An Empirical Study of the Use of Neuroscientific Evidence in Sentencing in New South Wales, Australia
Armin Alimardani
A thesis in fulfilment of the requirements for the degree of Doctor of Philosophy
School of Law Faculty of Law
March 2019 THE UNIVERSITY OF NEW SOUTH WALES – Thesis/Dissertation Sheet
Surname or Family name: Alimardani First name: Armin Other name/s: N/A Abbreviation for degree as given in the University calendar: PhD School: Law Faculty: Law Title: An Empirical Study of the Use of Neuroscientific Evidence in Sentencing in New South Wales, Australia
ABSTRACT In the neurolaw literature there are many debates and claims associated with the increasing use of neuroscience in law. These claims can be sorted into three broad categories: predictive claims that assert how neuroscientific evidence will be used or will influence the application of criminal law principles in the future; descriptive claims about how neuroscientific evidence is being used in practice; and normative claims which assert how neuroscientific evidence ought to be used in practice. While it has been at least 40 years since the first use of neuroscience in Australian courts, neurolaw claims have not been substantiated by comprehensive and systematic empirical research on how neuroscience is used in practice. Thus prior to this thesis it was unclear whether the actual approach of the courts is consistent with claims of various neurolaw scholars. To fill this substantial void, this thesis offers an empirically oriented review of seven claims about the introduction and use of neuroscientific evidence applied to sentencing courts in New South Wales, Australia. To pursue this research, a comprehensive and systematic search was conducted on three databases finding 214 decisions before 2016 in which neuroscientific evidence was discussed. These cases were then coded, and analysed. The findings of the study suggest that there are some inconsistencies with the neurolaw claims and how neuroscientific evidence is being used in practice. Although not all these inconsistencies warrant concern, there appears to be a need for practice-oriented instruments and guidelines about the use of neuroscience in practice.
DECLARATION RELATING TO DISPOSITION OF PROJECT THESIS/DISSERTATION I HEREBY GRANT TO THE UNIVERSITY OF NEW SOUTH WALES OR ITS AGENTS THE RIGHT TO ARCHIVE AND TO MAKE AVAILABLE MY THESIS OR DISSERTATION IN WHOLE OR IN PART IN THE UNIVERSITY LIBRARIES IN ALL FORMS OF MEDIA, NOW OR HERE AFTER KNOWN, SUBJECT TO THE PROVISIONS OF THE COPYRIGHT ACT 1968. I RETAIN ALL PROPERTY RIGHTS, SUCH AS PATENT RIGHTS. I ALSO RETAIN THE RIGHT TO USE IN FUTURE WORKS (SUCH AS ARTICLES OR BOOKS) ALL OR PART OF THIS THESIS OR DISSERTATION. I ALSO AUTHORISE UNIVERSITY MICROFILMS TO USE THE 350 WORD ABSTRACT OF MY THESIS IN DISSERTATION ABSTRACTS INTERNATIONAL (THIS IS APPLICABLE TO DOCTORAL THESES ONLY).
23 April 2019 ……………………………………………………. ……………..…...... ……………..…...... ………...... Signature Witness Signature Date
The University recognises that there may be exceptional circumstances requiring restrictions on copying or conditions on use. Requests for restriction for a period of up to 2 years must be made in writing. Requests for a longer period of restriction may be considered in exceptional circumstances and require the approval of the Dean of Graduate Research. FOR OFFICE USE ONLY Date of completion of requirements for Award:
I
ORIGINALITY STATEMENT
‘I hereby declare that this submission is my own work and to the best of my knowledge it contains no materials previously published or written by another person, or substantial proportions of material which have been accepted for the award of any other degree or diploma at UNSW or any other educational institution, except where due acknowledgement is made in the thesis. Any contribution made to the research by others, with whom I have worked at UNSW or elsewhere, is explicitly acknowledged in the thesis. I also declare that the intellectual content of this thesis is the product of my own work, except to the extent that assistance from others in the project's design and conception or in style, presentation and linguistic expression is acknowledged.’
Signed ......
Date 23/04/2019
II
23/Jul/2019
23/Jul/2019
III
INCLUSION OF PUBLICATIONS STATEMENT
UNSW is supportive of candidates publishing their research results during their candidature as detailed in the UNSW Thesis Examination Procedure.
Publications can be used in their thesis in lieu of a Chapter if: • The student contributed greater than 50% of the content in the publication and is the “primary author”, ie. the student was responsible primarily for the planning, execution and preparation of the work for publication • The student has approval to include the publication in their thesis in lieu of a Chapter from their supervisor and Postgraduate Coordinator. • The publication is not subject to any obligations or contractual agreements with a third party that would constrain its inclusion in the thesis
Please indicate whether this thesis contains published material or not. ☐ This thesis contains no publications, either published or submitted for publication
Some of the work described in this thesis has been published and it has been ☒ documented in the relevant Chapters with acknowledgement
This thesis has publications (either published or submitted for publication) ☐ incorporated into it in lieu of a chapter and the details are presented below
CANDIDATE’S DECLARATION I declare that: • I have complied with the Thesis Examination Procedure • where I have used a publication in lieu of a Chapter, the listed publication(s) below meet(s) the requirements to be included in the thesis. Name Signature Date (dd/mm/yy)
Armin Alimardani 23/04/2019
Postgraduate Coordinator’s Declaration I declare that: • the information below is accurate • where listed publication(s) have been used in lieu of Chapter(s), their use complies with the Thesis Examination Procedure • the minimum requirements for the format of the thesis have been met. PGC’s Name PGC’s Signature Date (dd/mm/yy)
IV
Publications and presentations arising from this dissertation
Publications
Alimardani, A, Chin, J (2019) Neurolaw in Australia: The Use of Neuroscience in Australian Criminal Proceedings, Neuroethics https://doi.org/10.1007/s12152-018-09395-z
Alimardani, A (2018) Neuroscience, Criminal Responsibility and Sentencing in an Islamic Country: Iran, Journal of Law and Biosciences, Oxford University Press, doi:10.1093/jlb/lsy024
Presentations
Alimardani, A (2018) A Shift from Retribution and Towards Consequentialism: An Empirical Study of the Use of Neuroscience in Australian Courts. Presented in The 31th annual Australian and New Zealand Society of Criminology (ANZSOC) conference, Melbourne, Australia.
Alimardani, A, Vincent, N. A (2018) Doing Time: Neuroscience and Custodial Sentencing, Presented in The 12th annual ANZSOC Postgraduate and Early Career Research Conference, Melbourne, Australia.
Alimardani, A, Vincent, N. A. (2018), Doing Time: Neuroscience and Custodial Sentencing, Poster session in NEUROSCIENCE & SOCIETY: Ethics, Law, and Technology Conference, Sydney, Australia
Alimardani, A (2018) Neurolaw in Australia: An Empirical Study of the Use of Neuroscientific Evidence in Sentencing, Presented at NEUROSCIENCE & SOCIETY: Ethics, Law, and Technology Conference, Sydney, Australia
Alimardani, A (2017) Neuro-time-relativity and Sentencing: How Subjective Perception of Time may Change the Length of the Prison Sentence, presented at Law, Technology and Innovation (LTI) Junior Scholars Forum, Faculty of Law, UNSW and accepted in The 10th Melbourne Doctoral Forum on Legal Theory Melbourne Law School – this paper only presented at LTI Junior Scholars Forum UNSW.
V
Armin Alimardani, Jason M. Chin (2017) Neurolaw in Australia: The Use of Neuroscientific Evidence in Criminal Proceedings, Presented in The 30th annual Australian and New Zealand Society of Criminology (ANZSOC) conference, Canberra, Australia.
Armin Alimardani, Jason M. Chin, (2017) Neurolaw in Australia: The Use of Neuroscientific Evidence in Criminal Proceedings, Poster session presented at the annual meeting of Neuroscience and Society: Ethical, Legal & Clinical Implications of Neuroscience Research Conference, Sydney, Australia.
Alimardani, A (2017) The Emergence of Neuroscience as a Commonly Admitted Type of Evidence in Australian Criminal Courts: How it Happened and Why We Should Care, The 30th annual Australian and New Zealand Society of Criminology (ANZSOC) conference, Canberra, Australia.
Alimardani, A (2017) Uncovering Infamous Biological Theories Buried in the Criminal Justice System: An Empirical Study of the Use of Neuroscience in Australian Criminal Courts, Presented in The 11th annual ANZSOC Postgraduate and Early Career Research Conference, Canberra, Australia.
Alimardani, A (2016) Explaining crime in criminal proceedings: Towards a new approach, Presented in The 29th Annual Australian and New Zealand Society of Criminology conference, Hobart, Australia
Alimardani, A (2016) Explaining Crime: Contributing Factors within a Multilayered Network (CFMN), Presented in The 10th ANZSOC Postgraduate & Early Career Researcher Conference (PECRC), Hobart, Australia
VI
TABLE OF CONTENTS DEDICATIONS ...... XI
ABSTRACT ...... XII
CHAPTER 1 INTRODUCTION ...... 1 1. RESEARCH QUESTIONS ...... 11 2. RESEARCH ORIGINALITY AND SIGNIFICANCE ...... 12 3. DISSERTATION OUTLINE ...... 13
CHAPTER 2 LITERATURE REVIEW AND BACKGROUND ...... 18 Introduction ...... 19 1. NEUROSCIENCE: BACKGROUND AND TECHNIQUES ...... 19 BRAIN IMAGING TOOLS ...... 24
NEUROSCIENTIFIC EVIDENCE ...... 31 2. NEUROSCIENCE AND LAW: NEUROLAW...... 35 BACKGROUND TO NEUROSCIENCE AND CRIME ...... 37 3. BACKGROUND TO NEUROLAW CLAIMS ...... 41 NEUROSCIENCE, FREE WILL AND CRIMINAL RESPONSIBILITY ...... 42
NEUROSCIENCE AND EXPLAINING CRIME ...... 45
NEUROSCIENCE AND REHABILITATION ...... 46
LIMITATIONS OF NEUROSCIENCE AND CONVERGENCE OF EVIDENCE ...... 51
THE PERSUASIVE POWER OF NEUROSCIENTIFIC EVIDENCE ...... 55
EXPERT WITNESSES ...... 57 4. NEUROSCIENCE IN COURTS: EMPIRICAL NEUROLAW STUDIES ...... 59 Conclusion...... 62
CHAPTER 3 LEGAL BACKGROUND ...... 63 Introduction ...... 64 1. THE FRAMEWORK OF THE CRIMINAL JUSTICE SYSTEM ...... 64 2. SENTENCING FACTORS ...... 67 THE AIMS OF SENTENCING ...... 67
MITIGATING AND AGGRAVATING FACTORS IN SENTENCING ...... 70
OTHER FACTORS ...... 71 3. DECIDING ON THE PROPER SENTENCE ...... 71 TRANSPARENCY IN SENTENCING DECISIONS ...... 73 4. CRIMINAL COURTS IN NSW...... 74
VII
Conclusion...... 75
CHAPTER 4 AN INTRODUCTION TO EMPIRICAL RESEARCH ...... 76 Introduction ...... 77 1. SOURCE OF DATA ...... 77 2. SAMPLING AND DATA COLLECTION ...... 82 LIMITATIONS OF SAMPLING AND DATA COLLECTION ...... 89
PRELIMINARY ANALYSIS OF DATASET (CASE SAMPLES) ...... 90 3. DATA ANALYSIS ...... 94 Conclusion...... 101
CHAPTER 5 DOUBLE-EDGED SWORD EFFECT OF NEUROSCIENCE ...... 102 Introduction ...... 103 1. AN OVERVIEW OF SENTENCING ELEMENTS IN NSW THAT ARE POTENTIALLY RELEVANT TO NEUROSCIENCE EVIDENCE ...... 107 2. A QUALITATIVE ANALYSIS OF NEUROSCIENTIFIC EVIDENCE IN SENTENCING ...... 109 NEUROSCIENCE POINTS TOWARDS A SHORTER SENTENCE ...... 109
NEUROSCIENCE POINTS TOWARDS A LONGER SENTENCE ...... 113
NEUROSCIENTIFIC EVIDENCE SUGGESTING A REDUCED RISK OF RE-OFFENDING ...... 119 3. A QUANTITATIVE ANALYSIS OF ‘DOUBLE-EDGED SWORD’ CLAIM IN SENTENCING ...... 121 RESULTS AND FINDINGS ...... 123
UNCLEAR CASES ...... 127 4. DOUBLE-EDGED SWORD EFFECT, OTHER CONSIDERATIONS ...... 128 SENTENCING FACTORS: WEIGHT AND LIMITS ...... 128
CASES BIASED TOWARDS MITIGATION ...... 132
LEGISLATIONS AND RISK OF A LONGER SENTENCE ...... 135 5. DOUBLE-EDGED SWORD CLAIM IN OTHER JURISDICTIONS ...... 139 CONCLUSION ...... 141
CHAPTER 6 THE GAP BETWEEN NEUROSCIENCE AND THE LAW’S LANGUAGE ...... 142 Introduction ...... 143 1. THE PURPORTED GAP BETWEEN NEUROSCIENCE AND LAW ...... 144 2. DIFFICULTIES IN EXAMINING THE CLAIM ...... 148 3. NEUROSCIENTIFIC LANGUAGE IN PRACTICE: ANALYSIS OF CASES ...... 152 4. NEUROSCIENCE AND FOLK PSYCHOLOGY: INTERCHANGEABLE LANGUAGES ...... 160 Conclusion...... 161
VIII CHAPTER 7 CONCERNS ABOUT THE RELIABILITY OF NEUROSCIENTIFIC EVIDENCE IN COURTS ...... 163 Introduction ...... 164 1. AN OVERVIEW OF THE RELIABILITY OF EVIDENCE IN NSW SENTENCING PROCEDURES ...... 168 2. RELIABILITY OF NEUROSCIENCE: CONVERGENCE OF EVIDENCE...... 169 COURTS’ CONSIDERATION OF THE RELIABILITY OF NEUROSCIENTIFIC EVIDENCE IN PRACTICE...... 172
CONVERGENCE OF EVIDENCE: AN EMPIRICAL ANALYSIS ...... 173 3. RELIABILITY OF NEUROSCIENCE: THE PERSUASIVE POWER OF NEUROSCIENCE ...... 178 BEHAVIOURAL EVIDENCE VS NEUROSCIENTIFIC EVIDENCE ...... 182
BRAIN SCANS ARE NOT OBJECTIVE EVIDENCE ...... 184
VEEN AND PREJUDICIAL NEUROSCIENCE: A CAUTIONARY TALE ABOUT NEUROLAW ...... 186 4. RELIABILITY OF NEUROSCIENCE: EXPERT WITNESSES...... 189 EXPERT WITNESSES AND THEIR AREA OF EXPERTISE IN COURTS ...... 192
COURTS’ REACTION TO AREAS OF EXPERTISE ...... 196 CONCLUSION ...... 199
CHAPTER 8 NEUROSCIENCE AND REHABILITATION OF THE OFFENDER ...... 201 Introduction ...... 202 1. A QUALITATIVE ANALYSIS OF NEUROSCIENCE AND REHABILITATION ...... 203 2. ANALYSIS OF THE NEURO-REHABILITATION CLAIM...... 212 Conclusion...... 214
CHAPTER 9 NEUROLAW OF THE MECHANICAL BRAIN ...... 216 Introduction ...... 217 1. A MECHANICAL BRAIN: LEGAL IMPLICATION ...... 218 2. RETRIBUTIVE AND CONSEQUENTIALIST SENTENCING FACTORS ...... 223 3. EVALUATING GREENE AND COHEN’S CLAIM IN NSW SENTENCING PRACTICE ...... 226 EVIDENCE OF A SHIFT FROM RETRIBUTION TOWARDS CONSEQUENTIALISM ...... 227 4. PREVENTIVE DETENTION: PUSHING THE SCALE TOWARDS CONSEQUENTIALISM ...... 230 Conclusion...... 231
CHAPTER 10 CONCLUSION ...... 233 INTRODUCTION ...... 234 1. NEUROLAW CLAIMS: THEORY VS PRACTICE ...... 235 2. CONTRIBUTIONS OF THE STUDY, RECOMMENDATIONS, AND FUTURE DIRECTIONS ...... 248
BIBLIOGRAPHY ...... 252 ARTICLES/BOOKS...... 252
IX
CASES ...... 265
WEB SOURCES ...... 267
X
Dedications
I would like to express my gratitude to my supervisors Professor Gary Edmond, Dr. Allan McCay and Dr. Marc De Leeuw for their support in the development of ideas and the encouragement they provided in the writing of the thesis, and for their kindness during my periods of bereavement. Gary, I will never forget when you picked me up during my worst time and encouraged me to finish this thesis.
I must acknowledge the love and support I received from my family. You had a difficult life, but you always did more than what you should do for my education. Thank you!
Further, I cannot imagine going through these days without the support of Professor Nicole Vincent. You have been my best friend for the last couple of years and you were always there for me.
I am grateful to Dr. Jenny Jarrett for her support, you are the best Jenny!
I am grateful to Dr. Jane Bolitho for giving me the opportunity to teach at UNSW and making a great teacher.
I would like to thank Warrane College residents, especially the dean of Warrane, Arthur Escamilla, for their warm support throughout these years,
My fellow HDR students, thanks for everything. You were always there for me.
I am grateful to IceFrog and Valve for developing Dota2. Without Dota, I could not survive all the stress and anxiety of the PhD.
I am grateful to my friends (Kasakhele group chat) from National Organisation of Developing Exceptional Talents (NODET) for being my greatest friends for the last 15 years and cheering me up whenever I needed the most.
I am also immensely grateful to Antonia Xue and Oliver Ray for editorial assistance.
And finally, my dear friend, MohammadReza Behrouzih, thanks for everything you did for me.
XI
Abstract
In the neurolaw literature, there are many debates and claims associated with the increasing use of neuroscience in law. These claims can be sorted into three broad categories: predictive claims that assert how neuroscientific evidence will be used or will influence the application of criminal law principles in the future; descriptive claims about how neuroscientific evidence is being used in practice; and normative claims which assert how neuroscientific evidence ought to be used in practice.
While it has been at least 40 years since the first use of neuroscience in Australian courts, neurolaw claims have not been substantiated by comprehensive and systematic empirical research on how neuroscience is used in practice. Thus prior to this thesis, it was unclear whether the actual approach of the courts is consistent with claims of various neurolaw scholars. To fill this substantial void, this thesis offers an empirically oriented review of seven claims about the introduction and use of neuroscientific evidence applied to sentencing courts in New South Wales, Australia. To pursue this research, a comprehensive and systematic search was conducted on three databases finding 214 decisions before 2016 in which neuroscientific evidence was discussed. These cases were then coded and analysed.
The findings of the study suggest that there are some inconsistencies with the neurolaw claims and how neuroscientific evidence is being used in practice. Although not all these inconsistencies warrant concern, there appears to be a need for practice-oriented instruments and guidelines about the use of neuroscience in practice.
XII
Chapter 1 Introduction
1
On the night of March 12, 2011, in Springwood, NSW, Australia, Mr McCann strangled Mr Morgan with a nylon cord.1 Mr Morgan was one of Mr McCann’s lodgers who was supposed to have left the premises two nights before the incident. McCann was charged with murder. The defendant claimed that the deceased threatened to kill him, and thus his act was a matter of self- defence. However, the jury concluded that the killing was unnecessary and strangling the deceased was excessive self-defence. On this account, the jury found the offender guilty of manslaughter.
In sentencing, with respect to the offender’s mental condition, several reports by expert witnesses were submitted. Among these reports was a reference to two brain images – a neuroscientific technique – that is normally used in hospitals rather than a criminal court. A Computed Tomography (CT) scan and a Magnetic Resonance Imaging (MRI) of the offender’s head were adduced in the proceedings. Expert witnesses explained that the defendant’s brain scans showed some abnormalities that may have caused issues such as misinterpretation of events, poor decision-making and impulsive behaviour.2
The brain imaging evidence was used to support a lower punishment, but the court found that ‘the irreversible condition of [the brain], raises the question of his dangerousness in the future’.3 The court concluded that since McCann posed a continuing danger to society once released, his brain condition required that he should be kept in prison for a longer period.4 In effect, this was a harsher punishment.
This case was not the first case in which neuroscientific techniques were presented in court as evidence and it has now been at least 40 years since the first use of neuroscience in Australian courts.5 This early episode embodies some of the uncertainties inherent in the legal uptake of what might be described as neuroscientific evidence. In response to such evidence, some scholars have claimed that brain imaging evidence, or more generally, neuroscientific evidence, has a dual potential: it may serve to either aggravate or mitigate punishment. In consequence, the decision about whether to adduce and rely on neuroscientific evidence is risky.6 This contention about the uncertain implications of reliance on neuroscience for guilt, therapy and punishment is known
1 [2012] NSWSC 1462. 2 Id. at [27]. 3 Id. at [36]. 4 Id. at [37] - [40]. 5 According to this study. 6 See, AS Barth, ‘A double-edged sword: The role of neuroimaging in federal capital sentencing’, in American journal of law & medicine, vol. 33, 2007, 501–522; DW Denno, ‘The Myth of the Double- Edged Sword: An Empirical Study of Neuroscience Evidence in Criminal Cases’, in Boston College Law Review, vol. 56, 2015, 493–551. 2
among attentive scholars as the ‘double-edged sword’ effect of neuroscientific evidence. It is one of the many claims associated with the increasing use of neuroscience in law.
This thesis offers an empirically oriented review of the ‘double-edged sword’ and six additional claims about the introduction and use of neuroscientific evidence during sentencing in New South Wales courts. Before moving to introduce the ‘claims’ at the centre of this empirical analysis, it is useful to explain what is meant by neuroscience, how neuroscience and criminal law are connected and what are some of the issues and shortcomings with existing research in this field.
Neuroscience, ‘the scientific study of the nervous system and the brain’,7 is a relatively new area of research that involves using tools, such as brain scans, to visualise the structure and function of the brain and to explain the relationship between the brain, mind and behaviour.8 Neuroscience entered the legal sphere and became an area of increasing legal interest by providing novel ways of explaining the mental states and behaviours of individuals, and a new field, neurolaw, emerged. Neurolaw is a multidisciplinary area of research that investigates the application of neuroscientific discoveries to law. This application is evident particularly in criminal law, where considerable emphasis is placed on regulating criminal behaviour and understanding the mental state of persons. Several experimental studies suggested that neuroscientific evidence can detect abnormalities of the brain that might affect an individual’s behaviour and help to explain the criminal conduct at the time a crime is perpetrated and predicting future mental states in relation to recidivism.9
7 Cambridge University Press, ‘Cambridge Dictionary’,
Forty years after the first use of neuroscience in Australian courts,10 it is timely to reflect on how this relatively recent and developing scientific area has been used. What types of neuroscientific evidence have been introduced in courts? Do courts treat neuroscientific evidence different to other forms of evidence (e.g. psychological assessment)? Further, in a broader sense, how do claims about the use of neuroscience appear in practice? For instance, with respect to double- edged sword claim, does neuroscientific evidence, in fact, contribute to either a harsher or more lenient sentence?
The dearth of empirical research prevents us from understanding the actual use of neuroscience in Australian criminal courts and from examining many questions and claims posed in the neurolaw literature. To fill this gap, the aim of this thesis is to investigate how neuroscientific evidence is being used in sentencing matters in New South Wales and, more specifically, to determine whether the use of neuroscience in courts is consistent with the prominent neurolaw claims considered in this study.
The neurolaw claims in this study (and in general) can be sorted into three basic categories. First, there are predictive claims that assert how neuroscientific evidence will be used or will influence the application of criminal law principles in the future. Secondly, there are descriptive claims about how neuroscientific evidence is being used in practice. Third are normative claims that assert how neuroscientific evidence ought to be used in practice.11 It is noteworthy that this study
2004, 27–42; MP Laakso et al., ‘Prefrontal volumes in habitually violent subjects with antisocial personality disorder and type 2 alcoholism’, in Psychiatry Research: Neuroimaging, vol. 114, 2002, 95– 102; MA Bobes et al., ‘Linkage of functional and structural anomalies in the left amygdala of reactive- aggressive men’, in Social Cognitive and Affective Neuroscience, vol. 8, 2013, 928–936; SM Bannon, KL Salis & K Daniel O’Leary, ‘Structural brain abnormalities in aggression and violent behavior’, in Aggression and Violent Behavior, vol. 25, 2015, 323–331; DA Pardini et al., ‘Lower amygdala volume in men is associated with childhood aggression, early psychopathic traits, and future violence’, in Biological Psychiatry, vol. 75, 2014; V Leutgeb et al., ‘Brain abnormalities in high-risk violent offenders and their association with psychopathic traits and criminal recidivism’, in Neuroscience, vol. 308, 2015; K Schiltz et al., ‘High prevalence of brain pathology in violent prisoners: A qualitative CT and MRI scan study’, in European Archives of Psychiatry and Clinical Neuroscience, vol. 263, 2013, 607–616; A Dawson et al., ‘Capacity, control and responsibility in Parkinson’s disease patients with impulse control disorders: Views of neurological and psychiatric experts’, in International Journal of Law and Psychiatry, 2018,
is not adopting an empirical approach in order to determine whether normative claims have ethical merit (and can be normatively justified). The empirical approach in relation to the normative claims is to examine whether the use of neuroscience in NSW sentencing accords with the prescriptions of the normative claim. In other words, this thesis aims to empirically — by reviewing sentencing decisions — understand whether judicial officers are using neuroscience in ways that are consistent with the prescriptions associated with normative claims.
Below, I will outline the seven neurolaw claims considered in this study and the empirical questions that this study asks about each claim to test them.
First, the double-edged sword effect of neuroscientific evidence is a descriptive claim that asserts the same piece of neuroscientific evidence – such as proof of brain impairment – can support either a more severe or more lenient sentence. The neuroscientific evidence may indicate that the offender has a lower capacity to control their behaviour or understanding the wrongfulness of their act and consequently the court may reduce the punishment. Conversely, it may also lead the court to increase the sentence in order to protect the community from an individual whose brain defect may result in recidivism. Accordingly, this thesis examines whether the use of neuroscience in practice, in fact, contributes to either increases or decreases in the sentence.
Secondly, some scholars make the normative claim that neuroscience should be used for the purposes of promoting rehabilitation (neuro-rehabilitation).12 David Hodgson suggests that the greatest potential of neuroscience is in the area of treatment and rehabilitation of offenders: ‘as scientific research produces insights into how best to go about rehabilitating offenders, we should be prepared to accept and apply these insights’.13 As such, where evidence indicates brain abnormalities that may contribute to criminal behaviour, neuroscience should be used to treat the brain impairment and reduce the risk of recidivism. This study, therefore, examines whether in sentencing, neuroscience is used for the purposes of promoting the rehabilitation of offenders – i.e. whether courts’ decision accords with prescriptions of Hodgson’s normative claim.
Thirdly, is a predictive claim that concerns about how neuroscience will change the application of criminal law principles. Joshua Greene and Jonathan Cohen argue that the growing neuroscientific knowledge will indicate that behaviour is a result of the mechanistic function of
12 D Hodgson, ‘Guilty mind or guilty brain? Criminal responsibility in the age of neuroscience’, in Australian Law Journal, vol. 74, 2000, 661–680 (p. 676). 13 Hodgson, 661–680 (p. 676). Hodgson also mentions the early recognition and minimization of risk factors as part of the greatest contributions of neuroscience in criminal law, however, as they are not related to this research they were not discussed. 5
the brain, rather than of our free will. In their view, the brain automatically processes our decisions and actions and free will is merely an illusion.14 This view suggests that offenders do not deserve to be punished for conduct that is the result of the workings of a mechanical brain, and all conduct results from such workings. Greene and Cohen have predicted that it is likely neuroscience will change principles of criminal law and how the justice system deals with offenders in the future.15 This study examines whether the use of neuroscience in practice has changed the way courts sentence offenders, and if there is any sign that indicates the predicted transformation is taking place.
The fourth claim pertains to the challenges that arise due to neuroscience’s use of scientific language. Stephen Morse argues that there is a mismatch between the languages used respectively by neuroscience and law.16 To explain an individual’s behaviour, neuroscience uses mechanistic, cause and effect language, while the law focuses on people’s minds and their mental state variables (e.g. impaired judgement). According to Morse, the law’s view on criminality is that ‘brains do not commit crime; people commit crime’.17 Morse argues that for neuroscience to be compatible with a legal system, it should be translated into legal language. Otherwise, it may lead to misconceptions and misunderstandings by judicial officers.18 For instance, in the case where an offender commits crime and brain images show a brain tumour, some courts may mistakenly reduce the punishment due to the mere existence of a tumour, despite the fact that the tumour may not have interfered with mental faculties that are relevant to the consideration of criminal responsibility at law – such as capacity to control behaviour or understanding right from wrong. Therefore, according to Morse’s claim, the influence of a tumour should be translated into legal language such as how a tumour affects the capacity to control behaviour and understanding of the individual at the time of the crime.19 The related empirical question in this study asks where neuroscience is discussed in sentencing, whether neuroscience is in fact translated into legal language.
14 J Greene & J Cohen, ‘For the law, neuroscience changes nothing and everything’, in Philosophical Transactions of the Royal Society B: Biological Sciences, vol. 359, 2004, 1775–1785. 15 Greene and Cohen, 1775–1785. 16 S Morse, ‘Lost in Translation?: An Essay on Law and Neuroscience’ in Law and Neuroscience, Current Legal Issues, M Freeman (ed), Oxford University Press, 2011, pp. 529–562. Also see, Hodgson, 661–680. 17 SJ Morse, ‘Brain overclaim syndrome and criminal responsibility: A diagnostic note’, in Ohio St. J. Crim. L., vol. 3, 2005, 397–412 (p. 397). 18 S Morse, ‘Lost in Translation?: An Essay on Law and Neuroscience’ in Law and Neuroscience, Current Legal Issues, M Freeman (ed), 2011, XIII, 529. 19 Some may argue that Morse’s normative claim involves both moral and legal components. As previously discussed, whether these are classified as normative claims do not affect the evaluation of neurolaw claims in practice and are hence not relevant to aims of this study. As such, I will not elaborate further on these arguments. 6
The last three claims are concerned with the reliability of neuroscientific evidence: the first claim provides that brain imaging evidence is subject to many limitations and it is not reliable enough to prove a fact about a mental state (e.g. the offender could not control their act).20 As such, imaging evidence should not be taken as sole proof of a mental state (e.g. poor judgement and impulsive behaviour).21 Instead, neuroscientific evidence should be admitted and relied upon only where it is supported by other forms of evidence, such as behavioural and psychological evidence. (convergence of evidence)22 This empirical study investigates cases where neuroscience is introduced in sentencing decisions and asks whether in fact it is considered in conjunction with other forms of evidence to prove facts on the mental state of the offender.
The second claim about the reliability of neuroscientific evidence is a descriptive claim and suggests that judicial officers may be unduly persuaded by neuroscientific evidence and may inadvertently overestimate the strength of such evidence—taking it to be more convincing and more valuable than can be justified scientifically. That is, they may overlook uncertainty in neuroscientific findings, particularly with imagining techniques.23 For instance, courts may treat brain images as objective evidence even though the meaning depends on the subjective interpretations of expert witnesses—who may draw different inferences from the same brain image.24 Hence, courts may give undue weight to neuroscience in determining the severity of the sentence and they may also find neuroscientific evidence to be more credible than is warranted and unreasonably prefer it where it is in conflict with other evidence. To empirically evaluate this claim, it can be challenging (and sometimes impossible) to determine if the court is ascribing undue weight to neuroscientific evidence. Such assessment requires a normative justification of the appropriate weight that should be given to neuroscientific evidence and to ascertain how much
20 See, B Voytek et al., ‘Dynamic neuroplasticity after human prefrontal cortex damage’, in Neuron, vol. 68, 2010, 401–408; D Szucs & JPA Ioannidis, ‘Empirical assessment of published effect sizes and power in the recent cognitive neuroscience and psychology literature’, in PLoS biology, vol. 15, 2017, e2000797; S Morse, ‘Neuroscience in Forensic Contexts: Ethical Concerns’ in Ethics Challenges in Forensic Psychiatry and Psychology Practice, EEH Griffith (ed), New York, Columbia University Press, 2018, pp. 132–158. 21 W Glannon, ‘The Limitations and Potential of Neuroimaging in the Criminal Law’, in The Journal of Ethics, vol. 18, 2014, 153–170; TR Brown & ER Murphy, ‘Through a scanner darkly: functional neuroimaging as evidence of a criminal defendant’s past mental states’, in Stanford law review, vol. 62, 2010; N Aggarwal & E Ford, ‘The neuroethics and neurolaw of brain injury’, in Behavioral sciences & the law, vol. 31, 2013, 789–802; C Scarpazza et al., ‘The charm of structural neuroimaging in insanity evaluations: guidelines to avoid misinterpretation of the findings’, in Translational Psychiatry, vol. 8, 2018, 227–237. 22 C Scarpazza et al., ‘The charm of structural neuroimaging in insanity evaluations: guidelines to avoid misinterpretation of the findings’, in Translational Psychiatry, vol. 8, 2018, 227. 23 DP McCabe & AD Castel, ‘Seeing is believing: The effect of brain images on judgments of scientific reasoning’, in Cognition, vol. 107, 2008, 343–352; DS Weisberg et al., ‘The seductive allure of neuroscience explanations’, in Journal of cognitive neuroscience, vol. 20, 2008, 470–477; Morse, pp. 132–158. 24 J Dumit, ‘Objective brains, prejudicial images’, in Science in Context, vol. 12, 1999, 173–201. 7
weight the court actually gave to this evidence. This study, therefore, looks for clearer examples of the persuasive power of neuroscience, such as examining whether the court refers to brain imaging as an objective form of evidence.
The third claim with respect to the reliability of neuroscience questions, whether particular types of experts and specialties assist the court. Since brain images rely on subjective interpretations by expert witnesses, courts play an important role in the production, reception and use of this evidence. Further, a qualified expert can provide information about the boundaries and limitations of imaging techniques, methods and error rates. This normative claim suggests that psychiatrists, psychologists and social workers do not have the requisite expertise enabling them to discuss brain imaging evidence and thus their reports and testimony are not credible or sufficiently reliable (area of expertise claim).25 The empirical question at the base of this claim is whether psychiatrists, psychologists and social workers interpret brain imaging evidence and present the results in courts?
In order to examine the neurolaw claims, this thesis presents a systematic, large-scale empirical analysis of 214 sentencing judgements that referenced neuroscientific evidence before 2016. The study also provides some qualitative analysis of case studies that illustrate how neuroscience is used in these judgements, and also presents quantitative data on the use of neuroscience in criminal cases (e.g. prevalence of different types of neuroscientific evidence) and more specifically with respect to neurolaw claims (e.g. for the double-edged sword claim, how often neuroscience supports a harsher or a more lenient sentence).
25 JD Bremner, Brain imaging handbook, WW Norton & Co, 2005; D Amen & L Flaherty, ‘Is brain imaging clinically useful for psychiatrists?’, in Clinical Psychiatry News, vol. 34, 2006, 11; JA Silva, ‘Forensic psychiatry, neuroscience, and the law’, in Journal of the American Academy of Psychiatry and the Law, vol. 37, 2009, 489–502; JC Moriarty & DD Langleben, ‘Who Speaks for Neuroscience- Neuroimaging Evidence and Courtroom Expertise’, in Case W. Res. L. Rev., vol. 68, 2017, 783–804. 8
Figure 1-1: There are seven claims that are examined in this study. Prior to this research none of these claims was examined in Australia based on large-scale empirical data.
In the process of selecting the neurolaw claims for this study (discussed above), several factors were considered. There was an intention to collect claims that have been the centre of scholarly discussions and concerns. As such, the outcome of this research will be relevant to the work of many neurolaw scholars. Further, it was not possible to evaluate some neurolaw claims that were only relevant to the legal rules of particular jurisdictions. For instance, this study could not evaluate claims concerning the use of neuroscience in capital punishment cases, as these do not exist in Australia.26 Another consideration was the unity of the claims and that they were fundamentally connected. All the claims in this study are associated with ‘criminal law’ (rather than Civil law), ‘sentencing’ (after the trial) and ‘offenders’ (not the victims).
The evaluation of each neurolaw claim in this study has application both in isolation and with relation to other work. For instance, this study discusses whether neuroscience has a double-edged sword effect and lawyers might use this knowledge in designing strategies and legal arguments. This study may also show how much consideration courts are giving to the rehabilitation of offenders, and further studies might follow up this line of work and make a normative argument about the extent to which rehabilitation ought to be pursued. This study can also examine whether neuroscience presents special challenges, such as whether findings suggest that courts overestimate the weight of such evidence, or whether neuroscientific evidence is not translated into legal language and may result in misconceptions and misunderstandings. Finally, this study can show if neuroscientific evidence has stimulated changes in the application of legal rules.
26 For instance, see, NA Farahany, ‘Neuroscience and behavioral genetics in US criminal law: an empirical analysis’, in Journal of Law and the Biosciences, vol. 2, 2016, 485–509. 9
These conclusions may call for new policies and inform guidelines on the use of neuroscience in courts.
While neuroscientific evidence has been admitted and relied upon for decades in cases across various jurisdictions, neurolaw claims have often not been substantiated by reports of how neuroscience is used in practice.27 In the absence of such empirical investigation, the academic discussion has been limited to discussions of theories based on critics’ personal experiences from interacting with the legal system, speculation, and anecdotal evidence on criminal cases or media reports of cases where neuroscience has been discussed.28 Since 2016, a small number of studies across a few jurisdictions, namely Canada, England and Wales, the Netherlands and the US have utilised empirical strategies in large-scale studies to analyse the use of neuroscience in courts (referred to hereafter as empirical neurolaw studies).29 These studies, unlike previous studies which utilised only qualitative analysis, present both quantitative and qualitative assessments of the use and influence of neuroscientific evidence in criminal courts. Apart from one study in the US, which focuses on testing a neurolaw claim (the double-edged sword effect of neuroscience),30 most of these empirical studies have only attempted to provide general understanding on the use of neuroscience in courts with a brief reflection on neurolaw claims.
In Australia, the few studies that have analysed the use of neuroscientific evidence in criminal proceedings tend to be focused on its potential use.31 To date, there are four studies that have drawn on a small number of cases, to examine the use of neuroscience in practice. These four studies are limited by the fact that they have only utilised qualitative techniques to analyse a small
27 In this regard, Owen D Jones et al. (2013) refers to the difficulty of finding clear statistics on the application of neuroscience in courtrooms. OD Jones et al., ‘Law and Neuroscience', in Journal of Neuroscience, vol. 33, 2013, 17624–17630; cited in P Catley & L Claydon, ‘The use of neuroscientific evidence in the courtroom by those accused of criminal offenses in England and Wales’, 2015, 1–40. 28 Farahany, 485–509 (p. 485). For examples of studies that evaluate the potential use of neuroscience in courts, see A Alimardani, ‘Neuroscience, criminal responsibility and sentencing in an islamic country: Iran’, in Journal of Law and the Biosciences, 2018; L Houston & A Vierboom, ‘Neuroscience and law: Australia’, in International Neurolaw, 2012, 11–42; International neurolaw: a comparative analysis, in Springer, TM Spranger (ed), Springer Science & Business Media, 2012. 29 See Denno, 493–551; CH de Kogel & EJMC Westgeest, ‘Neuroscientific and behavioral genetic information in criminal cases in the Netherlands’, in Journal of Law and the Biosciences, vol. 2, 2015, 580–605; Farahany, 485–509; JA Chandler, ‘The use of neuroscientific evidence in Canadian criminal proceedings’, in Journal of Law and the Biosciences, vol. 2, 2016, 550–579; Catley and Claydon, 510– 549; LM Gaudet & GE Marchant, ‘Under the radar: Neuroimaging evidence in the criminal courtroom’, in Drake Law Review, vol. 64, 2016, 577–661. 30 Denno, 493–551. 31 D Chau, ‘Neurolaw – potential applications of fMRI in courts (Master’s thesis)’, Sydney, Australia: Macquarie University, 2015, Retrieved from http://hdl.handle.net/1959.14/1068255; Houston and Vierboom, 11–42. 10
number of cases. Two of these studies have specifically considered neurolaw claims;32 the third provides an overview of the uses of neuroscience at different stages of the criminal law process, and provides a brief analysis of some neurolaw claims.33 The fourth presents an analysis on the use of neuroscience in sentencing.34
Concerns about our limited understanding on the use of neuroscience in practice, and gaps in the literature more generally, may be significant in light of the fact that results from empirical studies, in particular those large-scale studies conducted in other countries, have identified inconsistencies between the use of neuroscience in practice and the premises of neurolaw claims.35
Taking into account the neurolaw claims discussed above, along with the gap in the literature, I will introduce my research questions; the originality and significance of the research; as well as the thesis outline.
1. Research questions
The main research question to this study asks:
How and for what purposes is neuroscientific evidence being used in the sentencing procedures in the criminal courts of New South Wales?
This main question is addressed by considering the following sub-questions of claims from the neurolaw literature:
Double-edged sword
Does neuroscientific evidence serve to either aggravate or mitigate sentence?
Neuroscience and legal language
Do the sentencing decisions suggest that the results of neuroscientific assessments are being translated into legal language?
32 See, A McCay & J Kennett, ‘Neuroscience and Punishment: From Theory to Practice’, in Neuroethics, 2019; A McCay & CJ Ryan, ‘Issues pertaining to expert evidence and the reasoning about punishment in a neuroscience-based sentencing appeal’, in International Journal of Law and Psychiatry, 2018. 33 See, A Alimardani & J Chin, ‘Neurolaw in Australia: The Use of Neuroscience in Australian Criminal Proceedings’, in Neuroethics, 2019. 34 Page, 659–691. 35 See, for example, Denno, 493–551. 11
Reliability of neuroscience
In cases where neuroscientific evidence is admitted and relied upon, is this evidence corroborated by other forms of evidence? (convergence of evidence)
Do psychiatrists, psychologists and social workers interpret brain imaging evidence and present these results in courts? (area of expertise claim)
Do the sentencing decisions suggest that courts find neuroscientific evidence overly persuasive?36 (persuasive neuroscience)
Neuro-rehabilitation
Do sentencing courts use neuroscientific evidence for the purposes of promoting rehabilitation?
Green and Cohen
Is there evidence to suggest that Greene and Cohen's prediction – that neuroscience will revolutionise the way the criminal justice system deals with offenders – has already occurred or is taking place?
2. Research originality and significance
The use of neuroscientific evidence in criminal courts has been a topic of substantial debate, and it is important to understand how and in what way this evidence is being used in practice. A systematic examination of the use of neuroscientific evidence in Australian courts has yet to be conducted. Accordingly, this project is the first of its kind in Australia, and one of few studies worldwide, that uses an empirical methodology to systematically collect and analyse criminal cases where neuroscientific evidence is discussed, rather than theorising about the use of neuroscience in criminal courts. Further, it investigates several claims that have not been examined by other empirical studies.
The outcome of this research will generate a firmer empirical basis for neurolaw discussions and lead to a clearer understanding of the use of neuroscientific evidence in courts, especially in New
36 This study does not try to normatively justify what is the appropriate value that should be given to neuroscientific evidence. I will only focus on the clearer instances where judges reasonings suggest they find neuroscientific evidence overly persuasive. For instance, whether courts find brain images as objective evidence. 12
South Wales. This study provides evidence for neurolaw scholars to understand the application of neuroscience in practice, begin to formulate new claims and to give them the opportunity to reject or downplay any empirical claim with a questionable basis that they may have otherwise been inclined to rely on in their scholarly work.
Legal scholars are not the only people who may benefit from this project. Lawyers and judges may gain insight into the process of using neuroscience in courts. For instance, lawyers can use the results of this study to inform their practice (and procedural strategies) and, similarly, judges may draw from the discussion in understanding the use of the evidence as well as the performance of legal institutions.
The results of this study may encourage scholars who are interested in other states in Australia to examine if and how courts’ decisions are affected by neuroscience. Further, it provides a basis for comparative research on the use of neuroscience in different jurisdictions (e.g. USA, Canada, Netherlands and England and Wales) where empirical studies have also been conducted.
The significance of this study lies not only in its informative value, but also in its potential to influence legal reform and practice-oriented guidelines on the use of neuroscience in court.
3. Dissertation outline
Apart from this introduction, this dissertation is divided into nine Chapters. To contextualise the interdisciplinary discussions in this project, an outline of basic neuroscientific concepts and how they are relevant to the law is required. In Chapter 2, I provide a brief overview to the background of the field of neuroscience, different neuroscientific tools and their limitations, the difficulty of (and arguments around) defining neuroscience – as a field of study – and neuroscientific evidence, and how neuroscience became involved in the study of criminal behaviour conducted by criminologists, criminal lawyers, and social scientists. Then, I will explain in detail the literature associated with the neurolaw claims considered in this study. Chapter 2 will then move on to acknowledge the previous empirical neurolaw studies conducted in other jurisdictions, to explore the shortcomings of existing empirical research in Australia, and to explain how the present study will contribute to filling the identified lacuna.
Chapter 3 provides an overview of the Australian criminal justice system and discusses some of the legal rules and concepts relevant to this study (e.g., sentencing legislation and case law pertaining to the aims of, and the factors relevant to, sentencing). This information will provide
13
context to the examination of neurolaw claims in the NSW criminal justice system, particularly for the case analysis phase, and the interpretation of the findings of the research.
Chapter 4 outlines the methodological approach used in this research and its limitations. It begins with a comparison between different sources of data that could be used — e.g. interviews with judges and lawyers, online surveys, court transcripts, and court judgements — and follows by assessing the strengths and weaknesses of each source. Chapter 4 will then justify the decision made to use court judgements involving neuroscientific evidence in examining the research question. This Chapter will then discuss the types of neuroscientific techniques that are deemed to be neuroscientific evidence for the purposes of this study. Chapter 4 will then outline the sampling method, including the key terms used in the search for cases, and justify why some specific cases were then selected or excluded from the dataset. This Chapter will then discuss the textual analysis approach adopted in examining the cases in the dataset and argue for the importance of using computer software to facilitate data analysis in an empirical study of this size.
Chapters 5 to 9 then present a critical analysis of the neurolaw claims against the results of the empirical research discussed in Chapter 4. Each Chapter will be dedicated to one research question and at the beginning of each Chapter the corresponding neurolaw claim will be discussed in more detail. Further, within these five Chapters (5-9), given the paucity of empirical data on the use of neuroscience in Australian courts, and that the main aim of this thesis is to discuss the ways in which neuroscientific evidence is being used in sentencing, and for what purposes such evidence is adduced, this project will attempt to present a more holistic, contextually-based picture of how neuroscience is used in courts concerning the neurolaw claims. For example, with respect to the double-edged sword claim, I will not merely discuss whether neuroscience points towards either harsher or more lenient sentences, but I will also provide specific examples of the factors relevant to sentencing to which this evidence pertains (e.g., future dangerousness and prospects of rehabilitation) and support a harsher or a more lenient sentence. This provides tangible results for neurolaw scholars and judicial officers, increasing the transparency of the study (how judgements were analysed and conclusions were made); and illustrates the complexity of the empirical analysis.
Chapter 5 will discuss the double-edged sword effect of neuroscience. This Chapter first presents a qualitative analysis of how neuroscience contributes to sentencing and more specifically, discusses cases where neuroscience supports a more lenient or a more harsh sentence. It will then discuss whether the quantitative analysis of cases and the use of neuroscience in courts is consistent with what proponents of the double-edged sword claim argue. This Chapter will further 14
discuss how other factors, such as legislation, may influence the conclusion drawn about the double-edged sword claim.
Chapter 6 presents Stephen Morse’s claim about the challenge faced when using neuroscience evidence in courts arising from the difference between languages used by neuroscientists and lawyers. This Chapter will first discuss Morse’s view, and what he means by a mismatch between the language of neuroscience and law, and why he believes there is a need to translate neuroscientific concepts into legal language. I will then argue that some aspects of his claim are too vague and simplistic to be examined in relation to the complexity of judgements. I will limit the scope of the inquiry to accommodate the ambiguities of the claim. The next part of the Chapter presents cases exemplifying neuroscience concepts to varying degrees translated into legal language, and then outlines the overall findings of case analysis with respect to Morse’s claim.
Chapter 7 discusses three claims about the reliability of neuroscientific evidence in court. In the first section of this Chapter, it will provide a brief overview of the evidence rules with respect to the reliability of evidence in the sentencing procedure in NSW to provide context for the examination of these claims. In the second section, it will discuss the convergence of evidence claim and explain why it is important for neuroscientific evidence to be corroborated with other types of evidence in drawing conclusions on the mental state of offenders. It then proceeds to studies and criticisms of the convergence of evidence claim. As this claim concerns the reliability of neuroscientific evidence, this Chapter considers how courts generally treat neuroscientific evidence in sentencing decisions and whether judgements suggest that courts are aware of the criticisms of neuroscientific techniques. The Chapter will then discuss the findings of this study with respect to the convergence of evidence claim. The third section of this Chapter will discuss the claim on the persuasive power of neuroscience — whether courts give undue weight to neuroscientific evidence in sentencing. To avoid the need to declare a normative stance of the appropriate weight that should be ascribed to neuroscientific evidence, the examination of this claim will be limited to highlighting more obvious examples of where neuroscience appears to be used inappropriately, such as cases where the court refers to brain imaging as objective evidence. The fourth and final section of this Chapter is concerned with the expert witnesses’ area of expertise claim. To examine this claim, this Chapter outlines the areas of expertise of those experts who reported on different types of neuroscientific evidence. This section will also discuss whether courts show any concern about expert’s competency during the criminal procedure are aware of the requisite expertise of expert witnesses who are called upon to discuss different forms of neuroscientific evidence
15
Chapter 8 discusses whether neuroscience is used to promote the rehabilitative purposes of sentencing. It will describe reasons why some scholars claim the rehabilitative aspect of neuroscience is important. This Chapter then presents some judgements to demonstrate how neuroscience has been associated with the rehabilitation of offenders in court. It then proceeds to discuss whether findings are consistent with the prescriptions in the neuro-rehabilitation claim.
Chapter 9 reviews Greene and Cohen’s predictive claim that neuroscience will change the application of legal principles. To better understanding this claim, it begins with a brief discussion of the scholars supporting the view that the brain is of a mechanical nature and that free will is an illusion. This Chapter then discusses arguments on the legal implications that these findings may have on criminal law, followed by some conclusions in relation to whether Green and Cohen’s prediction has already occurred or is taking place in the NSW criminal justice system.
Finally, Chapter 10 consolidates the findings from these Chapters and offers additional discussion on whether the neurolaw claims central to this study are consistent with the use of neuroscience in practice. This Chapter also reflects on the contributions that this study provides to the neurolaw literature, and implications and recommendations for future practice, policy reform, and study. Figure 1-2 (below) summarises the process and structure of this thesis.
Figure 1-2: Illustration of the thesis’ structure and how Chapters are connected.
16
Overall, the present study is designed to investigate how neuroscientific evidence is being used in sentencing procedures in New South Wales, and to reflect on whether the use of neuroscience in practice accords with the premises of neurolaw claims. To do so, 214 judgements were collected and analysed. The findings of this study suggest a mixed outcome: two neurolaw claims are consistent with the way that neuroscientific evidence is used in practice — the convergence of evidence claim and the Morse’s claim on the need to translate between the languages of neuroscience and law. Five neurolaw claims, however, are inconsistent with the use of neuroscientific evidence in courts — the double-edged sword claim, the claim on the persuasive power of neuroscience, the neuro-rehabilitation claim, the area of expertise claim, and Greene and Cohen’s predictive claim. A range of methodological limitations are discussed in Chapter 4, therefore these conclusions should be interpreted with a degree of caution.
Examination of the 214 judgements further indicated that there is much more academic work required to evaluate the use of neuroscience in NSW criminal courts. The scope of this research in evaluating some neurolaw claims has provided only a narrow view of the vast interaction of law and neuroscience in courts. Further studies on this topic are therefore required.
17
Chapter 2 Literature Review and Background
18
Introduction
This Chapter has three aims: to provide a background to neuroscience and basic knowledge of relevant neuroscientific concepts and how they are relevant to law; to review scholarly discussions around neurolaw claims considered in this study; and to outline empirical studies conducted in other countries and the existing neurolaw literature in Australia.
The first section of this Chapter will examine the definition and background of neuroscience. Reviewing the history of neuroscience and how contemporary neuroscience was formed is useful for some of the discussions about neurolaw claims and the difficulty of defining neuroscientific evidence. As many of the controversies on the intersection between neuroscience and law centre around brain imaging techniques, I will provide a brief analysis of brain imaging tools; and their differences and limitations. Then, I will move on to discuss other neuroscientific techniques that may be introduced as evidence to the court. The second section will provide a brief background to neurolaw. Then, because some of the neurolaw claims are associated with the brain and criminal behaviour, I will discuss some of the criminological studies that have examined this relationship. In the third section, I will review a number of scholarly discussions of the neurolaw claims considered in this study. Arguments around some of these claims are extensive and occasionally require sophisticated discussion of philosophical concepts. Since this thesis aims to conduct an empirical analysis of the use of neuroscience in courts, I will avoid discussions that might distract from the main objective of this study. Lastly, in the fourth section, I will first refer to the empirical neurolaw studies conducted in other countries and review some of the similarities and differences in their methods. This will help to explain how and where the present study is distinct from other empirical neurolaw studies but to allow cautious comparisons between the conclusions made in those studies and this thesis. Then, I will move to discuss the current gap in Australia’s neurolaw discourse.
1. Neuroscience: background and techniques
There are many different accounts of the origin of what we now know as contemporary neuroscience, including various key figures, institutional foundations and important discoveries. The term neuroscience was coined by Francis O. Schmitt in 1962, which is the year proposed by some historians as the birth of contemporary neuroscience.1 Neuroscience at its core was founded
1 N Rose & JM Abi-Rached, Neuro: The new brain sciences and the management of the mind, in Princeton University Press, EBSCO Publ, Princeton University Press, 2013. 19
based on the unification/connection of different disciplines.2 Schmitt believed that understanding the brain and its function was not a matter that could be explained within a single area of study, and he brought together different disciplines including physics, biology, mathematics, computer sciences, psychology, psychiatry and neural science in order to better understand the links between the brain, the mind, and behaviour.3
The study of the brain and its relationship with the mind and behaviour were not new matters. Nicolas Rose and Joelle Abi-Rached recognise that earlier studies of nerves, the brain, and mental disorders,4 and the connections between the three together led to founding contemporary neuroscience by Schmitt in 1962.5 I briefly discuss Rose and Abi-Rached’s view on the development of these three areas to contemporary neuroscience.
The term ‘Neuro’ originates from Ancient Greek and means ‘of or relating to nerves or the nervous system’.6 In 1873, Camillo Golgi, who was a key historical figure in the understanding and research on neurons, invented a method to observe neurons for the first time.7 Unlike Golgi, who believed that the nervous system was a single continuous network of neurons, Ramon y Cajal, by developing Golgi’s method, claimed that the nervous system consisted of discrete cells — neurons.8 During the 20th century, many studies shed light on how neurons communicate through electrical and chemical neurotransmitters. As a result of these studies, many fields of study based on the function of neurons were formed. For instance, in 1900, the term neurochemistry — the study of chemicals that influence neuronal function — was used for the first time, and by 1906, neurobiology — the study of the physiology of the nervous system and the brain — had emerged.9 This shows that the neuro prefixed disciplines had appeared long before the term neuroscience was coined.10
2 WM Cowan, DH Harter & ER Kandel, ‘The emergence of modern neuroscience: Some implications for neurology and psychiatry’, in Annual Review of Neuroscience, vol. 23, 2000, 343–391. 3 G Adelman, ‘The neurosciences research program at MIT and the beginning of the modern field of neuroscience.’, in Journal of the History of the Neurosciences: Basic and Clinical Perspectives, vol. 19, 2010, 15–23. 4 Rose and Abi-Rached refer to it as madness. It appears mental disorders and their variety were recognised in the last centuries and before that they all fell under category of madness. 5 Rose and Abi-Rached. 6 OED (online), ‘neurology’, in OED Online. Oxford University Press,
The study of the brain was the other route that led to contemporary neurosciences. Many historical, and even in some cases ancient, documents refer to the cerebral cortex and the role of brain injury in abnormalities such as aphasia and seizures.11 Current studies in neuroscience are based on research on the brain that developed in Europe during the 19th century. This was when the brain became gradually recognised as an organ, similar to other organs, where different parts were responsible for different functions, rather than what it was thought of as previously – a mysterious mass of fat hiding behind the skull.12 During the 20th century especially after technologies of visualisation (brain imaging tools) developed, which allowed the study of a living brain many scholars centred their studies on finding the associations between different locations of the brain, the mind and behaviour (I will discuss imaging tools later in this Chapter).13
The third route to contemporary neuroscience was the understanding of mental illness through the brain. Development in understanding that there is a connection between organic problems and symptoms such as mania, impaired speech and memory began in the 19th century. Some scholars studied the brains of cadavers from asylums to find the link between mental afflictions and the brain. Others studied injured brains such as Phineas Gage, whose behaviour and personality experienced changes after an accident where a rod pierced his skull through his eye.14 Despite these studies, the location of specific disorders in the brain was not clear. For instance, in the 20th century, while physical therapies – like psychosurgery and shock therapies – were becoming popular in asylums, it was not clear which, and how, different areas of the brain were responsible for troubled minds (the problem of localisation).15 Some studies focused on biological treatments for psychiatric disorders within the brain (psychopharmacology), and psychiatric drugs came to replace many physical approaches to brain interventions.16
Through history, scientists from different disciplines exposed features of the brain: understanding the role of chemicals in the brain and how neurons communicate (path through the neurons),
from neuroscience as it focuses on diseases (what is wrong in the brain) and how to deal with those problems. OED (online). 11 C Gross, ‘Neuroscience, early history of’, in Journal of Enclyclopedia of Neuroscience, 1987, 843–847,
understanding that the brain consists of different areas that are responsible for different functions such as cognitive, emotional, or volitional states (path through the brain), and understanding that disorders of the mind have physical underpinnings in the brain (path through mental disorder). Contemporary neuroscience is a recognition of a field study emerging from the development and intersection of the scientific study of neurons, the brain and mental disorders arising from the brain through the lens of different disciplines.
Following the work of Schmitt, the human brain drew increasing interest from different backgrounds to neuroscience. The Society for Neuroscience (SfN) which was founded in 1969, experienced dramatic growth in the number of members incorporating research from a variety of disciplines including science, mathematics, psychology, medicine, and computer sciences.17 Further, many disciplines were subsequently implicated by the results of neuroscientific studies and a variety of interdisciplinary areas formed including social neuroscience,18 Neuro-ethics,19 Neurophilosophy, Neuro-economics,20 Neuro-education,21 neuropsychiatry — the study of mental disorders associated with neurology – 22 and neuropsychology — the study of mental function and behaviours that are attributable to the brain’s structure and function.23
Development of neuroscience has also garnered much interest from the media, society, and politicians.24 For instance, President George H.W. Bush proclaimed the 1990s as the ‘Decade of the Brain’.25 Indeed, the last two decades witnessed some of the greatest initiatives accelerating our understanding of the brain. These projects received considerable support and funding (e.g.
17 Society for Neuroscience, Mission and Strategic Plan
the Human Brain Project received $1.3 billion USD)26 which aimed to understand the complex interactions of the brain.27 Some of these projects provided free data for other researchers to access in order to expedite brain research in different areas (e.g. The Allen Human Brain Atlas).28
Despite the effort and the expenditure on neuroscientific research, high hopes in solving the mystery of the brain were not as successful as was expected. From when Cajal invented a technique to visualise individual cells in the brain, to know where modern neuroscientific tools can help scientists to map the function of the brain, it has become more and more apparent ‘that the brain was a system of complexity for which we had no equivalent, and no language to capture it’.29 The brain, (purportedly) has 100 billion neurons 30 and trillions of connections.31 This is why some refer to the brain as the most complex machine in the world.32 Our growing knowledge of the brain did not result in solving many of the brain puzzles. Instead, it unveiled more complex aspects of them. As Rose puts it, ‘the more we know, the more we know we don’t know’.33
Nonetheless, our understanding of the brain that has developed within the last few decades is vastly superior to what was known in preceding. A considerable aspect of the recent progress made has been the revolutionising impact that imaging tools have had on the field of study. Before the advent of these tools, the study of the brain was limited to autopsies of dead people, exceptional cases such as Gage, and influence of brain injuries caused by stroke or war on behaviour, which provided some clues about the anatomy and function of the brain.34 Research on the living brain in the absence of proper tools was not practical. During the 20th century, particularly following technological developments in the 1970s, neuroimaging tools have allowed scientists to shift their examination of the brain from those in dead people and exceptional cases to the study of the brain of the living.35
26 See, The Human Brain Project (https://www.humanbrainproject.eu) 27 For example see, The Blue Brain Project (http://bluebrain.epfl.ch) and the Brain Research through Advancing Innovative Neurotechnologies (BRAIN) Initiative (https://www.braininitiative.nih.gov), 28 EH Shen, CC Overly & AR Jones, ‘The Allen Human Brain Atlas: comprehensive gene expression mapping of the human brain’, in Trends in neurosciences, vol. 35, 2012, 711–714. 29 D Eagleman, The Brain: The Story of You, in Canongate, Pantheon, 2015, p. 71. 30 S Herculano-Houzel, ‘The human brain in numbers: a linearly scaled-up primate brain’, in Frontiers in human neuroscience, vol. 3, 2009, 1–11. 31 Eagleman, The Brain: The Story of You. 32 HT Greely & AD Wagner, ‘Reference guide on neuroscience’ in Reference Manual on Scientific Evidence, 3rd ed., Washington, DC, National Academies Press, 2011. 33 N Rose, Our Psychiatric Future: the politics of mental health, in Polity, Cambridge, John Wiley & Sons, 2018, p. 113. 34 OR Goodenough & M Tucker, ‘Neuroscience Basics for Lawyers’, in Mercer L. Rev., vol. 62, 2011, 945–1335. 35 Goodenough and Tucker, 945–1335. 23
Brain imaging tools
We truly live in the golden age of neuroscience. Advances in technology over the past 20 years have given modern neuro-researchers tools of unprecedented power to probe the workings of the most complex machine in the universe (as far as we know). —Octavio S. Choi.36
One of the most common methods of studying the brain is through imaging. Looking into the living brain non-invasively, was more or less a figment of the imagination, but it became possible when sophisticated tools emerged. Imaging techniques can be divided into two basic categories. Structural and functional. Structural imaging looks at the anatomy of the organ. A good illustration of this is the X-ray of a hand where doctors can see, for instance, if any bones are broken. Structural imaging of the brain, can also allow us to detect abnormalities such as brain tumours.
Functional techniques are more advanced and sophisticated ways of studying the brain. Put simply, brain cells, depending on their level of activity (function), use a different degree of energy and oxygen (brain metabolism), which is reflected in changes in blood flow.37 As such, the more an area of the brain is activated, the more blood circulates in that area to deliver energy and oxygen. By measuring variation in brain metabolism or blood flow, we can identify which areas of the brain are functioning more than other areas.38 Functional images are particularly useful in cases where, for example, damage to the brain does not change the structure of the brain, but does impair the function of some tissues. Many types of brain damage can produce this result.39
In the following, I will provide a simplified explanation of three structural imaging techniques, namely X-ray, CT and MRI; and five functional techniques, electroencephalogram (EEG), Magnetoencephalography (MEG), functional Magnetic Resonance Imaging (fMRI), Positron Emission Tomography (PET), and Single Photon Emission Computed Tomography (SPECT) in chronological order of their developments.
36 OS Choi, ‘What Neuroscience Can and Cannot Answer’, in Journal of the American Academy of Psychiatry and the Law Online, vol. 45, 2017, 278–285. 37 ME Raichle & MA Mintun, ‘Brain work and brain imaging’, in Annu. Rev. Neurosci., vol. 29, 2006, 449–476. 38 C Asbury, ‘Brain imaging technologies and their applications in neuroscience’, in The Dana Foundation, 2011. 39 LM Gaudet & GE Marchant, ‘Under the radar: Neuroimaging evidence in the criminal courtroom’, in Drake Law Review, vol. 64, 2016, 577–661. 24
X-ray imaging was invented in 1895.40 X-rays evaluate the density of tissues by throwing particles (photons) at the organ and recording what passes through the body. X-ray imaging was widely used for generating images of the body parts (e.g. bones).41 The brain, however was not an appropriate organ for X-rays due to the density of the skull, which particles cannot pass through. Later, through technological developments and improved techniques, it became possible to visualise blood and other fluids in the brain.42 This made X-ray useful as a technique for capturing images of vascular abnormalities and tumours in the brain.43 Nonetheless, this only provided limited information about the brain’s structure.44
From the 1920s onward, functional techniques such as EEGs started to be used as a non-invasive technique to directly evaluate the brain’s activity by recording the brain’s electrical signals.45 MEG is a similar method that records the magnetic fields produced by the brain’s electrical activity. In these techniques, subjects wear a medical cap that measures the brain’s electrical signals and magnetic fields.46
For clinical purposes, EEG and MEG are currently used for diagnosing and managing conditions such as Alzheimer’s disease, depression and seizure disorders (e.g. epilepsy).47 For research purposes, these techniques are used to assess the EEG or MEG signals (i.e. size and pattern) that arise from performing a task, or exposure to a stimulus such as when subjects are shown happy, sad and angry faces.48
Some of the limitations of EEGs and MEGs are that they are not useful in assessing the function of areas deep in the brain that are far from the skull.49 They are also weak at indicating the sources of the electrical activity or magnetic fields.50 Some strengths of both techniques, however, are
40 M Castillo, ‘History and Evolution of Brain Tumor Imaging : Insights through’, in Radiology, vol. 273, 2014, 111–125. 41 Asbury. 42 ME Raichle, ‘A brief history of human brain mapping’, in Trends in Neurosciences, vol. 32, 2009, 118–126. 43 Rose and Abi-Rached. 44 Greely and Wagner. 45 M Rugg, D & M Coles, Electrophysiology of mind: Event-related brain potentials and cognition, in Oxford University Press, 1995. Developments of EEG dates back to 19th century. 46 Greely and Wagner. 47 A Ray & SM Bowyer, ‘Clinical applications of magnetoencephalography in epilepsy’, in Annals of Indian Academy of Neurology, vol. 13, 2010, 14–22; T Montez et al., ‘Altered temporal correlations in parietal alpha and prefrontal theta oscillations in early-stage Alzheimer disease’, in Proceedings of the National Academy of Sciences, vol. 106, 2009, 1614–1619; MA Williams & PS Sachdev, ‘Magnetoencephalography in neuropsychiatry: ready for application?’, in Current opinion in psychiatry, vol. 23, 2010, 273–277. 48 Greely and Wagner. 49 TM Spranger, ‘Neurosciences and the Law: An Introduction’ in International Neurolaw, TM Spranger (ed), Springer, 2012, pp. 1–10. 50 Williams and Sachdev, 273–277; Ray and Bowyer, 14–22; Greely and Wagner. 25
that they are quiet and safe. Though MEG is more expensive than EEG and it is less portable.51 Further, when subjects use MEG techniques, they need to remain still to prevent disrupting in the examination results. EEG, however, is less sensitive to head movement.52 Unlike other functional imaging techniques, EEG and MEG do not produce an actual image of the brain, rather they produce waveforms of electrical and magnetic activities in the brain.53 Nonetheless, these techniques are included with other imaging techniques in many neurolaw studies,54 possibly because they are non-invasive approaches that can extract information about the brain function.
Another structural imaging technique is CT scanning,55 which was introduced in 1971.56 CT scanning uses X-ray technology but captures many images of the brain while rotating around the person. A computer is used to generate a three-dimensional perspective of the brain. CT scanners are relatively quick at producing an image and inexpensive, which makes it a suitable method of imaging in emergency situations to detect life-threatening brain abnormalities – for example, swelling and bleeding.57 CT scanners are generally useful for capturing structural abnormalities such as haemorrhages, tumours, atrophy and lesions in the brain.58
MRI was invented after the CT scan and received approval for clinical use in 1985.59 MRI involves minimal risk in comparison to other structural imaging techniques since it is not dependent on X-rays passing through the body. Instead, it uses strong magnetic fields to generate images of the structure of the brain.60 One advantage of MRI machines in comparison to CT scans is that it can distinguish similar soft tissues and generate higher-resolution images of the brain structure. Since MRI images are detailed and can detect large and small abnormalities of the brain (e.g. tumours, strokes, traumatic injuries, and lesions),61 it is valuable in clinical settings and is considered ‘the most important diagnostic imaging modality in neuroscience’.62 From a research perspective, MRI images are useful in identifying differences between normal and abnormal subjects and their brain developments.63
51 Greely and Wagner. 52 Greely and Wagner. 53 Greely and Wagner. 54 See, JC Moriarty, ‘Flickering admissibility: neuroimaging evidence in the US courts’, in Behavioral sciences & the law, vol. 26, 2008, 29–49. 55 Also known as Computer Assisted Tomography (CAT) 56 Raichle, 118–126. 57 Asbury. 58 Moriarty, 29–49. 59 LS Gold et al., ‘The emergence of diagnostic imaging technologies in breast cancer: Discovery, regulatory approval, reimbursement, and adoption in clinical guidelines’, in Cancer Imaging, vol. 12, 2012, 13–24. It was approved by Food and Drug Administration (FDA). 60 Greely and Wagner. 61 Asbury. 62 Asbury; Rose and Abi-Rached. 63 Greely and Wagner. 26
There are further functional imaging techniques that are different to the ones discussed so far. They are unlike structural imaging techniques (X-ray, CT and MRI), as they can visualise brain activity. However, they are also unlike functional techniques such as EEG, in that they can provide an image of the brain. PET, a functional imaging technique, was invented in 1975.64 PET images assess different levels of metabolic activity in the brain.65 To do this, a substance (such as glucose) is radio-activated and injected into the patient’s body. Then the machine, by tracing the radioactive particles, can illustrate areas of the brain where the radio-activated subject has been metabolised, and thus depleted. The distribution of radio-activated particles is then used as a marker of the level of functioning in different brain areas, which is correlated with mental activity.66 Nowadays, however, PET is primarily used for purposes such as tracking chemicals (i.e. neurotransmitters) in the brain and to determine how drugs influence brain function.67 PET has the potential to assess some forms of brain abnormalities such as stroke, traumatic brain injury, epilepsy68 and Alzheimer’s disease.69
SPECT, a similar device to PET, was introduced in the early 1980s.70 In comparison to PET, SPECT is less accurate and has a limited resolution, however, is less expensive and relatively more commonly used for clinical purposes.71 Further, the metabolic components used in SPECT take longer to metabolise, and therefore, the technique can measure brain activity over a longer period of time, and can provide more information about the brain’s activities.72 An issue with radiographic methods of neuroimaging, such as PET and SPECT, is the differences between individuals in metabolising substances that may add a further source of variance to the analysis of the results of the examination.73 Further, these techniques do not directly measure neural activity. Instead, they generate images of areas in the brain based on levels of radio-activated substances.74
64 LH Portnow, DE Vaillancourt & MS Okun, ‘The history of cerebral PET scanning: from physiology to cutting-edge technology’, in Neurology, vol. 80, 2013, 952–956. The invention of PET has a long history and some of the device’s early iterations predate CT scanning, which was developed in 1971. See, Portnow, Vaillancourt and Okun, 952–956; Rose and Abi-Rached. 65 Asbury. 66 O Jones, J Buckholtz & J Schall, ‘Brain imaging for legal thinkers: A guide for the perplexed’, in Stanford Technology Law, 2009,
After PET and SPECT scans, fMRI, another functional imaging technique, emerged in 1990. fMRI uses the same technology as MRI machines,75 however, it measures the brain’s cellular function as a correlate of mental activities in different areas of the brain.76 Unlike PET and SPECT, fMRI is not dependent on radioactive particles and hence, it is considered a safer approach to studying the brain. In this technique, a subject is placed in the fMRI machine and invited to engage with a specific task. The fMRI machine, using magnets, is able to record changes in oxygen levels in certain areas of the brain. The assumption is that more oxygen will be delivered to areas of the brain that are undergoing higher levels of mental activity.77 Thus, the machine uses measures of blood oxygen level variation to estimate in real time, with a slight delay, a colourful image of areas of the brain that are supposedly activated.78
Since this technique is based on the level of oxygen carried in the blood, it is called ‘blood oxygen level dependent’ or BOLD.79 As such, fMRI machines do not directly measure the brain cells’ activity correlated with mental activity but measures the oxygen consumed in those areas. In other words, fMRI machines visualise neural activity through indirect markers. In addition, fMRI can only assess brain function at a neural level, while there are a variety of different brain functions,80 which it fails to consider.81 Another weakness of this technique is that fMRI scans are expensive, hence studies are usually limited to a small number of participants and might not represent the general population like some of the other brain imaging studies.82
Despite these limitations, fMRI can generate high-resolution images of neural activity in real time (while participants conduct a specific task), hence it provides invaluable information for research on brain function. The technique has been utilised in numerous research studies. In 2008 it was found that a search in a database (ISI/Web of Science) using the keyword ‘fMRI’83 resulted in over 19,000 peer-reviewed studies.84 By 2010 this number had increased to over 54,000 articles.85 These included studies measuring the brain activity of individuals in response to work of different painters, music producers, television commercials, love, hate, an act of will (while individuals are
75 S Ogawa et al., ‘Oxygenation-sensitive contrast in magnetic resonance image of rodent brain at high magnetic fields’, in Magnetic resonance in medicine, vol. 14, 1990, 68–78. 76 SA Tovino, ‘The confidentiality and privacy implications of functional magnetic resonance imaging’, in The Journal of Law, Medicine & Ethics, vol. 33, 2005, 844–850; Jones, Buckholtz and Schall. 77 Jones, Buckholtz and Schall. 78 Raichle, 118–126. 79 Raichle, 118–126. 80 Such as synaptic pruning. 81 LR Tancredi & JD Brodie, ‘The brain and behavior: limitations in the legal use of functional magnetic resonance imaging’, in American journal of law & medicine, vol. 33, 2007, 271–294. 82 Greely and Wagner. 83 The terms ‘functional MRI’ and ‘functional magnetic resonance imaging’ were also used. 84 NK Logothetis, ‘What we can do and what we cannot do with fMRI’, in Nature, vol. 453, 2008, 869– 878, cited in Rose and Abi-Rached. 85 Rose and Abi-Rached. 28
conducting simple tasks), and those of individuals with illnesses such as addiction and schizophrenia.86
Although brain imaging techniques provide fascinating and useful images of the structure and function of the brain, they have many limitations. I have discussed some of the limitations specific to each of the imaging techniques above. There are some limitations, however, that are also commonly shared between most brain imaging techniques.
Brain imaging studies are generally based on the concept of localisation of function — that different areas (locations) of the brain have different responsibilities (functions).87 Accordingly, if one part of the brain is impaired (either in structure or function), we may expect that that area of the brain cannot carry out its role. For instance, it is believed that the prefrontal cortex (an area of the brain behind the forehead) is responsible for decision-making and managing social behaviour. An impairment in that area may result in behavioural problems such as impulse control disorders and violent behaviour.88
But the notion of localisation of function is not an entirely accurate explanation of brain function. The brain is a very complicated organ, and some brain areas are associated with multiple and overlapping mental functions.89 Consequently, many mental activities of the brain do not take place in one region of the brain, but instead occur through a pattern or patterns of activity in different areas.90 Consequently, impairment in a particular area of the brain does not necessarily indicate an overall malfunction.
Another issue related to neuroimaging is the group to individual (G2i) inferences. Neuroscientific evidence is mostly derived from experiments performed on groups of people, and the result reflects the group’s average.91 For instance, a neuroscientific experiment may be conducted on healthy individuals with supposedly no brain abnormalities to determine the average normal brain
86 Rose and Abi-Rached. 87 See N Rafter, The criminal brain: Understanding biological theories of crime, NYU Press, 2008; FJ Gall, On the Function of the Brain and Each of Its Parts, in Boston: Marsh, Capen and Lyon, 1835, I. 88 Y Yang & A Raine, ‘Prefrontal structural and functional brain imaging findings in antisocial, violent, and psychopathic individuals: a meta-analysis’, in Psychiatry Research: Neuroimaging, vol. 174, 2009, 81–88. 89 M Mesulam, ‘Large-scale neurocognitive networks and distributed processing for attention, language, and memory’, in Annals of Neurology: Official Journal of the American Neurological Association and the Child Neurology Society, vol. 28, 1990, 597–613; W Glannon, ‘The Limitations and Potential of Neuroimaging in the Criminal Law’, in The Journal of Ethics, vol. 18, 2014, 153–170. 90 Glannon, 153–170. 91 DL Faigman et al., ‘G2i Knowledge Brief: A Knowledge Brief of the MacArthur Foundation Research Network on Law and Neuroscience’, in MacArthur Foundation Research Network on Law and Neuroscience, 2017. 29
activation level corresponding to a specific cognitive task. The problem arises when participants that exhibit marked variation from the group average are regarded as normal. As such, if an individual’s brain function with respect to a cognitive task differs from the group average, it does not necessarily mean they have abnormal brain function.92
There is often a considerable amount of overlap between the performance of experimental subjects and the control group,93 and each individual’s brain is formed and influenced by many factors (i.e. genetics and environment) that results in wide inter-individual variability even within normal participants. As such, determining what is abnormal and what is a normal function of the brain is a challenging task.94
The differences between brains may have various causes. Differences between brain function/structure are not necessarily indicative of cognitive impairment. An example of where this occurs is cognitive reserve – a situation where some ‘areas of brain tissue have degenerated, other areas have … compensated or taken over those functions’ and while the brain may appear impaired, it functions normally.95
Individual differences also cause methodological problems. When comparing different brains through complex mathematical analysis, often images of the brain need to be stretched or compressed to allow comparison.96 These techniques may produce false positive or negative results — where the brain structure or function is normal, but the results report abnormalities and vice versa.97
Science, in general, provides aggregate data on empirical research conducted across groups of individuals. Thus, using data from individual cases to make a diagnosis that is relied upon in the courtroom may be problematic.98
92 Faigman et al. 93 SJ Morse, ‘Brain overclaim syndrome and criminal responsibility: A diagnostic note’, in Ohio St. J. Crim. L., vol. 3, 2005, 397–412. 94 D Reeves, MJ Mills & SB Billick, ‘Limitations of brain imaging in forensic psychiatry’, in The journal of the American Academy of Psychiatry and the Law, vol. 31, 2003, 89–96; SB Bullick & SP Sullivan, ‘Neuroimaging in child and adolescent psychiatry’ in Principles and Practice of Forensic Psychiatry, R Posner (ed), 2nd ed., CRC Press, 2003, pp. 473–481; Jones, Buckholtz and Schall. 95 D Eagleman, The Brain: The Story of You, in Canongate, Pantheon, 2015, p. 29. 96 D Linden, Neuroimaging and Neurophysiology in Psychiatry, Oxford, United Kingdom, Oxford University Press, 2016. 97 ST Grafton et al., ‘Brain scans go legal’, in Scientific American Mind, vol. 17, 2006, 30–37. 98 Faigman et al. 30
The reliability of functional brain imaging evidence has been subjected to more criticism than structural brain imaging evidence. Functional imaging not only shares some of the limitations of structural brain imaging but has, in addition, other limitations. As noted before, functional machines measure, for instance, neurons’ oxygen or energy consumption. This means that they do not directly examine neuron activity, but rather their oxygen and energy usage as a surrogate marker. Additionally, functional machines such as fMRI, scan over millions of voxels (a unit of measurement – each voxel can contain hundreds of thousands of neurons) of data.99 An fMRI scanner does not calculate the activity of individual neurons, but rather the average activity of all neurons within each voxel. While there might be zero activity in some neurons and significant activity in others, data on the variation between individual neurons are lost.100 Another issue with functional imaging is that there is much activity in the brain that may not be relevant to the focus of a particular neuroscientific experiment, and can confound results. To address this problem, a threshold is often defined to eliminate ‘irrelevant’ activities. The issue with this approach, however, is that some neural activities relevant to the experiment might also be excluded from the evaluation.101
There are different methodological issues in functional imaging experiments. For instance, a statistical analysis comparing the differences between subjects and control groups is challenging and requires different strategies, each of which may alter the accuracy and validity of the result in different ways.102
Overall, despite their limitations, brain imaging techniques, have provided us with invaluable data on the relation between the structure and function of the brain, and nature of human cognition and volition. These findings have been implemented in various fields, one of which is law. Since the law is often interested in understanding an individual’s mind and behaviour, neuroscience draws the attention of many legal scholars.
Neuroscientific evidence
Although the majority of scholarly discussions in the neurolaw literature are centred on brain imaging techniques, some scholars use a broader definition for neuroscientific evidence. Defining
99 GK Aguirre, ‘Functional Neuroimaging: Technical, Logical, and Social Perspectives’, in Hastings Center Report, vol. 44, 2014, S8–S18. 100 A Mechelli et al., ‘Voxel-based morphometry of the human brain: methods and applications’, in Current medical imaging reviews, vol. 1, 2005, 105–113. 101 Aguirre, S8–S18. 102 NJ Schweitzer & MJ Saks, ‘Neuroimage evidence and the insanity defense’, in Behavioral sciences & the law, vol. 29, 2011, 592–607. 31
what constitutes neuroscientific evidence is not a simple task, which could explain why there are differing opinions in this regard. From looking at the history of neuroscience, it can be observed that neuroscience is a multidisciplinary area. Notably, some scientists from areas not especially concerned with imaging (e.g. psychology and psychiatry), while using the methods from their own disciplines to study the brain, have also identified themselves as ‘neuroscientists’.103 This integration makes distinguishing between different approaches to studying the brain, and whether they should be considered as ‘neuroscientific evidence’ potentially challenging.
There are two broad categories of neuroscientific evidence: framework evidence and diagnostic evidence. Framework evidence is general knowledge that has been deduced by neuroscientific studies.104 This type of evidence, for example, includes the hundreds of brain imaging studies that have found that the prefrontal cortex is associated with controlling impulses.105 Diagnostic evidence refers to the data used in a specific case to diagnose an abnormality (e.g. impaired prefrontal cortex). This is conducted based on knowledge provided through framework evidence. In other words, diagnostic evidence relies on framework evidence.106 Evidence from both categories is introduced in courts. A form of framework evidence is the case where, for instance, an expert refers to neuroscientific studies to suggest that juveniles’ brains are not yet fully developed and have a lower capacity for rational behaviour compared to adults. In cases where an expert explains that a brain scan shows that a particular individual has a lower capacity to control their behaviour, they are using diagnostic evidence.
Neurolaw scholarly discussions are mainly focused on two forms of diagnostic evidence: imaging evidence and non-imaging evidence. Put simply techniques associate behavioural issues based on brain images. Non-imaging evidence, by using other measurements (e.g. psychiatric medical examination) predicts whether brain abnormalities exist and how those issues may be associated with the behaviour.
Imaging evidence is the type of evidence most widely and conventionally regarded by scholars as neuroscientific evidence. Notwithstanding this agreement, there remains contention around the definition of ‘imaging evidence’. These competing views can be explained using the categories of the narrow and the broad view.107 The narrow view only considers techniques that generate
103 Rose and Abi-Rached. 104 A Alimardani & J Chin, ‘Neurolaw in Australia: The Use of Neuroscience in Australian Criminal Proceedings’, in Neuroethics, 2019,
actual images of the brain such as CT, MRI, PET and SPECT scans, as imaging techniques. The broad view also considers scanning techniques such as EEG and MEG that provide brainwave forms of brain activity as imaging techniques.108 Nonetheless, both forms of imaging tools — those that fall under the narrow and broad views — are commonly recognised as neuroscientific evidence in neurolaw studies.109
The further controversy arises regarding non-imaging evidence. These approaches use neuroscience’s framework knowledge to identify a brain-related issue.110 Paul Catley and Lisa Claydon explain that ‘tests developed in whole or in part on the basis of brain scan findings [framework evidence] but which do not themselves involve scans could arguably fall either side of the line. Artificial distinctions are difficult.’111 Non-imaging evidence includes tests by medical professionals that generally measure how an individual’s brain functions,112 or uncovers brain damage, for instance, by collecting medical history.113 Medical professionals then make conclusions/suggestions about the mental abilities and behaviour of the individual based on the diagnosed brain abnormality.
Despite the fact that non-brain imaging approaches do not involve scanning techniques, they still seek to understand the brain’s function and its influence on behaviour. Further, these approaches are at least partially based on neuroscientific knowledge — i.e. framework evidence. In this way, the findings of other disciplines involved in the study of the brain, as they are informed by neuroscientific knowledge, can also be incorporated into ‘neuroscientific evidence’. It appears that according to this account, some neurolaw studies have considered both brain imaging and non-brain imaging approaches as neuroscientific evidence.114
One common example of a non-imaging technique is neuropsychological testing (including executive function, IQ and memory), which seeks to identify abnormalities in cognitive functioning that are commonly linked to brain impairment.115 Although, as compared with brain imaging tools, neuropsychology is not the focus of neurolaw studies, it is a developing area. One
108 Kellmeyer, 530–554. 109 See, DW Denno, ‘The Myth of the Double-Edged Sword: An Empirical Study of Neuroscience Evidence in Criminal Cases’, in Boston College Law Review, vol. 56, 2015, 493–551; NA Farahany, ‘Neuroscience and behavioral genetics in US criminal law: an empirical analysis’, in Journal of Law and the Biosciences, vol. 2, 2016, 485–509. 110 P Catley & L Claydon, ‘The use of neuroscientific evidence in the courtroom by those accused of criminal offenses in England and Wales’, in Journal of Law and the Biosciences, vol. 2, 2015, 510–549. 111 Catley and Claydon, 510–549. 112 Denno, 493–551. 113 See, Farahany, 485–509. 114 For example, see, Denno, 493–551; Farahany, 485–509. 115 See, JA Chandler, ‘The use of neuroscientific evidence in Canadian criminal proceedings’, in Journal of Law and the Biosciences, vol. 2, 2016, 550–579. 33
recent developing area in neuropsychology is the relationship between impairments of executive function and impulsive behaviour.116 Executive function is associated with the ability to direct behaviour and thoughts, control impulses and emotions, and adapt behaviour based on environmental stimuli and planning.117 In these studies, neuropsychological tests are sometimes more useful than brain imaging techniques. For instance, MRI scans may be able to accurately show that juveniles in general do not have a fully developed brain. In individual cases, however, neuropsychological tests are more reliable than MRI scans in predicting high risk behaviour.118 Further, some scholars claim that unlike brain imaging evidence, neuropsychological testing in assessing the cognition of individuals is widely validated and ‘well characterized in terms of specificity, sensitivity, and predictive values’.119
Despite the utility of categorisation, this system merely draws artificial lines between different types of neuroscientific evidence, which may create further, more complex issues. For instance, when neuroscience is used in courts, the perception that lay people have of the definition of different forms of neuroscientific evidence also needs to be considered. This is an important consideration because they may value different forms of neuroscientific evidence differently. Some may argue that non-imaging evidence can be further divided into two sub-categories: neuro- tool evidence, such as neuropsychology and neuropsychiatric tests; and non-neuro-tool evidence, such as a historical record of a head injury resulting from a motor vehicle accident, or a psychological report showing brain damage as a consequence of long-term alcohol abuse. This further categorisation, while may overlap at some points, arose from the fact that courts may find neuro-tool evidence (e.g. neuropsychological evidence) to be more a type of neuroscientific evidence than non-neuro-tool evidence (e.g. historical report); or some courts may not find non- neuro-tool evidence as neuroscientific evidence but rather a form of, for example, psychological evidence.
In this study, I will use the broad definition of neuroscience that includes all forms of evidence discussed above. I will return to this point in Chapter 4 (an introduction to empirical research) to explain why the broad definition is applied.
Having discussed the background to the field of neuroscience, brain imaging tools and having
116 CH de Kogel & EJMC Westgeest, ‘Neuroscientific and behavioral genetic information in criminal cases in the Netherlands’, in Journal of Law and the Biosciences, vol. 2, 2015, 580–605. 117 SS Ishikawa & A Raine, ‘Prefrontal deficits and antisocial behavior: A causal model.’ in Causes of Conduct Disorder and Juvenile Delinquency, BB Lahey, TE Moffitt & A Caspi (eds), Guilford Press, 2003. 118 de Kogel and Westgeest, 580–605. 119 Choi, 278–285 (p. 282). 34
defined what is meant by neuroscientific evidence, I will now discuss the history of the connection between neuroscience and law, and how neuroscientific evidence found its way into the criminal courts.
2. Neuroscience and law: Neurolaw
Neurolaw is a term referring to the intersection between neuroscience and law and was first used by Taylor et al. in 1991.120 More specifically, the law is more focused on cognitive neuroscience, a sub-discipline of neuroscience. That is ‘a coherent and systematic brain-based approach to mental function’.121
Although the implication of neuroscientific studies in legal areas is recent, the use of neuroscience within the law is progressing rapidly, and research interest is increasing. According to the MacArthur Foundation Research Network on Law and Neuroscience (2018), there have been more than 1,700 publications about neurolaw between 1984 and 2017.122 The rate of growth in neurolaw scholarly publications is impressive. For instance, there were only about 100 articles published in 2005-2006. This number increased to 132 publications in 2007-2008 and to more than 250 publications in 2009-2010.123
The broad breadth of the literature on the intersection of law and neuroscience is shown in a study that conducted a systemic review of the 157 neurolaw articles published in 2016.124 They presented the results under seven thematic headings: criminal law (n=54) — where articles were associated with questions including juveniles’ brains development and criminal responsibility, criminal responsibly and sentencing of adults, procedural and correctional implications including competency to plead, voluntariness of confessions and waivers of the right to remain silent and forms of rehabilitative therapies; health law (n=27) — where the main focus of articles were on the health issues of addiction and ‘the regulation of neurotechnologies being used for therapeutic and enhancement purposes’; legal concepts and decision-making (n=23) — which included articles discussing conceptual and theoretical issues including legal responsibility and the insanity
120J Taylor, J Harp & T Elliott, ‘Neuropsychologists and neurolawyers’, in Neuropsychology, vol. 5, 1991, 293–305. 121 Cowan, Harter and Kandel, 343–391 (p. 344). 122 MacArthur Foundation Research Network on Law and Neuroscience, ‘Cumulative Total of Law and Neuroscience Publications: 1984–2017’,
defence; evidence law (n=23) — which included discussions about the limitations of neuroscience and the admissibly of such evidence and the systematic review of neuroscientific evidence use in courts; Human Rights and public law (n=17) —which included articles on disability rights, freedom of thought and mental privacy; Civil law (n=15) — including articles on alternative dispute resolution, tort, commercial and employment legal problems; and general discussions (n=4).125
Different countries have investigated the theoretical application of neuroscience in their respective legal systems.126 Scholars have also realised that neuroscience has been used in courts and that there are many judgements that can provide the basis for evidence-based analysis of how neuroscience is used in such proceedings. Consequently, some studies have examined the use of neuroscience in practice.127 The emergence of neuroscience in the courtroom has led to the establishment of neurolaw databases. In Australia, for instance, the Australian Neurolaw Database collects and makes publicly available, summaries of Civil and Criminal cases where neuroscientific evidence has been used. This database aims to enhance knowledge of the use of neuroscience in courts for the public, legal scholars, judges and lawyers.128
Analysis of the implications of using neuroscience in law has not been limited to scholarly studies. Teaching in universities has also been encouraged, and there appears to be significant investment in the future of law and neuroscience. The Royal Society in London has recommended that ‘[u]niversity law degrees should incorporate an introduction to the basic principles of how science is conducted and to key areas of science such as neuroscience and behavioural genetics’.129 Some law schools have already provided courses on law and neuroscience in the
125 Goodenough and Tucker find that the areas which neurolaw scholars are more interested in are: ‘(a) techniques for the objective investigation of subjective states such as pain, memory, and truth-telling; (b) evidentiary issues for admitting neuroscience facts and approaches into a court proceeding; (c) free will, responsibility, moral judgment, and punishment; (d) juvenile offenders; (e) addiction; (f) mental health; (g) bias; (h) emotion; and (i) the neuroeconomics of decision making and cooperation’. OR Goodenough and M Tucker, ‘Law and Cognitive Neuroscience’ (2010) 6 Annual Review of Law and Social Sciences 61, 61 126 Jones and Shen, pp. 349–380; R Schweizer & S Bischof, ‘Switzerland: Brain Research and the Law’, in International Neurolaw, 2012, 269–287; H Prata & M de Freitas, ‘Brainzil Imaging: Challenges for the Largest Latin American Country’, in International Neurolaw, 2012, 67–88; K Kai, ‘Neurolaw in Japan’, in International Neurolaw, 2012; L Houston & A Vierboom, ‘Neuroscience and law: Australia’, in International Neurolaw, 2012, 11–42; M Henaghan & K Rouch, ‘Neuroscience and the Law in New Zealand’, in International Neurolaw, 2012, 257–267; B Arda & A Acıduman, ‘Neuroethics and Neurolaw in Turkey’, in International Neurolaw, 2012, 289–303; G-M Gkotsi, V Moulin & J Gasser, ‘Neuroscience in the Courtroom: From responsibility to dangerousness, ethical issues raised by the new French law’, in L’Encephale, vol. 41, 2014, 385–393. 127 See, Farahany, 485–509. 128 See, https://neurolaw.edu.au 129 The Royal Society, Brain waves 4: neuroscience and the law, in The Royal Society, London, UK, 2011, p. VI, cited in NA Vincent, ‘Neurolaw and direct brain interventions’, in Criminal Law and Philosophy, vol. 8, 2014, 43–50. 36
US130 and Australia.131
Several institutions and projects that investigate the intersection between law and neuroscience have attracted scholarly attention. The MacArthur Foundation's $10 million USD investment in commencing the Law and Neuroscience Project has increased opportunities in the neurolaw field.132 Remarkably, it is not only law scholars who study neurolaw, some scientists have also focused their studies on this field. David Eagleman, an American neuroscientist, is the founder and director of the Centre for Science and Law in the US. He explains that their ‘initiative in neuroscience and law brings together neuroscientists, judges, lawyers, policymakers, ethicists, computer scientists, and statisticians. [They] aim to determine how modern neuroscience will navigate the way we make social policy’.133
Although the initial emergence of neuroscience in legal territory was in Civil cases — personal injury claims or litigation involving individuals suffering a traumatic brain injury as a result of an accident134 — it did not take long for scholars to call for the use of neuroscience in criminal law. Many of the early discussions about the intersection of neuroscience and criminal law were inspired by criminology and the discipline’s study on biological causes of crime.
Background to neuroscience and crime
Franz Joseph Gall, in the 18th century, was the first modern scholar to recognise the brain as an organ with distinct regions underpinning different mental faculties. Gall, based on the fact that different areas of the brain were responsible for different functions — known as the localisation of function — hypothesised that by studying the size and shape of different parts of the skull, which represented the brain (as a glove fits a hand),135 he could understand mental faculties and personalities,136 including areas associated with criminality.137
130 Jones and Shen, pp. 349–380. 131 Theories of Law and Biology, University of New South Wales: http://www.handbook.unsw.edu.au/undergraduate/courses/2016/LAWS3149.html; The University of Sydney, ‘LAWS6335 - Neurolaw: Brain, Mind, Law and Ethics - Future Students - The University of Sydney’, Faculty of Law,
After Gall, in the second half of the 19th century, Cesar Lombroso, one of the most prominent figures in criminology, discussed the correlation between the brain and criminality. Lombroso, while conducting an autopsy on an infamous criminal, identified an unusual issue in his brain. From this abnormality, and partly based on Gall and Charles Darwin’s ideas, he argued that the brain was the origin of criminal behaviour and that criminals, as a result of their genes, had not reached the same level of evolution as non-criminals.138 Lombroso called these people ‘atavistic criminals’ and believed that they could be identified by the size and shape of different body parts.139 While Lombroso’s approach to detecting criminality by measuring the size and shape of body parts has been scientifically discredited, his idea about the role of the brain and genes in criminality was pursued.140 However, due to a lack of technological capability, scientists at the same time were unable to study the brains of criminals until they were dead. Alternatively, some decided to use disreputable approaches similar to those of Gall and Lombroso.141
By the 1980s, neuroimaging, as an advanced and new style of studying and understanding the criminal brain, had emerged. Scientists began to study the brains of criminals, or individuals susceptible to criminal behaviour, particularly those with a tendency towards violence. These studies were conducted through the use of machines such as EEG, CT and PET scanners. One of the early studies in this area was conducted by Nora D. Volkow and Laurence Tancredi, who by using EEG, CT and PET scans, examined four individuals with repetitive violent behaviour. They affirmed the utility of PET scans in assessing brain impairment that could result in violent behaviour.142 In 1990, Antonio Damasio and his colleagues published their results about the relationship between frontal lobe abnormalities and impulse control disorders.143 A year later, Joseph Tonkonogy reported on his 4-year study using CT and MRI scans. He concluded that his research on five patients (he conceded the limited number of subjects), indicated a correlation between violent behaviour and tissue loss in a specific area of the brain.144
138 A Raine. 139 R White, F Haines & N Asquith, Crime and criminology, 5th ed., Oxford University Press, 2012. 140 Rose and Abi-Rached. Of note, Lombroso’s claim about the role of criminals’ brain which is valued, not his methodology in detecting criminals. 141 See, White, Haines and Asquith. 142 Rose and Abi-Rached. While EEG scans showed abnormal brain activity in three subjects, CT scan indicated abnormalities in only two, and the reports from PET scans showed abnormalities in all subjects. Perhaps surprisingly, while significant abnormalities were indicated by the PET scan in two individuals, their CT scan recorded no abnormalities. See, ND Volkow & L Tancredi, ‘Neural substrates of violent behaviour a preliminary study with positron emission tomography’, in The British Journal of Psychiatry, vol. 151, 1987, 668–673. 143 SW Anderson et al., ‘Impairment of social and moral behavior related to early damage in human prefrontal cortex.’, in Nature Neuroscience, vol. 2, 1999, 1032–7; Rose and Abi-Rached. 144 JM Tonkonogy, ‘Violence and temporal lobe lesion: Head CT and MRI data’, in The Journal of neuropsychiatry and clinical neurosciences, vol. 3, 1991, 189–196. 38
As a result of these studies, the field of neurocriminology was developed.145 Neurocriminology is a field of study that uses neuroscientific tools to analyse, prevent and predict criminal behaviour.146 A key figure in this area is Adrian Raine, who refers to Lombroso as the founder of neurocriminology.147 Raine, in several studies, has discussed imaging results from brains of murders and has concluded that there were specific abnormalities in the function and structure of their brains.148 Neurocriminology, however, is not limited to the study of the brain, as it also studies how the brain interacts with social and genetic factors to contribute to criminality.
Conversely, genetic behaviourist researchers — those who use genetics-based methods to understand differences in individual behaviour — had previously searched for a direct relationship between an individual’s genes and their behaviour. However, growing knowledge about genetics and behaviour led genetic behaviourists to shift their research direction towards understanding how genes impact upon the brain, and by extension, behaviour. These studies claimed that environmental factors (including social factors) interact with genes and influence the brain’s structure and function, and consequently behaviour.149 Genetic studies from around 1990
145 It appears there is no reliable source determining who established the field of neuroimmunology. 146 A Glenn & A Raine, ‘Neurocriminology: implications for the punishment, prediction and prevention of criminal behaviour’, in Nature Reviews Neuroscience, vol. 15, 2014, 54–63. 147 A Raine. 148 A Raine et al., ‘Selective reductions in prefrontal glucose metabolism in murderers’, in Biological Psychiatry, vol. 36, 1994, 365–373; A Raine, M Buchsbaum & L Lacasse, ‘Brain abnormalities in murderers indicated by positron emission tomography’, in Biological Psychiatry, vol. 42, 1997, 495–508; Rose and Abi-Rached. 149 A Caspi & TE Moffitt, ‘Gene-environment interactions in psychiatry: joining forces with neuroscience’, in Nat Rev Neurosci, vol. 7, 2006, 583–590; MG Vaughn et al., ‘DAT1 and 5HTT Are Associated With Pathological Criminal Behavior in a Nationally Representative Sample of Youth’, in Criminal Justice and Behavior, vol. 36, 2009, 1113–1124; CJ Ferguson & KM Beaver, ‘Natural born killers: The genetic origins of extreme violence’, in Aggression and Violent Behavior, vol. 14, 2009, 286– 294; M DeLisi, ‘Genetics: L’Enfant Terrible of Criminology’, in Journal of Criminal Justice, vol. 40, 2012, 515–516; A Alimardani, AM Ardekani & LR Astany, ‘A Bio-Social Review to Mitigate the Punishment of Unwanted Acts’, in International Journal of Criminology and Sociology, vol. 3, 2014, 348–359. One of the most famous studies about the relationship between genetics and criminality is in deficiency in the Monoamine Oxidase A gene (MAOA). In 1993, Han Brunner and his team found abnormal behavior (e.g. aggressive behaviour) among people who had a mutation in the structure of this gene.149 While the public, press and many researchers were excited about this discovery, and began referring to MAOA as the ‘aggression gene’ or the ‘warrior gene’ in the middle of the 1990s, Brunner explained that his research did not support the idea of an ‘aggression gene’ by stating that ‘[T]he notion of an ‘aggression gene’ does not make sense, because it belies the fact that behavior should and does arise at the highest level of cortical organization, where individual genes are only distantly reflected in the anatomical structure, as well as in the various neurophysiological and biochemical functions of the brain ... [A]lthough a multitude of genes must be involved in shaping brain function, none of these genes by itself encodes behavior’,149 quoted in N Rose, The Politics of Life Itself: Biomedicine, Power, and Subjectivity in the twenty-first century, in Princeton University Press, 2007, p. 239. Not long after that, while some studies had begun to disprove the relation between MAOA and aggression, a more comprehensive study by Avshalom Caspi and Terrie Moffitt and their colleagues149 concluded that the factor that in some cases confounded the relationship between MAOA and aggressive behavior, was the environment. They suggested that insufficient MAOA, when coupled with childhood maltreatment, highly influenced the neurotransmitter (neural) system, and consequently behavior. Rose 39
onwards promoted a link between neurobiology (i.e. neuroscience and genetics) and criminal behaviour.150 These studies pointed towards understanding crime as a behaviour, which had a multifaceted nature; the implication being that understanding crime requires consideration of different factors that arise from a variety of sources (i.e. genes, brain and environment).
This way of understanding crime has also been adopted in legal studies, including in some of the recent empirical neurolaw studies, where the implication of neuroscience and genetics in law are examined together — neurobiology and law.151 This was a move away from previous studies that examined the role of genetics and neuroscience separately to criminal law.152 Nita Farahany, for instance, in an empirical analysis of the use of neuroscience and behavioural genetics in the US criminal courts explains that ‘[m]uch of the scholarship on neurobiological evidence in criminal law has focused exclusively on either behavioral genetics or neuroscience. This study looks at these sciences together because the claims are inextricably intertwined’.153 Then she goes on to support this claim by referring to a special issue in a scientific journal on the relationship between genes, the brain, and behaviour,
The 15 articles included in the volume represented the diversity of methodologies applied to the complexity of pathways linking genes, brain, and behavior. The introductory Chapter concluded that the breadth of studies proves that tracing the pathways between biology and behavior requires expertise in genetics,
and Abi-Rached; A Caspi et al., ‘Role of genotype in the cycle of violence in maltreated children’, in Science, vol. 297, 2002, 851–854. 150 Rose and Abi-Rached. In the US, not long after the report about MAOA and aggression in the murder trial of Stephen Mobley, his lawyer attempted to use this evidence to diminish his punishment. He was not successful. The case of Stephen Mobley in 1994 was one of the earliest uses of MAOA evidence as a factor in diminishing criminal responsibility. Since then, lawyers in some countries have raised the possible role of genetic predisposition in criminal behaviour. See, Mobley v. State, 455 S.E.2d 61 (Ga. 1995). N Eastman & C Campbell, ‘Neuroscience and legal determination of criminal responsibility’, in Nature reviews neuroscience, vol. 7, 2006, 311–318. 151 See, for example, M Farisco & C Petrini, ‘The impact of neuroscience and genetics on the law: A recent Italian case’, in Neuroethics, vol. 5, 2012, 317–319. 152 For example see, C Gavaghan & A Bastani, ‘Genes, Blame and Loss of Control: is there a Place in Criminal Law for a “Genetic Defense”?’, in Recent Advances in DNA & Gene Sequences (Formerly Recent Patents on DNA & Gene Sequences), vol. 8, 2014, 119–125; A McCay, ‘Behavioural Genetics, Moral Agency and Retributive Sentencing: The Case for Mitigation (Doctoral dissertation)’, Sydney University, 2013; SJ Morse, ‘Gene-Environment interactions, criminal responsibility, and sentencing’, in Gene-environment interactions in developmental psychopathology, 2011, 207–234; Cowan, Harter and Kandel, 343–391; KM Beaver, ‘Genetic influences on being processed through the criminal justice system: Results from a sample of adoptees’, in Biological Psychiatry, vol. 69, 2011, 282–287,
neuroscience, psychology, and psychiatry.154
The understanding of behaviour as having a multifaced nature is also evident in some of the neurolaw claims (I will return to this point). In NSW criminal courts, while discussions about the use of genetic evidence seem to be rare,155 the interaction between the brain, environmental factors, psychology and psychiatry is common.
Given this introduction to the developing area of neurolaw, and the intersection between neuroscience and criminal law, I will consider scholarly discussions of neurolaw claims considered in this study.
3. Background to neurolaw claims
Neuroscientific studies and their potential impact on criminal law is the subject of controversial debate. Some of these arguments are extensive and have complicated philosophical bases that do not fit within the scope of this research. As such, in this section, I will focus on discussions that provide a background to the extent that they are helpful for examining these claims in practice.
In this section, first, I will discuss scholarly arguments about how neuroscience may influence criminal law by suggesting that free will is an illusion and individuals are not criminally responsible for their conduct. In this part, I will review Greene and Cohen’s predictive claim (see Chapter 9). Then, I will refer to those academics who advocate for the use of neuroscience within the conventional and established rules of criminal law. Here, I will also discuss the double-edged sword claim (see Chapter 5) and Morse’s argument about why neuroscientific evidence should be translated into the legal language (see Chapter 6). Following that, I will discuss different viewpoints on using neuroscience to promote rehabilitation of offenders (see Chapter 8). Next, I discuss the perspective of those scholars who criticise the reliability of neuroscientific evidence and its use in courts, and those who accept these limitations, however find converging forms of evidence a proper approach to address this issue (see Chapter 7). Then, I will outline different perspective on whether neuroscience has a persuasive power and undue influence on the legal decisions (see Chapter 7). Finally, I will present discussions about the different areas of expertise held by those who provide expert reports and testimony on neuroscientific evidence (see Chapter 7).
154 Farahany, 485–509 (p. 487). 155 See, McCay. Unlike Australia, it appears that in other countries, genetic evidence for the purpose of explaining criminal behaviour has been introduced to the court. See, for example, ML Baum, ‘The monoamine oxidase A (MAOA) genetic predisposition to impulsive violence: Is it relevant to criminal trials?’, in Neuroethics, vol. 6, 2013, 287–306. 41
Neuroscience, free will and criminal responsibility
A much debated question in the neurolaw literature is whether studies on the impact of biological factors, in particular the brain, on behaviour, will revolutionise criminal law and its conception of criminal responsibility.156 Proponents of this claim refer to studies suggesting that the brain functions in a mechanical way and free will in an illusion. For instance, in 1983, Libet and his colleagues published their controversial experiment, which purported to show that the brain plan actions (e.g. raising a hand) before an individual become consciously aware of that decision.157 Such studies suggest that what we know of as the ‘self’ — the decision maker — may be an illusion created by the brain.158
Further, some scholars refer to studies that suggest genetic makeup and environment shapes who we are. Eagleman explains that,
When it comes to nature and nurture, the important point is that you choose neither one. We are each constructed from a genetic blueprint and born into a world of circumstances about which we have no choice in our most formative years. The complex interactions of genes and environment means that the citizens of our society possess different perspectives, dissimilar personalities, and varied capacities for decision making. These are not free-will choices of the citizens; these are the hands of cards we're dealt.159
Eagleman goes on to complete the picture of such a mechanistic nature of human behaviour by explaining that ‘in our current understanding of science, we cannot find the physical gap in which to slip free will—the uncaused causer—because there seems to be no part of the machinery [brain] that does not follow in a causal relationship from the other parts.’160 He claims that the implication of these discoveries and experiments in the context of the criminal law means that ‘[b]ecause we did not choose the factors that affected the formation and structure of our brain, the concepts of free will and personal responsibility begin to sprout with question marks’.161
156 See, FX Shen, ‘The overlooked history of neurolaw’, in Fordham Law Review, vol. 85, 2016, 667– 695. 157 B Libet et al., ‘Time of conscious intention to act in relation to onset of cerebral activity (readiness- potential) the unconscious initiation of a freely voluntary act’, in Brain, vol. 106, 1983, 623–642. 158 T Metzinger, The ego tunnel: The science of the mind and the myth of the self, Basic Books (AZ), 2009. 159 D Eagleman, Incognito: The Secret Lives of the Brain, New York City, U.S., Pantheon, 2011, p. 159 and 160. 160 D Eagleman, Incognito: The Secret Lives of the Brain, New York City, U.S., Pantheon, 2011, p. 166. 161 Eagleman, Incognito: The Secret Lives of the Brain, p. 160. 42
Stephen Morse, on the other hand, claims that neuroscience will not change anything for the law. He argues that the law’s criteria are folk psychological. In simple terms, it means that the law explains human behaviour through the use of mental state variables.162 The law does not say that if an individual has a brain impairment, they are less responsible for their criminal behaviour. Instead, the law asks whether the offender could ‘understand’ their actions or ‘control’ their behaviour, meaning whether individuals were rational at the time of the crime.163 Even if biological sciences were to prove that the brain is a machine and that individuals do not possess free will, this would not change legal principles. The law is not concerned with an individual’s possession of free will, it asks whether they are rational at the time of the offence.164 Morse adds that it is ‘the fundamental psychological error’ if it is assumed causation is per se an excusing condition.165 If that assumption were true, as all behaviour is caused by the brain, then all criminal acts would result in a legal excuse from criminal responsibility. Hence, unless a brain abnormality diminishes an individual’s rationality, impairments of the brain are not relevant to criminal responsibility.166
Morse criticises many of the claims made about the relationship between neuroscience and criminal responsibility, and terms this phenomenon the ‘Brain Overclaim Syndrome’,
New, powerful scientific findings about the correlates and causes of behavior often have a potent and, alas, rationality-unhinging effect on the thinking of Potential Commentators [PCs]. Most flow from misunderstanding the relation between brains and responsibility … more legal implications are claimed for the brain science than can be justified.167
Joshua Greene and Jonathan Cohen, in support of scholars such as Eagleman, hold the opinion that all our actions are determined by the interaction between genes and the environment, and that those who commit crimes are merely victims of unfortunate neuronal circumstances in a mechanical brain.168 They further claim that our sense of free will is an illusion and that we should not punish offenders for acts they cannot actually control.169 While they also accept Morse’s
162 S Morse, ‘Neuroscience in Forensic Contexts: Ethical Concerns’ in Ethics Challenges in Forensic Psychiatry and Psychology Practice, EEH Griffith (ed), New York, Columbia University Press, 2018, pp. 132–158 (p. 134). 163 Morse, pp. 132–158 (p. 134). 164 S Morse, ‘Avoiding Irrational NeuroLaw Exuberance: A Plea for Neuromodesty’, in Law, Innovation and Technology, vol. 3, 2011, 209–228. 165 Morse, pp. 157–198 (p. 180) quoted in J Greene & J Cohen, ‘For the law, neuroscience changes nothing and everything’, in Philosophical Transactions of the Royal Society B: Biological Sciences, vol. 359, 2004, 1775–1785. 166 Morse, pp. 157–198. 167 Morse, 397–412 (p. 403). 168 Greene and Cohen, 1775–1785. 169 Greene and Cohen, 1775–1785. 43
objection that the law is not concerned with free will, they claim that in the future, when more studies support a mechanical understanding of the brain, people’s intuition about criminal responsibility will change. Because the legitimacy of the law depends on society’s intuitions and commitment, courts will not punish offenders because they are criminally responsible for their crime, but rather punishment will be applied to the extent that it is beneficial for society. For instance, if the courts find that punishment would deter offenders from future offending or deter others in society from committing similar crimes, then it is justifiable to punish the offender.170
There are many studies arguing the question of whether neuroscience will or will not change the concept of criminal responsibility.171 Some argue that disbelief in free will collides with long- standing and deeply-rooted perceptions of individual responsibility and self-control, and that advances in scientific knowledge cannot trigger fundamental changes in the way we conceive of such concepts.172 As Georg Lichtenberg puts it, ‘Man is a masterpiece of creation, if only because no amount of determinism can prevent him from believing that he acts as a free being.’173 Further, some academics argue that the available evidence is not yet convincing enough to change these commitments,174 and several scientists have pointed out limitations in the empirical studies claiming human actions are deterministic.175
There are also some academics that do not claim neuroscience will change criminal law, but rather can assist criminal procedure by providing more information to a fact-finder operating within established principles of the criminal justice system. They suggest that neuroscience can assist in
170 Greene and Cohen, 1775–1785. 171 A Moir & D Jessel, A Mind to Crime: the controversial link between the mind and criminal behaviour, in Penguin, London, Michael Joseph, 1995; R Wright, The moral animal: Evolutionary psychology and everyday life, Vintage, 1995; Morse, 209–228; C Willmott, Biological determinism, free will and moral responsibility: insights from genetics and neuroscience, Springer, 2016; MS Gazzaniga & MS Steven, ‘Free will in the twenty-first century: a discussion of neuroscience and the law’ in Neuroscience and the Law, 2004, pp. 51–70; AC Pustilnik, ‘Violence on the brain: a critique of neuroscience in criminal law’, in Wake Forest L. Rev., vol. 44, 2009, 183–237. As previously discussed, many of these arguments have philosophical basis. For more information on these arguments, see D Hodgson, Rationality+ consciousness= free will, Oxford University Press, 2012; H Walter, ‘Contributions of neuroscience to the free will debate: from random movement to intelligible action’ in The Oxford Handbook of Free Will, R Kane (ed), 2nd ed., USA, Oxford University Press, 2011; A Mele, ‘Free will and neuroscience’, in Philosophic Exchange, vol. 43, 2013; M McKenna & D Pereboom, Free will: A contemporary introduction, Routledge, 2016; C Gavaghan, ‘No gene for fate?’ in Bioethics at the Movies, S Shapshay (ed), Johns Hopkins University Press, 2009, p. 75. 172 C O’Connor & H Joffe, ‘How has neuroscience affected lay understandings of personhood? A review of the evidence’, in Public understanding of science, vol. 22, 2013, 254–268. Also see, A Roskies, ‘Neuroscientific challenges to free will and responsibility’, in Trends in cognitive sciences, vol. 10, 2006, 419–423. 173 Quoted in Eagleman, Incognito: The Secret Lives of the Brain. page 160 174 O’Connor and Joffe, 254–268. 175 See A Lavazza & M De Caro, ‘Not so fast. On some bold neuroscientific claims concerning human agency’, in Neuroethics, vol. 3, 2010, 23–41; WR Klemm, ‘Free will debates: Simple experiments are not so simple’, in Advances in Cognitive Psychology, vol. 6, 2010, 47. 44
explaining criminal behaviour.
Neuroscience and explaining crime
Some scholars assume that individuals are criminally responsible for their conduct, unless there are factors that slightly or significantly reduce that individual’s capacity. They consider that some brain abnormalities may interrupt a person’s decision making, ability to understanding their own behaviour, or control over their acts. For instance, an individual may commit a crime because of an impaired prefrontal cortex, which is normally associated with controlling intentional and emotional behaviour.176 Evidence from brain imaging , which shows a structural or functional abnormality, may suggest that an offender had a lower capacity to control or understand their behaviour, therefore warranting a mitigated sentence.177
Interestingly, such brain imaging evidence may have unfavourable results for the defendant. If a brain image provides an explanation for a person’s violent behaviour at the time of the offence, it may also suggest that the offender poses a continuous risk of violence to the community upon release. Several studies have discussed whether proof of a brain impairment may result in either a more lenient, or harsher punishment.178 This claim is called the double-edged sword effect of neuroscience. Owen D. Jones and Francis X. Shen summarise the claim by stating that, ‘a brain too broken may be simply too dangerous to have at large, even if it is somehow less culpable.’179
This type of claim is not new, and some scholars have discussed the same issue where genetic evidence has been argued to mitigate culpability in violent behaviour.180 Hence, some refer to the claim as the ‘double-edged sword of biological explanation of criminal behaviour’.181
On the use of neuroscience in assessing criminal responsibility, Morse argues that ‘law’s criteria are not biological’182 and therefore any neuroscientific evidence submitted to the court should be
176 Ishikawa and Raine; Yang and Raine, 81–88. 177 See, L Klaming & B Koops, ‘Neuroscientific Evidence and Criminal Responsibility in the Netherlands’ in International Neurolaw: A Comparative Analysis, TM Spranger (ed), 2012, pp. 227–256; T Vidalis & G-M Gkotsi, ‘Neurolaw in Greece: an overview’ in International Neurolaw, TM Spranger (ed), Springer, 2012, pp. 179–196; Alimardani and Chin. 178 See, NA Farahany & JE Coleman Jr, ‘Genetics, Neuroscience, and Criminal Responsibility’ in The Impact of Behavioral Sciences on Criminal Law, NA Farahany (ed), New York, Oxford University Press, 2009, pp. 183–240; AS Barth, ‘A double-edged sword: The role of neuroimaging in federal capital sentencing’, in American journal of law & medicine, vol. 33, 2007, 501–522; OC Snead, ‘Neuroimaging and the Complexity of Capital Punishment’, in NYUL Rev., vol. 82, 2007, 1265–1339. 179 Jones and Shen, pp. 349–380 (p. 362). 180 DW Denno, ‘Courts’ increasing consideration of behavioral genetics evidence in criminal cases: Results of a longitudinal study’, in Mich. St. L. Rev., 2011, 967–1047. 181 Chandler, 550–579. 182 Morse, pp. 132–158 (p. 144). 45
translated into legal language — i.e. the law’s folk psychology.183 For instance, evidence of an impaired brain alone does not mean anything to the law, because law is concerned with the mental states of individuals, such as the capacity to control behaviour, or understanding of the wrongfulness of an act. As such, evidence of severe brain damage is not legally relevant unless it is explained how such damage would influence an individual’s mental capacity (e.g. understanding that certain behaviour was wrong).184 Morse warns that without this translation, while courts may find evidence of an impaired brain legally relevant in diminishing punishment, the brain impairment may not interfere with mental states in a way that is considered relevant in determining criminal responsibility, e.g., capacity to conduct intentional behaviour.185
Neuroscience and rehabilitation
Predicting the possibility of future offending and of rehabilitation through the use of neuro- interventions, have been discussed by several scholars.186 In general, the views of scholars supporting the use of neuroscience to promote rehabilitation can be divided into two broad categories: those who question human free will and criminal responsibly and advocate for a legal system with a focus on rehabilitation. Scholars holding views from the second group support the use of neuroscientific rehabilitative approaches within the traditional principles of criminal law.
Scholars that hold views from the first group, who have been inspired by our understanding of the neurobiological causes of crime, support a medical model for responding to crime. Anne Moir and David Jessel argue that,
So if great swathes of crime are a biological disorder, how should it be treated if not
183 S Morse, ‘Lost in Translation?: An Essay on Law and Neuroscience’ in Law and Neuroscience, Current Legal Issues, M Freeman (ed), Oxford University Press, 2011, pp. 529–562. 184 Morse, pp. 132–158 (p. 134). 185 S Morse, ‘Lost in Translation?: An Essay on Law and Neuroscience’ in Law and Neuroscience, Current Legal Issues, M Freeman (ed), 2011, XIII, 529. 186 C Gavaghan, ‘Neuroscience, deviant appetites, and the criminal law’ in Neuroscience and Legal Responsibility, NA Vincent (ed), Oxford University Press, 2013, p. 205; DM Eagleman, ‘Neuroscience and the law’, in Houston Lawyer, vol. 16, 2008, 36–40; HT Greely, ‘Neuroscience and criminal justice: not responsibility but treatment’, in University of Kansas Law Review, vol. 56, 2008, 1103–1138; HT Greely, ‘Direct brain interventions to “treat” disfavored human behaviors: ethical and social issues’, in Clinical Pharmacology & Therapeutics, vol. 91, 2012, 163–165; AR Baskin-Sommers & K Fonteneau, ‘Correctional change through neuroscience’, in Fordham L. Rev., vol. 85, 2016, 423–436; AL Glenn & KE McCauley, ‘How biosocial research can improve interventions for antisocial behavior’, in Journal of Contemporary Criminal Justice, vol. 35, 2019, 103–119; P Koi, S Uusitalo & J Tuominen, ‘Self-Control in Responsibility Enhancement and Criminal Rehabilitation’, in Criminal Law and Philosophy, vol. 12, 2018, 227–244; RA Poldrack et al., ‘Predicting Violent Behavior: What Can Neuroscience Add?’, in Trends in Cognitive Sciences, vol. 22, 2018, 111–123; KA Kiehl et al., ‘Age of gray matters: Neuroprediction of recidivism’, in NeuroImage: Clinical, vol. 19, 2018, 813–823; W Hall & A Carter, ‘Is deep brain stimulation for addiction an acceptable crime control measure?’, in Addiction, vol. 107, 2012, 1360–1360. 46
by killing or incarceration? Perhaps we can make things easier for ourselves by not seeking to acquit or pardon such individuals, but by confining them for the protection of society. With the introduction of PET scans we should be able more easily to identify those with, for instance, dysfunctional control mechanisms. Identifying the cause, however, loads us with the responsibility of doing something about it - treating the offender. It is easier simply to condemn and lock away - although of course more expensive when the offender re-emerges from prison and inevitably offends again.187
Similar to Greene and Cohen who argue that a justice system should not be based on guilt and punishment, Moir and Jessel do not find incarceration justifiable, and they view it as ‘an expensive and wasteful reaction [to crime], which does very little good apart from providing a secure hotel facility for a limited amount of time.’188 Instead, they propose a medical model of criminal justice that is based on prevention, diagnosis and treatment.189
Scholars holding the second view, however, advocate for treatment, prediction and prevention, while staying loyal to established principles of criminal responsibility. David Hodgson190 specifically argues against a justice system that does not recognise criminal responsibility and focuses solely on rehabilitation.191 Nonetheless, he encourages treatment and rehabilitation of offenders within conventional rules of criminal law that is based on guilt and punishment:
It is to be expected that increasing knowledge of how the brain works will assist in devising programs for rehabilitation of offenders and programs for addressing genetic and environmental factors contributing to criminal conduct, and will contribute to the development and implementation of educational strategies to discourage criminal conduct.192
Hodgson emphasises that we should consider principles of Human Rights (e.g. consensual treatment rather than coercive) when using advanced technology to diagnose and treat abnormalities that may contribute to criminal behaviour.193
187 Moir and Jessel, pp. 296–298, quoted in D Hodgson, ‘Guilty mind or guilty brain? Criminal responsibility in the age of neuroscience’, in Australian Law Journal, vol. 74, 2000, 661–680. 188 Moir and Jessel, p. 303. 189 Also see, Greely, 1103–1138. 190 Hodgson is a significant figure in this thesis as he worked as a judge in the highest court in NSW (Australia) and has discussed the use of neuroscience in a legal context in his academic publications. For example see, Hodgson. For more information about David Hodgson see, The Sydney Morning Herald, ‘Judge blessed with flawless logic’, 2012,
There are some scholars who suggest more controversial approaches to the prediction, prevention and treatment of brain function abnormalities in offenders.194 Raine claims that some forms of criminal behaviour are similar to the behaviours exhibited in clinical disorders, and that just as psychiatric disorders are diagnosable, criminality can also be diagnosed, predicted, and treated.195 He argues for the implementation of a program, LOMBROSO—Legal Offensive on Murder: Brain Research Operation for the Screening of Offenders, and explains,
Under LOMBROSO, all males in society aged eighteen and over have to register at their local hospital for a quick brain scan and DNA testing. One simple finger prick for one drop of blood that takes ten seconds. Then a five-minute brain scan for [different measurements including assessing] expression of 23,000 different genes at the cellular level. The computerization of all medical, school, psychological, census, and neighborhood data makes it easy to combine these traditional variables alongside the vast amount of DNA and brain data to form an all-encompassing bio-social data set.196
If the program finds individuals that are at risk of offending, it proposes that they be kept in indefinite detention. The system attempts to reduce the risk of criminal wrongdoings, and uses periodical assessments to determine if an individual is ready to be released back into society.197
Some scholars have focused on less controversial approaches to predicting whether individuals will most likely become offenders, non-offenders, or re-offenders.198 For instance, a group of scientists (including Eagleman) have designed an assessment program that uses a simple, game- like, interactive tablet software based on neuropsychological tools to measure a range of cognitive traits including empathy, reduced impulse control, aggression, executive function and risk-taking, and correlates this with the risk of future offending. They claim that their approach, compared to
194 For a review see, C Spivakovsky, K Seear & A Carter, Critical Perspectives on Coercive Interventions: Law, Medicine and Society, Routledge, 2018. C Spivakovsky, K Seear & A Carter, ‘Coercive interventions in law and medicine: Setting the scene’ in Critical Perspectives on Coercive Interventions, Routledge, 2018, pp. 13–22. 195 See Adrian Raine, The psychopathology of crime: Criminal behavior as a clinical disorder., in Academic Press, Elsevier Science, 1993. There have also been some studies that connect brain abnormalities to personality disorders. See PF Goyer et al., ‘Positron-emission tomography and personality disorders’, in Neuropsychopharmacology, vol. 10, 1994, 21–28. Also see, A Raine, ‘The biological basis of crime’, in Crime: Public policies for crime control, vol. 43, 2002, 74. 196 A Raine, p. 342. 197 A Raine. 198 For example, see, MG Berman et al., ‘Dimensionality of brain networks linked to life-long individual differences in self-control’, in Nature Communications, vol. 4, 2013, 1373–1379. 48
the current interview-based risk assessment, is faster, more precise, and generates consistent measurements across individuals.199
Eagleman has also generated novel ways to rehabilitate offenders. For instance, he is developing a brain-imaging based approach to rehabilitate offenders. In one of the experiments, he conducted fMRI scans on a participant with a crack cocaine addiction. They were shown images of crack cocaine and were asked to entertain their cravings. Areas of the brain that were activated — craving network — were recorded. The participant was then asked to suppress their cravings200 and the areas that were then activated were recorded — suppression network. Subsequently, the participant was shown pictures of more crack cocaine, and was provided with real-time visual feedback of their brain activity — i.e. craving and suppression networks — to practice suppressing the craving, and to reinforce thoughts and approaches which helped them to suppress their craving.201
Generally, there are two categories of approaches to rehabilitating offenders with disordered brain function.202 Sometimes, the rehabilitative program is indirect and involves prescribing psychotherapy (e.g. cognitive behavioural treatment).203 A principle underlying these measures is neuroplasticity—that the brain’s structure can change as a result of experiences, memorisation, and learning.204 Adverse experiences and brain injuries may negatively shape the brain’s structure, which, by extension, may affect behaviour. Neurobiological elements may assist in determining who would benefit from psychotherapy to address these changes or for example, how to control behavioural issues.205 This is the rationale behind Eagleman’s technique for
199 PA Ormachea et al., ‘The role of tablet-based psychological tasks in risk assessment’, in Criminal Justice and Behavior, vol. 44, 2017, 993–1008; PA Ormachea et al., ‘Enabling Individualized Criminal Sentencing While Reducing Subjectivity: A Tablet-Based Assessment of Recidivism Risk’, in AMA journal of ethics, vol. 18, 2016, 243–251; Center for Science & Law, ‘Assessing offender risk’, in Center for Science & Law,
rehabilitating offenders with substance abuse issues – to make suppression networks stronger than craving networks.
Direct rehabilitative approaches include deep brain stimulation—surgically implanted electrodes in the brain that provide electrical stimulation to change the brain’s function206 — psychosurgery,207 radioisotope therapy, transcranial magnetic stimulation and electroconvulsive therapy.208 These approaches, apart from exceptional cases where they were used to treat offenders,209 are primarily used in clinical contexts and were developed to treat diseases such as Alzheimer’s and Parkinson’s disease. Nonetheless, scholars have expressed interest in these techniques and have been discussing the possibility of using direct interventions to treat brain- related causes of criminal behaviour.210
Direct neuro-interventions are controversial approaches to treating offenders, not only because they may have grave side effects or involve unsafe techniques,211 but they also involve many ethical dilemmas.212 Psychiatric drugs by balancing neurochemicals in the brain are another direct approach to treating psychiatry-related antisocial behaviours such as ADHD, depression and schizophrenia. This is a less controversial method than other discussed approaches.213
Overall, there are different viewpoints on the promotion of rehabilitation through neuroscience. Some scholars suggest a more controversial approach to rehabilitation and risk assessment, while others are cautious about the principles of Human Rights. Despite their diverse views, they have a similar motivation for adopting neuro-interventions. Elizabeth Shaw summarises this motive as ‘a pressing need to develop more effective ways of re-integrating offenders back into society’214
the literature’, in International journal of offender therapy and comparative criminology, vol. 58, 2014, 1279–1296, cited in de Kogel and Westgeest, 580–605. 206 See, Greely, 1103–1138; AM Lozano et al., ‘Subcallosal cingulate gyrus deep brain stimulation for treatment-resistant depression’, in Biological psychiatry, vol. 64, 2008, 461–467; M Rizzi & CE Marras, ‘Deep Brain Stimulation for the Treatment of Aggressive Behaviour: Considerations on Pathophysiology and Target Choice’, in Stereotactic and Functional Neurosurgery, vol. 95, 2017, 114–116. 207 Greely, 1103–1138. 208 Vincent, 43–50. 209 See, A Franzini et al., ‘Stimulation of the posterior hypothalamus for medically intractable impulsive and violent behavior’, in Stereotactic and functional neurosurgery, vol. 83, 2005, 63–66. 210 Vincent, 43–50. 211 Egas Moniz won the Nobel Prize for inventing frontal leucotomy in 1949. In less than three decades, however, this procedure was reviled. See Greely, 1103–1138. 212 See, L Bomann-Larsen, ‘Voluntary rehabilitation? On neurotechnological behavioural treatment, valid consent and (in) appropriate offers’, in Neuroethics, vol. 6, 2013, 65–77; L Forsberg & T Douglas, ‘Anti- Libidinal Interventions in Sex Offenders: Medical or Correctional?’, in Medical Law Review, vol. 24, 2016, 453–473; E Shaw, ‘Direct Brain Interventions and Responsibility Enhancement’, in Criminal Law and Philosophy, vol. 8, 2014, 1–20. 213 See, Alimardani, Ardekani and Astany, 348–359. 214 Shaw, E. (2014). Direct Brain Interventions and Responsibility Enhancement. Criminal Law and Philosophy 8(1): 1–20. 50
due to the ineffectiveness of the criminal justice system in rehabilitating criminals and preventing recidivism.215
Limitations of neuroscience and convergence of evidence
Despite academic interest and excitement about the growing neurolaw field, there are many who criticise the limitations of brain images (see above), particularly functional imaging techniques (i.e. PET, SPECT and fMRI), and the utility of such evidence for courts attempting to interpret the mental state of individuals. Jane Campbell Moriarty, claims that,
while courts have routinely admitted some neuroimages, such as CT scans and MRI, as proof of trauma and disease, they have been more circumspect about admitting the PET and SPECT scans and fMRI evidence. With the latter technologies, courts have often expressed reservations about what can be inferred from the images. Moreover, courts seem unwilling to find neuroimaging sufficient to prove either insanity or incompetency, but are relatively lenient about admitting neuroimages in death penalty hearings.216
She goes on to explain that functional imaging techniques ‘have remarkable potential for forensic use, but they are too new, too uncertain, and too laden with troubling questions to earn easy admission to the courts’.217
Hank Greely, on the other hand, argues that in the US, ‘[a]nybody can try to admit neuroscience evidence in any case, in any court’, because judges that are less familiar with this type of evidence would not be prepared to make appropriate decisions anyway.218 In Australia, Leanne Houston and Amy Vierboom also claim that courts in the criminal justice system are reluctant in accepting neuroscientific evidence such as fMRI and EEG, due to concerns about the reliability of brain imaging . Nonetheless, they predict that in the future, when brain imaging techniques improve, courts may change their position on these imaging tools.219
215 In NSW, it was found that 41 per cent of adult prisoners, and about 66 per cent of juveniles who were released in 2015, re-offended within 12 months. The rate of recidivism could be higher if a longer period is considered. Bureau of Crime Statistics and Research (BOCSAR), ‘Re-offending statistics for NSW’, in BOCSAR, 2017,
Some scholars, despite admitting the limitations of neuroscientific evidence, nonetheless support its use in courts. Eyal Aharoni et al. argue that,
Chemists perhaps cannot determine if a cake was baked with love, but they can determine if it was baked with cyanide, which in turn provides circumstantial evidence against the love hypothesis. Likewise, although neuroscience cannot locate responsibility in the brain, perhaps it can identify maladies that provide at least circumstantial evidence against guilt or liability.220
Frederick Schauer similarly notes that the current legal stance is that not every piece of admissible evidence is required to independently have the capacity to determine the final legal position. Decisions are made in courts as a result of different pieces of evidence and factual matters.221 Schauer’s point aligns with a wave of studies that support the convergence of evidence claim: that neuroscientific evidence should be admitted and relied upon only where it is supported by other forms of evidence, used as a kind of informal corroboration. Octavio S. Choi elaborates on this view:
Although neuroscience’s proper role in the courts is limited … I also believe that neuroscience evidence can be very useful … it may be helpful as one component of an analysis that integrates psychological and behavioral perspectives … they can compensate for each other’s relative weaknesses, while combining their strengths.222
Choi unlike many scholars that merely point out the limitations of brain imaging techniques explains that behavioural and psychological assessments also have their own shortcomings, and that by converging these forms of evidence, we can reach a more convincing outcome.223
The use of this form of corroboration to understand criminal behaviour and the connection between different sources of behaviour, was also evident in Farahany’s study referred to previously. This study considers how genetics, neuroscience, psychology and psychiatry could be integrated into understanding human behaviour.224 Raine similarly discusses violent crime as being like a ‘biosocial jigsaw puzzle’, and that understanding it requires combining social sciences and biology as complementary pieces of information.225
220 Aharoni et al, 157 221 N Aggarwal & E Ford, ‘The neuroethics and neurolaw of brain injury’, in Behavioral sciences & the law, vol. 31, 2013, 789–802. 222 Choi, 278–285 (p. 282). 223 For more on the limitations of psychological science, see D Lakens, ‘Estimating the reproducibility of psychological science’, in Science, vol. 349, 2015, 1–10. 224 Farahany, 485–509. 225 A Raine, p. 270. 52
Overall, the propositions by these scholars can be summarised as follows: neuroscientific evidence alone is rarely dispositive of a mental state, and should be used to direct and support other types of evidence such as psychological and behavioural evidence — i.e. the convergence of evidence.226
However, if neuroscience should only be used to corroborate other forms of evidence where they point in the same direction, what is the usefulness of neuroscientific evidence?
Morse criticises the usefulness of neuroscientific evidence in courts. To simplify his argument, I will outline the three possible situations that he discusses where neuroscientific evidence might be introduced in the courtroom.227 First, is a case where neuroscience is the sole form of evidence presented to the court. There is no behavioural evidence to explain the offender’s conduct at the time of the offence, and other types of evidence such as psychological and psychiatric examinations are not provided to the court. In such conditions, due to the limitations of neuroscientific evidence discussed above, brain imaging evidence would not have sufficient probative value. For instance, a brain with a structural or functional abnormality does not necessarily show that there is impaired mental and behavioural function.228 The second type of case is where other forms of evidence are provided, and they are clear. In these cases, the introduction of neuroscientific evidence in courts is ‘cumulative and unnecessary’;229 it is merely additive and does not contribute much useful information to what can already be provided by other forms of evidence, e.g. behavioural and psychological evidence. More specifically, Morse finds neuroscientific evidence in these conditions to be rhetorical — neuroscientific evidence is not legally relevant, but makes the case more persuasive.230 The third type of case is where different forms of evidence are in conflict, or they not sufficiently strong enough to meet the standard of proof and are thus merely suggestive. In this case, brain imaging evidence that supports the same conclusion made by other types of evidence can
226 Choi, 278–285; C Scarpazza et al., ‘The charm of structural neuroimaging in insanity evaluations: guidelines to avoid misinterpretation of the findings’, in Translational Psychiatry, vol. 8, 2018, 227–237; JA Silva, ‘Forensic psychiatry, neuroscience, and the law’, in Journal of the American Academy of Psychiatry and the Law, vol. 37, 2009, 489–502. Also see, OD Jones, ‘Seven Ways Neuroscience Aids Law’ in Neurosciences and the Human Person: New Perspectives on Human Activities, AM Battro et al. (eds), Vatican City, The Pontifical Academy of Sciences, 2013, pp. 181–194. 227 Morse, pp. 132–158. 228 See, Morse, pp. 132–158. 229 Morse, pp. 132–158 (p. 145). 230 Morse, pp. 529–562. 53
be potentially useful. Morse, however, claims that these cases are very rare.231
Jones appears to disagree. With respect to the first situation, Jones claims that neuroscientific evidence is rarely, if ever, the sole relevant evidence to the mental state of the defendant.232 The second and third conditions are where the convergence of evidence claim emerges. Jones explains that as both parties should try to make the strongest argument possible, they would admit different types of evidence, despite the fact that even without neuroscientific, the finding of fact would have still been made. He elaborates that even in cases where neuroscience points in the same direction as other types of evidence, it still has a buttressing function in that it supports other presented evidence in forming a more robust conclusion.233 Even if neuroscience does not change the outcome of the court decision, it changes the context, ‘[a]nd that can provide an important advantage to the legal system – in the same way that four different and independent methods for reaching a similar conclusion can provide better support for that conclusion than would just one or two methods alone’. 234
Morse, however, argues that brain imaging evidence is provided at a cost. Brain imaging is time- consuming and expensive, and when weighed up with its limited utility, it would result in a waste of the court’s time and resources. Morse also refers to empirical neurolaw studies235 and argues that:
A critical reader of the empirical studies will be repeatedly struck by how many of the expanded cases either used irrelevant or weak (or nonexistent) neuroscience—for example, to assess competence or whether a defendant suffered from a mental illness—or could have been fully resolved with more careful behavioral evaluation.236
Overall, Morse argues that behavioural evidence will always be more valuable than brain imaging evidence in explaining a person’s mental state,
if there is a disjunction between the subject’s behavior and the neuroevidence, actions always speak louder than images … If the defendant’s brain appears broken, but he is a rational agent, he is rational for legal purposes. If the brain appears normal, but the agent is clearly psychotic, the agent is not rational for legal purposes.237
231 Morse, pp. 132–158. 232 Jones, pp. 181–194. 233 See, Jones, pp. 181–194. 234 See, Jones, pp. 181–194 (p. 186). 235 For example, Farahany, 485–509; Catley and Claydon, 510–549; Gaudet and Marchant, 577–661; de Kogel and Westgeest, 580–605. 236 Morse, pp. 132–158 (p. 149). 237 SJ Morse, ‘Actions speak louder than images: the use of neuroscientific evidence in criminal cases’, in Journal of Law and the Biosciences, vol. 3, 2016, 336–342 (p. 339). Also see, SJ Morse, ‘Indispensable 54
Those that support the convergence of evidence approach also support the claim that behavioural evidence is more credible than brain imaging evidence. Choi explains that,
When behavioral evidence conflicts with neuroimaging findings, in general the high percentage move will be to side with the behavioral, because neuroscience is so poor at predicting individual outcomes of brain defects. In other words, at this point, in most cases careful behavioral analysis continues to be more reliable than neuroimaging in ascertaining the relevant mental states, capacities, and behaviors that form the actual basis of legal criteria.238
This does not mean that advocates of the convergence of evidence approach agree with Morse in that neuroscientific evidence is not valuable to the courtroom. Cristina Scarpazza et al., for example, argues that while it is a logical error to use neuroscientific data in isolation, it is a similar error to ignore its potentially helpful applications.239
Unlike the convergence of evidence claim, the usefulness of neuroscientific evidence is not one of the claims that I will examine in this study. However, I will return to it in the Conclusion of this thesis.
