Wilfred Ong
Introduction
The offence of attempt is part of a larger category of crimes collectively called the inchoate offences, which also include the offences of encouraging and assisting; and conspiracy. It seems to me that the time has come for the law to recognise an altogether separate inchoate offence of endangerment. This is because attempts are failed attacks,[1] and attacks and endangerments are fundamentally distinct wrongs eliciting different moral responses.[2]
Viewing recklessness, which is the paradigm fault of endangerment,[3] as sufficient mens rea for attempts goes against the principle of fair labelling and distorts the ordinary meaning of attempts. It widens the scope of attempts in order to criminalise what are really endangerments. I argue that we should uphold the attack-endangerment distinction because accurate offence labels are central to effective deterrence. By criminalising endangerments separately, we would be doing more to highlight the wrongfulness of endangerments by highlighting the indifference that gives rise to such actions.[4] Therefore, a new separate inchoate endangerment offence should be created and the mens rea for attempts should be confined to direct intention.
I shall divide this work into three parts. First, I will introduce the issue and the law as it stands. Second, I will introduce Duff’s and Yang’s views. I will then resolve their differences to create a unified approach and apply that to the more difficult cases of intended endangerment and oblique intention. Third, I will suggest guidelines for reform and provide justifications for them.
I: The Law as it Stands
The crucial question is what should be the mens rea for attempts, particularly for substantive offences which do not require intention. The Criminal Attempts Act 1981 speaks of requiring intention to commit the substantive offence.[5] However, it is not explicit on whether the same requirement applies to substantive offences which do not require intention. For instance, the offence of criminal damage requires a mens rea of ‘intending to destroy or damage any such property or being reckless as to whether any such property would be destroyed or damaged’.[6] In such cases, should recklessness suffice for attempts?
There is no clear answer in case law. In the context of rape, Khan[7] held that it did. As long as there was intention to have intercourse, recklessness as to consent was enough as mens rea for attempted rape. In the context of converting criminal property however, Pace[8] held that it did not. Mere recklessness as to whether the property was stolen was not enough. Knowledge or belief as to the circumstance element was required. Thus, a possible interpretation is that whether or not recklessness suffices is based on the severity of the substantive offence.
The Law Commission reports also offer limited guidance. In 1980, the Law Commision recommended that ‘an intent to bring about each of the constituent elements of the offence attempted’[9] was required. This was because it was thought that attempts involved intention to commit the complete offence as opposed to some element(s) of the offence.[10] However, the position was reversed in 1989 and the Khan approach was adopted.[11] This was due to concerns that the 1980 report would make the scope of criminal attempts too narrow. The reasoning was that by requiring intent as to every element of the offence, defendants who should be culpable would escape criminal liability altogether.[12] While the policy considerations here are compelling, I think we ought to strive for a more principled approach to this issue.
It seems to me that, in viewing recklessness as being sufficient, the scope of attempts was unsatisfactorily expanded beyond the ordinary meaning of an attempt in order to criminalise culpable unconsummated acts such as culpable endangerment. We should not ‘water down’ the meaning of ‘intent’ in the Criminal Attempts Act and should confine the scope of attempts to direct intention.
There is, however, a principled way out of this untidy compromise. By viewing attempts as failed attacks, I argue that the problems of adequate criminalisation and accurate labelling can be solved. First, the law needs to recognise the distinction between attacks and endangerments. Second, reform is needed to recognise an altogether separate inchoate offence of endangerment (alongside the offences of attempt, encouragement and assistance, and conspiracy).
