In 2023, a new Macrossan lecture series was established by The University of Queensland’s TC Beirne School of Law and the UQ Law Association (UQLA). The annual public lectures are given in the Banco Court of the Supreme Court of Queensland to celebrate the commencement of each academic year. In the spirit of the original lecture series, distinguished Australians will be invited to address varied topics of public importance. The 2024 lecturer was Geoff Davies AO KC – a former justice of the Supreme Court of Queensland, sitting in the Appeal Division, and before that appointment having served as Solicitor – General of Queensland and President of the Bar Association of Queensland. The lecture was delivered on 7 March 2024. The lecture synopsis was this:
“There are two serious problems with our civil justice system. Both involve cost.
The first is that the vast majority of actions which are properly amenable to resolution before a judgment by a court tend to be resolved only at trial by which time huge and unnecessary costs have already been incurred by all parties.
And the second is that a trial under that system is beyond the financial capacity of most of those who would use it. And consequently they have to drop out.
The lecture discusses both of these problems and suggests reasons for them and ways in which they can be eliminated or at least substantially ameliorated. In both cases these involve radical reform: the first involving much greater judicial control over the pretrial process; and the second requiring, at least for most litigants, a simpler and cheaper system of trial.”
An expanded version of the lecture will soon be reported in the University of Queensland Law Journal. When published, undoubtedly, it will repay careful reading and consideration.
The lecture was as follows:
The Macrossan Lecture 2024
Is the Civil Trial System Past its Use-by Date?
Thursday 7 March 2024
I am honored to be giving the Macrossan Lecture for 2024, which celebrates the contribution of the Macrossan family to the law in this state.
As some of you did, I knew the last of the Chief Justices, the Honorable John Macrossan. I was junior to him at the Bar. On many occasions, I appeared before him when he became a judge. And I later served with him in the Court of Appeal.
He was an outstanding lawyer and he had a very quick mind. But his mild and polite manner often concealed, from those appearing before him, how quick he had been in perceiving their argument, and identifying its weaknesses, often to the cost of the barristers concerned.
I found him a wonderful person to appear with in a court. I found him difficult as a judge, because I had to prepare more for him than I had with some other judges. And he was a joy to sit with in the Court of Appeal, because he could do all the work for me. I learned a lot from him, although I never managed to master his politeness or his felicity of expression.
Because of the topic of my lecture tonight, and what I propose to say, I should add that John Macrossan was a conservative and a traditionalist – and I am not sure how much he would enjoy about what I am going to say.
… John Macrossan was a conservative and a traditionalist – and I am not sure how much he would enjoy about what I am going to say.
Introduction
In delivering the Gerard Brennan Lecture in 2011, the Hon John Doyle AC, Chief Justice of South Australia, a person not given to exaggeration, said this: “I believe that, during the time of the next generation of legal practitioners, those now being admitted to practice, civil litigation as we know it in the higher courts will come to an end”. And he subtitled his lecture: “the demise of civil litigation.”
There is no doubt, in my opinion, that the cost of a trial of an action in the Supreme Court, even for a successful party, is, in most cases, beyond the means of the average citizen, or even of a small company. For a losing party of average means, that is almost certainly the case. And there is nearly always a risk, for either party, of doing worse by judgment than that party expects. That has been the position now for a considerable time. That is why over 95% of actions commenced in the Supreme Court are resolved before judgment.
The cost of going to trial and judgment is, for many parties, the effective deterrent to proceeding to an action, especially where the amount or value in dispute is less than, say, $1 million. In those cases, costs may be a substantial proportion of that amount or the value in dispute. So the system of civil litigation by trial and judgment is not working for most potential litigants. That is, no doubt, why Chief Justice Doyle said what he did.
… civil litigation as we know it in the higher courts will come to an end.
There are two aspects of cost which, in my opinion, are of concern. The first is that, of those cases which are resolved before judgment, a massive proportion are resolved only at or close to trial, after all pre-trial costs have been incurred. And that is, unfortunately, at a cost which most litigants can’t afford and, as it turns out, a cost which was, in large part, unnecessarily incurred.
And the second is that, of those cases which are resolved before judgment, it is likely that a substantial portion are resolved only or principally because the litigants, or one of them, cannot afford to proceed. There are, I think, many cases resolved by agreement, in which one or both parties to the dispute would reasonably have preferred to have their dispute resolved by an independent arbiter at a reasonable cost than by agreement.
