Law vs science: scientific evidence and causation in class action litigation

Lessons on judicial approaches to complex scientific disputes from the Roundup litigation 6 min read

In class action litigation, where claims of loss and damage can reach into the hundreds of millions, it is increasingly common for parties to rely on highly technical and specialised scientific evidence on causation. But what happens when the science is 'all not one way'?¹

In this Insight, we consider how class action judges have approached cases with highly contested scientific evidence, the key takeaways that emerge and what it all means for the parties involved in class actions.

Key takeaways

• The court will make a decision about causation based on the weight of all the evidence, including scientific evidence. A positive finding of causation does not require scientific certainty, but parties should carefully consider whether the state of existing scientific knowledge is sufficient to support a finding of causation.
• Scientific evidence is only one component of all the evidence that will be considered by the court. In cases where scientific evidence is unsettled or inconclusive, lay evidence may be determinative.
• Courts have developed practical mechanisms to assist them to deal with complex scientific evidence, including the use of court-appointed experts, assessors or special referees. As class action filings continue at pace, it is likely that the use of these mechanisms will increase.

The Roundup class action

The recent case of McNickle v Huntsman Chemical Company Australia Pty Ltd [2024] FCA 807 (the Roundup class action) is a useful case study to consider the treatment of scientific evidence about causation in class actions.

Background

The Roundup class action was brought on behalf of sufferers of non-Hodgkin lymphoma (NHL) who alleged their cancer was caused by the weedkiller Roundup. The lead applicant was a frequent user of Roundup who was diagnosed with NHL in 2018. The lead applicant claimed that the use of, or exposure to, Roundup increased the risk of an individual developing NHL. The manufacturer denied the claim.

The question of general causation

In an initial trial to decide the question of general causation, the court had to grapple with scientific evidence from more than 10 experts across three different scientific disciplines to determine whether, on the balance of probabilities, exposure to Roundup increases the risk of (and/or causes an individual to develop) NHL. Reflecting on the evidence before the court, Justice Lee noted it was 'an exercise in considerable understatement' to describe the scientific evidence in the case as 'dense, complex, and voluminous'.²

The court's conclusion

With the assistance of a court-appointed assessor, Justice Lee concluded that the evidence was not sufficient to establish that Roundup causes or increases the risk of NHL in humans. In coming to this conclusion, His Honour observed that 'the whole of the evidence in the three streams revealed some not insignificant issues with the [lead applicant's] evidence'.³

We set out below the key points that emerge from Justice Lee's decision, and some considerations that parties should bear in mind when adducing scientific evidence about causation in class action proceedings.

Implications for parties in class action proceedings

The difference between legal and scientific concepts of causation

In the Roundup case, the question of causation being considered was general and factual in nature: did Roundup cause or increase the risk of developing NHL in humans? The respondent could not be held legally liable unless the applicant could prove on the balance of probabilities that the answer to this question was yes.

There is an evident tension between the time-bound, conclusive approach to causation in law and the scientific approach. Writing extra-curially, Justice Beach has described scientific theory as 'necessarily tentative and revisable'.⁴ Scientific understandings are constantly evolving and developing as new theories are tested and new technologies are developed, making scientific certainty somewhat of an oxymoron. However, the Roundup class action emphasises that '[a] judge (unlike a scientist) must decide, one way or another, whether the burden of proof is satisfied by the party positing a general causation hypothesis based upon scientific evidence'.⁵

This is consistent with the purpose of legal causation: to assign legal responsibility for harm. Doing so requires a judge to reach a static, point-in-time decision about a state of facts. Courts must decide cases in a timely manner, and in a manner that brings finality to the litigants involved—they cannot wait until the science is 'certain'.

Grappling with this in the Roundup class action, Justice Lee was at pains to stress that his decision was an answer to a legal question, not a scientific one. He emphasised that the science may change and develop over time.

The legal framework for considering scientific evidence

Near the beginning of his judgment, Justice Lee observed that 'it is trite that my job as a fact finder is to determine the central issue, being a factual issue, according to law, that is, by having regard to all the evidence admitted (and only the evidence admitted)'.⁶ Trite or not, this observation captures the foundational elements of the legal framework within which scientific evidence will be assessed in class actions.

First, the court will make a decision based only on the evidence that is admitted. Scientific developments outside of the evidential record will not be considered. Justice Lee observed that his decision was based on 'the evidence selected by the parties and which can only represent a snapshot of the state of scientific knowledge as at the time the evidence closed'.⁷ For the parties in class actions, this underscores the importance of leading reliable and credible scientific evidence from experts with the relevant expertise about the key matters in dispute.

Second, the court will have regard to the whole of the evidence admitted when reaching a conclusion about causation. In class action proceedings, like all proceedings, scientific evidence remains but one component of all the evidence. While no lay witnesses were called in the Roundup class action, lay evidence may be determinative in circumstances where the scientific evidence is unsettled or inconclusive.⁸

The challenge for the parties relying on scientific evidence is to satisfy the burden of proof on the basis of the evidence that exists at the time, which may be unsettled, contradictory or imperfect. When assessing prospects in litigation, potential litigants will need to assess the state of the scientific evidence and how that may translate to the legal standard of proof—'the balance of probabilities'. As Justice Lee emphasised, this does not involve 'a simple estimate of probabilities': rather, the whole of the evidence must allow for the court to be subjectively persuaded that the causal connection exists.⁹

Mechanisms to assist the court to deal with complex scientific evidence

To assist him to understand the volume and complexity of the scientific evidence adduced in the Roundup class action, Justice Lee ordered that an assessor be appointed pursuant to s23 and/or 33ZF of the Federal Court of Australia Act 1976 (Cth). An assessor is an expert in a particular topic who is tasked to assist the court to resolve contested factual issues.¹⁰

The assessor observed the concurrent evidence sessions and was available to consult with, and advise his Honour on, scientific and technical matters. His Honour also provided a draft copy of his reasons for judgment to the assessor to ensure he had not 'committed any scientific "solecisms" caused by any fundamental misapprehension of the scientific evidence'.¹¹

While the use of assessors by the court is not novel,¹² there has been some criticism, particularly in relation to procedural fairness and the delegation of judicial power.¹³ Addressing these criticisms in his judgment, Justice Lee was careful to explain the precise role of the assessor and the parameters around his dealings with the court. This included that his Honour would consult with the assessor in chambers about matters raised by the parties' experts, but that any opinion raised by the assessor that had not been previously identified by the parties would be disclosed and discussed with counsel.¹⁴ In relying on the assistance of the assessor, Justice Lee emphasised that he remained the sole decision-maker and tribunal of fact.¹⁵

Court-appointed assessors are just one of the mechanisms that have developed to assist judges to understand and interpret specialised technical evidence in fields outside of their expertise. Each of these methods offers its own benefits and disadvantages. Some other mechanisms that may be deployed in class actions include:

• Court-appointed experts: the court may appoint its own expert to report on a particular question.¹⁶ While this approach may allow for greater independence of the expert, it may lack the benefits of the more traditional adversarial approach (such as allowing the parties' experts to debate and narrow the scope of the issues in dispute).
• Special referees: the court may appoint a referee to inquire into one or more questions and provide a report to the court, which may be adopted by the court in whole or in part.¹⁷ Practically, this mechanism could be used to carve out a scientific question of fact (to be dealt with by a technically qualified referee) from a question of law.

In November 2022, the Federal Court issued the Referee and Assessor Practice Note (GPN-REF), which outlines the Federal Court's procedures for appointing referees and assessors. While this practice note affords greater clarity about the court's use of these mechanisms, practical examples of their use in class actions have been relatively rare. With complex scientific evidence becoming par for the course and class action filings showing no sign of abating, it is likely that the use of these mechanisms to assist the court to deal with highly technical scientific evidence in class action litigation will only increase.