II: Duff’s and Yang’s Views Unified
Duff’s Views
Duff was one of the pioneers of the distinction between attacks and endangerments. Attacks are the intentional harming of protected interests. For instance, if I shoot you intending to injure you or start a fire intending to damage your property, I attack your physical integrity or your property.[13] Meanwhile, endangerments are the non-intentional harming of protected interests. For instance, if I shoot in your general direction without intending to injure you or start a fire very near to your property without intending to damage your property, I endanger your physical integrity or your property.[14]
Duff also argued that attempts are failed, unconsummated attacks.[15] Therefore, attempts and endangerments can be distinguished because while the former is intrinsically or essentially harmful, the latter is only potentially harmful.[16] Because intent is the paradigm mens rea for attacks, it is also by extension the paradigm mens rea for attempts. In contrast, the paradigm mens rea for culpable endangerments is recklessness.[17]
Interestingly, Duff is of the view that intended endangerments should count as attacks,[18] but cases of oblique intention, where harm is not directly intended but there is foresight that harm will be caused as a side-effect, should count as an extreme type of endangerment.[19] This is apparently because the former manifests hostility rather than mere indifference.[20] However, it seems to me that the hostility requirement for attacks is problematic as there is an argument to be made that certain culpable endangerments manifest hostility. For instance, hostility is present, if during a road rage incident, an agent swerves their car toward another vehicle aiming to force that vehicle off the road but there was no direct intention to cause injury.
It is hard to justify why some difficult cases should count as attacks while others should count as endangerments when it is difficult to meaningfully distinguish between these cases based on hostility. I will expand more on this in the later section where I attempt to unify Duff’s and Yang’s views.
Yang’s Views
Yang’s work built on the views expressed by Duff. He offered justification for why viewing recklessness as being insufficient mens rea for attempts is consistent with ordinary language, by arguing that attempts, as failed attacks, elicit different types of attitudes from those who have been wronged as compared to endangerments.[21] We demand that the agent of an attack should not have tried to harm some protected interest. In contrast, we demand that the agent of an endangerment should have taken more care, so as not to risk harm to others.[22] Yang uses examples of responses in an attack versus an endangerment on someone’s tree to illustrate his point: a typical response in the former case would be ‘Why are you trying to harm my tree?’ while in the latter case it would be ‘You should be careful! You could have harmed my tree!’.[23]
Yang also argued that the law should follow our ordinary understanding of attempts and endangerments in order to serve its deterrent function. A crucial part of addressing and preventing any type of wrong surely has to do with recognising the wrongful action for what it really is.[24] If we care to differentiate between attacks and endangerments, it is not good enough to subsume an act of endangerment under the law of attempts.[25]
Unlike Duff, Yang took a more hardline stance on whether recklessness sufficed for attempts. He directly challenged the decision in Khan and argued for extending the approach in Pace to attempts in general.[26] He also viewed both intended endangerments[27] and cases of oblique intention[28] as endangerments, not attacks.
Duff and Yang Unified
Having introduced Duff’s and Yang’s views, I will attempt to resolve their differences in favour of Yang’s stance. While they agree that attacks and endangerments should be recognised as distinct concepts, they disagree about the classification of some of the more difficult cases, particularly in the case of intended endangerments.
There are two fundamental issues that need to be addressed in order to engage with this debate meaningfully: 1) at which point on the recklessness-to-intention scale do we decide if someone is culpable for an attempt, and 2) on what basis should we draw that line.
It has been suggested that we should instead draw the line based on hostility. Indeed, Duff has suggested that hostility is intrinsic to attacks,[29] and therefore attempts. However, while hostility is a useful criterion to distinguish between typical cases of attacks and endangerments, it does not fare so well in the more difficult cases of endangerments which come close to direct intention, such as instances of intended endangerments and oblique intention. This is because, as Duff has recognised, there can be normative disagreements about what constitutes hostility.[30] The argument as to whether difficult cases should be considered as attacks or endangerments can genuinely go both ways depending on how hostility is construed. This inserts an element of subjectivity into the exercise which would lead to considerable uncertainty. A test that works well in typical cases but struggles in the more difficult cases is not very useful.
We need a test that brings clarity in difficult cases. I will argue instead that the test should be whether the defendant tried to harm the protected interest or was merely indifferent to it. Only direct intention should be sufficient mens rea for attempts generally.