So first, in this lecture, I propose to say why I think that our system, and the way those within it operate, impede early resolution of a dispute.
Secondly, I will explain what I think should be done to change this. There are two aspects of this. The first is to show how I think that those cases which are resolved by agreement may be resolved before the majority of pre-trial costs have been incurred. And the second is to explain why I think that there are some cases which are incapable of resolution by agreement, at least initially, but amenable to early neutral evaluation; and how that may be achieved.
Thirdly, I want say why I think that the present system with respect to experts is misconceived, and how radical change to that system will ensure earlier, cheaper and fairer resolution of questions requiring expertise – and consequently in many cases, earlier resolution of an action.
And finally, I want to say something about whether, and if so how, an alternative cheaper system of deciding disputes can be devised. For if it can’t, our system of trial and judgment will remain effectively accessible only to persons or entities to whom the cost is not an effective deterrent. In that case, it would be past its use-by date.
Why the adversarial system, and the way those within it operate, impede early resolution of a dispute
An analysis of actions commenced in the Supreme Court over the past decade yields two relevant percentages. The first is that, of actions commenced, more than 95% are resolved without a judgment. And the second is that, while approximately one half of defended actions are listed for trial, of those which are so listed, less than 10% go on to trial and judgment. That is, more than 90% of those are resolved by settlement or abandonment only after most, or possibly even all, pre-trial costs have been incurred. And that percentage appears to be increasing over the last few years.
The reality is that our system is one of party autonomy or, more accurately, the autonomy of the parties’ lawyers, and that lawyers tend to do too much before attempting settlement.
There are three impediments to early resolution of disputes which I think can be eliminated or at least mitigated.
The first is the way in which the system requires all actions to proceed under the existing rules of court as if they are to go to trial and judgment, notwithstanding that the reality is that the vast majority end before then. Under the Uniform Civil Procedure Rules, ADR is relatively new. It came into the UCPR only in 1999. And in those Rules it appears to arise only after virtually all of the trial preparation has taken place. In other words, the system itself still assumes a likely trial and judgment, with ADR only as a last minute exception.
The second and third impediments are of greater concern.
The second is the tendency of lawyers, notwithstanding that they could seek ADR early in proceedings, to take every step in those proceedings, almost up to trial, before attempting to do so. An explanation for that tendency, by some lawyers, is that, unless they proceed up to at least disclosure before seeking resolution by agreement, they may not uncover all of their opponent’s relevant documents. And that if they proceed to settle an action before then, they leave their client unprotected, and themselves open to action by their client, if a later contention or discovered document might have changed the client’s decision to settle.
The reality is that our system is one of party autonomy or, more accurately, the autonomy of the parties’ lawyers, and that lawyers tend to do too much before attempting settlement. Party autonomy is a leftover from the time when the judges of fact were juries. Under such a system it never occurred to lawyers or to judges that control of the litigation process up to the time of trial, its substantive issues, its form and its speed should be other than entirely in the hands of the parties’ lawyers. But it persists, notwithstanding the change to judge-only trials, due to the reluctance of lawyers to forego control of the pace and, especially, the shape of litigation – and the reluctance of judges to seize it.
As Chief Justice Doyle pointed out: “The judge does not determine the issues that will be contested between the parties. The judge does not determine how the parties will present their respective case or defence. The judge has no significant control over the quantity or quality of evidence presented on either side, nor over the relationship between the significance of the matter at issue and the effort or resources deployed by the parties. They can commit more or less resources than the issue warrants. Often it is more.”
There are incentives to a litigation lawyer under our existing system to do too much, rather than too little. In theory, the more work that is done in preparation of a case, the better are the client’s prospects of success; the better is the lawyer protected against later being sued by the client; and the more that the lawyer earns. And there are no disincentives. Moreover, in most cases it is difficult to judge how much is too much.
Whilst party autonomy remains the essence of our civil justice system, it will remain one in which lawyers do too much and, consequently, costs are too high. Moreover, party autonomy ensures that litigation generally proceeds at the pace of the party least inclined to bring it to trial. And it may introduce unfairness into litigation between parties of unequal bargaining power. The richer litigant may use that to its advantage, obliging the other to commit more resources than it can afford.
