McNickle v Huntsman Chemical Company Australia Pty Ltd (Initial Trial)

[2024] FCA 807

FEDERAL COURT OF AUSTRALIA

McNickle v Huntsman Chemical Company Australia Pty Ltd (Initial Trial) [2024] FCA 807

File number:

VID 243 of 2020

Judgment of:

LEE J

Date of judgment:

25 July 2024

Catchwords:

REPRESENTATIVE PROCEEDINGS – class action concerning the alleged carcinogenic effects of Roundup produced by the Monsanto Company – initial trial of common questions – general causation – whether use of or exposure to Roundup products which contained glyphosate increases risk of developing non-Hodgkin lymphoma (NHL) – where ten expert conclaves ordered – where three streams of scientific evidence – epidemiological stream – animal studies stream – mechanistic evidence stream – where evidence as a whole does not establish that use and/or exposure to Roundup Products which contain glyphosate cause NHL in humans – wisps of smoke but no fire on the evidence adduced – central common question framed and answered – orders made

EVIDENCE – proper approach to fact finding – s 140 of the Evidence Act 1995 (Cth) – where all evidence was expert or documentary – evidence of general causation – onus of proof – observations as to “weight of evidence” approach – assessing the whole of the evidence – proper approach to evidence of scientific articles and studies – Karpik v Carnival plc (The Ruby Princess) (Evidential Ruling) [2022] FCA 1318 – where agreed position as to admissibility –opinions to be assessed on their merits

EVIDENCE – expert evidence – “hired guns” – credibility findings – where allegations that various expert witnesses were not independent – Expert Evidence Practice Note (GPN-EXPT) – where unnecessary credit findings should be avoided – where Court’s task is to reason its conclusions on the basis of contemporary materials – where allegations of partisanship liable to distract from substantive merits of scientific materials  

EVIDENCE – Jones v Dunkel inferences – where Monsanto made forensic decision not to call expert witness in conclave – relevant principles – where inference cannot be used to fill gaps in the evidence or to convert conjecture and suspicion into inference

PRACTICE AND PROCEDURE – Merck orders – identification of issues the subject of order under s 33ZB of the Federal Court of Australia Act 1976 (Cth) – whether it is necessary or appropriate to answer questions as initially framed – form of question – central common question framed and answered

Legislation:

Evidence Act 1995 (Cth) Pts 3.2, 3.3; ss 55, 56, 59, 76, 131(2)(h), 136, 140(1), 140(2), 140(2)(c), 144, 190(1), 190(1)(c), 191

Federal Court of Australia Act 1976 (Cth) Pts IVA, VB; ss 23, 33J(1), 33K, 33K(1), 37M(3), 37N(2), 37P(2), 33ZB, 33ZF

Cases cited:

ACQ Pty Ltd v Cook [2009] HCA 28; (2009) 237 CLR 656

Amaca v Booth Pty Ltd [2011] HCA 53; (2011) 246 CLR 36

Australian Securities and Investments Commission v Hellicar [2012] HCA 17; (2012) 247 CLR 345

Axon v Axon (1937) 59 CLR 395

Bellamy’s Australia Limited v Basil [2019] FCAFC 147; (2019) 372 ALR 638

Born Brands Pty Ltd v Nine Network Australia Pty Ltd [2013] NSWSC 1646

Born Brands Pty Ltd v Nine Network Australia Pty Ltd [2014] NSWCA 369; (2014) 88 NSWLR 421

Briginshaw v Briginshaw (1938) 60 CLR 336

CPB Contractors Pty Limited v Celsus Pty Limited (formerly known as SA Health Partnership Nominees Pty Ltd) (No 2) [2018] FCA 2112; (2018) 268 FCR 590

Dasreef Pty Limited v Hawchar [2011] HCA 21; (2011) 243 CLR 588

Daubert v Merrell Dow Pharmaceuticals Inc 43 F 3d 1311 (9th Cir, 1995)

Ethicon Sàrl v Gill [2018] FCAFC 137; (2018) 264 FCR 394

Foots v Southern Cross Mine Management Pty Ltd [2007] HCA 56; (2007) 234 CLR 52

Fox v Percy [2003] HCA 22; (2003) 214 CLR 118

Furber v Stacey [2005] NSWCA 242

GEC Marconi Systems Pty Ltd v BHP Information Technology Pty Ltd [2003] FCA 688

Gill v Ethicon Sàrl (No 3) [2019] FCA 587

GLJ v The Trustees of the Roman Catholic Church for the Diocese of Lismore [2023] HCA 32; (2023) 97 ALJR 857

Hockey v Fairfax Media Publications Pty Ltd (No 2) [2015] FCA 750

Jones v Dunkel (1959) 101 CLR 298

Karpik v Carnival plc (The Ruby Princess) (Evidential Ruling) [2022] FCA 1318

Kuligowski v Metrobus [2004] HCA 34; (2004) 220 CLR 363

Kumova v Davison (No 2) [2023] FCA 1

Lehrmann v Network Ten Pty Limited [2024] FCA 369

LFDB v SM No 2 [2017] FCAFC 207

Lloyd v Belconnen Lakeview Pty Ltd [2019] FCA 2177; (2019) 377 ALR 234

Lombard Insurance Co (Australia) Ltd v Pastro (1994) 175 LSJS 448

Matthews v SPI Electricity Pty Ltd (No 32) [2013] VSC 630

McNickle v Huntsman Chemical Company Australia Pty Ltd (Additional Expert Conclave) [2022] FCA 1596

McNickle v Huntsman Chemical Company Australia Pty Ltd (Assessors) [2021] FCA 780; (2021) 285 FCR 244

McNickle v Huntsman Chemical Company Australia Pty Ltd (Common Questions) [2023] FCA 662

McNickle v Huntsman Chemical Company Australia Pty Ltd (Evidentiary Ruling) [2023] FCA 1268

McNickle v Huntsman Chemical Company Australia Pty Ltd (Expert Evidence) [2021] FCA 370

McNickle v Huntsman Chemical Company Australia Pty Ltd (Hearing Vacation) [2022] FCA 133

Merck Sharp & Dohme (Australia) Pty Ltd v Peterson [2009] FCAFC 26; (2009) 355 ALR 20

Oshlack v Richmond River Council (1998) 193 CLR 72

Perera v GetSwift Limited [2018] FCA 732; (2018) 263 FCR 1

Perry v Novartis 564 F Supp 2d 452 (2008)

Qantas Airways Limited v Gama [2008] FCAFC 69; (2008) 167 FCR 537

Ramsay v Watson (1961) 108 CLR 642

Re Australian Federation of Consumer Organisations Inc v Tobacco Institute of Australia Ltd (1991) 27 FCR 149

Re Joint Eastern and Southern District Asbestos Litigation 758 F Supp 199 (1991)

Regent Holdings Pty Ltd v State of Victoria [2012] VSCA 221; (2012) 36 VR 424

Rhesa Shipping Co SA v Edmunds (The Popi M) [1985] 1 WLR 948

Roberts-Smith v Fairfax Media Publications Pty Limited (No 41) [2023] FCA 555

Russell v Australian Broadcasting Corporation (No 3) [2023] FCA 1223

Seltsam Pty Ltd v McGuiness [2000] NSWCA 29; (2000) 49 NSWLR 262

Transport Workers’ Union of Australia v Qantas Airways Limited [2021] FCA 873; (2021) 308 IR 244

Turner v MyBudget Pty Limited (No 2) [2018] FCA 1509

Williams v Lewer [1974] 2 NSWLR 91

Access to Justice, Interim Report to the Lord Chancellor on the Civil Justice System in England and Wales (HMSO, London, 1995)

Beuscher J H, “The Use of Experts by the Courts” (1941) 54(7) Harvard Law Review 1105

Brennan, T A “Causal Chains and Statistical Links: The Role of Scientific Uncertainty in Hazardous-Substance Litigation” (1988) 73 Cornell L. Rev. 469

Freckleton I, Expert Evidence: Law, Practice, Procedure and Advocacy (Thomson Reuters, 6th ed, 2021)

Geistfeld, M “Scientific Uncertainty and Causation in Tort Law” (2001) 54 Vanderbilt Law Review 1011

Korn, H L “Law, Fact, and Science in the Courts” (1966) 66 Columbia Law Review 1080

Pound N R, The Spirit of the Common Law (Marshall Jones Company, 1921)

Restatement Third, Torts: Liability for Physical and Emotional Harm §28, Comment c(3)

Justice Jonathan Beach “Scientific Evidence: A Need for Caution in Decision-Making” (2010) 42(1) Australian Journal of Forensic Sciences 49

Justice Jonathan Beach “Indeterminacy: The Uncertainty Principle of Negligence” (2005) 13 Torts Law Journal 129

Justice Jonathan Beach “Causation: The Interface Between the Scientific and Legal Methods” (2022) 49(1) University of Western Australia Law Review 113

Victorian Law Reform Commission, Civil Justice Review (Report, May 2008)

Wigmore J H, Wigmore on Evidence (3rd ed) Vol 2

Division:

General Division  

Registry:

Victoria

National Practice Area:

Commercial and Corporations

Sub-area:

Regulator and Consumer Protection

Number of paragraphs:

1203

Date of hearing:

4–7, 11–13, 18–21 and 25–28 September 2023; 2–5, 9–10 and 19 October 2023; 29 and 30 January 2024

Date of last submissions:

15 February 2024

Counsel for the applicant:

Mr A Clements KC with Ms M Szydzik SC, Ms R Howe and Ms R Singleton

Solicitor for the applicant:

Maurice Blackburn  

Counsel for the respondents:

Mr S Finch SC with Mr R Craig KC, Ms K O’Gorman SC, Mr R Ajzensztat and Mr D Habashy

Solicitor for the respondents:

Herbert Smith Freehills

ORDERS

VID 243 of 2020

BETWEEN:

KELVIN MCNICKLE

Applicant

AND:

HUNTSMAN CHEMICAL COMPANY AUSTRALIA PTY LTD (ACN 004 146 338)

First Respondent

MONSANTO AUSTRALIA PTY LTD (ACN 006 725 560)

Second Respondent

MONSANTO COMPANY

Third Respondent

ORDER MADE BY:	LEE J
DATE OF ORDER:	25 JULY 2024

THE COURT ORDERS THAT:

1.Pursuant to s 33ZF of the Federal Court of Australia Act 1976 (Cth) (FCA Act) (and notwithstanding that no application has been made by the representative party in accordance with s 33K(1) of the FCA Act) leave be granted for the group membership to be amended nunc pro tunc so that it is defined as it is in the last version of the statement of claim filed prior to the fixing of the date for opt out (being the second further amended statement of claim filed on 19 October 2020).

2.Pursuant to ss 33ZF and 37P(2) of the FCA Act, the following question, common to the claim of the applicant and all group members (Central Common Question) be answered separately and before any other question or issue in the proceeding:

Throughout the period between July 1976 and 19 October 2020 (relevant period), did or could use of and/or exposure to the herbicide product or products, which contained glyphosate and were branded as “Roundup”, or which contained glyphosate and were otherwise branded with the name “Monsanto” (Roundup Products) increase an individual’s risk of developing non-Hodgkin lymphoma (NHL); and/or cause an individual to develop NHL?