The persuasive power of neuroscientific evidence
A number of researchers have claimed that laypersons may be unduly persuaded by neuroscientific evidence. Courts may find such evidence more convincing and scientifically credible than can actually be justified, and by extension give the evidence undue weight. Brown and Murphy argue that ‘[t]he mere presentation of brain images can make a bad argument appear more sensible’.240 David McCabe and Alan Castel demonstrated that adding brain images to other scientific arguments would make those arguments seem more credible compared to arguments that were not supported by these images, or supported by graphs.241 Similarly, a study by Weisberg et al. suggested that jurors are more likely to find explanations of an individual’s behaviour persuasive when they are expressed in terms of neuroscientific evidence. 242 In a study with 396 mock jurors,
Forensic Psychiatry and Psychology: The (non) challenge from neuroscience’ in The Evolution of Forensic Psychiatry: History, Current Developments, Future Directions, LR Sadoff (ed), New York, Oxford University Press, USA, 2015, pp. 339–358. 238 Choi, 278–285 (p. 283). 239 Scarpazza et al., 227–237. 240 Brown and Murphy quoted in J Aronson, ‘The law’s use of brain evidence’, in Annual Review of Law and Social Science, vol. 6, 2010, 93–108. 241 DP McCabe & AD Castel, ‘Seeing is believing: The effect of brain images on judgments of scientific reasoning’, in Cognition, vol. 107, 2008, 343–352. 242 DS Weisberg et al., ‘The seductive allure of neuroscience explanations’, in Journal of cognitive neuroscience, vol. 20, 2008, 470–477. 55
it was demonstrated that defendants in cases where neuroscientific evidence was introduced were six times more likely to receive an order of not guilty by reason of insanity.243 Some scholars believe that judicial actors’ abilities to analyse neuroimaging studies are inadequate, and that education is required to enable a better understanding of this evidence.244
Although it appears that neuroscience, or in other words, discussions about the brain, are seductive to some,245 other scholars see a problem in public fascination with images in particular. For instance, some point to the ‘Christmas tree effect’ and explain that colourful brain images can inappropriately affect judges and jurors decisions.246 Walter Sinnott-Armstrong et al. refer to a study showing that jurors may be affected by photographs (not brain images) more than its appropriate.247 They then suggest that as functional brain images (e.g. fMRI) look similar to photographs, jurors may also find these more informative and persuasive than is warranted. Although functional brain images may seem to portray brain function, these images do not capture a visual field, rather they are generated from abstract numerical data and visualisation effects.248
Concurrently, there are scholars who claim that courts do not give undue weight to brain images. Nicholas Schweitzer et al., conducted a meta-analysis of four experiments that included 1476 jury-eligible participants overall, to measure the extent to which neuroscientific explanations and neuroimages affected jurors’ decisions.249 They concluded that brain imaging evidence did not unduly influence judgement. They found that when it came to determining the extent of the defendant’s capacity to control their actions was reduced due to a disorder, neuroscientific evidence was more persuasive than clinical psychological evidence. However, this did not appear to influence the verdicts.250 Valerie Gray Hardcastle claims that the proportion of appellate cases where the court order is in favour of the defendant with brain imaging evidence is equal to the proportion of cases where brain imaging evidence is not introduced to the court.251
243 JR Gurley & DK Marcus, ‘The effects of neuroimaging and brain injury on insanity defenses’, in Behavioral sciences & the law, vol. 26, 2008, 85–97. 244 D Horstkötter, C van El & M Kempes, ‘Neuroimaging in the Courtroom: Normative Frameworks and Consensual Practices’, in AJOB Neuroscience, vol. 5, 2014, 37–39. 245 D Keltner & P Ekman, ‘The Science of ‘Inside Out’, in New York Times, vol. 3, 2015, cited in Choi, 278–285. Also see, Crockett. 246 See, Aggarwal and Ford, 789–802. 247 W Sinnott-Armstrong et al., ‘Brain images as legal evidence’, in Episteme, vol. 5, 2008, 359–373; DA Bright & J Goodman-Delahunty, ‘Gruesome evidence and emotion: anger, blame, and jury decision- making’, in Law and human behavior, vol. 30, 2006, 183–202. 248 Sinnott-Armstrong et al., 359–373. Also see, A Roskies & W Sinnott-Armstrong, ‘Brain images as evidence in the criminal law’, in Law and Neuroscience, 2011. 249 NJ Schweitzer et al., ‘Neuroimages as evidence in a mens rea defense: No impact’, in Psychology, Public Policy, and Law, vol. 17, 2011, 357–393. 250 Schweitzer et al., 357–393. 251 VG Hardcastle, ‘Why Brain Images (Probably) Should Not Be Used in US Criminal Trials’ in The Palgrave Handbook of Philosophy and Public Policy, D Boonin (ed), Palgrave Macmillan, Cham, 2018, pp. 25–37. 56
Further, some studies concluded that the persuasive power of neuroscientific evidence emerges under some circumstances and contextual matters such as an individual’s pre-existing beliefs and their view and attitudes toward scientific studies.252 Overall, these conflicting evaluations of the influence of neuroscientific evidence on judgements paint a more complex picture of whether such evidence is potentially overly persuasive.
Expert witnesses
Arguments have also been made about the types of expertise that those who report on neuroscientific should hold, and the resulting value of such reports to legal decision-makers. Rose and Abi-Rached explain that brain images ‘do not speak for themselves and must be interpreted by experts’. 253 In other words, the courts’ understanding of brain imaging evidence depends on what expert witnesses report. Some argue that expert witnesses who do not possess the requisite expertise enabling them to discuss brain imaging evidence render their reports and testimonies not credible or sufficiently reliable.254 For instance, Daniel G. Amen and Lois T. Flaherty believe that psychiatrists do not have expertise in areas of brain imaging, as they have not received training in these areas.255 Douglas Bremner claims that many mental health professionals, such as social workers, psychologists and psychiatrists, do not have the required knowledge of brain anatomy and physiology to apprehend the neurobiology of mental disorders and to interpret brain images.256 Jennifer Kulynych (1996) finds that neurologists or psychiatrists may not have the technical expertise to understand or conduct brain scans. Brain-imaging is an extremely complicated area that can engage radiologists, psychiatrists, physicists, and biochemists. However, while the interpretation of scans requires multidisciplinary expertise when a brain scan is admitted in court, there may only be one expert (a psychiatrist or neurologist) responsible for diagnosing impairments.257
However, the determination of which professions are qualified to express opinions about neuroscientific evidence in courts is difficult and not well-characterised in the neurolaw literature. Moriarty criticises the current neurolaw literature for primarily focusing on the reliability and
252 DA Baker et al., ‘Making sense of research on the neuroimage bias’, in Public Understanding of Science, vol. 26, 2017, 251–258. 253 Rose and Abi-Rached. 254 JC Moriarty & DD Langleben, ‘Who Speaks for Neuroscience-Neuroimaging Evidence and Courtroom Expertise’, in Case W. Res. L. Rev., vol. 68, 2017, 783–804. 255 D Amen & L Flaherty, ‘Is brain imaging clinically useful for psychiatrists?’, in Clinical Psychiatry News, vol. 34, 2006, 11. 256 JD Bremner, Brain imaging handbook, WW Norton & Co, 2005. 257 J Kulynych, ‘Brain, mind, and criminal behavior: neuroimages as scientific evidence’, in Jurimetrics, 1996, 235–244. 57
relevance of brain imaging in courts while neglecting discussion about the imaging qualifications of expert witnesses.258 She argues that this may be due to the complexity of the field of brain imaging,
The wide-ranging pursuits of the neuroscience field include intersecting and often overlapping areas of expertise among psychologists, research scientists, and various categories of physicians. Yet, lawyers and judges are not always aware—nor could they be—of the general boundaries of expertise and the precise boundaries of expertise in a given matter on a specific issue.
A major complexity of an expert witness’ qualification arises from the fact that title or formal education attributable to a particular expert (e.g. a psychologist or psychiatrist) does not necessarily determine that expert’s competency. Arturo J. Silva explains that forensic neuropsychiatric competency may be acquired by receiving training in forensic psychiatry and neuropsychiatry. There are many psychiatrists, however, who may not have had those options in their formal training.259 Further, in a study conducted by David Linden, it was demonstrated that about 70% of neuropsychologists in the study had received training in neuroimaging, the majority of which through clinical conferences, individualised mentoring and independent study.260
As such, despite the important position of expert witnesses and their knowledge in understanding and interpreting brain images, there are unclear lines in determining which expert witnesses are qualified to interpret neuroscientific evidence.
Consideration of the neurolaw claims in this section makes it seem as if the use of neuroscientific evidence in criminal proceedings presents new and potentially difficult conceptual and doctrinal challenges. However, many of the neurolaw claims about the use of neuroscience in courts are based on a limited number of cases, subjective experience of interacting with the legal system, speculation, or media reports, rather than a comprehensive analysis of judgements involving neuroscientific evidence. Farahany warns that ‘the fundamental assumptions guiding the current ethical, legal and social inquiry into the use of neurobiological evidence in criminal law are limited and potentially flawed’.261 Similarly, Denno argues that neurolaw discussions must be
258 Moriarty and Langleben, 783–804. Moriarty also outlines a few articles discussing experts’ area of expertise regarding neuroscientific evidence including, P Patel et al., ‘The role of imaging in United States courtrooms’, in Neuroimaging Clinics of North America, vol. 17, 2007, 557–567; NK Aggarwal, ‘Neuroimaging, culture, and forensic psychiatry’, in The journal of the American Academy of Psychiatry and the Law, vol. 37, 2009, 239–244. 259 Silva, 489–502. 260 J Hassenstab & KJ Bangen, ‘Neuroimaging Training Among Neuropsychologists: A Survey of the State of Current Training and Recommendations for Trainees’, in The Clinical Neuropsychologist, vol. 28, 2014, 600–613. 261 Farahany, 485–509 (p. 491). 58
‘grounded in fact rather than hyperbole’.262 In the next section, I will review all the empirical studies that discuss the use of neuroscience in courts based on a comprehensive and systematic analysis of criminal cases, and outline the existing gaps in such research in Australia.
4. Neuroscience in courts: empirical neurolaw studies
Cases in which neuroscientific evidence has been introduced in criminal courts go back decades. In the US, neuroscientific evidence was tendered in the high-profile case of John Hinckley, who was accused of attempting to assassinate President Ronald Reagan in 1982. During Hinckley’s trial, a CT scan, along with other evidence, was used to prove schizophrenia, which resulted in the successful defence of not guilty by reason of insanity.263 In England, one of the earliest uses of neuroscientific evidence was in 1958, where a defendant used EEG evidence to support his claim of involuntary behaviour.264 Despite the relative longevity of the use of neuroscience in courts, and the fact that it has been raised in different judicial systems in various countries, only six studies have adopted an empirical strategy to explore and understand the use of neuroscientific evidence in courts.265 Perhaps surprisingly, all of these studies were published after 2015, three of which examine the US jurisdiction, while the other three examine Canada, Netherlands and England and Wales jurisdictions (see Table 2-1).266
Table 2-1: Empirical studies.
Year of Study Country Author/s publication
The use of neuroscientific evidence in Canadian criminal 2015 Canada Chandler, J. A. proceedings267
262 Denno, 493–551 (p. 494). 263 L Gaudet, J Lushing & K Kiehl, ‘Functional Magnetic Resonance Imaging in Court’, in AJOB Neuroscience, vol. 5, 2014, 43–45. 264 C Willmott, ‘Use of Genetic and Neuroscientific Evidence in Criminal Cases: A Brief History of “Neurolaw”’ in Biological Determinism, Free Will and Moral Responsibility: Insights from Genetics and Neuroscience, Springer, 2016, pp. 41–63. 265 In this regard, Owen D Jones et al. (2013) noted the difficulty of finding clear statistics for the application of neuroscience in courtrooms. OD Jones et al., ‘Law and Neuroscience', in Journal of Neuroscience, vol. 33, 2013, 17624–17630; cited in P Catley & L Claydon, ‘The use of neuroscientific evidence in the courtroom by those accused of criminal offenses in England and Wales’, 2015, 1–40. 266 The study by Calvin Ho titled ‘The use of neuroscientific evidence by defendants in criminal trials in Singapore and Malaysia 2000-2012’ is not considered in this research as the paper is a forthcoming publication and sufficient information of its content is not available. 267 An alternative citation with different year is: Chandler, J. A. (2016). The use of neuroscientific evidence in Canadian criminal proceedings. Journal of Law and the Biosciences, 2(3), 550-579. 59
The use of neuroscientific evidence in the courtroom by England Catley, P. and 2015 those accused of criminal offences in England and Wales and Wales Claydon, L.
De Kogel, C.H. and Neuroscientific and behavioural genetic information in 2016 Netherlands Westgeest, criminal cases in the Netherlands E.J.M.C.
Under the radar: Neuroimaging evidence in the criminal United Gaudet, L. M. and 2016 courtroom States Marchant, G. E.
Neuroscience and behavioural genetics in US criminal United 2016 Farahany, N. A. law: an empirical analysis States
The myth of the double-edged sword: an empirical study United 2015 Denno, D. W. of neuroscience evidence in criminal cases States
All of these empirical neurolaw studies use a similar methodology. By using legal databases, they collected and analysed criminal judgements that introduced neuroscience in court. The general objectives of these studies were to provide fact-based evidence of the use of neuroscience in criminal proceedings. There are, however, some differences across the studies. For example, two of these studies examined the use of genetic evidence in courts, along with neuroscientific evidence. Most of them used the broad definition of neuroscience (i.e. both imagining and non- imaging evidence), but some only focused on brain imaging evidence.268 Further, the number of cases analysed in these studies and the time period considered are also different.
While the number of empirical studies is limited, these studies have revealed that the use of neuroscience in practice is in contradiction with some of the neurolaw claims. For instance, it appears that neuroscience, in the majority of cases, is only used to mitigate the effects on the sentence and therefore it does not have a double-edged sword effect in the US courts.269 Further, contrary to what some scholars have claimed, courts have been receptive toward neuroscientific evidence.270 And, courts did not appear to have been misled by neuroscientific evidence.271
In Australia, the very first demonstration (found to this date) of the use of neuroscience through
268 Gaudet and Marchant, 577–661. 269 DW Denno, ‘How Prosecutors and Defense Attorneys Differ in Their Use of Neuroscience Evidence’, in Fordham Law Review, vol. 85, 2016, 453–479. 270 For example see, Gaudet and Marchant, 577–661. 271 See, Kellmeyer, 530–554. 60
expert evidence to examine a defendant’s mental state dates back to 1975.272 Despite this, there has been no large-scale empirical study that demonstrates the practical use of this evidence in Australian criminal proceedings. There are, however, four studies that consider a small number of cases to examine the use of neuroscience in practice. The first study, by Ellie A. Page (2017), discusses the use of neuroscience as a mitigating factor in NSW sentencing courts.273 The second study by Allan McCay and Christopher J. Ryan (2018), examines an appeal case and judges’ reasons around the expert witness and the neuroscientific evidence.274 The third study, conducted by McCay and Jeanette Kennett (2019), examines Greene and Cohen’s claim about whether neuroscience has revolutionised the Australian criminal justice system.275 The fourth study, by Armin Alimardani and Jason Chin (2019), discusses the different uses of neuroscientific evidence in the pre-trial, trial, and sentencing phases of the Australian criminal justice system.276 While Alimardani and Chin produced the most comprehensive examination of the use of neuroscience in Australia, their study was much less comprehensive and systematic compared to empirical neurolaw studies from other jurisdictions and, further, did not provide any quantitative analysis of judgements.277
The Australian Neurolaw Database project was launched in December 2015, and is a public database that aims to collect neuroscientific cases within Australia,278 to provide a comprehensive picture of the use of neuroscience in courts. Unfortunately, results cannot be provided in the near future due to the lengthy process of data collection and analysis.279
To fill the gap this thesis attempts to systematically collect and analyse criminal cases where neuroscientific evidence has been introduced in criminal courts in New South Wales, Australia. The aim is to provide a fact-based discussion of the use of neuroscience with respect to the neurolaw claims considered in this study.
272 Veen v R in [1979] HCA 7; (1979) 143 CLR 458; The initial court was held in NSW, in 1975. 273 EA Page, ‘The Criminal Mind: Neuroscientific Evidence as a Mitigating Factor in Sentencing in New South Wales, Australia’, in Pac. Rim L. & Pol’y J., vol. 26, 2017, 659–691. 274 A McCay & CJ Ryan, ‘Issues pertaining to expert evidence and the reasoning about punishment in a neuroscience-based sentencing appeal’, in International Journal of Law and Psychiatry, vol. Forthcomin, 2018. 275 A McCay & J Kennett, ‘Neuroscience and Punishment: From Theory to Practice’, in Neuroethics, 2019. 276 Alimardani and Chin. 277 McCay and Ryan. 278 Both civil and criminal cases. 279 See, https://neurolaw.edu.au. As a member of this project, I have provided further insights about the use of neuroscience in Australia, and have shared my experiences with collecting and analysing these judgments. 61
Conclusion
The relationship between neuroscience and the law has been the centre of many scholarly studies and resulted in a variety of claims about the use of neuroscientific evidence in criminal courts. Some of these arguments centre around how neuroscience will alter the concept of criminal responsibility. On the other hand, some advocate for the use of neuroscience to provide useful information to accommodate established principles of criminal law. This includes predicting recidivism and explaining the mental state of the defendant at the time of the crime. Further, there are some who have been more sceptical and cautious about the use of neuroscience in courts, claiming that neuroscience is still a new science in its formative stages and that hasty application of its findings may have unintended or even dangerous results. The court may be misled by neuroscientific evidence, or they may overlook the complex limitations of such evidence. Concurrently, others embrace the use of neuroscience in courts and have sought to find approaches that deal with some of the shortcomings of this evidence (e.g. convergence of evidence approach).
Despite the use of neuroscience in courts and the wide array of arguments about it, neurolaw claims are often not substantiated by the empirical analysis of the use of neuroscience in courts. In some countries, a few empirical studies have provided a systematic comparative analysis on how and for what purposes neuroscientific evidence is being used in courtrooms. Since such examination in Australian courts has not yet been conducted, the aim of this thesis is to fill this gap in the neurolaw literature.
62
Chapter 3 Legal Background
63
Introduction
This Chapter aims to introduce some of the key aspects of the NSW legal framework. This review is useful in order to better understand some of the methodological bases of the empirical research and discussions about findings of this study with respect to neurolaw claims. It is especially useful to readers who are unfamiliar with the NSW criminal justice system.
The Chapter begins with an overview of the NSW criminal justice system and then moves on to outline different sentencing factors that a court may consider in imposing a sentence. In the third section, I will discuss the complexity and technicality of the criminal justice system in determining a proper sentence. The final section of this Chapter discusses the different trial and appellate criminal courts and their jurisdiction for different types of crimes.
To avoid undue complexity and provide meaningful discussions of NSW legal rules with respect to neurolaw claims, I will present further information on legislative provisions in later Chapters as the appropriate context arises.
1. The framework of the criminal justice system
NSW’s criminal system is adversarial1 (rather than inquisitorial). In this system, parties – the defendant and the prosecutor – are responsible for finding evidence and proving their own versions of the facts by providing the strongest argument possible to the court. The defence lawyer is obliged to advise their clients on any matter that may benefit them (e.g. reduce their sentence),2 and the prosecutor is obliged to provide any relevant materials to criminal procedure, whether they support or damage the defendant’s argument.3
The judge and the jury are neutral and their decisions are based on the adduced evidence.4 The court is not permitted to request evidence (except in exceptional situations) nor can it take an
1 Also known as the adversarial system. 2 Judicial Commission of New South Wales, ‘Sentencing Bench Book’, in Judicial Commission of New South Wales,
inquisitorial role if there is an ‘unsatisfactory statement of facts’.5 The adversarial system has been criticised for failing to adequately prioritise the search for the truth: judges cannot use their experience or knowledge to guide parties toward the truth.6 Therefore, if none of the parties submits evidence that is expected by the jury or the judge, the court is not entitled to ask for that evidence.7 Judges can only ask for clarification of issues raised in evidence.8
Accordingly, in the adversarial system lawyers have a considerable forensic responsibility.9 For instance, whether they call an expert witness to discuss the defendant’s mental state; or whether lawyers can identify an expert whose area of expertise is suitable for medical assessment of the defendant. For instance, a psychologist may not be able to provide an accurate and credible psychiatrist assessment.
Evidence with respect to the mental state of individuals may be submitted in three phases of criminal procedure: pre-trial, trial (guilt phase) and sentencing (penalty phase). During pre-trial, evidence may be submitted to prove that the defendant is not fit to stand trial. That is, whether the defendant would be able to understand matters such as the nature of the charge, comprehend the evidence that being presented, and communicate with judicial officers during the trial.10 If the person is found unfit to stand trial, the accused is referred to as a ‘forensic patient’11 and a special hearing will take place:12 a hearing with the aim of ensuring that although the accused is found unfit, they are tried ‘in accordance with the normal procedures, that the person is acquitted unless it can be proved to the requisite criminal standard of proof that, on the limited evidence available, the person committed the offence charged or any other offence available as an alternative to the offence charged.’13 If the court finds the accused guilty of a crime, it will order a limiting term – ‘the best estimate of the sentence’ that the court would have ordered for an offender following a normal trial.14 This procedure is normally determined by a judge alone (that is, without a jury).15
5 Odgers. 6 W van Caenegem, ‘Advantages and disadvantages of the adversarial system in criminal proceedings’, 1999. 7 See O’Neill-Shaw v R [2010] NSWCCA 42 8 Judicial Commission of New South Wales. 9 van Caenegem, 69–102. 10 A Samuels, C O’Driscoll & S Allnutt, ‘Fitness issues in the context of judicial proceedings’, in Australasian Psychiatry, vol. 15, 2007, 212–216. 11 In simple terms, any person with mental illness in contact with the criminal justice system is called a forensic patient. 12 If the person is found to be unfit, he/she is referred to the Tribunal. If the Tribunal finds that within 12 months the accused will not become fit to be tried, then the court will hold a special hearing. For more information, see, Mental Health (Forensic Provisions) Act 1990 (NSW) 13 S 19(2) Mental Health (Forensic Provisions) Act 1990 (NSW) 14 S 23 Mental Health (Forensic Provisions) Act 1990 (NSW) (emphasis added). 15 S 11 Mental Health (Forensic Provisions) Act 1990 (NSW) 65
The trial is the only criminal phase in which the jury participates.16 During the trial, other than deciding whether the accused is guilty or not guilty, there are two other orders available to the jury. First, the defence of mental impairment/illness (also known as the insanity defence): a successful defence requires the accused to demonstrate that, at the time of the crime, he/she did not understand either the nature and quality of their act or if their act was wrong.17 Second, the defence of substantial impairment by abnormality of the mind (also known as the defence of diminished responsibility), a partial defence to murder, where the charge of murder is reduced to manslaughter. For a successful defence of substantial impairment, the defence should prove that, at the time of the crime, the accused’s capacity to understand events, judge their actions, or control their behaviour was substantially impaired by an abnormality of the mind.18
In the sentencing phase, the judge determines the appropriate sentence based on the available evidence and sentencing rules where the jury has no authority. The relevant evidence to sentencing may have been submitted during pre-trial, trial, or sentencing. However, the judge’s consideration of facts in determining the sentence must be consistent with the jury’s decision made in the trial.19 For instance, if the jury found the offender guilty of manslaughter based on the defence of substantial impairment by abnormality of the mind, the judge, in determining the sentence, should rely on the evidence that appears to be in accordance with the jury’s decision; even if the evidence is submitted in the sentencing phase – after the trial.
With this introduction to the criminal system, I will move on to briefly discuss the sentencing factors that the judge may consider in determining the appropriate sentence, with a focus on matters that may be relevant to neuroscience.
16 As such, unlike the US courts, in sentencing procedure there is no jury and the judge orders the sentence. In NSW, however, the Local Court, where the least serious crimes are dealt with, has no juries. I will return to this point in the following section. 17 S Allnutt, A Samuels & C O’driscoll, ‘The insanity defence: from wild beasts to M’Naghten.’, in Australasian psychiatry : bulletin of Royal Australian and New Zealand College of Psychiatrists, vol. 15, 2007, 292–298; D Howard & B Westmore, Crime and Mental Health Law in New South Wales: A Practical Guide for Lawyers and Health Care Professionals, 2nd ed., LexisNexis, AU, 2010. NSW adopted this defence from England, known as the M’Naghten Rules. There are some differences between the states of Australia regarding establishing the defense of mental illness. For more information about these differences, see, S Bronitt & B McSherry, Principles of criminal law, in Thomson Reuters, L Weeks & NT Rozelle (eds), 3rd ed., 2010, pp. 240–241. 18 S Bronitt & B McSherry, Principles of Criminal Law, in Thomson Reuters (Professional) Australia Pty Limited, 4th ed., 2017. 19 Odgers. Also see, R v Isaacs (1997) 41 NSWLR 374 66
2. Sentencing factors
These sentencing factors are originated in both legislation and the common law. The statutory provisions are not intended to contradict common law rules. In practice, common law dominates sentencing and the statutory provisions are a means to accommodate the common law rules.20 There are potentially three categories of sentencing factors in determining the sentence: aims of sentencing, mitigating and aggravating factors, and other factors not included in the last two sections. Sentencing in general is a technical and complicated process and some sentencing factors are interconnected or overlap with others.
The aims of sentencing
The aims of sentencing are explained both in s 3A Crimes (Sentencing Procedure) Act 1999 (NSW) (hereafter Sentencing Act) and the common law. Although some common law aims of sentencing are not mentioned in s 3A, courts should consider them in the sentencing process.
Proportionality is an aim of sentencing and it is referred to in both the common law and s 3A(a) of Sentencing Act where its purpose is defined as being: ‘to ensure that the offender is adequately punished for the offence’. This principle is designed to avoid a sentence which exceeds or falls short of the seriousness of the offence. The seriousness of an offence is determined by objective factors: those that are associated with the circumstances of the offences (e.g. physical attack), and not the characteristics of the offender at the time of the crime (e.g. their age).21 There are, however, some specific subjective factors that may have an influence on the objective seriousness of the crime. One of these factors is the moral culpability of the offender. In R v Israil, the court with respect to where the mental illness of the offender explains the offence and its influence on the sentence, explains that:
An offender’s inability to understand the wrongfulness of his action, or to make reasonable judgements, or to control his or her faculties and emotions, will impact on the level of culpability of the offender.22
The reason that moral culpability affects the seriousness of the offence is that it is causally
20 Odgers. 21 See Markarian v R (2005) 228 CLR 357 at [51] per McHugh J; R v Scott (2005) NSWCCA 152 at [15] 22 (2002) NSWCCA 255 at [23]. Also in Mudlrock [2011] HCA 39, 244 CLR 120 at [58], it was noted that ‘punishment, in the sense of retribution, and denunciation did not require significant emphasis in light of the appellant’s limited moral culpability for his offence’. 67
associated with committing the offence. As such, those matters that influence the level of moral culpability ‘can be classified as circumstances of the offence and not merely circumstances of the offender’.23 If, for example, an offender’s lower capacity to make reasonable judgements as the result of some mental issues is causally connected to the offence, they are less blameworthy and deserve less punishment than offenders with no mental problem.
Retribution is another aim of sentencing that is defined in common law. This aim overlaps with proportionality. In R v Gordon (1994), retribution was considered as ‘the taking of vengeance for the injury which was done by the offender’.24 However, ‘vengeance’ is not the only purpose of retribution. Justice McHugh in Ryan v R argued that the community has to be satisfied that the criminal is receiving their ‘just deserts’.25 According to the ‘just deserts’ approach, the punishment must be no more or less than is proportionate to the seriousness of the offence and the culpability of the offender.26
Denunciation is expressed in s 3A(f) Sentencing Act as intended ‘to denounce the conduct of the offender’. An example of how denunciation is considered in sentencing is ‘where mental illness contributes to the commission of the offence in a material way, the offender’s moral culpability may be reduced; there may not then be the same call for denunciation and the punishment may accordingly be reduced’.27 A similar aim of sentencing is accountability that is listed in s 3A(e) of the Sentencing Act as follows: ‘to make the offender accountable for his or her actions’. Stephen Odgers argues that accountability does not add anything new to the aims of sentencing and assumes that ‘it might be thought that holding an offender ‘accountable’ for his or her actions and recognition of any harm done is a necessary prelude to denunciation, as well as guiding the appropriate retribution’.28 As such, similarly to denunciation, an offender whose mental
23 Kennedy v R (2008) 181 A Crim R 185; [2008] NSWCCA 21 at [38] and [39]. 24 In R v Gordon (1994) 71 A Crim R 459 Hunt CJ at CL said at 468 (citations are omitted). 25 Ryan v The Queen (2001) 206 CLR 267 per McHugh J at [46]. McHugh also referred to the history of how retribution became more important in CJS after the period in around mid-20th century that rehabilitation was more stressed see: Ryan v The Queen (2001) 206 CLR 267 per McHugh J at [46] 26 R White, F Haines & N Asquith, Crime and criminology, 5th ed., Oxford University Press, 2012. Regarding just deserts and moral culpability Odgers states that: ‘retribution tends to focus on the harm done by the offender but, as the term “just deserts” indicates, it also recognizes that an important consideration is the offender’s level of culpability’ Odgers, p. 73. Also Allan McCay in this regard concludes that ‘the “just deserts” movement could be thought of as a retributive movement, being a movement that advocated a focus upon retributive desert by way of the notion of proportionality’. A McCay, ‘Behavioural Genetics, Moral Agency and Retributive Sentencing: The Case for Mitigation (Doctoral dissertation)’, Sydney University, 2013, p. 47. 27 R v Hemsley (2004) NSWCCA 228; BC200404239 Per Sperling J at [33-36] (citations are omitted), quoted in Howard and Westmore. 28 Odgers, p. 79. Some legal scholars may have a different interpretation of this section. However, those arguments are outside of the scope of this research. 68
abnormality causally influenced the commission of an offence is held to be less accountable than an individual without such an abnormality.29
Promoting rehabilitation of the offender is another aim of sentencing that is recognised at common law and also outlined in s 3A(d) of the Sentencing Act. This aim of sentencing includes recognising and addressing underlying issues that may increase the risk of re-offending.30 It is further closely associated with another aim of sentencing, the protection of society. Section 3A(c) of the Sentencing Act articulates this factor as intended ‘to protect the community from the offender’. That means that during the period the offender is imprisoned society is kept safe from any further offending. Prospects of rehabilitation and protection of the community may have a substantial effect on one another. That is, a person with good prospects of rehabilitation will be considered less of a threat to society once they are released, while an individual with poor prospects of rehabilitation may pose a greater risk and the protection of society will require higher consideration.31 As such, prospects of rehabilitation and protection of the community are two particularly important factors in cases where an abnormality (e.g. a mental disorder) contributes to the criminal behaviour and increases the risk of recidivism.32
It is of note that rehabilitation has a broader implication than merely treating a mental abnormality or preventing re-offending. In R v Pogson, the court explained that:
rehabilitation has as its purpose the remodelling of a person’s thinking and behaviour so that they will, notwithstanding their past offending, re-establish themselves in the community with a conscious determination to renounce their wrongdoing and establish or re-establish themselves as an honourable law abiding citizen. In this sense, every offender is in need of rehabilitation. Some may need greater assistance than others.33
An offender whose mental problems contributed to the crime may receive proper treatment and the protection of society will no longer be a significant consideration in sentencing. However, he or she may not admit their accountability or may show no remorse for their past offending, and the court may consider this to be evidence of poor prospects of rehabilitation.
Section 3A(b) of the Sentencing Act identifies deterrence as another aim of sentencing: ‘to prevent crime by deterring the offender and other persons from committing similar offences’. This section is referring to two types of deterrence: general deterrence (which means to deter the public from
29 See, McCay, p. 200. 30 R v Pogson (2012) 82 NSWLR 60 at [103] 31 See R v Z (2006) 167 A Crim R 436; [2006] NSWCCA 342 at [50] 32 See Id. 33 (2012) 82 NSWLR 60 per McClellan CJ at CL and Johnson J at [124-125], Price, RA Hulme and Button JJ agreeing at [152], [155] and [156] (citations are omitted). 69
engaging in criminal behaviour by punishing the offender); and specific deterrence (which is punishing the offender specifically to deter him or her from recidivism).
The aim of deterrence may receive lower or no weight during sentencing in different circumstances. For instance, it has commonly been stated that an offender with mental problems may not be a suitable vessel for general deterrence.34 The degree to which the significance of general deterrence is lowered or eliminated depends on ‘the nature and severity of the symptoms exhibited by the offender, and the effect of the condition on the mental capacity of the offender’.35 Further, if there is no causal link between the mental problems and the crime or any risk of recidivism, the significance of specific deterrence in sentencing might be mitigated.36 Conversely, if ‘it is not shown to be likely that the offender will accept the need for treatment in the future’, specific deterrence may receive greater consideration.37
Mitigating and aggravating factors in sentencing In s 21A of the Sentencing Act, aggravating and mitigating factors are listed. The difference between this section and s 3(A) of the Sentencing Act is that while all the purposes of sentencing must be considered in reaching an appropriate sentence,38 courts should consider the matters under s 21(A) only when they are relevant to conditions of the case.39
There are no aggravating factors under this section which would be of potential relevance to neuroscience or any of the general neurolaw claims considered in this study; however, there are four relevant mitigating factors. The first mitigating factor is that the offender could not fully appreciate the consequences of their conduct.40 The second mitigating factor states that ‘the offence was not part of a planned or organised criminal activity’.41 Both these factors can be potentially relevant to sentencing in a case where, for example, some abnormality of mind is identified which suggests that the offender could not fully understand the consequences of their criminal behaviour, or that the offender is unable to plan criminal activity and their behaviour is
34 See Engert (1995) 84 A Crim R 67 at 71 per Gleeson CJ; R v Hemsley (2004) NSWCCA 228 35 R v Tsiaras (1996) 1 VR 398 36 Odgers. 37 Clay (2007) NSWCCA 106 at [25]-[26] quoted in Odgers, p. 277. 38 Odgers. Markarian (2005) HCA 25, 228 CLR 357, Gleeson CJ, Gummow, Hayne and Calllinan JJ at [26]. A purpose of sentencing that is not included in section 3A is the ‘advancement of systemic goals’ which help with the effectiveness and efficiency of the criminal justice system. For example, a discount in sentencing for a guilty plea or assistance due to the utilitarian value of such actions. See Odgers, pp. 70, 71 and 73. 39 Judicial Commission of New South Wales. See DBW v R (2007) NSWCCA 236 at [33], [36]. Van Can Ha v R (2008) NSWCCA 141 at [4]. 40 Section 21A(3)(j) Crimes (Sentencing Procedure) Act 1999 41 Section 21A(3)(b) Crimes (Sentencing Procedure) Act 1999 70
instead impulsive and spontaneous.
The third mitigating factor is if ‘the offender has good prospects of rehabilitation, whether by reason of the offender’s age or otherwise’.42 ‘Otherwise’ can be interpreted to include medical evidence that suggests the abnormality of mind which contributed to criminal behaviour is treatable, and therefore the offender has good prospects of rehabilitation. The fourth mitigating factor is whether ‘the offender is unlikely to re-offend’.43 Similarly to the previously mentioned mitigating factor, an expert witness may present evidence suggesting that what caused the criminal behaviour has been treated and it is therefore unlikely that the offender will commit a further offence. These last two mitigating factors are similar to the two aims of sentencing – promoting rehabilitation of the offender and the protection of society. Despite the similarity, there are differences in the application of these factors during sentencing. For instance, if the court is not positive about the offender’s prospects of rehabilitation, it does not then consider the mitigating factor of good prospects of rehabilitation under s 21(A) of the Sentencing Act. Nonetheless, since an aim of sentencing is the promotion of rehabilitation (s 3(A) of the Sentencing Act), the court must take into consideration what can be done to rehabilitate the offender (e.g. by ordering a rehabilitative program).
Other factors Two other sentencing factors are only referred in the common law and are not listed under the aims of sentencing or in the aggravating and mitigating factors in s 21A of the Sentencing Act. These are circumstances in which a mental or physical abnormality makes a custodial sentence more onerous than for an average offender;44 or where imprisonment would have a significant impact on the offender’s health.45 In such cases the court may reduce the sentence.
In the following section, I will discuss how courts consider these sentencing factors in imposing sentence.
3. Deciding on the proper sentence
Sentencing is a technical process, and there are two main factors that contribute to this complexity. First is the difficulty of identifying relevant factors and assigning them appropriate
42 Section 21A(3)(h) Crimes (Sentencing Procedure) Act 1999 43 Section 21A(3)(g) Crimes (Sentencing Procedure) Act 1999 44 Freckelton and Selby. Also see, R v Hemsley (2004) NSWCCA 228 45 King CJ in R v Smith (1987) 44 SASR 587 at 589. 71
weight in each particular case. The High Court of Australia, with respect to this difficult process explained that:
sentencing is not a purely logical exercise, and the troublesome nature of the sentencing discretion arises in large measure from unavoidable difficulty in giving weight to each of the purposes of punishment. The purposes of criminal punishment are various ... The purposes overlap and none of them can be considered in isolation from the others when determining what is an appropriate sentence in a particular case.46
Second is the judges’ discretion in sentencing. While there are some mechanisms to help formulate decisions regarding an appropriate sentence; aside from limited exceptions, sentencing is a matter for the court’s discretion. In determining the sentence, courts exercise the process of ‘instinctive synthesis’. This means that the judge considers all the factors relevant to the sentence, outlines their significance, and makes a simultaneous decision with regard to all the factors in the case. This has given courts significant flexibility and many law scholars have criticised the difficulty of explaining how courts determine the proper sentence, claiming that ‘there remains something mysterious or magical about how the process of sentencing works’.47
These issues make sentencing considerably subjective and two courts may impose different sentences for the same case. In this regard, the High Court has noted that ‘there is no single correct sentence’.48 That said, it does not mean that courts necessarily impose sentences within an unjustifiable range.
One of the mechanisms that helps courts to keep the sentencing in a justifiable range is through standard non-parole periods. A sentence is divided into parole and non-parole periods. The statutory norms of non-parole and parole period are 3 to 1 respectively (for instance, when the total sentence is 4 years, the normal order for the non-parole period is 3 years and the parole period is 1 year). In Part 4, Div 1A of the Sentencing Act, a table of standard non-parole periods corresponding to the middle range of objective seriousness is provided for various offences. The maximum sentence is also provided in the legislation for each crime, representing the upper limit for cases of the gravest objective seriousness. This helps the courts to achieve consistency of approach in sentencing.49
46 Veen v R (No. 2) (1988) 164 CLR 465 per Mason CJ, Brennan, Dawson and Toohey JJ, p 476 47 I Potas, ‘The Principles of Sentencing Violent Offenders: Towards a More Structured Approach’, in Serious Violent Offenders: Sentencing, Psychiatry and Law Reform, Australian Institute of Criminology, 1991, p. 98. 48 Markarian v R (2005) 228 CLR 357 at [27] (citations are removed) 49 I will further discuss how courts determine the appropriate sentence in Chapter 5, the double-edged sword effect of neuroscience. 72
It is notable that the statutory norm for the ratio of the non-parole to the parole period (that is, 3 to 1) is sometimes modified in cases with special circumstances – the non-parole period is reduced and added to the parole period without changing the total sentence.50 Generally, special circumstances are imposed where:
offenders would be likely to benefit from extended supervision in the community, and require a longer period on parole in order to reintegrate more successfully. Similarly, a shorter non-parole period may be warranted where an offender would benefit from rehabilitation or treatment services in the community.51
Overall, despite the standard non-parole period defined in the legislation which may assist with consistency in sentencing in NSW, the process remains discretionary: the objective seriousness of the crime is subjected to the ultimate determination of the court. Consideration of subjective sentencing factors is a further discretionary matter, and after determination of the total sentence a finding that there are special circumstances may result in changes to the standard non-parole to parole ratio – this depends on how the court views the significance of factors such as the rehabilitation of the offender.
Transparency in sentencing decisions
Despite the discretionary process of sentencing in NSW, it has been noted in a number of statutory provisions52 and in the common law that the law encourages transparency and clear reasoning in sentencing.53 In a decision of the New South Wales Court of Criminal Appeal it was held that: Sentencing judges are under an obligation to give reasons for their decisions. … This is a manifestation of the fundamental principle of the common law that justice must not only be done but must manifestly be seen to be done: It is desirable that sentencing judges … set out their findings in relation to all matters taken into account in mitigation or aggravation of sentence as well as the reasoning which leads to the sentence imposed.54 Procedural fairness also encourages transparency in different ways. It requires the court to inform parties if some agreed statements of fact (or some aspects of them) is not going to be considered in sentencing, for instance, because they are inconsistent with other facts;55 or if the evidence submitted by the offender is uncontested, they will be notified by the court if that evidence is
50 Odgers. 51 NSWLRC, Consultation Paper 6 (2010), note 4 above, at [8.51] - [8.54] (notes are omitted), quoted in Howard and Westmore, p. 546. 52 For example, see S 5(2), s 44(2) and s 45(2) of Crimes (Sentencing Procedure) Act 1999 (NSW), cited in Odgers. 53 See Markarian v The Queen (2005) 228 CLR 357 54 Thomas v R (Commonwealth) (2006) NSWCCA 313 at [16] (citations were removed). 55 See Uzabeaga (2000) NSWCCA 381, 119 A Crim R 452 at [35], cited in Odgers. 73
unlikely to be accepted.56 Despite the encouragement of transparency in sentencing, many courts do not clearly outline all the relevant sentencing factors. They may even orally refer to some sentencing consideration but not include these in the final judgement.
A further issue is that even if a court includes all the relevant sentencing matters in the judgement, this does not mean it is possible to measure the weight given to each of those factors. For instance, with respect to the aggravating and mitigating factors outlined in s 21A of the Sentencing Act, the court is obliged to indicate if a factor is taken into consideration.57 In R v Mills58 the court required the judge to recognise ‘the relevant factors, the weight given to them, and their role’.59 Yet, it is an error for the court to determine exactly how much a sentence would be discounted or increased based on those aggravating and mitigating factors.60
Taking the discussion in this section into account, sentencing is a complicated and technical process that may have different outcomes based on the discretion of the judge. I will return to this point in Chapter 4, an introduction to empirical research, where I discuss the challenges in analysing criminal judgements.
4. Criminal Courts in NSW
Australia has eight State or Territory systems and one federal system: a total of nine legal systems. Each State and Territory have an independent criminal justice system. Nonetheless, there are many similarities between these jurisdictions.
Most criminal matters are heard in State and Territory courts. There are three levels of trial courts in NSW. The lowest level of criminal courts is the Local Court, where the least serious crimes (such as possession of drugs and shoplifting) are dealt with. These courts have no juries and the magistrate determines whether or not the accused is guilty. The District Court is an intermediate court that deals with indictable offences (except murder, piracy and treason). The Supreme Court of New South Wales is the highest trial court where the most serious crimes (including murder and manslaughter) are handled. The highest court for criminal matters is the New South Wales Court of Criminal Appeal, which is part of the Supreme Court of New South Wales. The New
56 Odgers. 57 R v McNamara (2005) NSWCCA 195 at [37]. Also see, BW v R (2007) NSWCCA 236 at [33], [36], cited in Judicial Commission of New South Wales. 58 (2005) 154 A Crim R 40 at [49] 59 Cited in Judicial Commission of New South Wales. Also see Doolan v R (2006) 160 A Crim R 54 60 Judicial Commission of New South Wales. 74
South Wales Court of Criminal Appeal determines appeals from the Supreme Court and District Court.61
Appeals from the New South Wales Court of Criminal Appeal are decided in the High Court of Australia.62 Decisions by the High Court are final, meaning that no further appeal can be granted after a decision of the High Court. An important feature of the High Court is that it interprets the law, and decisions made in this court are binding on all other courts in Australia.63
Conclusion
This Chapter aimed to provide the legal background to the discussions in later Chapters. The first section of this Chapter reviewed the legal system of NSW and the role of judges, juries, and lawyers. It further considered three phases to the criminal procedure: pre-trial, trial, and sentencing. The second section presented an overview of three categories of sentencing factors that are potentially relevant to neuroscience: aims of sentencing, mitigating and aggravating factors, and other factors. The third section of this Chapter went on to explain the challenges to the process by which courts consider different sentencing factors and impose an appropriate sentence. The fourth section outlined the jurisdiction of different criminal courts in NSW.
Having discussed the gap in the literature, aim of this thesis, the neurolaw claims considered in this study, and the legal background, in the next Chapter I will discuss the methodological approach adopted to conduct the empirical research that underlies this thesis.
61 Australian Bureau of Statistics, ‘Australian Social Trends, 1997’, in Australian Bureau of Statistics, 1997,
Chapter 4 An Introduction to Empirical Research
76
Introduction
This study overlaps to some extent with neuroscience, criminology, and criminal law. It aims to examine whether suggestions by proponents of specific neurolaw claims are consistent with how neuroscience is actually being used in courts. The source of the data for this study is a set of criminal cases that involve neuroscientific evidence during the sentencing phase. The research methodology adopted for this thesis – in order to meet the objectives of the research and to be suitable for analysing criminal cases as the source of data – was a ‘classical content analysis’. This analytical approach uses both qualitative and quantitative approaches to analysing data.1 Qualitative content analysis is adopted to interpret textual data based on the contextual circumstances (i.e., understanding the judgement based on the legal rules), and thus gain an in- depth understanding of how neuroscientific evidence is used, and why it is used in particular ways. This analysis also allows the examination of themes and core ideas concerning neurolaw claims in judgements.2 This study also analysed the content of criminal cases to produce quantitative data through a systematic classification process that allowed the analysis of a large dataset of criminal cases in order to evaluate and draw conclusions regarding the neurolaw claims.3
The following Chapter describes the methodology used to conduct this research, explains why this methodology was chosen and assesses its strengths and weaknesses. It begins with a discussion about different sources of data, and why criminal cases are the appropriate data source to meet the objectives of this thesis. It then moves on to discuss how the criminal cases used in this research were collected and sampled. Finally, this Chapter describes how the selected cases were analysed to examine the neurolaw claims considered in this study.
1. Source of data
Since this study is an attempt to evaluate the neurolaw claims in practice, an empirical and data- based method is required. Sources of data that are potentially suitable for examining neurolaw claims are: documentary sources (e.g., court judgements and transcripts that provide the court’s
1 L Webley, ‘Qualitative approaches to empirical legal research’, in The Oxford handbook of empirical legal research, 2010, 926–948. 2 H-F Hsieh & SE Shannon, ‘Three approaches to qualitative content analysis’, in Qualitative health research, vol. 15, 2005, 1277–1288; P Mayring, ‘Qualitative content analysis: theoretical foundation, basic procedures and software solution’, 2014. 3 Webley, 926–948; Y Zhang & BM Wildemuth, ‘Qualitative analysis of content’, in Applications of social research methods to questions in information and library science, vol. 318, 2016. 77
reasons as to why a particular order was decided), interviews, and surveys. I will review these sources of data and their limitations as well as how they may be useful in an examination of the neurolaw claims.
The first source of data to review is data from criminal cases (judgements). Criminal cases are authoritative records of a criminal proceeding that contain the arguments made by the parties, the evidence submitted to the court, the decisions made about that evidence (i.e. conclusions drawn about expert reports), and how different sentencing factors are considered in the sentencing decision by the judge. This source of data can provide valuable evidence-based qualitative data in order to explore themes and patterns in the use of neuroscientific evidence in courts, and to evaluate whether they support the prescriptions of the neurolaw claims.
One of the main limitations in analysing criminal cases is that judges do not always clearly articulate their considerations of neuroscientific evidence when writing or delivering sentencing judgements. For instance, judges may be silent as to how particular evidence has influenced the sentence; or they may explain their conclusion through general reference to multiple pieces of evidence (i.e. psychological and neuroscientific evidence) that provide information about the same matter (e.g., the offender’s mental state), making it difficult to determine how court have considered each piece of evidence in a sentencing decision.
Additionally, judges usually summarise legal arguments and expert reports. This may exclude information that is important for this study from the judgement, including the nature of the neuroscientific evidence used by the expert. Judges may also misinterpret expert opinion and this may have an effect on how neuroscientific evidence is represented during sentencing.
Likewise, in cases where neuroscientific evidence and related arguments are raised during the trial, the judge may summarise the trial without referring to the information that is relevant to this study. For example, CT, MRI and fMRI imaging evidence can be tendered as evidence to the court, and all of these sources of evidence may indicate brain abnormalities associated with impulse control disorder. The judge may then summarise those reports, saying something like: ‘the offender’s brain images suggested that he is more likely to lose control and act impulsively’, rather than referring to which specific types of brain imaging techniques were used. Another example is that the court may summarise the reports by saying: ‘medical examinations suggested the offender has impulse control disorder’.
Another limitation of relying on criminal judgements is that many of the factors that can influence judges’ decisions are invisible to the researcher, such as the influence of the media on a
78
judgement,4 unconscious factors5 such as the race, gender and age of the accused/offender,6 and idiosyncratic approaches to the philosophy of penology.7
Overall, using criminal cases as the source of analysis for this study means that the insight provided into the use of neuroscientific evidence, and the assessment of the neurolaw claims, is limited to what judges include in their sentencing judgements.8
One approach that may help address this limitation is to review case files (records) or court transcripts. However, both approaches have their own particular disadvantages, such as difficulties in accessing a large number of case files or court transcripts as access to this information is restricted for third parties – even for research purposes.9
Another source of data would be to conduct interviews with judges and lawyers. Conducting interviews is better than using case judgements in the sense that the interview subjects are directly involved in the process of evaluating the relevant evidence and other material which may provide information that is not expressed in court records, and may also allow for a freer expression of ideas without concerns about political or public reaction.10
4 See R White, F Haines & N Asquith, Crime and criminology, 5th ed., Oxford University Press, 2012. 5 See, J Goodman-Delahunty & SL Sporer, ‘Unconscious influences in sentencing decisions: a research review of psychological sources of disparity’, in Australian Journal of Forensic Sciences, vol. 42, 2010, 19–36. 6 For instance, prisoner’s parole decision can be influenced by if the court procedure is right before lunch or after it. D Eagleman, The Brain: The Story of You, in Canongate, Pantheon, 2015. Also see, D Steffensmeier & S Demuth, ‘Does gender modify the effects of race–ethnicity on criminal sanctioning? Sentences for male and female white, black, and Hispanic defendants’, in Journal of Quantitative Criminology, vol. 22, 2006, 241–261; JK Doerner & S Demuth, ‘The independent and joint effects of race/ethnicity, gender, and age on sentencing outcomes in US federal courts’, in Justice Quarterly, vol. 27, 2010, 1–27; C Spohn, ‘Thirty years of sentencing reform: The quest for a racially neutral sentencing process’, in Criminal justice, vol. 3, 2000, 427–501; D Steffensmeier & S Demuth, ‘Ethnicity and sentencing outcomes in US federal courts: Who is punished more harshly?’, in American sociological review, 2000, 705–729; CA Franklin & NE Fearn, ‘Gender, race, and formal court decision-making outcomes: Chivalry/paternalism, conflict theory or gender conflict?’, in Journal of Criminal Justice, vol. 36, 2008, 279–290; MJ Leiber & KC Fox, ‘Race and the impact of detention on juvenile justice decision making’, in NCCD news, vol. 51, 2005, 470–497. 7 J Hogarth, Sentencing as a Human Process, Toronto, University of Toronto Press, 1971; A McCay, ‘Behavioural Genetics, Moral Agency and Retributive Sentencing: The Case for Mitigation (Doctoral dissertation)’, Sydney University, 2013. 8 H Al-Alosi, ‘Fantasy Crime’, University of New South Wales, 2016. 9 For instance see, Australian Law Reform Commission, ‘Access to court and tribunal records | ALRC’,
However, this approach also has several limitations. Access to judges and lawyers11 and ethical considerations12 in interviewing them pose difficulties that cannot be ignored. Further, interviewing judges and lawyers may not be helpful in examining some of the neurolaw claims considered in this study, primarily because of issues with memory and selective recall. For instance, judges and lawyers may be able to discuss whether neuroscience is used in courts to promote rehabilitation of offenders, however it is unlikely they would be able to recall whether brain imaging evidence is generally corroborated by other forms of evidence when proving a fact about an individual’s mental state (i.e., the convergence of evidence claim).
The other potential source of data is information obtained by conducting online surveys. An advantage of this approach is that it is be argued that unlike interviews, judges and lawyers may spend more time on a survey and may be able to answer questions about the convergence of evidence claim. Nonetheless, easier to access a large number of lawyers and judges. That said, it is unlikely that a sufficient number of judges and lawyers (i.e. a representative sample) would respond to the survey. Moreover, surveys – similarly to interviews – pose problems relating to the memory and recall of the participants. It may when it comes to the double-edged sword claim, it is unlikely they would be able to recall how frequently neuroscientific evidence (or specific types of neuroscientific evidence such as brain images) supports a more lenient or a harsher punishment.
At this point, it is useful to describe the sources of data that were used in empirical neurolaw studies and their research aims, as they may provide helpful information for this study in determining which source of data is most suitable. The aims of empirical neurolaw studies were mostly general, including ‘to gain more insight into the relationship between neuroscience and criminal law’13 and to examine ‘the extent of the use of neuroscientific evidence by those accused of criminal offenses’.14 There is only one study that has a more central aim, that is, to examine the double-edged sword claim.15 Interestingly, all of these studies used criminal cases (judgements) in conducting their research.
11 S Halliday & P Schmidt, Conducting law and society research: reflections on methods and practices, Cambridge University Press, 2009; D Richards, ‘Elite interviewing: Approaches and pitfalls’, in Politics, vol. 16, 1996, 199–204 Cited in; Al-Alosi. 12 Obtaining ethical approval 13 See JA Chandler, ‘The use of neuroscientific evidence in Canadian criminal proceedings’, in Journal of Law and the Biosciences, vol. 2, 2016, 550–579. 14 P Catley & L Claydon, ‘The use of neuroscientific evidence in the courtroom by those accused of criminal offenses in England and Wales’, in Journal of Law and the Biosciences, vol. 2, 2015, 510–549. 15 DW Denno, ‘The Myth of the Double-Edged Sword: An Empirical Study of Neuroscience Evidence in Criminal Cases’, in Boston College Law Review, vol. 56, 2015, 493–551. 80
Overall, taking into consideration the strengths and weaknesses of these different sources of data, this study, like other empirical studies, has used criminal cases as its source of data. An initial justification for this choice is the accessibility of criminal cases. This study aims to construct a relatively large dataset to provide a relatively credible evaluation of the neurolaw claims.16 In relation to interviews and surveys, approaching lawyers and judges who are appropriate for the purpose of this study and ensuring that the number of participants reaches a sufficient sample size may not be feasible. Similarly, as mentioned earlier, access to case files and court transcripts is a significant constraint. In contrast, legal cases are much more accessible, and a sample search (test) of legal databases demonstrated that there are many criminal cases available for inclusion in this study.
Unlike interview and surveys, criminal cases appear to be a better fit to examine all the neurolaw claims considered in this study – the research objective. Nonetheless, as I discussed above, this does not mean that criminal cases can provide a complete picture of the use of neuroscience with respect to the neurolaw claims, as judgements have their limitations and are not outlined and published based on these claims.
Although judgements are limited to published cases on databases, it is these published cases that are referenced in future cases within the common law system. As such, while judgements may not offer a complete picture of the use of neuroscience in courts, they have the advantage of providing cases that may, in the future, be referred to as precedent and of common procedures relating to the use of neuroscience in the criminal justice system.17
Another benefit of using judgements as the source of data is that it parallels other empirical neurolaw studies. Thus, it is possible to compare the results of this study with other jurisdictions and gain a deeper insight into the use of neuroscience around the world. In addition, empirical neurolaw studies provide beneficial information such as different aspects of the use of neuroscience in courts and methodological limitations that were not previously considered in this study.
Overall, considering all the sources of data outlined above, criminal cases, while not without limitations, are a more appropriate source of data for the purpose of this study.
16 I will elaborate on why this research is a large-scale study later in this section 17 See, Chandler, 550–579. 81
2. Sampling and data collection
The first step in sampling the data was to choose appropriate ‘search terms’ in order to produce criminal cases in which neuroscientific evidence was referred to. Therefore, a definition of neuroscientific evidence needed to be developed for this phase.
In Chapter 2 (literature review and background), I explained that scholars have discussed the difficulties posed by the artificial distinction between different forms of evidence that determines what constitutes ‘neuroscientific evidence’ in the courts. I noted that there are two forms of (diagnostic) neuroscientific evidence. First, imaging techniques, which are arguably divided into two categories. Those that generate images of the brain such as MRI, CT, and fMRI techniques; and those that generate brainwave forms of brain activity such as EEG and MEG. Second, non- imaging evidence, i.e. evidence that has been produced without the aid of imaging techniques, such as neuropsychological testings. I also discussed that based on what laypeople may consider neuroscientific evidence to be, non-imaging techniques may be divided into two sub-groups: neuro-tool techniques (such as neuropsychology and neuropsychiatric tests); and non-neuro-tool evidence (such as a historical records of head injury as the result of an accident, or a report by psychologist suggesting the possibility of brain damage as a consequence of long-term alcohol abuse).
For the purpose of this study, I took a more cautious approach to the definition of neuroscientific evidence. That is, I considered all forms of techniques discussed above to be neuroscientific evidence. I did this for two main reasons. First, this approach enables the study to analyse its results based on each category of evidence and discuss whether different ways of defining neuroscientific evidence result in different conclusions. Second, by considering all three forms of evidence, this study can satisfy scholars who define neuroscientific evidence differently. Nonetheless, since it appears that most scholars agree on imaging techniques (both ‘imaging of the brain’ and ‘brainwave forms of the brain activity’) to be neuroscientific evidence,18 and the majority of scholarly discussions in the neurolaw literature are centred on imaging evidence, this study primarily focuses on this form of evidence (the different types of neuroscientific evidence are shown in Figure 4-1).
18 See, Catley and Claydon, 510–549. 82
Figure 4-1: Different types of neuroscientific evidence.
After listing some search terms based on this broad understanding of neuroscientific evidence, I used empirical neurolaw studies and the ‘Australian Neurolaw Database’ project ‘keywords’ list19 to supplement my search terms. The main reason for this is their experience in identifying relevant cases. In these studies, occasionally, they updated their search term list to find more relevant cases.20 A simple example of this is a search conducted for cases containing the phrase ‘MRI scan’. Using ‘MRI’ (scan) as a target term may ignore cases where this neuroscientific evidence is referred to as ‘Magnetic Resonance Imaging’. 21 Key terms (including the stem words and synonyms) were outlined using this approach, and Boolean operators were used to refine the search.22 For instance, to find criminal cases where a reference was made to MRI, I searched ‘criminal law AND MRI’.
In Table 4-1, I have listed the key terms used to find cases. The search was much more comprehensive than this short list indicates: for instance, by searching ‘brain’, all the combinations with ‘brain’ including ‘brain injury’, ‘brain abnormality’, and ‘traumatic brain injury’ were located. I also used a variety of approaches to make sure that all of the cases that could possibly be related to this study would be found. One of these search functions is adding an asterisk after a word. For example, searching ‘neuro*’ will turn up all words beginning with neuro such as ‘neuroscience’, ‘neuropsychology’, ‘neurology’, and ‘neuroimaging’. Table 4-2 sets out a list of search terms that were produced by using these search functions.
Yet, as different brain imaging and tests in cases are mentioned in different forms (i.e. EEG and Electroencephalography), I had to list all the different ways of referring to these tests and
19 See https://neurolaw.edu.au/keywords 20 See, Catley and Claydon, 510–549. 21 Catley and Claydon, 510–549 (under Table 5). 22 Unfortunately, one of the databases I used, Caselaw, does not support sophisticated Boolean operations. 83
technologies separately. This can be seen in Table 4-3.
Table 4-1: Main key terms.
Magneto Ceretec Brain Halstead Bender- Encephalo Perfusion EEG Reitan Neuro∗ SPECT Scan Gestalt Test Graphy Study Battery (MEG)
Electrical Brain* Brainwave fMRI Hypox* MMPI PET Scan WAIS-R Activity
Wide Range Brixton Test CT Scan Fronto* Haematoma* MRI Scan qEEG Achievement Test
Diffusion General Tensor Adaptive Rorschach Cerebr* Ischaemi* NIRS X-RAY Scan Imaging Behaviour Inkblot Test (DTI) Testing
Table 4-2: Some examples of terms that were produced by *.
Neuropsychological Abnormal Brain Brain Impairment Closed Head Injury Frontemporal (test or examination)
Acquired Brain Congenital Brain Fronto-Temporal Neuropsychological Brain Infection Injury (ABI) Disorder Degeneration Assessment
Alcohol-Induced Fronto-Temporal Neuropsychological Brain Injury FASD Brain Damage Dementia Impairment
Alcohol-Related Frontal Assessment Fronto-Temporal Neuropsychological Brain Lesion Brain Damage Battery Lobe Test
Bilateral Frontal Frontotemporal Brain Mapping Frontal Atrophy Neuropsychology Lobe Abnormality Dementia
Bilateral Frontal Brain Scan Frontal Damage Haemorrhage Neuroscience Lobe Abnormality
Frontal Organic Brain Biological Brain Scanning Dysexecutive Head Injury (Damage or Injury) Syndrome
Frontal Lobe Biology Brain Swelling Hypoxia Parasomnia Abnormality
Frontal Lobe Parietal Birth Trauma Brain Syndrome Hypoxic Damage Haemorrhage
Frontal Lobe Ischaemia (or Blow Brain Brain Trauma Parietal Lobe Dysfunction Ischemia)
Brain Tumour (or Frontal Lobe Post Concussion Brain Abnormality Minor Head Injury Brain Tumor) Epilepsy Syndrome
Temporal Lobe Brain Activity Cerebral Damage Neuro-Disease Post Concussive Abnormality
Frontal Lobe Neurobiological Post-Offence Brain Brain Damage Cerebral Impairment Epilepsy (evidence) Injury
Frontal Lobe Severe Closed Head Brain Damage Cerebral Insult Neurofibromatosis Abnormality Injury
Frontal Lobe Subdural Brain Disorder Cerebral Lacunar Neuroimaging Dysfunction Haematoma
84
Cerebral Oedeama Neurological (tests Temporal (control or Brain Dysfunction Parietal Lobe (or Oedema) or examination) function)
Cerebro-Vascular Fronto-Temporal Brain Function Neurological Deficit Temporal Lobe Incident Dementia
Frontal Lobe Temporal Lobe Brain Image Cerebrovascular Neurologist Impairment Abnormality
Cerebrovascular Frontal Lobe Traumatic Brain Brain Imaging Neurology Disease Syndrome Injury (TBI)
Table 4-3: Different forms of referring to various tests.
Diffusion Tensor General Adaptive BEAM Perfusion Scan SPECT Scan Imaging Behaviour Testing
Halstead Reitan Bender-Gestalt Test DTI PET Scan Tomography Battery
Brain Electrical Magnetic Resonance EEG pet-scan WAIS Activity Mapping Imaging
Electrical brainwave Magnetoencephalogr Positron Emission Brixton Test WAIS-R activity aphy Tomography
Electro- Wechsler Adult CAT Scan MEG qEEG encephalography Intelligence Scale
Quantitative Wechsler Adult Cerebral Perfusion Electroencephalogra MMPI electroencephalogra Intelligence Scale- Study phy m Revised
Quantitative Ceretec Brain Electroencephalogra Wide Range MRI electroencephalograp Perfusion Study phy Achievement Test hy
Computed Near-infrared Rorschach Inkblot Wisconsin Card fMRI tomography spectroscopy Test Sorting Test
Functional Magnetic CT Scan NIRS S.P.E.C.T. Scan X-radiation Resonance Imaging
Single Photon ct-scan Functional MRI P.E.T. Emission Computed X-RAY Tomography
The search terms were used across three databases. The Australian Neurolaw Database23 (a database which contains only cases involving neuroscientific evidence), AustLII24 and CaseLaw.25 The searches also targeted the five levels of court most likely to hear criminal cases in which the mental state of offender is relevant and neuroscientific evidence would be discussed: the High Court of Australia (only those cases that involved appeals from NSW courts), the Court
23 https://neurolaw.edu.au 24 http://www.austlii.edu.au 25 https://www.caselaw.nsw.gov.au 85
of Criminal Appeal, the Supreme Court, the District Court, and the Local Court.26 The searches were conducted for judgements issued before 31 December 2016.
This study is confined to the criminal courts of New South Wales, Australia. In Chapter 3 (legal background), I discussed that each State and Territory in Australia has an independent criminal justice system and legal rules are occasionally different. Conducting research that involves all these states requires inter-jurisdictional assessment (examining criminal cases and systems in all the states) and poses the difficulty of dealing with legal inconsistencies. Further, since this study aims to provide a large-scale dataset of the use of neuroscience in courts, including all states of Australia would require the assessment of a larger number of cases than is feasible for this project or at best it would not allow for an in-depth analysis of the cases. As such, although restricting this study may have resulted in the loss of some degree of generalisability of the use of neuroscience in Australia, it provides a large-scale and credible account of the use of neuroscience in NSW courts. Nonetheless, this does not mean that the study is not relevant to other states in Australia and some other common law countries, as there are many similar legal principles, rules and practices across these jurisdictions.27
As a result of the searches, 4242 cases were located on CaseLaw, 2904 cases were found on AustLII, and 109 criminal cases were identified through the Australian Neurolaw Database.28 These cases were all reviewed and are on file with the author. Overall, the total number of cases reviewed was 7255 cases.
The main reason that CaseLaw generated about 1300 more cases than AustLII was the problem with its search engine. The engine does not support the ‘NOT’ searching operator. As such, it was not possible to exclude cases that contained specific terms (e.g. Civil) from the search result, and consequently many cases that were not part of the intended target group were identified.29 Also, in contrast to AustLII, CaseLaw did not allow the use of complex search queries. For AustLII, I used a single search with a complex search query, while for CaseLaw, I had to search each search term separately. As such, when searching for cases containing ‘CT scan’ and ‘brain damage’ separately, cases that contained both ‘CT scan’ and ‘brain damage’ were produced twice.
26 Other courts which are not targeted for this research include the Children’s Court, the Compensation Court, the Court of Appeal, the Drug Court, the Industrial Court, the Industrial Relations Commission (Commissioners), the Industrial Relations Commission (Judges), the Land and Environment Court (Commissioners), and the Land and Environment Court (Judges). 27 See, A Alimardani & J Chin, ‘Neurolaw in Australia: The Use of Neuroscience in Australian Criminal Proceedings’, in Neuroethics, 2019; McCay. 28 Some of these cases are not published yet and the public do not have access to those cases. 29 There are also some issues with ‘OR’ and ‘AND’ Caselaw search operators which made it even harder to narrow down the search result to the relevant cases. 86
AustLII’s search engine also has features that facilitate more efficient searches. One simple example of these features is a proximity operator: by searching ‘Frontal/2/lobe’, only cases where ‘frontal’ and ‘lobe’ appear within two words of each other were produced. Accordingly, searching ‘Frontal/2/lobe’ prevented the search from missing cases with relevant key terms such as ‘frontal parietal lobe’. The proximity operator also helped to exclude cases where irrelevant combinations existed in the text. For example, by searching PET/5/scan, cases that used ‘pet’ to refer to a ‘domestic animal’ were avoided.
Of the 7255 collected cases, a considerable number were excluded from the study for a variety of reasons. This included that the neuroscientific techniques in question were used to scan the body of the offender rather than their brain; the cases concerned pre-trial and trial phases including those discussing fitness to stand trial, the defence of mental illness, and the defence of substantial impairment;30 the search generated many Civil law cases;31 many duplicate cases had been collected due to the use of multiple databases; that some of the search terms were used to refer to an offender’s relative with a health condition (e.g., an offender’s father brain damage); there was an overlap between the name of a form of neuroscientific evidence and a name in the case (e.g. for ‘MEG’, CaseLaw produced cases where the name ‘Megan’ was mentioned in the text). Finally, cases where the neuroscientific evidence was related to the victim or deceased were excluded because the neurolaw claims considered in this study are only concerned with brain impairment in relation to the sentencing of an offender.
Following these exclusions, 494 cases remained. This number was much larger than the expected number of cases involving neuroscientific evidence in NSW courts, and analysing all the cases was not practical. Accordingly, I decided to sample some cases from this collection (within-case sampling).
First, since the main focus of this study is on brain imaging techniques, all cases that involved brain imaging evidence were selected: that is, 61 cases. In some of these cases, neuro-tool and non-neuro-tool evidence were also introduced to the court. After screening the rest of the cases, it turned out that there were 92 cases involving neuro-tool evidence. Again, there were some cases where non-neuro-tool evidence was also introduced. At this point, 153 cases from 495 cases were selected. That means there were 342 cases that exclusively contained non-neuro-tool evidence.
30 Where the accused is found unfit to stand trial, the trial is held without a jury (judge alone) and the judgment is published. 31 In such cases the term ‘criminal law’ was mentioned in the text, or these cases were generated as a weakness of CaseLaw search engine. 87
Of the final total of 153 selected cases, there were 147 cases that involved non-neuro-tool evidence. However, there was no selected case that exclusively discusses non-neuro-tool evidence (i.e., cases in which no imaging or neuro-tool evidence were introduced). This is because cases had been selected because they contained imaging or neuro-tool evidence. To allow the possibility of comparing cases that exclusively involved non-neuro-tool evidence with other cases (a control group), it was decided to add some cases from the remaining 342 cases in which only non-neuro- tool evidence was introduced. In order to ensure this group of cases was comparable with other groups of cases, in particular cases that contained imaging evidence,32 61 cases, equal to the number of cases involving brain imaging evidence, were randomly selected and added to the dataset. Cases were listed based on the year of judgement and were randomly selected (every fifth and sixth case). Accordingly, the total number of selected cases for the final data set of this study is 214 cases. A summary of case sampling is shown in Figure 4-2 below.
