Recklessness: Report

7. Proposed alternative definitions of recklessness

Overview

• Several stakeholders supported alternative definitions of recklessness during our inquiry.

• Some proposed definitions based on the tests used in other jurisdictions:

– The Office of Public Prosecutions (OPP) recommended a definition with the same threshold as is used in New South Wales and that has similarities to the definitions in England and Wales and New Zealand.

– Victoria Police supports a ‘possibility’ threshold based on the High Court’s decision in Aubrey v The Queen (‘Aubrey’).

– Steven Tudor proposed a definition influenced by the law in England and Wales, although his support for a new definition was qualified.

– Some students from the University of Melbourne proposed definitions similar to the Commonwealth definition, and others suggested keeping the current definition but adding a reasonableness test.

• Caterina Politi said recklessness should be based on ‘what a reasonable person’ would or could have foreseen.

• Each of the proposed definitions has disadvantages. None would achieve consistency between Victoria and other jurisdictions. Consistency is an impractical goal given the different path Victoria’s criminal law has taken and our distinctive and integrated architecture of offences.

The Office of Public Prosecutions

7.1 The submission from the OPP suggests inserting the following definition in the ‘offences against the person’ subdivision of the Crimes Act 1958 (Vic):

a person is reckless as to a result or circumstance if they were aware of the possibility that:

• their actions would bring about the result; or

• the circumstance existed; and

• having regard to the risk their actions were unreasonable.

7.2 That definition adopts the possibility threshold for recklessness used in New South Wales, but explicitly:

• distinguishes between foresight of a result or circumstance

• articulates the objective part of the test: the accused’s actions were unreasonable.

7.3 The OPP said a test that distinguishes between results and circumstances is necessary because for some offences against the person, being reckless about a circumstance is relevant. These are offences against on-duty emergency workers or other specified workers (see Chapter 4).

7.4 However, during our inquiry we did not hear about any difficulties applying the current Victorian test to offences against on-duty emergency workers and other specified workers.[1]

7.5 The OPP says its proposed definition is based on the common law of New Zealand and the United Kingdom and is a formulation that has ‘proven to be sufficiently flexible to cater for a broad range of offences’ in those jurisdictions.[2] But the definition of recklessness applicable to different categories of offences in New Zealand was contested until quite recently (see Chapter 6). Further, it remains unclear if England and Wales have one definition of recklessness applicable to offences in the Offences Against the Person Act 1861 and a separate definition for other offences (see Chapter 6).

7.6 The OPP says that the express incorporation of an objective ‘unreasonableness’ element is ‘broadly consistent’ with the recklessness definition in New South Wales, where the possibility test is implicitly limited by the concept of unreasonableness.[3]

7.7 Although advocating for an explicit objective element, the OPP says this ‘would not necessarily overcomplicate jury directions. Indeed, it may not always require an explicit direction at all.’[4]

7.8 The OPP points out that in New South Wales:

juries are not generally directed to consider social utility or [the] unreasonableness of the accused’s act in determining whether the accused was reckless. Rather, they are simply directed to consider whether the accused realised their act may possibly cause the result.[5]

7.9 The OPP says that for many offences against the person in Victoria, ‘the unreasonableness of the actions will not be in issue’ and it would not be necessary to direct a jury about the unreasonableness of the act.[6] For example, it says that ‘punching someone in the head’ or ‘kicking a prone person’ will ‘rarely if ever be reasonable’.[7]

7.10 As well as consistency with other jurisdictions, the OPP says there are additional reasons to adopt its proposed definition.[8] We address these in Chapters 8 and 9.

Victoria Police

7.11 Victoria Police expressed strong support for a new definition of recklessness based on the definition set out by the High Court in Aubrey: ‘foresight of the possibility of harm with an objective element of unreasonableness’.[9]

7.12 Victoria Police said the Crimes Act should be amended to include this definition for offences against the person in Part I, Division 1(4). It also said the same definition should be considered for other Crimes Act offences.[10]

7.13 In addition to the definition based on Aubrey, Victoria Police suggested that:

the element of recklessness that requires the prosecution to prove the accused acted without lawful justification or excuse, be incorporated into the statutory definition of recklessness.[11]

7.14 Victoria Police said this would ensure that:

people employed in high-risk occupations, such as surgeons, paramedics, police officers (where they may be required to take reasonable risks), are protected from prosecution.[12]

7.15 Many offences against the person have a separate ‘without lawful excuse’ element, requiring the prosecution to disprove any justifications, excuses or defences that are open on the evidence (see Chapter 4). This is not ‘an element of recklessness’ as such. The effect of incorporating it into the definition of recklessness could be to:

• alter the existing structure of recklessness offences, removing ‘without lawful excuse’ as a stand-alone element and instead making it part of recklessness, or

• create offences in which this element must be established twice.

7.16 The first option would lead to disparity between the structure of intentional and recklessness offences, creating unnecessary complexity without any obvious benefit. The second option would lead to unnecessary duplication.

Dr Steven Tudor

7.17 In his submission, Dr Steven Tudor is critical of both ‘probable’ and ‘possible’ definitions of recklessness. He says that both:

seem to present the task facing fact-finders as if the sole question was the degree of likelihood of the circumstance or result occurring. This makes it look like the issue … is simply a matter of trying to work out, in a quasi-mathematical way, just what that degree of likelihood was in a particular case…[13]

7.18 In his view, both thresholds implicitly rely on an objective assessment of the reasonableness or social utility of the act in question, resulting in a lack of clarity and ‘contortions of ordinary English language’. This, he says, is ‘clearly undesirable’ because the law needs to be ‘clear and honest’.[14]

7.19 The probability test in Victoria is subjective. In instances where foresight of risk is fleeting, it may be difficult for juries to avoid importing an objective assessment, but they can be cautioned that they should not do this. In our discussion of folk psychology, we discuss the artificiality inherent in attributing a state of mind to a person who acted on the spur of the moment (see Chapter 2). This is an issue that also arises in relation to intentional offences. Stakeholders told us that in practice it does not cause problems for the prosecution of criminal offences (see Chapters 2 and 8).

7.20 We do not consider that the issue is so serious as to justify a wholesale revision of fundamental principles of criminal responsibility. It certainly does not justify changing the test for recklessness in the absence of more comprehensive reform than we were asked to consider in our terms of reference.

7.21 Dr Tudor says that the Commonwealth test’s reference to ‘substantial’ and ‘unjustifiable’ risk in the Criminal Code Act 1995 (Cth) (the Commonwealth Code) ‘goes part of the way towards [the] goal’ of a test that explicitly assesses:

• the degree of risk

• the seriousness of the harm in relation to which the risk is taken

• the social utility or value of the conduct.[15]

7.22 He points out that:

there is more to the unjustifiability of taking [a] risk [than] the degree (or substantiality) of the risk. The other factors of seriousness of the risked harm and the social value or utility of the conduct need to go into the mix. The Commonwealth Code’s definition certainly allows for that. But it should be made explicit.[16]

7.23 For this reason, he supports a version of the test used in England and Wales, which he claims is explicit about containing an objective element.[17] His proposed definition is:

A person is reckless (or acts recklessly) with regard to a circumstance or result where:

(i) they are aware of a risk that the circumstance exists or will exist or that the result will occur; and

(ii) it is, in the circumstances known to them, unjustifiable to take that risk.

In assessing whether taking the risk was unjustifiable, the fact-finder must take into account, among any other relevant matters, the following:

(a) the degree of the risk; that is, how likely it is that the circumstance exists or will exist or that the result will occur;

(b) the seriousness of the circumstance or result, including the seriousness of any harm involved in that circumstance or result; and

(c) the social value or utility of the person’s act.

7.24 Dr Tudor qualifies his support for this test by saying that empirical research is required to work out if the test can be easily understood by juries. In his view, the test should only be adopted if research indicates juries are able to understand and apply it without difficulty.[18]

Other submissions

7.25 Some students from the University of Melbourne supported alternative definitions of recklessness. One submission suggested a definition very close to the Commonwealth Code definition but without reference to ‘circumstance’ or ‘result’, proposing that:

A person is reckless with respect to a situation if [they are] aware of a substantial risk and having regard to the circumstances known to [them], it is unjustifiable to take the risk.[19]

7.26 Another student submission proposed that recklessness for offences other than murder should be defined as a ‘conscious and unjustifiable disregard of a substantial risk of infliction of harm upon another person.’[20] This includes the requirement from the United States Model Penal Code that the recognition of risk be ‘conscious’ (see

Chapter 6).

7.27 Another student submission supported a test for recklessness with the current threshold of probability but including a reasonableness assessment: ‘the risk was unreasonable in the circumstances known to the accused.’[21]

7.28 We also received a submission from Caterina Politi, whose son David Cassai was killed in a ‘one-punch’ attack. Ms Politi suggested a purely objective test for recklessness:

based on what a reasonable person would have foreseen or could have foreseen. This involves considering whether the risks and potential consequences of the accused’s conduct were reasonably foreseeable and not justified in taking.[22]

7.29 Ms Politi said that ‘a clear definition of recklessness’ could reduce ‘the risk of uncertainty or confusion in legal proceedings’.[23]

The proposed definitions do not improve on the current test

7.30 The advantages of the proposed definitions do not outweigh the advantages of the current Victorian test which we discuss in Chapter 10. As we discuss further in Chapter 11, a new definition would create risks.

7.31 We have not found a ‘gap’ in the existing hierarchy of offences that requires a different test for recklessness (see Chapter 8).

7.32 Ms Politi proposed a purely objective test. The definitions proposed by the OPP, Victoria Police, Dr Tudor, and some students from the University of Melbourne include an objective component, which increases their complexity.

7.33 Dr Greg Byrne told us that having two limbs to assess rather than one is a ‘complication’ and including an objective limb ‘makes comprehension more difficult’ for jurors.[24]

7.34 The complexity involved in assessments of social utility and reasonableness is apparent in Ormerod and Laird’s discussion of the English test:

Whether it is justifiable to take a risk depends on the social value of the activity involved relative to the probability and the gravity of the harm which might be caused.[25]

7.35 An assessment of ‘social value’ ‘relative to the probability and the gravity of the harm’ introduces multiple concepts, each of which may be contentious. Different people might reasonably arrive at different conclusions about what behaviour has ‘social value’ when taking into account the likelihood of harm and its seriousness, ‘introduc[ing] complexities into juries’ considerations and room for juries to come to different views about the same conduct.’[26]

7.36 By comparison, an assessment of whether a person had foresight of probable harm is more straightforward (see Chapter 10).

The proposed definitions will not achieve consistency with other jurisdictions

7.37 The OPP says its proposed test is ‘broadly in line with the common law in New South Wales’ and would allow Victoria to ‘move into step with other jurisdictions’.[27]

7.38 However, others told us that given the different criminal law frameworks in New South Wales and Victoria, changing Victoria’s definition of recklessness for offences against the person from probability to possibility would not lead to consistency with New South Wales.

7.39 In the DPP Reference in the High Court, Justices Gageler, Gordon and Steward said that:

The DPP’s reliance on a policy preference for consistency in the meaning of like provisions in different States as a reason for this Court to alter the meaning of s 17 of the Crimes Act is misplaced … Each State has taken a different view on the criminality to be ascribed to the conduct.[28]

7.40 Dr Greg Byrne emphasised the ‘translation process’ that occurs as the common law develops within discrete jurisdictions, leading to inconsistencies between them.[29]

7.41 Dr Byrne and the Criminal Bar Association (CBA) said the quest for consistency between New South Wales and Victoria ignores the differences in the elements of their offences and between the definitions of serious injury and grievous bodily harm.[30]

7.42 Barrister Dermot Dann and solicitor Felix Ralph pointed out that recklessness has a central place ‘in the criminal justice landscape in Victoria’ and this landscape ‘is a finely calibrated mix between common law and statute.’[31] As such, the contours of our criminal justice landscape are quite different to those in New South Wales and other jurisdictions.

7.43 The CBA told us:

it is mistaken to think that Victoria cleaving to the definition of recklessness that prevails in New South Wales would bring about [consistency] …

merely ‘cutting and pasting’ the definition of ‘recklessness’ that prevails in New South Wales would not result in similar criminal laws applying in Albury and Moama—instead, it would result in more disparate treatment of equivalent conduct on each side of the Murray River.[32]

7.44 On the other hand, the County Court told us that inconsistency between Victoria and other jurisdictions makes it difficult to consider case law from other intermediate appellate courts.[33]

7.45 In her commentary on the DPP Reference, Michaela Puntillo says that the outcome in the High Court, leaving the decision as to whether to adopt the Aubrey test or keep the existing Campbell test (see Chapter 3) to Parliament:

generates the unattractive consequence of inconsistency and incoherence across different Australian states in respect of the concept of recklessness in similar statutory offences.[34]

7.46 But there are many inconsistencies, some stark and others subtle, between the criminal law in Victoria and other Australian jurisdictions (see Chapter 6). They occur at the macro level in the criminal law frameworks of each jurisdiction, as well as in the detail of their laws regarding recklessness and other fault elements.

7.47 As the CBA told us, consistency with other jurisdictions is a ‘radical objective’ that would require ‘a wholesale revision of the criminal laws of the various states’.[35]

Conclusion

7.48 It would not be a simple matter to import a definition of recklessness from another jurisdiction into Victorian law. No other jurisdiction has a model of recklessness that provides a clear improvement on the current law in Victoria, and the proposed definitions are more complex than Victoria’s definition. Consistency is in any event an impractical goal, because Victoria’s criminal law has taken a different trajectory and has its own integrated architecture of offences (see Chapters 3 and 4).


  1. A review of reforms that strengthened sentencing requirements for injury offences against emergency workers found the reforms were working as intended. While the focus of the review was not recklessness, at least two of the cases reviewed involved sentences for recklessly causing injury to an emergency worker on duty, suggesting that this offence is being successfully prosecuted: Victorian Government, Sentencing of Emergency Worker Harm Offences: Review into the Operation and Effectiveness of the Sentencing Amendment (Emergency Worker Harm) Act 2020 (Report, 2022) 6, 7 (case 2).

  2. Submission 10 (Office of Public Prosecutions).

  3. Ibid (citations omitted).

  4. Ibid.

  5. Ibid.

  6. Ibid; If an accused person claims they were acting in self-defence, this will raise the issue of whether their actions were ‘a reasonable response in the circumstances as [they] perceived them’, but that is so regardless of the definition of recklessness that applies: Crimes Act 1958 (Vic) s 322K.

  7. Submission 10 (Office of Public Prosecutions).

  8. Ibid. The OPP makes seven arguments for its proposed definition. The OPP says the definition: (1) would better align legal and moral culpability; (2) would be consistent with other jurisdictions; (3) is relatively simple; (4) will only infrequently require juries to apply a different threshold to other offences; (5) will not over-criminalise conduct; (6) is distinct from negligence; and (7) will correct a long-standing error in law.

  9. Submission 7 (Victoria Police) citing Aubrey v The Queen [2017] HCA 18; (2017) 260 CLR 305 (emphasis added).

  10. Submission 7 (Victoria Police).

  11. Ibid.

  12. Ibid.

  13. Submission 8 (Dr Steven Tudor) (emphasis in original).

  14. Ibid.

  15. Ibid.

  16. Ibid.

  17. In Chapter 6, we note the current characterisation of the test in England and Wales as purely subjective.

  18. Submission 8 (Dr Steven Tudor).

  19. Submission 4 (Waller, Herszberg, Muldoon (students at the University of Melbourne)).

  20. Submission 2 (Peck, Borchard, Carlei, Ciampoli (students at the University of Melbourne)).

  21. Submission 3 (Kavaleris, Shehnah, Aforozis, Vinci (students at the University of Melbourne)) (citations omitted).

  22. Submission 21 (Caterina Politi).

  23. Ibid.

  24. Consultation 9 (Dr Greg Byrne PSM).

  25. David Ormerod and Karl Laird, Smith, Hogan, and Ormerod’s Criminal Law (Oxford University Press, 16th ed, 2021) 106 [3.2.2.3].

  26. Consultation 4 (Criminal Bar Association).

  27. The OPP also said its proposed formulation is broadly in line with the Australian Capital Territory, the United Kingdom and New Zealand, and consistent with South Australia’s statutory offences of causing harm and serious harm: Submission 10 (Office of Public Prosecutions). See Chapter 6 for discussion of recklessness in these jurisdictions.

  28. Director of Public Prosecutions Reference No 1 of 2019 [2021] HCA 26 [58]; (2021) 274 CLR 177, 201 [58] (Gageler, Gordon and Steward JJ).

  29. Consultation 9 (Dr Greg Byrne PSM).

  30. Ibid; Submission 6 (Criminal Bar Association); See also Submissions 1 (Clifton, Liu, Neulinger, Wong (students at the University of Melbourne)); 5 (McGavin, Jenkins-Smales, NcNaughton, Allen (students at the University of Melbourne)).

  31. Submission 12 (Dermot Dann KC and Felix Ralph).

  32. Submission 6 (Criminal Bar Association) (emphasis in original).

  33. Submission 15 (County Court of Victoria).

  34. Michaela Puntillo, ‘What Does Parliament Want? Director of Public Prosecutions (Vic) Reference No 1 of 2019 (2021) 392 ALR 413’ (2022) 43(2) Adelaide Law Review 1003, 1013 citing Director of Public Prosecutions Reference No 1 of 2019 [2021] HCA 26; (2021) 274 CLR 177, 218 [101] (Edelman J).

  35. Submission 6 (Criminal Bar Association). Dermot Dann KC and Felix Ralph told us that ‘a radical overhaul of the criminal justice system is unnecessary and no good reason has been advanced in support of such a dramatic overhaul’: Submission 12 (Dermot Dann KC and Felix Ralph).