Cosmos Briefing: Science v law

The recent pardon petition by leading Australian scientists regarding the Kathleen Folbigg case – in which new scientific findings have shown her four children could have died of natural cases – has raised a number of questions about how specialised scientific evidence is regarded by our system of courts.

Cosmos has convened a webinar with Gary Edmond (law professor at University of NSW), Anna-Maria Arabia (chief executive of the Australian Academy of Science) and Richard O’Brien (endocrinology professor at the University of Melbourne and frequent expert witness) to explain why science and law struggle to communicate.

To watch this session, hosted by Professor Alan Duffy of The Royal Institution of Australia on 12pm (AEDT) Tuesday 9 March, register here.

What’s at stake?

As the Kathleen Folbigg petition went to NSW Governor Margaret Beazley, we spoke with Professor Gary Edmond about the reasons why scientific evidence has a difficult passage to court.

Edmond directs UNSW’s Program in Expertise, Evidence and Law. He originally trained in the history and philosophy of science, and later studied law at the University of Sydney and the University of Cambridge. Until this year he was a member of the Council of the Australian Academy of Forensic Sciences, and a member of Standards Australia’s forensic science committee. He remains a member of the editorial board of the Australian Journal of Forensic Sciences, a Fellow of the Royal Society of New South Wales, and served as an international adviser to the Goudge Inquiry into Pediatric Forensic Pathology in Ontario (2007–2008). He is Chair of the Evidence-based forensics initiative (EBFI) and is co-author of Australian Evidence: A principled approach to the common law and the uniform acts.

What are the rules around admissibility of scientific and other expert evidence in Australian courts?

It will be surprising to many people that, in terms of admissibility, courts are not concerned with whether a technique or a procedure used by, say, a forensic scientist, is valid and reliable. They don’t ask whether a person who claims to have some kind of ability and where the expertise is has a demonstrable ability where there’s evidence for that – they’re not part of the admissibility criteria. The basic approach is that judges should not attend to the reliability, validity, and other indications of the trustworthiness of the claim.

All the possible rules regulating the admission of expert evidence have explicitly repudiated a need to attend to reliability. So-called expert evidence really has a low bar, or there’s no gate stopping it getting in. And once it gets in, it’s up to the parties to enable the decision-maker – often a jury but sometimes the judge – to determine the value of evidence. And that’s a problem.

Your points about scientific validity and reliability – that there isn’t a test for those things ­– are probably the most surprising.

It’s surprising to me! And I shouldn’t say they’re surprising to everyone, except maybe the lawyers and judges in some of our jurisdictions. It’s inconsistent with international common law comparisons around the world. The United States has a formal reliability test – it’s now written into their rules of evidence, the Federal Rules of Evidence. England has introduced the need for reliability into rules of procedure and practice directions. And Canada has adopted a version of the United States’ reliability test, which has a focus on things like testing, error rates, standards being applied, and so forth – criteria that enable decision makers to rationally evaluate evidence.

When we talk about validity and reliability in relation to science in a court matter, we’re asking if it has been done properly, effectively, it can be tested, it’s been reviewed – all of these sorts of things?

Yes, we are. Now, the only time these questions matter is at trial, if they’re asked. And the Folbigg example shows that we can’t be sure that they’ll be asked, even when they’re important.

Many of us would tend to think that the courts would look for expertise towards people who work in an area where proven facts and validity and reliability are critical to the practice of their knowledge. That’s not the case?

It might even make the court’s job a lot easier, but unfortunately that’s not the case.

There’s a broad spectrum of types of expert evidence in the scientific, medical and technical domains, ranging [down] from those which are really well validated – for instance, many of the types of DNA processing have been validated in many ways and repeatedly, and they’re robust. But other types of forensic sciences have either been only tested recently, or are yet to be tested. And we don’t know about their uses in all sorts of situations, at least partially because the courts have never required it. DNA typing evidence has come out of mainstream science, and there was interest in its validity there. That has spilled over to its availability for forensic uses. Most of the other forensic sciences emerged out of police investigations and some are yet to be rigorously tested.

What about the people that are discussing or examining evidence in court – the judges and juries and barristers. Surely there must be rules in place that ensure that they understand what’s being presented to them if the evidence is technical or scientific?

There are studies suggesting that these types of evidence hadn’t been rigorously challenged for over 100 years, so I’d suggest maybe we’re not doing very well or don’t know the sorts of things we need to focus on. I think there’s a belief that lawyers, particularly barristers and judges, are very sceptical, critical people. But to become a lawyer, a barrister, or a judge dealing in criminal law or civil disputes involving patent or trademark disputes in technical areas, there’s no requirement to have any technical, scientific or medical or statistical training. Some people might acquire bits and pieces, but there’s no sense that in practice you acquire something similar to a formal education in these increasingly specialised disciplines.

One would think that in high-stakes cases such as appeals and inquiries that there would be specialists, if not the individual judges or commissioners, then perhaps specialist panels appointed to examine evidence. Does that occur?

