human rights, 'arranged' marriages and nullity law: when do 'force', parental


 

 

 

68 

Denning Law Journal 2017 Vol 29 pp 68- 105 

 

PUBLIC GOODS AND CRIMINALISATION 
 

James Slater* 
 

1. INTRODUCTION 
 

Theories of criminalisation seek to identify the criteria by which 

behaviour is legitimately criminalised. This article believes that their 

success in so doing is best assessed if they examine the question of 

criminalisation in light of four desirable features for any such theory. 

These desirable features, which this article will term desiderata for short, 

are as follows: 

 

 Desideratum 1: a theory of criminalisation should offer an 
evaluative framework that justifies the form of legal regulation 

known as the criminal law. 

 Desideratum 2: a theory of criminalisation’s evaluative framework 
under Desideratum 1 should allow for a coherent and defensible 

account of the criminal law as morally censorious, thereby 

articulating something distinctive about the criminal law as a form 

of legal regulation. 

 Desideratum 3: a theory of criminalisation should display a 
coherent understanding of how its evaluative framework under 

Desideratum 1 integrates with a theoretical account of the 

purpose, and legitimacy, of the state.  

 Desideratum 4: a theory of criminalisation’s evaluative framework 
under Desideratum 1 should distil criminal from non-criminal 

behaviour in a principled and defensible way.  

 

Given that the defence of each desideratum would arguably generate an 

article apiece, the aims of this article are consequently more modest. It is 

aimed at those who already accept one or more of them. It will 

demonstrate the success, in satisfying the desiderata, of a theory of 

criminalisation embedded in the notion of public goods. It shall call this 

theory the public goods account (the ‘PGA’). The PGA is not an entirely 

                                                      
* Senior Lecturer in Law, University of Buckingham. I am very grateful to John 

Gardner, bob Watt, March Stauch, Natalie Pratt and Sarah Sargent for their 

comments on earlier drafts of this article. The usual disclaimer applies.  



THE DENNING LAW JOURNAL 

 

69 

new theory, as elements of it can be found in the writings of a number of 

theorists.1 However, by expanding on, exploring and assessing these 

elements in light of the desiderata, this article offers further support to a 

theory of criminal law embedded in the notion of public goods.  

In order to understand the PGA, it is necessary to begin this article 

with a section outlining the nature of public goods. Subsequent sections 

will then address how the PGA satisfies each desideratum, in the order 

they are set out above.  

 

2. THE NATURE OF PUBLIC GOODS 
 

In order to understand the nature of public goods, it is first necessary 

to understand the nature of a good. A good is a defined instance of the 

exercise of valuable autonomy that manifests itself in a particular moral, 

political, social or economic context.2 Individual goods are those when 

only one person exercises the relevant form of valuable autonomy: 

examples include eating and sleeping. Shared goods are when the exercise 

of valuable autonomy requires two or more people to cooperate: team 

sports and many forms of sexual activity are examples. Finally, an 

important feature of goods is that their autonomy component requires that 

the good be freely chosen.  

Turning attention to the concept of a public good, Joseph Raz has 

provided the following definition: 

 

[A good] that refers not to the sum of the good of individuals but 

to those goods which, in a certain community, serve the interest of 

                                                      
1 See, for example: J Horder, Ashworth’s Principles of Criminal Law (8th edn, 

OUP 2016) Chapter 3 ‘Criminal Law Values’ and ‘Bribery as a Form of Criminal 

Wrongdoing’ (2011) 127 LQR 37; J Gardner and S Shute, ‘The Wrongness of 

Rape’ in Jeremy Horder (ed), Oxford Essays in Jurisprudence: Fourth Series 

(OUP 1998); reprinted with minor changes in Gardner, Offences and Defences: 

Selected Essays in the Philosophy of the Criminal Law (OUP 2007) 1, 31; AP 

Simester and A von Hirsch, Crimes, Harms, and Wrongs: On the Principles of 

Criminalisation (Hart Publishing 2011) 42-43.  
2 As such they may require social forms to exist, see J Raz, The Morality of 

Freedom (Clarendon Press 1986) 309-312. See also Raz, Ethics in the Public 

Domain: Essays in the Morality of Law and Politics (Clarendon Press 1994) 121. 

This is the notion of valuable autonomy being dependent on social forms that 

give that autonomy meaning. For reasons of space, this claim cannot be 

interrogated here. 



PUBLIC GOODS AND CRIMINALISATION 

 

70 

people generally in a conflict-free, non-exclusive, and non-

excludable way.3 

 

As Raz states, public goods possess the following characteristics: non-

competitiveness, that is to say that one person’s enjoyment of the public 

good does not diminish that of others, though different persons will 

benefit to a different degree, and non-excludability, that is to say the 

public good is available to all, and benefits all, without exclusion.4 Thus, 

for example, the systematic provision of clean air is a public good, as the 

benefits of clean air are then enjoyed by all, and one person enjoying the 

fruits of clean air does not diminish its availability to others.5 

Having outlined the nature of, on the one hand, individual and shared 

goods and, on the other, public goods, the connection between them can 

now be articulated: public goods are constituted by the non-exclusive and 

non-competitive availability, in any given society, of a range of individual 

and shared goods.6 In other words, the ability to exercise the valuable 

autonomy of the relevant individual or shared goods is enjoyed by all 

citizens, that is to say non-rivalrously and without discrimination.7  

                                                      
3 Raz, Ethics in the Public Domain: Essays in the Morality of Law and Politics 

ibid 52-55, where Raz sets out his theory of public goods. Raz also uses the terms 