There are incentives to a litigation lawyer under our existing system to do too much, rather than too little.
Although there have been some modifying changes, it remains fundamentally correct that the parties or, more accurately their lawyers, remain substantially in control of the pace and shape of litigation. And if we are to have a system which works for the majority, that must change. In each of these respects, there must be greater judicial control, or rule control, especially in the pre-trial process, than is presently the case.
And the third impediment to early resolution of a dispute is the likely failure of lawyers, before litigation commences, and thereafter if circumstances change or the situation becomes clearer, to advise their clients realistically both of the likely result of litigation and of its likely cost.
What parties to a dispute want, or more accurately what they need, when they first approach a litigation lawyer, and continuously thereafter, may be expressed in three questions:
What is the likely result if the dispute goes to judgment?
What will it cost me to get to that result?
Can I get to that result, or a result approximating it, more cheaply?
Without answers to each of these questions a party has no real prospect of weighing the relative advantages and disadvantages of, on the one hand, reaching an agreed early solution and, on the other, proceeding to judgment – or, indeed, of not proceeding at all.
At present, it seems to me, they are unlikely, in most cases, to get reliable answers to any of those questions, either when they first consult their lawyer or at any subsequent time before the imminence of trial.
As to the first of these – advice about the likely result – this is a common law duty. Unfortunately, there is no statutory statement of that duty, notwithstanding the existence of a statute stating lawyers’ other duties. There should be such a provision – and one stating that the lawyer could be deprived of her costs for giving and failing to correct unrealistic advice.
Absent such a disincentive, any such advice is quite likely to be unreliable and optimistic, at least until a trial is imminent.
As to the second and third questions – each requiring a realistic estimate of the cost of proceeding to trial and judgment – there is already a statutory obligation upon a lawyer, on receiving instructions in any matter, to give her client an estimate of the total legal costs, if reasonably practicable and, if not reasonably practicable, a range of estimates.
These provisions, in my opinion, are an unsatisfactory statement of the continuing duty of a litigation lawyer to advise her client of the likely costs of an action proceeding to trial and judgment – and of the consequences of failure to perform that duty.
How resolution by agreement may be achieved earlier
I want to turn now to how resolution by agreement can, in my opinion, be achieved earlier. But first, I would like to warn against introducing procedures aimed at resolving a dispute, which may in fact increase costs.
Here is an example. Following the Woolf Report in England, pre-action protocols were introduced by a practice direction under the Civil Procedure Rules, the aim being to encourage parties to settle their dispute without the need to issue proceedings. These require the parties to exchange correspondence and information sufficient to understand each others’ positions.
These requirements, or, more accurately, the way that they have been implemented by lawyers, involve too much cost to be incurred too soon in a system in which, statistically, the vast majority of cases will be resolved before judgment.
Whilst this is a laudable aim, it has tended to front-load costs because, it has been noted, the requirements generate time consuming and costly exchanges. These requirements, or, more accurately, the way that they have been implemented by lawyers, involve too much cost to be incurred too soon in a system in which, statistically, the vast majority of cases will be resolved before judgment. So it is not something I would recommend for adoption here.
I turn now to some measures which I think will have a positive effect.
The plaintiff’s lawyer should be obliged, shortly after the filing and service of the originating process, to provide to her client costs estimates and an opinion of the likely result if the action goes to trial and judgment. And at the same time, the lawyers of any persons so served, who propose to defend, should be obliged to provide the same to their clients.
Accordingly, I suggest, that within a short time, say 14 days, after filing and service of the originating process, each party’s lawyer should be obliged to provide her client with a costs statement containing:
details of legal costs payable by the party to the party’s lawyer up to that date; and
an estimate of the party’s likely legal costs, if the claim proceeds to trial and is determined by a judge;
and that a copy of that statement be provided to the court.
At the same time each party’s lawyer should be obliged to provide her client with a statement estimating the likely result of the action, if it proceeds to trial and judgment – and provide the court with a statement that she has done that. And the lawyer should be obliged to file a sealed copy of that advice to be opened only on a question of costs.
These obligations should be enforceable and enforced. The lawyer should have her costs, otherwise recoverable from her client, reduced if she fails to comply with these provisions. An unrealistic estimate of costs or of the likely result of the action should, in the absence of compelling reason to the contrary, be sufficient evidence of that failure. There should be provisions in the rules which provide for that enforcement.
At the same time the parties should be obliged to exchange a short number (say 10) of the principal documents upon which they rely.
Under the present regime, I am told, some parties, or, perhaps more accurately their lawyers, tend to treat early mediation not so much as an attempt to resolve the action by agreement, but more as a means of testing the strengths and weaknesses of the opponent’s case.
Once these obligations have been complied with, or the time for performing them has expired, mediation should automatically occur. By that I mean that mediation should occur without the necessity of a court order. There would, of course need to be provisions for how that takes place.
Under the present regime, I am told, some parties, or, perhaps more accurately their lawyers, tend to treat early mediation not so much as an attempt to resolve the action by agreement, but more as a means of testing the strengths and weaknesses of the opponent’s case. But if the parties are by this stage, armed with more realistic answers to the questions posed earlier, and knowledge of the opponent’s principal contentions and documents, there should be greater interest in the possibility of resolution by agreement. That may be especially so in cases in which, it can now be seen, costs are likely to assume a substantial part of the amount or value in issue.
What should happen next
So far I’ve discussed automatic procedures; that is, those which occur without the intervention of a judge.
But if the action has not been resolved by these procedures, I think the court should intervene, although it could do so by informal means. But it should have power to make orders for:
further steps;
further mediation;
neutral evaluation;
appointment of an expert to decide any question involving expertise;
deciding separately, and in advance of trial, any question of fact or law in the action;
most importantly, how the action is to be tried.
Now I appreciate that some of these may be dealt with under existing practice directions. However, I want to say a little more about some of these which may result in earlier resolution.
Resolution by early neutral evaluation
There are some cases which will not initially be capable of resolution by agreement, but which would be amenable to early neutral evaluation. One example is where a party, or that party’s lawyer, has an unrealistic view of the likely result upon judgment. Another is where, because of a substantial financial imbalance between the parties, the negotiating power of the poorer party may be impaired by her realisation that litigation against her richer opponent may lead to her own financial ruin – and where that imbalance is impeding a fair resolution by agreement.
I do not mean to imply that case appraisal should be limited to those cases. It is useful in many cases in which there appears to be a reluctance to agree. That is why early neutral evaluation should be an important part of the procedures used to resolve actions.
Case appraisal has been rarely used. I am told that the principal reason for this is that lawyers have tended to prepare for a case appraisal as if it were a trial and so, in effect, incurred the costs of a trial. That, of course, was never its intention. It was intended to be a summary procedure in which costs were limited. It is another example of lawyers tending to do too much, more than is warranted by the evident intent of the provisions, and therefore to cost more than the procedure intended.
If it is to be used, as I think it should, an order for it should be accompanied by an order limiting the costs of both parties in the case appraisal.
Resolving questions involving expertise more rationally, fairly and cheaply
I want to turn now to resolving questions involving expertise more rationally, more fairly and more cheaply than they are now.
If I were to start afresh, to design a means of deciding a question involving expertise, I would, having no expertise myself, ask an expert to decide it.
Sometimes the resolution of a question involving expertise will lead to resolution of the case or a shortening of a trial. It is therefore beneficial in all cases to resolve that question promptly. But under our present system that requires a trial of that question. That is because our system, in my opinion, misunderstands the true role of an expert in court proceedings.
If I were to start afresh, to design a means of deciding a question involving expertise, I would, having no expertise myself, ask an expert to decide it. Or if it appeared that there might reasonably be different views among experts about the question, I would appoint a panel of those experts holding different views and attempt to have them resolve those differences. And to the extent that they did not, I would accept the majority opinion. But again, having no expertise, I would not attempt to decide the question myself.
That is the way in which such questions are decided every day in business, in industry and in the professions.
If someone were to suggest to me that such a question could best be resolved by my hearing two or more experts, appointed by opposing parties, have each of them cross-examined by a person who had no expertise, and by then attempting to decide this question myself, I would think that suggestion irrational. And so would those who seek to have such questions decided every day in business, in industry and in the professions.
Yet that is the system which, with some minor modifications, we have now.