Figuring out which point on the recklessness-to-intention scale to draw the line for attempts is difficult if we focus solely on concepts relating to culpability (such as hostility). This is because endangerments are not necessarily less culpable than attacks. The attack-endangerment distinction is important not because it marks differences in the extent of culpability, but because it marks different types of moral wrong.[31]
In my view, those who view the more difficult cases such as intended endangerments as attacks are guilty of using culpability to determine whether the mens rea is that of an attack or endangerment. They operate under the mistaken assumption that attacks are necessarily more culpable than endangerments. While that is true most of the time, this view does not always hold, especially in the difficult cases where mens rea is close to, but not actually, direct intention. An agent who is extremely reckless as to the life of another may be just as culpable as someone who attacks another. However, an attack is an attack not because it is particularly culpable, but because it involves direct intention to harm a protected interest.
It seems to me that this misconception has been the source of much of our troubles in the law of attempts. Not wanting agents who endanger culpably to escape sanction, we criminalise at the expense of fair labelling. But I argue that it does not have to be one or the other. Both objectives could be achieved if inchoate endangerments were separately criminalised in the law. The first step in this has to be recognising and upholding the distinction between attacks and endangerments.
Therefore, the demarcation should be on the basis of the different characteristics of each type of wrong. The test for attempts should be whether the defendant tried to harm the protected interest or was merely indifferent to it. This approach lends much-needed simplicity and clarity to the exercise. I will now illustrate its application to the difficult cases of intended endangerment and oblique intention below, which are both extreme forms of endangerment.
Intended Endangerment
Intended endangerment refers to a case where an agent might intend not to cause substantive harm, but to create (or to expose another to) a risk of harm.[32] For instance, forcing a victim to play Russian roulette or swinging a golf club towards somebody’s precious vase to see how close they can get without hitting it are examples of intended endangerment.[33]
Oblique Intention
Oblique intention is where an agent might not intend harm directly but foresees that harm will be caused as a side-effect. A famous example of this is from Glanville Williams: ‘suppose that a villain of the deepest dye blows up an aircraft in flight with a time-bomb, merely for the purpose of collecting on insurance. It is not his aim to cause the people on board to perish, but he knows that success in his scheme will inevitably involve their deaths as a side-effect.’[34]
Applying the Unified Approach
In both these cases of extreme indifference, because we judge the agent’s actions to be at least equivalent to an attack in terms of culpability, the temptation is to label them as attacks. However, as mentioned above, endangerments are not so called because they are less culpable than attacks, but rather because they are a different kind of wrong. They involve indifference rather than intention. Thus, both intended endangerment and oblique intention should not be subsumed under attempts, but should be separately criminalised as endangerments. This does not mean that the agents are let off the hook. The extent of sanction will be commensurate to their culpability under this separate endangerment offence.
A key aspect that my argument turns on is the existence of a separate inchoate endangerment offence. It is to this aspect that I will turn now.
III: The Way Forward
The Need for Reform
The criminal law punishes certain forms of endangerments, but it has tended to focus mostly on consummated offences where the risk has actualised. For example, an agent might be guilty of wounding and of criminal damage if injury results and property is damaged.[35] However, if the endangerment is unconsummated and the risk has not actualised, it is not clear whether there is criminal liability. This is because there is no general offence of unconsummated endangerment (unlike unconsummated attacks in the form of attempts liability) in English law.[36] It seems to me that the time has come for the law to fill the gap and recognise an altogether separate inchoate offence of endangerment.
I will offer two guidelines for reform. They are not meant to be comprehensive, but are merely intended to serve as a starting point for legislators who have made the enlightened decision to uphold the attack-endangerment distinction in inchoate offences.
Guideline 1: Not A Wholly General Offence
The Criminal Attempts Act 1981 is of general application as it applies, subject to limited exceptions, to ‘any offence which, if it were completed, would be triable in England and Wales as an indictable offence’.[37] However, unlike attempts (unconsummated attacks), unconsummated endangerments should not be a wholly general offence. This is because in the case of failed attacks, they are generally serious enough to warrant criminal sanction: if a crime exists, it is generally accepted that it is also an offence to attempt to commit that crime.[38] But the same universal condemnation is not true of failed endangerments. They are one step removed from the harm that is intrinsic to a failed attack.[39]
We must ask ourselves, from the perspective of penal desert, whether the kinds of wrong in unconsummated endangerments are always serious enough to warrant the coercive attention of the criminal law, and the numerous costs involved in criminalisation.[40] Because endangerments are not intrinsically culpable, criminalisation will depend on the type of potential harm and the type of mens rea involved. Therefore, a wholly general offence of failed endangerments is undesirable. Instead, the offence needs to be subject to certain restrictions which focus on the more serious endangerments. For instance, in his draft Code of Conduct, Robinson proposes an offence of ‘act[ing] in a way that creates a substantial and unjustified risk of causing a result made criminal by this Code’.[41] However, I argue that even so, this restriction can be refined further, as it affords judges too wide a discretion.
Guideline 2: Criteria for Criminalising Unconsummated Endangerments
It is difficult to suggest criteria to specify which endangerments are serious and should therefore be criminalised without resorting to general descriptors such as ‘substantial’ and ‘unjustified’ mentioned above, which are, in my view, too broad to be helpful. There is a certain circularity in the process, the perennial tension inherent when proposing broad legislation of this sort: in searching for a coherent general principle, we cannot help but fall back on vague yardsticks which invoke our moral instincts instead of a formula of sorts. This is problematic especially since, as we have seen, there can be normative disagreements about these culpability concepts.
However, we can circumvent this problem by taking advantage of the good sense of centuries of legislation and common law. Some complete offences can be committed either as an attack (with intent) or as an endangerment (with recklessness or negligence).[42] For instance, a defendant can be guilty of criminal damage whether the damage to property is caused intentionally or recklessly.[43] Rape is another example where the offence can be committed either through intention or indifference. This is because, whether by Parliament or judges, it was thought that indifference in these contexts were serious crimes because they endangered important protected interests which merit criminalisation even if they were not intended. If it has been decided that a consummated endangerment is worth criminalising, then its unconsummated form is very likely worth criminalising as well. What matters is that an important protected interest is being endangered, and the agent, in not taking more care, is culpable. By using this as a proxy, we are able to specify which types of protected interests require enhanced protection with greater certainty and clarity. For instance, unconsummated endangerment in the context of property interests and sexual integrity should be criminalised.
Evidently, this is a rough proposition that will need to be stress-tested and refined, but it is, in my view, a promising starting point in view of the alternative criteria suggested.
Justifications for Reform
But why should we insist that endangerments be recognised as such in the law and create an altogether separate offence for it? In the final section, I will provide two justifications.
First, while I have thus far stressed the importance of fair labelling as a key reason why we should not be subsuming endangerments under attempts, I have not yet explained the reason for its importance. We should uphold the attack-endangerment distinction because fair labelling is central to effective deterrence. A crucial part of addressing and preventing any type of wrong surely has to do with recognising the wrongful action for what it really is.[44]
Offence labels matter because they serve a vital communicative function. They tell citizens that a certain kind of action is wrong and should not be done.[45] Yang argues that in the context of sexual integrity, for example, the point of an inchoate endangerment offence such as reckless sexual conduct is not to communicate that it is wrong to act with negligence as to consent in sexual encounters. The complete offence of rape already does that by criminalising non-consensual sex and doing so without a reasonable belief in consent. Rather, the purpose is to highlight the endangerment as a distinct wrong, different in kind from attempted rape.[46] This is because, while most people understand that rape is wrong and roughly why it is wrong, significantly fewer people appreciate the fact that we can cause serious harm to others by not taking sufficient care, or what it means to take sufficient care, especially in the context of sexual encounters.[47] In other words, a separate inchoate endangerment offence contributes to effective deterrence through education. We would be doing more to highlight the wrongfulness of acting with negligence or recklessness by highlighting the indifference that gives rise to such actions.[48] In contrast, to convict an agent who endangers another under attempts liability would distort the wrong committed and mislabel the kind of wrong as an attack.[49]
Second, from the point of view of parliamentary sovereignty and constitutional propriety, it is contrary to the language of the Criminal Attempts Act 1981 for judges to subsume endangerments under attempts. Because the statutory language requires intention to commit the substantive offence,[50] it is wrong in principle to convict endangerments, which involve mens rea of recklessness and negligence, under the category of attempts. If Parliament creates a separate offence of inchoate endangerment, it will achieve much-needed clarity in this area of the law by distinguishing between what are really two distinct wrongs. In turn, judges no longer have to go against statutory wording in order to criminalise endangerments.