“common” and “general” good. See also 122, where Raz states that “… the 

conditions of autonomy … depend on the common good, that is, on a good which 

if available to one is available to all and whose benefits can be had by all without 

competition or conflict.” See also George Klosko, ‘The Principle of Fairness and 

Political Obligation’ (1987) 7 Ethics 353. 
4 Klosko, ibid 353-354; Klosko notes: “[public goods] cannot be enjoyed by 

particular individuals without being made available to a much wider group of 

people, frequently to all members of the community.” Klosko therefore implicitly 

accepts, correctly in the view of this article, that there may be (a few) legitimate 

exclusions from public goods. These include the exclusion of children from the 

franchise and the legal right to enter contracts. Such exclusions must, 

nevertheless, be carefully justified. Perfectionist liberals, such as Raz, might do 

so for reasons different to those committed to impartial liberalism. 
5 Raz, ‘Rights and Politics’ (1995) 71 Indiana Law Journal 27, 35. 
6 Ibid 37: “an adequate range of goods in society is a common good.” 
7 Public goods may well be politically controversial: see Raz, ‘Rights and 

Politics’ (n 5) 38. Iseult Honohan also points out, correctly, that the language of 

common goods can mask political agendas, which suggests the need for political 

scrutiny: see her Civic Republicanism (Routledge 2002) 157. This article accepts 

that rights-based, and public reason type, arguments may be needed to nuance and 

supplement the provision of public goods; for example, the common good of a 

democratic polity benefits all, but, whilst foreign nationals might be legitimately 

excluded from the franchise, exclusions based on race are illegitimate. In this 

regard, see Horder, Ashworth’s Principles of Criminal Law (n 1) 49.  



THE DENNING LAW JOURNAL 

 

71 

A distinction should be drawn at this stage between, on the one hand, 

those public goods that are concerned with a single good and, on the 

other, those goods that embrace a wider variety of goods, termed framing 

goods by Raz.8 Public goods concerned with a single good universally 

provide a range of individual and shared goods based on that good and 

that good only. Sexual integrity is an example: it is constituted by the 

systematic availability of a sufficient range of the private and shared 

goods involving sex. Another example is a democratic polity, which 

provides the individual and shared goods that make up a citizen’s 

democratic identity, not least the individual good of the right to vote. By 

way of contrast, Raz has defined framing goods as those ‘…goods the 

existence of which is a precondition for the existence of an adequate range 

of other goods in the society.’9 Framing goods include the public peace, 

which provides for the many different kinds of private and shared goods 

that are themselves dependent on physical and psychic autonomy. The 

public purse is also a framing good: it provides general sustenance for 

valuable autonomy, by financing all the institutions of the state that 

contribute to the valuable autonomy of its citizens. Another example of a 

framing good is a clean environment, which provides a general enabling 

environment for a significant range of individual and shared goods. 

Public goods can be positive or negative.10 They are positive when 

their existence involves the creation, often at public expense, of a 

supportive environment for the exercise of valuable autonomy, such as 

green spaces, secondary education and a democratic polity. They are 

negative when they constitute protection from interferences with 

autonomy, such as a culture of tolerance, the public peace and freedom of 

expression.  

The fact that public goods provide for the systematic protection of 

individual and shared goods might lead some to think that public goods 

have no intrinsic value, but are merely instruments in the provision of 

individual and shared goods. There are two reasons why this is not so, and 

it will emerge in this article that these two reasons are central to the 

PGA’s conception of the nature, justification and limits of the criminal 

law.  

First, public goods have intrinsic value because of their characteristics 

as non-excludable and non-rivalrous. This characteristic means that public 

goods instantiate the notion of making the relevant private and shared 

goods available to all. It is therefore in the nature of public goods that all 

                                                      
8 Raz, ‘Rights and Politics’ (n 5). 
9 Ibid.  
10 As pointed out by Horder, ‘Bribery as a Form of Criminal Wrongdoing’ (n 1) 

44-45. 



PUBLIC GOODS AND CRIMINALISATION 

 

72 

citizens should enjoy the various forms of valuable autonomy made 

available by public goods; consequently, they reflect a fundamental 

commitment to the equality of citizens where the exercise of valuable 

autonomy is concerned. Second, public goods have intrinsic value because 

their contribution to valuable autonomy often transcends the sum of 

individual valuable autonomy they provide and protect. The framing 

goods of the property regime and a democratic polity illustrate this 

capacity.  

The property regime defines and protects those private and shared 

goods required to create the regime, including the shared good of contract. 

However, the economic benefits of the property regime, not least its 

wealth creating capacity, and the opportunities those economic benefits 

create, for example through taxation, transcend the sum of individual good 

each participant derives from the private and shared goods protected by 

the regime. And it is a significant reason why we value the public good of 

the property regime that it possesses these overarching benefits. 

In the same vein, a democratic polity protects those individual and 

shared goods required to create democratic government, including the 

individual good of the right to vote. However, a democratic polity also 

provides a range of benefits that transcend the sum of individual interests 

it protects. For example, it ensures the continued responsiveness of 

government to the wishes of the people and maintains, through inter-party 

competition, the quality of political debate. Thus the quality of the 

political environment is also a concern of a democratic polity, a concern 

that, once successfully realised, benefits all citizens.11  

The above two examples reveal that public goods are valuable not 

only because they provide valuable options to all citizens; they are also 

valuable because, by increasing the impact, benefits and consequences of 

the exercise of valuable autonomy, they enhance the value and effect of 

those options. Public goods are greater than the sum of their parts. This 

overarching quality is evidenced by the fact that both the above public 

goods benefit those who are too young to enter contracts or vote, i.e. are 

ineligible, for legitimate reasons, to exercise the valuable autonomy 

provided by the relevant individual and shared goods.12 It also explains 

the inalienability of the right to vote, since the systematic selling of that 

                                                      
11 This is the quality of dual harmony: see Raz (n 46) and accompanying text in 

this article.  
12 The point concerning children benefiting from the property regime is taken 

from Raz: see Ethics in the Public Domain (n 3) 53-54. On the question of 

legitimate exclusions, see (n 4) and (n 7).  



THE DENNING LAW JOURNAL 

 

73 

right would undermine the distinctive benefits provided by the public 

good of democratic government.13 

Having expanded on the nature of public goods, attention can now 

turn to how the PGA satisfies the desiderata.  

 

3. THE PGA AND DESIDERATUM 1 
 

Desideratum 1 (‘D1’) states that a theory of criminalisation should 

offer an evaluative framework that justifies the form of legal regulation 

known as the criminal law. Its reference to an evaluative framework 

incorporates explanatory, justificatory and critical strands. By 

explanatory, it means a theory of criminalisation must set out its criteria 

for the criminalisation of behaviour. By justificatory, it means a theory of 

criminalisation must defend those criteria in light of moral and/or political 

values. In other words, why is the criminal law, as that theory conceives 

it, a good thing? Finally, by critical, a theory of criminalisation should 

suggest reform and improvement to existing systems of criminal law, 

whilst possessing a measure of descriptive accuracy where such systems 

are concerned.14 For reasons of space, and because the critical strand of a 

theory of criminalisation flows from its explanatory and justificatory 

strands, this article will focus on the explanatory and justificatory strands.  

 

3.1 The explanatory strand of D1 
 

The PGA is a theory of criminalisation that shares with certain 

theories of criminalisation the notion that wrongfulness, conceived of in 

some way, is an intentional object of criminalisation. And so, in this 

section, the article will set out the PGA’s criteria of criminal 

wrongfulness. Before proceeding, however, it is necessary to expand 

somewhat on the familiar distinction between crimes that are mala in se 

and those that are mala prohibita. Admittedly, the nature of this 

distinction, or even the fact of making it, is controversial. The reason for 

                                                      
13 This point concerning the inalienability of the right to vote and the public good 

of a democratic polity is taken from Raz, ‘Rights and Politics’ (n 5) 34.  
14 The need for theories of criminalisation to have both descriptive and normative 

dimensions has been summarised by R Dagger: see his ‘Republicanism and the 

Foundations of Criminal Law’ in RA Duff and SP Green (eds), Philosophical 

Foundations of Criminal Law (OUP 2011) 44, 45, where he states that such 

theories “… must account for the leading features of the criminal law and point 

the way to its reform or further development.” It should be noted that there is a 

risk of a significant gap between, on the one hand, theories of criminalisation and 

their account of the rules and principles of the criminal law and, on the other, the 

implementation of those rules and principles in actual practice.   



PUBLIC GOODS AND CRIMINALISATION 

 

74 

offering an explanation will become clear: the distinction, as this article 

conceives it, serves as a necessary backdrop to the PGA’s account of 

criminalisation.15    

Moral wrongs are determinations that certain behaviour should not be 

performed in light of the moral reasons for and against that behaviour in 

any given context. On occasion the wrongs that emerge from such 

determinations will admit of relatively little disagreement as to their form: 

it is the crimes that seek to reflect such wrongs that are mala in se. It is for 

this reason that, when they accurately reflect their source morality, mala 

in se crimes such as murder and rape take much the same form across 

jurisdictions. It is the precision of the source morality where such crimes 

are concerned that accounts for their conventional definition as wrongful 

independent of law: in such cases, the criminal law is, to a significant 

degree, simply a formally posited definition of what morality already 

articulates. 

By way of contrast, mala prohibita crimes are concerned with those 

occasions when the morality from which the criminal wrong is derived is 

considerably more ambiguous. The reason for such ambiguity lies in the 

highly abstract nature of certain moral objectives, of which perhaps the 

most pervasive example is the objective to make the world a safer place. 

How abstract moral objectives of this nature are particularised in the 

context of complex projects of social and economic coordination admits 

of enormous variety in matters of detail. For this reason, a number of 

different definitions of any derivative criminal wrong will do justice to the 

source morality. Yet the principle of maximum certainty in the criminal 

law requires that a line be drawn somewhere: where exactly may be 

somewhat arbitrary or governed by local contingencies. Because the line 

between what is and is not criminal therefore depends largely on what the 

law says, as opposed to being governed by fidelity to an underlying 

morality, what is prohibited remains ‘unknown’ until a definition is 

supplied by the criminal law.16  It is this fact that accounts for the 

conventional definition of mala prohibita crimes as wrong only because 

                                                      
15 For an excellent exploration of the nature of this distinction, see the trilogy of 

papers (special topic) published in Dialogue: Canadian Philosophical Review 

(2016) 55 (1), as follows: S Dimock, ‘A Trilogy of Papers on the Malum 

prohibitum-Malum in se Distinction in Criminal Law: Introduction’ 1; S Dimock, 

‘The Malum prohibitum-Malum in se Distinction and the Wrongfulness 

Constraint on Criminalization’ 9; SP Green, ‘The Conceptual Utility of Malum 

prohibitum’ 33; C Flanders, ‘Public Wrongs and Public Reason’ 45.  See also 

Simester and von Hirsch, Crimes, Harms, and Wrongs: On the Principles of 

Criminalisation (n 1) 24-29. 
16 Or the criminal law in conjunction with the civil law. 



THE DENNING LAW JOURNAL 

 

75 

the law says so. Such crimes are therefore prohibita to the extent their 

final formulation could legitimately have taken different forms. However, 

they are mala to the extent that they do justice to their underlying moral 

objective. So no crime can be mala by dint of prohibition alone.17  

It is important to note that this article conceives of the difference 

between mala in se and mala prohibita as one of degree, rather than one 

of kind.18 This is because even those criminal wrongs conventionally seen 

as reflecting mala in se admit of some measure of reasonable discretion in 

the exact form they take. This may be due to reasonable differences in 

conceptions of the underlying morality or it can flow, as will be argued 

later in the article, from how the criminal law puts its legitimate political 

goals into effect.19 For example, whether only certain, or alternatively all, 

frauds in the inducement are constitutive of the wrong of rape, is a 

controversy within morality.20 When the criminal law settles on which 

frauds in the inducement to include and exclude, and for what reasons 

(moral, political or a mixture of both),21 it is making a choice that 

accounts for a measure of prohibita where the crime of rape is concerned. 

What prohibita therefore means in the context of criminalisation is 

discretion as to form when, in light of the relevant source morality, a 

crime can take more than one legitimate form.22 

                                                      
17 As Simester and von Hirsch point out, the state cannot make something wrong 

simply by declaring it so: see Simester and von Hirsch, Crimes, Harms, and 

Wrongs: On the Principles of Criminalisation (n 1) 24-27. 
18 This insight is taken from Green (n 15). 
19 Duff makes a similar point: see his Punishment, Communication and 

Community (OUP 2001) 64-67. 
20 For a conception of rape that conceives of the wrong as a violation of self-

possession, as opposed to sexual autonomy, and the implications of this 

conception for the criminalisation of frauds in the inducement, see J Rubenfeld, 

‘The Riddle of Rape-by-Deception and the Myth of Sexual Autonomy’ (2013) 

122 Yale Law Journal 1372. 
21 See the difference in opinion between Jonathan Herring and Hyman Gross 

played out in the following articles: Herring, ‘Mistaken Sex’ [2005] Crim LR 

511, Gross, ‘Rape, Moralism and Human Rights’ [2007] Crim LR 220 and 

Herring, ‘Human Rights and Rape: A Reply to Hyman Gross’ (2007) Crim LR 

228. 
22 It should be noted that, despite the fact that such lines can be reasonably drawn 

in different places where such crimes are concerned, their source morality means 

some places where those lines are drawn are, as Douglas Husak points out, 

indefensible: see his Overcriminalization: The Limits of the Criminal Law (OUP 

2008) 110. See also Dimock ‘Contractarian Criminal Law Theory and Mala 

Prohibita Offences’ in Duff, L Farmer, SE Marshall, M Renzo and V Tadros 

(eds), Criminalisation: The Political Morality of the Criminal Law (OUP 2014) 



PUBLIC GOODS AND CRIMINALISATION 

 

76 

In light of the distinction between mala in se and mala prohibita 

crimes as outlined above, the explanation of how the PGA identifies 

criminal wrongs can begin with those crimes that most saliently embody a 

concern with public goods, those that, as Duff expresses it, “wrong or 

harm the polity as a whole, rather than a distinct identifiable individual. 

These include such serious crimes against collective, shared goods as 

treason and attempting to pervert the course of justice; tax evasion [and] 

‘public nuisance’”.23 These crimes, which shall be termed public crimes 

for short, can be found at various points along the spectrum between mala 

in se and mala prohibita. According to the PGA, they are united by the 

fact that the criminal prohibition seeks to maintain one or more public 

goods, for example the environment, the public purse or health and safety.  

For those public crimes closer to the mala prohibita end of the 

spectrum, wrongfulness is governed predominantly by the fact that the 

individual and/or systematic commission of the crime reduces the 

availability of the particular form(s) of valuable autonomy provided, 

directly or indirectly, by the public good(s) concerned. Such crimes are 

mala in se to the extent that they constitute reasonable and carefully 

considered attempts to target behaviour that directly or indirectly, in 

isolation or cumulatively, sets back the provision of the valuable 

autonomy. But they are largely prohibita because the manner in which 

autonomy is protected in particular contexts, for example food, drug or 

road safety, admits of much discretion. This category accounts for those 

crimes of a regulatory nature, often strict and met with milder punishment, 

that are designed, in a fairly instrumental way, to maintain autonomy in 

given contexts, such as health and safety and the transport system.  

However, public goods are not exclusively concerned with the 

provision of autonomy per se: rather, they articulate and protect 

environments where the exercise of autonomy has moral value and 

significance. It is for this reason that they are characterised by a number of 

moral principles. For example, all public goods associated with the public 

good of the rule of law, such as the public purse and the systems of civil 

and criminal justice, are characterised by the notions of fairness, 

objectivity, honesty and integrity. In turn, for those public crimes closer to 

the mala in se end of the spectrum, the wrongfulness of the behaviour 

flows from the fact it violates the moral principles that characterise the 

                                                                                                                         
151, 175: ‘… not all ways of providing determinate content to mala in se wrongs 

are acceptable.’ 
23 Duff, Answering for Crime: Responsibility and Liability in the Criminal Law 

(Hart Publishing 2007) 140. Duff is expressing this notion within the context of 

what it means for a wrong to be public, for the sake of his ‘public wrong’ account 

of criminalisation: see (n 52).   



THE DENNING LAW JOURNAL 

 

77 

public good and make the existence of the public good valuable. 

Alternatively, such mala in se crimes violate the moral principles that 

assist in the provision of the valuable autonomy characteristic of the 

public good.  

Let us take, for example, public crimes that target tax evasion. Such 

crimes are designed to sustain the public purse. Of course, these are mala 

prohibita to the extent that their reach is dependent on the contingencies 

of how the targets, and levels, of taxation are decided within any given 

jurisdiction. But the public purse is more than simply the methodological 

notion of money gathered through taxes: it is characterised by a number of 

moral principles. These include fairness in distribution and the presence of 

honesty and integrity in those public officials tasked with gathering and 

spending the sums raised through taxation; the latter requires that tax is 

gathered for, and spent on, the business of government, not syphoned 

away for personal advantage. Offences that are designed to maintain the 

public purse will therefore be mala in se to the extent that they constitute 

violations of these moral principles. For example, when the burden of 

taxation is justly distributed, tax evasion is rightly characterised as 

wrongful, as a form of cheating and free riding. In the same vein, some 

instances of the electoral offence of personation, that is voting as another 

person, violate the democratic principle of ‘one-person, one vote’, which 

is itself derived from the principle of political equality.24 Thus the 

wrongfulness of (some forms of) personation is constituted by the 

violation of important values that constitute the public good of a 

democratic polity.  

However, as will be defended in greater detail in the next subsection, 

what completes the case for the criminalisation of these more mala in se 

public crimes, as well as their more mala prohibita cousins, is the 

damaging impact their systematic commission would have on the various 

forms of valuable autonomy provided by the relevant public good. This 

impact can take different forms. Where positive public goods are 

concerned, it may take the form of a reduction in the means by which the 

state creates a supportive environment for valuable autonomy. For 

example, though a single instance of tax evasion may have no practical 

effect on the public purse in its capacity as a framing good, systematic tax 

evasion does. This is because the consequent loss of funds to the state will 

reduce its capacity to provide numerous public goods, such as the public 

                                                      
24 For a defence of this argument, see J Slater and b Watt, ‘In Defence of 

Democracy: The Criminalization of Impersonation’ (2015) 14(2) Election Law 

Journal: Rules, Politics, and Policy 165.  



PUBLIC GOODS AND CRIMINALISATION 

 

78 

peace and secondary education.25 Alternatively, systematic commission of 

the offence may undermine the overarching benefits of a public good, 

thereby eroding the valuable nature of the autonomy it provides. To 

illustrate, though a single instance of personation may not alter an election 

result, systematic personation will eventually erode the provision of the 

valuable autonomy characteristic of a democratic polity. It will do so 

principally by undermining the overarching benefits of a democratic 

polity, such as the confidence of the electorate in the electoral system and 

the responsiveness of political parties, and government, to the wishes of 

the people.26   

On the other hand, the relationship between public goods and mala in 

se crimes such as murder, theft and rape, which are wrongs against 

individuals as opposed to the public, seems less obvious. The PGA’s 

concern with public goods suggests the individual does not feature in the 

PGA’s account of criminalisation, and, consequently, that it would 

struggle to give a satisfactory account of such crimes. The nature of this 

(potential) flaw is outlined by Duff:  

 

If we are going to say that such actions as murder, rape and theft 

should be criminal because they injure or threaten some common 

good, are we not then ignoring, and thus denigrating, the wrong 

done to the individual victims of such actions-a wrong which 

surely should be central to the law’s concerns?27 

 

                                                      
25 Green points out that it is in the nature of tax evasion that it is ’significant only 

in aggregate’: see his ‘What is Wrong with Tax Evasion?’ (2009) Houston 

Business and Tax Law Journal 220, 226. 
26 See Slater and Watt, ‘In Defence of Democracy: The Criminalisation of 

Impersonation’ (n 24). 
27 Marshall and Duff, ‘Criminalization and Sharing Wrongs’ (1998) 11 Canadian 

Journal of Law and Jurisprudence 7, 12. Similarly in Answering for Crime (n 23) 

141, Duff states: “If we … argue that such mala in se as murder and rape count as 

public wrongs only because they too have a harmful or wrongful impact on ‘the 

public’, as well as on their individual victims, we are likely to distort the 

wrongfulness that makes them criminalisable. Even if a rapist takes unfair 

advantage over the law-abiding (which is at best arguable), or creates ‘social 

volatility’ or undermines trust, that is not what is central to the criminal 

wrongfulness of his action; what he is properly convicted and punished for is the 

wrong done to his victim.” This point is a common theme in Duff’s work: see, for 

example, Punishment Communication and Responsibility (n 19) 60 and Public 

and Private Wrongs (again with Marshall) in J Chalmers, F Leverick and Farmer 

(eds) Essays in Criminal Law in Honour of Sir Gerald Gordon (Vol 5 Edinburgh 

Studies in Law, Edinburgh University Press 2010) 71.  



THE DENNING LAW JOURNAL 

 

79 

As Duff explains further, the problem with accounts that exhibit an 

exclusive concern with the public, or common, good, is that they …: 

 

… seem to subordinate the individual victim (a concern for their 

good, or for the wrong done to them) to some supposedly larger 

social good. The offender's conduct is counted as criminal, and he 

is to be punished, for the sake of that larger good: to which it is 

appropriate to object that his conduct should be criminalised 

because of, and he should be punished for, the wrong he does to 

the individual victim. We do not criminalise rape, and punish 

rapists, because rape causes social volatility; or because the rapist 

takes an unfair advantage over his law-abiding fellow citizens: but 

because of the nature of the wrong that the rapist does to his 

victims.28 

 

Duff is making two closely related claims here. First, he is suggesting that 

theories of criminalisation that justify the criminalisation of wrongs 

against individuals purely in terms of some benefit to the public, for 

example because they suppress social volatility, fall into error by ignoring 

the wrong done to the individual.29 Second, he is suggesting a theory’s 

account of such crimes should be reacting, in some fairly fine-grained 

way, to the moral mala committed against the victim. 

The position of this article is that the first claim is self-evidently true: 

some conception of what happens to the individual must play its part in 

the nature and justification of such crimes. In this section, the article will 

therefore demonstrate how the PGA does not suffer from this flaw. It will 

also address the second claim, by demonstrating how the PGA is 

sufficiently fine grained to respond appropriately to the moral mala 

experienced by the individual. In the next section, which addresses 

Desideratum 2, it will then defend the claim that this is what the criminal 

law should be doing, to the extent that a distinctive, that is to say 

                                                      
28 Marshall and Duff, ‘Criminalization and Sharing Wrongs’ ibid. 
29 Duff’s own ‘public wrong’ account of criminalisation does not fall foul of this 

error. This is because he fuses concern with the moral wrong done to the 

individual with the quality of publicness, by arguing that the moral values that 

define the behaviour as wrong where the individual is concerned are, 

simultaneously, part of the fabric of shared values that define the political 

community: see his ‘Criminalization and Sharing Wrongs’ (n 27) 20, where Duff 

states: “ … wrongs against individual citizens can be understood as shared 

wrongs, as wrongs against the whole community, insofar as the individual goods 

which are attacked are goods in terms of which the community identifies and 

understands itself.” 



PUBLIC GOODS AND CRIMINALISATION 

 

80 

censorious, account of the criminal law is not possible without so 

responding.  

With regard to the first claim, the reason why the maintenance of 

public goods does not ignore the individual in the name of a distinct social 

goal is because creating and/or maintaining a public good frequently 

involves directly protecting the capacity of individual citizens to enjoy the 

valuable autonomy of various individual and shared goods. The public 

goods of the public peace and the property regime illustrate this.  

The public peace is a public good because it assists all members of 

society in partaking in a wide range of individual and shared goods, by 

freeing them of physical interference and violence, and one member of 

society benefiting from the public peace does not diminish its availability 

to other members.30 However, it can only be provided by systematically 

protecting individual members of society from physical and psychic 

assault; in other words, it is through the protection of the individual goods 

of bodily and psychic integrity, in the form of fatal and non-fatal offences 

against the person, that the public good is created. The core of the 

criminal wrong therefore consists of the invasion of the valuable 

autonomy of the individual. 

The property regime is a public good because it provides, for all 

citizens, “… opportunities for personal and social advancement through 

reliable coordinated economic activity, and for other forms of welfare and 

personal realisation that only the peaceful ownership and possession of 

property can deliver.”31 The foundation of peaceful ownership and 

possession of property, along with its resulting benefits, is individual 

dominium over assets, with the nature of that individual dominium, and 

how it can be shared and exchanged, largely defined by the civil law. In 

turn, for the public good of the property regime to materialise, that system 

of individual dominium must also be protected. This is why the crime of 

theft, along with other property offences, target all those who seek to 

ignore or violate the rules of individual ownership, possession and 

transfer.  

The above two examples reveal that, where certain public goods are 

concerned, there is a symbiotic relationship between, on the one hand, the 

systematic protection of the valuable autonomy of individuals and, on the 

other, the maintenance of the relevant public good. This symbiosis means 

that, where wrongs against individuals are concerned, the PGA conceives 

of the criminal law as possessing what Blackstone termed a ‘double 

view’: an interest in both protecting the individual’s valuable autonomy 

                                                      
30 Horder, ‘Bribery as a Form of Criminal Wrongdoing’ (n 1) 44. 
31 See Simester and von Hirsch Crimes, Harms, and Wrongs: On the Principles 

of Criminalisation (n 1) 42. 



THE DENNING LAW JOURNAL 

 

81 

and protecting the public.32 It is this ‘double view’ that means the PGA 

does not ignore the individual in the name of a distinct public good, 

because concern with the valuable autonomy of the individual is the 

fundamental building block of the public good. The PGA then simply 

demands that that protection be offered systematically to all, such that a 

distinct, non-rivalrous and non-excludable good is created.  

As for the nature of the wrong done to the individual, does the PGA 

have the capacity to articulate the moral mala experienced by the 

individual? The answer is yes and flows, as with public crimes, from the 

fact that public goods are characterised by a number of moral principles 

that mean the autonomy they foster and protect is valuable in nature.  

For example, the public peace is more than the existence of freedom 

from physical and psychic attack. It concerned with the human dignity of 

each citizen and the public’s conception of their security and social 

environment. This concern with human dignity means it is interested in 

marking the difference between, say, murder, a deliberate attack, and 

gross negligent manslaughter, a morally culpable failure to meet a 

standard of conduct. It also accounts, it is suggested, for offences against 

the person defined in part by racial or religious animus. These various 

wrongs threaten the public peace in different ways: proclaiming them as 

crimes, and prosecuting in their name, gives the public peace its value, its 

moral character.33  

This moral articulacy is also illustrated by sexual offences. There is a 

public good in the form of sexual integrity, a general sphere of valuable 

autonomy made up of the universal availability of various forms of human 

flourishing constituted by private and shared goods involving sex.34 Craig 

has listed these various forms of flourishing as follows:  

                                                      
32 Commentaries on the Laws of England (Clarendon Press 1769) Book IV, Ch 1, 

7: “Upon the whole we may observe that in taking cognizance of all wrongs, or 

unlawful acts, the law has a double view: viz. not only to redress the party injured 

… but also to secure to the public the benefit of society, by preventing or 

punishing every breach and violation of those laws, which the sovereign power 

has thought proper to establish for the government and tranquillity of the whole.” 

See also Book IV, Ch 1, 5: “ … every public offence is also a private wrong, and 

somewhat more; it affects the individual, and it likewise affects the community.” 
33 Where racial or religious animus is concerned, see, for example, the Crime and 

Disorder Act 1998, sections 29 and 32. Fair labelling is therefore not just a 

concern for the accused: because of the interest of all citizens in the relevant 

public good, it is a concern for all citizens.  
34 The article will not address whether and, if so, in what ways, a good is 

dependent on existing social forms and, additionally, grounded in independent 

criteria of value: the aim here is more limited, which is simply to demonstrate 

how the notion of public goods has moral granularity.  



PUBLIC GOODS AND CRIMINALISATION 

 

82 

 

… not only … freedom from violations of sexual integrity but also 

promoting and protecting the ‘conditions for’ sexual fulfilment, 

sexual diversity, sexual literacy, the safety necessary for sexual 

exploration, the freedom to say ‘yes,’ and the capacity to gain 

sexual benefit (whether that benefit be physical, emotional, 

financial or social).35  

 

These forms of human flourishing are clearly dependent on autonomy per 

se. But the various private and shared goods of sex listed above are also 

morally and socially meaningful, in other words valuable, and sexual 

offences target that value in many different ways. For example, if 

consensual sex is joyful or recreational or an expression of deep love and 

commitment, or all three at the same time, the horror and exploitative 

nature of rape negates these qualities in a particular way. This is worthy of 

discrete articulation by a conception of the criminal law that seeks to 

foster the public good of sexual integrity.   

Finally, the PGA has significant explanatory and evaluative power 

where the concept of criminal recklessness is concerned. Though 

recklessness has a role as a determinant of culpability, the requirement 

that the risk taken is unjustified means it is also concerned with the 

boundary between criminal and non-criminal behaviour. Whether a risk is 

unjustified is a complex socio-moral, that is to say normative, judgment. It 

requires balancing the social utility of taking the risk against the gravity of 

the harm it might cause, as well as its likelihood to cause that harm, with a 

view to establishing whether the risk is worth taking. In other words, 

whether an activity is reckless depends on how that conclusion integrates 

with the promotion of valuable autonomy generally, the net gain in 

valuable autonomy from the toleration of risk. Because the raison d’être 

of the PGA is the promotion valuable autonomy, rather than autonomy per 

se, it provides a normative context that engages directly with how those 

risks are assessed and tolerated.  

 

3.2 The justificatory strand of D1 
 

The justificatory strand of D1 is concerned with the question why the 

decision to criminalise behaviour is a good thing. The PGA sees this 

justification in the criminal law’s impact on practical reason, with the 

result that citizens desist or are deterred from committing criminal 

wrongs, with a view to creating and/or maintaining public goods. This 

                                                      
35 E Craig, Troubling Sex: Towards a Legal Theory of Sexual Integrity (UBC 

Press 2012) 136. 



THE DENNING LAW JOURNAL 

 

83 

justification of the criminal law is derived from Raz’s conception of the 

purpose, and legitimacy, of the state. This purpose is to create, through 

law and other measures, the social conditions under which citizens can 

lead autonomous lives, through the provision of a sufficient range of 

individual and shared goods. The PGA’s theory of criminalisation is 

therefore embedded in, rather than independent of, a political theory of the 

state.36 The reasons for criminalisation are fundamentally political in 

nature. The implications of this approach for the definitions of criminal 

wrongs and the rules of attribution will be examined in Section 5, when 

this article explores the implications of the PGA’s political approach to 

the criminal law. This section will address why such an approach justifies 

the criminal law, as the PGA sees it.   

The support for autonomy within this conception of the state’s role is 

perfectionist in nature: this means that personal autonomy is worthy of 

state support only to the extent that it is exercised in favour of valuable 

options.37 Raz explains:  

 

…the autonomy principle is a perfectionist principle. Autonomous 

life is valuable only if it is spent in the pursuit of acceptable and 

valuable projects and relationships. The autonomy principle 

permits and even requires governments to create morally valuable 

opportunities …38 

                                                      
36 For a discussion of the whether the starting points of a theory of criminalisation 

should be political or moral, see Duff et al, Criminalization: The Political 

Morality of the Criminal Law (n 22) 17-26. 
37 Raz, The Morality of Freedom (n 2) Ch 15, ‘Freedom and Autonomy’. For a 

critical analysis of Raz’s perfectionist liberalism, see P Neal, ‘Perfectionism with 

a Liberal Face? Nervous Liberals and Raz's Political Theory’ (1994) 20 Social 

Theory and Practice 25. 
38 Raz, The Morality of Freedom (n 2) 417; see also 133, where Raz states: “… it 

is the goal of all political action to enable individuals to pursue valid conceptions 

of the good …” Raz also believes that the state is under a duty to discourage or 

eliminate empty or evil options, as these quotes go on to state. According to Raz, 

there are no barriers, in principle, preventing the state from targeting worthless 

options through coercion because, according to Raz, there is no value in 

autonomy per se (418). However, he acknowledges both pragmatic and 

contingent objections. The pragmatic objections consist of, first, the possibility of 

error, by government, as to what is valuable (427) and, second, the fact that the 

pursuit of too many perfectionist policies may lead to civil strife (429). The 

contingent objection is that coercion will have a collateral effect on the exercise 

of valuable autonomy (418-419). Contra Raz, this article believes that the 

coercive effect of the criminal law means it should be confined to promoting, and 

protecting, valuable autonomy, as matter of principle. Autonomy per se has 

sufficient value such that the criminal law cannot be used to suppress valueless 



PUBLIC GOODS AND CRIMINALISATION 

 

84 

 

Additionally, and equally importantly, the state’s support of autonomy 

requires it to provide a wide range of individual and shared goods: it is 

only in the face of a sufficient range of valuable options that the pursuit of 

any given valuable option is freely chosen, that is to say autonomous.39 As 

explained above, the provision of public goods is fundamental to the state 

fulfilling this role because, as Raz states: “Public goods lie at the 

foundation of most options.”40 And so, when the state provides a wide 

range of public goods, this ensures that all citizens have an adequate range 

of individual and shared goods to choose from, which also ensures that the 

goods are chosen freely. Public goods are therefore fundamental to the 

ability of citizens to lead autonomous lives.  

According to the PGA, the criminal law is necessary for the creation 

and maintenance of public goods. This is because, echoing the concept of 

public goods in economic theory, the creation of public goods cannot be 

left exclusively to the ‘market’ of enforcement by aggrieved individuals 

pursuing private law remedies. Such enforcement would not occur 

systematically enough to create the system of deterrence necessary to 

bring public goods into existence. Market failure will result for a number 

of reasons: one is that victims often lack the resources required to mount a 

prosecution; another is that many crimes do not produce individual 

victims in any event. Accordingly, public goods need the systematic 

support from the state, in its capacity as police authority and prosecutor, to 

materialise. Of course, according such a purpose to the criminal law is 

subject to an efficacy condition, that the criminal law will, indeed, be 

effective as a deterrent. 

However, just because autonomous lives cannot exist without the 

criminal law does not address a more fundamental question, which is 

whether the state is justified in using the form of regulation known as the 

criminal law, as the PGA conceives it, to achieve this objective? The 

                                                                                                                         
autonomy, though the state is under no duty to provide it, or protect it, through 

the criminal law, or other measures.  
39 Raz, The Morality of Freedom (n 2) 204, 408, 410, 417-18, 425. See also his 

Ethics in The Public Domain (n 3) 121, where he states: “One is autonomous only 

if one lives in an environment rich with possibilities.” 
40 Raz, Ethics in the Public Domain (n 3) 121. See also The Morality of Freedom 

(n 2) 207: “The provision of many collective goods is constitutive of the very 

possibility of autonomy …”. 



THE DENNING LAW JOURNAL 

 

85 

answer is yes, because such a theory engages directly with, first, the state 

administered and, second, the coercive nature of the criminal law.41  

With regard to the fact that the criminal law is a state created and 

administered practice, the criminal law is a resource intensive enterprise.42 

The use of public resources requires a justification that speaks to the 

polity as a whole: the vindication of moral values, though a reason to 

criminalise, cannot justify, by itself, the expense and logistical efforts 

required to put the criminal law into effect. The PGA offers such a reason: 

the creation of valuable autonomy for all, via the means of public goods.   

As for the question of coercion, the criminal law’s transgression 

generates the risk of prosecution and punishment, including the possibility 

of incarceration.43 These risks create prudential, content independent 

reasons to comply that, on occasion, override the preferences of the 

individual concerned, and so are coercive in nature. For some, the bloody-

minded and recalcitrant, such prudential reasons may be the only reasons 

for which they comply.44 Autonomy has sufficient value, even when 

exercised in the name of valueless options, that coercion requires 

justification. Furthermore, the impact on valuable autonomy of a 

conviction, for example its impact on reputation and employability, means 

the criminal law must, overall, have a positive effect on people’s lives in 

order to be justified.45 According to the PGA, it is the maintenance of 

public goods, and the consequential beneficial impact such maintenance 

has on the valuable autonomy of all citizens, so that they may lead 

                                                      
41 Simester and von Hirsch make similar point: see Crimes, Harms, and Wrongs 

(n 1) 118, repeated in ