearing dealing with technical issues, that is more than enough justification.

Joint Reports Before Conferencing: Essential, Not Optional

In our view, a meaningful joint report is ordinarily essential before witness conferencing. In our view, a meaningful joint report is ordinarily essential before witness conferencing. That is consistent with the thinking reflected in our earlier articles in this series^[6]See Part 1 dealing with effective delivery of expert evidence By ‘ordinarily essential’, we mean that without a joint report (or equivalent schedule): (i) the issues are unlikely to be properly crystallised, (ii) the tribunal must identify areas of disagreement in real time, and (iii) the efficiency and comparative value of any concurrent evidence is materially reduced.

The reason is straightforward. A witness conference only works properly if the issues have already been fully crystallised. The tribunal needs to know what is agreed, what is not agreed, and — critically — why those matters remain disagreed. A proper joint report does that work in advance. It identifies the areas of consensus so that hearing time is not wasted rediscovering them. It identifies the genuine points of disagreement.^[7]And preferably in easy to follow/understand Scott Schedule format It also helps expose whether the disagreement is really about expert witness opinion, or about different factual assumptions, different instructions, different legal contingencies, or simply expert witnesses talking past each other from different starting points.^[8]See Part 2 dealing with “figures-as-figures” assessments which addresses a three-limb model by which different “figures-as-figures” assessments are produced

That approach is reflected in the Ciarb Guideline standard directions, which provide that witnesses giving concurrent evidence should jointly prepare a schedule identifying areas of agreement and disagreement and summarising their respective positions. In practice, where such a joint report is not prepared, the consequences are usually immediate: the areas of disagreement are not clearly articulated, counsel are required to identify them in real time, and the conference becomes materially more difficult for the tribunal to follow and use.

A witness conference without a proper joint report can quickly become untidy. The discussion drifts. The agenda has to be found in the room rather than before the hearing. Points that ought to have been narrowed in advance are instead explored live, often at greater length and with less clarity. One may still call that conferencing, but that does not necessarily make it useful.

Absent such a report, the comparative value of conferencing is materially reduced.

A joint report, by contrast, gives the conference shape. It tells the tribunal where it needs to spend time and where it does not. It also usually improves the quality of the cross-examination that follows, because counsel are then testing real points of difference rather than using hearing time to work out where they are.

Why Conferencing Should Usually Come Before Cross-Examination

In our experience, where witness conferencing is to be used, it is usually best undertaken before cross-examination. That is a conclusion we have reached in practice and one that sits comfortably with the sequence reflected in the tribunal-led Ciarb model, where the tribunal’s questioning in conference is followed by questioning from counsel.

Put simply, the better sequence is often: joint report first, witness conference second, cross-examination third.

There are several reasons for that. First, conferencing before cross-examination allows the tribunal to identify (and/or confirm) the real points of disagreement before the more adversarial testing begins. That means cross-examination can be shorter, cleaner and more focused. Instead of exploring every possible issue on the off chance that something useful turns up, counsel can concentrate on the points the conference has shown actually matter.

There is also a practical benefit that carries through beyond the hearing itself. Where issues are addressed sequentially in conference, the transcript tends to provide a clear, self-contained record of each discrete point of disagreement. That can materially assist the tribunal at the award-writing stage, where the task is often to work issue by issue through competing analyses rather than to reconstruct them from fragmented exchanges.

In furtherance of that efficiency, it may also be worth considering whether, for particular issues, cross-examination should flow directly from the concurrent discussion of that issue. That is not a universal solution, but where the agenda has been clearly defined in a joint report, it can allow the tribunal to deal with each issue in a single, coherent sequence: identify the disagreement, explore it concurrently, and then test it immediately through focused questioning.

Second, conferencing before cross-examination often preserves a more open and more technically useful discussion. Once extensive cross-examination has taken place, positions do not always become more nuanced. Expert witnesses, like everyone else, are not invariably at their most reflective immediately after being challenged for an hour or two.

Third, this sequencing fits the expert witness’s role. An expert witness is there to assist the tribunal. A tribunal-led discussion of the real points of disagreement, based on a proper joint report, often allows that assistance to be given more directly and more clearly than if the first substantial oral engagement is adversarial. Cross-examination can then do what it does best: test the points that have been clearly identified.

It is also logically preferable. Ideally, the disagreement should first be identified and understood and then tested. Too often in practice, cross-examination is used to discover the disagreement when it should really be used to test it. That is not necessarily fatal, but it is not the best use of time, and arbitrations are not generally improved by treating hearing time as though it were limitless simply because it has been paid for.

In practical terms, witness conferencing is less likely to assist where:

• the issues are highly segmented and do not lend themselves to comparative discussion;
• one expert witness’ evidence turns materially on matters of credibility rather than analysis; or
• the imbalance in experience or communication style between experts risks distorting the process. In such cases, sequential testing may provide a more reliable structure for the tribunal.

The Tribunal’s Role Matters More Than the Label

None of this means conferencing is always appropriate. It plainly is not. Nor does it mean that every process labelled “witness conferencing” will be helpful. The tribunal’s role is critical.

The Ciarb Report emphasises that expert witness evidence is more effective where tribunals are proactive, particularly in relation to issue identification, sequencing and procedural expectations. The witness conferencing guideline makes the same point in practical terms. It identifies a range of factors that may affect whether conferencing is appropriate at all, including credibility issues, differences in experience, cultural dynamics and interpersonal relationships between witnesses. It also makes clear that the tribunal may need to adapt, vary or even dispense with the process if it is not working.

That is entirely consistent with our experience. A well-prepared tribunal can make witness conferencing very effective by focusing on what it really needs from expert witnesses. A disengaged tribunal can make it feel like a slightly earnest panel event at the end of a conference day, with the same risk that one or two people do most of the talking while everybody else wonders when it will conclude.

Where conferencing is used, the tribunal should know what it wants from it. It should identify the issues on which concurrent evidence is likely to help. It should require a proper joint report or schedule. It should choose the right format — tribunal-led, witness-led, counsel-led, or a hybrid — for a reason, not by habit. And it should be willing to intervene if one expert witness is dominating or going off a tangent, if the dynamic is unhelpful, or if the discussion is generating more words than insight.

In construction and technical disputes, our view is that tribunal-led conferencing will often be the most effective format. It gives the tribunal direct control of the issues, allows both expert witnesses to answer on the same point at the same time, and usually provides the clearest foundation (and potentially reduces time) for any cross-examination that follows.

Presentations and Joint Presentations

Presentations can also be useful, but only if they are kept in their proper place.

The Ciarb Guideline recognises that presentations and demonstratives may be particularly helpful where issues are technical or complex, and that tribunals may direct separate presentations or, less commonly, a joint presentation dealing with areas of agreement and disagreement. Our experience has been similar. Views on presentations tend to evolve. They can be genuinely helpful where they are disciplined, issue-led and used to assist the tribunal rather than to re-argue the case with more slides.

In our view, any presentation should be anchored to the structure and issues already established in the joint report. If there is a joint presentation, it should deal neutrally with what is agreed and identify clearly why certain matters remain disagreed between the experts. If there are individual presentations, they should still follow that agreed issue structure. The purpose is to help the tribunal navigate the issues, not to create an additional opportunity for adducing further and/or new evidence under the polite label of “orientation”.

Conclusion

Cross-examination and witness conferencing are not rival philosophies. They are tools. Both have value. Both can also be used badly.

Cross-examination remains an important and often indispensable way of rigorously testing expert witness evidence. Witness conferencing, however, can offer tribunals a clearer and more immediate understanding of the issues that separate the expert witnesses, particularly where the evidence is technical and the disagreement is best understood side by side.

In our experience, where witness conferencing is to be used, it is usually best done after a proper joint report and before cross-examination. The joint report is essential because it crystallises the issues. The conference is then useful because it allows those issues to be explored clearly and comparatively, and with more control over keeping both experts on the same topic/issue. Cross-examination after that is more focused and usually more productive.

Ultimately, the question is not whether one prefers the style of cross-examination or the style of conferencing. The question is what will most help the tribunal in the particular case before it.

If the tribunal leaves the hearing understanding the real points of agreement and disagreement, the procedure has done its job. If it does not, no amount of procedural elegance is likely to rescue the position afterwards. Hearing formats, however well intentioned, are not self-justifying. And “hot tubbing”, despite the memorable label, is not in itself a guarantee of warmth, clarity or relaxation.

About the authors

Michael Tonkin is a Chartered Quantity Surveyor, Chartered Arbitrator, RICS Accredited Expert Witness and CEDR Accredited Mediator with over 35 years of construction industry experience. He is an effective dispute resolver and has acted as quantum expert on around 80 occasions.

Igor Corelj is a Chartered Quantity Surveyor and structural engineer with nearly 20 years of international construction industry experience. He has acted as a quantum expert six times and has assisted the named expert on numerous occasions.

---