3.The Central Common Question be answered as follows:   

It is not proven in this proceeding on the balance of probabilities (in accordance with s 140(1) of the Evidence Act 1995 (Cth)), that throughout the relevant period, use of and/or exposure to Roundup Products increased an individual’s risk of developing NHL; and/or caused an individual to develop NHL.

4.The proceeding be dismissed upon finalisation of any issues relating to costs.

5.Any further evidence as to costs (as contemplated in the reasons (at [1202])) and any submissions as to costs be filed by 12 noon on 30 July 2024.

6.For the purposes of s 33ZB of the FCA Act, Orders 1 to 3 above affect (and hence bind) all current group members in the class action (being those named in the second further amended statement of claim who did not opt out).

7.The proceeding be adjourned for the parties to be heard on the appropriate costs orders in conformity with these reasons at 10:15am on 31 July 2024.

AND THE COURT NOTES THAT:

8.The leave granted after the date of opt out had passed to amend the statement of claim so as to file the third and fourth statement of claim did not include leave being granted on application made by the representative party to alter the description of the group pursuant to s 33K(1) of the FCA Act.

Note:   Entry of orders is dealt with in Rule 39.32 of the Federal Court Rules 2011.

REASONS FOR JUDGMENT

A        INTRODUCTION	[1]
B        PROPER APPROACH TO FACT FINDING AND THE LEGAL METHOD	[11]
B.1      Relevant Procedural Background and the Role of the Assessor	[11]
B.2      Expert Assistance Generally	[25]
B.3      Fact Finding Generally in this Case	[36]
B.4      The “Weight of Evidence” and the Scientific Studies	[49]
B.5      The Scientific and Legal Method and Causation	[66]
C        FACTUAL BACKGROUND	[73]
C.1     Monsanto	[75]
C.2     Glyphosate and Roundup	[82]
C.3     Roundup Herbicide and Roundup Biactive in Australia	[85]
D        STRUCTURE OF CONSIDERATION OF THE SCIENTIFIC EVIDENCE	[90]
E        THE FIRST STREAM: EPIDEMIOLOGY (CONCLAVES C, D AND E)	[93]
E.1      Nature of the Evidence	[93]
I          Introduction	[93]
II        Methodology	[103]
III       Terminology	[109]
Relative risk, odds ratios and confidence	[109]
Types of studies	[117]
1.        Cohort studies	[118]
2.        Case-control studies	[120]
3.        Pooled studies	[123]
Other relevant terminology	[124]
E.2      The Relevant Experts	[126]
I          Overview	[126]
II        Credit Findings	[131]
Professor Checkoway	[131]
Associate Professor Harris	[142]
Professor Gordon	[155]
Dr Crump	[162]
E.3      The Epidemiological and Biostatistical Evidence	[169]
I          Primary case-control studies	[173]
Hardell and Eriksson (1999)	[173]
McDuffie (2001)	[188]
Eriksson (2008)	[199]
Meloni (2021)	[212]
II        Pooled case-control studies	[228]
Hardell (2002)	[231]
De Roos (2003)	[238]
Pahwa (2019)	[256]
Lee (2004)	[270]
Hardell (2023)	[278]
De Roos (2022)	[289]
III       Other studies	[294]
Orsi (2009)	[296]
Hohenadel (2011)	[299]
Cantor (1992)	[302]
Leon (2019)	[306]
IV       Agricultural Health Study – De Roos (2005) and Andreotti (2018)	[309]
Introduction	[309]
The AHS papers	[317]
Mr McNickle’s contentions	[326]
1.        Exposure misclassification	[330]
2.        Multiple imputation	[348]
3.        No positive association	[358]
Consideration	[362]
Conclusion	[370]
V         Conclave E – Recall and selection bias	[371]
Introduction	[371]
Mr McNickle’s contentions	[375]
Consideration	[387]
Conclusion	[399]
VI       Conclusion	[400]
F        THE SECOND STREAM: ANIMAL STUDIES (CONCLAVES H, I AND J)	[410]
F.1      Nature of the Evidence	[410]
I          Introduction	[410]
II        Methodology	[414]
F.2      The Relevant Experts	[429]
I          Overview	[429]
II        Credit Findings	[433]
Dr Bayard	[433]
Dr Juberg	[446]
Professor Prince	[463]
Dr Crump	[468]
F.3      The Animal Studies Evidence	[470]
I          Dr Bayard’s alleged flawed statistical approach	[475]
Multiple comparisons	[477]
Pooling	[503]
II        Dr Bayard’s alleged limited expertise	[508]
III       Animal bioassays	[517]
Mice	[519]
1.        Malignant lymphomas	[521]
2.        Hemangiomas and hemangiosarcomas	[540]
3.        Kidney tumours	[550]
4.        Lung adenocarcinomas	[565]
Rats	[573]
1.        Thyroid C-cell carcinomas and adenomas	[576]
2.        Liver hepatocellular adenomas and carcinomas in male rats	[595]
3.        Kidney adenomas	[605]
4.        Other tumours	[608]
IV       Wang (2019)	[615]
V         Conclave I and the relevance of animal studies to humans	[623]
VI       Conclusion	[634]
G        THE THIRD STREAM: MECHANISTIC EVIDENCE (CONCLAVES A, B AND G)	[641]
G.1     Nature of the Evidence	[641]
I          Introduction	[641]
II        Methodology	[643]
Cancer biology and NHL	[643]
Key characteristics	[656]
Mechanistic toxicology	[673]
G.2     The Relevant Experts	[684]
I          Overview	[684]
II        Credit findings	[687]
Dr Flecknoe-Brown	[687]
Professor Prince	[705]
Professor Smith	[723]
Dr Juberg	[750]
G.3     The Mechanistic Evidence	[760]
I          KC2 (genotoxicity)	[766]
The in vitro studies	[770]
1.          The regulatory studies	[771]
2.        The in vitro studies	[778]
Santovito (2018)	[779]
Wozniak (2018)	[789]
Kwiatkowska (2017)	[796]
Nagy (2021)	[801]
Mladinic (2009)	[808]
Alvarez-Moya (2023)	[813]
Roma (2023)	[819]
Dr Flecknoe-Brown (Conclave A)	[826]
Conclusion	[832]
The in vivo studies	[834]
1.        Paz-y-Mino (2007)	[835]
2.        Bolognesi (2009)	[847]
3.        Paz-y-Mino (2011)	[856]
4.        Chang and Andreotti (2023)	[863]
Conclusion	[871]
The real world significance of the in vitro studies	[874]
1.        Introduction	[874]
2.        Zouaoui (2013) and Kwiatkowska (2020)	[880]
3.        Consideration	[890]
4.        Conclusion	[901]
Conclusion	[903]
II        KC4 (induces epigenetic alterations)	[918]
Relevant studies	[918]
1.        Rossetti (2021)	[918]
2.        Lucia (2022)	[921]
Consideration	[924]
Conclusion	[935]
III       KC5 (induces oxidative stress)	[936]
IV       KC6 (induces chronic inflammation)	[950]
Relevant studies	[953]
Consideration	[960]
Conclusion	[965]
V         KC8 (modulates receptor-mediated effects)	[966]
VI       Hypothesised biologically plausible mechanisms	[976]
First hypothesised mechanism	[979]
Second hypothesised mechanism	[987]
1.        Introduction	[987]
2.        Wang (2019)	[989]
3.        Suárez-Larois (2017)	[1021]
4.        Conclusion	[1024]
Third hypothesised mechanism	[1025]
Fourth hypothesised mechanism	[1029]
Fifth hypothesised mechanism	[1038]
VII      Conclusion	[1042]
H        CONCLAVE F (ABSORPTION AND EXPOSURE) AND DOSE	[1048]
H.1     Nature of the Evidence	[1048]
I          Introduction	[1048]
II        Methodology	[1053]
Absorption	[1053]
Distribution	[1056]
Metabolism	[1058]
Excretion	[1059]
H.2     The Absorption and Exposure Evidence	[1062]
H.3     Independence of Dr Sawyer	[1067]
I          Introduction	[1067]
II        Background	[1072]
III       Credit Findings	[1077]
IV       Consideration	[1088]
V         Conclusion	[1092]
H.4     Dose	[1095]
I          Introduction	[1095]
II        Mr McNickle’s submissions	[1101]
III       Consideration	[1110]
IV       Alternative dose proposition	[1120]
V         Conclusion	[1127]
H.5     Jones v Dunkel	[1128]
I         DRAWING THE THREADS TOGETHER	[1133]
J         CONCLUSION AND ORDERS	[1147]
J.1      The Merck Order Questions	[1147]
J.2      The Form of Questions	[1153]
J.3      The Appropriate Answer	[1163]
J.4      Group Membership	[1167]
J.5      Costs	[1177]
I          The Nature of the Discretion	[1177]
II        The Parties’ Conduct and the Overarching Purpose	[1182]
III       Public Interest Litigation	[1197]
IV       The Position of Mr McNickle	[1199]
V         Future Submissions	[1202]
J.6      Orders	[1203]

LEE J:

A        INTRODUCTION

1. Mr Kelvin McNickle lived what might be thought by a young lad to be an Australian idyll. Born in September 1982, he grew up on a rural property on the beautiful Mid North Coast of New South Wales. His father operated a vegetation management business, and, during school holidays, Mr McNickle assisted his dad clearing vegetation.

2. But this was no passing stage. After completing school, Mr McNickle began working full-time for his father. No doubt it was hard work. Almost every working day, Mr McNickle and his father sprayed a weedkiller called “Roundup” on weeds and other unwanted growth using a knapsack sprayer or a drenching unit. As might be expected, drips or leaks of the Roundup soaked through his clothes and onto his skin. Mist or spray would go all over the place and get into his eyes and onto his face and fumes were inhaled.

1. Some years later, after marrying, Mr and Mrs McNickle left Coffs Harbour for Darwin. Mr McNickle is an apparently devout man. The Apostle Paul said: “let him do hard work, doing with his hands what is good work” (Ephesians 4:28). Apparently motivated by the perceived moral value of good and hard manual work, Mr McNickle volunteered to maintain the lawns and gardens of the Kingdom Hall of Jehovah’s Witnesses. His evidence is that in doing so he used Roundup.

2. In May 2018, Mr McNickle was diagnosed with a type of cancer known as non-Hodgkin lymphoma (NHL). Just over a year later, he was informed by a haematologist that he was in ongoing remission. Unfortunately, about six weeks prior to the commencement of the initial trial, his NHL recurred. As a result, Mr McNickle was unable to attend the hearing.

3. I have introduced Mr McNickle, who is the lead applicant in this class action. But this judgment is not, at least directly, about him. This is not the occasion to deal with questions of individual causation.    

4. Rather, when this initial trial was fixed, it was contemplated I would determine a set of common questions relating to Mr McNickle’s contention that during a period from July 1976 until an end date defined at the time I made the orders (Relevant Period), glyphosate, glyphosate-based formulations (GBFs), and Roundup Herbicide and Roundup Biactive (Roundup Products) (which have glyphosate as a component), were carcinogenic to humans. The respondents (collectively, Monsanto) reject this contention completely.

5. The common questions were identified in orders made by consent following a case management hearing in April 2023. I set out the terms of those questions in Section J below and why I do not propose to answer them in the form initially proposed.

6. As it happened, issues were refined during the initial trial. In the end, for reasons that will be explained, the determinative question is whether Mr McNickle has discharged his legal onus of proving on the evidence adduced that the use of and/or exposure to Roundup Products can increase an individual’s risk of developing NHL or cause an individual to develop NHL. I will call this, unimaginatively, the central issue.  

7. There is no need for me to explain the issues in this proceeding any further. This is a case which has been mired in interlocutory skirmishes and, save for matters which I will mention shortly, I do not propose to rehearse that background here. There have been several judgments of the Court which identify the central issue, the utility of determining it initially, and set out what might be described as the superstructure of the proceeding: see McNickle v Huntsman Chemical Company Australia Pty Ltd (Common Questions) [2023] FCA 662; McNickle v Huntsman Chemical Company Australia Pty Ltd (Additional Expert Conclave) [2022] FCA 1596; McNickle v Huntsman Chemical Company Australia Pty Ltd (Hearing Vacation) [2022] FCA 133; McNickle v Huntsman Chemical Company Australia Pty Ltd (Assessors) [2021] FCA 780; (2021) 285 FCR 244; McNickle v Huntsman Chemical Company Australia Pty Ltd (Expert Evidence) [2021] FCA 370.

8. I will return to the common questions and the utility of answering them below, but at the outset, it is worth saying something about the proper approach to fact finding and the legal method.  

   B        PROPER APPROACH TO FACT FINDING AND THE LEGAL METHOD

   B.1     Relevant Procedural Background and the Role of the Assessor

9. Enough has been said to indicate that this case presented significant challenges in fact finding. In the light of the central issue, it is an exercise in considerable understatement to describe the expert opinion evidence and underlying scientific material as dense, complex, and voluminous.

10. As I recorded in McNickle v Huntsman Chemical Company Australia Pty Ltd (Expert Evidence) (at [4]), when this matter first came before me for case management, I indicated my preliminary view was that the Court should adopt the process of appointing referees in a number of scientific disciplines to inquire into, and to report upon, questions concerning the characteristics of the Roundup Products and their alleged carcinogenic effects.

11. The parties vigorously opposed this proposed course. 

12. An array of arguments were marshalled and then deployed to oppose referees, including that to proceed along these lines would be constitutionally invalid (an argument wholly devoid of merit and decisively rejected in CPB Contractors Pty Limited v Celsus Pty Limited (formerly known as SA Health Partnership Nominees Pty Ltd) (No 2) [2018] FCA 2112; (2018) 268 FCR 590).

13. With that said, there was one contention made by Mr McNickle that persuaded me (reluctantly) to adopt the course of allowing expert evidence to be deployed in a more traditional way. Mr McNickle’s argument was that prevailing scientific opinion concerning the alleged carcinogenic effects of Roundup had been skewed or somehow manipulated by Monsanto. I concluded that the risk of referees assessing material said to be infected by this allegation of scientific manipulation created a prospect of complicating the inquiry process. Further, the allegation was one of some seriousness and is of a character that meant it should be resolved in open court.

14. Shortly after rejecting the process of references, and faced with what I knew was going to be a raft of highly complex opinion evidence, I noted in McNickle v Huntsman Chemical Company Australia Pty Ltd (Assessors) (at 246 [7]) that almost 100 years ago, Dean Roscoe Pound in his famous work The Spirit of the Common Law (Marshall Jones Company, 1921) (at 214–215) suggested that steps be taken to equip courts with research staffs comparable to those employed by administrative agencies in order to provide judges with specialised investigations, information, and advice. Yet, despite what some have described as centuries of legal history dotted with proof to the contrary, the traditional common law confidence that courts can handle any dispute without assistance has long persisted: see Beuscher, J H “The Use of Experts by the Courts” (1941) 54(7) Harvard Law Review 1105.  

15. I also observed that in saying this, there have been attempts by courts to provide, in different ways, expert aid to determination. There are three obvious examples: (1) special juries; (2) referees (properly seen as a version of a special jury); and (3) assessors.  

16. Having determined that there was to be no form of special jury, I went on to explain (at 247–253 [11]–[38]) the long history of appointment of assessors tracing back to the sixteenth century; the basis of the Court’s power to order an assessor; and the scope of the role of an assessor. In this last regard, I referred to the judgment of J Forrest J in Matthews v SPI Electricity Pty Ltd (No 32) [2013] VSC 630 and relevantly noted (at 252 [37]) that the role of any assessor would be to assist the judge, but any decision would be that of the tribunal of fact, being the judge alone. Conscious of the need to avoid any procedural unfairness and define the role of the assessor precisely, I explained, with the consent of the parties, that if an assessor was appointed:

    (1)the assessor would sit with me during the concurrent evidence sessions, and could question the experts (or counsel) but such questioning would be limited to clarification of the evidence: that is, where the evidence given is perceived by the assessor or me to be ambiguous, unclear or incomplete;

    (2)I would consult with the assessor if I found a point of evidence unclear;

    (3)I would consult with the assessor in chambers on matters raised by the experts in their oral evidence and in their individual and joint reports, including advice as to any questions the assessor thinks should be asked;

    (4)I would seek the guidance of the assessor on scientific and technical matters upon which I lack the requisite knowledge to understand without qualified assistance;

    (5)if the assessor raised a theory or opinion that had not previously been identified by the parties, I would disclose this fact and discuss it with counsel; and

    (6)the assessor would, from time to time, provide me with advice on matters over which there is dispute between the experts but that such advice is not binding and the determination of a particular issue rests with me.

17. After receiving suggestions from the parties, on 16 August 2021, I made an order pursuant to s 23 and/or s 33ZF of the Federal Court of Australia Act 1976 (Cth) (FCA Act) that Professor Sir John Stewart Savill BA, MBChB, PhD, FRCP, FRCPE, FRCSEd(Hon), FRCPCH(Hon), FASN, FRSE, FMedSci, FAHMS, FRS be appointed as Assessor. In dealing with the Assessor, care has been taken not to depart from the six points identified above and, more generally, to ensure the findings I will make are mine and mine alone, notwithstanding the assistance received from the Assessor in understanding the complex evidence adduced in this case. For completeness, in my dealings with the Assessor, I record he did not raise a theory or opinion privately that had not previously been identified by the parties through their retained experts (see [18(5)] above).

18. Further, and again with the express consent of the parties, when I had completed an initial draft of these reasons without the assistance of the Assessor, I provided a copy of the draft reasons to the Assessor in order to ensure, as it was explained to the parties, that I had not committed any scientific “solecisms” caused by any fundamental misapprehension of the scientific evidence. I only adopted this course after obtaining confirmation by both parties that they regarded any communication between my chambers and the Assessor as being equivalent to an “in chambers” communication and hence wholly confidential, and that they disavowed seeking a copy of any communication of any type between my chambers and the Assessor, including on any appeal.  

19. I should record that these reasons would have been significantly impoverished, and the delivery of judgment delayed, without the assistance of Sir John. It would have been much more difficult to identify and then understand the nuances of the opinion evidence without an Assessor, particularly one like Sir John with his happy facility for explaining complex scientific concepts in a way that can be comprehended by an interested layman. On behalf of myself and the Court, I express my appreciation to Sir John for his great assistance.

20. But this was not all the assistance I received.

21. As is well-known, Australia’s version of concurrent evidence typically involves two interrelated processes. First, there is a pre-trial joint expert conferral or conclave phase during which the parties’ experts meet to clarify the areas of agreement and/or disagreement to produce a joint report. The second part of the process is the giving of concurrent evidence.

22. In the present case, the first of these processes became particularly important and I appointed an experienced facilitator and independent barrister, Mr Edward Cowpe, who maintained the integrity of the conclave process and generally ensured the conclave process proceeded effectively and the various joint reports emerging from any conclave were expressed clearly, admissibly, and in a form that would best assist the Court. I am also grateful for his assistance.

    B.2     Expert Assistance Generally

23. One of the reasons I was reluctant to abandon a reference and proceed in the time-honoured way in this case is because of my concerns in navigating potentially polarised expert views. In McNickle v Huntsman Chemical Company Australia Pty Ltd (Expert Evidence), I observed (at [1]) that controversy as to how opinion evidence is often deployed by litigants is not new, and some best placed to judge have expressed their criticism with some force. As Lord Woolf MR observed in the report, Access to Justice, Interim Report to the Lord Chancellor on the Civil Justice System in England and Wales (HMSO, London, 1995) (at 183):

    Expert witnesses used to be genuinely independent experts. Men of outstanding eminence in their field. Today they are, in practice, hired guns. There is a new breed of litigation hangers-on, whose main expertise is to craft reports which will conceal anything that might be to the disadvantage of their clients.

24. This might be thought as putting the point too highly in this country, but reflecting similar concerns, in Australia there has been extensive law reform discussion concerning the role, deficiencies and remedial responses to perceived difficulties with expert evidence. Most relevantly for present purposes, in the Victorian Law Reform Commission Civil Justice Review (2008) (at 484), reference was made by the Victorian Law Reform Commission (VLRC) to the notion of “adversarial bias”. That conception falls into three varieties:

    (1)deliberate partisanship – an expert who deliberately tailors evidence to support the retaining client;

    (2)unconscious partisanship – the expert does not intentionally mislead the Court, but is influenced by the situation to give evidence in a way that supports the retaining client; and

    (3)selection bias – litigants choose as their expert witnesses persons whose views are known to support their case.  

25. Given the evident polarisation of scientific views and the possibility of unconscious partisanship and selection bias, from the moment it became apparent there was to be no special jury, I explained I considered it was likely I would be best assisted if the parties could seek out and retain experts in various disciplines who might be described as “cleanskins”: that is, experts who were not wedded to any particular concluded view and who could approach the questions posed informed by the Expert Evidence Practice Note (GPN-EXPT) (Expert Evidence Practice Note) (see also Harmonised Expert Witness Practice Note (Expert Witness Code of Conduct)) and without feeling the need to justify earlier expressed concluded views.  

26. My suggestion was met with a mixed response. Indeed, the implicit (and sometimes explicit) suggestion made in Mr McNickle’s submissions was that my suggestions or concerns in this regard were wrongheaded. I will deal with each of the experts below in detail, but it is worthwhile making four related and general points concerning the criticism that entreating the parties to engage experts who did not hold fixed views was somehow heterodox.

27. First, I accept we are dealing with abstruse areas of specialised knowledge. The relatively limited pool of qualified persons to give opinion evidence and the polarisation of scientific view as to the central issue (and matters adjectival to the central issue) would have perhaps caused some difficulty in identifying appropriate experts; a fortiori experts who had not already formed views.

28. Secondly, notwithstanding the first point, I did not come down in the last shower: like any judge who has been an experienced barrister, I understand the forensic attractions that lead to what the VLRC described as selection bias. I hope I am not being unfair in remarking that my distinct impression throughout the whole of the interlocutory stages of this case was that Mr McNickle’s legal team, on one level understandably, resisted any independent referee process and eschewed engaging “cleanskins” in some areas of specialised knowledge because they felt more comfortable in the traditional approach of fixing upon experts they considered would likely reflect their case theory.   

29. Thirdly, in the end, the role of the advocates acting for Mr McNickle was to persuade. As a general proposition, it stands to reason that the more objective, dispassionate, and disinterested an expert is, the more the opinions of the expert are likely to be perceived by the tribunal of fact as being: (a) unaffected by unconscious bias; and (b) persuasive. Of course, I stress this does not mean one assesses the admissibility and then reliability of an expert opinion by reference to some a priori view that it would be better to get a “cleanskin”. Disputed questions of fact must be decided according to the evidence that the parties adduce, not according to some speculation about what other evidence might possibly have been led: Australian Securities and Investments Commission v Hellicar [2012] HCA 17; (2012) 247 CLR 345 (at 412–413 [165]–[167] per French CJ, Gummow, Hayne, Crennan, Kiefel and Bell JJ).

30. Fourthly, Mr McNickle in his submissions repeatedly made the point that the mere fact that an expert witness has a pre-existing opinion on a key issue is an infirm basis for rejecting the expert’s opinion evidence “or placing less weight on it”. He cited Born Brands Pty Ltd v Nine Network Australia Pty Ltd [2013] NSWSC 1646 (and on appeal Born Brands Pty Ltd v Nine Network Australia Pty Ltd [2014] NSWCA 369; (2014) 88 NSWLR 421), to assert a pre-existing opinion does not mean an expert is not open-minded or biased but can rather indicate an expert has come to a considered opinion after years of relevant experience. In other words, preconceived principles can inform opinion evidence, if the principles are scientifically based.

31. Of course, Born Brands was dealing with admissibility and not weight: two different concepts. But I accept the underlying point made by Mr McNickle has real merit. The mere fact that the opinion an expert has expressed in this case as to whether, for example, glyphosate is carcinogenic, is the same as the opinion previously expressed on the same issue in other cases against Monsanto in the United States, is neither a complete nor satisfactory basis for the Court to reduce the weight to be afforded to the expert’s opinion evidence; let alone reject its admission. But depending upon the witness and the characteristics of the evidence, it can be part of the constellation of matters that may be relevant to assessing the persuasiveness of the evidence (depending upon how the evidence of the witness is challenged and whether procedural fairness has been provided to the witness to respond to whether the witness is approaching matters with an open mind or is fixed upon defending a pre-existing opinion). 

32. In the end, each witness and the opinions expressed by the witness must be judged on their overall substantive merits, not by reference to the preconceived notion that it might have been thought more compelling to adduce evidence from an expert who first formed opinions through the prism of adherence to the Expert Witness Code of Conduct and approached the task without having formed pre-existing views. Recognising the merit of Mr McNickle’s submissions, I have been careful to approach consideration of the expert evidence in this way.

33. As it happened, there was extensive cross-examination by each party alleging partisanship by some experts. Having foreshadowed this as a potential issue early on, I facilitated the experts engaging usefully with each other under the supervision of an independent barrister and settled the questions to be asked of them following extensive argument. I will deal below with the competing contentions of unconscious bias and partisanship; but I will do so as is appropriate: not by reference to sweeping generalisations; but rather informed by my impressions of the relevant expert’s overall demeanour and credit and, most importantly, the perceived underlying scientific merit of the evidence given by the expert.

    B.3     Fact Finding Generally in this Case

34. No lay witnesses were called. As noted above, the spectre of an attack on Monsanto for manipulating the science (decisive in my reasoning to not appoint referees) proved a damp squib, and although it was raised by one expert witness called by Mr McNickle directly, minimal evidence was adduced at this hearing directed to this allegation.   

35. Leaving aside agreed facts, all the evidence was either expert or documentary. In respect of the latter, what might be described as a tsunami of scientific articles, studies and other materials was placed before me. I deal below with how, by the consent of the parties, I have approached the admissibility and weight to be given to the representations contained in that material.

1. But it must be emphasised at the outset that the Court’s task in determining the central issue is not an exercise of choice between the evidence of the respective parties’ expert witnesses. 

2. It is trite to observe that in accordance with s 140(1) of the Evidence Act 1995 (Cth) (EA), the relevant standard of proof for the Court to apply is on the balance of probabilities, and that the applicant bears the onus of establishing that the evidence satisfies that test.   

3. I will return to differences between the approach of both science and the law to causation enquiry below, but as I explained in some detail in Transport Workers’ Union of Australia v Qantas Airways Limited [2021] FCA 873; (2021) 308 IR 244 (at 324–325 [284]–[288]), when the law requires proof of any fact, the tribunal of fact must feel actual persuasion as to its occurrence or existence before it can be found. A party bearing the onus will not succeed unless the whole of the evidence establishes a reasonable satisfaction on the preponderance of probabilities such as to sustain the relevant issue: Axon v Axon (1937) 59 CLR 395 (at 403 per Dixon J). In this way, the facts proved must form a reasonable basis for a definite conclusion affirmatively drawn of the truth of which the tribunal of fact may reasonably be satisfied: Jones v Dunkel (1959) 101 CLR 298 (at 305).

4. As I have also noted, despite the criticisms referred to in Transport Workers’ Union of Australia v Qantas (at 325 [286]), this approach is well entrenched and unquestionably represents the current state of the law. This was reinforced by Kiefel CJ, Gageler and Jagot JJ in GLJ v The Trustees of the Roman Catholic Church for the Diocese of Lismore [2023] HCA 32; (2023) 97 ALJR 857, where their Honours observed (at 875 [60]):

   [60]… “[t]o satisfy an onus of proof on the balance of probabilities is not simply a matter of asking whether the evidence supporting that conclusion has greater weight than any opposing evidence ... It is perfectly possible for there to be a scrap of evidence that favours one contention, and no countervailing evidence, but for the judge to not regard the scrap of evidence as enough to persuade him or her that the contention is correct.” The evidence must “give rise to a reasonable and definite inference” to enable a factual finding to be made; mere conjecture based on "conflicting inferences of equal degrees of probability" is insufficient. As Dixon CJ said in Jones v Dunkel, the law:

   does not authorise a court to choose between guesses, where the possibilities are not unlimited, on the ground that one guess seems more likely than another or the others. The facts proved must form a reasonable basis for a definite conclusion affirmatively drawn of the truth of which the tribunal of fact may reasonably be satisfied.

   (Footnotes omitted)

5. Before passing from these self-directions, it is useful to make two further points.  

6. First, in final submissions, Monsanto asserted that the seriousness of Mr McNickle’s allegation, being an allegation that Roundup Products cause NHL, must be considered in deciding whether Mr McNickle has discharged the onus of proof (referring to s 140(2) EA and Briginshaw v Briginshaw (1938) 60 CLR 336 (at 361–362 per Dixon J)).

7. With respect, at best, this is a distraction. Section 140(2) needs to be faithfully applied, but in considering “the gravity of the matters alleged” in s 140(2)(c), the primary focus is upon the nature of the factual allegations in the case (although the consequences of any finding is also a relevant matter). This makes sense when one considers the focus on the gravity of the finding is linked to the notion that the Court takes into account the inherent unlikelihood of alleged conduct, and common law principles concerning weighing evidence: see Qantas Airways Limited v Gama [2008] FCAFC 69; (2008) 167 FCR 537 (at 576 [137]–[138] per Branson J). There is no suggestion of knowing wrongdoing or of other intuitively unlikely conduct; here I am dealing with a question of general legal causation based upon the scientific evidence adduced and Monsanto’s submission is less than helpful when the rationale of Briginshaw is understood.

8. Secondly, on the one hand, Mr McNickle asserts the evidence of general factual causation is pellucid; and on the other hand, Monsanto denies its existence completely. Often these submissions elided the concepts of legal and scientific causation, but it suffices to note for present purposes that the principal case theories adopted are polar opposites. But, as Besanko J observed in Roberts-Smith v Fairfax Media Publications Pty Limited (No 41) [2023] FCA 555 (at [117]–[118]), there is no doubt that a court is not bound to accept the case of one or other of the parties.

9. In this regard, his Honour referred to the “great clarity” of the speech of Lord Brandon of Oakbrook in Rhesa Shipping Co SA v Edmunds (The Popi M) [1985] 1 WLR 948, where his Lordship said the following (at 955):

   My Lords, the late Sir Arthur Conan Doyle in his book The Sign of Four, describes his hero, Mr Sherlock Holmes, as saying to the latter's friend, Dr. Watson: “How often have I said to you that, when you have eliminated the impossible, whatever remains, however improbable, must be the truth?” …

   In my view there are three reasons why it is inappropriate to apply the dictum of Mr Sherlock Holmes, to which I have just referred, to the process of fact-finding which a judge of first instance has to perform at the conclusion of a case of the kind here concerned.

   The first reason is one which I have already sought to emphasise as being of great importance, namely, that the judge is not bound always to make a finding one way or the other with regard to the facts averred by the parties. He has open to him the third alternative of saying that the party on whom the burden of proof lies in relation to any averment made by him has failed to discharge that burden. No judge likes to decide cases on burden of proof if he can legitimately avoid having to do so. There are cases, however, in which, owing to the unsatisfactory state of the evidence or otherwise, deciding on the burden of proof is the only just course for him to take….

10. This reflects the legal reality of how the onus of proof works and the logical reality that a failure to prove a state of affairs does not necessarily mean the state of affairs does not exist. Hence, an acquittal by a jury of a person accused of a criminal charge does not amount to a declaration by the jury of the accused’s innocence (although this state of affairs is taken at law to be the case by reason of the operation of a different presumption of the criminal law).  

11. I come back to this important notion in Section J when I deal with the appropriateness of answering the common questions as initially framed.   

    B.4     The “Weight of Evidence” and the Scientific Studies

12. With metronomic regularity, Mr McNickle asserted that in determining whether glyphosate and/or GBFs can cause or increase the risk of NHL, “it is appropriate to take a weight of evidence approach” to the following three streams of scientific evidence, examined in detail below, being: (1) epidemiological studies of the effects of glyphosate and/or GBFs on humans; (2) long-term studies on experimental animals relating to whether glyphosate and/or GBFs can cause cancer in animals; and (3) mechanistic evidence, that is, evidence relating to whether there is a biologically plausible mechanism by which glyphosate and/or GBFs can induce cancer in humans.

13. The concept of the “weight of evidence” is a common term in published scientific literature, most often seen in the context of risk assessment. The concept is referred to in various conclave reports: Conclave A Joint Report (JRA) (at [19], [125]); Conclave B Joint Report (JRB) (at [72]–[78]); Conclave G Joint Report (JRG) (at [35], [39], [49]–[50], [102]); and Conclave I Joint Report  (JRI) (at [13]). But this does not mean the concept is entirely clear; nor does its mere recitation explain how it is called in aid by Mr McNickle.   

14. Dr Douglas Weed, an epidemiologist (who was the former Chief, Office of Preventive Oncology, and the Director of the Cancer Prevention Fellowship Program[me] at the National Cancer Institute (NCI), being an agency of the Executive Government of the United States) has published a systematic review of the scientific literature to characterise the “weight of evidence” concept. As I noted during final oral submissions, it is used apparently in three senses:

    (1)metaphorical, where weight of evidence refers to a collection of studies or to an unspecified methodological approach;

    (2)methodological, where weight of evidence points to established interpretative methodologies or where weight of evidence means that “all” rather than some subset of the evidence is examined, or rarely, where it points to methods using quantitative weights for evidence; and  

    (3)theoretical, where weight of evidence serves as a label for a conceptual framework. Dr Weed identifies several problems associated with the frequent lack of definition of the term and a lack of consensus about its meaning: see Weed, D L “Weight of Evidence: A Review of Concept and Methods” (2005) 25(6) Risk Analysis 1545 (at 1545–57).

15. Following clarification made during the course of oral submissions, it became apparent that the parties employed the term in the second, methodological sense in that the principled approach requires attention to be given to “all”, rather than some subset, of the scientific evidence.

16. Further, it became apparent during oral submissions that the assessment is suggested to take place on two different levels.

17. First, as is evident from the various conclave reports within each scientific stream, it is common ground that an expert cannot cherry-pick from the peer-reviewed studies those that support the expert’s opinion, but rather carry out a comprehensive assessment of the weight of evidence of regulatory and peer-reviewed studies within that stream of scientific evidence.

18. Secondly, Mr McNickle submits, and Monsanto expressly accepts, that the Court needs to carry out a weight of evidence analysis to “weigh up the three streams of evidence”.

19. I understand the way the concept is used within each scientific stream and its utility, but to the extent the concept is used in this second way, it needs to be approached with some caution and its discrete analytical utility is elusive. As I will explain, scientific causation and the legal question of general causation are different notions and 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).   

20. Subject to one matter to which I will come, evidence was admitted because the parties, through their non-objection to its adduction, accepted it was relevant because, if it was accepted, it could rationally affect (directly or indirectly) the assessment of the probability of the existence of a fact in issue at the initial trial: see ss 55 and 56 EA. As explained above, in assessing whether I have reached the level of reasonable satisfaction on the preponderance of probabilities such as to sustain Mr McNickle’s case on the central issue, statute and authority demand I have regard, and only have regard, to the whole of the evidence admitted. My reasoning, inferential and otherwise, can only be based on the evidence admitted and must take account of the entirety of the evidence in reaching conclusions.  

21. There is one qualification to what I have said about admissibility. I referred earlier to the scientific articles and studies placed before the Court. There is a lot of law about how a court is to deal with representations of opinion made in such material. As Heydon J explained in Dasreef Pty Limited v Hawchar [2011] HCA 21; (2011) 243 CLR 588 (at 615 [69], 631–632 [110]), there has long existed a qualification to the proof of assumption rule (under a common law exception to the hearsay rule), that experts may give evidence of hearsay matters which go to demonstrate their expertise – what is said in the writings of others in the relevant area of expertise they have read as a basis for their opinion, or what has been said to them in discussions they have had with colleagues and taken into account.

22. In short, experts are generally entitled to rely upon publications and material produced by others in the area in which they have expertise as a basis for their opinions and may give evidence of fact which is based on them. As is explained in Wigmore on Evidence (3rd ed) Vol 2 (at 784–785):   

    The data of every science are enormous in scope and variety. No one professional man can know from personal observation more than a minute fraction of the data which he must every day treat as working truths. Hence a reliance on the reported data of fellow-scientists, learned by perusing their reports in books and journals. The law must and does accept this kind of knowledge from scientific men. On the one hand, a mere layman, who comes to court and alleges a fact which he has learned only by reading a medical or a mathematical book, cannot be heard. But, on the other hand, to reject a professional physician or mathematician because the fact or some facts to which he testifies are known to him only upon the authority of others would be to ignore the accepted methods of professional work and to insist on finical and impossible standards. Yet it is not easy to express in usable form that element of professional competency which distinguishes the latter case from the former. In general, the considerations which define the latter are (a) a professional experience, giving the witness a knowledge of the trustworthy authorities and the proper source of information, (b) an extent of personal observation on the general subject, enabling him to estimate the general plausibility, or probability of soundness, of the views expressed, and (c) the impossibility of obtaining information on the particular technical detail except through reported data in part or entirely.

23. In Karpik v Carnival plc (The Ruby Princess) (Evidential Ruling) [2022] FCA 1318, Stewart J addressed this topic and, after referring to the above extract, noted (at [5]) that “[i]t can thus be seen that expert reliance on such learned publications is not excluded by the rule against hearsay (s 59) or the opinion rule (s 76)”. His Honour was dealing with a case, like here, where the expert reports were replete with references to scientific articles. Having noted the need for caution in making an order under s 136 EA limiting the use of such materials, such a limitation order was not made generally but was made in relation to an article which was: (1) not referred to in any of the expert reports; (2) only referred to in cross-examination; and (3) “deal[t] with an area of expertise not possessed by any of the many expert witnesses in the case” and therefore “there is no other evidence against which to compare, or weigh, the opinions and conclusions expressed in the article”, leaving the Court “in an invidious position with regard to trying to assess the weight to give to those opinions and conclusions” (at [24]).

24. His Honour explained that “[f]or that reason, taken together with the absence of any opportunity for the respondents the test or challenge those opinions and conclusions, I am satisfied that to not limit the use to which the article can be put under s 136 would be unfairly prejudicial to the respondents” (at [24]).

25. To their great credit, the parties to this case (well advised by experienced solicitors and junior counsel), took a sensible and constructive approach to all issues as to admissibility. One aspect of their agreement related to the reception and use of the scientific articles. Taking a slightly different approach to Ruby Princess, I had proposed to the parties a limitation that would restrict reception of material to those articles and studies in the experts’ relevant area of expertise that were read and were said to form a basis for otherwise admissible opinion evidence or were used in cross-examination, accompanied by a general limitation under s 136 EA that the evidence was to be used for the purposes of providing a basis for, or understanding, the expert opinions adduced by the witnesses called.

26. The parties did not embrace this approach. On 19 October 2023, I was informed by the parties that a “joint position” had been reached to the following effect:

    All reports or decisions of regulatory authorities and authoritative scientific bodies, as well as scientific articles and reviews and other learned scientific publications, which have been referred to or relied upon by the expert witnesses who participated in the joint conclave should be admitted into evidence, with the parties being at liberty to make submissions in the course of closing submissions as to the weight that the court should place on any such document admitted into evidence.

27. As I noted when informed of this agreement, to the extent it operates to expand the admissibility and use of these scientific materials beyond the principled application of the Pt 3.3 EA exception to the Pt 3.2 hearsay rule, it was open to the parties to dispense, by consent, with the application of Pts 3.2 and 3.3 in relation to particular evidence (see s 190(1)(c) EA).

28. Accordingly, I made orders in accordance with this agreement and will proceed to deal with this aspect of the evidence in accordance with the express agreement of the parties.  

    B.5     The Scientific and Legal Method and Causation

29. Before passing from fact finding and causation, it is important to make a further point that was somewhat obscured during submissions.  

30. In ACQ Pty Ltd v Cook [2009] HCA 28; (2009) 237 CLR 656 (at 661 [14]), French CJ, Gummow, Heydon, Crennan and Bell JJ observed, albeit in a different context, that causation is one of the most difficult fields of debate in the law and one about which “abstract discussion is seldom valuable”. Perhaps recognising this, abstract discussion about causation was largely absent from the parties’ submissions. But one must not gloss over important differences between the scientific and legal method in relation to causal connexions between facts.

31. These differences have been explored in many articles including, by way of example, Korn, H L “Law, Fact, and Science in the Courts” (1966) 66 Columbia Law Review 1080; Brennan, T A “Causal Chains and Statistical Links: The Role of Scientific Uncertainty in Hazardous-Substance Litigation” (1988) 73 Cornell L. Rev. 469; and Geistfeld, M “Scientific Uncertainty and Causation in Tort Law” (2001) 54 Vanderbilt Law Review 1011. It is beyond the scope of this judgment to canvass such a large topic in anything other than the most superficial way and this judgment should not be unnecessarily diverted into epistemological or philosophical questions, but the nature of differences must be understood, at least in broad terms, so the scientific evidence is properly understood and its relationship to the legal issue of general causation is not distorted. 

32. Justice Jonathan Beach has written extra-curially on this topic in a way that repays close attention. Drawing upon his earlier work (“Scientific Evidence: A Need for Caution in Decision-Making” (2010) 42(1) Australian Journal of Forensic Sciences 49 and “Indeterminacy: The Uncertainty Principle of Negligence” (2005) 13 Torts Law Journal 129), in a recent article “Causation: The Interface Between the Scientific and Legal Methods” (2022) 49(1) University of Western Australia Law Review 113, Justice Beach explained (at 113) how the objectives of legal and scientific inquiry differ as science “searches for increased knowledge, with truth as its ideal; its coverage [being] more comprehensive and not time sensitive. But the law’s “objective is to resolve conflicts; its coverage is limited and time sensitive”.  In this way, science describes explains and predicts while the law “is concerned with regulating human affairs in accordance with values and objectives”.

1. It is simplistic to say there is one “scientific method” because it depends upon the nature of the scientific task, but speaking generally, these different objectives mean there are critical contextual differences between science and the law concerning methodologies and in the expression of conclusions of their differently focused enquires. Moreover, when it comes to causation, the role of the law might be said to involve an optimistic search for truth but, as my self-directions make clear, the aim is not to discover objective truth, but rather whether the relevant factual issue (in this case of general factual causation) has been proved to the requisite standard.

2. As Justice Beach, with respect, cogently explains in his recent article (at 117–8), while recognising the differences between the legal method and the scientific method:

   … when lawyers analyse scientific evidence, the epistemic values that need to be applied correlate with values underpinning the scientific method. A lens must be used which provides a sophisticated picture of the content and reliability of the scientific evidence. And not to analyse scientific evidence through the appropriate lens can lead to outcomes ‘determined by intuitive perceptions of the weight of authority rather than by reasoning from evidence’.

   Of course, the language, premises and analytical styles between the scientific method and the legal method have their differences. But when the latter is required to evaluate the former, significant correspondence of epistemic values arises. Lawyers must ‘retrace and evaluate the technical analysts’ logic, from empirical data to subjective judgment’. And complex issues of science require reasoning among purely technical facts analysed through complex models, statistical inference and mathematical instruments that have no comparator in legal reasoning.

   Now there are different modes for judicial analysis of scientific evidence. The judge might hear expert evidence in the usual way or through the mechanism of concurrent evidence sessions, which is my preferred option in patent litigation. The judge might appoint his own expert. The judge might obtain assistance from sitting with an assessor who may assist with any explanation of the scientific evidence. The judge might receive a report on the scientific question from a special referee, although this last option is remote from any direct judicial analysis of the scientific question. But whatever the procedural mode, this does not change the epistemic values and methods needed to assess scientific evidence. Moreover, judges are not obliged to accept the ipse dixit of the expert. They are obliged to test propositions, to assess one expert’s opinion against another, and to synthesise a position taking into account all expert, lay and documentary evidence.

3. This is what I have attempted to do. To rely acontextually upon general statements made by scientists as to causation, or to assume that scientific expressions as to possibility or probability necessarily translate to the legal task of considering probability through the lens of general factual causation at law is simplistic. Put in more simple terms, the factual issue before me is, at bottom, a common law jury question and, as was explained by Dixon CJ, McTiernan, Kitto, Taylor and Windeyer JJ in Ramsay v Watson (1961) 108 CLR 642 (at 645):

   That some medical witnesses should go into the witness box and say only that in his opinion something is more probable than not does not conclude the case. A qualified medical practitioner may, as an expert, express his opinion as to the nature and cause, or probable cause, of an ailment. But it is for the jury to weigh and determine the probability. In so doing they may be assisted by the medical evidence. But they are not simply to transfer their task to the witnesses. They must ask themselves ‘are we on the whole of the evidence satisfied on the balance of probabilities of the fact?’

   C        FACTUAL BACKGROUND

4. To limit the scope of the evidence, I required the legal representatives for both parties to attempt to agree upon all relevant non-contentious facts prior to preparing evidence that would be the subject of dispute.   

5. Although for reasons that are unnecessary to canvass, this process was somewhat delayed, a statement of agreed facts was prepared, and a final version was provided to the Court. That finalised version of the statement of agreed facts document was admitted into evidence (Agreed Facts or AF) with each fact being an “agreed fact” within the meaning of s 191 EA. The factual findings are made upon an admixture of the Agreed Facts, material drawn from contemporaneous documents, as well as inferences drawn from any communications and documents. Most facts agreed are irrelevant to the scope of the initial trial. I set out some below for the purpose of recording matters which are of relevance (at times, marginal at best relevance) to the present trial.

   C.1     Monsanto

6. Between around 1974 and 1987, the first respondent, Huntsman Chemical Company Australia Pty Limited (previously called Monsanto Australia Limited, which I will describe as Monsanto Australia (Old)), was a wholly-owned subsidiary of the fourth respondent (Monsanto Company (Old)), and together with associated and related entities conducted an agricultural chemicals business in Australia.

7. In about March 1988, the agricultural chemicals business of Monsanto Australia (Old) was transferred to the second respondent, Monsanto Australia Pty Ltd, (Monsanto Australia (New)) and was thereafter conducted by Monsanto Australia (New) and associated and related entities. Monsanto Australia (Old) ceased to conduct that business. 

8. Pursuant to a Share Sale Agreement dated 21 March 1988, all of the assets and liabilities of Monsanto Australia (Old) (which were primarily used prior to 31 March 1988 in the agricultural chemicals business conducted by it) were transferred to an entity which became known as Monsanto Australia Pty Ltd (that is, Monsanto Australia (New)). 

9. Until September 2000, Monsanto Australia (New) was a wholly-owned subsidiary of Monsanto Company (Old). From September 2000 until 2018, it was a wholly-owned subsidiary of the third respondent (Monsanto Company (New)). In 2018 Monsanto Australia (New) became a wholly-owned subsidiary of Bayer Aktiengesellschaft (Bayer AG).  

10. Monsanto Company (Old) and Monsanto Company (New) are entities incorporated in Delaware in the United States. Until approximately September 2000, Monsanto Company (Old) conducted, among other things, an agricultural chemicals business. Following a corporate restructure, that business including as to Roundup Products, was conducted by Monsanto Company (New) after approximately 1 September 2000.

11. In or around early 2000, Monsanto Company (Old) merged with a publicly owned pharmaceuticals company. Monsanto Company (Old) was the surviving corporation from the merger.

12. Monsanto Company (New) was created on 9 February 2000 as a wholly-owned subsidiary of Monsanto Company (Old). Effective on 1 September 2000, Monsanto Company (New) received the operations, assets and liabilities of the agricultural business previously conducted by Monsanto Company (Old), including in respect of the Roundup Products. Upon completion of the merger Monsanto Company (Old) changed its name to Pharmacia Corporation.

    C.2     Glyphosate and Roundup

13. The International Union of Pure and Applied Chemistry name for glyphosate is N- (phosphonomethyl) glycine.

14. Glyphosate was first synthesised by a chemist employed by Monsanto Company (Old) in 1970 and patented by Monsanto Company (Old) in 1974 for use as a herbicide.

15. The manufacture of Roundup Products involves the conversion of intermediate products to glyphosate acid (also known as glyphosate technical) which, in turn, is further converted to glyphosate salts for use in the formulation of Roundup Products.

    C.3     Roundup Herbicide and Roundup Biactive in Australia

16. Roundup Herbicide is a herbicide product which has been available for sale in Australia since 1976. In the time periods set out below, the entity identified manufactured Roundup Herbicide and/or Roundup Biactive:

    (1)Roundup Herbicide

    (a)1979 to 1988 – Monsanto Australia (Old);

    (b)1988 to 2000 – Monsanto Australia (New) and Monsanto Company (Old);

    (c)2000 to 2019 – Monsanto Australia (New) and Monsanto Company (New)

    (2)Roundup Biactive

    (a)1996 and 2000 – Monsanto Australia (New) and Monsanto Company (Old);

    (b)2000 to 4 February 2002 – Monsanto Australia (New) and Monsanto Company (New);

    (c)2 February 2002 to 2010 – Monsanto Australia (New);

    (d)2010 to 2019 – Monsanto Australia (New) and Monsanto Company (New).

17. The Roundup Herbicide and Roundup Biactive used in Australia during the Relevant Period was manufactured in accordance with specifications (including as to the formulations) and quality assurances manual set by Monsanto Company (Old) for the period 1976 to 2000, and by Monsanto Company (New) for the period 2000 to 2022.

18. The Roundup Herbicide and Roundup Biactive used in Australia during the Relevant Period was packaged (including labelling) in accordance with specifications and quality assurances manuals set by Monsanto Company (Old) for the period 1976 to 2000, and by the Monsanto Company (New) for the period 2000 to 2020. 

19. During the claim period:

    (1)Monsanto Company (Old) for the period 1976 to 2000 in respect of Roundup Herbicide and for the period 1996 to 2000 in respect of Roundup Biactive;

    (2)Monsanto Company (New) for the period 2000 to 2019;

    (3)Monsanto Australia (Old) for the period 1976 to 1988 in respect of Roundup Herbicide; and  

    (4)Monsanto Australia (New) for the period 1988 to 2020 in respect of Roundup Herbicide and for the period 1996 to 2020 in respect of Roundup Biactive,

    directly or through other parties and agencies promoted and marketed Roundup Herbicide and Roundup Biactive in Australia and engaged in activities related to marketing including conducting promotions, educational publications and programmes, product demonstration programmes and print advertisements and in accordance with product use and safety information provided by Monsanto Company (Old) for the period 1976 to 2000 and by Monsanto Company (New) for the period 2000 to 2020.

20. Information concerning the safe use of Roundup Herbicide and Roundup Biactive was communicated by Monsanto to consumers in Australia during the Relevant Period by two predominant means:

    (1)the product label including the “Directions for Use” booklet, both affixed to the container of the products; and

    (2)the safety data sheets.

    D        STRUCTURE OF CONSIDERATION OF THE SCIENTIFIC EVIDENCE

21. As noted above (at [49]), the scientific evidence relevant to determining the central issue was divided into three streams, broadly as follows:

    (1)epidemiological studies of the effects of glyphosate and/or GBFs on humans (Conclaves C, D and E) (epidemiology stream);

    (2)long-term studies on experimental animals relating to whether glyphosate and/or GBFs can cause cancer in animals (Conclaves H, I and J) (animal studies stream); and

    (3)mechanistic evidence, that is evidence relating to whether there is a biologically plausible mechanism by which glyphosate and/or GBFs can induce cancer in humans (Conclaves A, B, G) (mechanistic evidence stream).

22. Consistently with the need to have regard to the whole of the evidence, the parties accepted that it is necessary to make factual findings in relation to the evidence adduced in each conclave. It is then appropriate to consider each conclave in the context of the relevant scientific streams and then have regard to all the evidence from each stream. For reasons that will become clear, I will deal separately with Conclave F, which concerned absorption and exposure and the discrete question of dose.  

23. Accordingly, I will address each stream separately (although there is necessarily some overlap). In doing so, I will first explain the nature of the stream of scientific evidence; secondly, make general credit findings with respect to each the relevant experts; thirdly, summarise the evidence given in each conclave relevant to each stream; and fourthly, summarise my conclusions in relation to each stream of evidence building upon and supplementing the findings that have been made during the process of setting out the conclave evidence. I will then draw all the threads together in reaching my conclusions.  

    E        THE FIRST STREAM: EPIDEMIOLOGY (CONCLAVES C, D AND E)

    E.1     Nature of the Evidence

    I          Introduction

24. In 1854, Dr John Snow witnessed an outbreak of cholera in London. Determined to discover its source, he plotted the location of cholera deaths on a dot-map. The dots congregated a short distance from Broad Street (now Broadwick Street) in Soho. There, the local council had installed a water pump. Suspecting that the pump was the source of the outbreak, Dr Snow persuaded the council to remove the handle. The outbreak subsided.   

25. But that was not the end of the story: had the handle been contaminated? Was it the water itself? Or was the outbreak due to something else? Dr Snow thought it was the water. He continued to collect data on cholera deaths in the city and identified three water companies that supplied water to districts with the highest mortality rates. It was not clear within each district, however, which company supplied water to households where a cholera death occurred. By collecting data from individual households, Dr Snow inferred that two of the water companies were responsible for transmitting cholera poison and, by 1857, legislation had been enacted to mandate the filtering of all water in London (see Harris, S A “Epidemiology: Theory, Study Design and Planning for Education” (2000) 20 Journal of Continuing Education in the Health Profession 133).

26. Albeit in nascent form, Dr Snow was engaged in epidemiology. For his efforts, he enjoys the not inconsiderable distinction of having a pub named after him.

27. Expressed in broad terms, epidemiology is the study of distribution and determinants of disease in humans. As Spigelman CJ explained in Seltsam Pty Ltd v McGuiness [2000] NSWCA 29; (2000) 49 NSWLR 262 (at 271–272 [59]–[62]), epidemiology is based on the assumption that a disease is not distributed among populations at random: subgroups may be identified which are at increased risk of contracting particular diseases. Epidemiological evidence identifies associations between specific forms of exposure and the risk of disease in groups of individuals.

28. Epidemiologists make judgments about whether a statistical association represents a cause-effect relationship. However, those judgments focus on what is called in the epidemiological literature (as well as the law) general causation, that is, whether the factor is capable of causing the disease. Epidemiologists are not concerned with specific causation, that is, whether the factor caused the disease in an individual case.

29. For the purposes of determining whether exposure to a particular substance is the legal cause of a particular disease, epidemiology only provides evidence of possibility, which (for reasons I have explained), when expressed in opinion form and in epidemiological research, is admissible and part of the whole of the evidence to which the court must have regard: Seltsam v McGuiness (at 274–275 [78]–[79] per Spigelman CJ).

30. At bottom, this is a trial directed to the issue of general factual causation. Obviously enough, given its nature, epidemiological evidence is of real importance. Indeed, as it turns out, successfully demonstrating that the epidemiological evidence adduced in this proceeding supports the conclusion that glyphosate and/or GBFs can increase the risk of and/or cause NHL in humans is a hurdle in accepting Mr McNickle’s case.

31. But it remains one hurdle. As senior counsel for Mr McNickle, Mr Clements KC, submitted in his opening (T10.32–37):

    Despite there being very broad agreement between the experts retained by each side, as to the appropriateness of the weight of evidence approach, the respondents, in their written opening, have submitted that the epidemiological evidence should be given primacy by the court. We contend that approach should be rejected. We contend that it’s important for the court to closely consider and weigh up the evidence in each of the three separate streams of scientific evidence.

32. Put another way, the importance of the epidemiological stream does not mean that my conclusion as to whether I have reached a level of reasonable satisfaction on the preponderance of probabilities to sustain Mr McNickle’s case on the central issue is the product of giving primacy to one aspect of the evidence and reducing others to some form of secondary role or, more specifically, placing unwavering and disproportionate focus upon the epidemiological evidence in a vacuum. It is not. Indeed, focussing too intently on one part of the evidence at the expense of proper consideration of another part and then assessing the accumulated whole, would not only be a legal but a logical error.  

33. Of course, some parts of the evidence may emerge as being more persuasive or important than other aspects, but any conclusion on the central issue must be made with regard to the overall weight of all of the evidence.

    II        Methodology

34. It is largely uncontroversial among epidemiologists as to the principles or postulates which should be applied to assess evidence of a statistical correlation or association between the exposure in question and health outcome of interest: see Seltsam v McGuiness (at 285 [139] per Spigelman CJ).

35. The first step in assessing the epidemiological evidence and whether it indicates an etiologic relationship (that is, a causal relationship) between exposure to glyphosate and/or GBFs and development of NHL in humans is to identify the relevant peer-reviewed literature. The second step is to identify strengths and limitations of each publication, which, in turn, allows an identification of publications that are most relevant to the specific topic of concern, and an evaluation of the weight that should be placed on such publications (Conclave C Joint Report (JRC) (at [63])).  

36. In evaluating the relevant literature, epidemiologists may attempt to distinguish between causal and non-causal explanations (at JRC [64]–[65]). While there is no agreed set of criteria to assess causation that exists in the field, lists of criteria have become popular. The most well-known of these among epidemiologists is one devised by Sir Austin Bradford Hill, then Professor Emeritus of Medical Statistics at the University of London, in his Presidential address to the Section of Occupational Medicine: “The Environment and Disease: Association or Causation” (1965) 58 Proc R Soc Medicine 295.

37. The following summary of the “Bradford Hill criteria” is extracted from French CJ’s judgment in Amaca v Booth Pty Ltd [2011] HCA 53; (2011) 246 CLR 36 (at 54–55 [44]):

    [44]… The [Bradford Hill] criteria were expressed as the aspects of an association between two variables that should be considered before inferring that the most likely interpretation of the association is causation. In summary, they are:

    strength of association – eg, reflected in the ratio of the death rates between groups exposed to a suspected agent and those not so exposed;

    consistency in the observed association – eg, has it been repeatedly observed by different persons in different places, circumstances and times;

    the specificity of the association – if the association is limited to specific workers and particular sites and types of disease and there is no association between the work and other modes of dying, that is a strong argument in favour of causation;

    temporality – the temporal relationship of the variables;

    biological gradient – whether the association reveals a biological gradient or dose-response curve;

    plausibility – whether the expected causation is biologically plausible – a consideration which depends upon the biological knowledge of the day;

    coherence – the cause and effect interpretation of the data should not seriously conflict with the generally known facts of the natural history and biology of the disease;

    experiment – whether experimental or semi-experimental evidence supports a causation hypothesis;

    analogy – eg, given the effects of thalidomide and rubella it is easier to accept slighter but similar evidence with another drug or another viral disease in pregnancy.

1. When the matter next came before me on 12 August 2020, when the spectre of scientific manipulation had been raised and the parties were united in their possession to any reference process, I observed (T39):

   HIS HONOUR: … common ground ha[s] broken out about the folly of the course that I had proposed… I thought it might – there might be – this is relevant for the exercise of discretion either by me or by a full court [as to a reference], you’ve suggested some alternatives, but there may be others. I mean, one of the complications about this case which is similar to the complications about the PFAS case is that – and this presented some challenges in PFAS but was able to be overcome – is that there are aspects of the questions which reflected a significant divide in the scientific community, and that is people were of particular camps.

   And … we don’t want this like – a District Court personal injuries case in the 1980’s where there’s plaintiff’s doctors and defendant’s doctors, and one knows exactly what they’re going to say, and with all the unsatisfactory aspects of that. So even if we didn’t go down the reference course, it may be that at least one alternative is the idea of one or two court appointed experts together with perhaps an assessor. Anyway, I raise this because it – I accept the force of the submissions that have been made by the parties, and it’s something which does cause on pause, and you have to think through these things about what’s best for not only the parties and the Court, and the just determination of the issues.

   But, I’m just not going to let this case proceed down the path where there is entrenched experts on one side, entrenched experts on the other, and I’m sit here with this tsunami of expert material having to make a determination of which I think, in some respects, courts are ill-equipped to do. Anyway, I say that apropos of nothing other than just to entreat you to start thinking about these issues even if the reference process isn’t to proceed, because I think those sorts of issues need to be thought through very carefully before we were to make decisions about processes for conclaves and the like.

2. On 22 March 2021, when issues had been raised about the extent of discovery, upon which it became evident an enormous amount of money was being spent, a roundtable conference was held and attended by counsel, solicitors and the eDiscovery and technology mangers for the solicitors for the parties, during which I observed (T101):

   HIS HONOUR: … I must say, when I see the limited relevance of these documents in the overall case, I can’t help getting the feeling that a vast amount of money is being spent in relation to this discovery process – like in a large number of other class actions, which is not consistent with the overarching purpose. This case is likely to be determined on – and which has been conceded before in response to my proposal we have a referee – the assessment of the expert evidence about the properties of… these products. It’s not going to be determined – I can see some issues in the case to which the relevance of Monsanto Australia – of the carcinogenic – of the alleged carcinogenic property – that character of these products to be relevant, but – arguably irrelevant, but it’s the tail – I just get the feeling it’s just the tail wagging the dog – that a vast amount of money is being spent…which is not directed to the primary issue about what these – what – whether or not these two products were toxic and carcinogenic.  

3. These entreaties again fell on deaf ears. After coming back to the need to conduct this litigation in accordance with the overarching purpose on several occasions in 2021 and 2022, by 9 December 2022 (and after having vacated an earlier hearing of the initial trial in February 2022) and perhaps revealing my exasperation that the resolution of the case had been made unnecessarily costly and complicated, the following exchange occurred (T20):

   HIS HONOUR: The whole case comes down to whether or not this thing is carcinogenic. And I just want to entreat you again to think about ways that we could make this trial simpler.

   MS SZYDZIK: We will do so.

   HIS HONOUR: Yes. I know I was talked out of having a separate question about whether or not it’s carcinogenic. I can’t quite understand how I was talked out of that, but I was.

   MR FINCH: ...

   HIS HONOUR: … I decided it was a good idea after hearing submissions, but not I’ve seen the expert reports and I still can’t understand why that’s not a preferable course.

   MS SZYDZIK: We maintain that it isn’t, but perhaps we don’t really want to re-enliven that.

   HIS HONOUR: Well, either it’s me or somebody else that’s going to have to sit down at a trial in separate next year and decide whether or not this is – you win or you lose. And that’s going to turn on the question of carcinogenicity, it seems to me.

4. After delivering an ex tempore judgment on an unrelated matter, later that day the following exchange occurred (T30):  

   HIS HONOUR: … Now, just remind me why – just remind me again – humour me by reminding me again as to why it’s still not sensible to deal with just the issues of carcinogenicity first?

   MS SZYDZIK: Because it’s not as straightforward as that. Because it is an issue that is bound up with questions relating to the respondent’s knowledge and, as your Honour has also said, questions relating to the tainting conduct that has been set out in the reply.

   HIS HONOUR: So peripheral this tainting business for reasons I’ve explained, I just – even more that I’ve seen when I’ve read the joint expert reports. It just does not seem to me – we have really real misgivings about this. I was talked out of a reference process. Having seen the joint reports, I don’t know why. This spectre of scientific fraud has been – which doesn’t seem to have anything to do with the way in which the experts have expressed their views. It only goes to the question of what the knowledge was of Monsanto at relevant times for the negligence case if you’re going to win on carcinogenicity anyway. And it can’t be determined. It’s fairyland to think that’s going to be determined if it seems to me it’s a safety defect point. So anyway, all I’m saying I will ask you to think about it again. And I will entertain – if anyone wanted to make an application again now that I’ve had the benefit of the expert reports and after I’ve seen the rejoinders, as to some other more efficient way of dealing with this which would not involve a judge of the court having to spend a vast, vast amount of its time on the balance of issues in the case if the view was formed that it’s not carcinogenic. And if a view is formed that it is carcinogenic, then order the parties to a mediation after that finding has been made. Then I will entertain it on 2 February.

   But again, look, you may be able to convince me again that it’s inappropriate. It’s just that I’m finding it really hard to reconcile the amount of time that’s allocated to this when there seems to me a fairly confined question which is going to leave the parties, irrespective of which ways it goes, in a much better position. And if we then have to have a subsequent hearing, we have a subsequent hearing. But I know I have expressed these views before and it must be an element of frustration, but I don’t think we should just necessarily continue to going down the same path if the path becomes increasingly rocky and the alternative path seems to lead to the broad sunlit uplands of making life easier for the person deciding the case.

5. Indeed, as it turned out, it was “as straightforward as that”. Finally, on 2 February 2023, senior counsel for Monsanto said (T3–4):

   MR FINCH: On two occasions on 9 December, your Honour entreated the parties to give some thought about ways in which the hearing can be made more simple - - -

   HIS HONOUR: I did.

   MR FINCH: - - - and shorter, not to put too fine a point on it, with particular reference to the issue of the causation of NHL/cancer/carcinogenicity issue more generally. Can I be disarmingly frank about this. There’s a great deal of attraction in the debate that your Honour has instigated on a number of occasions about focusing on that. There are some difficulties, so far as we can tell for the moment, in having a traditional separate question concerning that, and this is an issue your Honour has heard about before, because our learned friends wish to, amongst other things, put to our expert witnesses that their conclusions might be different had they been aware of certain scientific misconduct as it affects some of the basal data. Scientific misconduct, of course, doesn’t affect the witnesses; it affects some of the data, some of the articles, if it be established.  