It happens far less often than you might think. I think the Folbigg inquiry is a nice example. It’s remarkable that you have a case with complex forensic pathological evidence, and now additional genetics and statistics. It’s probably the most complex evidence you’ll ever come across in a criminal proceeding. Instead a retired judge from the Supreme Court was appointed to hear and decide these incredibly complex questions. I’m not saying that a judge shouldn’t be involved, but I think the judge, and perhaps a forensic pathologist, and a geneticist or a statistician should have sat as a panel to hear the evidence.

When we have technical, medical, scientific, evidence, it’s hard to understand why we don’t have more multidisciplinary groups making the decision. It’s a hubris of law, or maybe a failure to realise the complexity of the issue. Judges do have scope for appointing court-appointed experts, or for special advisors during proceedings, especially complex or large-scale proceedings. The litigation over the damages for the Victorian bushfires were very large-scale proceedings, with hundreds of millions of dollars of claims, and an advisor was appointed to assist the judge. But it happens quite exceptionally, when it probably should happen much more often.

The other thing that I’ve argued … is that it’s also hard to understand where we have types of forensic science evidence that we use routinely – whether it’s fingerprints, or the emergence of algorithms dealing with identifying faces from CCTV, or comparing voices on voice intercepts – why we allow judges in individual proceedings to decide on eligibility, as opposed to having some kind of commission or agency that does a formal review for the courts.

You could have legal people [involved], but you actually want technical people to say, “this is okay”, or “this is okay and here are some of the limitations, so be careful with the way you use it”. But leaving it as, “oh, this seems okay to me, and I’ll let the jury make sense of it, or make whatever they want of it, depending on what might happen to trial” seems inadequate, especially given that many defendants are limited by resources to “test” the evidence.

We know that the body of human knowledge is expanding and continues to expand, and that science in particular – as it learns faster through technological advantage – is teaching us things we never knew. Surely the law should take steps to keep pace.

It hasn’t really done that. I think the adversarial method, and the case-based nature of common law – where you just focus on the individual case – means there’s no ability for the courts to have some kind of systematic or synoptic view of developments in science and how they might adapt legal rules in response. Judges are unable to do their own research or investigations; it’s all on the parties. Appellate judges are particularly constrained by what is brought to them by the parties. They can’t say, “where’s this?” or, “actually I’ve done some reading and I think this is missing”.

The issue about the expansion of the sciences and of increasing specialisation and complexity manifests in my work. Increasingly I work with scientists because it’s too hard to do myself – you have multidisciplinary teams to assist. But it raises issues in terms of the progress or development in sciences, about the law’s quite strong commitment to this principle of finality. It’s very concerned about ending things, particularly civil litigation, but also in criminal litigation. So you have a trial, then you have rights of appeal. But once your normal appeal rights, which tend to be based on questions of law and procedure – whether things were done correctly, or appropriately, at trial – are exhausted, that’s it.

There’s some scope for fact review but that’s more limited. A couple of Australian jurisdictions have expanded the scope for opening this up, but it’s still really quite formalistic. If there’s strong evidence that suggests that a person might be innocent, courts should be much more open. Some jurisdictions have formal institutions – England, Canada is going to have one, New Zealand’s going to have one, Scotland has one. An independent commission, which often does some investigation, reviews the evidence, and makes a recommendation whether something should go back to a court or not. The court still makes the decision, but the commission is an important [independent] player.

In the case of those jurisdictions, that review commission is where you would present new scientific evidence? For instance, new scientific research that didn’t exist at the time of the original trial?

That’s right. And they would review the evidence and make a recommendation. It’s still going to go back to a court to consider the evidence – that’s appropriate; I don’t think there’s a problem with that necessarily, though maybe the court should be a bit more open to having some advisors sit with them in some of these cases where the evidence is complicated. I can’t see that in our society, with this increasing specialisation, that they should be too reluctant to more actively incorporate non-legal participants, even if courts are ultimately responsible for the decisions.

Based on what you’re saying about the reasons – procedural and matters of law – that prompt appeals and reviews, presenting new evidence sounds like a tough game.

Very tough. These things take a long time. And don’t forget, the issue is whether this person has been appropriately convicted. In reality, that’s the question – should this person be in jail or not? And this might take years, or even decades. It takes a long time for the institutions to unravel their mistakes. For institutions that claim to be very concerned with fairness and rectitude, once they’ve made these decisions, it’s very difficult for them to go back on them. I also think that judges, especially where there’s fresh evidence but even generally, are possibly too deferential to jury decisions. It’s important in our system that we respect jury decisions, but we ask too much of juries with complicated forms of expert evidence.

In the Folbigg 2019 judicial inquiry, the science is very complex. Surely it’s important that it’s presented and understood?

Yes, presentation and comprehension are really important issues. You can see that there are practical issues about how many appeals there can be. Courts are pragmatic in regulating access to very expensive and time-consuming processes, but that needs to be counterposed with genuine developments or new evidence or fresh evidence, which might suggest that people shouldn’t be considered guilty or probably shouldn’t be in jail. That’s fundamental. Historically, we used to say that it’s better that hundreds of people go free than one innocent person sits in jail. That number’s come down over the centuries. Those high principles don’t seem to be held in a way they once were.