Conclusion
In this work, building on the work of Duff and Yang on the importance of the attack-endangerment distinction, I have argued for a clearer test to differentiate the two categories in the first section. The demarcation should not be based on culpability concepts, but should instead be whether the defendant tried to harm the protected interest or was indifferent to it. Accordingly, only direct intention should be a sufficient mens rea for attempts generally.
In the second section, I suggested guidelines to serve as a starting point for reform if we are to embark on the project of creating an inchoate endangerment offence. It should not be the same kind of wholly general offence as criminal attempts, but should be specified by using completed offences as a proxy. I also provided justifications for such reform in accordance with the objective of effective deterrence and constitutional principle.
Looking ahead, the suggested guideline of using completed offences as a proxy to decide which types of unconsummated endangerment should be criminalised needs to be scrutinised and refined. This is because the argument of not subsuming endangerments under attempts rests heavily on the premise that there exists an inchoate endangerment offence which is capable of satisfactory criminalisation of culpable endangerments.
[1] R. A. Duff, Criminal Attempts (1996), at chs. 8.5 & 13.3.
[2] Di Yang, ‘Recklessness and Circumstances in Criminal Attempts.’ Criminal Law and Philosophy, vol. 17, no. 2, 2023, at 370.
[3] R. A. Duff, ‘Criminalizing Endangerment’ (2005) 65 La L Rev, at 944.
[4] Supra note 2, at 377.
[5] Criminal Attempts Act 1981 1(1)
[6] Criminal Damage Act 1971 1(1)
[7] Khan [1990] 1 W.L.R. 813.
[8] Pace [2014] EWCA Crim. 186.
[9] Law Com. No. 102, Attempt, and Impossibility in Relation to Attempt, Conspiracy and Incitement (1980), at para. 2.14.
[10] Ibid, at para. 2.17.
[11] Law Com. No. 177, A Criminal Code for England and Wales (1989), at paras. 13.44-45.
[12] Law Com. No. 318, Conspiracy and Attempts (2009), at paras. 8.133 & 8.137.
[13] Supra note 3, at 941.
[14] Ibid.
[15] Ibid.
[16] Ibid, at 958.
[17] Ibid, at 944.
[18] Ibid, at 950.
[19] Ibid, at 951.
[20] Ibid, at 950.
[21] Supra note 2, at 363.
[22] Ibid, at 369-370.
[23] Ibid, at 371.
[24] Ibid, at 377.
[25] Ibid, at 367.
[26] Ibid, at 362.
[27] Ibid, at 367.
[28] Ibid, at 366.
[29] Supra note 3, at 943.
[30] Ibid.
[31] A.P. Simester, “Why Distinguish Intention from Foresight?” in Harm and Culpability (A.P. Simester & A.T.H. Smith eds., 1996), at 71.
[32] Supra note 3, at 950.
[33] Ibid.
[34] G. Williams, ‘Oblique Intention’ (1987) 46 CLJ 417, at 423
[35] Supra note 3, at 942.
[36] Ibid.
[37] Criminal Attempts Act 1981 1(4)
[38] Stephen, History of the Criminal Law of England (1883) vol. 2, at 221-3.
[39]Supra note 3, at 958.
[40] Ibid.
[41] Paul H. Robinson, Structure and Function in Criminal Law (1997), at 218.
[42] Supra note 3, at 949.
[43] A. P. Simester et. al., Criminal Law: Theory and Doctrine (7th ed. 2019), at 461.
[44] Supra note 2, at 377.
[45] Ibid, at 376.
[46] Ibid, at 376-7.
[47] Ibid, at 377.
[48] Ibid.
[49] Ibid, at 376.
[50] Supra note 5.
Comments