Figure 4-2: Summary of case sampling.
32 As noted earlier, neuroscientific evidence is considered by the majority of scholars to involve imaging techniques and neurolaw literature is specifically focused on this type of evidence. 88
While analysing cases, it became apparent that in some cases containing neuro-tool evidence there was more than one offender who was the subject of neuroscientific examination. As a result, the analytical approach was changed from case-based to offender-based; and instead of considering the results by reference to 214 cases, the research analysed the 221 offenders in the sample. Nonetheless, in order to prevent any confusion, I will use the term ‘case’ instead of ‘offender’ (i.e. 221 cases).
In Table 4-4, I have outlined the number of cases in which there was at least one example of imaging evidence, neuro-tool evidence, and non-neuro-tool evidence.
Table 4-4: Number of cases based on evidence type.
Limitations of sampling and data collection
To begin with, while the intention was to find all criminal cases in which neuroscientific evidence was introduced to the court, not all cases are transcribed or uploaded to electronic databases. This means that some cases of relevance to this study were impossible to identify and analyse.
Another related limitation is that the databases provide an incomplete collection of cases. For instance, the AustLII database does not contain any Local Court decisions,33 and only provides three decisions of Supreme Court of New South Wales delivered prior to 199534 (see footnote for AustLII and CaseLaw case collection of different courts).35
33 See http://www.austlii.edu.au/databases.html 34 Only selected decisions of the Court are available. See, http://www.austlii.edu.au/au/cases/nsw/NSWSC/ 35 AustLII provides decisions for the following courts in these specific periods: The Supreme Court of New South Wales: 1995-present (three cases are available before 1995). Only selected decisions of the Court are available. See, http://www7.austlii.edu.au/cgi- bin/viewdb/au/cases/nsw/NSWSC/ The Supreme Court of New South Wales, Court of Criminal Appeal: 1999 – present (five cases are available before 1999). Only selected decisions of the Court are available. See, http://www6.austlii.edu.au/cgi-bin/viewdb/au/cases/nsw/NSWCCA/ The District Court of New South Wales: only selected decisions of the Court are available. According to the AustLII website, ‘[b]efore 2004, these decisions were exclusively appeals made by pharmacists against the disciplinary findings and/or orders made by the Pharmacy Board of New South Wales. The headnote to each appeal decision has a link back to the relevant Pharmacy Board decision.’ There are only 15 cases of this nature are available on this database before 2005. http://www.austlii.edu.au/au/cases/nsw/NSWDC/ Caselaw provides decisions for the following courts in these specific periods: The Court of Criminal Appeal: period of case collection: 1999 – present 89
With regard to the criteria for publishing a decision, AustLII notes that it ‘does not choose which cases will be reported. For some Courts/Tribunals AustLII holds every case decided, for others we hold significant decisions (as selected by the Court), and in others we hold only reported cases’.36 On CaseLaw, it is explained that the publication of decisions is not compulsory in some courts and tribunals, and only selected decisions are published.37 In relation to the Local Court, CaseLaw notes that only a small number of decisions ‘that provide interpretations of legislation and legal principles relevant to criminal, Civil and other matters’ are published.38 In the District Court, the publication of a decision is left to the discretion of the judge. As such, a decision may not be published for a variety of reasons, including the private nature of the case.39
The information about how courts select and provide these specific cases for publication on databases is limited and inadequate. Subject to the limitations of the cases selected for inclusion in the databases, there are few reasons to believe these cases are not representative or systematically skewed.
While searching through different databases increases the chance of finding published cases, the number of relevant but unpublished cases cannot be determined, and this means that making definitive conclusions as to how neuroscientific evidence is used in NSW courts is not possible. That said, the number of collected cases in the final dataset is capable of generating relatively reliable conclusions that create a useful picture of how neuroscientific evidence is used in courts.
Since the neurolaw claims are examined based on collected judgements, before I move on to discuss how I analysed the cases, it is useful to provide some general information on the distribution of cases based on the level of the court, year of judgement, and prevalence of different types of neuroscientific evidence.
Preliminary analysis of dataset (case samples)
The District Court: period of case collection: 2005 - present (incomplete collection) The Local Court: period of case collection: 2002 - present (incomplete collection) The Supreme Court: period of case collection: 1999 – present See, https://www.caselaw.nsw.gov.au/about 36 http://www.austlii.edu.au/faq.html#q1.9 37 https://www.caselaw.nsw.gov.au/about 38 https://www.caselaw.nsw.gov.au/policy#release 39 However, it is not clear why this may happen in sentencing. See, https://www.caselaw.nsw.gov.au/policy#release 90
Figure 4-3 reveals that neuroscientific evidence is most commonly used in Supreme Court cases, whilst this evidence is least likely to be introduced in Local Court cases. This result may suggest that the use of neuroscientific evidence is associated with the severity of the crime/punishment.
Figure 4-3: Distribution of cases based on court level.
With respect to the distribution of cases based on the year of judgement, there are two important matters to be considered: first, due to the fact that not all judgements are electronically published on legal databases, the figures do not represent the actual number of judgements made each year. This issue also skews the examination of the use of neuroscience over time. Second, cases that contain non-neuro-tool evidence are randomly selected from a bigger pool of cases and therefore these figures do not represent the actual number of cases in the collected dataset of this study. Bearing these issues in mind, the result of the comparison between the three types of neuroscientific evidence (imaging, neuro-tool, and non-neuro-tool evidence) is presented in Table 4-5 and Figure 4-4.
These three categories of cases are not mutually exclusive. By this, I mean that since categories are based on the availability of at least one form of neuroscientific evidence in the cases — imaging, neuro- tool and non-neuro tool — and some cases involved the introduction of more than one type of neuroscientific evidence, some cases are represented in more than one category.
The majority of cases in the dataset of this study are dated between 1997 and 2016. There are only two cases, Veen v R [1979]40 and Veen v R [1988],41 that was decided before 1997. These cases are currently understood to be the earliest published cases relevant to neuroscience in
40 HCA 7; 143 CLR 458 (28 February 1979) 41 HCA 14; 164 CLR 465 (29 March 1988) 91
Australia, and both were appealed to High Court of Australia from the New South Wealth Court of Criminal Appeal (in the Supreme Court of NSW). I excluded these two cases from Figures below since their decision dates would result in a gap that is due to the incomplete collection of cases on the databases utilised.
Table 4-5: Distribution of cases based on the year of judgement by three types of evidence.
92
Figure 4-4: Trends in the introduction of neuroscientific evidence from 1997 to 2016 by three types of evidence.
I will provide further information about cases containing imaging evidence because, as I previously discussed, it appears that most scholars find imaging techniques to be neuroscientific evidence and the majority of neurolaw discourse focuses on this type of evidence.
Figure 4-5 illustrates that structural imaging techniques are much more commonly raised in courts than functional neuroimaging techniques (the categories are not mutually exclusive)
Figure 4-5: Frequency of the use of structural and functional neuroimaging evidence in 61 cases.
A breakdown of structural and functional neuroimaging in Figure 4-6 (below) also indicates that two structural imaging techniques, MRI and CT scanning, are the most commonly used types of imaging techniques in courts. Among functional imaging techniques, EEG testing was used in seven cases while MEG and fMRI were not raised in any of the cases.
93
Figure 4-6: Frequency of different types of imaging techniques.
*Cerebral Perfusion Study
3. Data analysis
As discussed earlier, each neurolaw claim is focused upon a different proposition. The first step in analysing the judgements is to unpack the neurolaw claims and understand what information (themes) in the cases can assist in evaluating each claim. The next step is to identify and record different parts of the judgements (text) that are relevant to each neurolaw claim. The final step is to evaluate each neurolaw claim using the information extracted from the judgements.
Consider, for example, the double-edged sword claim. This claim states that neuroscientific evidence may either mitigate or aggravate the sentence. Four possible themes emerge from the case analysis: neuroscientific evidence tends towards aggravation of the sentence (theme 1), mitigation of the sentence (theme 2), both aggravation and mitigation (theme 3), or neither (theme 4) (see Figure 4-7).
94
Figure 4-7: Breaking down the double-edged sword claim to themes.
Neuro evidence tends towards aggravation Theme 1 of the sentence
Neuro evidence tends towards mitigation Theme 2 of the sentence
Neuro evidence tends towards both Theme 3 mitigaton and aggravaton of the sentence edged Sword (DES) claim - Neuro evidence tends towards neither Theme 4 mitigaton nor aggravation of the sentence Double
The judgements themselves do not discuss whether neuroscientific evidence has contributed to aggravation or mitigation of the sentence. Instead, the discussion involves a list (category) of sentencing factors that may aggravate or mitigate the sentence. For instance, as discussed in Chapter 3 (legal background), one of the aims of sentencing is ‘to protect the community from the offender’.42 As such, if the evidence suggests that an offender’s brain abnormality increases their risk of recidivism, the court may give greater consideration to that aim of sentencing and increases the sentence. Accordingly, in this case, neuroscientific evidence is connected to theme 1 – neuroscientific evidence tends towards aggravation of the sentence.
Next, I created a list of ‘key factors’ for each theme. For instance, for theme 1, I outlined a list of aggravating sentencing factors (theme 1 key factors). Similarly, for theme 2 (neuroscientific evidence tends towards mitigation of the sentence), I created a list of mitigating factors (theme 2 key factors) (see Figure 4-8).43
42 s 3A(c) of Crimes (Sentencing Procedure) Act 1999 43 Please note that themes, key factors, and the connection between them are sophisticated maters and examples are simplified in this section. 95
Figure 4-8: A neurolaw claim is broken down to themes and then themes are broken down into a list of key factors. If one of the key factors is identified in the case, that part of the text is labelled (coded) with that key factor.
As I reviewed a case, when I identified a sentencing factor associated with neuroscientific evidence, I linked it to (labelled it as) the corresponding key factor in the theme 1 and 2 key factor lists (e.g. risk of future offending). Each label that represents a sentencing factor is called ‘code’ and labelling different parts of the text that links a code to a ‘key factor’ is called ‘coding’44 (see Figure 4-9).
Figure 4-9: Part of a case that is coded as ‘risk of future offending’. This code is linked to a list of key factors titled ‘aggravating factors’. This means the requirement of theme 1 is met in this case (this paragraph is copied from R v McCann [2012] NSWSC 1462).
In coding the cases, determining if the content of the case is related to one of the neurolaw claims and how that content fits into the ‘key factors’ requires qualitative interpretation.45 To do this, I used qualitative content analysis. In qualitative content analysis, the researcher analyses the text (case materials) ‘with attention to the content or contextual meaning of the text’.46 For this research, it means I interpreted judges’ explanations based on the legal rules of NSW and the circumstances of the case. This analytical approach is suitable for analysing the judgements for different reasons. For instance, courts sometimes use different expressions in explaining how particular evidence is relevant to sentencing. A simple example to illustrate this issue is that instead of directly referring to the need ‘to protect the community from the offender’ as an aim
44 See, Webley, 926–948 (p. 941). 45 Webley, 926–948. 46 Mayring. 96
of sentencing,47 the court may provide any number of different expressions: e.g., the offender’s brain impairment may result in future offending; may increase the risk of recidivism; may require control to keep society safe; may make him/her a danger to the society; or may mean that he or she is violent once released. Connecting evidence of the brain impairment to the relevant aim of sentencing (in this case, the protection of the community from the offender) requires careful analysis of the content, as well as knowledge of how NSW sentencing rules are represented in the case, and – on that basis – how a brain abnormality may be associated with different sentencing factors. Further, where the judge did not clearly explain whether he or she has accepted expert evidence regarding the offender’s brain injury, I reviewed other parts of the case in order to find out whether there was any other reference to the brain injury. I identified some cases where the judge did not comment on the expert’s evaluation of the offender’s brain injury; however, in sentencing remarks they referred to the brain injury and how it contributed to the sentencing decision.
In circumstances where the content did not provide sufficient information as to whether neuroscientific evidence was connected to one of the theme 1 or 2 key factors, I labelled it as ‘unclear’.
To illustrate this issue, I refer to one of the cases in the dataset of this study where the judge was unclear about the influence of neuroscientific evidence on the sentence. In Chapter 3 (legal background), I explained that where a mental issue is causally connected with the crime, the court may reduce the consideration given to the moral culpability of the offender. In R v Sankey,48 the applicant sought leave to appeal against the severity of the sentence for drug and sexual offences. He had a history of severe closed head injury that led to behavioural and cognitive changes. His medical records confirmed the existence of frontal lobe impairment and cognitive deficits. Expert witnesses in the trial court also reported issues with appellant’s rational decision making and the ability to monitor his behaviour. While it seems that the court found a causal relationship between his medical issues and the offence, it did not discuss whether the appellant’s brain impairment had reduced his moral culpability.49 In such cases, the relationship between neuroscientific evidence and moral culpability as a sentencing factors (in this case as a key factor under theme 2 — mitigating the sentence) is unclear.
Another example where content analysis does not help to interpret the court’s vague expression
47 s 3A(c) of Crimes (Sentencing Procedure) Act 1999 48 [2002] NSWCCA 135 49 The court discussed the significance of brain injury to general and specific deterrence. However, that is not relevant in this discussion. 97
is in Iglesias v R.50 In this case the appeal court, referring to the trial court, explains that ‘[h]is Honour accepted [the expert’s] opinions that the applicant’s brain had been damaged by substance abuse and that he had borderline intellectual capacity which contributed to his violent behaviour.’51 This statement can be analysed in different ways. First, that the applicant’s violent behaviour is due to his intellectual capacity which was reduced by brain damage. Or second, that the applicant’s intellectual capacity is an independent factor to the violent behaviour and it is unclear whether his brain damage had anything with violent behaviour. In this case, the connection between the offender’s brain damage and his criminality is unclear and the content of the judgement does not provide any further information to address this issue. As such, I labelled this case as ‘unclear’.
The issues in the above examples indicate how much coding relies on the researcher’s judgement – it requires the researcher to decide how to interpret unclear matters. As such, it is possible that two researchers may take different approaches to the interpretation of a matter in the case, and this may lead to different conclusions. Unlike this project, some other empirical neurolaw studies had multiple coders that provided an inter-rater reliability approach (in which different coders reassess the cases; for instance, with two coders per case), to increase the reliability of the research outcome. This is one of the limitations of this study and one of the drawbacks of coding.
The coding process for a large dataset requires systematic, accurate and case-by-case text analysis. A research platform called Qualtrics52 was used to facilitate this process. Qualtrics functions similarly to survey software, and the researcher can list a set of questions and then analyse the results of the survey for a large number of participants. In this study, I converted key factors to questions. For instance, for the key factor one in Figure 4-9 (above), I generated the following question: ‘does neuroscientific evidence contribute to the finding of future risk of re- offending?’. Once I had read and coded the judgement, I answered the questions listed on Qualtrics and if I had labelled any part of the text as ‘risk of future offending’ (see Figure 4-10 below), then the answer to the Qualtrics question would be ‘yes’.
50 [2006] NSWCCA 261 51 Id. at [17] 52 https://www.qualtrics.com/au/ 98
Figure 4-10: A question that represents a key factor of double-edged sword claim in Qualtrics.
There are different computer programs/platforms for recording the results of coding, such as Excel and Nvivo. However, some features of Qualtrics made it a preferable program for this study. For instance, Qualtrics allows the researcher to design conditional questions: e.g., for the double- edged sword claim, I created a condition where if a question corresponding to one of the aggravating key factors was answered in the affirmative, (e.g. in Figure 4-10 above, question 90 about risk of future offending), then Qualtrics automatically connected it to the double-edged sword claim theme 1: neuroscientific evidence tends towards a harsher punishment; removed the irrelevant choices; and left only the correct choice (see Figure 4-11). Accordingly, Qualtrics made the coding process swifter and reduced the potential for subjective errors by the researcher. There are also other features of Qualtrics that make it possible to design a sophisticated coding system.53
Figure 4-11: Conditional questioning removes irrelevant choices and leaves the correct answer in that case.
I should note that although the computer program facilitates the coding, it does not mean that it analyses the text and the results of the study. It is the researcher who undertakes the qualitative interpretation of the text, develops the codes, and makes sense of the results in accordance with the research objectives.54
By entering all the key factors produced for this study into Qualtrics, the first coding platform was created. Before coding all the cases, I decided to code 10 cases as a pilot to examine whether
53 Qualtrics is available for free for UNSW students 54 Webley, 926–948; Zhang and Wildemuth. 99
key factors in the coding platform covered all aspects of the neurolaw claims and identify whether there were any errors. Following the pilot reading, I identified many errors and matters that I did not initially consider. Accordingly, in light of the pilot coding, some key factors were modified or removed, and some new key factors were added or replaced. For instance, I had not considered how to code cases where neuroscientific evidence was associated with more than one offender in a judgement, or where the expert witness’ area of expertise was unclear. By coding each new case, I realised that there were many other matters that had not been considered in the coding platform. I decided to test more cases and update the coding platform until sufficient consistency was achieved. This process was repeated for 60 cases until that conclusion was reached. Once the pilots were concluded, all 60 cases were re-coded in order to apply the new updates to all cases. That said, some other minor updates were required later, but there was no need to re-code all the questions in all the cases from the beginning, and only the updated questions in each case were separately re-coded (Qualtrics facilitates such changes). Following adjustments, the total number of questions listed in the coding platform reached 259.
After coding all the cases, questions in Qualtrics corresponding to the key factors under each neurolaw claim were answered. Once I carried out this process for all the judgements, Qualtrics transformed the answers to all the questions into quantitative data in a systematic way that could be easily sorted and compared. This data was suitable for drawing conclusion with regards to the large dataset. This allowed me to have an overview of the dataset, organise the qualitative data, and measure and compare the extent of the occurrence/existence of the specific themes of neurolaw claims. For instance, I could measure the proportion of appellate court cases which supported the double-edged sword claim and compare the result with trial (lower) court cases. It also enabled me to examine whether there are differences in conclusions as to the double-edged sword claim where different types of neuroscientific evidence (i.e. neuro-tool evidence vs non- neuro-tool evidence) were introduced in the judgements.
Although in this study quantitative data is used to examine the neurolaw claims, the study remained largely qualitative because quantitative analysis overlooks an understanding of how neuroscientific evidence is being used or why it is used in a specific way. For instance, quantitative analysis can indicate how commonly neuroscientific evidence leads to a finding that the offender is a danger to society and points towards a harsher punishment. However, quantitative analysis cannot explain why, in similar cases, the court may not find this to be a reason to increase the sentence. Qualitative analysis, on the other hand, can assist in better understanding such discrepancies. It can indicate, for example, that if an expert finds a brain abnormality treatable, then the court does not consider the offender to pose a danger to society and will consequently not use the evidence to support a harsher punishment. A further reason that this study favours
100
qualitative analysis is that the contextual inferences of the case content in this study are highly qualitative (qualitative content analysis), and this comprises a big proportion of this research.
Conclusion
The aim of this Chapter was to discuss the methodological approach to this research. In the first part of the Chapter, I discussed different sources of data and explained why criminal cases (judgements) appear to suit the objectives of this research more than other sources, such as surveys and interviews with judges and lawyers. In the second part of the Chapter – sampling and data collection – I first defined the three types of neuroscientific evidence considered in this study (i.e. imaging evidence, neuro-tool evidence, and non-neuro-tool evidence). Then, I moved on to review the process of collecting criminal cases, excluding irrelevant cases, and selecting the final sample of cases that comprised the dataset of this study. In the final part of this Chapter, I presented the analytical approach taken to analysing these criminal cases. I explained how different parts of judgements were labelled and connected to neurolaw claims, and then sorted using a computer program.
In the next five Chapters I use the results of this empirical research to draw conclusions regarding the neurolaw claims.
101
Chapter 5 Double-edged Sword Effect of Neuroscience
102
Introduction
This part of the thesis, Chapters 5 to 9, will focus on the five neurolaw claims. The first claim that I will discuss is the ‘double-edged sword’ effect of neuroscientific evidence (also known as the ‘double-edged sword of biological explanation of criminal behaviour’).1 According to this descriptive claim, neuroscientific evidence – such as proof of a brain impairment – may result in either a more lenient or a harsher punishment.2 For example, a brain impairment may indicate that the offender had a lower capacity to control their behaviour at the time of offence and is consequently less culpable (the neuroscientific evidence therefore having a mitigating effect), however the same evidence may lead the court to increase the sentence in order to protect the community from an individual whose brain impairment may result in a high risk of recidivism (the neuroscientific evidence therefore having an aggravating effect). As such, defence lawyers may be uncertain as to whether they should choose to introduce neuroscientific evidence and risk increasing the sentence, or instead choose not to introduce such evidence and thereby lose the opportunity to mitigate the sentence.3
Many scholars have discussed the double-edged sword claim.4 For instance, Owen D. Jones and Francis X. Shen summarise the claim by stating that, ‘a brain too broken may be simply too dangerous to have at large, even if it is somehow less culpable.’5 Similarly, Henry T. Greely, drew on his experience organising a seminar for federal judges and observing their reaction to different topics of legal discussion on neuroscience and genetics to observe that ‘[i]ronically, what might be set out as a mitigating factor for a defendant in terms of responsibility is likely to increase the sentence for a convicted criminal.’6
In general, proponents of the double-edged sword effect of neuroscience claim that neuroscientific evidence may tend towards either mitigation or aggravation of the sentence. For
1 JA Chandler, ‘The use of neuroscientific evidence in Canadian criminal proceedings’, in Journal of Law and the Biosciences, vol. 2, 2016, 550–579. 2 See, NA Farahany & JE Coleman Jr, ‘Genetics, Neuroscience, and Criminal Responsibility’ in The Impact of Behavioral Sciences on Criminal Law, NA Farahany (ed), New York, Oxford University Press, 2009, pp. 183–240; AS Barth, ‘A double-edged sword: The role of neuroimaging in federal capital sentencing’, in American journal of law & medicine, vol. 33, 2007, 501–522; OC Snead, ‘Neuroimaging and the Complexity of Capital Punishment’, in NYUL Rev., vol. 82, 2007, 1265–1339. 3 See DW Denno, ‘The Myth of the Double-Edged Sword: An Empirical Study of Neuroscience Evidence in Criminal Cases’, in Boston College Law Review, vol. 56, 2015, 493–551. 4 See, Farahany and Coleman Jr, pp. 183–240; Barth, 501–522; Snead, 1265–1339; BY Cheung & SJ Heine, ‘The Double-Edged Sword of Genetic Accounts of Criminality Causal Attributions From Genetic Ascriptions Affect Legal Decision Making’, in Personality and Social Psychology Bulletin, vol. 41, 2015, 1723–1738. 5 O Jones & F Shen, ‘Law and neuroscience in the United States’ in International Neurolaw: A Comparative Analysis, TM Spranger (ed), Springer, 2012, pp. 349–380 (chap. 362). 6 HT Greely, ‘Neuroscience and Criminal Justice: Not Responsibility but Treatment’, in Kan. L. Rev., vol. 56, 2008, 1103–1159 (footnote, p. 1104). 103
the use of neuroscience in courts to be consistent with this claim, the quantitative analysis should reveal a close to an equal number of cases where the introduction of neuroscientific evidence points towards either mitigating or aggravating the offender’s punishment. In other words, the closer the contribution of neuroscience is to a fifty-fifty distribution of cases between aggravation and mitigation of the sentence, the more consistent the double-edged sword claim is with the use of neuroscience in Australian courts. As such, cases where neuroscience only supports a more lenient sentence, or a harsher punishment, do not independently and directly support the double- edged sword claim; rather it is the overall distribution of cases that will indicate when the claim is consistent with practice — i.e. when considering all cases in the dataset of this study, the results suggest that neuroscience either points towards mitigating or aggravating the sentence. For example, if 10 per cent of cases show neuroscience being used as an argument for aggravating the sentence, and 90 per cent of cases show neuroscience being used to mitigating the sentence, then the double-edged sword claim is not consistent with the use of neuroscience in courts. However, cases in which neuroscience is considered to tend towards both aggravating and mitigating the sentence simultaneously, provide direct support for the double-edged sword effect. Nevertheless, a few cases that fit this condition cannot be extended to draw a general conclusion that the double-edged sword claim is consistent with the use of neuroscience in courts.
Empirical neurolaw studies have also discussed this claim.7 In particular, American legal scholar Deborah Denno, in ‘The Myth of the Double-Edged Sword: An Empirical Study of Neuroscience Evidence in Criminal Cases’,8 found that neuroscientific evidence is usually introduced for the purpose of mitigating the sentence, and it is rarely a function of neuroscientific evidence to support claim that the offender poses a potential future risk. She concludes that in the US criminal justice system, the double-edged sword claim is not consistent with the use of neuroscience in 9 practice.
In Australia, however, no study has examined the double-edged sword claim. As such, the main aim of this Chapter is to fill this gap. By the end of this Chapter, I will conclude that empirical analysis does not suggest that the use of neuroscience in sentencing is consistent with the double- edged sword claim. In the majority of cases where neuroscience evidence appears it points towards mitigating the punishment and only tends towards a harsher punishment in a limited number of cases.
7 See, Barth, 501–522; Farahany and Coleman Jr, pp. 183–240. 8 Denno, 493–551. 9 Also see CH de Kogel & EJMC Westgeest, ‘Neuroscientific and behavioral genetic information in criminal cases in the Netherlands’, in Journal of Law and the Biosciences, 2015, 1–26. 104
An important consideration in the examination of this claim is an argument against the usefulness of neuroscience in criminal procedure. One example is where psychological examination indicates an offender has impulse control disorder, and a brain scan further shows impairment to an area of the brain associated with impulse control. In such cases, some scholars argue that neuroscience merely duplicates the psychological analysis and therefore should have no actual influence on sentencing (having merely rhetorical relevance).10 Hence, it may be argued that in such cases neuroscientific evidence should not be considered in evaluating the double-edged sword claim. To minimise any conflicting argument in this Chapter regarding the double-edged sword effect of neuroscience, I will provide a brief explanation: in the majority of judgements, it is generally not possible to distinguish between the influence of different types of evidence during sentencing. Judges make a general conclusion concerning different types of evidence tendered to the court and it is not possible, for instance, to indicate whether a brain scan was necessary in determining the existence of an impulse control disorder, or whether psychological evidence provided sufficient information to reach that conclusion. Consequently, in some cases, neuroscience may have no actual effect on considerations in sentencing. This raises a limitation of this study, which is that it cannot determine whether the admittance of neuroscientific evidence had a causal effect in altering the length of the sentence. Accordingly, in this Chapter, my analysis is based on the assumption that where neuroscientific evidence is adduced and considered by the court, it contributed to the sentencing decision, regardless of whether other evidence was available. This contribution may be significant, slight or ineffective. Of course, in some cases the court may not consider neuroscientific evidence for a variety of reasons, such as where the brain image does not indicate any abnormality. In these cases, I do not consider neuroscience to have contributed to the sentencing decision.11
Further, some may argue that where there are other factors that nullify or outweigh the influence of neuroscience on sentence, neuroscientific evidence should not be considered in the examination of the double-edged sword claim. For instance, neuroscience may appear to support a more severe sentence, however, once the court has also considered the mitigating influence of the offender’s lack of criminal record the aggravating influence of neuroscience on the sentence is nullified. In sentencing, there are usually several mitigating and aggravating factors, but this does not necessarily result in an increased or diminished sentence, as the factors may balance each other out and therefore have no influence on the ultimate sentence. Nonetheless, this does not deny the general relevance and contribution of neuroscience or any other evidence in
10 S Morse, ‘Lost in Translation?: An Essay on Law and Neuroscience’ in Law and Neuroscience, Current Legal Issues, M Freeman (ed), Oxford University Press, 2011, pp. 529–562. 11 For a similar discussion, see DW Denno, ‘Courts’ increasing consideration of behavioral genetics evidence in criminal cases: Results of a longitudinal study’, in Mich. St. L. Rev., 2011, 967–1047 (p. 975). 105
sentencing that supports a mitigating or an aggravating consideration. It is the totality of the contribution of different evidence in a case that should determine the sentence.12
Further, if the aggravating effects of neuroscientific evidence were to be balanced out by its mitigating effects, this would indicate that neuroscience has no influence on sentencing. However, it is unclear the extent to which a single piece of evidence would influence considerations in sentencing, and the actual influence that neuroscientific evidence has on the sentence cannot be measured. Indeed, this poses another limitation for this study, which is that it cannot determine whether neuroscience balances out its mitigating and aggravating effects. This study can merely suggest whether neuroscience has contributed to the sentence, regardless of its actual influence.
Taking these issues into consideration, this Chapter cannot discuss whether neuroscience is useful or determine whether it actually increases or decreases the sentence, but it aims to examine whether neuroscience, when adduced in proceedings, is used to support a more lenient or harsher punishment. As such, throughout this Chapter, where I indicate that neuroscience contributes to/points towards aggravating or mitigating the sentence, I only mean that neuroscience was relevant to the consideration of whether a sentence should be extended or mitigated, rather than it actually makes the sentence longer or shorter.13
This Chapter is divided into five sections. In the first section, I will briefly review sentencing factors in NSW to which neuroscience may in theory be relevant. Then, to gain more insight into the ways neuroscientific evidence may contribute to sentencing and subsequently the double- edged sword effect, I will present a qualitative analysis of several representative cases that
12 David Hodgson, discussing the contribution of evidence that may support a harsher or a more lenient sentence explains that, ‘[a]dmittedly, some of these factors may not work unequivocally in reducing the penalty. For example, if genetic or environmental factors have resulted in a mental abnormality which makes the offender more prone to violence, this may diminish moral culpability and thus point towards a shorter sentence of imprisonment; but at the same time point towards a longer sentence because of considerations of deterrence (greater penalties may be thought necessary to deter such persons from offending) and protection of society from risk. However, this does not deny the general relevance of diminished moral culpability as a mitigating factor’. D Hodgson, ‘Guilty mind or guilty brain? Criminal responsibility in the age of neuroscience’, in Australian Law Journal, vol. 74, 2000, 661–680 (p. 665). 13 An alternative method for investigating this research question may be to compare cases involving neuroscientific evidence with cases that do not contain this evidence (control group). This may allow us to determine whether neuroscientific evidence has an actual effect on sentencing. However, this approach requires sampling judgments that are comparable. This means that for each judgment that contains neuroscientific evidence, a parallel judgment as a control group should be identified, which includes very similar variables including, the type and circumstances of the offence, and subjective variables of offender (e.g. age, sex and background) and victim (e.g. severity of injury, age, sex and relationship with offender). This degree of similarity between judgments is unlikely. Even if such judgments were found, as I have explained in Chapter 3 (legal background), sentencing is strongly dependent on the subjective discretion of judges and two courts may impose different sentences for the same case. On this, the High Court has noted that ‘there is no single correct sentence’ Markarian v R [2005] 228 CLR 357 at [27] (citations were removed). 106
demonstrate how neuroscientific evidence may support a more lenient or a harsher punishment in courts. This provides a particularly meaningful context for the third section of this Chapter where I will turn to present an analysis of the quantitative data and discuss the number of cases in which neuroscience point towards a shorter or a longer sentence. I will conclude that the findings do suggest that the double-edged sword claim is inconsistent with the actual use of neuroscience in NSW and the number of cases where neuroscience supports a more lenient sentence is considerably greater than the number of cases where this evidence supports a harsher punishment. That means that submitting neuroscientific evidence is likely to be a beneficial strategy for the defendant. In the fourth section of this Chapter, I will make three observations and explain how they may change the conclusion I have reached about the double-edged sword claim and posit that defence lawyers still need to be cautious about the risk of adducing neuroscientific evidence. First, I will posit that for lawyers, it is both important to consider how frequently neuroscience tends towards a harsher or more lenient punishment, and also the weight that the neuroscientific evidence carries in influencing sentencing outcomes. Second, I will consider the possibility that lawyers may abstain from the use of neuroscientific evidence where the content of evidence suggests the possibility that the evidence may contribute to a harsher punishment. Therefore, the dataset of the study is biased towards cases where lawyers were confident that neuroscience would result in a more lenient sentence. Third, I will discuss two Acts, the application of which defence lawyers need to be cautious of in cases where they adduce neuroscientific evidence. The fifth section of this Chapter will compare the findings of this study of the double-edged sword claim with studies from other jurisdictions.
1. An overview of sentencing elements in NSW that are potentially relevant to neuroscience evidence
As discussed in Chapter 3, an introduction to legal background, factors that courts consider in the sentencing considerations are recognised in legislation – section 3A (aims of sentencing) and section 21A (mitigating and aggravating factors) of the Crimes (Sentencing Procedure) Act 1999 – and the common law. I will briefly review these factors.
Considering the main objective of this Chapter, I have divided the sentencing factors that are potentially relevant to neuroscience into two categories. I will first outline sentencing factors that may lead to a more lenient sentence and then those factors that may contribute to a harsher punishment.
107
If brain impairment affects the offender’s judgement, his/her ability to understand the wrongfulness of criminal behaviour, or his/her capacity to control his/her acts and emotions, then this may reduce the moral culpability of the offender. Meaning, if brain impairment is causally connected to the commission of the offence (or it can explain the crime), the offender is less blameworthy14 and the court may reduce their view of the objective seriousness of the crime,15 as well as the associated retribution16 and denunciation.17 Another sentencing consideration that is relevant to neuroscience is deterrence. Deterrence uses the punishment of the offender to dissuade the public (general deterrence) and the offender (specific deterrence) from committing a similar crime. However, a brain impairment may mean that the offender is not a suitable vessel for deterring other individuals from committing an offence,18 and this may also reduce the value of specific deterrence. This is because punishment or the threat of punishment may not deter a person with a brain abnormality from offending in the same way it may be expected to deter a person without a brain abnormality, or because the offence was caused by a brain impairment which has subsequently been treated.19 If neuroscientific evidence suggests that the brain impairment can be treated, and that the offender can be rehabilitated, it may mitigate the sentence because the offender has good prospects of rehabilitation.20 Further, the court may reduce the sentence if the impairment makes the sentence more onerous for the offender than for other prisoners.21
Conversely, brain abnormality may increase the risk of re-offending and increase the need for the protection of society through an aggravated sentence.22 Similarly, the court may give more significant consideration to specific deterrence to prevent the offender from further offending.23
Before I move on to the next section, it is useful to make it clear how I will refer to these sentencing factors where they serve to increase or decrease the allocated punishment. In section 21A of the Crimes (Sentencing Procedure) Act 1999 (NSW), some aggravating and mitigating sentencing factors are outlined. The court, based on the circumstance of the case, may or may not consider factors in section 21A when making sentencing decisions. Unlike section 21A, the aims
14 R v Israil [2002] NSWCCA 255; R v Hemsley [2004] NSWCCA 228 at [33] – [36] 15 Kennedy v T [2008] 181 A Crim R 185; [2008] NSWCCA 21 at [39] 16 R v Israil [2002] NSWCCA 255 17 D Howard & B Westmore, Crime and Mental Health Law in New South Wales: A Practical Guide for Lawyers and Health Care Professionals, 2nd ed., LexisNexis, AU, 2010. Also see R v Tsiaras [1996] 1 VR 398 and Courtney v R [2007] 172 Crim R 371; [2007] NSWCCA 195 at [14]; Mudlrock [2011] HCA 39, 244 CLR 120 at [58] 18 Engert [1995] 84 A Crim R 67 at 71 per Gleeson CJ; R v Hemsley [2004] NSWCCA 228 19 Director of Public Prosecutions (Cth) v De La Rosa [2010] NSWCCA 194; also see, R v Henry [1999] 46 NSWLR 346, (Wood CJ at CL, Adam and Kirby JJ) at [46] 20 S 21A(3)(h) Crimes (Sentencing Procedure) Act 1999 21 Freckelton and Selby. Also see, R v Hemsley [2004] NSWCCA 228 22 Benitez v R [2006] 160 A Crim R 166; De La Rosa [2010] NSWCCA 194 23 SJ Odgers, Sentence : the law of sentencing in NSW courts for State and Federal offences, 3rd ed., Haberfield, N.S.W, Longueville Media, 2015. 108
of sentencing (provided in section 3A of the same Act) must be considered during sentencing and the court, depending on the circumstances of the case, may give more or less weight to each of these aims (e.g., deterrence). This consequently results in a shorter or longer sentence of imprisonment. As such, unlike section 21A factors, these aims of sentencing are not mitigating or aggravating sentencing factors in and of themselves. That said, in order to make my discussion throughout this thesis more concise, both sentencing aims and section 21A factors, will be referred to as mitigating/aggravating factors where the court, based on the circumstance of the case, gives more or less weight to that aim and on that basis mitigates or aggravates the sentence. For instance, where I note that specific deterrence is a mitigating sentencing factor, it means, lower weight is given to this sentencing aim and therefore it has contributed to a shorter sentence.
2. A qualitative analysis of neuroscientific evidence in sentencing
In the previous section, I explained how neuroscientific evidence may theoretically contribute to sentencing outcomes. To provide a meaningful conclusion as to the relevance of the double-edged sword claim, before I present quantitative analysis of cases and draw conclusions as to how frequently neuroscience tends towards a more lenient or a harsher punishment, in this section I will discuss some cases that illustrate four general ways that neuroscience is associated with sentencing in relation to the double-edged sword claim. These cases will illuminate circumstances in which neuroscientific evidence has contributed to either mitigation or aggravation in sentencing. Then, I will move on to discuss cases where neuroscience supports both a more lenient and a harsher punishment simultaneously. Further, while one aspect of neuroscience is that it may suggest an increased risk of future offending, I will present cases where neuroscience may suggest otherwise. In other words, where neuroscience shows that an offender may no longer pose a potential danger to society.
Neuroscience points towards a shorter sentence
Neuroscience may tend towards a more lenient sentence when considered in relation to several sentencing factors. In this section, I will provide some exemplary cases that help to illustrate how judges determine that neuroscience is relevant to mitigating sentencing factors in different circumstances.
Brain impairment may explain criminal behaviour, and this explanation is important in criminal law as it can be relevant to sentencing by showing that the offender is less culpable (or less 109
blameworthy). Therefore, it is necessary to show that the impairment existed before the crime happened and also that it had an impact on the offender’s behaviour at the time of the offence (i.e. causal relationship). Accordingly, even if neuroscientific tools prove the existence of some brain issues, experts still need to explain the reasons for their belief that the impairment existed before the offending, and how it contributed to the crime.
In R v Lea-Caton,24 the offender pleaded guilty to a especially aggravated form of kidnapping and the murder of two victims (four offences in total). The offender was accompanied by two co- offenders. On the day of the incident, the offender's role was to guard the victims; he was not involved in attacking or tying them up. He also was not involved in killing the victims but assisted in moving the bodies to some remote area where they were burned. The offender surrendered himself shortly after the crime.
With respect to his medical condition, the offender had a serious motor vehicle accident where he ‘sustained severe injuries including a depressed fracture of the left frontal skull’. His sister stated that his personality and mood changed following the accident: the offender went into a cycle of depression and also attempted to commit suicide; moreover, following the accident he frequently committed crimes, including breaking and entering, dangerous driving and drug crimes.25 From the judgement it appears that before the accident he was not convicted of any offence.
After studying an MRI scan, a clinical psychologist gave the opinion that the offender suffered from frontal lobe syndrome as the result of the accident. The offender's sister noted that he found it difficult to make decisions in situations where he is under pressure. The expert found this to be consistent with the offender’s psychological assessment.
The judge acknowledged that the frontal lobe condition was a factor which ‘somewhat diminished [his] capacity … to have extricated himself from the situation in which he found himself on that fateful day’.26 In determining the objective seriousness of the crime, the judge considered that the offender’s ‘psychological condition compromised his capacity to react appropriately to the circumstances in which he found himself’.27 The court imposed a sentence of 22 years of imprisonment with a non-parole period of 16 years and 6 months.
24 [2007] NSWSC 1294 25 Id. at [36] - [38] 26 Id. at [45] (emphasis added). 27 Id. at [59] 110
This case suggests that the court accepted that the brain condition partly explained the criminal behaviour. The evidence showed that the offender’s capacity was reduced due to his brain injury and therefore neuroscience was found to be relevant to mitigating the sentence by way of reducing the objective seriousness of the offence.28
In order to more clearly indicate the contribution of neuroscience to a more lenient sentence, I describe a case where the sentencing court did not give sufficient consideration to brain impairment and this error was recognised in the Court of Criminal Appeal (Supreme Court of NSW).29 In Carroll v R, 30 the offender slashed his victim’s forearm and face with a knife following a fight. He confessed to his crime a few weeks later, after surrendering to the police. The maximum penalty for this offence is 25 years imprisonment with a standard non-parole period of 7 years.
According to the report submitted by a psychiatrist on behalf of the offender, he sustained two head injuries in 1999 and 2002 that resulted in loss of consciousness, and mild amnesia. The hospital records indicated that following the head injuries he suffered from seizures. According to the medical records, the offender was almost always violent during the post-ictal — the 5 to 10 minutes following his seizures. In addition, the side effects of seizure medication included a worsened mood and poor temper control. However, the expert did not believe that seizures had a direct connection with the offence. Instead, the psychiatrist noted that the offender ‘probably suffered frontal lobe damage and that this would have contributed to a diminished capacity for self-control and increased his vulnerability to acting in an impulsive and aggressive manner when angered’.31
The lower court in sentencing the offender made no reference to the psychiatrist’s opinion about the possible causal contribution of frontal lobe damage to the offending and the only mitigating factor considered was the guilty plea. The court sentenced the offender to 8 years and 3 months imprisonment with a non-parole period of 6 years.
28 Neuroscience was not the sole factor in proving the offender’s mental condition; a psychological assessment and behavioural evidence provided by his sister also contributed to proving his diminished capacity. This combination of evidence indicates the difficulty that can be faced in distinguishing the influence of neuroscience from other sources in sentencing decisions that I discussed in the introduction of this chapter. 29 Later in the thesis, I will discus a more comprehensive discussion of the courts error regarding the consideration of neuroscientific evidence. 30 [2012] NSWCCA 118. 31 Id. at [30] 111
The offender appealed against the sentence on various grounds, including the failure to properly consider the offender’s mental condition during sentencing. The Court of Appeal noted that the sentencing court had apparently rejected the contribution of the offender’s cognitive deficits to the offending, whereas the applicant’s brain injuries, his seizures, and the side effects of his medication should have been considered. This ground of appeal was upheld and alongside other grounds, led to resentencing. In resentencing the offender, the Court of Criminal Appeal stated that:
It seems to me that significant weight must be given when determining a sentence to the impact that the applicant's head injuries have had on his behaviour, including the fact that he has poor temper control … it seems obvious, and I would conclude, that poor temper control played a significant role in the events that occurred.32
By considering the effect of the offender’s head injuries on his moral culpability alongside other sentencing matters, the appeal court decided unanimously to re-sentence the offender to a non- parole period of 4 years with a remaining balance of 2 years. While it appears that there was no neuro-tool evidence (such as brain scan or neuropsychological testing), the court gave considerable attention to the offender’s head injuries.
In these cases, and many others, the discussions are centred around the contribution of brain abnormality to criminality, whereas sometimes that is not the case: either the impairment does not exist at the time of offending, or it is not causally connected to offending. Nonetheless, it does not mean the brain impairment is not relevant to sentencing. An example involves the sentence of three offenders for their involvement in a physical conflict which led to a death.33
One of the offenders, Lonsdale, pleaded guilty to manslaughter (maximum penalty 25 years), and the other two offenders pleaded guilty to manslaughter and affray. According to a psychiatrist’s report, following a head injury in football, Lonsdale was diagnosed with schizoaffective disorder. His CT and MRI brain scans indicated no abnormality, however, EEG (electroencephalogram) showed some brain abnormalities. Also, a SPECT scan indicated ‘abnormal perfusion in areas of the brain consistent with the presence of a psychotic episode.’34 About four years before the crime, he was admitted to hospital involuntarily which seemed to be a typical drug-induced psychosis. After his recovery, there was no further episode of mental illness. There was no sign of recurrence of ‘symptoms of schizoaffective disorder’ or drug-induced psychosis around the time of the crime.
32 Id. at [102]; [103] per Garling J 33 R v Clay, Lonsdale and JM [2006] NSWSC 1220 34 Id. at [38] 112
The defence sought a more lenient sentence based on the offender’s psychotic illness and brain function abnormalities. The court stated that there was no causal connection between the crime and the offender’s mental condition, and he was sufficiently aware of his acts. Nevertheless, the court noted that because of his condition, ‘some modest adjustment of the otherwise appropriate sentence should be made’35 based on general deterrence and the fact that the sentence may weigh more heavily upon him.36
So far, I have discussed how neuroscience may support a more lenient sentence, i.e. the mitigating side of the double-edged sword claim. I will now move on to discuss how neuroscience may contribute to a harsher punishment.
Neuroscience points towards a longer sentence
The other side of the double-edged sword claim is that neuroscience may point towards a longer sentence. The aggravating sentencing factors to which neuroscience is relevant in criminal cases are almost always associated with protection of the community from the offender, and only occasionally associated with specific deterrence to prevent further offending.
The only case in the dataset where neuroscientific evidence was only relevant to aggravating the sentence and not mitigating it was Ngati v R.37 In this case, the applicant pleaded guilty to two counts of robbery and the two Form 1 offences associated with each offence (i.e. robbery in company and detaining a person for advantage in company). The robberies involved serious threats and the use of significant violence against the victims. The applicant appealed against the severity of the sentence on the basis that the sentencing judge erred in mitigating the sentence and did not give due consideration to the causal relationship between the applicant's impaired intelligence and his offending.
The offender had a substantial criminal history that included robbery and arson. He had started using heroin at the age of 17. It was noted that he had committed the most recent offence to provide money for his heroin addiction. This substance abuse was also accompanied by mental impairment. A psychologist conducted psychometric testing and reported significant impaired intellectual functioning (that is, an IQ score ranging between 52 and 66). A psychiatrist also examined the applicant and reported impaired intelligence and cognitive capacity. The applicant had sustained a head injury when he was in Year 7, however, both the psychologist and the
35 Id. at [65] 36 Id. at [64]; [65] 37 [2014] NSWCCA 125 113
psychiatrist could not conclude whether the head injury had had any considerable impact on the applicant.38 Neither the appellate court nor the sentencing court found a causal connection between impaired intellect and the offence. This was because the crime had involved a degree of planning, the applicant had been fully aware that his conduct was wrong, and the crime had not been the result of an impulsive act.
Counsel for the applicant referred to the psychologist's report and the result of psychometric testing to support their claim of a causal relationship between the applicant’s mental condition and offending. Counsel stated that ‘[t]he likelihood that [the applicant] may return to substance use and related offending in times of stress may be aggravated by below average intelligence, which can be associated with decreased control of [offending] behaviours and capacities for consequential reasoning’.39 However, the appellate court noted that this section of the expert report was directed to applicant's likelihood of future offending and did not relate to whether this condition may have contributed to the offences he was being punished for. The sentencing court concluded that the applicant did not have good prospects of rehabilitation, and on the contrary, he had a real risk of re-offending. As such, the result of psychometric testing that indicated significant impaired intellectual functioning only tend towards aggravating the sentence.
The cases discussed in this section and the previous section illustrate how neuroscience may contribute to mitigating or aggravating an offender’s sentence in different circumstances. In the section that follows, I will discuss cases that illustrate how neuroscience may support a longer and shorter sentence simultaneously.
Neuroscience cut both ways: supporting a harsher and a more lenient sentence simultaneously
As discussed, the double-edged sword claim asserts that neuroscience may contribute to either mitigation or aggravation of the sentence. I explained that for this assertion to be consistent with the use of neuroscience in court, the quantitative analysis should reveal a close to fifty-fifty distribution of cases in which neuroscience aggravates and mitigates the sentence. As such, the cases that were discussed above do not independently and directly support the claim; rather it is the overall distribution of cases that will indicate whether the claim is consistent with practice. Unlike those examples discussed so far, cases in which neuroscience is considered to contribute to both shortening and lengthening the sentence simultaneously, provide direct support for the
38 Id. at [2] 39 Id. at [35] 114
double-edged sword effect. Nonetheless, a few cases of this nature cannot support a general conclusion that this claim is consistent with the use of neuroscience in sentencing courts of NSW.
In 2012, in R v McCann,40 the jury found McCann guilty of manslaughter for strangling the deceased. Five experts provided evidence regarding the offender's mental condition. A clinical psychologist found that the offender displayed cognitive impairment consistent with frontal lobe damage that could cause issues such as poor planning and evaluation of behavioural responses, mental inflexibility, and impairment of social judgements. A neuropsychologist reported that CT and MRI scans showed focal abnormalities in the frontal lobes and significant cerebral atrophy. Also, the neuropsychological assessment revealed problems with working memory, learning and remembering complex new information, and impairment in creating novel ideas.
Three psychiatrists who read the neuropsychologist’s reports (including the reports on neuro- images), came up with a similar conclusion: that the offender had ‘atrophy of the frontal lobes and general brain shrinkage’ at the time of the crime that resulted in cognitive impairment (an impairment of the brain executive function) and escalated the possibility of misunderstanding events, engaging in impulsive behaviour, and disturbed decision-making.
Although the sentencing judge acknowledged that the offender’s mental condition meant that lower weight should be attached to elements of retribution and general deterrence, his Honour also added that, ‘[h]owever, … the irreversible condition of the frontal lobes and the brain generally, raises the question of his dangerousness in the future’.41 Although two psychiatrists believed that the risk of re-offending was fairly low, the judge accepted the third psychiatrist’s opinion that the offender posed a risk of violence to the community once released. As such, the offender’s brain condition simultaneously supported mitigating and aggravating the sentence.
There are also two High Court decisions that discussed where neuroscience supports a more lenient and a harsher punishment simultaneously. What makes these cases interesting is that they were decided more than three decades ago: Veen v R [1979]42 and Veen v R [1988]43 (hereafter, I will refer to them Veen (No.1) and Veen (No.2) respectively).
Veen was adopted at age 2, and from age 11 his foster parents found it difficult to control him. By the age of 15, he was removed from their custody. When Veen was 16, he stabbed his landlady
40 [2012] NSWSC 1462. 41 Id. at [36] 42 HCA 7; [1979] 143 CLR 458 (28 February 1979) 43 [1988] HCA 14; [1988] 164 CLR 465 (29 March 1988) 115
multiple times after consuming a large amount of alcohol. The victim survived, and Veen pleaded guilty. He was sentenced for the crime of malicious wounding.
Veen’s second violent offence occurred when he was 20. He was invited by the victim to his house. Following some homosexual activities and heavy drinking, the applicant expected money from the victim who resisted payment and made a racial slur against him (Veen was aboriginal). Veen fatally stabbed the victim. He left the premises taking a number of items with him.
A psychologist (for the defence), after conducting a number of tests on one occasion, opined that Veen had diminished emotional control, and concluded that he suffered from a type of brain damage which if coupled with severe provocation and alcohol, could result in an ‘uncontrollable aggressive manner’. While admitting that Veen could understand what he was doing, the psychologist argued that he could not control his behaviour. On the other hand, an experienced psychiatrist (for the prosecution) met him on several occasions and reported that Veen had no mental impairment for his behaviour and concluded that ‘he had a personality defect in the form of a character disorder’.44 He also ‘denied the ability of any psychologist to make a diagnosis, in this case, criticised aspects of the psychologist’s test procedures and rejected the conclusion that there was any brain damage’.45
The jury found Veen guilty of manslaughter due to the abnormality of mind caused by his brain damage. The Supreme Court of NSW concluded that Veen undoubtedly had a brain abnormality at the time of both violent crimes. Accordingly, the court found that there was little doubt that Veen, while continuing to suffer from brain damage, would kill or seriously injure other people if he was allowed to return to society. No evidence of a curable condition or any prospects of rehabilitation were found. The court, based on Veen’s mental condition and the fact that there was no institution available to provide treatment, concluded that he should be imprisoned for life for the purpose of protecting society from his uncontrollable behaviour.46 As such, although Veen’s brain damage reduced his criminal responsibility and changed the murder conviction to manslaughter, it also leads to the imposition of a harsher punishment. This decision dates back to 1975 and is the first example among the collected cases in the dataset of this study in which neuroscientific evidence resulted in both mitigation and aggravation of the sentence.
Veen appealed to the Court of Criminal Appeal against the sentence of life imprisonment, however his appeal was unsuccessful. He then sought leave to appeal to the High Court (see, Veen
44 Id. p 464 45 Id. pp 464 & 487 46 Id. p 489 116
(No. 1)). In the High Court, arguments were raised as to the permissibility of life imprisonment on the basis of legal rules and the specific facts of this case. Beyond those discussions, a majority of the High Court criticised the trial judge’s conclusion of future dangerousness based, among other things, on Veen’s brain injury. The reasons for this included: the existence of a psychiatrist’s evidence to the contrary; the fact that the psychologist’s evidence was based on an examination that occurred on only one occasion; that determining future dangerousness requires compelling evidence which was not available; and that the applicant had a record of only one violent crime which was not sufficient evidence to allow the conclusion that there was significant future risk of violent re-offending.47 The High Court, by the majority, ordered that the term of imprisonment be decreased to 12 years’ imprisonment, with no fixed non-parole period.
Veen was released on licence in 1983 and after about 9 months killed another person. Similar to the previous killing, he stabbed the victim repeatedly after being invited to engage in homosexual activity. He was charged with murder but the plea of manslaughter was accepted by the Crown on the grounds of his brain damage and diminished responsibility. The court sentenced Veen to life imprisonment for the protection of society, as the trial judge accepted the brain damage on the basis of ample medical evidence and the fact that when the applicant was drunk or under stress, he would commit similar crimes. The applicant appealed against the sentence of life imprisonment. The judgement was upheld by the Court of Criminal Appeal (Supreme Court of New South Wales). Veen then obtained special leave to appeal his sentence in the High Court, i.e. Veen (No.2).
The High Court in Veen (No.2) noted that contrary to Veen (No.1), the previous killing and medical evidence sufficiently supported the probability of future offending. As such, although Veen’s mental abnormality caused by his brain injury reduced the murder charge to manslaughter, it made him a grave danger to the society. Therefore, special leave to appeal was granted and the appeal was dismissed (I will return to this point that in exceptional circumstances, how brain impairment may result in life imprisonment).48
47 Id. per Jacobs J, Stephen J and Murphy J. Another matter discussed was the possibility that the jury’s verdict of manslaughter was based on the defence of provocation rather than Veen’s abnormality of mind. The trial court, to ascertain which one of these defences was determinative, asked the jury whether their judgment (the finding of manslaughter) was based on the offender’s abnormality of the mind. They responded in the affirmative. On appeal, the High Court noted that the question posed to the jury did not provide them with the opportunity to explain whether the verdict of manslaughter was based on the defence of abnormality of the mind or the defence of provocation. Instead, the court’s question was only directed at the defence of abnormality of the mind. It is unusual in an adversarial justice system for the court to ask a question directly of the jury during the trial. 48 While there were some contrary opinions regarding the permissibility of life imprisonment between High Court judges, the majority concluded that the circumstances of this case (i.e. seriousness of the offence) fall within the worst category of cases where the most severe punishment is reserved. 117
In this case, the influence of brain impairment on the sentence as a double-edged sword is clearly recognised. In particular where the majority with respect to Veen’s brain injury notes that,
[The purposes of sentencing] are guideposts to the appropriate sentence but sometimes they point in different directions. And so a mental abnormality which makes an offender a danger to society when he is at large but which diminishes his moral culpability for a particular crime is a factor which has two countervailing effects: one which tends towards a longer custodial sentence, the other towards a shorter.49
Although the High Court refers to mental abnormality, the condition was caused by Veen’s brain injury. It is noteworthy that the double-edged sword effect is not limited to neuroscientific evidence (or more generally, biological explanations of criminal behaviour) and in NSW other forms of evidence are recognised to have a similar potential effect on sentencing. For instance, in Bugmy v R,50 the High Court referred to the influence of social deprivation on sentencing:
An offender’s childhood exposure to extreme violence and alcohol abuse may explain the offender’s recourse to violence when frustrated such that the offender’s moral culpability for the inability to control that impulse may be substantially reduced. However, the inability to control the violent response to frustration may increase the importance of protecting the community from the offender.51
Generally, since cases where neuroscience contributes to both aggravation and mitigation of the sentence are consistent with the double-edged sword claim, the Veen cases may suggest that the use of neuroscientific evidence in Australian criminal courts supports the proposition that neuroscientific evidence may act as a double-edged sword. In particular, in Veen (No. 2), the High Court indirectly affirmed the double-edged sword effect of neuroscience by dismissing the appeal. Further, as these cases appear to be two of the very first cases in which neuroscientific evidence was relied upon in an Australian court, this would also suggest that over the preceding three decades, neuroscience had continued to have a dual effect in many other cases. However, as I will discuss in section three (quantitative analysis of the double-edged sword claim), findings reveal the opposite
49 Veen v R (No.2) at [13] per Mason C.J., Brennan, Dawson and Toohey JJ, 50 [2013] HCA 37 at [44]. Bugmy’s case was an appeal from the Supreme Court of New South Wales and it was initially tried in the New South Wales Local Court. 51 Also, in R v Hemsley [2004] NSWCCA 228; BC200404239 Sperling J at 33-36 explained that, ‘[m]ental illness may be relevant … in three ways. First, where mental illness contributes to the commission of the offence in a material way, the offender’s moral culpability may be reduced; there may not then be the same call for denunciation and the punishment warranted may accordingly be reduced … Secondly, mental illness may render the offender an inappropriate vehicle for general deterrence and moderate that consideration … Thirdly, a custodial sentence may weigh more heavily on a mentally ill person … A fourth, and countervailing, consideration may arise, namely, the level of danger which the offender presents to the community. That may sound in special deterrence’. 118
In this section, I illustrated three cases where evidence of a brain abnormality explained the criminal behaviour and reduced moral culpability of the offender, concurrently, the same brain abnormality suggested risk of future offending and increased the consideration of protection of the society.
Perhaps remarkably, in some cases, a brain abnormality may suggest why the offender would not pose a risk in the future. These cases are against the claim of the double-edged sword effect of neuroscience because they suggest neuroscience may only reduce the punishment.
Neuroscientific evidence suggesting a reduced risk of re- offending
The relationship between neuroscience and sentencing with respect to the double-edged sword claim has another aspect: the brain impairment may be taken by the court to reduce or eliminate the risk of future offending. In R v Goodridge (No 2),52 the offender committed murder by forcing his arm violently into the victim's vagina and rectum, resulting in heavy bleeding and death. At the age of three, the offender was run over by a car and experience a head injury and could not remember his childhood. In 2010, he also suffered another head injury following an assault in jail. He started drinking from age 12 and consumed a significant amount of cannabis thereafter.
Four expert reports, consisting of three psychiatrists and a psychologist, were submitted to the court. The psychologist reported that the offender had marked cognitive deficits that were associated with his history of cerebral insult and a lengthy history of substance and alcohol abuse, and found his function at the mentally retarded range. An MRI scan by one of the psychiatrists indicated ‘cerebral atrophy and lesions in the white matter consistent with vascular disease’53 and led to the offender being diagnosed with dementia. The diagnosis of dementia was less than three years after the crime. The other psychiatrist also reported alcohol-induced brain damage and possible brain damage following previous head injuries with low baseline IQ, poor judgement and cognitive impairment. A cognitive test also confirmed the offender’s dementia; he could not remember the crime, the trial, or the reason he was in prison. The psychiatrist reported that if the cognitive decline continues the offender's lifespan would be not more than 6 - 7 years.54
The sentencing judge concluded that ‘His long-standing history of alcohol abuse and the concomitant brain damage meant that he was prone to binge-drinking which tended to make him
52 [2012] NSWSC 1180 53 Id. at [31] 54 Id. at [34]; [36] 119
behave erratically and aggressively’.55 With impaired judgement, his moral culpability was represented as reduced.56 Additionally, his mental condition significantly reduced the need for general and specific deterrence. It was concluded that his compromised mental condition and physical health meant that he no longer posed a potential danger to society, and he was likely to die before the end of the sentence (limiting term).
While the offender’s brain condition contributed to the commission of a violent murder, neuroscientific evidence was only used to mitigate the sentence as the possibility of aggravating the sentence for the protection of society was eliminated by virtue of his terminal illness.
In another criminal case,57 the offender’s brain injury occurred after the crime had taken place and so it could not have been considered as a factor contributing to the offending – however, the injury reduced the risk of re-offending. Moreover, in this case, the court’s comparison of sentencing circumstances before and after the brain injury clearly illustrates the extent to which neuroscience may mitigate the sentence by reducing the likelihood of future offending.
In this case, the offender was found guilty of two counts of attempt to choke with the intention of enabling himself to commit an indictable offence (i.e. intimidate the victim), and one count of sexual intercourse without consent. About a week after the trial the offender was the victim of an assault and suffered from a head injury while in custody. Different experts reported a major brain injury that led to problems such as difficulty remembering having committed the crime (amnesia), changes of personality, stress, and anxiety. While the brain injury occurred after the crime and was therefore not relevant to moral culpability, the judge noted that general and specific deterrence had a lesser role than where the offender had not suffered a brain injury. Also, the court considered the increased difficulty of holding the offender in custody and gave considerable weight to rehabilitation due to the offender’s brain injury: ‘It must be recognised that the longer he spends in custody the greater is the risk to that rehabilitation.’58 With regard to the risk of re- offending, despite the brain injury, some risk factors (dynamic risk areas) such as deficits in sexual- and self-regulation, negative emotional responses, and antisocial personality constraints were still present. In addition, the offender’s criminal record added to the concerns about recidivism. Despite that, the court concluded that due to offender’s amnesia and the severity of his injuries, neither personal deterrence, nor the need for protection of the community was as significant as before he was assaulted.
55 Id. at [37] 56 Id. at [18]; [37] 57 R v O'Connor [2013] NSWDC 272 (aka John Coble) 58 Id. at [100] 120
Although the offender committed serious offences, and he sustained the brain injury after the crime, the court concluded that the sentence imposed upon the offender should be significantly mitigated:
Often sentences can be used as an indicator of the seriousness with which the court views the crime committed. Sometimes as here, particular and extraordinary considerations mean such a direct correlation is impossible. The victim should, as I said, never equate her injury with the punishment actually inflicted. For these reasons I feel it is appropriate to note that the sentence that I would have imposed had the [brain injury] not occurred would have been significantly more severe. The sentence I intend to impose is much shorter and the non-parole period considerably shorter.59
As the brain injury occurred after the crime, it could not be relied upon to explain the criminal behaviour and this is the main difference between this case and the previous case (R v McCann). In this case, brain impairment could not be considered to be one of the causes of crime. Further, as a consequence of his subsequent injury, the factors that contributed to the crime in this case could not contribute to another offence. In other words, in these circumstances, not only was the brain impairment not one of the causes of the criminal behaviour, but it also interrupted the contribution of existing factors to future re-offending.
Unlike cases discussed in the previous section (in which neuroscience cut both ways: supporting a harsher and a more lenient sentence simultaneously), both cases in this section indicate that neuroscience can in some circumstances have a purely mitigating effect, reducing the significance of aggravating factors. These cases provide strong evidence against the claims that neuroscience operates as a double-edged sword.
Now that I have canvassed the diverse ways in which neuroscience may contribute to mitigating or aggravating the sentence, in the following section I will move on to discuss the results of my analysis of the case law relating to the double-edged sword claim.
3. A quantitative analysis of ‘Double-edged Sword’ claim in sentencing
As discussed, the double-edged sword claim asserts that introducing neuroscience may either aggravate or mitigate the sentence. That means that in some cases neuroscientific evidence contributes to a harsher punishment and in other cases, it contributes to a more lenient sentence.
59 Id. at [107] 121
The claim would be supported if the distribution of cases in which neuroscience supports an aggravated and mitigated sentence is such that defence lawyers are likely to be uncertain as to whether they should introduce this evidence and take the risk of increasing the sentence rather than withhold it. As such, if the number of cases where neuroscience points towards a more lenient sentence is considerably higher than cases where neuroscience tends towards a harsher punishment, or vice versa, the use of neuroscience in courts would not be consistent with the double-edged sword claim.
Following my discussions in Chapters 2 (literature review and background) and 4 (an introduction to empirical research), which explained that there are some uncertainties over what should be considered neuroscientific evidence, I have created three categories of cases to help evaluate the double-edged sword claim. The intention is to make sure that the results of my study are suitable to both those scholars who define neuroscientific evidence narrowly (i.e. those who only consider neuroscientific evidence to extend to imaging tools) and those who have a broader definition of neuroscientific evidence.60 Before outlining the categories, I will briefly review the different types of neuroscientific evidence that I noted in the previous Chapter.
There are three types of evidence produced through the use of neuroscientific tools: Type 1, imaging tools (for instance, PET, MRI and CT scan); Type 2, neuro tools (such as neuropsychological evidence); and Type 3 is non-neuro-tool evidence. This is evidence that is produced without the aid of the neuroscientific tools outlined above. For instance, psychiatric evidence that a history of alcohol abuse combined with behavioural evidence suggests some form of brain abnormality (the different types of neuroscientific evidence are shown in Figure 4-1).
Now, I will turn to three categories of cases. This categorising system is based on the types of neuroscientific evidence outlined above (referred to as Type 1, Type 2 and Type 3). The first category is made up of cases in which at least one ‘imaging’ evidence (Type 1) was presented to the court (61 cases). There are some cases in which both brain imaging evidence and other types of evidence (such as Type 2 or Type 3 evidence) were presented to the court, however due to methodological difficulties and uncertainties, it was not possible to distinguish the effect that each type of evidence available to the court had on the sentence. The second category contains cases where at least one instance of either imaging evidence (Type 1) or neuro-tool evidence (Type 2), or both were introduced to the court (160 cases). The third category includes all types of evidence
60 As I have already discussed in the methodology chapter, this study is mainly interested in analysing cases that involve imaging and non-imaging tools, only a limited number of cases that do not involve neuro-tool evidence are randomly selected. A comparison between these categories would give us a better understanding of the influence of different types of neuroscientific evidence on the sentence.
122
(Type 1, Type 2 and Type 3). That is, imaging evidence, neuro-tool evidence, and non-neuro-tool evidence (221 cases). As such, the first category suits those scholars who define the scope of neuroscience narrowly, (only including Type 1 evidence); the second category suits those who define both Type 1 and Type 2 evidence as neuroscientific; and the third category suits those scholars who use a broader definition of neuroscience, to include Type 1, Type 2 and Type 3 evidence.
I have designed another approach to categorising cases, which is based on the type of court involved: trial (lower) courts and appellate courts. According to my experience in analysing these cases, trial courts review various aspects of the case, while the scope of an appellate court’s review is confined to those matters relevant to the appeal. The fact that some matters relating to neuroscientific evidence may not have been discussed in appellate courts may skew the results of this study. By separating cases based on whether they were at the trial or appellate level, allowed me to draw conclusions about whether there were differences between decisions in these courts where neuroscientific evidence was introduced (the summary of these categories is shown in Figure 5-1 below).
Figure 5-1: Four categories of cases.
Results and findings
In Table 5-1, for each category of cases based on the type of evidence introduced in the case, I have noted whether neuroscientific evidence tends to aggravate or mitigate the sentence in
123
relation to any of the sentencing elements. Further, since cases where neuroscientific evidence contributes to both a more lenient and a harsher punishment support the double-edged sword claim, I have included those cases in the analysis. I also listed cases where neuroscience did not contribute to the sentence — i.e. neither mitigate, nor aggravate the sentence — because such cases are contrary to the premises of the double-edged sword claim — i.e. neuroscience either mitigates or aggravates the sentence.
The last two rows of the Table show the total number of cases in each category and the number of cases in each category where neuroscientific evidence contributed to the sentence. Accordingly, by comparing these two rows, it is possible to assess the proportion of cases in which neuroscience contributed to either a harsher or more lenient sentence.
Table 5-1 shows that in the first category (Type 1 evidence — at least one ‘imaging’ evidence) there are three out of the 61 total cases in which neuroscience supported an aggravated sentence, while there are 27 cases in which neuroscience contributed to a more lenient sentence. In three cases, neuroscientific evidence contributed to both. In the second category (Type 1 and/or Type 2 evidence — at least one ‘imaging’ and/or ‘neuro-tool’ evidence), neuroscientific evidence contributed to a harsher sentence in 12 out of a total of 160 cases. However, there were 84 cases in this category in which neuroscientific evidence supported a more lenient sentence and 11 cases where this evidence contributed to both sides. In the third category, where all types of evidence are considered, 13 cases involved neuroscience aggravating the sentence, 111 cases involved neuroscience mitigating the sentence, and 12 cases involved neuroscience contributing to both sides.
Table 5-1: Double-edged sword claim, comparison by the number of cases (categories are not mutually exclusive).
Overall, the results in the Table above suggest that the double-edged sword claim is not consistent with the use of neuroscience in practice. The number of cases where neuroscientific evidence contributed to a harsher sentence is much lower than the number of cases where this evidence contributed to a more lenient sentence. Further, in almost all cases in which neuroscience
124
contributed to the sentence it did so to achieve a more lenient sentence. There is only one case where this evidence supports a harsher sentence but did not support a more lenient sentence (Ngati v R,61 as previously discussed).
A better way to compare different categories and assess the double-edged sword claim is to calculate the ratio of cases in each row against the total number of cases in that category and convert the ratio into the form of a percentage. For instance, in the third category, the number of cases that aggravate the sentence (13) was divided by the total number of cases in that category (221) and then multiplied the result by 100 (5.9%) (the comparison between percentages within the set of cases with aggravating and mitigating factors contributing to sentence is illustrated in Table 5-2 and Figure 5-2 below).
Table 5-2: Double-edged sword claim, comparison of three categories of evidence by percentage (categories are not mutually exclusive).
Figure 5-2: Double-edged sword claim, comparison of three categories of evidence by percentage (categories are not mutually exclusive).
61 [2014] NSWCCA 125 125
From the Figure above, a comparison of the different categories more clearly reveals how frequently neuroscience contributed to either a more lenient or a harsher sentence and why the results in this study suggest that the double-edged sword claim is inconsistent with the use of neuroscience in NSW criminal courts. For instance, in category 3 (where all types of neuroscientific evidence were considered) the empirical results suggest that in about 50% of cases neuroscience did not contribute to the sentence. Within the remaining 50% or so of cases where neuroscientific evidence contributed to the sentence, in almost all these cases it supported a more lenient sentence, and only in about 6% of cases did neuroscience contribute to a harsher punishment. Further, the number of cases where neuroscience contributed to both aggravating and mitigating the sentence is slight, 5.4%. These results do not seem to suggest that neuroscientific evidence, where adduced to courts, either aggravates or mitigates the sentence in roughly equal cases – the double-edged sword claim.
As shown in Table 5-3, a comparison between trial and appellate courts suggests that the way neuroscientific evidence influenced sentencing decisions in lower and appellate courts is not substantially different. In both tiers of courts, neuroscientific evidence mitigates the sentence more frequently than it aggravates the sentence. Further, the number of cases where neuroscience aggravates and mitigates the evidence simultaneously is similar in both court tiers. These findings are clearer in Table 5-4 and Figure 5-3, where these statistics are compared across the lower and appellate courts. Of note, is that in many appeal cases, the legal issue does not involve neuroscientific evidence, and this evidence is only mentioned where the appellate court reviews the case facts concluded by the lower court. Because of the fact that some matters related to neuroscientific evidence may not have been discussed in appellate courts, this analysis was carried out to ensure that this issue did not skew the results of this study with respect to the double-edged sword claim.
Table 5-3: Contribution of neuroscience to sentence: trial (lower) vs Appeal courts.
Table 5-4: Contribution of neuroscience to sentence: trial (lower) vs Appeal courts, comparison by percentage.
126
Figure 5-3: Contribution of neuroscience to sentence: trial (lower) vs Appeal courts, comparison by percentage.
Overall, the findings in this section suggests that the use of neuroscience in courts is not consistent with the double-edged sword claim and this conclusion does not differ based on the different categories of evidence (i.e. case that only brain imaging evidence was introduced or all cases regardless of types neuroscientific evidence presented to the court), nor it is influenced by whether sentencing or appellate court.
Unclear cases
A note of caution is due in this analysis. In some cases, it was unclear whether neuroscience contributed to sentencing and it is arguable that if it was possible to make this determination, the conclusion regarding the double-edged sword claim could be different.
Across all 221 cases (i.e. category 3), the number of cases where it was unclear whether neuroscience contributed to mitigating the sentence is 19, while there are only six cases in which it was unclear whether neuroscience supported a harsher punishment. Assuming that neuroscience
127
in these cases contributed to the sentence, there would be 130 cases where neuroscience supported a more lenient sentence and 19 cases where neuroscience supported a harsher punishment, that is, 59% mitigating, and 8.6% aggravating. These figures do not change the conclusion I have made about the double-edged sword claim. Even if I assumed that in all unclear cases neuroscience did not contribute to mitigating the sentence, however, it contributed to a harsher punishment, this would not change my conclusion as to the double-edged sword claim (i.e. 54% mitigating vs 8.6% aggravating). These results are similar for category 1 and category 2 cases.
Overall, this study suggests that in the majority of cases neuroscience supports a more lenient sentence and therefore, the use of neuroscience in courts is not consistent with the double-edged sword claim. These findings suggest that it is in the benefit of the defendant to adduce the neuroscientific evidence. Despite that, there are other matters that may change this conclusion made about the double-edged sword claim.
4. Double-edged sword effect, other considerations
There are three issues that deserve consideration with respect to the double-edged sword claim and whether defence lawyers still need to be cautious about the risk of adducing neuroscientific evidence. First, it is plausible that defence lawyers need to consider not just how frequently neuroscience supports a harsher or more lenient punishment, but also the weight with which neuroscience may contribute to a longer sentence, in designing their strategies. It is important to consider whether that means defence lawyers should still be cautious about the outcome of adducing neuroscientific evidence. Second, it is possible that in cases where there is a possibility that neuroscience may contribute to a harsher punishment defence counsel withhold this evidence and judgements are therefore biased towards cases in which neuroscience has contributed to a more lenient sentence. Third, there are two pieces of legislation that may further influence the defence lawyers’ consideration of adducing neuroscientific evidence. In this section I will discuss these three issues.
Sentencing factors: weight and limits
I noted that the influence of the double-edged sword claim in practice is that defence lawyers may be uncertain as to whether they should introduce this evidence and take the risk of increasing the sentence, or whether they should instead withhold it. From a defence lawyer’s perspective, it should be considered that it is not only important to consider how frequently neuroscience supports either a harsher or a more lenient punishment, but also which sentencing factors
128
contribute to aggravating or mitigating the sentence. This is because some sentencing factors are given more weight and are considered more influential during sentencing. Further, there are limits as to how much some sentencing factor may influence the sentence.
In Chapter 3 (legal background), I discussed the complexity of sentencing and the difficulty of estimating the extent to which neuroscientific evidence influences the sentence. Accordingly, it is not possible for this research to fully ascertain the extent to which mitigating and aggravating factors influence the sentence. However, I will discuss what sentencing factors according to law potentially have the most weight in sentencing and to what extent they may change the severity of the sentence.
In determining an appropriate sentence, an offender’s punishment should not be disproportionate to the seriousness of the offence. A proportional sentence is determined by factors that objectively affect the seriousness of the offence, i.e. those factors that are associated with the offence and not with a particular offender. For instance, when determining the seriousness of an offence, courts assess the gravity of the crime in order to place it on a scale, or spectrum, of seriousness and thus determine the proportional punishment. Based on the maximum penalties for particular categories of offence (as defined by statute), the most severe punishment is reserved for the worst (most serious) kinds of offending. Courts may determine the relative seriousness of a specific offence by comparing it to the hypothetical worst possible offending and apportioning punishment accordingly.62 Notably, these objective factors do not determine a specific sentence, but rather a range of proportionate sentences.63 Other factors that the court must consider when determining where a sentence should sit within this proportionate range of punishment are known as ‘subjective factors’, i.e. factors that are specific to the offender (e.g. the offender’s prospects of rehabilitation).
Nonetheless, if an offender’s specific factors are causally relevant to the commission of the crime, it may be relevant to the proportionality of the punishment. Where an offender’s mental illness (e.g. due to brain impairment) at the time of the offence is causally related to the commission of the crime, it may reduce the offender’s moral culpability.64 Moral culpability is an objective factor
62 Odgers. 63 BW v R [2011] NSWCCA 176 at [70] 64 In Kennedy v R the court explained that ‘[s]ome of the relevant circumstances which can be said ‘objectively’ to affect the ‘seriousness’ of the offence will be personal to the offender at the time of the offence but become relevant because of their causal connection with its commission. This would extend to matters [including] … mental illness, or intellectual disability, where that is causally related to the commission of the offence, in so far as the offender’s capacity to reason, or to appreciate fully the rightness or wrongness of a particular act, or to exercise appropriate powers of control has been affected … Such matters can be classified as circumstances of the offence and not merely circumstances of the offender that might go to the appropriate level of punishment’. The court then added that such subjective 129
that is considered in assessing the seriousness of the offence. That means an offender’s mental condition, by reducing the moral culpability of the offender, may in turn, reduce the seriousness of the offence, requiring a less proportionate punishment.65
Concurrently, the same mental illness might be associated with the offender’s prospects of rehabilitation, which is a subjective factor.66 As such, prospects of rehabilitation will not contribute to determining the proportional punishment and courts consider this sentencing element in reaching an appropriate sentence within the proportionate sentencing range. Similar to rehabilitation, the protection of society67 is a subjective sentencing factor and cannot extend the sentence beyond what is proportionate to the seriousness of the offence.68 In Veen v R (No 2)69 the High Court stated that:
The principal for proportionality is now firmly established in this country. It was the unanimous view of the Court in Veen (No 1) that a sentence should not be increased beyond what is proportionate to the crime in order merely to extend the period of protection of society from the risk of recidivism on the part of the offender.70
Accordingly, subjective factors such as the protection of society cannot ‘extend’ the sentence by reason of preventive detention (protecting the society) beyond what would have been imposed had the mental condition not existed and the protection of society can only be considered alongside other sentencing factors to reach the appropriate sentence within the range of proportionate sentences.71 However, objective factors such as the moral culpability of an offender
factors include ‘the seriousness of the physical attack. It also includes the factors which ‘may explain why [the offence] was committed’ (citations are omitted) (2008) 181 A Crim R 185; [2008] NSWCCA 21 at [38] and [39]. 65 Hodgson J noted that ‘[i]n my understanding, considerations of retribution direct attention to what the offender deserves; and in my opinion, where emotional immaturity or a young person's less-than-fully- developed capacity to control impulsive behaviour contributes to the offending, this may be seen as mitigating culpability and thus as reducing what is suggested by considerations of retribution’ (Rothman J agreeing; Johnson J dissenting in the principal judgment). BP [2010] NSWCCA 159 at [4] quoted in Odgers. 66 Odgers. 67 Section 3A Crimes (Sentencing Procedure) Act 1999, NSW 68 Veen v R (No. 2) p 473: ‘an extension merely by way of preventive detention’ is not permitted while, ‘… an exercise of the sentencing discretion having regard to the protection of society among other factors ...’ is allowed. In other words ‘a sentence may not be “extended” to protect society but society’s protection may be taken into account in determining the appropriate sentence’, p. 277. 69 [1988] HCA 14; 164 CLR 465, 70 Id. at [472] 71 Majority in Veen v R (No.2) at [9] per Mason C.J., Brennan, Dawson and Toohey JJ, ‘It is one thing to say that the principle of proportionality precludes the imposition of a sentence extended beyond what is appropriate to the crime merely to protect society; it is another thing to say that the protection of society is not a material factor in fixing an appropriate sentence. The distinction in principle is clear between an extension merely by way of preventative detention, which is impermissible, and an exercise of a 130
can result in a sentence that would otherwise be outside the lower limit of the range of sentences proportionate to the gravity of the offence.
Now that we know what types of sentencing factors generally have more weight in sentencing – i.e. objective factors – in the following section I will discuss which sentencing factors neuroscientific evidence is associated with based on the results of the case analysis, as well as which of them are objective and subjective.
Sentencing factors associated with neuroscience and their frequency
Figure 5-4 illustrates sentencing factors that neuroscience contributes to and how frequently. In some cases, neuroscience contributed to more than one sentencing factor and consequently, those cases are represented more than once. I have divided the sentencing factors into two sets of groups. The first set divides the sentencing factors into those that are aggravating and those that are mitigating. As noted previously, that does not mean these factors are essentially aggravating or mitigating. For instance, adducing neuroscience may result in the court giving greater consideration to specific deterrence, therefore resulting in a longer sentence. In these circumstances, specific deterrence is an aggravating factor. Neuroscientific evidence may also reduce the consideration given to specific deterrence where it is not possible to deter an offender with mental illness, therefore resulting in a shorter sentence. In this case, specific deterrence is a mitigating factor. The second set divides the sentencing factors into objective and subjective categories.
As you can see in Figure 5-4, with respect to the contribution of neuroscience to sentencing, most of the relevant sentencing factors are mitigating, while there are only two relevant aggravating factors. The most common relevant elements that contributed to sentencing are moral culpability and general deterrence. The next most common factor is custodial hardship. The specific importance of this element is that it does not require the brain impairment to exist before the offending or to be causally connected to the commission of the crime (see O'Connor, discussed above). The four cases in the last category appearing under the heading of ‘general mitigation’ are those cases where it was noted in the judgement (mostly appellate courts) that neuroscientific evidence mitigates the sentence, but it was not clear which sentencing element was relevant.
On the aggravating side, there are 12 cases where protection of the community was relevant, while
sentencing discretion having regard to the protection of society among other factors which is permissible’. 131
in three cases specific deterrence was invoked.
Figure 5-4: Contribution of neuroscientific evidence to sentencing elements. Categories are not mutually exclusive.
This Figure also illustrates that both aggravating factors are subjective factors, while there are five mitigating sentencing factors that are objective, one of which (moral culpability) is most commonly relevant to neuroscience. As discussed, it can be said that, in general, objective factors have a greater influence on the sentence than subjective factors. Since both aggravating factors are subjective, they cannot increase the sentence beyond the proportionate punishment. However, since the mitigating factors include objective factors, neuroscientific evidence is capable of reducing the sentence in relation to what may otherwise be considered proportionate to the severity of the offence. Therefore, considering the weight and limits of sentencing factors, adducing neuroscientific evidence appears to be a reasonable strategy for defence lawyers.
Cases biased towards mitigation
In Chapter 3 (legal background), I discussed that in an adversarial system, defence lawyers are obliged to advise their clients on any matter that may benefit them (e.g. reduce their punishment). Contrastingly, prosecutors should present any relevant materials to the criminal procedure,
132
whether they support or damage the defendant’s argument.72 There is a possibility that defence lawyers withhold neuroscientific evidence to avoid any potential risk that the contribution of the evidence would be to aggravate the sentence. On the other hand, prosecutors should provide neuroscientific evidence to the court regardless of the impact that it may have on the sentence — that it may mitigate or aggravate the sentence. This issue may explain the reason why there are many judgements where neuroscientific evidence only points towards a more lenient sentence, which may change the conclusion made about the double-edged sword claim.
In Sumpton v R,73 the defence decided not to tender an expert report regarding the offender’s brain impairment. Sumpton was found guilty of two counts of arson and one count of murder. The offender struck the deceased with a statue before stabbing her 24 times and then setting the house on fire. He then claimed that he had discovered the fire and informed neighbours about it.
One ground of appeal alleged a miscarriage of justice on the basis that the judge was unable to consider the psychiatrist’s report during sentencing because the applicant’s previous representatives had decided not to tender this evidence.74 It was submitted that the expert report indicated that the applicant potentially suffered from alcohol-related brain damage. It was noted that the brain damage affected the frontal lobe of the brain, which is associated with judgement. It was also noted that such brain damage would reduce tolerance to the disinhibiting effect of alcohol. Finally, it was submitted that according to the opinions expressed in this report, the sentencing judge was ‘dealing with an offender with brain damage which may reasonably be supposed to have impacted on the offence committed’.75 The defence argued that the trial court would have considered this matter during sentencing if it had been submitted to the court.
The Crown argued that there was no miscarriage of justice since the ‘applicant’s then legal representatives had a number of legitimate concerns arising from other observations made by [the expert] in his report, and had made a forensic decision, in accordance with the applicant’s instructions, not to tender the report on sentence’.76 The Crown also submitted that as the report was available at the time of sentencing, it was not fresh evidence.
72 Odgers; Judicial Commission of New South Wales, ‘Sentencing Bench Book’, in Judicial Commission of New South Wales,
The appellate court stated that the psychiatrist did not comment on the causal relationship between the brain injury and the offences committed by the applicant. Instead, the expert report stated that if the applicant did not stop drinking alcohol there would be a risk of re-offending. The court found that this was one of the main reasons why the report had not been submitted during sentencing. It was concluded that no miscarriage of justice was established and the appeal was dismissed.
Cases similar to Sumpton, where the defence does not adduce neuroscientific evidence, cast doubt on the conclusion that the use of neuroscience in NSW sentencing courts is not consistent with the double-edged sword claim. Such cases do not indicate the decision made by the defence to withhold the relevant evidence unless they happen to be appealed on that ground. It is, however, unclear how many cases similar to Sumpton exist, and to what extent this issue may throw into question the conclusion drawn about the double-edged sword claim in this study.
If we imagine that there are 60 cases in which neuroscience has contributed to the protection of society (five times more than the current figure), and neuroscience has not contributed to any of the mitigating factors outlined in the Figure 5-4 except moral culpability, in 60 cases (the current figure), does that mean lawyers should not take the risk of adducing neuroscientific evidence? In this situation, the double-edged sword claim would be consistent with the use of neuroscience in courts.
Since the moral culpability of an offender is an objective sentencing factor and has more weight in sentencing than the protection of society (which is a subjective sentencing factor), taking the risk of adducing neuroscientific evidence appears to be a more appropriate decision for the defence. In particular, in cases where the circumstances of the case suggest that the offender’s mental condition may support both a shorter and a longer sentence simultaneously, it is expected that the objective mitigating factors will outweigh the subjective aggravating factors. To support this conclusion, I refer to Odgers claim concerning cases where evidence of mental abnormality may support a more lenient sentence and a harsher punishment simultaneously: ‘more commonly, a mental abnormality will result in a more lenient sentence than would otherwise have been imposed.’77
Generally, it appears that adducing neuroscientific evidence is beneficial to the defendant, even if we assume that the number of cases where neuroscience contributes to the protection of society (an aggravating factor) is five times greater than what this study concluded. Of course, this study
77 Odgers, p. 278. 134
does not suggest lawyers should unconditionally adduce neuroscientific evidence, and the circumstances of the case can be important. For instance, there are two Acts that lawyers may need to consider regarding the contribution of neuroscience to a harsher punishment: Section 61(1) of the Crimes (Sentencing Procedure) Act 1999 (NSW) and the Crimes (High Risk Offenders) Act 2006 (NSW). I will briefly discuss them in the following section.
Legislations and risk of a longer sentence
Section 61(1) of the Crimes (Sentencing Procedure) Act 1999 (NSW) provides that mandatory life sentences can be given for murder:
if the court is satisfied that the level of culpability in the commission of the offence is so extreme that the community interest in retribution, punishment, community protection and deterrence can only be met through the imposition of that sentence.
Similar to section 61(1) for murder cases, the same order may be imposed for serious heroin or cocaine trafficking offences provided that additional criteria are met.78 For example, the offence must involve ‘a high degree of planning and organisation.’79
Where the criteria in section 61 are met, the court must impose imprisonment on the offender for the term of their natural life (with no non-parole or parole period). That said, for murder cases section 61 (3) also provides that the offender’s subjective circumstances may lead to a more lenient sentence than that of life imprisonment.80 Conversely, in some cases, the offender’s level of culpability (objective circumstances) is so extreme that the court may disregard the offender’s subjective circumstances (e.g. their prospects of rehabilitation).81
78 Section 61 (2) Crimes (Sentencing Procedure) Act 1999 (NSW) 79 Section 61 (2) (a) (i) Crimes (Sentencing Procedure) Act 1999 (NSW) According to Section 61 (7), ‘“serious heroin or cocaine trafficking offence” means an offence under section 25 (2) or (2A) of the Drug Misuse and Trafficking Act 1985 involving heroin or cocaine, and being an offence to which section 33 (subsection (2) excepted) of that Act applies’ (emphasis is original). 80 S 61(3) Crimes (Sentencing Procedure) Act 1999 (NSW) provides that, ‘Nothing in subsection (1) affects section 21 (1).’ S 21 of this Act, general power to reduce penalties, states that ‘(1) If by any provision of an Act an offender is made liable to imprisonment for life, a court may nevertheless impose a sentence of imprisonment for a specified term.’ R v Merritt (2004) at [36] cited in Judicial Commission of New South Wales. Also see, Odgers, pp. 449 and 450. There is a tension between applicability of s 21(1) that permits discretion to reduce the sentence while s 61(1) requires a mandatory imprisonment for life if the criteria set in that section are met. This tension seems to be addressed later in favouring of s 21(1). R v Harris [2000] 50 NSWLR 409 at [93]; R v Merritt [2004] at [36] cited in Judicial Commission of New South Wales. 81 R v Miles at [203]; R v Ngo (No 3) [2001] 125 A Crim R 495 at [42] cited in Judicial Commission of New South Wales. Moral culpability, or the degree to which an individual should be blamed for what he or she has done can be altered by a number of factors. It may be reduced because of brain impairment, or because the offender’s experience of social deprivation has contributed to their criminality. Moral culpability, however, may instead be increased due to aggravating factors, such as where an offender has committed sexual assault in the company of others, or where they are aware of the degree of harm that the 135
The court does not need to be satisfied that community interest has been met in relation to retribution, punishment, community protection, and deterrence, individually. Rather the court is concerned with the combined effect of these four considerations, which may, in some circumstances, result in an order for life imprisonment.82
Since one of the components of section 61(1) of the Crimes (Sentencing Procedure) Act 1999 (NSW) is protection of the society, considering other circumstances of the case, the defendant may need to evaluate whether it is worth submitting neuroscientific evidence for the purpose of a more lenient sentence, concurrently taking the risk life imprisonment, or withhold neuroscientific evidence to avoid the risk that the evidence leads to an order under the Act.
While life imprisonment is the worst-case scenario for the offender, there is also the possibility that the court may accept the connection between the brain impairment and the offending but still increase the offender’s term of imprisonment or otherwise restrict their liberty. According to the Crimes (High Risk Offenders) Act 2006 (NSW),83 the Supreme Court may impose post-sentence continuing (preventive) detention and ongoing supervision on high risk sex and violent offenders.84 The purpose of the Act is to ‘ensure the safety and protection of the community’ from high risk sex and violent offenders by preventing further serious offences following their release, as well as to encourage rehabilitation.85 As noted, sex-related and violent offences must be serious. For instance, the Act defines serious violence offence as
offence may cause. Certain circumstances can result in the most extreme level of moral culpability, such as the murder and mutilation of a child by a relative. See, Bugmy v R [2013] 249 CLR 571 at [40]; Section 52A Crimes Act 1900 NSW; the guideline judgment in R v Whyte (2002) 55 NSWLR 252, cited in sentencing benchmark; R v Way [2004] 60 NSWLR at [107]; R v Lewis [2001] NSWCCA 448 at [67]; R v Coulter [2005] NSWSC 101 cited in Judicial Commission of New South Wales. 82 R v Merritt [2004] 59 NSWLR 557 at [52] cited in Judicial Commission of New South Wales. Also see Odgers, pp. 449 and 450. However, this does mean that the absence of one or more of the four considerations would make it more difficult to justify such a severe punishment. For instance, if an offender has been treated and is no longer a risk to society, community protection is not an applicable justification for punishment and the court must be satisfied that the other three considerations are of such interest to the community that the imposition of life imprisonment is required. 83 There were amendments in 2013 and the name of the act from Crimes (Serious Sex Offenders) Act 2006 (NSW) was changed to Crimes (High Risk Offenders) Act 2006 (NSW). Further amendments occurred in 2014 and 2016. The latest version of the act Crimes (High Risk Offenders) Amendment Act 2017 No 54 is available at https://legislation.nsw.gov.au/#/view/act/2017/54 (accessed 30 December 2017 at 12:49). 84 This order can be interim or final. For instance, see, SS 10A and 18A Crimes (High Risk Offenders) Act 2006 (NSW). Also, Crimes (High Risk Offenders) Amendment Act 2014 (NSW) added emergency detention order to be imposed against offenders on interim or extended supervision order. For instance, see, 18CB Crimes (High Risk Offenders) Act 2006 (NSW). Also see T Tulich, ‘Post-Sentence Preventative Detention and Extended Supervision of High Risk Offenders in New South Wales’, in UNSWLJ, vol. 38, 2015, 823–853. 85 S 3 Crimes (High Risk Offenders) Act 2006 (NSW) This Act has been criticized for a variety of reasons such as being against Human Rights and Legal Principles. However, these arguments are not relevant to this chapter and I would not discuss it any 136
engaging in conduct that causes the death of another person or grievous bodily harm to another person, with the intention of causing, or while being reckless as to causing, the death of another person or grievous or actual bodily harm to another person. 86
An extended supervision order may require the offender ‘to make periodic reports to a corrective services officer’, ‘to participate in treatment and rehabilitation programs’, or ‘to wear electronic monitoring equipment’,87 while a continuing detention order obliges the offender to remain in custody after the sentence has ended or where the offender cannot be adequately supervised under an extended supervision order.88 These orders will be made if ‘the Supreme Court is satisfied to a high degree of probability that the offender poses an unacceptable risk of committing another serious offence if not kept’ under supervision or in detention under the order.89
In one of the cases,90 the judge warns the offender about the application of this Act. In this case, the offender was found guilty of murder. While he did not intend to murder the victim, he did ‘inflict really serious bodily harm’ on the victim in a savage unarmed assault.91 The offender had an extended criminal record that included violent offences against women.
After considering the offender’s history and the result of the neuropsychological test, a neuropsychologist concluded that the offender’s ‘behaviour and type of cognitive deficits are consistent with those seen in patients with damage to [a region of the brain that] plays a key role in impulse control.’92 He continued, explaining that ‘[d]amage to this region can result in disinhibition and impulsivity with such associated behaviour problems as aggressive outbursts and sexual promiscuity.’93
There was no cure for the offender’s brain damage, instead, it was suggested that participating in intensive rehabilitation programs may help, otherwise he would remain a danger to the society in particular to women. Based on the offender’s criminal history, previous failure to pursue
further. For more information see, L Bartels, J Walvisch & K Richards, ‘More, longer, tougher… or is it finally time for a different approach to the post-sentence management of sex offenders in Australia?’, in Criminal Law Journal, vol. 43, 2019, 41–57; B McSherry, ‘Throwing Away the Key: The Ethics of Risk Assessment for Preventive Detention Schemes: RG Myers Memorial Lecture 2013’, in Psychiatry, Psychology and Law, vol. 21, 2014, 779–790; G Williams, ‘The legal assault on Australian democracy’, in QUT L. Rev., vol. 16, 2016, 19. 86 5A(1)(a) Crimes (High Risk Offenders) Act 2006 (NSW) 87 S 11 Crimes (High Risk Offenders) Act 2006 (NSW). 88 See ss 13B and 18 Crimes (High Risk Offenders) Act 2006 (NSW). 89 Crimes (High Risk Offenders) Act 2006 (NSW) 5B and 5C 90 R v Ray [2013] NSWSC 767 91 Id. at [2] 92 Id. at [69] 93 Id. at [69] 137
rehabilitation, and lack of respect for court orders, the court concluded that the offender did not have good prospects for rehabilitation.
Although the neuropsychologist had some reservations about the brain injury, as there was no brain scan to confirm the results, the court ‘generously’ reduced the offender’s moral culpability. The most interesting aspect of this case is that the judge, at the end of the judgement, stated that:
I am required to warn the offender that the Crimes (High Risk Offenders) Act 2006 (NSW) applies to him because of the nature of the offence … this means that the State may make an application at the end of his sentence for him to be made the subject of a continuing detention order or an extended supervision order for up to 5 years and that the State may continue to make applications for further such orders … it would be in his interests to engage with rehabilitation programs whilst serving his sentence. If he does not, there is the potential for him to remain in custody for the rest of his life.94
One important matter to discuss about Crimes (High Risk Offenders) Act 2006 (NSW) is that if orders from this Act can be considered sentence and relevant to the double-edged sword claim. For instance, a possible argument is that since an order from this Act is not made at the time of sentencing, but at the end of the custodial sentence (or at the end of an existing detention order or extended supervision)95 any order under this Act cannot be considered as ‘sentence’ or part of the original sentence and involve in evaluation of double-edged sword claim (for more discussion see footnote).96
Regardless, the uncertain application of orders under this Act would not change the results of this study with respect to the double-edged sword claim. This is because there are very few cases in this study that refer to the Crimes (High Risk Offenders) Act 2006 (NSW), and these cases are inconclusive. For instance, in most of these cases the results of neuro-examinations did not
94 Id. at [80] 95 ‘An application for an extended supervision order against an offender may not be made until the last 9 months of the offender’s current custody or supervision’ Crimes (High Risk Offenders) Act 2006 (NSW) s 6 (1). ‘An application for a continuing detention order in respect of a detained offender may not be made more than 9 months before: (a) the end of the offender’s total sentence, or (b) the expiry of the existing continuing detention order, as appropriate’ Crimes (High Risk Offenders) Act 2006 (NSW) s 13B (3). 96 In cases like this, it is not clear what will happen at the end of the offender’s sentence. It could be argued that the Crimes (High Risk Offenders) Act 2006 (NSW) is most likely to be applied in cases where the offender is a danger to the society and that is an aggravating factor already considered in the sentence, therefore, a subsequent order under this Act would not change anything in relation to the double-edged sword claim. There are other reasons against this consideration: while an extended supervision order restrains the offender’s liberty, it is not clear if it is a type of ‘punishment’. Another possible argument is that in cases that the offender may receive an order under this Act, we should wait until the completion of their sentence; otherwise conclusions made regarding double-edged sword claim at the time of sentencing are not credible because a later order upon this Act may change that conclusion. 138 indicate brain abnormality,97 or the judge made an interim order rather than a final decision and asked for further mental examinations.98
This is a matter that requires further discussion, particularly because prior to 2014 this Act only applied to sex offenders, but following amendments in 2014 the Act’s title was changed from the Crimes (Serious Sex Offenders) Act to the Crimes (High Risk Offenders) Act 2006 (NSW) and was amended to include high risk violent offenders.99 In short, this means that there may be more cases in the future where neuroscientific evidence contributes to an order under this Act. This study, however, collected cases published prior to the end of 2016 and therefore cannot draw any conclusions about the number of cases to which this Act may apply. Regardless, since the findings of this study may be used in practice, it remains important to discuss the potential long-term outcome of adducing neuroscientific evidence.
In this section, I discussed three considerations that may have an influence on the findings made concerning the double-edged sword claim. First, I discussed the extent to which some sentencing factors may influence the sentence and whether that should affect the decisions made by defence lawyers as to the submission of neuroscientific evidence. Next, I went on to consider the extent to which the dataset of this study may be biased towards cases in which neuroscience supports a more lenient sentence, and whether if there were the same number of cases where neuroscience contributed to mitigation and aggravation of the sentence that would change the conclusions drawn. For both of these considerations, I concluded that it is still likely to be beneficial for the defence to submit neuroscientific evidence. Finally, I referred to two pieces of legislation that may make adducing neuroscientific evidence risky, and noted that lawyers should consider whether the circumstances of the case may lead to an order under these Acts.
In the last section of this Chapter, I will compare the results of this study with the empirical neurolaw studies in other jurisdiction.
5. Double-edged sword claim in other jurisdictions
97 See State of New South Wales v Thomas (Final) [2009] NSWSC 1410 98 State of New South Wales v Hill [2014] NSWSC 1803 (28 November 2014); State of New South Wales v Scott [2013] NSWSC 1834 99 Tulich, 823–853. 139
The double-edged sword claim has been analysed in three of the empirical studies in other countries: the Netherlands,100 Canada101 and the US.102 Before I make a comparison between the result of this study and the neurolaw studies in other jurisdictions, I should note that there are some methodological and legal differences between these studies. For instance, in the Netherlands study, the research encompassed both neuroscientific and behavioural genetic evidence.103 The number of cases analysed in the Canadian study was 133, while in the US study 553 cases were considered. Other differences include the criminal procedure of these jurisdictions, and whether the trial phase was considered in the study.
The U.S. study, ‘The Myth of the Double-Edged Sword: An Empirical Study of Neuroscience Evidence in Criminal Cases’104 is undoubtedly the most prominent empirical neurolaw study in evaluating the double-edged sword claim. As is clear from the title, the main purpose of the study was to examine the double-edged sword claim. The results indicated that this claim is not consistent with the US criminal justice system and is instead a ‘myth’. The study concluded that neuroscientific evidence is usually used to mitigate the sentence and is rarely used as evidence for the offender’s future dangerousness. These results are in accordance with the Netherlands study, which indicated that neuroscientific evidence point towards a more lenient sentence in most cases and was only considered as relevant to measuring an individual’s future dangerousness in some cases.105
By comparison, neuroscientific evidence is more likely to have an aggravating influence on the sentence in Canada.106 In the Canadian empirical study, the results showed that while neuroscientific evidence was a diminishing factor in 35 cases, in 25 cases it had a mixed influence – both aggravating and mitigating. Comparison with this study indicates that in Canada, the possibility of witnessing the double-edged sword effect in courts is much higher than in NSW criminal courts.
Overall, a comparison between the empirical studies in other countries and this study indicates that the double-edged sword effect of neuroscience is inconsistent with practice in Australia, the US and Netherlands’ jurisdictions, whereas this claim is comparatively more consistent with practice in Canada.
100 de Kogel and Westgeest, 580–605. 101 Chandler, 550–579. 102 Denno, 493–551. 103 That said, majority of cases involve neuroscientific evidence. 104 Denno, 493–551. 105 de Kogel and Westgeest, 580–605. 106 Chandler, 550–579. 140
Conclusion
The purpose of this Chapter was to examine whether the double-edged sword claim – that neuroscientific evidence may just as likely contribute to either aggravation or mitigation of the sentence – is consistent with court practice in NSW. The results suggest that neuroscientific evidence contributed to the sentence (112 cases) in about half of the cases. Of these, the number of cases where neuroscience tends towards a more lenient sentence is significantly more than the number of cases where neuroscience tends towards a harsher punishment. These findings suggest that the use of neuroscience in NSW courts is not consistent with the double-edged sword claim.
A similar conclusion was reached when cases were examined based on the type of court involved (i.e. sentencing courts and appellate courts) and where results were differentiated based on the types of neuroscientific evidence adduced (e.g. cases that involve only brain imaging tools versus cases involving any type of neuroscientific evidence). Similarly to this study, the results of empirical neurolaw studies in the US and the Netherlands suggest that the double-edged sword claim is inconsistent with the use of neuroscience in those jurisdictions. Notably, this claim appears to be comparatively more consistent with practice in Canada.
Overall, the practical application of the conclusions in this Chapter is that defendants should raise neuroscientific evidence to mitigate sentencing. This may be particularly efficacious, given the previous discussions on NSW criminal laws, that aggravating factors cannot justify a sentence that is harsher than what is proportional to the seriousness of the offence, whereas mitigating factors may justify a more lenient sentence than what is proportional. However, it is of significance that the defence lawyers should consider the possibility of the introduction of the Crimes (High Risk Offenders) Act 2006 (NSW), which an individual with an untreated brain impairment that indicates a high risk of re-offending may face an additional custodial sentence or an extended supervision order. Also, according to section 61(1) of the Crimes (Sentencing Procedure) Act 1999 (NSW) one of the reasons that an offender may receive a sentence of imprisonment for life is proof of future dangerousness that may raise by reason of brain damage.
141
Chapter 6 The Gap Between Neuroscience and the Law’s Language
142
Introduction
This Chapter discusses the challenges that arise due to the gap between the languages used by the law and neuroscience to define crimes, criminal responsibility, and criminal defences (e.g. defence of mental illness). Stephen Morse argues that neuroscience and law use different languages.1 Neuroscience explains how neurons communicate and how different areas of the brain are responsible for different functions (e.g. cognitive abilities), while the categories used by the law to define crimes and defences are concerned with pre-scientific concepts such as the intention and knowledge of individuals, or more generally the mental state of people. Morse claims that for neuroscience to be compatible with a legal system, it ought to be translated (explained) into legal language, i.e. instead of asking whether a brain dysfunction causes criminal behaviour, it should be asked whether a brain dysfunction affects mental state, which is associated with legal criteria of responsibility (e.g. as capacity to control behaviour or understanding of the wrongfulness of an act). Otherwise, according to Morse, neuroscience is not relevant to legal principles, and this can result in misconceptions and misunderstandings by judicial officers.2 For instance, where a brain tumour is discovered, some judicial officers may find the mere existence of the tumour relevant to an offender’s criminal responsibility. However, strictly speaking, the brain tumour may not interfere with mental states that are considered relevant in determining the sentence, such capacity to conduct intentional behaviour.
The purpose of this Chapter is to explore whether neuroscientific evidence, where discussed in court, is actually translated into legal language (as Morse claims it ought to be). The first part of the Chapter discusses Morse’s claim in detail, drawing on the differences between legal and neuroscientific languages to explain why according to Morse neuroscience needs to be translated into legal language. The second part of the Chapter moves on to review the difficulties and limitations in examining the claim — whether neuroscience is actually translated into legal language — and how this Chapter is going to address those issues. The third part of the Chapter outlines the findings of some analysis leading to the conclusion that in the majority of judgements, where neuroscience contributes to the determination of an appropriate sentence, it is translated into legal language—in accordance with Morse’s claim. The last part of the Chapter suggests courts use neuroscience and legal concepts interchangeably, so the gap between neuroscience and legal language is not as wide as it is suggested.
1 S Morse, ‘Lost in Translation?: An Essay on Law and Neuroscience’ in Law and Neuroscience, Current Legal Issues, M Freeman (ed), Oxford University Press, 2011, pp. 529–562. 2 S Morse, ‘Lost in Translation?: An Essay on Law and Neuroscience’ in Law and Neuroscience, Current Legal Issues, M Freeman (ed), 2011, XIII, 529. 143
1. The purported gap between neuroscience and law
Morse argues that neuroscience is not generally useful and relevant to law until it is translated into legal language.3 He explains that law’s understanding of the person and behaviour is a folk psychological view—that is, ‘an explanatory causal theory that explains human behaviour in part using mental state variables, such as desires, beliefs, plans, and intentions’.4 For instance, from a folk psychological viewpoint, the reason someone reads and studies a neurolaw paper is that they want to expand their knowledge about the relation between neuroscience and law and they believe that reading that paper will be helpful in order to meet that goal. He goes on to examine the criteria of criminal responsibility (elements of offence), which he notes is also based on two mental states; a voluntary act, which is an intentional movement of the body, and a culpable mental state, e.g. purpose and knowledge which the offender has while committing the crime.5
Morse explains that unlike folk psychology, neuroscience is a mechanistic science of cause and effect. ‘Neurons, neural networks, and the connectome do not have reasons. They have no aspirations, no sense of past, present, and future’,6 while law and ethics are attributable to agents, a ‘self’, that is ‘capable of acting for reasons’.7 More specifically, law’s view of criminality is that ‘brains do not commit crime; people commit crime’.8
In summary, he notes that:
law's psychology must be a folk-psychological theory, a view of the person as a conscious-and potentially self-conscious-creature who forms and acts on intentions that are the product of the person's other mental states. We are the sort of creatures who can act for and respond to reason. The law treats persons generally as intentional creatures and not simply as mechanistic forces of nature.9
Morse argues that the structure and function of brains are not matters that the law is fundamentally concerned about. The existence of an abnormal biological variable may not be legally relevant to
3 Morse, pp. 529–562. 4 S Morse, ‘Neuroscience in Forensic Contexts: Ethical Concerns’ in Ethics Challenges in Forensic Psychiatry and Psychology Practice, EEH Griffith (ed), New York, Columbia University Press, 2018, pp. 132–158 (p. 134). 5 Morse, pp. 529–562. Morse’s Claim applies to all stages of criminal proceeding and it is not limited to the trial. 6 Morse, pp. 132–158 (p. 143). 7 S Morse, ‘Avoiding Irrational NeuroLaw Exuberance: A Plea for Neuromodesty’, in Law, Innovation and Technology, vol. 3, 2011, 209–228 (p. 840). 8 SJ Morse, ‘Brain overclaim syndrome and criminal responsibility: A diagnostic note’, in Ohio St. J. Crim. L., vol. 3, 2005, 397–412 (p. 397). 9 Morse, 209–228 (p. 840). 144
explaining the cause of a crime. For Morse, ‘law’s criteria are not biological’.10 In the case where a person with a brain tumour has paedophilic tendencies and subsequently commits sexual assault, this person is only legally responsible if it can be proven that they could understand that their behaviour was wrong, and was able to control their actions. As such, the brain tumour, regardless of its influence on the person, may not be legally relevant if it does not answer the legal question of whether the individual was rational and could control their behaviour.11
Further, Morse notes that the law is not concerned about what has caused the mental state, but simply what mental state the individual was in at the time of the crime.12 This means it is not important whether there is a biological variable that causes a crime, or any other factor, the law is concerned about whether the outcome of that abnormality (e.g. poor judgement) is relevant to the elements of an offence, i.e. the criminal law’s folk-psychological criteria (I will return to this point in Chapter 9, Neurolaw of the mechanical brain)
Similarly, David Hodgson, New South Wales Court of Appeal Judge and philosopher, recognises the problem of the gap between neuroscience and folk psychology. He explains that in law, categories of ‘belief, intention, consent, unsoundness of mind, knowing what one is doing, knowing that what one is doing is wrong, willed or voluntary action, ability to control one’s actions, loss of self-control, impairment of mental responsibility’13 are all pre-scientific, folk psychological concepts: ‘the language which describes them is non-scientific, and, except in the clearest cases, the existence or otherwise of the psychological state which they describe is not susceptible to scientific proof’.14 Law presumes a conscious agent that has the ability to freely make decisions, while neuroscience presumes scientific and physical cause and effect, and focuses on the brain’s function. Unlike the law, which considers the mind to be the entity that controls an individual’s intentional acts, neuroscience does not have a place for the mind.15 He adds that:
Apart from some areas of psychology and psychiatry which continue to use folk- psychological concepts, but which tend increasingly to be regarded as unscientific, neuroscience has no place for the concept of an holistic entity, the mind, which according to the law is supposed to be in control of a person's voluntary conduct.16
10 Morse, pp. 132–158 (p. 144). 11 Morse, pp. 132–158. 12 Morse, pp. 529–562. 13 D Hodgson, ‘Guilty mind or guilty brain? Criminal responsibility in the age of neuroscience’, in Australian Law Journal, vol. 74, 2000, 661–680 (p. 670). 14 Hodgson, 661–680 (p. 670). 15 Hodgson, 661–680. 16 Hodgson, 661–680 (p. 670). 145
Overall, Hodgson considers that human conduct can be described in two languages:
the language of neuroscience, dealing in scientific cause and effect, and the language of folk psychology, dealing in the beliefs, intentions, choices, and actions of persons. There is not at present any accepted overall theoretical framework that can make sense together of these two languages and thereby facilitate translation from one to the other.17
With respect to psychology and psychiatry (the psychological sciences), Morse notes that these fields will sometimes treat individuals mechanistically, but will also at times treat them as agents, and are thus to some extent folk psychological. He does not explain what exactly he means by the fact that psychological sciences treat people as mechanisms.18 In any event, he claims that in most cases, psychological sciences have a significantly more similar approach to that of folk psychology, compared to the purely mechanistic approach of neuroscience (i.e. ‘neural structure and function’).19
Morse further explains that the brain enables the mind, and thus learning about the brain can offer beneficial information about. As such, if neuroscientific data is translated into the folk psychological framework of law, then such evidence would be legally relevant and potentially useful.20 For example, a brain scan that proves an impairment to the prefrontal cortex is meaningless in the legal context, as the law cares about issues such as the ‘intention’ and ‘knowledge’ of the person. Therefore, neuroscientific data should be explained in a legal folk psychological framework: evidence of ‘impaired prefrontal cortex’ must be directly related to legal concepts and translated into an issue that reduces the individual’s capacity to control their behaviour or capacity to form an intention to act.21
Morse’s claim about the gap between neuroscience and the law’s language can be summarised as follows: law’s criteria for responsibility and competence are behavioural, i.e. acts and mental states (law’s folk psychology), and neuroscience, rather, is mechanistic (cause and effect); and that is a mismatch between legal and neuroscientific languages. In order to make this mismatch compatible with the law’s folk psychology we should ‘translate the biological evidence into the criminal law’s folk psychological criteria’.22 In other words, ‘[i]f neuroscience is to have any influence on current law and legal decision-making, it must be through [law's folk psychological
17 Hodgson, 661–680 (p. 670). 18 Morse, pp. 529–562 (p. 537). 19 Morse, pp. 529–562 (p. 537). 20 Morse, pp. 529–562. 21 For more information about deficits arising from lesions of Prefrontal Cortex see SM Szczepanski & RT Knight, ‘Insights into human behavior from lesions to the prefrontal cortex’, in Neuron, vol. 83, 2014, 1002–1018. 22 Morse, pp. 529–562 (p. 537). 146
criteria]’.23 For instance, the claim that some brain impairment causes a crime should not influence the offender’s sentence – as the brain (whether impaired or not) always causes a person’s particular behaviour. Instead, it is necessary to explain how the brain impairment influenced the individual’s mental state (e.g. understanding and intention). Discussions of legal implications would then follow.24 As Morse explains, ‘[t]he chain of inference from the purely mechanical neurodata to the law’s act and mental state criteria must be clear. Unless the neuroevidence can help answer these questions, it is not legally relevant, even if it is scientifically valid’.25
Morse explains the importance of translation by noting that many claims that find neuroscience relevant to law are rhetorical.26 That is, neuroscientific evidence is not legally relevant, but makes the case more persuasive. He illustrates this issue by pointing to his experience in a conference presenting a criminal case before a number of federal judges. In the case, the offender27 strangled his wife and then threw her dead body out of the window of their apartment located on the twelfth floor. After he was charged with murder, a brain image indicated ‘a huge subarachnoid cyst pressing on and displacing a large amount of his left prefrontal lobe’.28 Consequently, he claimed that he could not control his conduct and raised the defence of insanity.29 However, his behavioural history and other evidence in the case were considered inconsistent with this version, and all of the judges rejected his claim. Then Morse asked judges whether they would find the cyst a mitigating factor in sentencing. About one-third of judges noted that they would consider it because the offender had ‘a proverbial “hole in his head”’.30 Morse responded that if the brain issue did not affect the offender’s behaviour and legally relevant mental states (e.g. capacity to form intent), why should it be considered as a mitigating factor? In this case, the judges’ observations about the cyst were mainly rhetorical.31 This case shows the importance of translating neuroscientific data into legal folk psychology, which can help to understand whether the evidence is directly related to legal concepts or not.32
23 Morse, pp. 529–562 (p. 533). 24 Morse, pp. 529–562 (p. 533). 25 SJ Morse, ‘Actions speak louder than images: the use of neuroscientific evidence in criminal cases’, in Journal of Law and the Biosciences, vol. 3, 2016, 336–342 (p. 4). 26 Morse, pp. 132–158. 27 Spyder Cystkopf/Herbert Weinstein 28 Morse, pp. 132–158 (p. 151). 29 i.e. not guilty by reason of mental illness. 30 Morse, pp. 132–158 (p. 152); Morse, pp. 529–562 (p. 549). 31 Morse, pp. 132–158 (p. 152). 32 In this case, the defendant first raised insanity defence and then he changed his decision pled guilty to manslaughter and was sentenced to prison. However, it appears Morse questioned judges with respect to the insanity defence, to avoid confusion, I did not discuss the defendant’s plea to manslaughter. 147
Overall, the main aim of this Chapter is to explore whether neuroscientific evidence, when discussed in court, is translated into criminal law’s folk psychology. There are a number of considerations in the analysis of this aim that I will discuss in the next section.
2. Difficulties in examining the claim
According to Hodgson and Morse’s explanation of law’s folk psychology, it is relatively clear that some terms such as the offender’s judgement and intention are law’s folk psychological terms; however, the status of many other terms and concepts that is less clear. For instance, is ‘intellectual capacity’ a folk psychological term, scientific psychology, or neuroscientific language? Further, as mentioned previously, the fact that Morse does not define what he means by the mechanistic way of explaining matters in psychological sciences (i.e. psychology and psychiatry), and Hodgson similarly does not clarify which areas of psychology and psychiatry explain things in folk psychology, make examining Morse’s claim more difficult.
In order to illustrate the difficulty of examining this claim in criminal cases, I will analyse the following statement from one of the judgements and discuss which terms are expressed in folk psychology, psychological sciences and neuroscientific language. In DPP v Houn, the trial judge states:
The presence of a depressive illness, developmental delay and cortical atrophy would [make a person] more vulnerable to the disinheriting33 effects of alcohol ...34
A ‘depressive illness’ appears to be a psychiatric and psychological term. Although both Hodgson and Morse find psychology and psychiatry to be close to folk psychology, and it may mean an individual with poor judgement, it is unclear whether Hodgson and Morse consider it a pre- scientific term that is used in law’s folk psychology.
‘Developmental delay’, is a concept that may be used in some areas of neuroscience, psychiatry and psychology and folk psychology.35 For instance, some may consider developmental delay a neuroscientific term because it is relevant to brain development. Others may understand it as a
33 From the context of the case, it appears that the judge was actually referring to the ‘disinhibiting’ effects of alcohol. 34 DPP v Houn [2008] NSWLC 16 (original italics are removed and emphasis added). I will discuss this judgment in more detail later in this chapter. 35 See AP Association, Diagnostic and statistical manual of mental disorders (DSM-5®), American Psychiatric Pub, 2013. 148
term from the psychological sciences referring to an individual with lower intellectual capacity. Developmental delay may also be considered a folk psychological term used to describe a person who is not clever and make poor judgements, or a person that bears an ‘unsoundness of mind’ — which according to Hodgson, is a folk psychological term used by the law.36
Concepts such as ‘cortical atrophy’ are more neuroscientific, and it is easier to distinguish them from folk psychological and psychological sciences’ languages.
Being ‘vulnerable to the disinhibiting effects of alcohol’ does not directly speak to the law’s folk psychological criteria, but its relevance to the law’s folk psychology (i.e. prone to lose control) and may be more readily understandable for lay people (e.g. the jury). Yet, it is unclear how Morse and Hodgson would consider this concept.
As such, there are some terms where it is difficult to determine whether they are compatible with law’s folk psychology, psychological sciences or neuroscience. The boundaries are not always clear and there are often differences, and inconsistencies, in classification and use.
To accommodate the ambiguities of this claim, I will define and limit the scope of the present inquiry to four categories of languages: first, the law’s folk psychology, which strictly speaking, includes only those terms that Morse and Hodgson have specifically referred to and defined as the law’s folk psychology language This includes concepts such as cognitive and rationality defects,37 impulsivity,38 ability to make free choices, impaired judgement, knowing wrong from right, lack of knowledge and loss of self-control.39 Thus, although some terms may have a close connection to those of the law’s folk psychology (e.g. depressive illness may indicate poor judgement, or development delay may indicate unsoundness of mind), strictly speaking, they are not defined by Hodgson and Morse as such. Second, concepts in the psychological sciences are referred to in terms such as depression, anxiety and schizophrenia. Third, terms such as brain injury and cortical atrophy are categorised as being neuroscientific language. Fourth, are unspecified terms that fall ambiguously between the categories of folk psychological, psychological sciences and neuroscientific languages, such as developmental delay and poor intellectual capacity (see Figure 6-1, categories of languages).
36 Hodgson, 661–680 (p. 670). 37 Morse, 336–342 (p. 4). 38 Morse, 209–228 (p. 847). 39 Hodgson, 661–680 (p. 670). 149
Figure 6-1: Categories of languages considered in this Chapter and their examples.
[
*Attention-Deficit Hyperactivity Disorder
I will briefly outline four types of cases with respect to Morse’s claim — neuroscience ought to be translated into law’s folk psychology to be compatible with law:
First, judgements that neuroscientific evidence is translated into legal folk psychology (e.g. how the brain impairment influences considerations such as the offender’s understanding and intention). In these cases, the use of neuroscience is consistent with prescriptions of Morse’s claim.
Second and third are types of cases where neuroscientific evidence is left untranslated, or it is translated into psychological sciences:
• Neuroscientific evidence affects the sentence (directly through discussions in neuroscientific language or indirectly through translation into psychological sciences language), then the judgement does not substantiate Morse’s claim.
• Neuroscience does not affect the sentence. This is consistent with Morse’s claim as he explains that if neuroscience is going to have any influence on decision-making, it should be through the law’s folk psychology; and in such cases, neuroscience is not translated into the law’s folk psychology, and it does not affect the sentence.
150
• Fourth, unspecified language – one that cannot be clearly categorised as being from the law’s folk psychology, psychological sciences, or neuroscientific language. In these situations, it is unclear whether the case is consistent with prescriptions of Morse’s claim.
Figure 6-2: This Figure illustrates the four types of cases with respect to the results of examining Morse’s claim.
Neuroscientific evidence Use of neuroscience in is translated into law's this case is consistent folk psychology with Morse's claim
Court finds neuroscientific evidence Use of neuroscience in to be relevant to this case is inconsistent Neuroscientific evidence is sentencing with Morse's claim Morse's claim discussed in psychological sciences or neuroscientific language Court does not find neuroscientific evidence Use of neuroscience in this case is consistent to be relevant to with Morse's claim sentencing
Neuroscientific evidence is discussed in Morse's claim is unclear unspecified language in this case
Despite these classifications, the translation process can still be ambiguous. What occurs if the data is only partially translated? For instance, evidence may indicate that a brain abnormality reduces the offender’s capacity to control their behaviour (i.e. law’s folk psychology), but also that it affects the neural activity of some brain area (neuroscientific language), which together causes the offence. Does this partial translation accord with Morse’s claim about the need to translate neuroscience into folk psychology? In such cases, I have determined that it is unclear whether the judgement is consistent with Morse’s claim. However, if neuroscience if it is explained that brain abnormality increases the offender’s anxiety (i.e. psychological sciences language), which causes the offender to have lower capacity to control their behaviour (i.e. law’s folk psychology), then the case is consistent with Morse’s claim because neuroscience is eventually translated into law’s folk psychology.
Another issue to consider in analysing this claim is the question of to what extent the accuracy of the translation matters. Is Morse concerned with the person who does this translation (e.g. an expert or a layperson)? Morse mentions that
151
forensic practitioners [expert witnesses] must explain precisely how the neuroevidence bears on whether the subject was capable of meeting the law’s functional criteria. If the evidence is not directly relevant, the advocate should be able to explain the chain of inference from the indirect evidence to the law’s criteria.40
It appears that Morse is mainly concerned with the translation itself, and the connection between the evidence and the legal criteria, rather than the accuracy. Even if Morse is concerned with the accuracy of the translation, and who conducts the translation, in many judgements it is not clear whether it was the judge, the expert witness, or the advocate that translated or inferred from the neuroscientific data. It is also possible that the expert’s report is comprised of a mix of neuroscientific and folk psychological language, and the judge makes assumptions to explain everything in folk psychological language in the judgement. Further, even if the expert witness is quoted in the judgement explaining how the brain condition affects the mental state of the individual (i.e. in folk psychological terms), the court may draw different, legally-inflected, inferences from those elaborated by the witness.
A further instance of ambiguity is the possible case where the expert translates the neuroscientific evidence into the law’s folk psychology, but the judge only uses neuroscientific terms in the judgement. Does this mean the court order is based on neuroscientific language rather than the law’s folk psychology? Or does it mean that as the judge understands the outcome of the neuroscientific issue in law’s folk psychology, he/she uses these two languages interchangeably and the case accords with Morse’s claim?
As one of the limitations of this study is that experts’ reports are not included in the dataset, it is not possible to delve into some of these uncertain circumstances, or even recognise where they arise (e.g. where the expert in the report discusses neuroscientific matters in the neuroscientific language and the judge translates them into folk psychology in the judgement). In circumstances where the uncertainty is recognisable (such as where there is a partial translation), I consider it to be unclear whether Morse’s claim is substantiated in that case.
3. Neuroscientific language in practice: analysis of
cases
40 Morse, pp. 132–158; Morse, pp. 529–562 (emphasis added). 152
In analysing the cases, I explored Morse’s claim from both the expert witnesses’ and judges’ perspectives: where the expert witness is quoted, how he/she explains neuroscientific matters, and how judges refer to neuroscientific evidence (in neuroscience language, law’s folk psychology, or psychological sciences).41
Analysis of cases suggests that where expert witnesses are quoted, they often translate neuroscientific evidence into the language of law’s folk psychology. For instance, in one of the cases, the expert outlined the offender’s mental state as follows,42
It would hence appear that in many ways a " perfect storm " of clinical issues have converged with this man, inclusive of underlying psychogenic problems, possible brain damage and the aggravating impact of protracted substance abuse over the years which in turn have had a dramatic impact upon his judgement, impulse control and decision making.43
Judgement, impulse control and decision-making are folk psychological concepts that are translated from the influence of underlying psychogenic problems, possible brain damage and the impact of protracted substance abuse on the offender’s mental state.
In some cases, however, the matter is more complicated, and experts use sophisticated neuroscientific terms. In DPP v Houn,44 the expert diagnosed problems including alcoholism and depressive moods. On the basis of the result of CT and MRI scans the expert reported that the offender had ‘an abnormal brain’, explained by ‘“the folia of the cerebellum being a little prominent and of there being an subarachnoid space in the posterior foca with a prominent cisterna magna”’. So far, the expert’s explanation of brain abnormality is in pure neuroscientific language. He further stated that,
The presence of a depressive illness, developmental delay and cortical atrophy would however interact to cause a circumstance in which a person would be more vulnerable to the [disinhibiting] effects of alcohol than if those conditions were not present.45
As discussed previously, vulnerability to the disinhibiting effects of alcohol does not directly speak to law’s folk psychological language, rather it is a concept from the psychological sciences. Yet, due uncertainties of Morse’s claim, this concept has been considered unspecified in this Chapter.
41 It is presumed that the judge’s explanation of neuroscientific matters is based on what the experts outline in their report. I will discuss this issue on the following page. 42 Caristo v R [2011] NSWCCA 7 43 Id. at [18] expert witness is quoted by the Supreme Court (Court of Criminal Appeal) (italic is original). 44 [2008] NSWLC 16 45 Id. (emphasis is removed). 153
The court’s expression of the offender’s mental state drew a clearer picture of his mental state in folk psychology. The judge explained that the existence of ‘an altered mental state’ at the time of the offence due to underlying conditions resulted in ‘cognitive impairment, impairment of judgement and poor volitional control’.46 As such, the overall expression of the judge was in law’s folk psychological language and this case accords with Morse’s claim.
In some judgements, the court paraphrases the expert witness’s report. In the majority of these cases, the court's explanation of the offender's mental state was in law’s folk psychology. This means that either judges had a great understanding of relevant areas in neuroscience, or experts explained everything in their report in folk psychological language, and judges were able to rely directly on these reports. The former seems to be more plausible. Admittedly, one of the limitations of this study is that the experts’ full reports cannot be accessed to gain further insight into this issue.
In one case, the appellant suffered from brain injury and some psychological issues that were caused by a motor vehicle accident.47 He had pled guilty to three offences two of which were committed one month after his accident. The Court of Criminal Appeal referred to neuropsychological testing conducted by an expert witness and noted that at the time of the crime ‘the applicant was continuing to experience the effects of the brain injury’.48 It seems the expert’s report is paraphrased, however there was no explanation in law’s folk psychological terms.49
The Court refers to the lower court’s conclusion about the offender's mental state,
by reason of [applicant] emotional and depressive state, the applicant was not fully aware of the consequences of his actions. However, [the sentencing judge] said: “this is not to say that he was not capable of rationally sitting down and considering them but rather that the combination of mental health issues and the sequelae of his motor vehicle accident some one month before the first offences was likely to have had a significant impact on his ability to think clearly at the time of the first two offences”.50
As such, the sentencing court’s representation of the expert report about the appellant’s mental state was in folk psychological language.
There are some instances where there are several expert witnesses who provide reports using a mix of neuroscientific, psychological sciences and folk psychological languages. In a special
46 Id. 47 Ionatana v R [2008] NSWCCA 95 48 Id. at [12]. 49 It is of note that there was another expert witness who reported, ‘the applicant’s [offences] were out of character and reflected an individual who was “under considerable stress, more impulsive and erratic in mood and generally acting in desperation for a ‘quick fix’”’. However, the expert witness could not determine if the offences were relevant to the sequelae of applicant’s brain injury. Id. at [13]. 50 Id. at [18] (emphasis added). 154
hearing51 in the Supreme Court,52 one of the experts with respect to the Forensic Patient’s mental state reported that an MRI scan showed ‘cerebral atrophy and lesions in the white matter consistent with vascular disease’.53 In another examination, the expert noted that the Forensic Patient’s mental state deteriorated, partly due to the head injury. The expert did not elaborate on the mental state.
A second expert witness found that the Forensic Patient had
significant cognitive deficits associated with a history of cerebral insult as well as a long history of severe poly-substance and alcohol dependence. [The expert witness] considered the Forensic Patient to be functioning in the mentally retarded range … He also obtained a history of seizures consequent on alcohol withdrawal.’54
Another expert witness diagnosed the Forensic Patient with ‘Axis III alcohol-induced brain damage and possible brain damage associated with previous head injuries’55 but did not explain what the (legal) consequences of those issues were.
A fourth expert witness reported the Forensic Patient to have:
a long history of heavy alcohol, cannabis and benzodiazepine use. The substance abuse has been complicated by seizures. He has a history of head injury and a baseline low IQ. His cognition appears to have deteriorated [in 2 years] since he was first assessed by [two other expert witnesses]. He [had] no memory of the crime and was unable to provide details of many historical aspects of his life … whereas he had previously been able to give a fuller history. His behaviour in the [hospital] is significantly disturbed ... He poses a risk to his safety as a result of his poor judgement and cognitive impairment.56
This expert’s report on the Forensic Patient’s mental state was in neuroscientific, psychological sciences and folk psychological language. However, he does not explain whether, for instance, the offender’s poor judgement and cognitive impairment were the sequala of his head injury or alcohol and substance use, or both.
51 As discussed in Chapter 3 (legal background), in special hearing, sentencing for individuals who are found guilty is referred ‘limiting term’. 52 R v Goodridge (No 2) [2012] NSWSC 1180 53 Id. at [31]. 54 Id. at [32]. 55 Id. at [33]. 56 Id. at [35] the court quoted the expert witness. 155
The judge, however, clearly distinguished the influence of alcohol and brain condition on Forensic Patient’s behaviour in law’s folk psychology,
At the time [of the crime] he was grossly intoxicated and he also had an underlying condition of alcohol-related brain damage. His judgement was substantially impaired by reason of his self-induced intoxication, which tended to make him behave aggressively and erratically. His judgement was also impaired, but not substantially so, by the underlying brain damage.57
Overall, in this case, while there were several expert witnesses who reported on the Forensic Patient’s mental state in different languages, the court clearly outlined the influence of Forensic Patient’s condition in law’s folk psychology.58
These cases, however, do not represent the entirety of the use of neuroscience in courts with respect to Morse’s claim. There are cases where neither the expert nor the judge has translated the neuroscientific evidence into folk psychological language. In the majority of these cases, neuroscientific data appears to have no effect on the sentence because the data was inconclusive, or deemed irrelevant to the case. These cases, as previously discussed (see Figure 6-2), are consistent with Morse’s claim. This is because Morse explains that if neuroscience is going to have any influence on legal decision-making, it should be through law’s folk psychology, and in these cases, the untranslated neuroscience does not appear to have contributed to the sentencing decision.
For example, in one of the cases,59 the result of a CT scan was found inconclusive and an expert witness reported it as follows: ‘There is apparent hypo density in the pons image six measuring up to 4.5mm. It is difficult to determine whether this is a genuine lesion or an artefact’. An MRI was conducted to resolve the issue, however, it was not submitted to the court. The judge concluded that if there was something wrong with the offender’s brain the defence would have reported the result to the court. As such, while the CT scan report was in neuroscientific language, it was not translated into law’s folk psychology (by experts or by the judge). This neuroscientific evidence did not appear to affect the determination of the sentence.60
57 Id. at [18]. 58 The court also at [50] stated that ‘The period of premeditation was very short; any thought processes on his part which led him to form the relevant intent would have been affected by the substantial quantities of alcohol he had ingested and his underlying mental impairment.’ 59 R v Paredes, R v Barillaro, R v Donevski [2011] NSWDC 48 60 Also see, R v Basilious [1999] NSWCCA 219 at [14] 156
In another instance, the offender sustained brain damage and the expert witness reported, 61
[The offender] demonstrated average non-verbal abilities and intact verbal memory function but he showed compromised performance in tasks of visual memory and more complex processing speed and working memory tasks. That comparison was strongly suggestive of brain damage.62
In this case, neuroscientific evidence appears to have been translated into the language of the psychological sciences.63 However, as with the previous case, this condition did not seem to contribute to sentencing and therefore, it is consistent with Morse’s claim.64
Conversely, there are some cases where no translation is apparent and it seems that neuroscience somehow contributed to the sentence and therefore, such cases do not accord with Morse’s claim. In an appeal case65 (Court of Criminal Appeal) one of the expert witness observed that while the appellant in his previous examination did not indicate any mental illness, ‘[he] presented in a way that was typical of someone with damage to the frontal lobes, with slightly eccentric appearance and a wide eyed stare’.66 Another expert witness also reported that ‘the bizarre aspects of the applicant’s behaviour were possibly related to organic brain dysfunction of the frontal lobe.’67 Despite the fact that none of the experts explained what the outcome of the damage to the frontal lobe was, and how it was legally relevant,68 the trial court took into account the appellant’s brain condition in determining the appropriate sentence.69
In another case, Iglesias v R,70 when referring to the earlier decision the Court of Criminal Appeal noted that ‘His Honour accepted [the expert’s] opinions that the applicant’s brain had been damaged by substance abuse and that he had borderline intellectual capacity which contributed to his violent behaviour’.71
61 R v Do (No. 4) [2015] NSWSC 512 62 Id. at [6]. 63 R v Do (No. 4) [2015] NSWSC 512 64 Also see, R v Wallace [2006] NSWSC 897 65 Chaplin v R [2006] NSWCCA 40 66 Id. at [12] 67 Id. at [13] 68 The second expert witness also referred to a number of other factors such as his alcohol and drug consumption, disturbed sleeping pattern. He noted that ‘the applicant had significant problems with thinking and concentration, at times accompanied by prominent distress and dysphoria'. However, it does not seem this observation is with respect to the frontal lobe damage. Id. at 13. 69 Id. at [19]. 70 Iglesias v R [2006] NSWCCA 261 71 Id. at [17] 157
The uncertainties in this case suggest two possible interpretations of the court’s statement: first, there was no translation from neuroscience (brain damage) to law’s folk psychological language, and it was merely noted that the brain damage contributed to violent behaviour (i.e. the crime). Causation, per se, as Morse explains, is not a mitigating or an excusing factor.72 A psychopath may also be considered more violent due to some structural brain differences that result in a lowered ability to empathise with their victims and apathy towards hurting other people. But the law may not find that to be an excusing or mitigating factor, since the inability to share empathy with others is not relevant to considerations of knowing right from wrong, and control over one’s actions – factors the criminal law is concerned with.73 Thus, in this case, the use of neuroscience does not accord with Morse’s claim.
The second possible interpretation of the court’s statement is that brain damage resulted in borderline intellectual capacity,74 which contributed to violent behaviour. Based on this view, the influence of brain damage is translated into borderline intellectual capacity. Intellectual capacity may be interpreted as a neuroscientific concept, since it is relevant to a neurodevelopmental problem. Alternatively, it may also be argued that it falls under psychological sciences because it is associated with an individual’s IQ and ability to adapt to environmental factors. Its relation with law’s folk psychological language is relatively clear: a low intellectual capacity means someone may lack the requisite judgement skills, which explains the offender’s violent behaviour at the time of crime. As outlined in Figure 6-1, borderline intellectual capacity remains one of the concepts that may be difficult to demarcate. Therefore, it is unclear whether this case is consistent with Morse’s claim if the second interpretation is taken.
It is noteworthy in many appeal cases courts only summarise the lower court’s judgements and may not repeat the lower court’s translation of neuroscience into folk psychology. In Iglesias, specifically, the appeal did not turn on the applicant’s mental state, rather, it focused on the applicant’s physical condition (cancer) and therefore even if the lower court provided a more detailed explanation of brain damage and what it meant in law’s folk psychology, it was not relevant to the appeal and was not discussed.
72 Morse, pp. 529–562. 73 See SJ Morse, ‘Psychopathy and criminal responsibility’, in Neuroethics, vol. 1, 2008, 205–212 (pp. 207 & 208). 74 Borderline intellectual capacity as a developmental disorder does not seem to be caused by brain damage resulting from drug abuse. However, as a non-scientist, I cannot make a judgment on the relationship between different disorders. For information about intellectual capacity see, JVM Jenkins et al., ‘Direct and indirect effects of brain volume, socioeconomic status and family stress on child IQ’, in Journal of child and adolescent behavior, vol. 1, 2013; F Baglio et al., ‘Abnormal development of sensory-motor, visual temporal and parahippocampal cortex in children with learning disabilities and borderline intellectual functioning’, in Frontiers in human neuroscience, vol. 8, 2014, 806. 158
Some courts appear to be cautious about the importance of translation of neuroscientific issues (or more broadly biological disorders) and how they are relevant to legal criteria. In a District Court case, the judge explained that, ‘[o]f course he may have a damaged mind, or a genetic brain malfunction, … but it does not demonstrate itself in the prisoner's day to day drug dealing activities.’75 As such, the court did not find the mere existence of a genetic brain malfunction relevant to sentencing because it was not relevant to the offender’s criminal behaviour.76
Overall, the empirical analysis suggests that from the 112 cases where neuroscience contributed to the determination of an appropriate sentence, 100 involved some form of translation between the neuroscientific and legal folk psychological languages. Of these 100, 93 involved a direct translation of the neuroscientific evidence into the law’s folk psychology. In the remaining seven, the expert and the judge did not explain the neuroscientific evidence in law’s folk psychology, however, the judge in sentencing remarks summarised the offender’s medical issues in the law’s folk psychology. It is important to note that in these 100 cases, other types of evidence were also submitted to the court, along with the neuroscientific evidence. At times, the implications of all the evidence were combined to draw inferences on the offender’s mental state, which was explained in the law’s folk psychology. In the remaining 12 cases, neuroscientific evidence was translated only into the language of the psychological sciences (seven cases) or an unspecified language (five cases). This included terms such as low range of intellectual functioning, developmental delay, ADHD, poor memory, comprehensive and expressive language difficulties and poor attention.
From the remaining cases in the dataset of this study (109), there were 27 cases that it was unclear whether the court considered neuroscientific evidence in sentencing. From the 82 remaining cases in which neuroscientific evidence did not appear to be considered in determining an appropriate sentence (e.g. because the brain imaging did not show any abnormality), there were 60 cases in which neuroscience was left untranslated or was translated into the language of the psychological sciences. Nonetheless, these cases are consistent with Morse’s claim since they did not influence the sentence. This is because, as Morse explains, if neuroscience is going to have any influence on legal decision-making, it should be through the law’s folk psychology.
75 R v Orchard [2011] NSWDC 151 at [58]. 76 Similarly, in Ngati v R [2014] NSWCCA 125 the court concerning applicant's possible brain injury and its consequences stated that ‘They are not specifically directed to the applicant's "reasoning and decision making capacity" in relation to the subject offences. Otherwise, [the expert witness's] report did not expressly address the existence of any link between his impaired intelligence and his commission of the three armed robberies.' at [37]. 159
Taking these findings into consideration, it appears that in the majority of cases, the use of neuroscience in courts is consistent with the prescriptions of Morse’s claim. It is important to note that due to the ambiguities of this claim and the unclear boundaries between the different languages (i.e. law’s folk psychology, psychological sciences, and neuroscience), this study was required to make determinations on the scope of each category, and the terms that fell within them. As these decisions may not be shared by others, the results should be interpreted cautiously and appropriately.
Analysing judgements with the purpose of evaluating Morse’s claim revealed an ancillary finding that I will discuss in the next section.
4. Neuroscience and folk psychology: interchangeable languages
Another finding emerging from the analysis of neuro-language in the cases is that in some situations, while the expert or the judge translated neuroscientific matters into law’s folk psychology, the court later used neuroscientific terms to refer to brain issues. This might suggest that some courts are becoming sufficiently familiar with neuroscience to use neuroscientific and law’s folk psychological language interchangeably. In other words, courts may not find it necessary to explain matters in the language of folk psychology (once matters are translated) but find neuroscientific terms equivalent to those of folk psychology (e.g. they use the original neuroscientific language such as ‘the influence of prefrontal cortex impairment’ instead of the law’s folk psychological translation, ‘problems with controlling behaviour’).
The interchangeable use of languages suggests the possibility that in the future, judges and expert witnesses may not find it necessary to translate neuroscience (or so much neuroscience) into the law’s folk psychology and will only explain an individual’s condition in neuroscientific terms.
This may also suggest the development of folk-neuroscience. While there is no pre-determined theoretical framework linking neuroscience and the law’s folk psychology, the regular use of these two languages in courts seems to be creating a connection (that is using these two languages interchangeably).77 Similar to discussions about ‘gravity’ that once were purely scientific topics
77 This connection may not be valid as it is created in the court room rather than being analysed by experts in this area. That said, Hodgson also finds that there are some areas of psychology and psychiatry that use folk-psychological concepts, ‘but tend increasingly to be regarded as unscientific’ (emphasis added). Hodgson, 661–680 (p. 670). 160
that now seems to be a folk science, a common use of neuroscience in courts linked to folk- psychological concepts may have the same result, i.e. folk-neuroscience. Further, the neuroscientific language that experts seem to use in courts is different to the level of specificity experts’ (neuroscientists) use in everyday communication in the lab. Rather, it is a simplified version. However, issues remain including the fact that the connection formed in courts between neuroscientific and folk psychological concepts may be inaccurate or even incorrect. Further, while folk-neuroscience, by creating an interchangeable language between neuroscience and the law, may to some extent address Morse’s concerns about neuroscience’s relevance to legal criteria — that neuroscience and law use different languages — it may still fall short in addressing the problem that neuroscience is a mechanical science which does not consider the mind, a holistic entity that the law believes is in control of an individual’s voluntary acts.78
Another possible outcome of the use of neuroscientific and folk psychological languages interchangeably is that courts may become more positively disposed to the submission of neuroscientific evidence and other more recent technologies. Judicial officers who read judgements will become more familiar with the relationship between neuroscience and the relevant legal principles, and consequently, we may observe more submissions of, and references to, neuroscience in courts.
Conclusion
Morse claims that neuroscientific evidence is not legally relevant until it is translated into the criminal law's folk psychological criteria. For example, a brain scan that proves an impairment to the prefrontal cortex is meaningless in the legal context and it should be explained the way the impairment influenced the individual’s mental state at the time of crime and how it is relevant to legal criteria, such as the capacity to make intent. The aim of this Chapter was to examine whether discussions of neuroscience in judgements are translated into law’s folk psychology — i.e. whether the use of neuroscience is consistent with the prescriptions of Morse’s claim.
The analysis of cases indicated that in the majority of cases where neuroscientific evidence appears to influence the sentence, it is translated into the legal framework of folk psychology. In most cases where neuroscience is not translated, or translated into psychological sciences, neuroscience does not appear to have any influence on the sentencing decision.
78 See Hodgson, 661–680 (p. 670). 161
It is difficult to evaluate Morse’s claim due to the difficulty of determining the boundaries of concepts that fall under law’s folk psychology in specific circumstances, along with the complexity and limitations of many judgements. It is of note that none of the studies from other jurisdictions that have empirically analysed the use of neuroscience in courts has evaluated this claim. If Morse provides more explanation and clarification on this claim, empirical studies may then be able to examine the claim in different jurisdictions. The results of these studies then can potentially be used as feedback for judicial officers on appropriate ways to use neuroscientific evidence, and how to address problems of translation when they arise.
162
Chapter 7 Concerns about the Reliability of Neuroscientific Evidence in Courts
163
Introduction
In Chapter 5, I explained that neuroscientific evidence has been used in courts from as early as 1975, and I discussed its wide range of uses. However, despite the relative longevity, many still harbour reservations about the reliability of such evidence. This Chapter is centred around three claims regarding the reliability of neuroscientific evidence in sentencing.
The first claim contends that brain imaging evidence (specifically functional imaging) is subject to many limitations and it is not reliable enough to prove a fact about a mental state (e.g. the offender could not control their act).1 For instance, a brain with a defect does not necessarily imply impaired mental and behavioural function,2 and the use of this evidence may misguide the criminal procedure. Some scholars, while acknowledging the limitations of neuroscientific evidence, argue that each single piece of admissible evidence in criminal proceedings should not be required to meet the legal standard of proof. Instead, an argument about the mental state of an individual only needs to be supported by a combination of different pieces of evidence, which together, sufficiently satisfy the legal standard of proof.3 As such, some scholars make a normative claim—the convergence of evidence—about how neuroscience ought to be used: neuroscientific evidence alone is rarely dispositive of a mental state, and it should be used to direct and support (or buttress) other types of evidence such as psychiatric, neuropsychological and behavioural evidence.4
Second, the claim of the persuasive power of neuroscience presents that courts may place undue
1 See W Glannon, ‘The Limitations and Potential of Neuroimaging in the Criminal Law’, in The Journal of Ethics, vol. 18, 2014, 153–170; TR Brown & ER Murphy, ‘Through a scanner darkly: functional neuroimaging as evidence of a criminal defendant’s past mental states’, in Stanford law review, vol. 62, 2010; JC Moriarty, ‘Flickering admissibility: neuroimaging evidence in the US courts’, in Behavioral sciences & the law, vol. 26, 2008, 29–49. 2 See, S Morse, ‘Neuroscience in Forensic Contexts: Ethical Concerns’ in Ethics Challenges in Forensic Psychiatry and Psychology Practice, EEH Griffith (ed), New York, Columbia University Press, 2018, pp. 132–158. 3 See, N Aggarwal & E Ford, ‘The neuroethics and neurolaw of brain injury’, in Behavioral sciences & the law, vol. 31, 2013, 789–802. Also see F Schauer, ‘Neuroscience, lie-detection, and the law: Contrary to the prevailing view, the suitability of brain-based lie-detection for courtroom or forensic use should be determined according to legal and not scientific standards’, in Trends in Cognitive Sciences, vol. 14, 2010, 101–103. 4 OS Choi, ‘What Neuroscience Can and Cannot Answer’, in Journal of the American Academy of Psychiatry and the Law Online, vol. 45, 2017, 278–285; C Scarpazza et al., ‘The charm of structural neuroimaging in insanity evaluations: guidelines to avoid misinterpretation of the findings’, in Translational Psychiatry, vol. 8, 2018, 227–237; JA Silva, ‘Forensic psychiatry, neuroscience, and the law’, in Journal of the American Academy of Psychiatry and the Law, vol. 37, 2009, 489–502; G Sartori, S Pellegrini & A Mechelli, ‘Forensic neurosciences: from basic research to applications and pitfalls’, in Current Opinion in Neurology, vol. 24, 2011, 371–377; D Amen & L Flaherty, ‘Is brain imaging clinically useful for psychiatrists?’, in Clinical Psychiatry News, vol. 34, 2006, 11. Also see, OD Jones, ‘Seven Ways Neuroscience Aids Law’ in Neurosciences and the Human Person: New Perspectives on Human Activities, AM Battro et al. (eds), Vatican City, The Pontifical Academy of Sciences, 2013, pp. 181–194. 164
weight on neuroscientific evidence, and find it more convincing and scientifically credible than can actually be justified.5 For example, Joseph Dumit argues that courts may refer to brain images as objective evidence, despite the fact that the interpretation of this evidence relies on the subjective analysis of expert witnesses.6 Consequently, the outcome of introducing brain images as evidence may vary depending on who interprets it.7 Decisions influenced by the persuasive power of neuroscience may have two consequences: one is that neuroscientific evidence may be found more credible than other evidence, where this is unwarranted (e.g. behavioural evidence), and will result in excluding conflicting evidence, and two, it may have undue influence on sentencing decision (i.e. unjustified harsher or more lenient sentence).
An empirical evaluation of this claim requires a normative stance of the appropriate weight that should be ascribed to neuroscientific evidence, which would act as a point of comparison on whether courts are in practice giving undue weight to neuroscientific evidence. To avoid the need to make normative justifications, the examination of this claim will be limited to highlighting more distinguishable examples of where neuroscience appears to be used inappropriately, such as cases where the court refers to brain imaging as objective evidence.
Third, arguments have been made about some types of expertise and its value to legal decision- makers. More specifically, it is claimed that some expert witnesses do not have the requisite expertise enabling them to discuss brain imaging evidence and thus their reports and testimony are not credible or sufficiently reliable8 (area of expertise claim). For instance, Amen and Flaherty argue that psychiatrists are not qualified to interpret brain images as they are not trained in clinical imaging. Consequently, they are not experts on when brain imaging is appropriate, or how to interpret and use the results of imaging.9 Similarly, Douglas Bremner claims that many mental health professionals, such as social workers or psychologists, do not have the background needed in brain anatomy and physiology to understand the neurobiology of mental disorders and brain imaging.10
5 DP McCabe & AD Castel, ‘Seeing is believing: The effect of brain images on judgments of scientific reasoning’, in Cognition, vol. 107, 2008, 343–352; DS Weisberg et al., ‘The seductive allure of neuroscience explanations’, in Journal of cognitive neuroscience, vol. 20, 2008, 470–477. As discussed in the literature review, there are other studies that claim otherwise. Nonetheless, this issue is not addressed empirically. See, NJ Schweitzer et al., ‘Neuroimages as evidence in a mens rea defense: No impact’, in Psychology, Public Policy, and Law, vol. 17, 2011, 357–393; Morse, pp. 132–158. 6 J Dumit, ‘Objective brains, prejudicial images’, in Science in Context, vol. 12, 1999, 173–201. 7 Dumit, 173–201. 8 JC Moriarty & DD Langleben, ‘Who Speaks for Neuroscience-Neuroimaging Evidence and Courtroom Expertise’, in Case W. Res. L. Rev., vol. 68, 2017, 783–804. 9 Amen and Flaherty, 11; Silva, 489–502. Also see D Linden, Neuroimaging and Neurophysiology in Psychiatry, Oxford, United Kingdom, Oxford University Press, 2016. 10 JD Bremner, Brain imaging handbook, WW Norton & Co, 2005; Amen and Flaherty, 11; Silva, 489– 502.Bremner; Amen and Flaherty, 11; Silva, 489–502. 165
Notwithstanding this call for relevant expertise, there are no guidelines stipulating what qualifications and experiences are sufficient to satisfy the competency of an expert witness on different forms of neuroscientific evidence (e.g. MRI, SPECT, neuropsychological testing). Rather, this issue is not addressed in the empirical neurolaw studies conducted in other countries.11
As it is beyond the scope of this research to design such a guideline, the following analysis will provide a descriptive assessment of expert witnesses’ areas of expertise when they are called upon to discuss neuroscientific evidence, and whether courts are concerned about expertise and competency. This will allow this study, first, to assess whether psychiatrists, psychologists and social workers interpret brain imaging evidence and present the results in courts, i.e. to test the area of expertise claim.12 Second, it provides a more comprehensive description of experts who present neuroscientific evidence to courts and how courts react to their experience and qualification. The latter assessment is conducted with the aim of encouraging and providing an opportunity for future studies to build upon the results of this section and discuss whether experts’ area of expertise is a matter of concern in Australia. Further, this assessment is in line with the broad aim of this thesis — to understand how neuroscientific evidence is being used in practice.
Accordingly, this Chapter has three main aims corresponding to the three claims relating to the reliability of neuroscience evidence. It aims to examine whether, in proving a legal claim (e.g. the offender could not control their behaviour), neuroscience is a dispositive proof, or it is used in conjunction with other forms of evidence (convergence of evidence claim). Second, it aims to discuss whether the interpretation of neuroscientific evidence and how it is weighted is misguided (claim of the persuasive power of neuroscience). Finally, it also aims to identify expert witnesses’ areas of expertise when they are called upon to discuss neuroscientific evidence, and whether courts are concerned about expert witnesses’ competency. More specifically, it aims to determine whether psychiatrists, psychologists and social workers interpret brain imaging evidence and present the results in courts (area of expertise claim).
11 For instance, in the England and Wales study outlines who the people that appear in courts discussing neuroscientific evidence are and does not discuss whether these experts are qualified to discuss neuroscientific evidence, ‘[t]he experts who appear in the case reports are not described as “neuroscientists”. They are described as: “neuropsychiatrists”, “neuropsychologists”, “neurologists”, “neurosurgeons”, “neuroconsultants”, “neuroradiologists”, and “neuropathologists”’. P Catley & L Claydon, ‘The use of neuroscientific evidence in the courtroom by those accused of criminal offenses in England and Wales’, in Journal of Law and the Biosciences, vol. 2, 2015, 510–549 (p. 514). 12 The claim of areas of expertise is confined to three areas of psychiatry, psychology and social work as these are the only areas that apparently concern scholars. 166
In the first part of this Chapter, in order to provide clarity on the discussions of these three claims in the NSW legal context, I will discuss the laws of evidence associated with these claims, including the safeguards that are currently in place to prevent unreliable evidence from entering the courtroom.
In the second part of this Chapter, I will turn to discuss the convergence of evidence claim. I will first draw upon studies and criticisms of the convergence of evidence claim, which deserve consideration. Further, since this claim concerns the reliability of neuroscientific evidence, it is useful to know how courts generally treat neuroscience as a field of study and as evidence. That is, whether they criticise neuroscience as a field that is subject to limitations, and whether they treat neuroscientific evidence differently compared to other types of evidence. Following this, I will discuss results from the analysis of cases with respect to the convergence of evidence claim, and conclude that the use of neuroscience in courts accords with the prescriptions of this claim. That is, the use of neuroscientific evidence in practice is consistent with the suggestion that where this evidence is admitted and relied upon, it should be coupled with other types of evidence in determining a mental state rather than be taken as dispositive proof of a mental state.
In the third part of this Chapter, I will discuss the claim of the persuasive power of neuroscience. I will first review the limitations of examining this claim, such as the difficulty of determining how much weight judges assign to neuroscientific evidence and why I will limit the evaluation of this claim to more objective approaches. Then, I will assess where neuroscientific evidence and behavioural evidence are in conflict, which one the court finds more credible. I will move on to examine whether courts refer to neuroscientific evidence as an objective form of evidence. Finally, I will discuss Veen’s cases where the courts’ reasoning suggests judges gave undue weight to neuroscientific evidence. Bearing in mind that evaluation of this claim was limited to specific approaches, I will conclude that apart from a few cases, the overall examination of cases does not suggest that courts are prejudicial towards neuroscientific evidence (i.e. the use of neuroscience in courts is not consistent with the claim of prejudicial neuroscience).
In the fourth part of this Chapter, after a brief explanation of the important role of expert witnesses in presenting neuroscientific evidence to courts (via reports and testimonies), I will outline and compare the area of expertise of those experts who reported on neuroscientific evidence in the cases reviewed. I will conclude that the results suggest psychiatrists repeatedly, and psychologists in a few instances, reported on brain imaging evidence. These findings are in contrast with what the proponents of areas of expertise claim recommend. Further, I will explain that with only a few exceptions courts do not appear to be particularly concerned (at least in the judgements) with the expert’s area of expertise in reporting neuroscientific evidence.
167
1. An overview of the reliability of evidence in NSW sentencing procedures
In NSW reliability – in the sense of trustworthiness – does not play an important part in the admissibility of expert opinion evidence.13 To be admissible under the Evidence Act 1995 (NSW), opinion evidence must be accommodated within an exception to the exclusionary opinion rule. Section 79(1), the exception for the expert opinion evidence, states: ‘If a person has specialised knowledge based on the person's training, study or experience, the opinion rule does not apply to evidence of an opinion of that person that is wholly or substantially based on that knowledge’. Courts, including the NSW Court of Criminal Appeal in Tang and Chen, determined that trial judges should not consider the reliability of expert opinion when considering admissibility under this section.14 The other potential means of exclusion, s137 of the same Act states: ‘In a criminal proceeding, the court must refuse to admit evidence adduced by the prosecutor if its probative value is outweighed by the danger of unfair prejudice to the defendant’. In IMM, the High Court explained that trial judges should not consider the reliability of the evidence or the credibility of the witness when determining the probative value of the impugned evidence.15 In practice, these approaches mean that the reliability of evidence, including expert opinion evidence, must be explored during the trial and is ultimately a decision made by the tribunal of fact; whether trial judge or jury.
It should not come as a surprise, given this orientation, that during sentencing the judge must determine the appropriate weight to attributed to expert opinion. Accommodating admissibility rules are not usually applied, or are applied even more leniently, such that the sentencing judge will usually receive expert opinion evidence offered by the parties.16 The value or weight attributed to that evidence is a matter for the judge in the context of the available evidence. For reasons of fairness to the parties (particularly the offender), sentencing judges are generally open to admitting expert opinion, perhaps confident in the belief that they can effectively manage its influence through their overall assessment. In this regard, Odgers comments on the outcome of admitted, but fully or partially unreliable evidence that parties do not object to, stating that: ‘Of
13 G Edmond & MS Roque, ‘The Cool Crucible: Forensic Science and the Frailty of the Criminal Trial’(2012)’, in Current Issues in Criminal Justice, vol. 24, 2012, 51; Australian Law Reform Commission, Evidence, ALRC 26 (Interim) Vol 1 (1985), at [356] and [743]. 14 R v Tang (2006) 65 NSWLR 681, [137]; Chen v R [2018] NSWCCA 106. See also Tuite v The Queen [2015] VSCA 148, [58]-[59]. 15 IMM v The Queen [2016] HCA 14. 16 In R v Bourchas (2002) 133 A Crim R 413 at [69] the court stated that ‘most material will be put before the sentencing judge by consent, either express consent or because the Crown or the offender does not object to inadmissible material or even assertion’. 168
course, there may be no objection taken to evidence that is either strictly inadmissible or may not be used for a particular purpose. In those circumstances, it is for the sentencing court to determine the weight that should be given to the material’.17 He provides an example of an expert witness proffering an opinion that extends beyond their legitimate area of expertise, and concludes that this evidence ‘would be strictly inadmissible – little or no weight may be given to such evidence’ by the court.18
With this brief introduction to the liberal admissibility jurisprudence of NSW, I will move on to discuss the first of three claims concerning the reliability of neuroscientific evidence.
2. Reliability of neuroscience: convergence of evidence
We may need to be cautious about the use of neuroscientific evidence where the underlying scientific procedures or interpretations have not been adequately evaluated.19 While there are few rules or mechanisms preventing the admission of and potential reliance on neuroscientific evidence, those contemplating reliance should be interested in its validity and accuracy.20
The epistemic value of neuroscientific evidence is sometimes contested and some of its methods and claims are quite recent in origin.21 Proponents of the convergence of evidence claim suggest that while neuroimaging evidence may include some contentious/controversial claims and methods, it should be admitted and relied upon where it is supported by other forms of evidence (a kind of informal corroboration).22
While this might appear to be reasonable, it raises issues that deserve consideration. One of the important issues associated with corroborating neuroscientific evidence is the question of whether
17 SJ Odgers, Sentence : the law of sentencing in NSW courts for State and Federal offences, 3rd ed., Haberfield, N.S.W, Longueville Media, 2015 (emphasis added). 18 Odgers. Also see WW v R [2012] NSWCCA 165 at [55]-[60]; Lam v R [2015] NSWCCA 143 at [74]- [83]. Also, in the recent case Heath v R [2016] NSWCCA 24 at [40] the appellate court confirmed that ‘a sentencing court is not obliged to accept unchallenged evidence’ for instance because ‘it is inherently improbable’. 19 This seems to be a particular danger where procedures and opinions of unknown value are developed in circumstances where the experts are exposed to a great deal of contextual information. 20 G Edmond, ‘Forensic science evidence and the conditions for rational (jury) evaluation’, in Melbourne University Law Review, vol. 39, 2015, 77. 21 See, Morse, pp. 132–158. 22 Choi, 278–285. This claim is mainly directed at the use of brain imaging evidence, particularly functional imaging because of the many criticisms of these approaches. See, Brown and Murphy; Glannon, 153–170. 169
the expert’s opinion is made independently from the other evidence in the case. Did, for example, the expert consider the other evidence (or the beliefs of investigators) in forming their opinion. Consider the following example, where an individual has experienced a serious motor vehicle accident followed by behavioural changes, evidence will often be received from a psychiatrist. If an MRI is interpreted by an expert to suggest a brain injury that corresponds with the individual’s behavioural history, this may represent a type of corroboration – the corroborative effect extends insofar as where the behavioural history and the MRI diagnosis are consistent. However, an important issue here that may threaten the value of the neuroscientific evidence – and an issue that reinforces the need to attend to the validity and reliability of methods – is the extent to which the interpretation of the MRI is made independent of the history. There is a risk that exposing the expert that interprets the offender’s scan to his or her history may bias the interpretation. While this may seem to be a minor risk, numerous studies suggest that biased interpretation that is influenced by exposure to contextual information is serious, and may lead experts to interpret the same image in different ways.23 That is, experts who are shown images (such as an MRI) may interpret them differently depending on what they know about the offender. These risks are not peculiar to the interpretation of neuroscience, but have been studied across the forensic sciences and human behaviour more generally.24 Moreover, they may arise in circumstances where experts are acting with integrity and honesty, and even when experts are aware of the risks. Indeed, this is the rationale behind double-blind clinical trials in biomedical studies.25
So two issues emerge from this discussion. The first concerns the known value of the procedures (their validity) and accuracy. The second concerns the extent to which dangers from cognitive bias are managed or avoided. Both of these issues pose serious threats to the convergence of evidence claim because we cannot assume that merely because two experts appear to agree, that their evidence is genuinely corroborative. It may be the case, but agreement between expert witnesses may also be the result of suggestion, priming and other insidious influences on their perception. These dangers require attention to whether the procedure works, and how well, and
23 For more information on how contextual information may influence an expert witnesses’ decision- making see, G Edmond & J Tangen, ‘Contextual bias and cross-contamination in the forensic sciences: the corrosive implications for investigations, plea bargains, trials and appeals’, in Law, Probability and Risk, vol. 14, 2015, 1–25. 24 See JJ Lockhart & S Satya-Murti, ‘Diagnosing crime and diagnosing disease: bias reduction strategies in the forensic and clinical sciences’, in Journal of forensic sciences, vol. 62, 2017, 1534–1541; PA Zapf & IE Dror, ‘Understanding and mitigating bias in forensic evaluation: Lessons from forensic science’, in International Journal of Forensic Mental Health, vol. 16, 2017, 227–238; DM Risinger et al., ‘The Daubert/Kumho implications of observer effects in forensic science: Hidden problems of expectation and suggestion’, in California Law Review, vol. 90, 2002, 1–56. 25 PJ Karanicolas, F Farrokhyar & M Bhandari, ‘Blinding: Who, what, when, why, how?’, in Canadian Journal of Surgery, vol. 53, 2010, 345–348. 170
whether the interpretation was managed to protect the expert from information that may unnecessarily threaten their interpretation.
These concerns were expressed by an interdisciplinary committee in the US considering the forensic sciences more generally. They advised:
Two very important questions should underlie the law’s admission of and reliance upon forensic evidence in criminal trials: (1) the extent to which a particular forensic discipline is founded on a reliable scientific methodology that gives it the capacity to accurately analyze evidence and report findings and (2) the extent to which practitioners in a particular forensic discipline rely on human interpretation that could be tainted by error, the threat of bias, or the absence of sound operational procedures and robust performance standards.26
I am not suggesting that neuroscientific evidence cannot corroborate other types of evidence used in sentencing. But problems arise where, for example, the opinion is not based on a validated procedure. Further, if an opinion is developed in circumstances where the expert is exposed unnecessarily to information about the offender or the case, there is a real danger that any interpretation may be influenced by this information – even where the underlying procedure has been validated. Interestingly, there is little evidence that awareness of the risks of bias or experience is effective at countering or overcoming the dangers.
On balance, this claim is more complicated than some of the proponents seem to appreciate.27 The question of whether neuroscientific evidence offers support or corroboration will often depend on the value of the particular procedure/technique and the ways threats to unbiased interpretation are managed. In general, where the procedure is valid and where contextual information is managed, corroboration will be most valuable. Where these conditions are not satisfied, the corroboration may have limited, if any, evidentiary value.
At the core of the convergence of evidence claim — whether courts rely on neuroscience where it is corroborated with other types of evidence — is the concern about the validity and reliability of neuroscience. Therefore, before I move on to assess this claim, it is useful to discuss how the courts conceive and seek the testimony of expert witnesses on the reliability of this evidence.
26 National Research Council, Strengthening Forensic Science in the United States: A Path Forward, Washington, DC, The National Academies Press, 2009, p. 87. 27 Scarpazza et al., 227–237 briefly refer to this issue (p. 232). 171
Courts’ consideration of the reliability of neuroscientific evidence in practice
As indicated, the reliability of neuroscientific evidence is an important issue both at trial and during sentencing. The admissibility rules and jurisprudence will not necessarily lead to the exclusion of procedures and opinions that may be contested and even unreliable. This means that decision makers – the jury during the trial and the judge during sentencing – are responsible for determining the value (or weight) of the neuroscientific evidence. And, judges appear to be more accommodating of evidence during sentencing than during the trial. This may be because there are no juries and judges believe they are capable of assessing complex evidentiary issues.28
Analysis of the cases suggests that unlike psychological and psychiatric examinations and reports, which judges assume to be associated with their respective fields of study (i.e. psychology and psychiatry), it is unclear whether judges find different types of neuroscientific evidence (MRI scan and neuropsychological testing) to be associated with neuroscience as a field of study. In fact, terms such as ‘neuroscience’ and ‘neuroscientific evidence’ are not mentioned in judgements. Courts only refer to different forms of neuroscientific techniques, e.g. brain imaging, MRI scan, PET scan and neuropsychological testing. It would not be surprising if courts found different types of neuroscientific evidence to be only associated with the fields of psychology and psychiatry. This assumption is supported by the fact that in many cases neuroscientific evidence is discussed by psychologists and psychiatrist (I will discuss the frequency of expert witnesses and their areas of expertise later in this Chapter). It follows that courts may fail to assess whether the relevant scientific community considers neuroscience to be a reliable body of scientific knowledge and they may rather refer the opinions of experts on neuroscientific evidence from the fields of psychology and psychiatry.
With respect to the different types of neuroscientific evidence (and not the field of neuroscience), it appears that courts treat this evidence similarly to other types of expert evidence, such as traditional psychiatric analysis. Courts do not show any concern about the validity of neuroscientific evidence. On the contrary, it seems that they welcome the use of neuroscientific evidence and consider it as admissible and potentially probative.
For instance, in several cases, the court accepted a request for an adjournment to conduct a neuro- examination (e.g. brain imaging), which would assist with determining the existence or extent of
28 Though, see C Guthrie, JJ Rachlinski & AJ Wistrich, ‘Inside the judicial mind’, in Cornell Law Review, vol. 86, 2001, 777–830. 172
any neurological issue.29 In some cases, the court has also specifically indicated that the result of a neuro-examination was important for the court. In one of these cases, for example, where the MRI scan was conducted but the result was not submitted, the court concluded that ‘[o]ne would expect that if those results showed some organic damage to the prisoner, this fact would have been established.’30 This suggests that the court would find that an MRI scan indicating a potential brain impairment has potential probative value in sentencing considerations.
Even in Veen (No. 1), which dates back to 1979, there was no issue raised by the High Court of Australia on the reliability of neuroscientific evidence.31 The positive attitude towards neuroscientific tools was further evidenced in Jacobs J’s explanation of why the psychologist’s report in Veen (No.1), regarding the existence of a brain injury, was not convincing: ‘It is hardly necessary to remark that there had been no neurological examination and no electro- encephalogram [EEG] to support the conclusion [of brain damage]’.32 Thus, neuroscientific evidence was not criticised for insufficient or limited reliability, but was instead welcomed by the High Court. Similarly, Veen (No. 2) supports the assumption that courts find neuroscientific evidence to be a potentially probative type of evidence as the High Court accepted the evidence concerning the existence of a brain impairment and thus its contribution to criminal behaviour and the risk of recidivism.
From the cases outlined above and the analysis of the cases in the dataset of this study, it can be understood that generally the admissibility of neuroscientific evidence broadly conceived does not seem to be a controversial matter in NSW courts and neuroscientific evidence is seen to be useful and relevant to sentencing. Now that we know how courts respond to the reliability of neuroscientific evidence, I will turn to discuss whether courts rely on neuroscience where it is coupled with other types of evidence — the claim of convergence of evidence.
Convergence of evidence: an empirical analysis
This normative claim suggests that brain imaging evidence should be admitted and relied upon where it is coupled with and supported by other types of evidence (i.e. psychiatric, neuropsychological and behavioural evidence).
29 See, Clinton v R [2014] NSWCCA 320 at [5]; R v Sullivan [2010] NSWSC 755; In R v Huon [2012] NSWSC 1092 the defendant did not enter a plea until the results of the MRI scan was determined. 30 R v Paredes, R v Barillaro, R v Donevski [2011] NSWDC 48 at [55] 31 In this case, the conclusion of future dangerousness based on the brain injury and available evidence was criticised for other reasons, e.g. the fact that the psychologist who supported the existence of a brain impairment and the possibility of recidivism examined Veen on only one occasion; whereas the psychiatrists who believed there was no brain impairment met Veen a number of times. 32 Veen (No. 1), p 489 173
The main question guiding analysis of this claim was: ‘does brain imaging evidence independently prove a factual matter?’ That is, whether brain imaging evidence without any support from other types of evidence, prove a fact about a mental state and contributes to the sentencing decision.
Of the 61 cases in which at least one brain imaging evidence was referred to in the judgement, there are 27 cases where neuroscience contributed to the sentence. The analysis of these cases suggests that in all the judgements, brain imaging tools were supported by at least one other type of evidence. In other words, neuroscientific evidence functions as a contributory factor (i.e. buttressing)33 in understanding the mental condition of an individual, rather than as determinative proof (i.e. smoking gun).
In one of the cases,34 the offender and the co-offender were convicted of aggravated robbery after stealing a backpack containing approximately $51,000 from a private mail contractor with the use of (minor) violence.
An MRI scan of the offender35 indicated evidence of brain injury to the frontal lobes (bilaterally) that was caused by a motor vehicle accident.36 In sentencing, the court noted that the frontal lobe injury had an influence on the offender’s ‘personality and reduce[d] the level of inhibition which otherwise might have discouraged him from’ committing the offence.37
If this conclusion was merely based on the brain scan and the existence of a brain injury, then this case would not accord with the prescriptions of convergence of evidence claim. However, there was a variety of other forms of evidence, including behavioural evidence, that suggested and confirmed the influence of the brain injury on the criminal behaviour: following the accident, changes to the personality of the offender were observed.38 A psychiatrist who examined the offender explained his personality changes in the form of a tendency to frustration, demonstrating anger, being restless and overreacting in social situations that occasionally resulted in fights.39 The psychiatrist reported that the frontal brain injury was highly likely to have influenced the offender’s thinking and behaviour at the time of the offence ‘causing him to be swayed by
33 Jones, pp. 181–194. 34 R v Tortell, R v Tsegay [2007] NSWCCA 313 35 Mr Tortell 36 Id. at [32] 37 Id. at [46] 38 Id. at [29] 39 Id. at [32] 174
others’.40 The psychiatrist added that it was probable that after the accident the offender had a diagnosable mental health disorder that led to a neuropsychological impairment. The impact of impairment on the offender probably made him ‘unable to make a proper appraisal of his actions within a proper ethical framework or to make a balanced decision about whether or not to commit the offence in question’.41 Also, according to a neurologist’s report, it appeared that the personality changes and ‘outbursts of anger’ could be explained by the head injury.42
What is apparent from the previous paragraph is that the MRI scan, alongside consideration of the relation between the brain impairment and the personality changes, as well as the offender’s behaviour at the time of the offence, acted as evidence of a causal relationship between the brain impairment and the crime.
These findings raise intriguing questions about whether the evidence in this case other than the MRI scan could have independently met the standard of proof, and there was no need to raise the neuroscientific evidence.43 This assumption could be true, however, that is not the question I am evaluating in this section. This claim raises the question of whether brain imaging evidence independently proves a factual matter (i.e. neuroscience should be coupled with other types of evidence) rather than whether neuroscientific evidence is a necessary – but perhaps insufficient – component in proving an individual’s mental state.44
Another example where the court’s decisions appear to be consistent with the prescriptions of convergence of evidence claim is EG v R.45 In this case, the appellant pleaded guilty to aggravated sexual intercourse with a child under the age of 10 years. The appellant was about 65 years old at the time of the offence. A psychiatrist diagnosed him with mild dementia. This diagnosis was supported by an MRI scan that showed ‘widespread small vessel disease, as well as a stroke, affecting the temporal and parietal lobes of the brain’.46 Further, the psychiatrist presented a number of pieces of behavioural evidence that were consistent with a diagnoses of dementia including ‘the abnormal slowness of the applicant’s answers, his torpid emotional responses and his difficulty in spontaneous generation of a sentence despite prompting.’ The psychiatrist explained that as a consequence of the stroke and widespread cerebrovascular disease the
40 Id. at [33] 41 Id. at [33] 42 Id. at [32] 43 As discussed in Chapter 6 (the gap between neuroscience and the law’s language), Morse refers to this use of neuroscience as rhetorically relevant. That is, neuroscientific evidence is not legally relevant, but makes the case more persuasive. Morse, pp. 132–158. 44 I will briefly discuss this issue in Chapter 10, the conclusion. 45 [2015] NSWCCA 21 46 Id. at [13]. 175
appellant may have carried an impairment of judgement and inhibition that could explain his offending. These issues were consistent with his behaviour that was also observed in the court where he was giving evidence (i.e. he was perplexed by his own behaviour).47 Similar to the previous case, in EG the brain imaging evidence in conjunction with behavioural evidence supported the appellant’s impaired mental state.
As discussed in Chapter 2 (literature review and background), the reliability of functional brain imaging evidence (e.g., fMRI) is more criticised than structural brain imaging evidence (e.g., MRI).48 In Chapter 4 (an introduction to empirical research), I explained that functional brain imaging was only submitted to the court in 11 cases. None of these cases included fMRI scans. Other types of functional imaging include EEG, PET, SPECT and Cerebral Perfusion Study. Neuroscience contributed to sentencing in only four of these cases (involving EEG and PET scan). In four cases it was unclear whether neuroscience had influenced the sentence (involving EEG, PET scan, and Cerebral Perfusion Study) and in three cases neuroscience did not contribute to the sentence at all (involving EEG) (see Table 7-1).
Table 7-1: Types of functional imaging evidence submitted in NSW courts, and whether this evidence contributed to sentencing.
Case citation Type of functional Evidence contributed imaging to sentencing
Ross v R [2006] NSWCCA 65 EEG Yes
R v NAA [2009] NSWSC 1077 PET scan Yes
R v Kennedy [2008] NSWSC 703 PET scan Yes
Regina v Mabbott [2002] NSWSC 502 EEG Yes
Monteiro v R [2014] NSWCCA 277 EEG; PET scan Unclear
R v Elkassir [2013] NSWCCA 181 CPS* Unclear
R v Dunn [2000] NSWCCA 377 EEG Unclear
R v Clay, Lonsdale and JM [2006] EEG Unclear NSWSC 1220
47 Id. at [14]. 48 See, N Rose & JM Abi-Rached, Neuro: The new brain sciences and the management of the mind, in Princeton University Press, EBSCO Publ, Princeton University Press, 2013; GK Aguirre, ‘Functional Neuroimaging: Technical, Logical, and Social Perspectives’, in Hastings Center Report, vol. 44, 2014, S8–S18; A Mechelli et al., ‘Voxel-based morphometry of the human brain: methods and applications’, in Current medical imaging reviews, vol. 1, 2005, 105–113; NJ Schweitzer & MJ Saks, ‘Neuroimage evidence and the insanity defense’, in Behavioral sciences & the law, vol. 29, 2011, 592–607. 176
Philopos v R [2008] NSWCCA 66 EEG No
R v El-Chammas [2009] NSWCCA 154 EEG No
R v Taha [2012] NSWSC 903 EEG No
* Cerebral Perfusion Study
There are a variety of possible reasons as to why functional imaging evidence was submitted to the court in only a few cases: functional imaging is a more recent scanning technique and generally more expensive than structural imaging tests (except EEG). Another possibility is that because functional imaging evidence is subject to more limitations than structural imaging evidence, expert witnesses and lawyers avoid using functional imaging to support their claims. Further study in this area is needed to understand the reason why functional imaging tests are considerably less common in courts, despite the fact that they are considered more advanced imaging techniques.
Generally, analysis of cases suggests that the use of brain imaging evidence accords with the convergence of evidence claim. This means courts admit and rely on brain imaging evidence where it is coupled with other forms of evidence. It would be interesting to know if the courts are purposefully, relying on brain imaging evidence in conjunction with other forms of evidence. In the previous section, I explained that it is unclear whether courts recognise neuroscience as a field of study, or assume that there are other fields of study underpinning brain imaging techniques. I also discussed that generally courts treat brain imaging evidence as similar to other forms of evidence such as psychological and psychiatric evidence and I concluded that judgements do not suggest courts are concerned with the reliability of neuroscientific evidence. These findings suggest that courts are not purposefully relying on brain imaging evidence where it is corroborated by other types of evidence. In other words, it does not appear that the courts are intending to make decisions that are consistent with proponents of the convergence of evidence claim. It is possible that standard practice for legal practitioners requires them to adduce different types of evidence to the court, especially if courts find a legal claim more convincing when evidence is corroborated with other types of evidence. Further, corroboration is not limited to neuroscientific evidence and would be consistent with other forms of evidence as well (e.g. psychiatric evidence that is corroborated by behavioural evidence). This assumption is supported by the idea that each single piece of admissible evidence in criminal proceedings should not be required to meet the legal standard of proof, which allows for corroboration to be evident for different forms of evidence in courts.
177
3. Reliability of neuroscience: the persuasive power of neuroscience
One hazard of submitting neuroscience to the court is that the court may then place too much weight on this evidence and find it to be more probative than can actually be justified. This phenomenon is sometimes called ‘neuromania’.49 Some studies have found that explanations that include neuroscientific information or language are preferred by test subjects.50 Dumit argues that where testimonies were buttressed by brain images, judicial officers would find them to be more objective: ‘there is an undue risk in courtrooms that brain images will not be seen as prejudiced, stylized representations of correlation, but rather as straightforward, objective photographs of, for example, madness.’51 This is important because brain images are not objective evidence and the interpretation of each image may differ from one expert witness to another.52
This issue appears to be more concerning in light of the findings in the previous section, that neuroscience is supported by other types of evidence, as judicial officers may find that neuroscience is even more persuasive than is warranted. In the context of Australia, a study that interviewed judges about the use of neuroscience in reports on fitness to stand trial suggested that some judges find brain imaging evidence more objective than other types of evidence.53 A case study in Australia also found some problematic inferences made by the court concerning neuroscientific evidence.54
Hence, courts may find neuroscience to be more credible that is warranted, and prefer it when it is in conflict with other evidence (e.g. behavioural evidence). Courts may also give too much consideration to neuroscientific evidence in determining the severity of a sentence.
49 P Legrenzi & C Umiltà, Neuromania: On the limits of brain science, trans. Frances Anderson, Oxford University Press, 2011. 50 O Goodenough & M Tucker, ‘Law and cognitive neuroscience’, in Annual Review of Law and Social Sciences, vol. 6, 2010, 61–92; J Aronson, ‘The law’s use of brain evidence’, in Annual Review of Law and Social Science, vol. 6, 2010, 93–108; Brown and Murphy; McCabe and Castel, 343–352; Weisberg et al., 470–477; RE Rhodes, F Rodriguez & P Shah, ‘Explaining the alluring influence of neuroscience information on scientific reasoning.’, in Journal of Experimental Psychology: Learning, Memory, and Cognition, vol. 40, 2014, 1432. 51 J Dumit, ‘Objective brains, prejudicial images’, in Science in Context, vol. 12, 1999, 173–201 (p. 174), quoted in A Alimardani & J Chin, ‘Neurolaw in Australia: The Use of Neuroscience in Australian Criminal Proceedings’, in Neuroethics, 2019. 52 A Roskies & W Sinnott-Armstrong, ‘Brain images as evidence in the criminal law’, in Law and Neuroscience, 2011; Mechelli et al., 105–113. 53 D Rolfe, ‘It wasn’t me, it was my brain’, in LSJ: Law Society of NSW Journal, vol. May, 2016, 26–32 (p. 29). 54 A McCay & CJ Ryan, ‘Issues pertaining to expert evidence and the reasoning about punishment in a neuroscience-based sentencing appeal’, in International Journal of Law and Psychiatry, 2018. The authors examined an Appeal case in South Australia (R v Lepore [2013] SASCFC 13) 178
Examining judgements for the purpose of recognising where (and whether) courts have found neuroscientific evidence more probative than it should be is very difficult, and a number of limitations should be acknowledged in this section. Determining the probative value of neuroscientific evidence in general and in each case context is a challenging task. This is without mentioning the further difficulty of comparing this with the weight the court actually assigns to that evidence and considering factors such as social values and the complicated legal rules that courts are concerned with.55 Further, such assessment requires advanced knowledge in areas such as neuroscience, psychiatry, and behavioural sciences – and almost any conclusion might be controvertible. This is because if evidence from these areas are in conflict, only an individual with such diverse knowledge may be able to make an appropriate assessment of the differing evidence, and even in these situations their expertise and conclusions might be contested. For this study, analysis of these issues is also limited to the materials outlined in the judgement, which does not include the detailed reports from expert witnesses and whether judges misrepresent the evidence and its influence.
In order to illustrate the complexity of courts’ consideration of the probative value of neuroscientific evidence, I will outline different decisions made by an appeal court and trial court concerning neuroscientific evidence.
In R v Pitt,56 the sentencing court did not accept that the mental condition, including the unchallenged indication of frontal lobe impairment showed by CT scans, had a causal relationship with the defendant’s criminal behaviour.57 The sentencing judge stated that: ‘there is a body of evidence before me that supports the conclusion that the prisoner’s cognitive faculties and judgment were, at the relevant time impaired by long-standing alcohol and substance abuse and mental health sequelae.’58 However, the judge later came to the conclusion that, ‘I do not accept that the prisoner’s disorders, such as it might be, or such as they might be, are such that she acted without knowledge of the gravity of her actions or, to put it another way, that the prisoner’s intellectual function is insufficient to allow her a full understanding of the authority and requirements of the law.’59 The court’s only consideration of mental illness was with respect to rehabilitation.
The appellant submitted that the sentencing judge did not sufficiently consider the impact of mental illness. The Court of Criminal Appeal explained that at the time of the offence ‘the
55 See Chapter 3, literature review and background and Chapter 4, an introduction to empirical research. 56 [2005] NSWCCA 304 57 Id. at [27] 58 Id. at [15] appeal court quoting the sentencing court (page 3) (emphasis is removed). 59 Id. at [20] (emphasis is removed). 179
applicant was experiencing an altered mental state due to a number of underlying conditions which caused ‘cognitive impairment, impaired judgement and poor volitional control’.60 The appellate court found that the lower court’s decision did not adequately consider the influence of mental illness by reducing the appellant’s culpability, and therefore quashed the sentence. Similarly, in Hitchcock v R,61 the appellate court quashed the sentence. The leading judgement explained that ‘[m]y principal reason for imposing a lesser sentence is that I have given substantial weight to the causal effect of the applicant’s brain damage on the offending.’62
In both cases, the different inferences of the brain conditions may suggest that appeal courts placed too much weight on neuroscientific evidence (merely because neuroscience is more convincing), or lower courts gave insufficient weight to neuroscientific evidence. Since this study cannot determine the actual probative value of neuroscientific evidence in a particular case, it is not possible to conclude whether courts’ consideration of neuroscientific evidence was inappropriate. Further, the different weight given to neuroscientific evidence in these cases might be considered to be only more of a legal than an epistemic issue. In other words, courts may have made a legal error in their consideration of the moral culpability of the offender.
Another way in which it may be possible to assess whether courts are giving undue weight to neuroscience is to look at cases where neuroscientific evidence is adduced to the court as fresh evidence on appeal. Because in these cases, neuroscience (as fresh evidence) is distinguishable from other evidence, and if it does not affect the sentence, then the case may suggest that the court is not overly persuasive towards neuroscientific evidence. However, similar to the previous examples, the court’s decision may carry only a legal implication rather than epistemic.
In an appellate case,63 the expert, by referring to the fresh evidence, explains that, ‘[the applicant] had a mental incapacity at the time of his offences and that his illness is well known to affect prefrontal brain functions including social (dis)inhibition, moral judgement and other higher cognitive capacities. I have little doubt but that this was a biological contributing factor in the crimes in which he was involved.’64 However, the appellate court explained that the sentencing court had already considered the existence of mental illness and its influence on the sentence. The court continued:
60 Id. at [27] (emphasis is removed). 61 [2016] NSWCCA 226 62 Id. Hoeben CJ at [48]. 63 Ayoubi v R [2006] NSWCCA 364. 64 Id. at [19] 180
In order for this Court to be persuaded that the “fresh” evidence may have had a real bearing on the exercise of the sentencing discretion, something more than an assertion that the applicant’s mental illness was a “biological contributing factor” in the commission of the offences is required. That statement does not allow a court to determine the extent to which the mental illness contributed to the offending behaviour, that is, was it a major contribution, a minor contribution or something in between ? That was the dilemma facing his Honour and the “fresh” evidence does not resolve it.65
The fact that the fresh evidence suggests the offender’s condition had a biological foundation (namely, prefrontal brain impairment) did not persuade the court that the evidence has something new to add. Although this case suggests that neuroscientific evidence is not overly persuasive, it is difficult to make this conclusion because of the limitations of this study and the complexity of legal judgements.66
Accordingly, in this section, I will focus on some examples that require a less subjective assessment of how courts consider neuroscientific evidence. From these examples, it may be easier to determine whether courts are assigning an undue weight to neuroscientific evidence.
As behaviour is the outcome of brain function, and in many cases neuroscience does not correctly predict the outcome of a brain defect, behavioural evidence is generally considered more reliable than evidence from brain images.67 Morse also supports this view and explains that ‘[i]f the finding of any test or measurement of behaviour is contradicted by actual behavioral evidence, then we must believe the behavioral evidence because it is more direct and probative of the law’s behavioral criteria.’68 As such, in the following, I will first discuss where brain imaging evidence is in conflict with behavioural evidence, whether courts find neuroscientific evidence more credible than behavioural evidence.69 Then, I will move on to discuss whether courts refer to neuroscientific evidence as objective evidence, despite the fact that as discussed previously, the interpretation of brain images relies on the subjective analysis of expert witnesses. Further, I will argue Veen’s cases suggest that courts were overly influenced by the persuasive power of neuroscience. Although this part relies on a more subjective approach to examining this claim, discussions are based on several courts’ decisions that provide a plausible argument that suggests undue weight was given to neuroscientific evidence.
65 Id. at [22] 66 For a similar case see Ballard v R [2011] NSWCCA 193. 67 S Morse, ‘Actions speak louder than images’ in Using Imaging to Identify Deceit: Scientific and Ethical Questions, E Bizzi et al. (eds), Cambridge, American Academy of the Advancement of Science, 2009, pp. 23–34; Choi, 278–285. 68 Morse, pp. 132–158 (p. 145). 69 For example, I may not be able to determine the price of an apple and a computer, but I can say the computer worth more. As such, when these two products are offered for free, and recording people prefer which one of these two, I can examine whether their decision is appropriate. 181
At the end of this section I will conclude that through assessment of the cases, except in four instances, there is little to suggest that courts are unduly persuaded by neuroscientific evidence and find it to be more probative than can actually be justified. Nonetheless, the findings in this section should be treated cautiously as the assessment of this claim is limited by the specific approach taken.
Behavioural evidence vs neuroscientific evidence
The results in this study suggest that courts find behavioural evidence to be more reliable than brain images. In R v Blaikie,70 the offender received convictions including possession of housebreaking implements, breaking, entering and stealing, and break and entering the land with intent to steal. The offender sustained head injuries in 1964 and 1984 and a possible head injury in 1991. The court noted that ‘CT scanning and other medical imagining’ plus examinations by neurologists, psychologists and psychiatrists suggested that the damage to his frontal lobe was ‘irrefutable’.71
With regard to the brain damage and offending behaviour, a clinical psychologist reported that the offender’s behaviour was ‘related to his organic brain damage and specifically to his inability to plan and organise effectively and the discrepancy between what he says and what he actually does.’72 However, the court found this conclusion implausible as ‘the prisoner [was] a person who seem[ed] to be able to organise and plan effectively.’73 Other expert witnesses were also unable to find evidence that brain damage was the cause of the offending. Even an expert who acknowledged the existence of an ‘organic brain disorder’ did not propose it as a cause for the offending behaviour.74 The court noted that the submitted evidence, with the exception of that provided by the neuropsychologist, overwhelmingly suggested that ‘whilst there are frontal lobe anomalies they are not of a severity and sufficient magnitude to explain what is called his compulsive stealing, from which precisely he suffers.’75 Accordingly, the behavioural evidence and the evidence given by other experts suggested the offender had the ability to plan the crime, and this led the court to reject the argument that the existence of a brain abnormality had a significant impact on the offender’s behaviour, as stipulated by the neuropsychologist.
70 [2007] NSWDC 311 71 Id. at [43] Frontal lobe injury is associated with impulsive behaviour (see Chapter 2, literature review and background). 72 Id. at [44] 73 Id. at [45] 74 Id. at [48] 75 Id. at [43] 182
In R v Lewis,76 the offender and the co-offender approached the victims who were delivering around $62,000 to a bank. The offender threatened them, took the money, and ran away. Shortly afterwards, the victims saw the offender exiting a hotel with the co-offender, wearing a different shirt.77 The offender was arrested and convicted of robbery.78
A psychiatrist who considered the offender’s behaviour at the time of the offence concluded that there was ‘a substantial degree of impaired cognitive function’.79 An MRI scan was also conducted and this suggested that there was frontal lobe atrophy, which could be the result of either his head injury as an infant, substance abuse, or a combination of these factors. He explained that people ‘with frontal lobe injuries they tend to act on impulse without giving consideration to the consequences of their actions.’80
Based on the fact that the crime occurred during the day near an area where there were numerous police officers,81 the expert explained that such ‘impulsive stupid ill-planned [criminal] acts’ was influenced by drug use, which impaired his judgement and was ‘undoubtedly compounded’ by the existing frontal lobe damage. He concluded that these conditions impaired the offenders’ ability to understand the consequences of his actions, and limited his focus to the immediate gratification resulting from these acts; he was ‘in a state of mind where he would be unable to exercise consideration and restraint. That being said, the expert considered that the defendant had a sufficient level of understanding required to know that what he was doing was wrong. 82
The sentencing judge found that the expert’s suggestion did not fit with the narrative of the crime, ‘Mr Lewis appreciated what he was doing. Mr Lewis realised that this was the way he was going to satisfy his drug debt, and demonstrated his consciousness of his involvement by taking his shirt off, and emerging from the hotel in a blue T-shirt, thus making an effort to alter his appearance.’83
The defendant appealed against the sentence on the basis that the court had erred by failing to consider the influence of brain damage and drug addiction on his judgement, impulsivity and irrationality.84 The appellate court noted that the expert had been provided with evidence that was
76 [2004] NSWCCA 383 77 Id. at [2] to [5] 78 Id. at [1] 79 Id. at [21] 80 Id. at [21] 81 Id. at [21] 82 Id. at [20]; [21]. 83 Quoted by Court of Criminal Appeal at [23]. 84 It was on one of the three grounds of appeal in this case. The other two grounds were not relevant to this study. 183
materially different from what had been put provided to the judge. For instance, the applicant told the expert that he would have ‘blown’ the money if they were successful in the robbery attempt. While it turned out that the judge had received evidence that the applicant was trying to pay back his drug debt. The appellate court also explained that the sentencing judge found that the applicant had physical and emotional problems, irrational behaviour and that the crime was committed with limited planning.85 The appellate court in upholding the sentencing court’s decision concluded that the evidence suggested that it would be implausible to determine that the crime was impulsive.86 It is of note that in this case is that unlike the previous case, no expert witness criticised the results of the neuroscientific evidence. Rather, it was the sentencing judge who made this determination.
Overall, courts appear to find behavioural evidence more credible than neuroscientific evidence. However, since courts do not explicitly make a comparison between these two types of evidence, it is difficult to draw a stronger conclusion or to generalise.
Analysis of these two judgements reveals another interesting point: that if brain imaging evidence suggests an abnormality that may have contributed to the offending, even though this is supported by behavioural evidence, the court may still reject this conclusion or at least reduce the significance attached to the brain imaging evidence through consideration of other persuasive evidence. In other words, in these two judgements, courts did not find brain imaging evidence that was supported by behavioural evidence more convincing than other forms of evidence.
Brain scans are NOT objective evidence
As I have explained, commentators are concerned that the results of brain imaging are conceived by courts as objective rather than subjective evidence. Despite this concern, an overview of the cases in the dataset of this study suggests that there is only one case where it appears that the court has considered a brain scan to be objective evidence. In this case,87 a PET scan indicated organic brain damage, and expert witnesses explained how alcohol and drugs could have intensified the impact that this damage had on his behaviour. The court then, with regard to rehabilitation and the risk of recidivism, acknowledged that the offender was aware of his vulnerability to drug use and alcohol; ‘having been confronted with the objective evidence of his irreversible brain damage causally connected to sustained drug and alcohol abuse over many years’.88 Although this line
85 Id. at [27] 86 Id. at [28] 87 R v Kennedy [2008] NSWSC 703 88 Id. at [54] (emphasis added). 184
suggests that the court found a PET scan to be objective evidence, it also appears from the judgement that the court did not find that the brain damage was a significant factor in sentencing considerations. The court concluded that ‘the overwhelming explanation for the assault was the offender’s self-induced intoxication’.89 The main consideration of the impact of the offender’s brain damage was in relation to their rehabilitation.
There is another case, R v Hussein,90 where the judge considered a brain image to be objective evidence, however this case is not in the dataset of this study as the issue before the court was whether the accused was fit to stand trial. I briefly draw upon this case as it provides a further example of how a court may misinterpret brain imaging evidence as objective evidence.
In this case, several experts were in conflict about whether the accused was fit to stand trial. One of the experts (a consultant in neurology and clinical neurophysiology) conducted three brain imaging tests. An MRI scan and an EEG test did not show any abnormality, but the third test (a Ceretec Brain Perfusion Study), indicated some abnormalities. The expert concluded that the accused might not be able to understand the details of the court proceedings, and they would have problems remaining engaged and making decisions. The court found the accused was unfit to stand trial. The counsel for the Director of Public Prosecutions objected to this decision and explained that other experts with greater expertise in assessing fitness had found the accused fit to stand trial. The suggestion was also made that the accused was trying to deceive the experts. The court explained that the finding of fitness was based on the balance of probabilities, and that even the experts who had found the accused fit to stand trial, had reservations about his fitness. The judge noted that ‘[t]o my mind, there is clear objective evidence from [the neurophysiologist’s] testing of [the accused].’91
The fact that the court found brain imaging to be objective evidence is not the only striking issue in this case. Counsel made several objections to the court decision but did not make any argument on the objectivity of brain scans.
It is possible that even if the court did not find the brain images to be objective evidence, the court’s decision would have been the same. However, if Hussein had been deemed fit to stand trial and found guilty, interpreting the brain image to be objective evidence may influence the severity of the sentence. For instance, it may unduly reduce the consideration of moral culpability, or increase the risk of future dangerousness.
89 Id. at [42] 90 [2011] NSWDC 103 91 Id. at [40] (emphasis added). 185
In general, courts may find brain images to be objective evidence, however, they may not specifically refer to it as objective evidence in judgements. Therefore, the fact that brain images were referred to as objective evidence only in one sentencing procedure does not mean that this phenomenon does not occur in other cases.
Another example where it appeared the courts gave undue weight to neuroscience emerges from the Veen cases, however unlike the cases discussed in this section, this time, no brain imaging evidence was involved.
Veen and prejudicial neuroscience: a cautionary tale about neurolaw
Decisions regarding Veen’s sentencing (in both the first and second manslaughter cases) and medical evidence provided in a later decision where Veen applied for a redetermination of his sentence in 2000, suggest the potential persuasive power of neuroscience in the criminal procedure. While the materials in judgements do not reveal whether any brain scans were submitted to the courts, the significance of neuroscientific findings in explaining criminal behaviour seems to have been exaggerated.
In Veen’s first manslaughter case, despite that the evidence on the existence of any brain injury and high risk of re-offending was controversial, the court sentenced Veen to life imprisonment. Interestingly, this sentence was upheld in the Court of Criminal Appeal, but was later reduced significantly to only 12 years of imprisonment in the High Court. The trial court’s decision was criticised by the High Court in Veen (No. 1) on different grounds. First, the trial court accepted the suggestion that there was a brain impairment based on the report of a psychologist who examined the offender on only one occasion. An experienced psychiatrist (for the prosecution), on the other hand, had met the applicant on several occasions and indicated that the applicant had no impairment of the mind relevant to criminal responsibility and concluded that ‘he had a defect in the form of a character disorder’.92 He also ‘denied the ability of any psychologist to make a diagnosis, in this case, criticised aspects of the psychologist’s test procedures and rejected the conclusion that there was any brain damage’.93 Veen had only one record of violent crime before his first manslaughter, which was not sufficient to argue for a high risk of recidivism. The High Court also criticised the finding by the trial judge that the offender would be a danger to society:
92 Id. p. 464 93 Id. p. 464 186
No process of logical deduction from the legal implications of the jury’s finding is a substitute for the sentencing judge being himself satisfied on the whole of the evidence that the prisoner before him had brain damage which would in the future be likely to lead him to kill or seriously injure other human beings. 94
The High Court’s objection to the decision of life imprisonment suggests that although the existence of brain damage and risk of recidivism were controversial, the trial court and the Court of Criminal Appeal were possibly influenced by the persuasive power of neuroscience and hence found life imprisonment to be a suitable sentence for Veen. In particular, despite the contrary evidence, the trial judge explained that ‘[t]here can be little doubt that the prisoner, if and when released, will, whilst he suffers from this brain damage, be likely sooner or later to kill or seriously injure one or more other human beings’.95
The second occasion where it seems undue weight was given to neuroscience was in relation to Veen’s second manslaughter that occurred within 9 months of his release from prison. The trial court accepted his plea of guilty to manslaughter on the basis of diminished responsibility and the court sentenced him to life imprisonment. The court rejected the view ‘that the appellant was suffering from a paranoid type personality disturbance with the possibility of schizophrenic type episodes from time to time’.96 Instead, the judge accepted the medical evidence suggesting that Veen posed a continuing danger to society due to his brain damage and the fact that when he was drunk or under stress, he would commit a similar crime.97 This judgement was upheld by the Court of Criminal Appeal and Veen appealed to the High Court.
In Veen (No.2), the High Court noted that unlike in Veen (No.1), Veen’s current medical evidence of brain injury, as well as the previous violent killing supported a finding that he was at risk of recidivism. The majority concluded that the circumstances of this case fell within the worst category of cases and that Veen’s mental abnormality (due to brain injury), while reducing the charge from murder to manslaughter, would also mean that he would be a danger to society if he was set free.98
The types of neuroscientific evidence that supported the existence of a brain injury to the satisfaction of the Court of Criminal Appeal and the High Court are not outlined in the case. However, whatever it was, it would have been persuasive enough that there were no reservations about accepting it. Indeed, no objection was made by the majority of the High Court on the
94 Id. Mason J., p. 489 95 Veen (No.1) Jacobs J. quoted the trial judge, p. 487. 96 Veen [1986] CCANSW, p. 224 97 Veen (No. 2) per Mason C.J., Brennan, Dawson and Toohey JJ. 98 Id. per Mason C.J., Brennan, Dawson and Toohey JJ. 187
influence of the brain injury, when they made their conclusions regarding both manslaughter and future dangerousness.99
The controversy in this case arises from the medical evidence that was submitted to the Court of Criminal Appeal, where Veen applied for a redetermination of his sentence in 2000.100 Hereafter, I will refer to this decision as Veen (No. 3).
In the process of this application, the court considered a report submitted by a forensic psychiatrist who noted that:
I think it is reasonable to say that for practical purposes it has been established Mr. Veen does not suffer from brain damage but has a moderate impairment of verbal learning and memory and some difficulty to concept formation, probably for social and educational reasons.101
A neurologist and another expert (whose area of expertise was not specified) also ‘found no evidence for frontal lobe dysfunction and they excluded alcohol or head injury as causes for the impairment.’102 Further, various medical reports that were submitted by the Serious Offenders Review Council in 1997 suggested that Veen had a normal brain rather than an organically damaged brain.103
The judge concluded that: ‘[t]he evidence in the present application establishes, in my opinion, that the applicant does not suffer from brain damage; and probably did not suffer from brain damage at any of the times when he [committed the previous violent crimes]’.104 As such, the court granted his application and re-determined the sentence of life imprisonment to a minimum term of 20 years with the balance of natural life.
Although there are uncertainties relating to the types of examinations that experts had conducted since Veen’s first manslaughter, there are a number of reasons to suggest that the evidence supporting a finding that Veen never had any brain injury was more reliable. The experts and the court in Veen (No. 3) would likely have reviewed the previous medical evidence adduced to the courts during Veen’s first and second manslaughters’ charges, which supported the existence of a brain injury. Nonetheless, it appears that in Veen (No. 3) the court concluded that Veen never had a brain issue. Further, at the time that Veen applied for a redetermination of his sentence,
99 Id. per Mason C.J., Brennan, Dawson and Toohey JJ. 100 R v Veen [2000] NSWSC 656 (7 July 2000). 101 Id. at [117]. 102 Id. at [117]. 103 Id. at [75] – [77]. 104 Id. at [131]. 188
more neuroscientific studies had been carried out and neuroscientific tools had improved along with their potential for diagnosing brain issues.
Combined, these factors suggest that following the second manslaughter case, courts were influenced by the persuasive power of neuroscience. The finding as to the existence of brain damage was made despite other conflicting evidence about Veen’s mental condition. Admittedly, this conclusion has some limitations. It is possible that the medical reports provided to the court during the second manslaughter case strongly supported the existence of a brain injury. In other words, expert witnesses exaggerated Veen’s condition, rather than the court being influenced by prejudicial neuroscience.105 Further, since Veen had already been convicted for the murder of two people, reducing the sentence could have resulted in additional offences, and the courts were under social and political pressure to set a life sentence.
Overall, the findings in this section suggest that except in a few cases, it does not appear that courts place undue weight on neuroscientific evidence. However, the results of this section need to be interpreted with caution, because my assessment of the claim of the persuasive power of neuroscience is limited by the specific approaches taken. I could only evaluate whether courts find neuroscientific evidence to be more credible than behavioural evidence in cases were these two types of evidence were in conflict. Also, while courts may not refer to neuroscience as objective evidence in judgements, it is unclear whether in sentencing decisions, brain images were nonetheless considered as objective evidence.
Beyond the questions of whether courts consider neuroscientific evidence where it is corroborated with other forms of evidence, and whether neuroscientific evidence is overly persuasive, are questions of how those experts who report on neuroscientific evidence, and their qualifications (or lack thereof) may influence the reliability of neuroscience. I will discuss this in the following section.
4. Reliability of neuroscience: expert witnesses
105 For more information about biased conclusions by expert witnesses see, D Reeves, MJ Mills & SB Billick, ‘Limitations of brain imaging in forensic psychiatry’, in The journal of the American Academy of Psychiatry and the Law, vol. 31, 2003, 89–96. Neuroplasticity is another matter to consider, and it is possible that Veen’s brain had healed during his imprisonment where he could not access alcohol. That said, neuroplasticity is a complicated topic and it is unclear to me how a severely damaged brain can heal without any medical assistance. 189
While reliability is not typically an admissibility issue, particularly during sentencing, reliability emerges as a central feature with relation to the use of neuroscientific evidence, where sentencing judges rely may heavily on the opinions and perspectives of experts. Judges invariably draw upon the opinions of experts when trying to apply complicated sentencing rules to the circumstances of individual cases. These difficulties can be both assisted and accentuated by the involvement of experts, particularly where experts disagree and where experts from different fields express opinions. If the evidence is presented by an expert witness whose opinion is beyond their area of expertise, the court may give little or no weight to that evidence. 106
The third claim about the reliability of neuroscience concerns the eligibility of particular types of experts or fields and sub-fields (or specialties) to assist the court. Some scholars claim that some mental health professionals such as psychiatrists, psychologists and social workers are not sufficiently qualified in the area of brain imaging. This is because they are not educated in clinical imaging and consequently do not know when it is appropriate to order a brain imaging examination or how to interpret and use these results (area of expertise claim).107 Despite this claim, there are no guidelines that delineate what qualifications and experiences satisfy the required level of competency an expert witness must have relating to different forms of neuroscientific evidence.108 As discussed in the introduction, since the question of who is qualified to conduct a neuroscientific test, or interpret the results of such tests is a complicated matter requiring normative assessment, the main focus of the following discussion is to assess the areas of expertise of experts called upon to discuss neuroscientific evidence in NSW and, in the section that follows, I will examine whether courts are concerned with expert’s areas of expertise. This assessment will allow evaluation of whether psychiatrists, psychologists and social workers are interpreting brain imaging evidence to courts, and also present information that would provide an opportunity for future studies to further analysis in this area.
Results of assessments in this section are subject to some considerations: the mere title or formal education attributable to an expert (e.g. a psychologist or psychiatrist) does not necessarily determine the expert’s competency. For instance, while it is claimed that psychiatry may not be a suitable area of expertise for discussions on brain imaging evidence, it is important to consider
106 Odgers. Also see WW [2012] NSWCCA 165 at [55]-[60]; Lam [2015] NSWCCA 143 at [74]-[83]. Also, in the recent case Heath v Regina [2016] NSWCCA 24 at [40] the appellate court confirmed that ‘a sentencing court is not obliged to accept unchallenged evidence’, for instance, because ‘it is inherently improbable’. 107 Bremner; Amen and Flaherty, 11; Silva, 489–502. Also see Linden. 108 For instance, in a study examining neuropsychologists’ competency with neuroimaging it is argued that there are no guidelines on how a neuropsychologist might become competent enough to interpret neuroimaging data. See, J Hassenstab & KJ Bangen, ‘Neuroimaging Training Among Neuropsychologists: A Survey of the State of Current Training and Recommendations for Trainees’, in The Clinical Neuropsychologist, vol. 28, 2014, 600–613. 190
the possibility that some psychiatrists may have been trained in some extended courses that allow them to have sufficient working knowledge in relevant neuroscientific matters.109 This is supported by the results of a study examining neuropsychologists’ competency with neuroimaging, which showed that about 70% of neuropsychologists in the study had received training in neuroimaging, the majority through clinical conferences, individualised mentoring and independent study.110 Accordingly, experts may have some training and experience but that does not mean that they are necessarily competent to analyse neuroimages. This empirical study cannot access the expert witnesses’ educational or work background (i.e. CV) nor examine whether the expert who submitted the neuroscientific evidence has the required training. Even if that information was provided in judgements, determining what training programs would make an expert eligible in some areas of neuroscience is a matter that I cannot determine.
Further, in some circumstances, during the criminal procedure, the expert witness may ask other experts such as neurologists or radiologists to conduct or interpret a brain scan on their behalf and report their assessment to them – a matter that may not be mentioned in the expert report or judgement. Another issue associated with the analysis of the results in this section is that occasionally the court only discusses the neuroscientific evidence without reference to any experts or their area of expertise. Also, while sometimes the source of evidence is indicated (e.g. brain scan conducted by a hospital) it is unclear who conducted the neuro-examination, and what area of expertise they fell within. Accordingly, such unclear instances are not considered in the results of this section.
In the following section, first, I will describe types of expertise called upon to assist with different forms of neuroscientific evidence. Then, I will assess the area of expertise claim—whether psychiatrists, psychologists and social workers interpret brain imaging evidence in criminal procedures. Further, I will discuss whether courts express concern about the competency of an expert witness.
109 Linden. It was noted by Freckelton and Selby ‘There is much to be said for an individualised, case by case assessment of psychologists’ levels of specialized knowledge entitling a diagnostic opinion about some psychologists who, by reason of their postgraduate training and/or their experience have ample knowledge to be able to assist courts with their opinions. Others do not fall into that category because of the modest nature of their training or the particular characteristics of their experience, which may not have facilitated the acquisition of specialized knowledge about the diagnostic criteria for particular disorders’ Freckelton and Selby, p. 611. There are also areas where it is not determined in common law (or the uniform evidence scheme) whether psychologists are suitable experts to express opinions about issues such as diagnoses of pathology, and the admissibility of their evidence relating to those issues. See Freckelton and Selby, p. 611. 110 Hassenstab and Bangen, 600–613. 191
Expert witnesses and their area of expertise in courts
The results suggest that experts from 12 areas of expertise expressed their opinion about neuroscientific evidence. Eight types of experts are less frequently involved with neuroscience: neuropsychiatrist, cardiologist, a specialist in geriatric medicine, a specialist physician in rehabilitation medicine, a specialist in addiction medicine, general practitioner, rehabilitation specialist and social welfare worker. In the Table below (Table 7-2) you can see that although these experts discussed all three categories of neuroscientific evidence, i.e. imaging evidence, neuro-tool evidence (e.g. neuropsychological and neurological examination) and non-neuro-tool evidence (e.g. traditional psychiatric evidence and behavioural evidence),111 none of them had discussed neuroscience on more than six occasions. For instance, only three CT scans and three non-neuro-tool pieces of evidence were submitted by general practitioners.
Table 7-2: Areas of expertise with a low contribution to neuroscientific evidence.
However, experts in four areas: psychology, psychiatry, neuropsychology and neurology more frequently expressed their opinion on neuroscientific evidence. A comparison between these four areas indicates that with regard to neuroimaging evidence, psychiatrists had the greatest contribution, both in variety (i.e. CT scan, MRI, PET, SPECT, EEG and CPS) and frequency (see Table 7-3 and Figure 7-1).
111 See Figure 4-1 for a review of different categories of neuroscientific evidence. 192
Table 7-3: Comparison of different expert witnesses’ areas of expertise and the imaging evidence that they submit as evidence.
Figure 7-1: Comparison of different expert witnesses’ areas of expertise and the imaging evidence that they submit as evidence.
In the category of non-imaging tools, neuropsychologists were more frequently involved with neuropsychological and neurological examinations. This is contrasted with psychologists, who more commonly appeared in relation to psychometric evidence (see Table 7-4 and 7-2).
193
Table 7-4: Comparison of different expert witnesses’ area of expertise and the non-imaging results that they submit as evidence.
Figure 7-2: Comparison of different expert witnesses’ areas of expertise and the non-imaging results that they submit as evidence.
Finally, in the category of non-neuro-tool evidence, psychiatrists provided reports more often than other experts (see Table 7-5 and Figure 7-3).
194
Table 7-5: Comparison of different expert witnesses’ areas of expertise and the non-neuro-tool results that they submit as evidence.
Figure 7-3: Comparison of different expert witnesses’ area of expertise and the non-neuro-tool results that they submit as evidence.
The results in this section suggest, contrary to the area of expertise claim – stating that psychiatrists, psychologists and social workers do not have the requisite qualification to interpret brain images – that psychiatrists frequently and psychologists on four occasions appeared in courts to discuss brain imaging evidence.
More generally, I also outlined the areas of expertise that experts are being drawn from to express opinions about different types of neuroscientific evidence. If future studies find that some of the areas of expertise outlined above are not sufficiently specialised to interpret some types of
195
neuroscientific evidence, it would be useful to know whether and the extent to which courts are aware of the required qualifications that experts should hold. Another further area of study would be whether courts consider the reliability of neuroscientific evidence differently based on an expert’s areas of expertise.
Courts’ reaction to areas of expertise
In general, there are only a few cases where courts have expressed their concern about the area of expertise experts should hold when interpreting neuroscientific evidence. Similar to the previous section, I will not argue whether courts’ discussions about areas of expertise are justifiable.
In R v Almoustafa,112 based on the expert witness’s area of expertise, the court rejected (i.e. attached no weight to) their opinion on a psychosocial assessment of the offender, which found that he had an acquired brain injury: ‘From the qualifications supplied in [the witness’s] resume I am unable to determine the depth of her study into brain functioning. Clearly, she is not an expert. She simply comes to the view by examination of documents. Her background seems to be one of social welfare.’113 Instead, it appeared that the court accepted the opinion of a consultant psychiatrist, who reported that there was no evidence supporting the existence of the brain injury.114
In R v King,115 a psychologist, based on the history of an accident and hospital records, concluded that the offender had a severe head injury and that this was consistent with the behavioural changes, namely observed frustration and irritability. Despite giving the report, the psychologist suggested ‘it would be important to obtain the opinion of a neurologist concerning the consequences of traumatic brain injury.’ No neurological opinion was submitted, and the Court of Criminal Appeal concluded that the offender did not prove that he had a brain injury at the time of the offence. As such, the court disregarded the report by the psychologist, which indicated the existence of a severe head injury.
In another case,116 a psychiatrist relying on historical evidence concluded that the offender had a temporal lobe syndrome that could have impaired their ability to control their rage or violence. The psychiatrist found some support for this conclusion by way of an MRI scan provided to him.
112 [2009] NSWDC 217 113 Id. at [38] 114 R v Almoustafa [2009] NSWDC 217 115 R v King [2004] NSWCCA 444 revised - 7/12/2004 116 R v Toki [2003] NSWCCA 125 at [8]-[11] 196
That said, the judgement revealed that the psychiatrist said that he ‘would defer to the opinion of a neurologist in the interpretation of that scan’.117 The Crown called on a neurologist who disagreed with the interpretation of the MRI scan, and noted that it was doubtful that the offender was suffering from a temporal lobe syndrome.118 Subsequently, it appears that the court gave no weight to the opinion of the psychiatrist, in lieu of the neurologist’s report. It is difficult to know whether the court preferred the neurologist’s opinion based on their area of expertise or whether was a matter of the neurologist’s opinion being more convincing in the context of on the other evidence presented to the court.119
While the case discussed above provides an example where it seems the judge gave no weight to the witness’s report, in some instances the court appeared only to reduce the weight given to the expert opinion in consideration of their area of expertise. In one of the cases, the court explained that, ‘[a]lthough I would have welcomed a neurological assessment of [the offender’s] condition, rather than a psychologist’s report … I accept that [the offender] suffers from a number of significant disabilities’.120 As such, in this case the court acknowledged that the offender suffered from some disabilities, but the findings of the report appeared to have been less robust than one delivered by a neurologist.
It is noteworthy that in a few cases the expert witnesses themselves have asked other experts to review the imaging evidence. This is an important safeguard because even if the expert witness’ qualifications and experiences do not qualify them to interpret neuroscientific evidence, the consultant expert may meet those requirements. In R v Kennedy121 the court explained that the PET scan that was provided to the neuropsychologist was prepared by a neurologist and interpreted by a specialist radiologist. It seems that the specialist radiologist was not an expert involved in this case and was only asked to offer an interpretation of the brain image.122
In two of the cases the consultant’s area of expertise was not mentioned. In one of these, a psychiatrist discussed the results of an MRI scan with another expert whose area of expertise was
117 Id. at [10] 118 The Crown called another psychiatrist who, based on a detailed history of the offender, also disagreed with the diagnosis of a temporal lobe syndrome. 119 Id. at [11] 120 R v Clark [No 3] [2008] NSWSC 795 at [30] 121 [2008] NSWSC 703 122 Id at [31] 197
not revealed.123 The same psychiatrist, in another case,124 also consulted another expert whose area of expertise was also unclear.125
It is possible that the expertise of the consulting expert is elaborated in the experts’ original reports. It is also possible that expert witnesses consult with other experts, but judges do not find out or do not find it necessary to report this in the judgement. It was not possible to investigate this issue further, as one of the limitations of this study is the inaccessibility of experts’ reports.
Overall, it seems that none of the publications in the Australian Neurolaw discourse has expressly discussed the areas of expertise of experts that present and interpret neuroscience in court. In general, apart from one case that briefly elaborates on this issue (Almoustafa, discussed above), there are no cases in the dataset of this study where the court discusses the area of expertise required for before experts can express opinions regarding neuroscientific matters.
That said, in a case outside of the dataset of this study, the New South Wales Court of Criminal Appeal126 outlines some conditions about the eligibility of a clinical psychologist to submit evidence with regard to brain damage:
I consider it necessary to observe once again that it is important that clinical psychologists do not cross the barrier of their experience. It is appropriate for persons trained in the field of clinical psychology to give evidence of the results of psychometric and other psychological testing, and to explain the relevance of these results, and their significance so far as they reveal or support the existence of brain damage or other recognized mental states or disorders. It is not, however, appropriate for them to enter into the field of psychiatry.127
Looking back at Table 7-5, psychologists made the greatest contribution in the area of psychometric testing. Therefore, it seems that at least with regard to this type of neuroscientific evidence, the area of expertise of the chosen experts meets the requirements which the Court of Criminal Appeal outlines above.
123 R v Daniel Stephen Lewis [2004] NSWCCA 383 124 FD v R [2013] NSWCCA 139 125 My personal search of the consultant expert’s name showed that she worked at the Hunter Imaging Group where ‘diagnostic imaging and interventional services’ are provided, and that this expert was introduced as a consultant cardiologist. This website also provided that in 2012, the expert updated her MRI skills in areas including neurologic MRI. This suggests that in NSW there are some expert witnesses who receive extended trainings in neuroimaging courses. See https://www.hunterimaging.com.au/department/our-specialists 126 R v Peisley [1990] 54 A Crim R 42 at [52] (New South Wales Court of Criminal Appeal) cited in Freckelton and Selby, Expert Evidence : Law, Practice, Procedure and Advocacy. 127 Id. at [52] (emphasis added). 198
While acknowledging the limitations of this study, the judgements in this study and the information provided in this section suggest two possible outcomes:
It appears that despite the concerns of some scholars, courts are generally not concerned with the precise expertise of expert witnesses. This conclusion is supported by the fact that although the capacity of psychiatrists to express opinions regarding imaging evidence has been questioned, these experts have nevertheless made frequent contributions in interpreting evidence of this nature.
On the contrary, the results in this section may also suggest that courts carefully review the qualification and experience of expert witnesses, and find them satisfactory with relation to the presented evidence. This conclusion also implies that judges have extensive knowledge in areas of neuroscience, psychology and psychiatry required to be able to make that judgement. This does not seem plausible.
Therefore, the first conclusion appears more persuasive. For future studies, this conclusion suggests a need to design guidelines that can help judges to focus on the qualifications and experiences of particular witnesses, and the types of expertise they possess.
Conclusion
A variety of neurolaw studies have critically discussed the unreliability of neuroscience and the use of this evidence in courts. In this Chapter, I have analysed three of these claims:
Many scholars claim that brain images are rarely dispositive of a mental state (e.g. the individual could not control their behaviour) and should be considered in the criminal procedure only where it is corroborated by other types of evidence. This study suggested that in cases where brain imaging evidence contributes to sentencing, this evidence is never taken as sole proof of a mental state. There are always other forms of evidence that support the interpretation of the brain images with respect to the legal question of the mental state. Nonetheless, this study could not come to the conclusion that courts are aware of the limitations of brain imaging evidence and hence, they rely on brain imaging evidence in conjunction with other forms of evidence to be consistent with the prescriptions of the convergence of evidence claim.
Some academics also claim that some courts might find neuroscientific evidence overly persuasive and more probative than it can be justified, and therefore, neuroscientific evidence may unreasonably lead to the exclusion of conflicting evidence, thus exerting an undue influence
199
on the sentence. However, findings in this study suggest that apart from a few cases, courts did not seem to give undue weight to neuroscientific evidence. That said, these findings should be considered cautiously as the assessment of this claim was limited to only some instances that required less subjective analysis of how courts considered the probative value of neuroscientific evidence.
With respect to the area of expertise claim — that psychiatrists, psychologists and social workers are not qualified to interpret brain imaging evidence — the results of the case analysis showed that psychiatrists regularly, and psychologists in a few cases, appeared in court to report on brain imaging evidence. This seems to be more concerning considering that apart for a few cases, courts do not seem to be particularly concerned with the precise expertise of expert witnesses. Compounding this issue, is the fact that there are no guidelines on the qualifications and experiences that satisfy the level of sufficient competency required of an expert witness interpreting neuroscientific evidence. This study encourages scholars to pursue this line of research.
200
Chapter 8 Neuroscience and Rehabilitation of the Offender
201
Introduction
Many studies have discussed different methods through which neuroscience may help with the treatment of an impaired brain.1 The importance of the contribution of neuroscience can be understood in cases where criminal conduct is associated with brain impairment.2 In these cases, treatment may reduce the possibility of re-offending, which may contribute to the protection of society.3
David Hodgson, a former judge of Appeal in the Supreme Court of New South Wales,4 suggests that the greatest potential of neuroscience is in the area of treatment and rehabilitation of the offender:
As scientific research shows what are risk factors for criminal behaviour, how to identify such factors in particular individuals, and how to minimise such factors, we should be prepared to use and to apply this knowledge. Moreover, similarly, as scientific research produces insights into how best to go about rehabilitating offenders, we should be prepared to accept and apply these insights.5
In this Chapter, I examine whether courts’ decisions are consistent with prescriptions of Hodgson’s claim – i.e. whether in sentencing, neuroscience is used for the purposes of promoting the rehabilitation of offenders (neuro-rehabilitation). I conclude that in the majority of cases, courts do not discuss treating a brain abnormality and therefore, the neuro-rehabilitation claim is inconsistent with the use of neuroscience in courts.
This Chapter is divided into two parts. In the first part, I will outline a general review of the different ways that neuroscience is relevant to the rehabilitation of the offender. In the second part, I will move on to discuss the results of this study with respect to the neuro-rehabilitation claim.
1 See B Jancin, ‘Deep Brain Stimulation Helps Lift Refractory OCD Early data suggest that DBS is effective, but adjunctive therapy is indispensable.’, in Clinical Psychiatry News, vol. 36, 2008, 17; E V Cooter, ‘“ Neurodiversion”: incorporating frontal lobe rehabilitation treatment for impulse-control management in the rehabilitation of criminal offenders in Australia’, Bond University, 2015; J Adesegha, ‘The Role of Neuroscience and Psychology in the Criminal Justice System’, 2015. 2 As discussed in Chapter 3 (legal background), rehabilitation is one of the purposes of sentencing (Section 3A Crimes (Sentencing Procedure) Act 1999, NSW), and in Chapter 5 (double-edged sword effect of neuroscience’), I explained how neuroscience is relevant to this area. 3 See SJ Odgers, Sentence, 3rd ed., Longueville Media Pty Ltd, 2015. 4 For more information about David Hodgson see, The Sydney Morning Herald, ‘Judge blessed with flawless logic’, 2012,
1. A qualitative analysis of neuroscience and rehabilitation
Before I discuss the different ways that neuroscience is associated with the rehabilitation of offenders, I will briefly outline a number of considerations in the qualitative analysis of cases in this study.
Treatment of an individual’s disorder should not be confused with rehabilitation. As there may be several factors contributing to criminal behaviour, treating one disorder may not rehabilitate an offender. For example, neuroscientific tools may successfully address a brain condition, however, due to the antisocial personality of the offender, the prospects of rehabilitation may remain low. Nonetheless, because the purpose of this Chapter is to understand whether courts use neuroscience to promote rehabilitation of offenders, if there is evidence in the judgement to suggest that treatment of a brain abnormality may reduce the risk of recidivism, then I consider that judgement to be consistent with the neuro-rehabilitation claim.
Following from complexities in the definition of neuroscientific evidence (see Chapter 4, an introduction to empirical research), there may be argument around what should be considered a neuroscientific rehabilitative approach. Some may suggest that only direct rehabilitative techniques, such as psychosurgery, can be considered to be neuro-rehabilitation.6 Others may argue that indirect rehabilitative methods that are based on neuroscientific findings constitute neuro-rehabilitation.7 Neuro-rehabilitation may include cognitive behavioural therapy with a focus on impaired areas of the brain, or methods that can help offenders deal with their addiction, even where there is no brain injury (see Chapter 2, literature review and background for more information on these techniques).8 Hodgson seems to refer to a broad definition of neuro- rehabilitation. In fact, the neuroscientific method does not seem to concern Hodgson, rather the use of recent knowledge to improve the rehabilitation of offenders is his main focus: ‘scientific research produces insights into how best to go about rehabilitating offenders, we should be prepared to accept and apply these insights.’9
6 See, Greely, 1103–1138. 7 See, L Cozolino, The neuroscience of psychotherapy: Healing the social brain, WW Norton & Company, 2017. 8 For instance, by using functional brain imaging machines to help individuals overcome their addictions. See D Eagleman, The Brain: The Story of You, in Canongate, Pantheon, 2015, pp. 126–129. 9 Hodgson, 661–680 (p. 676). 203
As such, in this Chapter, I consider the broad definition of neuro-rehabilitation, and any discussion in judgements about the rehabilitation of offenders that suggests the use of neuroscientific studies accords with the neuro-rehabilitation claim. The cases used to evaluate the neuro-rehabilitation claim are mainly selected based on the existence of some brain abnormality that suggests future offending. This is because an aspect of the neuro-rehabilitation claim is concerned with treating disorders that reduce the risk of recidivism. Comparatively, neuroscience can also be used for the purpose of treating a physical disability (e.g. following a stroke)10 which is less likely to be relevant to recidivism.
There are several methodological limitations to the analysis in this Chapter, which are shared with the previous Chapters. For instance, in some cases an expert may suggest treatment of the offender's brain, but the court does not refer to it in the judgement and only provides that the expert witness's recommendation will be applied. In this case, the neuro-rehabilitation claim is met, but it is not possible to make such a conclusion on the basis of the judgement alone. Further, since I chose the broad definition of neuro-rehabilitation, which includes rehabilitative approaches that are similar to other methods appropriate for issues such as mental disorders (e.g. depression), confusion may arise in some cases. For instance, if in a case the court identifies both a brain impairment and a mental disorder, and orders a rehabilitation program (e.g. psychotherapy), it may not be possible to identify whether the program is intended to treat the brain or the mental disorder.
With this introduction, I will move on to discuss several cases that show different ways that neuroscience may contribute to promoting the rehabilitation of offenders.
In some cases, the court makes a direct reference to treating the brain. In R v Justin,11 the offender was found guilty of five counts including drug-related offences, possession of a firearm without a licence, and use of electricity without permission. The offender suffered from a traumatic brain injury, and consequently encountered a number of issues including anxiety, poor attention, memory deficits and some behavioural problems, which consequently ended his relationship.12 There was an inconclusive understanding of the offender’s condition and whether it could be fully treated.13 Due to his brain injury, the court gave lower weight to general deterrence, found that he would encounter more hardship than an average prisoner, and ordered special circumstances.14
10 R Lindenberg et al., ‘Bihemispheric brain stimulation facilitates motor recovery in chronic stroke patients’, in Neurology, vol. 75, 2010, 2176-2184. 11 [2013] NSWDC 163 12 Id. at [13-16] 13 Id. at [15] 14 Id. at [16-20]. 204
With regard to the offender’s brain injury and rehabilitation, the court noted that ‘he has good prospects of rehabilitation, provided he is supported over a significant period of time in relation to addressing his psychological and brain injury problems’ and concluded that there would be a low risk of recidivism if he received adequate support.15 Further, the court emphasised treatment of the offender’s brain by recommending that, ‘parole be subject to the conditions that the offender report to the Probation and Parole Service within two working days of release and submit to their supervision in relation to drug rehabilitation (particularly relapse prevention issues), treatment for the traumatic brain injury and other psychological problems.’16
There are few cases where the court gives such explicit consideration to the offender's brain condition and their rehabilitation. In the majority of cases, other factors along with the brain condition that contributed to criminal behaviour are considered. For example, in R v Naa17 the offender was charged with murder, assault occasioning actual bodily harm and use of an offensive weapon. During the incident, the offender stabbed the deceased in her stomach and assaulted other victims. A PET scan showed a brain abnormality that intensified under the influence of alcohol. Accordingly, the expert witness explained that this brain injury would make the offender more aggressive, in particular when he consumed alcohol. The court concluded that at the time of the crime ‘the offender suffered from depression and a brain injury that made him less able to control himself when under the influence of alcohol.’18 The court was uncertain of the offender's prospects of rehabilitation and explained that the offender was a danger to society at the time of the killing; however, the court added that if the offender’s depression was treated, and alcohol use was controlled, he might not be a danger to society anymore.19 As such, the main focuses of rehabilitation were his depression and alcohol use, while it appears attention was not drawn to his brain injury by the court in its approach to rehabilitation.
Another similar example is R v Santos20 where the offender was found guilty of manslaughter on the basis of substantial impairment by an abnormality of the mind due to depression and the sequelae of a brain damage event. The court accepted that general deterrence was not a considerable factor in sentencing this offender. With regard to future dangerousness, the court explained that the offender was unlikely to re-offend if his depression was properly treated.21
15 Id. at 21 16 Id. at [22] 17 [2009] NSWSC 1077 18 Id. at 15 19 Id. at 23 20 [2001] NSWSC 923 21 Id. at [31] 205
Based on the court decisions outlined above (and other similar cases in the dataset of this study) it is plausible to assume that the criminal justice system acknowledges two principles with respect to rehabilitation, or at least the decisions can be explained as follows. The first principle, contributory cause, is that crime is understood as a behaviour, which is multifaceted in nature. This means that crime does not occur due to one factor: instead, it is a result of the contribution of different factors.22 For example, in the case of an individual who has depression, a brain impairment and is alcoholic, the combination of a brain impairment and depression may not lead to a criminal act, but the contribution of all three factors may result in criminal behaviour. As such, instead of treating all contributory factors (brain impairment, alcohol issue and depression), experts may focus on addressing alcohol abuse to sufficiently prevent criminal behaviour.
The second principle, weight effect, is that the degree to which a factor is present may influence the effect of other factors. For example, an individual’s brain impairment may intensify the influence of alcohol on their behaviour. Therefore, not only is the existence of a factor important but the degree to which a factor is present, and the fact that it may affect other factors also need to be considered. As such, in the previous example (depression, alcohol and brain impairment), it is also possible that managing the depression and reducing its symptoms is sufficient in preventing criminal behaviour. Accordingly, it may not be necessary completely to eliminate a contributory factor, rather reducing its effect would be sufficient.
Considering these two assumed principles, the courts’ decision in Naa seems to be justifiable in not focusing on the brain condition, as it appears more pragmatic to address depression and alcohol use than to treat a brain impairment; or at least, that may be how judges view these abnormalities. Similarly, in Santos, the court found that treating the offender’s depression would reduce the influence of the brain abnormality on the risk of future offending (i.e. weight effect).
According to these two principles, it would be expected that in cases where problems such as depression or alcohol abuse cannot be addressed, the court would give more consideration to addressing the brain condition. This does not, however, seem to have been the case in R v McMillan.23 In this case, the offender was found guilty of two counts: aggravated dangerous driving occasioning grievous bodily harm, and aggravated dangerous driving causing death. The offender had previously suffered from a head injury while in prison for another crime.24 He had been convicted of a number of alcohol-related offences, one of which was manslaughter. The
22 Farahany also refers to the interactions between genetics, environment and the brain, and its influence on behaviour. See NA Farahany, ‘Neuroscience and behavioral genetics in US criminal law: an empirical analysis’, in Journal of Law and the Biosciences, vol. 2, 2016, 485–509 (pp. 487–489). 23 [2005] NSWCCA 28 24 Id. at [33] 206
prosecutor appealed against the inadequacy of the sentence, and in the appeal court, it was found that the respondent should have known that alcohol impacts his ability to drive more than other people due to his brain injury.25 The appeal court found that his prospects of rehabilitation were dependent on his ability to refrain from alcohol use. While the court noted that if the offender stayed sober, he was unlikely to offend, the court found that his previous criminal records indicated that he had not learned about being responsible for his alcohol use, and as such, his rehabilitation was unlikely.26
Since the court was not optimistic that the offender could address his alcohol issue, based on the two assumed principles discussed in the previous case (Naa), it would be expected that the court would give more consideration, or would at least discuss, the possibility of treating the offender's brain condition. A possible explanation for this lack of consideration is that treating the brain impairment was not a plausible option. For instance, the expert witness noted in their report that there are no medical facilities in prison capable of treating this issue and that it may not be possible to treat the impairment at all (irreversible brain injury). In such cases, while it might be expected that the court should at least refer to this in the judgement, judges do not discuss all the materials in expert witnesses’ reports (I discussed in Chapter 4 an introduction to empirical research, that this is one a limitation of this study). As such, it is not possible to draw an accurate conclusion on the reason why the court did not consider treating the brain impairment, and it remains a matter of speculation.
Another possible reason for the uncertainties of offender’s rehabilitation is because expert witnesses’ do not provide this information. In Tilyard v Regina27, the offender suffered from brain injuries possibly resulting from alcohol abuse. The appellant appealed against the sentence and one of the grounds was related to the offender's prospects of rehabilitation. The appellant submitted that he had good prospects of rehabilitation. However, the sentencing court found that the appellant had poor prospects of rehabilitation due to his lack of insight into alcohol abuse, stating that ‘[i]t seems people with brain damage or mental illness have very little help’.28
The Court of Criminal Appeal explained that the sentencing judge was entitled to come to that conclusion as two of the expert witnesses did not address the question of rehabilitation and while the third expert explained that the appellant needs a long-term community-based program of
25 Id. at [44] 26 Id. at [45] 27 (2007) NSWCCA 7 28 Id. at [27] 207
rehabilitation to avoid alcohol, he did not comment on the likelihood of rehabilitation in light of the offender’s brain damage.29
In general, there are several possible reasons why the mental condition of the offender, and consequently their prospects for rehabilitation remain undetermined. One is that similar to the case in Tilyard, the expert does not provide such information. This may be due to the narrow focus of the court at specific stages of the criminal procedure. For instance, if the question of fitness is raised in court, the expert will only conduct tests and provide a report that addresses that inquiry. Another possible explanation is that if tests are conducted, but the result is inconclusive, nothing else can be done at that stage. Contrary opinions between experts may also make the issue in question complicated. In particular, in cases where expert witnesses from different areas (e.g. psychologist, psychiatrists and neuropsychologist) provide conflicting opinions. A further issue arises where an expert witness (e.g. psychologist) suggests a possible brain injury that requires brain imaging tests to diagnose, and the defendant cannot afford those tests.
It appears that the most comprehensive evaluation of an offender's mental condition is in the application of Crimes (High Risk Offenders) Act 2006 (NSW) (see Chapter 5, double-edged sword effect of neuroscience). In one such case30 the court made a decision to extend the supervision order. Although there was no reference to the prisoner's brain injury in the case, the court stated that,
[t]he defendant must undergo a comprehensive assessment (and further assessments from time to time) including medical examination, pathological investigations, psychometric testing and radiological imaging to be conducted and/or arranged by the Community Forensic Mental Health Service (CFMHS) or the Area Mental Health Service (AMHS), to determine what is required for treatment, including in respect of the defendant’s potential for alcohol and/or drug abuse and potential for sex offending.31
It is somewhat convincing why the criminal justice system considers the rehabilitation of the offender more seriously where there is a high risk of re-offending, although it includes expensive brain imaging even where there is no indication of a brain abnormality. However, it also raises the question of why this procedure only takes place when the prisoner is about to be released into society? There is no similar comprehensive medical examination when an offender’s prison sentence commences, and there is still time to rehabilitate the offender before the sentence
29 Id. at [28] 30 The State of New South Wales v Johnson [2010] NSWSC 590 31 Id. at [28] (emphasis added). Also see, State of New South Wales v Harrison [2009] NSWSC 198 at 14. In this case, similarly, there is no reference to any brain-related issue and the court made an extension order based on Crimes (High Risk Offenders) Act 2006 (NSW). 208
terminates. In cases such as Tilyard, such practice may reduce the likelihood of an extension order once the sentence ends. However, it could be argued that some prisoners may rehabilitate during the sentence and therefore a costly medical procedure at the end of the sentence would be redundant. Yet, as I briefly referred in Chapter 5, there are critiques of the application of the Crimes (High Risk Offenders) Act 2006 (NSW) (e.g. with relation to principles of Human Rights),32 and the procedural benefit may not be strong enough to outweigh overruling ethical considerations within the criminal justice system.
A further complexity with an offender’s rehabilitation is where the brain issue is found irreversible. For example, in R v McCann,33 the offender was found guilty of manslaughter. A CT and MRI scan along with other tests conducted by expert witnesses showed some brain abnormalities that contributed to the crime. Accordingly, the judge reduced the weight given to general deterrence and retribution. However, the court also added that ‘the irreversible condition of the frontal lobes and the brain generally, raises the question of his dangerousness in the future.’34 With regard to offender’s rehabilitation, while the court found some prospects, due to uncertainties about the treatment of the brain condition (countervailing risk of further offence), it could not be concluded that the offender would not re-offend.35
It is noteworthy that in this case it does not appear that any of the expert witnesses made comments to the effect that this brain condition is irreversible, which would suggest that the judge made this conclusion.36 It is also possible that the court made this assumption based on expert witnesses' reports that were not discussed in the judgement. Analysis of the cases suggests that it is more common that expert witnesses make conclusions about a brain condition being irreversible. For example, in Hitchcock v R37 the expert witness made it clear that the offender’s brain condition is
32 See E Methven & D Carter, ‘Serious crime prevention orders’, in Current Issues Crim. Just., vol. 28, 2016, 227; B McSherry, ‘Throwing Away the Key: The Ethics of Risk Assessment for Preventive Detention Schemes: RG Myers Memorial Lecture 2013’, in Psychiatry, Psychology and Law, vol. 21, 2014, 779–790; G Williams, ‘The legal assault on Australian democracy’, in QUT L. Rev., vol. 16, 2016, 19. 33 [2012] NSWSC 1462. 34 Id. at [36] (emphasis added). There were two psychiatrists who came to the conclusion that the risk of re-offending was low. However, the judge, based on another psychiatrist's opinion, found that that the offender would be a risk to the community upon his release (Id. at [37-40]). 35 Id. at [44]. 36 One of the expert witnesses made notes about the offender’s cognitive condition, but there does not seem to be anything about the brain impairment being irreversible. The expert explained that ‘when a person has a dementing process, as the offender does, he can easily become paranoid and can interpret non-threatening behaviour in a hostile way. [The expert witness] has observed an association between brain atrophy and bipolar disorder. He says that a person with many years of bipolar disorder, especially when it is untreated or unresponsive to treatment, is at high risk of developing cognitive deficits. That does not mean that the offender's bipolar disorder cannot be treated, of course, but his condition has been described as resistant to treatment.’ Ibid at 36. 37 [2016] NSWCCA 226 209
probably untreatable (at [52]): ‘[the offender] has passed the point of maximum spontaneous neurological recovery ([due to his] traumatic brain injuries) and his cognitive functioning is unlikely to improve substantially over time.’38
In fact, while some brain conditions appear untreatable, it would be a concerning matter if judges generally find that there is nothing that can be done in such circumstances in terms of rehabilitating the offender. Brain Injury Australia also discusses this concern and refers to the concept of ‘therapeutic nihilism'. That is an attitude that nothing works for prisoners with acquired brain injury (ABI). Adult corrections and juvenile justice staff have found offenders with an ABI ‘“too hard” a client group’.39
Fortunately, some cases indicate otherwise. In R v Fisher,40 the offender was found guilty of manslaughter, which was due to substantial impairment caused by his damaged brain. The court noted that his condition was permanent, and could possibly deteriorate due to the likely early onset of dementia.41 The court explained that because of the brain injury, the offender’s level of culpability should be considered as low, however, the offender represents a danger in the future (the double-edged sword effect of neuroscience).42 Although the offender's brain condition was found to be irreversible, the expert witness suggested that through a psychiatric assessment and development of a risk management plan, it would be possible to reduce the potential risk of a future violent offence. The court was not certain on whether it was possible to arrange supervision and if the offender would participate in such services. But to facilitate this rehabilitative plan, the court gave some scope to the Parole Board to release the offender into a supervisory environment. Further, the court provided special circumstances under which to increase the parole period. Accordingly, it can be said that in cases where there is a brain impairment, even for permanent conditions, while the prospects of rehabilitation may be reduced, this does not mean that nothing can be done to rehabilitate the offender.
Perhaps surprisingly, contrary to the previous cases, in R v Kennedy,43 the offender’s diagnosis of permanent brain damage indirectly supported the conclusion that there was a higher likelihood of
38 In R v Ray [2013] NSWSC 767 at 71 the court discusses the offender's brain condition and notes that ‘The offender told [the expert witness] that he had received a number of knocks to the head during the course of his life which the doctor opined could have a cumulative effect in causing such injury. There is, however, no remedy or cure. The solution lies in the hope that with advancing age there will be a greater level of maturity, but more so if the offender was to engage in intensive rehabilitation programs in a controlled environment, something the doctor thought that he had the intellectual capacity to do.' 39 Rushworth. p. 27 40 [2009] NSWSC 348 41 Id. at [16] 42 Id. at [16-17] 43 [2008] NSWSC 703 210
rehabilitation. The offender was accused of murder while he was intoxicated at the time of the offence. Following neuropsychological and psychometric tests and a PET scan by expert witnesses, it was concluded that the offender was suffering from brain damage that had not been diagnosed before.44 It was also suggested that the brain damage likely resulted from his chronic use of intoxicants over many years.45 The offender was found guilty of manslaughter on the basis of a combination of his level of intoxication and pre-existing brain damage. The court concluded that self-induced intoxication was the main explanation of the offence and that there was only some degree of casualty between his brain damage and the crime.46 With respect to the offender’s rehabilitation, the court noted that ‘the offender is in the process of gaining a new and encouraging insight into his particular vulnerability to any resumption of drug and alcohol use on his release, having been confronted with the objective evidence of his irreversible brain damage causally connected to sustained drug and alcohol abuse over many years’.47 Subsequently, the court ordered that his prospects of rehabilitation were good and mitigated the sentence on that account.
As such, in this case, the court found that being aware of the existence of irreversible brain damage that was a probable outcome of alcohol and drug abuse, would help the offender to refrain from further use of those intoxicants.
Another interesting way that neuroscience is relevant to rehabilitation of offenders is that in some judgements, brain issues have been found to indirectly prevent the offender from rehabilitation. In other words, while in Kennedy brain issues helped the offender to manage his drug and alcohol use, in some cases brain issues may cause difficulty for individuals to manage those contributing factors to offending (e.g. their alcohol abuse).
This issue is illustrated in Munn v R.48 In this case, the expert witness concluded that a brain injury that had occurred at age 15 had a significant role in the offender’s repeated offending, and led to ‘poor impulse control, lack of ability to evaluate situations before acting upon them, difficulties in delaying gratification, difficulties in utilising feedback on his behaviour, that is failing to learn to modify his behaviour from past experience, difficulties with anger management and mood swings’.49 The judge considered that the expert witness was not optimistic about the offender’s ability to learn and modify his behaviour (i.e. his drug use and criminal history) based on his past conduct. As such, the brain impairment in two ways precluded offender’s rehabilitation: once
44 Id. at [5], [6], [31-37]. 45 Id. at [6]. 46 Id. at [42]. 47 Id. at [54]. 48 [2009] NSWCCA 218 49 Id. at [51] (emphasis added). 211
directly reduced the prospects of rehabilitation (e.g. resulting in poor impulse control), and once indirectly, by preventing the offender from learning how to control his conduct (e.g. abstain from drug use) and therefore to rehabilitate.
The detrimental impact of brain impairment on rehabilitation is a matter that has been recognised for more than two decades ago and it is not limited to brain issues and also include some psychiatric disorders. In R v Davis (1999),50 the expert witness clearly expressed this matter, as the offender ‘does not have a psychiatric disorder or brain damage likely to interfere with his rehabilitation.’51 This would also support the claim that courts view neuroscientific evidence similarly to other forms of evidence such as traditional psychiatric assessments (see Chapter 7, concerns about the reliability of neuroscientific evidence in courts).
Overall, analysis of the cases suggests neuroscience and rehabilitation intersect in different ways, and while in some cases there is a clear and direct link between them, in some cases this connection is more complicated. In the majority of cases in the dataset, rehabilitative approaches were discussed to address other factors (e.g. addiction) that along with a brain condition contribute to criminal behaviour. Consequently, there were no discussions of neuro-rehabilitative methods, in particular, those using advanced neuroscientific methods (i.e. invasive and non-invasive techniques). Further, it appeared that courts discuss using neuroscience in an offender’s rehabilitation only if there is some brain abnormality.52 Therefore, there were no judgements where, for example, courts discussed how neuroscience could help offenders with addiction.
Having reviewed some cases on the relevance of neuroscience in rehabilitation, I will now move on to discuss whether the Neuro-rehabilitation claim is satisfied in practice.
2. Analysis of the neuro-rehabilitation claim
Analysis of cases to examine whether neuroscience is used to promote the rehabilitation of offenders — i.e. whether the use of neuroscience in courts is consistent with the neuro- rehabilitation claim —suggests that in the majority of cases where a brain impairment exists and contributes to the risk of future offending, treating the brain is less likely to be the main focus of the offender’s rehabilitation. Instead, other factors that along with the brain conditions contribute to criminal behaviour are considered, and the neuro-rehabilitation claim is not consistent with the
50 [1999] NSWSC 876 51 Id. at [44], quoted by the court. 52 If neuroscience is used for other purposes, it is not discussed in the judgment 212
use of neuroscience in courts (for example, see Naa). 53 Notwithstanding, I discussed in the previous section that according to the two principles of contributory cause and weight effect, this approach to rehabilitating offenders is to some extent justifiable. Conversely, in cases such as McMillan, where it seems it was not possible to address other contributory factors, and the court did not discuss treating the brain condition, the use of neuroscience in court does not accord with prescriptions of the neuro-rehabilitation claim.
In cases where the court finds the brain condition to be irreversible, there is less clarity as to whether the use of neuroscience is consistent with neuro-rehabilitation claim. Whether the brain condition is in fact irreversible, or whether it is the expert witness’s opinion or the judge’s way of understanding a brain condition is a matter that I cannot assess. If they are in fact irreversible, then it could be concluded that the neuro-rehabilitation claim is not applicable; because the court could not order a treatment program for an irreversible condition. However, if the brain condition is reversible, but the court or the expert concludes otherwise, some may argue that we should not blame them for the lack of knowledge that leads to an incorrect diagnosis, or a court order that does not appreciate the Neuro-rehabilitation claim. On the other hand, some may argue that this claim is about embracing neuroscientific technology to rehabilitate criminals more effectively. If courts and experts do not attempt to keep pace with advancing technology and knowledge, then this claim is not consistent with the prescriptions of the neuro-rehabilitation claim.
We may be able to criticise expert witnesses for a misdiagnosis, as they are expected to disclose if the fact in issue is beyond their area of expertise. However, it is difficult to criticise judges who are responsible to address legal questions for not being familiar with a recent scientific area. It should then be the criminal justice system’s responsibility to educate judges, and consequently so that they can order decisions in accordance with the neuro-rehabilitation claim.
In some other cases, the court cannot make an order with respect to treatment of offender’s brain abnormality because the criminal justice system lacks the facilities to rehabilitate offenders with such conditions.54 In cases where the court recognises the necessity of treating the offender’s brain, but is also aware that the criminal justice system lacks the facilities to deal with that issue, courts can find that a special circumstance arises. The NSWLRC pointed out that,
[g]enerally, courts have found special circumstance to exist where offenders would be likely to benefit from extended supervision in the community, and require a longer period on parole in order to reintegrate more successfully.
53 [2009] NSWSC 1077 54 See N Rushworth, ‘Out of sight, out of mind: People with an acquired brain injury and the criminal justice system’, in Brain Injury Australia, 2011. 213
Similarly, a shorter non-parole period may be warranted where an offender would benefit from rehabilitation or treatment services in the community.55
Accordingly, finding that a special circumstance exists, which extends the parole period, and can allow prisoners with brain injuries to access suitable treatment programs, seems to be an approach used to bypass issues relating to insufficient services.56 For instance, in cases such as Justin57 (discussed above), the court found special circumstances existed, and recommended that the parole was subjected to the offender’s condition, including treatment for his traumatic brain injury.58 As such, cases similar to Justin are consistent with the neuro-rehabilitation claim. However, in many cases where courts find special circumstances to exist, they either do not outline, or only partly justify the reasons behind their decision, and therefore, it is unclear if these cases are consistent with neuro-rehabilitation claim.
Overall, the analysis of the cases with relation to the neuro-rehabilitation claim reveals that there are many cases where the court disregards the treatment of a brain impairment that may be a risk factor leading to an offender's future recidivism. However, blaming courts for not making rehabilitative orders is also inappropriate. In some cases, there are constraints that prevent the court from ordering rehabilitation (e.g. where a brain impairment is irreversible) resulting in the neuro-rehabilitation claim being not applicable. Also, as discussed in earlier Chapters, there are some limitations that may skew an accurate analysis of neurolaw claims. For instance, it is possible that while an expert witness may explain in detail that due to lack of facilities in prison, treatment of offender's brain damage requires community-based rehabilitation, the judgement provides for special circumstances without clarifying that it is to support the offender's brain treatment.
Conclusion
This Chapter aimed to examine whether courts use neuroscience with the purpose of promoting rehabilitation of offender — i.e. whether the use of neuroscience in courts is consistent with the neuro-rehabilitation claim.
55 Quoted in D Howard & B Westmore, Crime and Mental Health Law in New South Wales: A Practical Guide for Lawyers and Health Care Professionals, 2nd ed., LexisNexis, AU, 2010, p. 546; NSWLRC, Consultation Paper 6 (2010), note 4 above, at [8.51] - [8.54] (notes are omitted). 56 In R v El-Hayek (2004) 144 A Crim R 90 at [105] the court held that: ‘A finding of special circumstances is purposive in that it is a warrant for reducing the non-parole period below that arrived at by the statutory formula. While the considerations or the factors giving rise to special circumstances are not limited to the reform of the offender, that will often be the purpose in finding special circumstances’. 57 [2013] NSWDC 163 58 Id. at [22]. 214
The result of analysing the cases suggests that only in a few cases did courts directly consider rehabilitating the offender by treating the brain abnormality, and that these cases are consistent with the neuro-rehabilitation claim. The majority of the cases, however, were not consistent with the neuro-rehabilitation claim; meaning, courts do not discuss treating a brain abnormality and instead merely discuss the rehabilitation of the offender with regard to other risk factors (e.g. alcohol abuse). One reason for this decision is that it is less complicated and costly to address other risk factors rather than the brain impairment. Also, in some cases courts find the brain condition to be irreversible, and therefore the only option is to address other factors. In these cases, the neuro-rehabilitation claim is not applicable; because the aim of this claim is to increase rehabilitation and reduce recidivism by using neuroscience, it appears that in these cases there is nothing that can be done to treat the brain.
Generally, it seems that courts give the most consideration to rehabilitating criminals when the Crimes (High Risk Offenders) Act 2006 (NSW) applies. A possible explanation for this is that in some cases, it appears that the defence does not request a mental examination, for example, because of a defence strategy or the cost of those services. However, an order under this Act is a matter that is issued by the court and the medical examination is paid by the criminal justice system, which seems to be an important element resulting in its prevalence.
Overall, Hodgson argues that neuroscience—in accordance with traditional legal principles—can provide more information for courts, make rehabilitation of offenders more effective, and reduce rates of recidivism. However, while some scholars also encourage the use of neuroscience for the purpose of rehabilitation, unlike Hodgson, they argue that neuroscience will change the application of legal principles. In the next Chapter, I will discuss this claim; how neuroscience can change the application of legal rules.
215
Chapter 9 Neurolaw of the Mechanical Brain
216
The cosmos turned out to be larger than we had ever imagined from gazing at the night sky. Similarly, the universe inside our heads extends far beyond the reach of our conscious experience … At this moment, just like every moment of your life, networks in your brain are buzzing with activity: billions of electrical signals are racing along cells, triggering chemical pulses at trillions of connections between neurons. Simple acts are underpinned by a massive labor force of neurons. You remain blissfully unaware of all their activity, but your life is shaped and colored by what's happening under the hood: how you act, what matters to you, your reactions, your loves and desires, what you believe to be true and false. Your experience is the final output of these hidden networks. So who exactly is steering the ship? — David Eagleman1
Introduction
In this Chapter, I will discuss the claim by Joshua Greene and Jonathan Cohen (2004) which predicts neuroscience will change the application of legal principles with respect to responsibility and how the criminal justice system deals with offenders and punishment. They argue that free will is an illusion and the growing neuroscientific knowledge will change the society’s intuition about criminal responsibility by indicating that behaviour is a result of the mechanistic function of the brain, rather than of our free will. As such, offenders do not deserve to be punished for their conduct (retribution), and courts will merely focus on the consequentialist aspect of punishment such as preventing recidivism. This Chapter aims to reveal whether the predictive claim made by Greene and Cohen has already occurred, or is taking place.2
This Chapter is divided into four sections. First, in order to better understand Greene and Cohen’s claim, I will briefly introduce some underlying mechanisms of the brain, which explain how neuroscience supports the notion of human agency without free will. Then, I will discuss arguments about the legal implication of a mechanistic brain in some detail.
In the second section, I will consider the because Greene and Cohen’s claim is about a shift from retribution to a consequentialist approach to punishment, examining the claim requires determining which sentencing factors are retributive and consequentialist. Therefore, in the third section, I will turn to consider whether Greene and Cohen’s definition of retributive and consequentialist sentencing factors are in accordance with how they are defined in the NSW
1 D Eagleman, The Brain: The Story of You, in Canongate, Pantheon, 2015, p. 68. 2 For contradictory opinions see, S Morse, ‘Lost in Translation?: An Essay on Law and Neuroscience’ in Law and Neuroscience, Current Legal Issues, M Freeman (ed), Oxford University Press, 2011, pp. 529– 562. cited in JA Chandler, ‘The use of neuroscientific evidence in Canadian criminal proceedings’, in Journal of Law and the Biosciences, vol. 2, 2016, 550–579. Also see, HT Greely, ‘Neuroscience and criminal justice: not responsibility but treatment’, in University of Kansas Law Review, vol. 56, 2008, 1103–1138. 217
jurisdiction, and discuss any differences and ambiguities.
In the third section of this Chapter, I will introduce the evidence, which apart from one case that seems to support Greene and Cohen’s claim, suggests courts’ sentencing practices are in accordance with traditional legal principles and therefore, inconsistent with Greene and Cohen’s claim. As the claim is predictive, in the last section, I will also discuss whether there are any factors (e.g. Acts) that increase the possibility of Greene and Cohen’s predictive claim actualising in the future.
1. A mechanical brain: legal implication
Scientists claim that what we see, hear, smell, sense, and taste are attributable to internally generated impressions of the outside world.3 The real world has no colour. Our brains interpret different wavelengths of electromagnetic radiation to ascribe colour to objects. Further, through a process inside our brains, feedback from incoming visual information is supplemented. More specifically, our brains, outside of our conscious control, use previous experiences of that context to make guesses about the condition of the outside world.4
Our brain’s unconscious interpretation of the outside world is not limited to sensory perceptions. The brain plays a major role in decision-making. The extent to which the brain’s unconscious mechanics influence people’s everyday life is far beyond what we can imagine. It works to create new ideas by recalling memories and combining them before they even come into our awareness.5 A study that analysed 1,112 judicial decisions of Israeli parole boards by eight judges found that prisoners are up to 65% more likely to receive parole after judicial officers have a food break (such as a court recess for lunch) than they are at the end of the session.6 The influence of food breaks on judges’ decisions raises questions regarding the extent to which our brains unconsciously shape our decision-making, and what role our conscious mind plays in that decision-making processes. Are our decisions actually shaped, influenced or determined through uncontrollable mechanical processes in our brains that determine how we behave and feel, while we remain mere observers of the eventual outcome? Is free will merely an illusion created by our cognitive design?
3 C Frith, Making up the mind: How the brain creates our mental world, in Blackwell Publication, John Wiley & Sons, 2013, chap. five. 4 Eagleman, The Brain: The Story of You. 5 Eagleman, The Brain: The Story of You. 6 S Danziger, J Levav & L Avnaim-Pesso, ‘Extraneous factors in judicial decisions’, in Proceedings of the National Academy of Sciences, vol. 108, 2011, 6889–6892 cited in D Eagleman, Incognito: The Secret Lives of the Brain, New York City, U.S., Pantheon, 2011. 218
Different studies and discoveries throughout the history of neuroscience have supported this vision of the mechanical basis of the brain (see Chapter 2, literature review and background). These scientific studies have led to claims that neuroscience would change the way we understand our own humanity.7 Simultaneously, they stimulated questions such as whether we can still hold the individual responsible for actions determined by the machinery of the brain, and if the brain’s mechanistic process controls individual’s conduct (or even a portion of it), what consequences should this have for criminal law?
David Eagleman, a neuroscientist and founder and the head of the Centre for Science and Law (Houston, Texas), suggests that, ‘[f]ree will may exist—but if it does, it has very little room in which to operate … [it] is only a small factor riding on top of enormous automated machinery’8 and therefore, human behaviour is more closely related to the biological makeup of individuals, rather than their free will.9 Eagleman refers to Charles Whitman’s story to illustrate the extent to which our actions are dependent on our biology. In 1996, Whitman visited the observation deck of the University of Texas Tower in Austin where he went on a violent rampage that resulted in the death of 13 people and left 33 wounded. Police discovered that before the shooting, he had murdered his wife and his mother. In his suicide note, Whitman requested that an autopsy be conducted on his brain, as he had recently been conscious of unusual thoughts. His autopsy led to the discovery of a brain tumour in an area of the brain that is correlated with aggression, providing a possible explanation for his criminal conduct. Eagleman concludes that, in this case, it is arguable that blame should be attributed to the tumour, rather than Charles’s free will.10
In the context of criminal law and criminal responsibility, Eagleman explains that while many criminal acts cannot be explained by biology at this point in time, scientific inquiry has been
7 MJ Farah, ‘Neuroethics: the ethical, legal, and societal impact of neuroscience’, in Annual review of psychology, vol. 63, 2012, 571–591, cited in C O’Connor & H Joffe, ‘How has neuroscience affected lay understandings of personhood? A review of the evidence’, in Public understanding of science, vol. 22, 2013, 254–268. Also see, World Science Festival, ‘What It Means to be Human’, YouTube, 2014,
uncovering more and more biological explanations for criminal behaviour (i.e. tumours, schizophrenia, and epilepsy), and that such conditions may eventually mitigate the culpability of the individual. Accordingly, he claims that it is not reasonable to punish an individual for actions that may be explained by scientific insight in the future, and calls into question the morality of criminal punishment.11
Eagleman goes on to propose adopting a forward-looking response (i.e. consequentialism) to criminal behaviour in which rehabilitation, rather than retribution, is the main objective. Punishment should be utilised only where it is useful and has the capacity to modify the individual’s behaviour (e.g. specific deterrence). If there is no rehabilitative approach that can modify the individual’s behaviour, and punishment will not afford any benefit in this regard, then the state may have to keep the individual under control to incapacitate them for the protection of the individual and the society rather than as retribution.12
Similarly to Eagleman, Greene and Cohen (2004) have claimed that:
[A]dvances in neuroscience are likely to change the way people think about human action and criminal responsibility … Free will as we ordinarily understand it is an illusion generated by our cognitive architecture. Retributivist notions of criminal responsibility ultimately depend on this illusion, and, if we are lucky, they will give way to consequentialist ones, thus radically transforming our approach to criminal justice.13
One Australian scholar, Stephen Odgers, a barrister with a background in law reform, also draws on the potential for neuroscience to change the practice of sentencing. In his book on sentencing in NSW courts, he states that ‘[i]t is increasingly clear that developments in [neuroscience] are likely to provide clear links between neurobiological detriment and certain types of offending.’14 Considering neuroscience is still developing, and the law is typically conservative in following scientific developments, Odgers claims ‘[t]here are reasons to believe that sentencing will be a very different exercise a few decades from now’.15
In contrast, the legal scholar Stephen Morse claims that neuroscience will not change anything in terms of legal principles. In Chapter 6 (the gap between neuroscience and the law’s language), I outlined Morse’s view on the differences between neuroscientific and legal language and that ‘[i]f
11 Eagleman, Incognito: The Secret Lives of the Brain, p. 176. 12 Eagleman, Incognito: The Secret Lives of the Brain. 13 J Greene & J Cohen, ‘For the law, neuroscience changes nothing and everything’, in Philosophical Transactions of the Royal Society B: Biological Sciences, vol. 359, 2004, 1775–1785 (p. 1784). 14 SJ Odgers, Sentence : the law of sentencing in NSW courts for State and Federal offences, 3rd ed., Haberfield, N.S.W, Longueville Media, 2015, p. 594. 15 Odgers, p. 594. 220
any science is to have appropriate influence on current law and legal decision-making, the science must be relevant to and translated into the law’s folk-psychological framework’16 He argues that criminal law is indifferent to whether an individual has free will. It simply requires a minimal rationality, which makes individuals responsible for their actions (i.e. law’s folk psychological criteria).17 In other words, the law does not ask if a person had ‘free will’ at the time of crime. Instead, it asks, whether the individual could ‘understand’ their actions or ‘control’ their behaviour as a rational person could. Morse explains that neuroscientific studies do not prove that people are irrational, which is what the law is concerned with, rather neuroscience research suggests that our conduct is the result of mechanical processes in our brain that are beyond our conscious control. For Morse, neuroscience should not change relevant legal principles and practices.18
Morse adds that ‘the fundamental psychological error’ occurs where it is assumed ‘causation, especially abnormal causation, is per se an excusing condition’.19 Otherwise, since all behaviour is caused by the brain, all criminal acts would imply a legal excuse from criminal responsibility. As such, unless a brain condition sufficiently diminishes a person’s rationality, the brain cannot be a factor relevant to criminal responsibility.20
Greene and Cohen accept that criminal law is not concerned with physical causation in the brain. Their argument is from another perspective. Most people are not concerned with whether an individual was sufficiently rational at the time of the crime and diminished rationality is ‘just a presumed correlate of something deeper’.21 Instead, what most people are concerned with are the factors that cause a person to act in a specific way: was it the person who committed the crime, or was it the result of bad upbringing, their genes, or their brain?22
Greene and Cohen demonstrate the difference between legal categories, and what society judges as salient factors to a crime and punishment, in the hypothetical case of Mr. Puppet.23 A group of scientists design a person (Mr Puppet) by manipulating his genetics and environment (i.e. his upbringing, friends and the circumstances of his life) to create a specific personality type and behavioural profile. A scenario was then created and a question was asked ‘if Mr Puppet commits
16 S Morse, ‘Avoiding Irrational NeuroLaw Exuberance: A Plea for Neuromodesty’, in Law, Innovation and Technology, vol. 3, 2011, 209–228. 17 Morse, 209–228. 18 Morse, 209–228. 19 Morse, pp. 157–198 (p. 180) quoted in Greene and Cohen, 1775–1785. 20 Morse, pp. 157–198. 21 Greene and Cohen, 1775–1785 (p. 1778). 22 Greene and Cohen, 1775–1785. 23 Greene and Cohen, 1775–1785. 221
a crime, would we find him responsible?’ While Mr Puppet is a rational creature, his whole life has been designed and controlled by scientists, who had created the person he was at the time of the imagined crime. According to Morse, criminal law would find him responsible (because he was rational at the time of the crime), but Greene and Cohen argue that our society would not. Greene and Cohen claim that since the law is shaped by the moral intuitions and commitments of its society, neuroscience can change our approach to criminal justice by shifting those intuitions.24
Greene and Cohen do not make it clear what they mean by ‘society’. People in society can vary in terms of the social influences that impact upon them such as culture and background, and thus different groups of people with distinct ideologies would appreciate scientific discoveries and value scientific inquiries differently. For instance, as some religions value the notion of free will and moral responsibility,25 some religious groups might not turn away from these beliefs despite scientific findings showing otherwise.26 While Greene and Cohen attempt to justify such criticisms, their arguments are complicated, the analysis of which requires lengthy discussion that is not within the scope of this research or the aim of this Chapter. 27
Greene and Cohen’s claim has been one of the central arguments in the literature of neurolaw. This is evidenced by the considerable number of citations to their paper ‘For the law, neuroscience changes nothing and everything’.28 While other scholars consider how neuroscience may influence the punishment of the offender (e.g. double-edged sword), or how sentencing aims may have a more therapeutic outcome (e.g. rehabilitation), using traditional principles of criminal law, the significance of Greene and Cohen’s claim lies in the fact that they argue neuroscience can
24 Greene and Cohen, 1775–1785. 25 For example, see, A Alimardani, ‘Neuroscience, criminal responsibility and sentencing in an islamic country: Iran’, in Journal of Law and the Biosciences, 2018,
change the application of legal principles entirely. The resulting shift away from a retributive approach and towards consequentialism represents a radical change to criminal law policies that might be understood as revolutionary.
Although Greene and Cohen’s claim is controversial and subject to many criticisms, the significance of their predictive claim suggests merit in evaluating it in practice. Accordingly, this Chapter aims to gauge whether the predictive claim made by Greene and Cohen, that neuroscience would change the application of criminal law principles and that the retributive aim of punishment will give way to a justice system that is constructed only on consequentialism, has already occurred or is taking place in criminal courts of NSW.
To analyse Greene and Cohen’s claim, in judgements I should evaluate whether courts disregard retributive sentencing factors and only consider the consequentialist approach. Therefore, it is necessary first to outline which sentencing factors are ‘retributive’ (backward-looking) and which factors are ‘consequentialist’ (forward-looking).
2. Retributive and consequentialist sentencing factors
Greene and Cohen recognise blameworthiness, just deserts and proportionality as part of a retributive approach to punishment.29 With regard to a consequentialist approach, Greene and Cohen refer to the protection of society and deterrence. However, unlike Eagleman who emphasises the importance of rehabilitation, Greene and Cohen do not refer to rehabilitation, reform, or treatment of offenders.
All the sentencing factors that may be relevant to neuroscience in the Australian criminal justice system are outlined in Chapter 5, the double-edged sword effect of neuroscience (see Figure 5- 4). There are some sentencing factors that are not mentioned by Greene and Cohen, and it is unclear whether they should be considered retributive or consequentialist. I have divided these sentencing factors into three groups for the purpose of this Chapter (see Table 9-1).
29 Greene and Cohen, 1775–1785. 223
Table 9-1: Retributive and consequentialist sentencing factors according to Greene and Cohen (G & C) and the NSW criminal justice system.
The first group are those factors that certainly align with what Greene and Cohen define as retributive and consequentialist. Retributive factors include proportionality (ensuring the offender is adequately punished for the offence),30 the objective seriousness of the offence,31 and retribution32 itself. While Greene and Cohen do not mention ‘moral culpability’ as one of the retributive factors, as discussed in Chapter 3 (legal background), moral culpability directly correlates with retribution. Odgers with respect to retribution and the moral culpability of offender states that ‘retribution tends to focus on the harm done by the offender but … it also recognizes that an important consideration is the offender’s level of culpability’.33 Further, Greene and Cohen discuss the concept of ‘blameworthiness’ with relation to retribution, and I explained in Chapter 3 that blameworthiness directly correlates with moral culpability and retribution.34
The consequentialist approach to punishment includes factors such as the protection of society/the probability of re-offending,35 specific deterrence, and general deterrence.36
The second group are those sentencing factors that appear to be retributive and consequentialist in the NSW criminal justice system, but are not outlined by Greene and Cohen, and unlike ‘moral culpability’, it is not clear whether they would consider these sentencing factors as retributive or
30 s 3A(a) of the Crimes (Sentencing Procedure) Act 1999 31 Kennedy v T (2008) 181 A Crim R 185; [2008] NSWCCA 21 at [28] and [39]. 32 See Odgers. 33 Odgers, p. 73. 34 Also see, See Shi v R (2014) NSWCCA 276 at [21]; Ansari v R (2010) 241 CLR 299; [2010] HCA 18 plurality at [59]. 35 ss 3A(c) and 21A(3)(g) Crimes (Sentencing Procedure) Act 1999 36 s 3A(b) Crimes (Sentencing Procedure) Act 1999 224
consequentialist approaches. Accountability,37 denunciation,38 and the offender’s awareness of the consequences of their actions39 appear to be backward-looking or retributive but are not discussed by Greene and Cohen. Also, as mentioned previously, prospects of rehabilitation seem to be forward-looking or consequentialist40 but this is not discussed by Greene and Cohen.
The third group are those sentencing factors that are not discussed by Greene and Cohen, and in the NSW criminal justice system may be associated with either retributive or consequentialist approaches to punishment: custodial hardship,41 and the influence of punishment (i.e. imprisonment) on an offender’s health.42 For an individual with some brain impairment, for example, prison may be disproportionately onerous, and the court may reduce the sentence in an attempt to make the punishment proportionate to what is deserved (backward-looking). However, in another judgement, the court may reduce the sentence because imprisonment would have a considerably negative influence on the prisoner's health and reduce their prospects of rehabilitation (forward-looking). Therefore, whether these sentencing factors are considered retributive or consequentialist, will depend on the circumstances of the case and the offender.
One way to evaluate Greene and Cohen’s claim would be only to consider the first group of sentencing factors that are defined by Greene and Cohen. They do not consider the second and third group of sentencing factors, but this study focused on an NSW jurisdiction, seeks to test the claim against all of the factors. Accordingly, I test the claim in two stages. Initially with the first group of sentencing factors, and then with all the three groups. Since Greene and Cohen claim that there would be a change in the application of legal principles of disregarding retributive aspects in sentencing and solely considering consequentialist approaches to punishment, if in a judgement one of the retributive factors from group one is relevant to sentencing, then Greene and Cohen’s claim is not supported in that case and there is no need to test the judgement using factors from the second and third groups. In all the judgements, courts referred to at least one of the retributive factors listed in group one. As such, it was not necessary to re-evaluate any cases in the second stage (see Figure 9-1).
37 s 3A(e) Crimes (Sentencing Procedure) Act 1999. For the relation between accountability and retribution see Odgers, p. 79. 38 s 3A(f) Crimes (Sentencing Procedure) Act 1999. For the relation between denunciation and retribution see Odgers, p. 79. 39 s 21A(3)(j) Crimes (Sentencing Procedure) Act 1999 40 s 3A(d) Crimes (Sentencing Procedure) Act 1999 41 Freckelton and Selby. Also see, R v Hemsley (2004) NSWCCA 228 42 R v Smith (1987) 44 SASR 587 at [589] per King CJ. 225
Figure 9-1: The approach of examining Greene and Cohen’s claim based on three groups of retributive and consequentialist factors.
Testing Greene and Cohen's Claim
Is there any retributive factor from group 1 relevent to sentencig in this case? Stage one:
retributive factors Yes If only considering No Testing based on group 1 The claim is not met in group 1, Green and this case Cohen's claim is met
Is there any retributive factor from group 2 relevant to sentencing in this case?
Yes The claim is not met in No this case
Is there any sentencing
Stage two: factor from group 3 considered to be
retributive factors retributive in this case? Testing based on group 2 and 3 Yes No The claim is not met in Greene and Cohen's this case claim is supported in this case
Now that it is clear which sentencing factors are considered as retributive and consequentialist for the purposes of this analysis, in the next section, I will present the findings of my empirical research.
3. Evaluating Greene and Cohen’s claim in NSW sentencing practice
While Greene and Cohen made their prediction in 2004, I included cases before this time in the analysis to determine if there is evidence that suggests judicial attitudes may have been changing in relation to retributive and consequentialist considerations over time.
In general, the analysis of cases suggested that apart from one case, which I will discuss below,
226
no case in this study supports Greene and Cohen’s claim. Courts in NSW continue to consider and apply retributive sentencing factors. Further, it does not seem that with time there have been conspicuous differences in the way courts consider retributive and consequentialist approaches to punishment.
This result is consistent with findings by McCay and Kennett (2019), which assessed Greene and Cohen’s claim according to a few cases in Australia.43 They concluded that legal reasoning continuously implicated retributivism and therefore, Australian courts have not yet shifted to sentencing based on purely consequentialist factors. Denno, similarly, in her empirical study on the use of neuroscientific evidence in criminal courts in the US, concluded that this evidence is used according to traditional legal principles.44
In the following, I will discuss the trial judge’s order outlined in Veen (No. 1) as a case where in sentencing the judge appeared to disregard a retributive approach and instead adopted a solely consequentialist approach to punishment.
Evidence of a shift from retribution towards consequentialism
Interestingly, in 1975 when neuroscience and the implications of mechanical basis of the brain were relatively less widely discussed within a legal context, it appears that in the sentencing of Veen’s first manslaughter, the trial court’s reasoning was consistent with the prescriptions of Greene and Cohen’s claim. It is noteworthy that recourse to Veen might be criticised for pre- dating Greene and Cohen’s claim by several decades; thereby failing as a prediction. Without wanting to be whiggish, Greene and Cohen argued that over time scientific developments would become convincing enough that they would change society’s intuition about free will and criminal responsibility. However, their claim does not mean that our limited knowledge of the brain has been insufficient in convincing some people that free will is an illusion.45 As such, Veen affords an opportunity to consider sentencing practice over time, and that Greene and Cohen’s claim is plausible in the NSW justice system.
43 A McCay & J Kennett, ‘Neuroscience and Punishment: From Theory to Practice’, in Neuroethics, vol. Forthcoming, 2019. Also see, A McCay & J Kennett, ‘My brain made me do it: will neuroscience change the way we punish criminals?’, in The Conversation, 2016,
Since there is no transcript of the trial judge’s decision, the following discussion is based on the information provided in Veen (No.1), specifically the appeal reviewing the trial judge’s reasoning for imposing life imprisonment.46
The trial judge referred to Veen’s previous offending—at the age of 16 he stabbed his landlady and was convicted of malicious wounding. At the age of 20, he killed a person and received a conviction of manslaughter. The court concluded that both offences arose from some form of brain damage coupled with slight provocation and intoxication. The trial court also found that there was a substantial possibility of future offending, stating that: ‘There can be little doubt that the prisoner, if and when released, will, whilst he suffers from this brain damage, be likely sooner or later to kill or seriously injure one or more other human beings.’47 It appears that the judge viewed this case as one in which some brain abnormality combined with provocation and intoxication would result in at least a strong possibility of future offending.
The trial judge proceeded to find that punishment would not deter the offender, nor people like him, who could not control their behaviour and thus deterrence had no application in this case. While deterrence is a consequentialist approach to punishment, disregarding it in this case is congruent with consequentialism. As deterrence in the case of Veen was not beneficial (forward- looking approach), it would be against consequentialism to consider it in sentencing.
With respect to Veen’s sentence, the trial judge explained that although in the case of manslaughter ‘the penalty ranges from nominal punishment to penal servitude for life’, the punishment in circumstances where ‘the mental responsibility has been substantially impaired’ is normally lower than life imprisonment.48 It appears that the court was referring to the principle of proportionality—the adjustment of an offender’s moral culpability and punishment based on their brain impairment. However, the trial judge added that ‘in this case, I do not think the ordinary principles of punishment apply. Indeed, I do not think it can be properly said, as I interpret the jury’s verdict, that the prisoner should undergo punishment. He has to be imprisoned for the protection of the community from his own uncontrollable urges.’49
There are two salient points to draw from the explanation. First, the trial judge suggested that in cases like Veen, due to some factors (i.e. a combination of his brain impairment, intoxication and
46 Veen (No. 1), pp 483-487 Jacobs J quoting the trial judge. 47 Veen (No. 1), pp 483-487 Jacobs J quoting the trial judge. 48 Veen (No. 1), pp 483-487 Jacobs J quoting the trial judge. 49 Veen (No. 1), pp 483-487 Jacobs J quoting the trial judge (emphasis added). 228
provocation), the offender’s possibility of recidivism was high and the court should not follow ordinary principles of punishment. Second, it seems that where the judge referred to ‘punishment’, this was specifically a retributive approach to punishment. This is shown by the fact that his Honour found protection of the community, as a reason for imprisonment (a consequentialist approach) to be distinct from ‘punishment’. Both points support Greene and Cohen’s claim. The court decided not to apply ‘the ordinary principles of punishment’ and this is a change in the application of legal rules by disregarding the retributive aim of punishment.
There are other sentencing factors that can be considered part of a consequentialist approach, such as rehabilitation. But the court ruled out the possibility of rehabilitating the offender: ‘There is no suggestion that his condition is curable, or in any way responsive to treatment.’ Interestingly, what the judge added later is very similar to what Eagleman suggests: if imprisonment is not beneficial (i.e. it will not deter or rehabilitate the offender) and there is no treatment for the offender’s condition, then the state should keep the dangerous individual under control. And that is what the court seems to have intended: ‘There is no institution I can send him to’.50 Consequently, the trial judge found himself confronted with two alternatives. First, to release the offender. This was not a viable option because the offender was too dangerous to be allowed to remain in society. Secondly, to imprison him.
The trial judge’s subsequent findings confirm that retributivism was disregarded, and only consequentialist sentencing factors were considered:
I strongly recommend that the prisoner be given a full psychiatric assessment, and be provided with psychiatric treatment if upon assessment such treatment appears to be warranted. I should hope that the authorities will periodically review his case to determine whether he should be released, on a proper balance of his interests and the need to protect society … I believe there is only one sentence I can pass, and there will be no non-parole period.51
Overall, the court found Veen’s brain impairment to be the main cause of his criminal behaviour, and found that it was irreversible. The court also concluded that there was little doubt that when Veen was released, he would kill or seriously injure another person (suggesting a lack of free will). No determination of a non-parole period, which suggests that proportionality was not considered, and a retributive approach was not taken. His release was on the basis of his rehabilitation and the protection of society (consequentialist approach), and as such, it appears
50 Veen (No. 1), p 491, Murphy J quoting the trial judge. 51 Veen (No. 1), pp 483-487 Jacobs J quoting the trial judge. 229
that this decision is consistent with Greene and Cohen’s claim.52
Nonetheless, there are a number of considerations in this conclusion. This case was exceptional, and it is unclear if there are any other cases like it. More importantly, although the trial court’s order was upheld by the Supreme Court of New South Wales (Court of Criminal Appeal), the majority of the High Court in Veen (No. 1) quashed the sentence of life imprisonment and criticised trial judge for overlooking the principle of proportionality in favour of protecting society.53 More specifically, on appeal, the majority in the High Court emphasised the importance of proportionality and retribution along with other consequentialist factors.54
Further, it is arguable that Veen’s trial was a case where neuroscience was used to support a mechanistic brain theory, and it was merely a case where neuroscientific evidence pointed to abnormal brain function. Nonetheless, Veen is consistent with the prescriptions of Greene and Cohen’s claim because the trial court considered the interaction of Veen’s abnormal brain with other factors (i.e. stress and intoxication) a matter that would determine his future offending. As such, Veen suggests how courts may deal with offenders and punishment where they find that behaviour is a result of the mechanistic function of the brain, rather than of our free will.
4. Preventive detention: pushing the scale towards consequentialism
In the previous section I explained that apart from the case of Veen’s trial, there is no evidence in the dataset to support Greene and Cohen’s claim. In this section, I discuss the Crimes (High Risk Offenders) Act 2006 (NSW). Although this Act is not associated with the developments of neuroscience, it may facilitate the prediction made by Greene and Cohen to emerge in the future.
As discussed in Chapter 5 (the double-edged sword effect of neuroscience), the Crimes (High Risk Offenders) Act 2006 (NSW) has two main purposes that are both based on a consequentialist approach to punishment: to ‘ensure the safety and protection of the community’ and to encourage rehabilitation.55 As a consequence, an offender may receive a punishment that is disproportionate
52 As discussed earlier in this chapter, it is unclear if Greene and Cohen consider ‘rehabilitation’ to be a consequentialist approach. However, this does not change this conclusion. The court still disregarded retributivism and considered protection of the community to be the only consequentialist factor. 53 Veen (No. 1), p 467 per Stephen J. The majority ordered a sentence of twelve years imprisonment without determining a non-parole period. 54 Per Jacobs J, Stephen J and Murphy J. In Veen (No. 2) on the other hand, the majority found life imprisonment a proportional punishment, considering a variety of sentencing factors that included both retributivist and consequentialist approaches to punishment (for more information about this case see Chapter 5, double-edged sword effect of neuroscience). 55 S 3 Crimes (High Risk Offenders) Act 2006 (NSW) 230
to their crime and level of culpability, in order to prevent recidivism, i.e. overruling proportionality (a retributive approach) for the benefit of the society (a consequentialist approach). While this Act does not imply that courts should move away from traditional assumptions about free will and disregard retributivism by embracing consequentialism as the only approach to punishment, it does suggest that the criminal justice system has a greater potential to move toward the direction predicted by Greene and Cohen.
This Act was originally limited to High Risk Offenders (i.e. sex offenders and violent offenders). It is noteworthy that amendments in 2013 changed the title of the Act from the Crimes (Serious Sex Offenders) Act 2006 to the Crimes (High Risk Offenders) Act 2006 (NSW) and extended its application to include both sexual offenders and violent offenders. The introduction of this Act suggests that the protection of the community would reduce the consideration of proportionality where heinous crimes have been committed. In the future, community interest may require more crimes to be incorporated into this Act, or equivalent legislation.
The fact that orders under this Act are applied at the end of a prisoner’s sentence indicates that the criminal justice system acknowledges the difficulty of predicting future offending. Neuroscience, however, is providing more tools and more reliable approaches to understand the brain’s mechanisms, and the risk that a person may re-offend after they are released.56 Consequently, the more reliable approaches to predict the risk of recidivism could encourage courts to apply the Crimes (High Risk Offenders) Act 2006 (NSW) more frequently and for more types of crimes, and consequently the criminal justice system at large may move from a more retributivist approach toward consequentialism in sentencing.
Conclusion
Greene and Cohen predicted that neuroscientific studies will increasingly support the theory that mechanical processes in the brain cause behaviour, and subsequently change society’s moral intuition about free will and criminal responsibility. Accordingly, this would result in a shift away from the retributive aim of punishment, and towards a consequentialist approach. This Chapter was designed to examine whether this claim is consistent with the use of neuroscience evidence in sentencing practice in NSW.
56 PA Ormachea et al., ‘The role of tablet-based psychological tasks in risk assessment’, in Criminal Justice and Behavior, vol. 44, 2017, 993–1008; PA Ormachea et al., ‘Enabling Individualized Criminal Sentencing While Reducing Subjectivity: A Tablet-Based Assessment of Recidivism Risk’, in AMA journal of ethics, vol. 18, 2016, 243–251; Center for Science & Law, ‘Assessing offender risk’, in Center for Science & Law,
This study found that in all cases in the dataset except one, courts continue to consider retributive sentencing factors, and thus Greene and Cohen’s claim is not supported in practice. However, since the Greene and Cohen claim is predictive, this study and the temporally restricted dataset cannot falsify it outright. The future is uncertain, and there have been some trends identified in this thesis. I explained that in Veen, about 30 years before Greene and Cohen published their claim, the trial case indicated that the claim is not completely implausible. I also discussed the Crimes (High Risk Offenders) Act 2006 (NSW) and the potential for Greene and Cohen’s claim to manifest in practice.
A consideration to the examination of Greene and Cohen’s claim is that this study is limited to the sentencing phase, and hence was not able to evaluate trial (guild phase) decisions. During the trial, the court due to an understanding of the mechanistic brain function may find the offender not guilty. It is unclear what would happen in such conditions and whether the court would sentence the individual by solely considering consequentialist approaches to punishment or set them free.
232
Chapter 10 Conclusion
233
There is a significant disconnect between how neuroscience evidence is perceived and presented in actual courtrooms and what scholars seem to think are critical issues … While debates about the potential role of brain scanning and its purported value continue unabated, the practical reality is that brain scan technology is being used in the criminal courtroom, perhaps much more frequently than many observers appreciate
— Lyn M. Gaudet & Gary E. Marchant57
Introduction As has been demonstrated, researchers have shown an increased interest in the intersection between neuroscience and law, which has resulted in many scholarly discussions and claims on how neuroscience is being used, ought to be used, and will be used in courts. However, these claims are often based on small-scale studies, anecdotal recounts of the criminal justice system and speculation. These claims are rarely substantiated by comprehensive and systematic empirical studies on the use of neuroscience in courts, despite the fact that any analysis ‘must begin with a more accurate understanding of how [neuroscientific] evidence is currently being used’.58
Since 2015, a small number of empirical studies from different countries have begun to examine the use of neuroscience in courts both qualitatively and quantitatively. These studies suggest that neuroscience has a growing impact on criminal proceedings in ways that discredit some of the neurolaw claims.59 In Australia, while neuroscientific evidence was first used in criminal proceedings more than four decades ago, there is still little empirical data on the use of neuroscience in courts that can be used to assess neurolaw claims. To fill this gap, the present thesis aimed to examine the use of neuroscientific evidence in criminal courts in New South Wales, Australia, and to uncover whether the use of neuroscience in courts accorded with the seven neurolaw claims considered in this study. These claims are central to scholarly discussions and concerns in this area, and the results of this study have important implications, with the
57 LM Gaudet & GE Marchant, ‘Under the radar: Neuroimaging evidence in the criminal courtroom’, in Drake Law Review, vol. 64, 2016, 577–661 (emphasis is original). 58 NA Farahany, ‘Neuroscience and behavioral genetics in US criminal law: an empirical analysis’, in Journal of Law and the Biosciences, vol. 2, 2016, 485–509 (p. 486). 59 See DW Denno, ‘The Myth of the Double-Edged Sword: An Empirical Study of Neuroscience Evidence in Criminal Cases’, in Boston College Law Review, vol. 56, 2015, 493–551; CH de Kogel & EJMC Westgeest, ‘Neuroscientific and behavioral genetic information in criminal cases in the Netherlands’, in Journal of Law and the Biosciences, vol. 2, 2015, 580–605; Farahany, 485–509; JA Chandler, ‘The use of neuroscientific evidence in Canadian criminal proceedings’, in Journal of Law and the Biosciences, vol. 2, 2016, 550–579; P Catley & L Claydon, ‘The use of neuroscientific evidence in the courtroom by those accused of criminal offenses in England and Wales’, in Journal of Law and the Biosciences, vol. 2, 2015, 510–549; Gaudet and Marchant, 577–661. 234
potential to inform recommendations for practice, policy and future studies. To examine these claims, this thesis systematically collected and analysed criminal cases where neuroscientific evidence was introduced. Consequently, this thesis is distinguished from some other neurolaw discussions in that it is ‘grounded in fact rather than hyperbole’.60
The first four Chapters of this thesis provide a background to allow empirical examination of the research questions. Chapter 1 outlines the research questions, aims, which determined the direction of the thesis and materials required to conduct the empirical study and discussion of judgements. Chapters 2 and 3 provide the relevant theoretical background information. Chapter 2 reviews the background to neuroscience and explores the relevance of neuroscience to criminal law and criminology. It also reviews the existing scholarly literature central to this study and presents some empirical neurolaw publications from other countries. Further, it discusses the existing neurolaw literature in Australia and the dearth of empirical studies on the use of neuroscience in courts. Chapter 3 introduces some of the key aspects of the NSW legal framework, and presents the conceptual legal framework required to empirically examine neurolaw claims considered in this study. Chapter 4, based on the scope of the study determined in the first three Chapters, introduces criminal cases as the source of data through which the research questions would be examined. Chapter 4 also outlines the approach used to analyse the judgements, and discusses some of the shortcomings of the research methods used. Chapters 5 to 9 evaluate the neurolaw claims based on the results of the empirical research. Findings from the empirical work presented in these five Chapters are brought together in this concluding Chapter. A summary of the conclusions drawn in each of the Chapters examining neurolaw claims, is that there are some inconsistencies between the proponents of neurolaw claims and how neuroscience is being used in practice. The following presents the key findings and answers the research questions of this study.
1. Neurolaw claims: theory vs practice
Chapter 5: double-edged sword effect of neuroscience Neuroscientific evidence can potentially show that an offender’s brain impairment had contributed to the crime committed and therefore, that they are less culpable for their act. Concurrently, the same neuroscientific evidence can suggest that the offender presents an ongoing danger to society. This dual effect of neuroscience is referred to as the double-edged sword effect,
60 Denno, 493–551 (p. 494). 235
i.e. that neuroscientific evidence can either mitigate or aggravate the sentence. As has been shown, some scholars have discussed this potential effect of neuroscience and have suggested that lawyers may find it challenging to decide whether the evidence should be introduced at the risk of an increased sentence, or whether it should be withheld, relinquishing the opportunity to mitigate the sentence.61
The results of this study show that neuroscientific evidence contributed to sentencing in about half of the cases in the dataset. In almost all of those cases, this evidence tends towards a more lenient sentence. The number of cases where this evidence pointed towards a harsher punishment was much lower. These findings suggest that the ‘popular image of neuroscience evidence as a double-edged sword’,62 does not accord with the way neuroscience contributes to sentencing in courts. One practical implication of these findings with respect to the double-edged sword claim, is that defence lawyers may adduce neuroscientific evidence with a degree of confidence that in practice such evidence is likely to be used for mitigation and not likely to backfire on the offender.
One notable consideration in relation to this conclusion, however, is that the data from this study may over-represent the number of cases in which neuroscience mitigates sentencing. This is due to the possibility that defence lawyers may withhold neuroscientific evidence where they predict that this would tend towards a more severe punishment and the dataset is therefore biased towards cases in which neuroscience has contributed to a more lenient sentence. Nonetheless, since the aggravating sentencing factors relevant to neuroscientific evidence are less influential (have lower weight) in sentencing than the mitigating factors, even if there are more cases where neuroscience points towards aggravating the sentence, it is still worthwhile for the defence to adduce neuroscientific evidence. This conclusion may be more probative if considered in the light of future studies using other methods, such as interviews, which examine whether lawyers withhold neuroscientific evidence where they believe it would be likely that this evidence would point towards a more severe sentence.
Chapter 6: the gap between neuroscience and the law’s language
Steven J. Morse claims that there is a mismatch between the language of the law, and that of neuroscience, and he is concerned that the language used in courtrooms to describe brain
61 See, O Jones & F Shen, ‘Law and neuroscience in the United States’ in International Neurolaw: A Comparative Analysis, TM Spranger (ed), Springer, 2012, pp. 349–380; Denno, 493–551; AS Barth, ‘A double-edged sword: The role of neuroimaging in federal capital sentencing’, in American journal of law & medicine, vol. 33, 2007, 501–522. 62 See, Denno, 493–551 (p. 493). 236
abnormalities may be inappropriate.63 There are two main elements in Morse’s argument: first, neuroscience uses mechanistic, cause and effect language to describe how neurons communicate and how different areas of the brain are responsible for different functions (e.g. cognitive abilities). Law, however, does not assume that criminal behaviour is the product of a mechanistic cause and effect process that occurs in the brain. Law recognises the mind as the cause of a criminal act. In other words, law’s view of criminality is that ‘brains do not commit crime; people commit crime’.64 Morse’s second point is that to make this mismatch compatible, neuroscientific evidence should be translated into concepts that are compatible with the law’s folk psychology, i.e. the pre-scientific language that uses mental state variables to explain human behaviour (e.g. poor judgement). Therefore, instead of arguing that the damaged prefrontal cortex was the cause of a crime, it should be argued that the brain impairment influenced the offender’s judgement (i.e. law’s folk psychological language) at the time of crime. This translation distinguishes the mechanical cause and effect of brain processes from the mind and mental processes, which according to law, are causes of criminal behaviour. Morse claims that if this translation does not take place, some courts may find any brain abnormality relevant to criminal responsibility and sentencing. He explains that this is an error because an abnormality that does not affect an individual’s capacity to control their behaviour or understand right from wrong is not legally relevant.65
Unfortunately, there are some challenges to assessing this claim in practice. Morse does not clearly determine the scope of law’s folk psychology and consequently, it is unclear which concepts satisfy translation under this category. It seems that this claim is to some extent vague and simplistic to be examined in practice.
Bearing in mind these methodological difficulties, the findings of this study suggest that neuroscientific evidence, in the majority of cases, is translated into the language of law’s folk psychology before the court considers this evidence in sentencing. This is also evident where in most cases, the court provides their remarks on sentencing and refers to the underlying neuroscientific problems as mental issues or reviews them in folk psychological language. In the majority of cases where neuroscience is not explained in the law’s folk psychology, neuroscientific evidence was not considered in determining
63 S Morse, ‘Neuroscience in Forensic Contexts: Ethical Concerns’ in Ethics Challenges in Forensic Psychiatry and Psychology Practice, EEH Griffith (ed), New York, Columbia University Press, 2018, pp. 132–158; S Morse, ‘Lost in Translation?: An Essay on Law and Neuroscience’ in Law and Neuroscience, Current Legal Issues, M Freeman (ed), Oxford University Press, 2011, pp. 529–562. 64 SJ Morse, ‘Brain overclaim syndrome and criminal responsibility: A diagnostic note’, in Ohio St. J. Crim. L., vol. 3, 2005, 397–412 (p. 397). 65 Morse, pp. 132–158; Morse, pp. 529–562. 237
the appropriate sentence. This occurred in cases where, for instance, test results were inconclusive or did not suggest any brain abnormality.
This translation, or in other words, use of law’s folk psychological language in explaining neuroscientific evidence, is also evident in the cases examined in Chapter 5 (the double- edged sword effect of neuroscience). The majority in the High Court case of Veen (No. 2) (1988) discussed the double-edged sword effect of Veen’s brain impairment, and when referring to the effect of such evidence, they used the term mental abnormality—an example of the folk psychological language:
[The purposes of sentencing] are guideposts to the appropriate sentence but sometimes they point in different directions. And so a mental abnormality which makes an offender a danger to society when he is at large but which diminishes his moral culpability for a particular crime is a factor which has two countervailing effects: one which tends towards a longer custodial sentence, the other towards a shorter.66
Overall, it appears that NSW courts’ use of neuroscience is consistent with prescriptions of Morse’s claim since many years before Morse discusses the claim, supporting the conclusion that translation is not an issue in courts.
Chapter 7: concerns about the reliability of neuroscientific evidence in courts
Under this section, I will discuss findings around three claims regarding the reliability of neuroscientific evidence in sentencing.
The convergence of evidence (Chapter 7)
The reliability of brain imaging evidence has been criticised by many scholars. They contend that brain images cannot be relied upon as sole proof of an individual’s mental state.67 On the other hand, some argue that while brain imaging evidence may not be reliable enough to be considered dispositive proof of an individual’s mental state, it can
66 Veen v R (No. 2) at [13] per Mason C.J., Brennan, Dawson and Toohey JJ (emphasis added). 67 See, JC Moriarty, ‘Flickering admissibility: neuroimaging evidence in the US courts’, in Behavioral sciences & the law, vol. 26, 2008, 29–49; Morse, pp. 132–158; ST Grafton et al., ‘Brain scans go legal’, in Scientific American Mind, vol. 17, 2006, 30–37. 238
nonetheless be useful where it directs and supports (i.e. is converged with) other forms of evidence, such as psychiatric, neuropsychological and behavioural evidence.68
The findings in this study suggest that where brain imaging evidence appeared to contribute to sentencing, this evidence was always corroborated by other forms of evidence. This result, however, does not necessarily suggest that courts had intentionally made decisions to consider neuroscientific evidence where it is corroborated by other forms of evidence only to be consistent with the convergence of evidence claim — i.e. due to limitations of neuroscience, it should be relied upon where it is corroborated with other forms of evidence. Courts may in general find an argument about the mental state of an offender more persuasive where different pieces of evidence are corroborated. As such, where neuroscientific evidence was the sole evidence supporting an argument, judges did not rely on it not because its value is questioned. Rather, it is possible that judges preferred an opposing argument to neuroscientific evidence that was supported by the corroboration of other forms of evidence (hence more persuasive). That means, without opposing arguments supported by corroborative evidence, courts may have relied on brain imaging evidence as the sole proof of the mental state.
This conclusion is contrary to what some scholars, such as Jane Campbell Moriarty claim, ‘while courts have routinely admitted some neuroimages, such as CT scans and MRI, as proof of trauma and disease, they have been more circumspect about admitting the PET and SPECT scans and fMRI evidence.’69 Although functional imaging evidence was less prevalent than structural imaging evidence in the cases examined in this study, my assessment is that courts did not discuss the limitations of such evidence, and treated them in the same way as structural images. Hank Greely’s claim, on the other hand, appear to be more consistent with the findings of this study: ‘[a]nybody can try to admit
68 OS Choi, ‘What Neuroscience Can and Cannot Answer’, in Journal of the American Academy of Psychiatry and the Law Online, vol. 45, 2017, 278–285; C Scarpazza et al., ‘The charm of structural neuroimaging in insanity evaluations: guidelines to avoid misinterpretation of the findings’, in Translational Psychiatry, vol. 8, 2018, 227–237; JA Silva, ‘Forensic psychiatry, neuroscience, and the law’, in Journal of the American Academy of Psychiatry and the Law, vol. 37, 2009, 489–502; G Sartori, S Pellegrini & A Mechelli, ‘Forensic neurosciences: from basic research to applications and pitfalls’, in Current Opinion in Neurology, vol. 24, 2011, 371–377; D Amen & L Flaherty, ‘Is brain imaging clinically useful for psychiatrists?’, in Clinical Psychiatry News, vol. 34, 2006, 11. 69 Moriarty, 29–49 (p. 29). 239
neuroscience evidence in any case, in any court’, because judges are not fully aware of its limitations, and hence to make appropriate decision about it.70
It would be alarming if NSW courts were indeed unaware of the limitations of different forms of brain imaging evidence (I will return to this point in the following section). This thesis recommends that further research needs to be undertaken using methods such as interviews or surveys to investigate judges’ knowledge of the limitations of brain imaging techniques
Another aspect of the convergence of evidence claim relates to the usefulness of neuroscience. Although I did not examine any claims specifically concerning the usefulness of neuroscience, the following discussion is directly relevant to the convergence of neuroscientific evidence, and is hence deserving of consideration. In Chapter 2 (literature review and background), I explained that Morse believes there are two circumstances under which brain imaging evidence may be used in corroboration with other forms of evidence. The first circumstance is where brain images are merely ‘cumulative and unnecessary’.71 This refers to situations where evidence that is sufficient in proving an individual’s mental state already exists, and neuroscientific evidence is merely additive and its addition or subtraction would less likely alter the conclusion reached.72
The second circumstance occurs where other forms of evidence are in conflict or not convincing enough to meet the standard of proof. Morse argues that brain imaging evidence in such circumstances is useful, however, he claims that these cases are very rare.73 Morse continues, concluding that brain imaging is expensive and time consuming, and when considered against its limited utility, its use would be a waste of the court’s time and resources.74
It is unclear what basis Morse uses to come to his conclusion that the second circumstance — where neuroscience is useful — rarely occurs in courts. Morse has not conducted a systematic or comprehensive empirical study examining the usefulness of brain imaging evidence, nor does he refer to such studies to support his contentious.
The results and subsequent discussions of this study do not necessarily conclude that Morse’s
70 Quoted in I Chen, ‘The court will now call its expert witness: the brain’, in Stanford Report, 2009,
claims about the usefulness of neuroscience are wrong or inappropriate. Rather, it argues that the complexity of sentencing and the difficulty of estimating the extent to which neuroscientific evidence influences the sentence does not allow to fully ascertain whether neuroscience is cumulative or rarely useful.
If courts find Morse’s claim about the usefulness of brain imaging evidence persuasive, there could be costly consequences for both parties. For example, courts may not approve adjournments of the criminal proceedings to give defendants time to conduct brain imaging exams, despite the fact that these may actually be the deciding factor where other evidence is conflicting.
There are a variety of reasons for a court’s inability to predict whether brain images will be useful. One of these is that the weight given to different forms of evidence may change the following cross-examination. Subsequently, an offender’s brain image may become useful and necessary evidence in meeting the standard of proof. As discussed in Chapter 3, a feature of the adversarial system is that parties should provide the strongest argument possible to the court and may not wish to risk losing a more convincing argument only because evidence that was potentially cumulative and unnecessary was not adduced.75
Chapter 7: the persuasive power of neuroscientific evidence Some scholars argue that courts may place undue weight on neuroscientific evidence, and find it more convincing and scientifically credible, than can actually be justified.76 The challenge of examining this claim is determining the appropriate amount of probative value that should be given to different types of neuroscientific evidence both in general and in each case. This thesis is thus limited to examination of this claim through specific approaches that do not require determination of the probative value of neuroscientific evidence. First, since behavioural evidence is considered to be more credible than brain imaging evidence by scholars,77 I sampled cases where these two types of evidence were in conflict. Using these cases, I determined whether the court found imaging evidence to
75 For a similar opinion see, OD Jones, ‘Seven Ways Neuroscience Aids Law’ in Neurosciences and the Human Person: New Perspectives on Human Activities, AM Battro et al. (eds), Vatican City, The Pontifical Academy of Sciences, 2013, pp. 181–194. 76 Choi, 278–285; DP McCabe & AD Castel, ‘Seeing is believing: The effect of brain images on judgments of scientific reasoning’, in Cognition, vol. 107, 2008, 343–352; DS Weisberg et al., ‘The seductive allure of neuroscience explanations’, in Journal of cognitive neuroscience, vol. 20, 2008, 470– 477. As discussed in the literature review, there are other studies that claim otherwise. See, NJ Schweitzer et al., ‘Neuroimages as evidence in a mens rea defense: No impact’, in Psychology, Public Policy, and Law, vol. 17, 2011, 357–393; Morse, pp. 132–158. 77 Morse, pp. 132–158; Choi, 278–285. 241
be more credible than behavioural evidence. In using this approach, I was not required to determine an external benchmark on the probative value of any evidence, rather I was able to determine the court’s views on the relative value of evidence based on comparisons within a case. The results suggested that courts found behavioural evidence to have more probative value than brain imaging evidence. Second, I examined the claim that courts find brain images to be objective evidence despite the fact that the outcome of this evidence relies on the subjective interpretation of the expert witness that analyses it.78 The research results showed that except in one case, courts do not find neuroscientific evidence to be objective. Further, decisions regarding Veen’s sentencing (in both the first and second manslaughter cases) and medical evidence provided in a later decision where Veen applied for a redetermination of his sentence suggest courts gave undue weight to neuroscientific evidence.
However, considering the limited assessment of the claim by the specific approaches taken, an overall analysis of cases shows that except in a few cases it does not appear that courts give undue weight to neuroscientific evidence for their potential persuasiveness power (I will return to this conclusion in the following discussion).
Chapter 7: Expert witnesses’ area of expertise
Some scholars argue that psychiatrists, psychologists and social workers do not have the requisite expertise enabling them to interpret brain imaging evidence and that therefore, their reports and testimonies are not credible or sufficiently reliable.79 Despite concerns from the proponents of this claim, the results of this study indicated that psychiatrists frequently, and psychologists in a few occasions, appeared as expert witnesses to discuss brain imaging evidence.
Overall, from analysis of the judgements, it does not appear that courts are concerned with the precise expertise of expert witnesses. Further, there are also no guidelines to aid courts in assessing the qualifications and experiences of particular witnesses with respect to particular types of neuroscientific evidence. As such, this study recommends that more academic work is required to generate such guidelines. Two rationales are apparent for the timely provision of judicial
78 J Dumit, ‘Objective brains, prejudicial images’, in Science in Context, vol. 12, 1999, 173–201. 79 Amen and Flaherty, 11; Silva, 489–502; JD Bremner, Brain imaging handbook, WW Norton & Co, 2005. Also see D Linden, Neuroimaging and Neurophysiology in Psychiatry, Oxford, United Kingdom, Oxford University Press, 2016. 242
guidance on appropriate expert witness qualifications for neuroscientific evidence specific to the Australian criminal justice system. First, every criminal justice system has a different structure and therefore judges may differ in the types of assistance required. Second, Australian universities may offer differing subjects under what is ostensibly the same qualification. For example, a bachelor’s degree in forensic psychiatry may involve different skill sets in Australia compared to the US.
Before I discuss the next neurolaw claim, I will mention one conclusion that can be drawn from the discussions on neurolaw claims analysed thus far (Chapters 5-7), with respect to neurolaw claims examined in Chapter 7 (concerns about the reliability of neuroscientific evidence in courts).
Conclusions related to claims discussed in Chapter 7 There is a significant conclusion from the claims I have examined so far: it appears that courts find neuroscientific evidence value and influence similar to other types of expert evidence. There are several reasons for this conclusion, which has two important potential implications with respect to the reliability of neuroscientific evidence. The reasons for this conclusion are as follows. First, in the majority of cases analysed in this study, neuroscientific evidence had been translated into law’s folk psychological language, which involves concepts that are familiar to courts and are also used to describe other forms of evidence (e.g. report of psychological testing). Secondly, in the majority of cases, neuroscientific evidence had been presented to courts through expert witnesses that courts are familiar with, i.e. psychiatrists and psychologists. Thirdly, in Veen (No. 1), one of the earliest cases in Australia discussing neuroscientific evidence, the High Court does not criticise neuroscience evidence as having insufficient or limited reliability, but instead welcomed it, ‘[i]t is hardly necessary to remark that there had been no neurological examination and no electro-encephalogram [EEG] to support the conclusion [of brain damage]’.80 Similarly, in Veen (No. 2), the High Court in reviewing Veen (No. 1) did not make any objection to that evaluation nor make statements about the limitation of neuroscience in general. These are two significant cases in Australian criminal law because in Veen (No. 2), after discussing Veen (No. 1), the obiter outlined the aims of sentencing at common law, which has subsequently been codified by statute (s 3A Crimes (Sentencing Procedure) Act 1999).81 It is likely that the majority of NSW judges are familiar with these judgements and consider the High Court’s apparent
80 Veen (No. 1), p 489 81 See, SJ Odgers, Sentence : the law of sentencing in NSW courts for State and Federal offences, 3rd ed., Haberfield, N.S.W, Longueville Media, 2015. Also see, Muldrock v R (2011) 244 CLR 120 at [20]. 243
treatment of neuroscientific evidence in a similar way to other types of evidence for not criticising its value.82
There are two potential implications of treating neuroscientific evidence in a similar way to other forms of evidence: first, neuroscience may not be overly persuasive.83 There is no hype in the courtroom about neuroscientific evidence, it is rather an unremarkable form of evidence to the courts. As Allan McCay84 puts it ‘[p]eople don’t bat an eyelid at an expert witness report that refers to neuropsychological testing or brain scans … That’s not a strange intrusion on the legal system’.85 This conclusion is further supported by the findings in Chapter 5 (double-edged sword effect of neuroscience). Neuroscientific evidence does not influence the sentence in around half of the cases examined. The second potential implication of neuroscientific evidence being considered in a similar way as other evidence is that courts may overlook the limitations inherent to neuroscientific evidence. This is distinct from finding neuroscientific evidence more credible due to misplaced views of its persuasive power.
The conclusion that courts may overlook the limitations of neuroscience can be further illustrated through the Veen judgements and the concept of the ‘hype-cycle’. A ‘hype-cycle’ is a period of time in which technology has first emerged, and early expectations of what it can do far exceed its realistic potential. As the technology fails to deliver on initial expectations, it may be criticised and treated with appropriate caution. At this point, the technology’s actual capacity can be well- understood and defined within a more realistic frame of reference.86 At the time of Veen (No. 1) (1979) and Veen (No. 2) (1985), imaging techniques such as PET, SPECT and MRI were emerging,87 and there was hype surrounding the capacity of what neuroscience could tell us about the brain, mind and behaviour.88 The two Veen cases can be read as providing High Court
82 Veen (No.1) has been referred in Australian cases 245 times, and Veen (No. 2) despite being decided 9 years later has been referred in approximately 1800 cases. Retrieved from http://www.austlii.edu.au/cgi- bin/LawCite?cit=[1979]%20HCA%207&name=Veen%20v%20R on November 13, 2016. Retrieved from http://www.austlii.edu.au/cgi- bin/LawCite?cit=[1988]%20HCA%2014&name=Veen%20v%20R%20(No%202) on November 13, 2016. 83 See, for example, TR Brown & ER Murphy, ‘Through a scanner darkly: functional neuroimaging as evidence of a criminal defendant’s past mental states’, in Stanford law review, vol. 62, 2010; McCabe and Castel, 343–352; Weisberg et al., 470–477. 84 A former solicitor and legal researcher affiliated with Macquarie University’s Centre for Agency, Values and Ethics (CAVE) and a lecturer at the University of Sydney Law School 85 D Rolfe, ‘It wasn’t me, it was my brain’, in LSJ: Law Society of NSW Journal, vol. May, 2016, 26–32 (p. 28). 86 DA Baker et al., ‘Making sense of research on the neuroimage bias’, in Public Understanding of Science, vol. 26, 2017, 251–258 (p. 254). Also see, DE O’Leary, ‘Gartner’s hype cycle and information system research issues’, in International Journal of Accounting Information Systems, vol. 9, 2008, 240– 252. 87 Between 1975 and the early 1980s. 88 N Rose & JM Abi-Rached, Neuro: The new brain sciences and the management of the mind, in Princeton University Press, EBSCO Publ, Princeton University Press, 2013. 244
authority on the approach that should be taken when evaluating the probative value of neuroscientific evidence – to be an acceptable and probative form of evidence, and thus no discussion with respect to its limitations was raised.
Despite the fact that during the last decade or so scholars have been increasingly and openly acknowledging the limitations of neuroscience, courts are following the methods of dealing with neuroscientific evidence outlined in Veen. It can be argued that courts have not complied with the last phase of the hype-cycle when the technology’s actual capacity is well-understood and defined within a realistic frame of reference. This is because in a common law system, courts are bound by the principle of stare decisis – that subsequent decisions should be based on precedent.
In summary, two implications of courts finding neuroscientific evidence are, courts do not find neuroscience to be overly persuasive. Concurrently, courts do not examine the value and reliability of this evidence. That said, since this study suggested that where brain imaging evidence is considered in the sentence it is corroborated by other forms of evidence, the fact that courts are not aware of criticisms about the reliability of neuroscience, at least on the face of it, does not seem to be a significant issue. Further research should be undertaken to investigate the courts’ understanding of neuroscientific evidence and its influence on sentencing.
Chapter 8: Neuroscience and the rehabilitation of the offender Rehabilitation, an aim of sentencing in NSW, has several considerable benefits for a criminal justice system, including decreasing the rate of recidivism, which may lead to a safer society. However, the criminal justice system has been criticised for its ineffective function in rehabilitating offenders.89 A number of scholars have suggested that neuroscience should be pursued for the purposes of promoting rehabilitation as they present different approaches, despite the fact that some of these are controversial.90 For instance, David Hodgson suggests that whilst the courts should not reject retributivism,
89 See, for example, MW Lipsey & FT Cullen, ‘The effectiveness of correctional rehabilitation: A review of systematic reviews’, in Annual Review of Law and Social Science, vol. 3, 2007, 297–320; DP Mears et al., ‘Prison visitation and recidivism’, in Justice Quarterly, vol. 29, 2012, 888–918; DA Andrews & C Dowden, ‘Managing correctional treatment for reduced recidivism: A meta-analytic review of programme integrity’, in Legal and Criminological Psychology, vol. 10, 2005, 173–187. 90 DM Eagleman, ‘Neuroscience and the law’, in Houston Lawyer, vol. 16, 2008, 36–40; HT Greely, ‘Neuroscience and criminal justice: not responsibility but treatment’, in University of Kansas Law Review, vol. 56, 2008, 1103–1138; HT Greely, ‘Direct brain interventions to “treat” disfavored human behaviors: ethical and social issues’, in Clinical Pharmacology & Therapeutics, vol. 91, 2012, 163–165; A Raine, The anatomy of violence: The biological roots of crime, in Vintage, Vintage, 2013; P Koi, S Uusitalo & J Tuominen, ‘Self-Control in Responsibility Enhancement and Criminal Rehabilitation’, in 245
To the extent that science assists in identifying brain conditions that involve particular risks of criminal behaviour, and in devising methods to minimise these risks, this could properly be taken into account … and in guiding the treatment of that person during the period of that sentence.91
Contrary to these suggestions, the findings of this study have demonstrated that the application of different approaches to treating offenders with brain impairments are rarely discussed in judgements and that instead, it appears that courts focus on discussing how to address other risk factors such as addiction and alcohol abuse.
It is unclear why courts do not address the brain impairment of offenders. Some reasons for this may be that courts find brain impairments to be irreversible, and believe that, nothing would work in rehabilitating such an offender;92 or they may not know how to deal with this condition and thus what order would be appropriate. Brain Injury Australia criticises this nihilistic view on the rehabilitation of criminals with brain injuries by arguing that, ‘the greatest obstacle to using rehabilitation treatment effectively to reduce criminal behaviour is not a nothing-works research literature with nothing to offer but, rather, a correctional system that does not use the research available and has no history of doing so’.93 Considering the high rate of recidivism in NSW,94 the reasons behind courts’ lack of consideration of treatment options for impaired brains becomes a concerning issue and another area for future research.
A further matter with respect to the rehabilitation of offenders is the usefulness of brain imaging evidence. While Morse argues that neuroscientific evidence is cumulative or rarely useful in assessing criminal responsibility of offenders, he appears to find it a more valuable tool when determining appropriate rehabilitation for offenders.95 Neuroscientific evidence can assist with determining the origin of a risk factor and thus a target for treatment. For instance, psychological assessments suggest that an offender may have issues with impulse control. Brain scans can show that the offender may also have an impaired prefrontal cortex, which has consequently exacerbated their impulsivity. In this case, neuroscientific evidence, with respect to the treatment
Criminal Law and Philosophy, vol. 12, 2018, 227–244; D Eagleman, Incognito: The Secret Lives of the Brain, New York City, U.S., Pantheon, 2011. 91 D Hodgson, Rationality+ consciousness= free will, Oxford University Press, 2012, p. 225. 92 In chapter 8, I discussed the concept of ‘therapeutic nihilism', which refers to an attitude that no amount of rehabilitation would work for prisoners with acquired brain injuries. See, N Rushworth, Out of sight, out of mind: People with an acquired brain injury and the criminal justice system, in Brain Injury Australia, 2011, p. 27. 93 Lipsey and Cullen, 297–320 (p. 315), quoted in Rushworth, p. 28. 94 See, Bureau of Crime Statistics and Research (BOCSAR), ‘Re-offending statistics for NSW’, in BOCSAR, 2017,
of the offender, is not merely cumulative, because it indicates the origin of the problem and how the treatment should take place.96
Chapter 9: neurolaw of the mechanical brain Greene and Cohen have predicted that growing scientific knowledge will prove that an individual’s behaviour is the result of mechanistic processes in the brain and thus reject the idea of free will.97 They argue that as conceptions of criminal responsibility and retribution are based on free will, courts would, as a result, only focus on consequentialist approaches to punishment— punishing individuals for the purpose of beneficial consequences (i.e. deterrence and protection of society) rather than retribution (i.e. offender’s moral culpability). The results of this study, however, suggest that neuroscience has not revolutionised criminal justice system, in the way that Greene and Cohen have predicted, and with one exception (Veen), neuroscience has been used within the established principles of criminal law.
Veen’s first manslaughter was the only judgement where the trial judge, based on some factors specifically his brain impairment, found that the possibility of Veen re-offending was significant, and that the court should not follow ordinary principles of punishment such as proportionality. Instead, the court sentenced Veen to life in prison without determining a non-parole period, as the appropriate measure on the basis of the protecting wider society. On appeal, however, the majority of the High Court in Veen (No. 1) emphasised the importance of proportionality and retribution along with other consequentialist factors, and overturned the trial judge’s reasoning on this point.98
As I have previously discussed, Veen (No. 1) is an important case in the NSW criminal justice system and it is likely that many judges are familiar with the High Court’s decision, overruling the trial court’s order. For Greene and Cohen’s claim, this means that even if NSW courts were going to change the application of principles of criminal law, the High Court’s decision may have delayed this transition.
The results of this study are similar to Allan McCay and Jeanette Kennett’s findings in a study that assessed Greene and Cohen’s claim according to a few Australian cases.99 However, a
96 For more on suitable programs for individuals with brain injuries see, Rushworth, pp. 25–28. 97 J Greene & J Cohen, ‘For the law, neuroscience changes nothing and everything’, in Philosophical Transactions of the Royal Society B: Biological Sciences, vol. 359, 2004, 1775–1785. 98 Per Jacobs, Stephen and Murphy JJ. 99 A McCay & J Kennett, ‘Neuroscience and Punishment: From Theory to Practice’, in Neuroethics, 2019. 247
difference between that study and the present one is that this thesis has identified the exceptional judgement of Veen’s first manslaughter (the trial court), which is consistent with Greene and Cohen’s prescriptions. Further, by conducting a systematic and large-scale empirical analysis of sentencing judgements, this thesis has provided a more robust view of the NSW criminal justice system’s approach to sentencing offenders with respect to Greene and Cohen’s predictive claim.
Overall, Greene and Cohen’s claim is predictive, and this study can only suggest that neuroscience has not yet revolutionised the criminal justice system. Other studies should re-examine this claim in the future.
As I have discussed in Chapter 4 (an introduction to empirical research) and repeated in Chapters 5 to 9, the methodology of this thesis is subject to many limitations and hence these conclusions should be treated cautiously. I briefly outline some of the most important limitations of this study: first, are concerns about subjective errors in the conduct of this research. Unlike in empirical neurolaw studies where judgements are analysed and coded by more than one researcher, providing an inter-rater reliability approach (in which different coders reassess the cases; for instance, two coders per case), this study was conducted by the author only and is hence open to subjective errors and errors in the interpretation of judicial reasons. Secondly, in some cases, judges did not clearly articulate their reasoning and specific interpretation of neuroscientific evidence in their judgements, hence they may not have discussed what types of neuroscientific tests were conducted by experts, or may have misinterpreted expert witnesses’ reports on results of these tests. Thirdly, not all judgements discussing neuroscience were transcribed or uploaded onto the databases used in this study, and hence this study was not able to analyse all the collected cases and prominently focused on judgements with discussions on brain imaging evidence.
Notwithstanding these limitations, this thesis, as the first study in Australia and one of few studies worldwide to systematically collect and analyse criminal cases where neuroscientific evidence has been introduced, presents a number of important contributions to the neurolaw literature.
2. Contributions of the study, recommendations, and future directions
First, the findings of this study highlight that the NSW experience seems, at least ostensibly, to be at odds with some of the claims and arguments in contemporary neurolaw literature. Unlike the picture painted in the neurolaw literature, which suggests that the use of neuroscientific evidence in courts is new and potentially problematic, the findings of the study suggest that from
248
1975 the NSW criminal justice system has welcomed various forms of neuroscientific evidence (e.g. structural and functional imaging and neuropsychological testing), and those have been used for a wide range of purposes in sentencing.
The broad definition of neuroscience used in this study, and the diversity of neurolaw claims analysed, which are relevant to a wide range of uses of neuroscience in practice (e.g. relevance to sentencing elements, reliably of evidence, expert witnesses, changing application of criminal rules), have made the findings of this study useful to a wide range of neurolaw scholars with different areas of interest. The findings of this study are well placed to serve as a base for future studies and scholarly arguments. Indeed, this study should act as a new starting point for scholarly discussions in the neurolaw literature, as neurolaw scholars can now provide more practical and useful claims based on insights on the actual use of neuroscience in courts.
Further, the assessment of claims in this study may also be useful for some proponents of neurolaw claims. This thesis has demonstrated the importance of examining neurolaw claims using empirical observations, and concurrently, the difficulty of investigating some of these claims in practice (e.g. see Chapter 6, the gap between neuroscience and the law’s language). Since there is a new wave of empirical research on the use of neuroscience in practice, it is timely that scholars proposing neurolaw claims take into account the applicability of their claims in such studies.
This thesis, similar to the empirical neurolaw studies from other jurisdictions, has promoted the empirical recording of actual uses of neuroscience in courts, and opens the door to international, comparative research on the use of neuroscience in different jurisdictions, i.e. Australia, USA, Canada, Netherlands and England and Wales. The reflections on the findings and methodology of this thesis can also be used as a framework by other Australian jurisdictions conducting similar empirical studies. Subsequently, these studies may together go towards enhancing our understanding of and generating a comprehensive picture of the use of neuroscientific evidence in Australia.
This research also has several practical applications. First, for judges, it provides feedback on the performance of the legal system with respect to neuroscientific evidence and acts as a useful resource for precedents in moving toward consistent judicial interpretations of neuroscientific evidence. This study should also be potentially valuable to lawyers. Defence lawyers may use the findings on the relevance of neuroscientific evidence to different sentencing elements to incorporate the possible outcomes of submitting various evidence in designing their strategies.
249
For prosecutors and defence lawyers, it can be useful for cross-examination and questioning the reliability and value of neuroscientific evidence with respect to the fact in issue.
This study recommends a number of guidelines and practice-oriented instruments: first, the criminal justice system should ensure that judicial officers have basic knowledge on how different neuroscientific tests are conducted, the limitations of these techniques and the utility of such information for the fact finder, such as direction towards potential rehabilitative measures. In line with these guidelines, educational institutes should also incorporate courses that present key principles of neurolaw and neurocriminology to law and criminology students, which these students may bring to the workforce when they practice. Finally, while several neurolaw courses and seminars have been provided to judicial officers in the US,100 in Australia, the necessity of such educational programs has been largely overlooked. Such programs can act as quick fixes to some of the concerns raised in this thesis concerning the use of neuroscientific evidence in courts.
In line with some of the empirical neurolaw studies from other jurisdictions,101 this study recommends that future studies investigating the use of biological evidence in courts should examine the use of neuroscientific and genetic evidence together, rather than separately. This is because the current literature of the study of human behaviour is moving towards integrating examinations of neuroscientific, genetic, psychological and environmental factors and how they contribute to behaviour.102 Reflection on the findings of this study also shows that an individual’s mental assessment in courts is conducted through consideration of these factors together (e.g. see discussion about the convergence of evidence in Chapter 7).103
Other than those directions for future research that I have discussed in the previous sections, I outline a number of areas that also deserve more attention. This thesis by evaluating seven neurolaw claims has provided only a narrow view of the vast intersection of law and neuroscience in courts. During examining the criminal judgements, many other interesting and important themes relevant to neurolaw were identified including assessment of witness testimony and their memory (i.e. credibility); criminal responsibility and the juvenile’s brain; examination of accused’s fitness to stand for trial and defences of mental illness and substantial impairment by
100 See, https://www.aaas.org/programs/scientific-responsibility-human-rights-law/judicial-seminars- emerging-issues-neuroscience 101 See, de Kogel and Westgeest, 580–605; Farahany, 485–509. 102 See, Farahany, 485–509. 103 This comprehensive approach to an individual’s mental assessment in NSW criminal courts is also supported by Dan Howard and Westmore. D Howard & Bruce Westmore, Crime and Mental Health Law in New South Wales: A Practical Guide for Lawyers and Health Care Professionals, 2nd ed., LexisNexis, AU, 2010. 250
abnormality of mind (only in judge alone cases).104 The use of neuroscientific evidence is not limited to NSW criminal courts, and there are some Civil cases that neuroscientific evidence was introduced to the court for purposes such as assessing the extent of an individual’s brain impairment for insurance and compensation purposes.
This thesis and the recent empirical neurolaw studies have contributed to shifting the neurolaw discourse from theory to actual practice in courts. The findings from these studies have demonstrated that there are inconsistencies between the use of neuroscience in practice and the assumptions upon which proponents of neurolaw claims have based their claims. Further, these empirical studies have identified issues and raised questions which need to be addressed. These include areas where more caution is necessary, and education for judicial officers may be required. As these studies have shown, neuroscience has been used in courts for a longer period that many may have imagined and implementing changes based on findings of empirical research is timely. I hope that this thesis stimulates further discussion and research on the use of neuroscience in practice.
104 In Chapter 3 (legal background) I discussed that where the accused is found unfit, the trial is held without a jury. In these courts, the judgment is recorded and occasionally uploaded onto public databases. 251
Bibliography
Articles/books
Adelman, G, ‘The neurosciences research program at MIT and the beginning of the modern field of neuroscience.’in Journal of the History of the Neurosciences: Basic and Clinical Perspectives, 19, 2010, 15–23. Adesegha, J, ‘The Role of Neuroscience and Psychology in the Criminal Justice System’. 2015. Aggarwal, N, & E Ford, ‘The neuroethics and neurolaw of brain injury’.in Behavioral sciences & the law, 31, 2013, 789–802. Aggarwal, NK, ‘Neuroimaging, culture, and forensic psychiatry’.in The journal of the American Academy of Psychiatry and the Law, 37, 2009, 239–244. Aguirre, GK, ‘Functional Neuroimaging: Technical, Logical, and Social Perspectives’.in Hastings Center Report, 44, 2014, S8–S18. Aharoni, E, C Funk, W Sinnott-Armstrong, & M Gazzaniga, ‘Can neurological evidence help courts assess criminal responsibility? Lessons from law and neuroscience’.in Annals of the New York Academy of Sciences, 1124, 2008, 145–160. Aigner, M, R Eher, S Fruehwald, P Frottier, K Gutierrez-Lobos, & SM Dwyer, ‘Brain abnormalities and violent behavior’.in Journal of Psychology & Human Sexuality, 11, 2000, 57–64. Al-Alosi, H, ‘Fantasy Crime’.University of New South Wales, 2016. Alimardani, A, ‘Neuroscience, criminal responsibility and sentencing in an islamic country: Iran’.in Journal of Law and the Biosciences, 2018,
252
Australian Law Reform Commission, ‘Access to court and tribunal records | ALRC’.
253
and Practice of Forensic Psychiatry.R Posner (ed), 2nd ed., CRC Press, 2003, pp. 473– 481. Bureau of Crime Statistics and Research (BOCSAR), ‘Re-offending statistics for NSW’.in BOCSAR, 2017,
254
changes after intervention in individuals with anti-social behaviour: A literature review’.in Criminal Behaviour and Mental Health, 25, 2015, 10–27. Cornet, LJM, PH van der Laan, HLI Nijman, N Tollenaar, & CH de Kogel, ‘Neurobiological factors as predictors of prisoners’ response to a cognitive skills training’.in Journal of Criminal Justice, 43, 2015, 122–132. Cowan, WM, DH Harter, & ER Kandel, ‘The emergence of modern neuroscience: Some implications for neurology and psychiatry’.in Annual Review of Neuroscience, 23, 2000, 343–391. Cozolino, L, The neuroscience of psychotherapy: Healing the social brain.WW Norton & Company, 2017. Crockett, M, ‘Beware neuro-bunk | TED Talk | TED.com’.in TED.com, 2012,
255
Impact of Behavioral Sciences on Criminal Law.NA Farahany (ed), New York, Oxford University Press, 2009, pp. 183–240. Farisco, M, & C Petrini, ‘The impact of neuroscience and genetics on the law: A recent Italian case’.in Neuroethics, 5, 2012, 317–319. Ferguson, CJ, & KM Beaver, ‘Natural born killers: The genetic origins of extreme violence’.in Aggression and Violent Behavior, 14, 2009, 286–294. Fitzgerald, R, & T Graham, ‘Assessing the risk of domestic violence recidivism’.in Crime and Justice Bulletin, 189, 2016. Fodor, JA, The modularity of mind: An essay on faculty psychology.MIT press, 1983. Forsberg, L, & T Douglas, ‘Anti-Libidinal Interventions in Sex Offenders: Medical or Correctional?’ in Medical Law Review, 24, 2016, 453–473. Franklin, CA, & NE Fearn, ‘Gender, race, and formal court decision-making outcomes: Chivalry/paternalism, conflict theory or gender conflict?’ in Journal of Criminal Justice, 36, 2008, 279–290. Franzini, A, C Marras, P Ferroli, O Bugiani, & G Broggi, ‘Stimulation of the posterior hypothalamus for medically intractable impulsive and violent behavior’.in Stereotactic and functional neurosurgery, 83, 2005, 63–66. Freckelton, IR, & H Selby, Expert evidence : law, practice, procedure and advocacy.in Pyrmont, NSW : Lawbook Co, 5th ed., 2013. Friedland, DR, JG Los, & DK Ryugo, ‘A modified Golgi staining protocol for use in the human brain stem and cerebellum’.in Journal of Neuroscience Methods, 150, 2006, 90–95. Frith, C, Making up the mind: How the brain creates our mental world.in Blackwell Publication, John Wiley & Sons, 2013. Gall, FJ, On the Function of the Brain and Each of Its Parts.in Boston: Marsh, Capen and Lyon, 1835, I. Gaudet, L, J Lushing, & K Kiehl, ‘Functional Magnetic Resonance Imaging in Court’.in AJOB Neuroscience, 5, 2014, 43–45. Gaudet, LM, & GE Marchant, ‘Under the radar: Neuroimaging evidence in the criminal courtroom’.in Drake Law Review, 64, 2016, 577–661. Gavaghan, C, ‘Neuroscience, deviant appetites, and the criminal law’ in Neuroscience and Legal Responsibility.NA Vincent (ed), Oxford University Press, 2013, p. 205. ---, ‘No gene for fate?’ in Bioethics at the Movies.S Shapshay (ed), Johns Hopkins University Press, 2009, p. 75. Gavaghan, C, & A Bastani, ‘Genes, Blame and Loss of Control: is there a Place in Criminal Law for a “Genetic Defense”?’ in Recent Advances in DNA & Gene Sequences (Formerly Recent Patents on DNA & Gene Sequences), 8, 2014, 119–125. Gazzaniga, MS, & MS Steven, ‘Free will in the twenty-first century: a discussion of neuroscience and the law’ in Neuroscience and the Law. 2004, pp. 51–70. Gkotsi, G-M, V Moulin, & J Gasser, ‘Neuroscience in the Courtroom: From responsibility to dangerousness, ethical issues raised by the new French law’.in L’Encephale, 41, 2014, 385–393. Glannon, W, ‘The Limitations and Potential of Neuroimaging in the Criminal Law’.in The Journal of Ethics, 18, 2014, 153–170. Glenn, A, & A Raine, ‘Neurocriminology: implications for the punishment, prediction and prevention of criminal behaviour’.in Nature Reviews Neuroscience, 15, 2014, 54–63. Glenn, AL, & KE McCauley, ‘How biosocial research can improve interventions for antisocial behavior’.in Journal of Contemporary Criminal Justice, 35, 2019, 103–119. Gold, LS, G Klein, L Carr, L Kessler, & SD Sullivan, ‘The emergence of diagnostic imaging technologies in breast cancer: Discovery, regulatory approval, reimbursement, and adoption in clinical guidelines’.in Cancer Imaging, 12, 2012, 13–24. Goodenough, O, & M Tucker, ‘Law and cognitive neuroscience’.in Annual Review of Law and Social Sciences, 6, 2010, 61–92. Goodenough, OR, & M Tucker, ‘Neuroscience Basics for Lawyers’.in Mercer L. Rev., 62, 2011, 945–1335. Goodman-Delahunty, J, & SL Sporer, ‘Unconscious influences in sentencing decisions: a
256
research review of psychological sources of disparity’.in Australian Journal of Forensic Sciences, 42, 2010, 19–36. Google Scholar, ‘For the law, neuroscience changes nothing and everything’.in Google Scholar,
257
Ishikawa, SS, & A Raine, ‘Prefrontal deficits and antisocial behavior: A causal model.’ in Causes of Conduct Disorder and Juvenile Delinquency.BB Lahey, TE Moffitt & A Caspi (eds), Guilford Press, 2003. Jancin, B, ‘Deep Brain Stimulation Helps Lift Refractory OCD Early data suggest that DBS is effective, but adjunctive therapy is indispensable.’in Clinical Psychiatry News, 36, 2008, 17. Jenkins, JVM, DP Woolley, SR Hooper, & MD De Bellis, ‘Direct and indirect effects of brain volume, socioeconomic status and family stress on child IQ’.in Journal of child and adolescent behavior, 1, 2013. Jones, O, J Buckholtz, & J Schall, ‘Brain imaging for legal thinkers: A guide for the perplexed’.in Stanford Technology Law Review, 5, 2009. Jones, O, & F Shen, ‘Law and neuroscience in the United States’ in International Neurolaw: A Comparative Analysis.TM Spranger (ed), Springer, 2012, pp. 349–380. Jones, OD, ‘Seven Ways Neuroscience Aids Law’ in Neurosciences and the Human Person: New Perspectives on Human Activities.AM Battro, S Dehaene, MS Sorondo & WJ Singer (eds), Vatican City, The Pontifical Academy of Sciences, 2013, pp. 181–194. Jones, OD, R Marois, MJ Farah, & HT Greely, ‘Law and neuroscience’.in Journal of Neuroscience, 33, 2013, 17624–17630. Judicial Commission of New South Wales, ‘Sentencing Bench Book’.in Judicial Commission of New South Wales,
258
Lakens, D, ‘Estimating the reproducibility of psychological science’.in Science, 349, 2015, 1– 10. Lavazza, A, & M De Caro, ‘Not so fast. On some bold neuroscientific claims concerning human agency’.in Neuroethics, 3, 2010, 23–41. Legal Profession Uniform Conduct (Barristers) Rules 2015. . Legrenzi, P, & C Umiltà, Neuromania: On the limits of brain science.trans. Frances Anderson, Oxford University Press, 2011. Leiber, MJ, & KC Fox, ‘Race and the impact of detention on juvenile justice decision making’.in NCCD news, 51, 2005, 470–497. Leutgeb, V, M Leitner, A Wabnegger, D Klug, W Scharmüller, T Zussner, et al., ‘Brain abnormalities in high-risk violent offenders and their association with psychopathic traits and criminal recidivism’.in Neuroscience, 308, 2015. Libet, B, CA Gleason, EW Wright, & DK Pearl, ‘Time of conscious intention to act in relation to onset of cerebral activity (readiness-potential) the unconscious initiation of a freely voluntary act’.in Brain, 106, 1983, 623–642. Linden, D, Neuroimaging and Neurophysiology in Psychiatry.Oxford, United Kingdom, Oxford University Press, 2016. Lindenberg, R, V Renga, LL Zhu, D Nair, & G Schlaug, ‘Bihemispheric brain stimulation facilitates motor recovery in chronic stroke patients’.in Neurology, 75, 2010, 2176–2184. Lipsey, MW, & FT Cullen, ‘The effectiveness of correctional rehabilitation: A review of systematic reviews’.in Annual Review of Law and Social Science, 3, 2007, 297–320. Lockhart, JJ, & S Satya-Murti, ‘Diagnosing crime and diagnosing disease: bias reduction strategies in the forensic and clinical sciences’.in Journal of forensic sciences, 62, 2017, 1534–1541. Logothetis, NK, ‘What we can do and what we cannot do with fMRI’.in Nature, 453, 2008, 869–878. Lozano, AM, HS Mayberg, P Giacobbe, C Hamani, RC Craddock, & SH Kennedy, ‘Subcallosal cingulate gyrus deep brain stimulation for treatment-resistant depression’.in Biological psychiatry, 64, 2008, 461–467. MacArthur Foundation Research Network on Law and Neuroscience, ‘Cumulative Total of Law and Neuroscience Publications: 1984– 2017’.
259
Law, 21, 2014, 779–790. Mears, DP, JC Cochran, SE Siennick, & WD Bales, ‘Prison visitation and recidivism’.in Justice Quarterly, 29, 2012, 888–918. Mechelli, A, CJ Price, KJ Friston, & J Ashburner, ‘Voxel-based morphometry of the human brain: methods and applications’.in Current medical imaging reviews, 1, 2005, 105–113. Mehta, A, ‘“Neuroeducation” Emerges as Insights into Brain Development, Learning Abilities Grow’.in The Dana Foundation, 2009,
260
Forward.Washington, DC, The National Academies Press, 2009,
261
---, The psychopathology of crime: Criminal behavior as a clinical disorder.in Academic Press, Elsevier Science, 1993. Raine, A, M Buchsbaum, & L Lacasse, ‘Brain abnormalities in murderers indicated by positron emission tomography’.in Biological Psychiatry, 42, 1997, 495–508. Raine, A, MS Buchsbaum, J Stanley, S Lottenberg, L Abel, & J Stoddard, ‘Selective reductions in prefrontal glucose metabolism in murderers’.in Biological Psychiatry, 36, 1994, 365– 373. Ray, A, & SM Bowyer, ‘Clinical applications of magnetoencephalography in epilepsy’.in Annals of Indian Academy of Neurology, 13, 2010, 14–22. Reeves, D, MJ Mills, & SB Billick, ‘Limitations of brain imaging in forensic psychiatry’.in The journal of the American Academy of Psychiatry and the Law, 31, 2003, 89–96. Report of Therapeutics and Technology Assessment Subcommittee of the American Academy of Neurology, ‘Assessment of brain SPECT’.in Neurology, 46, 1996, 278–285. Rhodes, RE, F Rodriguez, & P Shah, ‘Explaining the alluring influence of neuroscience information on scientific reasoning.’in Journal of Experimental Psychology: Learning, Memory, and Cognition, 40, 2014, 1432. Richards, D, ‘Elite interviewing: Approaches and pitfalls’.in Politics, 16, 1996, 199–204. Risinger, DM, MJ Saks, WC Thompson, & R Rosenthal, ‘The Daubert/Kumho implications of observer effects in forensic science: Hidden problems of expectation and suggestion’.in California Law Review, 90, 2002, 1–56. Rizzi, M, & CE Marras, ‘Deep Brain Stimulation for the Treatment of Aggressive Behaviour: Considerations on Pathophysiology and Target Choice’.in Stereotactic and Functional Neurosurgery, 95, 2017, 114–116. Rolfe, D, ‘It wasn’t me, it was my brain’.in LSJ: Law Society of NSW Journal, May, 2016, 26– 32. Rose, N, Our Psychiatric Future: the politics of mental health.in Polity, Cambridge, John Wiley & Sons, 2018. ---, The Politics of Life Itself: Biomedicine, Power, and Subjectivity in the twenty-first century.in Princeton University Press, 2007. Rose, N, & JM Abi-Rached, Neuro: The new brain sciences and the management of the mind.in Princeton University Press, EBSCO Publ, Princeton University Press, 2013. Roskies, A, ‘Neuroethics for the new millenium’.in Neuron, 35, 2002, 21–23. ---, ‘Neuroscientific challenges to free will and responsibility’.in Trends in cognitive sciences, 10, 2006, 419–423. Roskies, A, & W Sinnott-Armstrong, ‘Brain images as evidence in the criminal law’.in Law and Neuroscience, 2011. Rothenberg, K, & A Wang, ‘The Scarlet Gene: Behavioral Genetics, Criminal Law, and Racial and Ethnic Stigma’.in Law and Contemporary Problems, 69, 2006, 343–365. Rugg, D, M, & M Coles, Electrophysiology of mind: Event-related brain potentials and cognition.in Oxford University Press, 1995. Rushworth, N, Out of sight, out of mind: People with an acquired brain injury and the criminal justice system.in Brain Injury Australia, 2011. Samuels, A, C O’Driscoll, & S Allnutt, ‘Fitness issues in the context of judicial proceedings’.in Australasian Psychiatry, 15, 2007, 212–216. Sartori, G, S Pellegrini, & A Mechelli, ‘Forensic neurosciences: from basic research to applications and pitfalls’.in Current Opinion in Neurology, 24, 2011, 371–377. Scarpazza, C, S Ferracuti, A Miolla, & G Sartori, ‘The charm of structural neuroimaging in insanity evaluations: guidelines to avoid misinterpretation of the findings’.in Translational Psychiatry, 8, 2018, 227–237. Schauer, F, ‘Neuroscience, lie-detection, and the law: Contrary to the prevailing view, the suitability of brain-based lie-detection for courtroom or forensic use should be determined according to legal and not scientific standards’.in Trends in Cognitive Sciences, 14, 2010, 101–103. Schiltz, K, JG Witzel, J Bausch-Hölterhoff, & B Bogerts, ‘High prevalence of brain pathology in violent prisoners: A qualitative CT and MRI scan study’.in European Archives of
262
Psychiatry and Clinical Neuroscience, 263, 2013, 607–616. Schleim, S, ‘Brains in context in the neurolaw debate: The examples of free will and “dangerous” brains’.in International Journal of Law and Psychiatry, 35, 2012, 104–111. Schweitzer, NJ, & MJ Saks, ‘Neuroimage evidence and the insanity defense’.in Behavioral sciences & the law, 29, 2011, 592–607. Schweitzer, NJ, MJ Saks, ER Murphy, AL Roskies, W Sinnott-Armstrong, & LM Gaudet, ‘Neuroimages as evidence in a mens rea defense: No impact’.in Psychology, Public Policy, and Law, 17, 2011, 357–393. Schweizer, R, & S Bischof, ‘Switzerland: Brain Research and the Law’.in International Neurolaw, 2012, 269–287. Shaw, E, ‘Direct Brain Interventions and Responsibility Enhancement’.in Criminal Law and Philosophy, 8, 2014, 1–20. Shen, EH, CC Overly, & AR Jones, ‘The Allen Human Brain Atlas: comprehensive gene expression mapping of the human brain’.in Trends in neurosciences, 35, 2012, 711–714. Shen, FX, ‘The Law and Neuroscience Bibliography: Navigating the Emerging Field of Neurolaw’.in International Journal of Legal Information, 38, 2010, 352–399. ---, ‘The overlooked history of neurolaw’.in Fordham Law Review, 85, 2016, 667–695. Silva, JA, ‘Forensic psychiatry, neuroscience, and the law’.in Journal of the American Academy of Psychiatry and the Law, 37, 2009, 489–502. Sinnott-Armstrong, W, A Roskies, T Brown, & E Murphy, ‘Brain images as legal evidence’.in Episteme, 5, 2008, 359–373. De Smedt, B, & RH Grabner, ‘Potential applications of cognitive neuroscience to mathematics education’.in Zdm, 48, 2016, 249–253. Snead, OC, ‘Neuroimaging and the Complexity of Capital Punishment’.in NYUL Rev., 82, 2007, 1265–1339. Space, P, A Network, & B Campaign, ‘Chapter I: Neuroscience Before Neuroscience, WWII to 1969’.in sfn.org,
263
1991, 293–305. The Royal Society, Brain waves 4: neuroscience and the law.in The Royal Society, London, UK, 2011. The Sydney Morning Herald, ‘Judge blessed with flawless logic’. 2012,
264
Yang, Y, & A Raine, ‘Prefrontal structural and functional brain imaging findings in antisocial, violent, and psychopathic individuals: a meta-analysis’.in Psychiatry Research: Neuroimaging, 174, 2009, 81–88. Zapf, PA, & IE Dror, ‘Understanding and mitigating bias in forensic evaluation: Lessons from forensic science’.in International Journal of Forensic Mental Health, 16, 2017, 227–238. Zhang, Y, & BM Wildemuth, ‘Qualitative analysis of content’.in Applications of social research methods to questions in information and library science, 318, 2016.
Cases
Ayoubi v R [2006] NSWCCA 364 Ballard v R [2011] NSWCCA 193 Benitez v R [2006] 160 A Crim R 166 BP [2010] NSWCCA 159 Bugmy v R [2013] 249 CLR 571 Bugmy v R [2013] HCA 37 BW v R (2007) NSWCCA 236 BW v R [2011] NSWCCA 176 Caristo v R [2011] NSWCCA 7 Carroll v R [2012] NSWCCA 118 Chaplin v R [2006] NSWCCA 40 Chen v R [2018] NSWCCA 106 Clay (2007) NSWCCA 106 Clinton v R [2014] NSWCCA 320 DBW v R (2007) NSWCCA 236 De La Rosa [2010] NSWCCA 194 Director of Public Prosecutions (Cth) v De La Rosa [2010] NSWCCA 194 Doolan v R (2006) 160 A Crim R 54 DPP v Houn [2008] NSWLC 16 EG v R [2015] NSWCCA 21 Engert [1995] 84 A Crim R 67 FD v R [2013] NSWCCA 139 Heath v R [2016] NSWCCA 24 Heath v Regina [2016] NSWCCA 24 Hitchcock v R [2016] NSWCCA 226 Iglesias v R [2006] NSWCCA 261 IMM v The Queen [2016] HCA 14 In R v Huon [2012] NSWSC 1092 Ionatana v R [2008] NSWCCA 95 Kennedy v R [2008] 181 A Crim R 185; [2008] NSWCCA 21 Lam v R [2015] NSWCCA 143 Markarian v R [2005] 228 CLR 357 Mudlrock [2011] HCA 39, 244 CLR 120 Munn v R [2009] NSWCCA 218 Ngati v R [2014] NSWCCA 125 O’Neill-Shaw v R [2010] NSWCCA 42 R v Almoustafa [2009] NSWDC 217 R v Basilious [1999] NSWCCA 219
265
R v Blaikie [2007] NSWDC 311 R v Bourchas (2002) 133 A Crim R 413 R v Clark [No 3] [2008] NSWSC 795 R v Clay, Lonsdale and JM [2006] NSWSC 1220 R v Coulter [2005] NSWSC 101 R v Daniel Stephen Lewis [2004] NSWCCA 383 R v Davis [1999] NSWSC 876 R v Do (No. 4) [2015] NSWSC 512 R v El-Hayek [2004] 144 A Crim R 90 R v Fisher [2009] NSWSC 348 R v Fisher [2016] NSWCCA 226 R v Goodridge (No 2) [2012] NSWSC 1180 R v Gordon [1994] 71 A Crim R 459 R v Harris [2000] 50 NSWLR 409 R v Hemsley [2004] NSWCCA 228 R v Henry [1999] 46 NSWLR 346 R v Hussein [2011] NSWDC 103 R v Isaacs [1997] 41 NSWLR 374 R v Israil [2002] NSWCCA 255 R v Justin [2013] NSWDC 163 R v Kennedy [2008] NSWSC 703 R v King [2004] NSWCCA 444 revised - 7/12/2004 R v Lea-Caton [2007] NSWSC 1294 R v Lewis [2001] NSWCCA 448 R v McCann [2012] NSWSC 1462 R v McMillan [2005] NSWCCA 28 R v McNamara (2005) NSWCCA 195 R v Merritt [2004] 59 NSWLR 557 R v Mills [2005] 154 A Crim R 40 at [49] R v Naa [2009] NSWSC 1077 R v Ngo (No 3) [2001] 125 A Crim R 495 R v O'Connor [2013] NSWDC 272 R v Orchard [2011] NSWDC 151 R v Paredes, R v Barillaro, R v Donevski [2011] NSWDC 48 R v Peisley [1990] 54 A Crim R 42 R v Pitt [2005] NSWCCA 304 R v Pogson [2012] 82 NSWLR 60 R v Ray [2013] NSWSC 767 R v Sankey [2002] NSWCCA 135 R v Santos [2001] NSWSC 923 R v Scott (2005) NSWCCA 152 R v Smith (1987) 44 SASR 587 R v Sullivan [2010] NSWSC 755 R v Tang (2006) 65 NSWLR 681 R v Toki [2003] NSWCCA 125 R v Tortell, R v Tsegay [2007] NSWCCA 313 R v Tsiaras (1996) 1 VR 398 R v Veen [2000] NSWSC 656
266
R v Wallace [2006] NSWSC 897 R v Way [2004] 60 NSWLR R v Whyte (2002) 55 NSWLR 252 R v Z [2006] 167 A Crim R 436 R v Z [2006] NSWCCA 342 Ryan v The Queen [2001] 206 CLR 267 Shi v R [2014] NSWCCA 276 State of New South Wales v Harrison [2009] NSWSC 198 State of New South Wales v Scott [2013] NSWSC 1834 State of New South Wales v Thomas (Final) [2009] NSWSC 1410 State of New South Wales v Hill [2014] NSWSC 1803 (28 November 2014) Sumpton v R [2016] NSWCCA 162 The State of New South Wales v Johnson [2010] NSWSC 590 Thomas v R (Commonwealth) (2006) NSWCCA 313 Tilyard v Regina (2007) NSWCCA 7 Tuite v The Queen [2015] VSCA 148 Uzabeaga (2000) NSWCCA 381, 119 A Crim R 452 Van Can Ha v R (2008) NSWCCA 141 Veen v R [1979] HCA 7; [1979] 143 CLR 458 (28 February 1979) Veen v R [1988] [1988] HCA 14; [1988] 164 CLR 465 (29 March 1988) WW v R [2012] NSWCCA 165
Web sources http://bluebrain.epfl.ch http://www.austlii.edu.au http://www.austlii.edu.au/au/cases/nsw/NSWDC/ http://www.austlii.edu.au/au/cases/nsw/NSWSC/ http://www.austlii.edu.au/databases.html http://www.austlii.edu.au/faq.html#q1.9 http://www.hcourt.gov.au/about/operation http://www.hcourt.gov.au/about/role-of-the-high-court http://www6.austlii.edu.au/cgi-bin/viewdb/au/cases/nsw/NSWCCA/ http://www7.austlii.edu.au/cgi-bin/viewdb/au/cases/nsw/NSWSC/ https://neurolaw.edu.au https://neurolaw.edu.au/keywords https://scilaw.org https://www.aaas.org/programs/scientific-responsibility-human-rights-law/judicial-seminars- emerging-issues-neuroscience https://www.braininitiative.nih.gov https://www.caselaw.nsw.gov.au https://www.caselaw.nsw.gov.au/about https://www.caselaw.nsw.gov.au/policy#release https://www.humanbrainproject.eu https://www.hunterimaging.com.au/department/our-specialists https://www.qualtrics.com/au/
267