   I have some problems, if I can be forgiven for speaking in the first person, with forcing the applicants not to be able to do that, because they say, “This is something which we want to say which vitiates the respondent’s scientific case about causation.” And, with respect, it seems to us to be dangerous to ask your Honour to force them not to do that either. But there is another way forward which we want to float now…

6. What was then suggested was an embryonic version of the trial plan eventually put in place. Unfortunately, those acting for Mr McNickle were not able to respond that day and said they required an adjournment, but this trial plan was eventually put in place. Why the parties resisted simplification of the initial trial and why a version of the final trial plan could not have been put in place years before is, on my present understanding of the material, difficult to fathom.

7. To repeat, my instinctive view, raised at the beginning of the case, is that it was possible to separate out potentially determinative issues. Subject to hearing submissions, it might be thought the reasons given from time to time as to why a simplification or staging of issues did not occur at a much earlier stage did not have any substance. Although it might be said those acting for Mr McNickle steadfastly refused any simplification of the case whenever it was raised (except at the heel of the hunt), Monsanto was also responsible for facilitating and giving effect to the overarching purpose. Notwithstanding my unfeigned respect for all those involved and my gratitude to the solicitors and all counsel for conducting the final hearing with such skill and courtesy, I do not consider this case has been conducted optimally or even efficiently.  

8. As noted above, these large class actions involve the consumption not only of the private resources of the parties, but also public resources. Everyone is entitled to their day in court, but they are not entitled to another litigant’s day in court. The resisting parties also have a separate obligation to identify and then promote the optimal way of resolving the dispute consistent with their separate obligations under Pt VB of the FCA Act.

9. Over and over in these large class actions, judges of the Court see huge process costs being expended which are ultimately wasted. As I have previously remarked, a respondent is entitled to spend as much money as it wants to defend litigation and, subject to ethical constraints, a solicitor is entitled to work as instructed, which may include doing work which could be later characterised as objectively unnecessary. But this does not mean that unnecessary costs should be recoverable against an unsuccessful litigant. A singular aspect of this case is that despite active case management and judicial intervention aimed towards constraining costs from the very outset, much more money has been spent on this litigation than in truth was necessary for a fair disposition of the issues – particularly relating to issues as to Monsanto’s knowledge.

10. I wish to receive submissions on how these issues inform any order for costs.  

    III       Public Interest Litigation

11. The second factor is that in Turner v MyBudget Pty Limited (No 2) [2018] FCA 1509, I dealt with a costs application made in a class action by a successful respondent. After noting the submission that the “usual rule that costs follow the event” is somewhat of an oversimplification, I observed (at [8]–[14]):

    [8]Although the fact that [the respondent] has been successful in resisting the claim by [the applicant] is a very powerful discretionary factor militating in favour of an award of costs, it is not necessarily determinative. While the discretion as to costs is a very broad one, it is restrained in the case of representative proceedings by the prohibition on making an award for costs against group members except in defined circumstances: see s 43(1A) of the [FCA Act]. Although this provision is not directly relevant in the case of costs sought against a representative party, it recognises the reality that the questions determined in representative proceedings have a public dimension which transcends ordinary inter partes litigation and the rights of the parties to the litigation inter se. 

    [9]The case brought by [the applicant] is an exemplar of the type of proceeding that the Australian Law Reform Commission had in mind when, in 1988, it first recommended the introduction of a class action regime in ALRC Report 46.  Here, a very large number of persons were affected by a common issue of law (in this case being the proper construction of the Interest Provision).  It would have made no sense whatsoever for any individual litigant to have commenced a proceeding seeking clarification of this issue given the very small amount of money at stake for any individual litigant.  The benefit of a grouping of claims by a Part IVA proceeding could only be achieved in the event that one person was prepared to come forward and act as the representative applicant.  Unless [the applicant] had been prepared to bring the claim… uncertainty would have continued to exist relating to the true effect of the Interest Provision.  Although I have reached a view that the proper construction is as [the respondent] contended, the arguments advanced by [the applicant] were far from being unarguable and, if I may say so, were advanced in an efficient and comprehensive manner by Counsel for the applicant, and in a way which was consistent with the overarching purpose. 

    [10]This is a world away from a commercial class action, where the applicant is part of some form of common enterprise which seeks to use the Court’s processes not only for the vindication of the applicant’s personal claim but also as a means by which a managed investment scheme is seeking to derive a significant financial advantage for the participants in the scheme including a litigation funder.   

    [11][The respondent] recognises the fact that a proceeding brought in the “public interest” may be a basis, at least in some circumstances, for not awarding against an unsuccessful applicant: see Ruddock v Vadarlis (No 2) [2001] FCA 1865; (2001) 115 FCR 229. It asserts, however, that there is no general principle that the usual “costs rule” should not apply if the subject matter of litigation is a matter of public importance. It also points to QANTAS Airways Limited v Cameron (No 3) (1996) 68 FCR 387, a case which involved the question as to whether an unsuccessful applicant in a class action should be ordered to pay costs. QANTAS Airways was a case brought to vindicate mixed public and private interests: see at 390.  Lindgren and Lehane JJ considered that the litigation served the “public interest” to the extent that it elucidated the duty of care owed by international airlines to passengers in relation to environmental tobacco smoke and that this should be given some weight, but damages were claimed and it was “impossible to view the proceedings as having been brought and pursued purely in the public interest”: see 389-390.  In those circumstances, the applicant was ordered to pay some costs.

    [12]Similarly, it is submitted, [the applicant] sought monetary relief and hence “his case cannot be viewed as having been pursued purely in the public interest”.

    [13]It may be accepted that this was not a “pure” public interest claim, but a striking feature of this case was that any claim of [the applicant] (or any group member) was very modest, yet the collective benefit was potentially large.  Although the class action did seek to vindicate [the applicant’s] claim, it had a very significant benefit transcending the parties relative to the stake of the personal financial claim of the applicant.  24,222 others have obtained certainty as to their position. This number includes, it is safe to infer, many persons likely to have some financial vulnerability.

    [14]Although I do take into account the important consideration that [the applicant’s] individual case has failed, he (together with his lawyers) has performed a valuable service for the benefit of others. In all the circumstances, while taking fully into account that the costs discretion is generally to be exercised in favour of the successful party, my view is that no order for costs should be made.  Although on one level this might be thought to operate unfairly on [the respondent] which has incurred not insignificant costs, it must be recalled that it has now received the benefit of quelling the controversy relating to the operation of the Interest Provision not only in relation to [the applicant], but with regard to all group members. Costs applications are not determined solely by identifying who was responsible for the litigation, but by a broad assessment as to whether the order does occasion an injustice...

12. Given the significant public interest in this litigation, the resolution of the central issue (at least for present purposes) transcends the interests of the parties. How this informs the discretion as to costs is also something in respect of which I seek assistance.

    IV       The Position of Mr McNickle

13. Thirdly, and related to the last point, as I noted in the introduction, Mr McNickle is a man of modest means; is presently unwell; and has a wife to support. However strongly I empathise with the position Mr McNickle finds himself in, I am aware, in ordinary, inter partes litigation, that any discretion exercised in such a way so as to deprive a successful party of their costs must be exercised judicially and “according to rules of reason and justice, not according to private opinion … or even benevolence … or sympathy”: Williams v Lewer [1974] 2 NSWLR 91 (at 95 per Rath J); Oshlack v Richmond River Council (1998) 193 CLR 72 (at 81 per Gaudron and Gummow JJ).

14. But this is not ordinary litigation, and he has relied on others to bring the claim and, despite being unwell, he has performed a signal service in acting as a representative applicant. Mr McNickle will no doubt be disappointed in the result, but his labours have not been in vain. His individual case has provided a means by which the evidence currently marshalled has been evaluated and assessed thus resolving the legal rights of a great many people. At the end of the day, any ordinary costs rules are subject to the ability of the Court to make further or other orders as required to achieve an overall just result: Lombard Insurance Co (Australia) Ltd v Pastro (1994) 175 LSJS 448; GEC Marconi Systems Pty Ltd v BHP Information Technology Pty Ltd [2003] FCA 688; Furber v Stacey [2005] NSWCA 242.

15. In this last respect, I am conscious that Monsanto has made submissions as to why the Court “should not be concerned as to the ramifications” of making a costs order against Mr McNickle but, so far as I can see, those submissions go beyond any evidence that was adduced at the hearing and, subject to any “without prejudice” material relevant to costs that may be admissible because of s 131(2)(h) EA, the parties agreed that the entire “universe of material” to which I am to have regard in making any final orders has already been adduced and is in evidence (T2087.40–2088.11).

    V        Future Submissions

1. If a costs order against Mr McNickle is pursued, subject to any application, I will allow for further evidence as to costs (restricted to material admissible because of the exception contained in s 131(2)(h) EA) and submissions to be made to address relevant considerations.

   J.6      Orders

2. For these reasons, I make the following orders and make the notation set out below:  

   1.Pursuant to s 33ZF of the Federal Court of Australia Act 1976 (Cth) (FCA Act) (and notwithstanding that no application has been made by the representative party in accordance with s 33K(1) of the FCA Act) leave be granted for the group membership to be amended nunc pro tunc so that it is defined as it is in the last version of the statement of claim filed prior to the fixing of the date for opt out (being the second further amended statement of claim filed on 19 October 2020).

   2.Pursuant to ss 33ZF and 37P(2) of the FCA Act, the following question, common to the claim of the applicant and all group members (Central Common Question) be answered separately and before any other question or issue in the proceeding:  

   Throughout the period between July 1976 and 19 October 2020 (relevant period), did or could use of and/or exposure to the herbicide product or products, which contained glyphosate and were branded as “Roundup”, or which contained glyphosate and were otherwise branded with the name “Monsanto” (Roundup Products) increase an individual’s risk of developing non-Hodgkin lymphoma (NHL); and/or cause an individual to develop NHL?

   3.        The Central Common Question be answered as follows:   

   It is not proven in this proceeding on the balance of probabilities (in accordance with s 140(1) of the Evidence Act 1995 (Cth)), that throughout the relevant period, use of and/or exposure to Roundup Products increased an individual’s risk of developing NHL; and/or caused an individual to develop NHL.

   4.        The proceeding be dismissed upon finalisation of any issues relating to costs.

   5.Any further evidence as to costs (as contemplated in the reasons (at [1202])) and any submissions as to costs be filed by 12 noon on 30 July 2024.

   6.For the purposes of s 33ZB of the FCA Act, Orders 1 to 3 above affect (and hence bind) all current group members in the class action (being those named in the second further amended statement of claim who did not opt out).

   7.The proceeding be adjourned for the parties to be heard on the appropriate costs orders in conformity with these reasons at 10:15am on 31 July 2024.

   AND THE COURT NOTES THAT:

   8.The leave granted after the date of opt out had passed to amend the statement of claim so as to file the third and fourth statement of claim did not include leave being granted on application made by the representative party to alter the description of the group pursuant to s 33K(1) of the FCA Act.

I certify that the preceding one thousand two-hundred and three (1203) numbered paragraphs are a true copy of the Reasons for Judgment of the Honourable Justice Lee.

Associate:

Dated: 25 July 2024