Judges should not attempt to decide questions involving expertise. Such questions are, by definition and in reality, beyond the competence of judges to decide.
To be fair to our distant forebears, there may well have been a time when questions involving expertise which came before a court were few and simple; and, consequently, questions which, with a little help, a judge, or more likely a jury, could decide. And it may be that jury trials could not have been conducted efficiently in any other way.
But if that were so, it has long ceased to be the case. Certainly, ever since I have been involved in the law, there have been many actions in which a question involving expertise arises which is beyond the understanding of a judge, let alone within her competence to decide.
Nor should a question involving expertise be turned into an adversarial contest. Especially one in which arguments on each side are made or controlled by non-experts. To want to do this requires an adversarial mindset.
Nor should a question involving expertise be turned into an adversarial contest.
The irrationality of the present system is highlighted by the rules relating to referees. Under those rules a judge may appoint a referee to decide any question of fact or law – including questions which the judge herself is eminently qualified to decide. But a judge may not appoint an expert to decide a question involving expertise, a question which is, by definition, beyond the competence of a judge to decide. It must be decided by a judge (or jury) after an adversarial contest. Of course, a judge may appoint an expert under the present system, but that is as a witness only in that adversarial contest.
The system therefore, in my opinion, misunderstands the real role of an expert. She should not be a witness in proceedings in the way in which witnesses of fact are. She is, in reality, a decision maker in the way in which a judge is, or perhaps a better analogy, as a referee is. The expert’s role is to decide a question which is, by definition, beyond the expertise of the judge to decide.
That is, in my opinion, the fallacy of the existing rules.
Some may say that that is what, in effect, occurs now – that the expert, or panel, in effect, decides the question. But if that is so, then we should face reality. We should not pretend that this is an adversarial contest which is decided by a judge.
An expert (or panel) should, in most if not all cases, be appointed by the judge to decide a question involving expertise and, once appointed, she (or they) should proceed to decide that question. No doubt the expert should be able to confer with the judge to ensure clarification of the question which she has to decide and the facts on which to decide it. But she should have no more contact with the parties or their lawyers than does a judge or a referee.
Nor should the resolution of a question involving expertise be turned into an adversarial contest. The expert should not be cross-examined by the parties. If there is any doubt as to the factual basis upon which the question is to be decided, or the question itself, the judge should have the power to seek the parties’ submissions on this.
So it seems to me that the rules with respect to experts should be radically changed. The existing rules should remain for exceptional cases only, and otherwise there should be rules along the lines I have suggested.
Such a system would have a number of advantages over the existing system.
In the first place, it would recognise the reality that an expert (or a panel of experts) is, in effect, the decision maker on the question involving expertise.
Secondly, by ensuring that the judge is appointing a decision maker, it would remove the risk, which presently exists, that the opinion of an adversarially-appointed expert, which may be accepted because she is more articulate or more persuasive than the opposing expert, will be biased in favour of the party who appoints her. “Whose bread I eat, his song I sing.” It is mere wishful thinking that statements in court rules, purporting to impose duties on expert witnesses, will change that.
Thirdly, it enables an effective decision on that question early in the litigation process.
Unfortunately, these rules are unlikely to be used by lawyers unless compelled to do so. They will be unwilling to surrender their own appointed experts.
And fourthly, it eliminates the costs incurred in having opposing experts proofed by non-experts and then cross-examined by non-experts.
Unfortunately, these rules are unlikely to be used by lawyers unless compelled to do so. They will be unwilling to surrender their own appointed experts. So these provisions should state, in my opinion, that they will apply, to the exclusion of Chapter 11 Part 5, other than in exceptional circumstances – and that the fact that a party has appointed an expert is not an exceptional circumstance.
Deciding on the manner of trial
If the purpose, or one of the principal purposes, of our civil justice system is to provide resolution of disputes by adjudication by a judge, then it does not fulfil that purpose if it provides access to such adjudication only to a small proportion of those who would want to use it – the very rich, or the funded, or those few other potential plaintiffs whose likelihood of success is all but guaranteed. And I think that that is the case.
There are, it seems to me, two related reasons for this.
The first is that, over many decades, the classes of litigants have increased substantially. Before, say, the 1950s, most litigants were either men of property (no women), or corporations. Then in the decades following Donoghue v Stevenson (1932), the law of negligence expanded exponentially. Added to that, social changes and legislative initiatives, especially in social welfare and economic regulation, which commenced in the 1950s and are still continuing, have made all of us potential litigants.
And the second, partly for the above reasons, is that relations between us, especially business relations, and consequently litigation, have become more complex and therefore more expensive.
The massively high proportion of cases commenced which are resolved, mostly by agreement, before judgment may therefore be seen in two lights. On the one hand, it is to be applauded that so many litigants have chosen to resolve their disputes by agreement. On the other, if it is the case as I think it is, it is to be regretted that so many litigants have been deprived of having their disputes resolved by a judge solely or principally because they cannot afford to go to trial.
There is little doubt that AI could and should reduce the work done by lawyers in basic routine tasks like disclosure.
If it is true as, at present, I think it is, that many litigants settle their actions solely or principally because they cannot afford to go to trial, there needs to be a more summary system of adjudication which is within the means of ordinary litigants and small companies.
The present situation may well change with the introduction of artificial intelligence into our system. There is little doubt that AI could and should reduce the work done by lawyers in basic routine tasks like disclosure. It should also enable more accurate prediction of costs and, in consequence, increase earlier resolution by agreement. It may enable more accurate prediction of the likely outcome of cases with the same result. And it may enable accurate prediction of those cases, or those kinds of cases, which are more likely to be resolved by agreement.
And, perhaps most importantly of all, it should assist in designing a cheaper and fairer system of deciding disputed questions of fact and law.
So the application of artificial intelligence to our system is likely ultimately to substantially change it. But whether, how, when and to what extent artificial intelligence will reduce the present costs of litigation, and change the present trial system, are, at present, unpredictable. I do not think that the changes which I have proposed should be postponed.
I shall not attempt to design the kind of trial system which I have in mind but, in the absence of artificial intelligence, it would be one along the lines of the present rules with respect to case appraisal but with a binding judgment. However I think that AI could be used to design a cheaper and more efficient way of deciding disputed questions of fact and possibly also of law.
Conclusion
I have said in this lecture, more than once, that our present system of trial and judgment is beyond the financial capacity of an ordinary litigant or small company, and that only large corporations, government entities, funded parties and those whose success at trial is all but guaranteed can afford to litigate to judgment in the Supreme Court.
If that is so, then our trial system is working for only a small percentage of potential litigants.
In this lecture, I have suggested two possible solutions. The first is one which, within the existing trial system, will better ensure that all of those disputes which can reasonably be resolved by agreement are resolved earlier and more cheaply than they now are.
The second is more radical, because it involves an alternative, cheaper trial system for litigants who should be able to have their disputes resolved by an independent arbiter, but who cannot afford to litigate under the present system. And these may well be the majority of litigants in defended cases.
Do I think that my proposals are likely to be adopted in the foreseeable future? No I don’t.
Hardly anyone likes changing the way they do things, especially if they have been doing them that way for a long time. And even less so if, as in this case, the way they have been doing it follows practices and traditions of highly respected forebears. The law is a conservative profession, as indeed, in many ways, it should be.
Do I think that my proposals are likely to be adopted in the foreseeable future? No I don’t.
This makes it difficult for us to accept that a system, which was once the subject of pride and admiration, has, by a substantial increase in the classes of litigants it serves, and a substantial increase in the complexity of our relations with one another, ceased to operate successfully; that a system which may have operated effectively in, say, the 1950s, has become, in its operation, too labour intensive, and so too expensive, for most of the classes of litigants it now serves.
But consider this.
Under a system substantially controlled by the parties’ lawyers, though 50% of defended actions go to trial, only 10% of those go on to hearing and judgment. That is, 90% of actions listed for trial are either settled or abandoned.
In addition to that, you have an overall dropout rate of 95% of actions. So together, a 95% dropout rate and a late dropout rate of 90% of actions listed for trial. Doesn’t that show that the trial system is not working? How many of those dropouts would have been better satisfied, and reasonably so, by a cheaper trial system?
So it seems to me that radical change to our system is not just desirable but essential. And if that includes a radically different system of deciding disputed questions of fact and law, it may fulfil the prediction – some may have said gloomy prediction – by Chief Justice Doyle.
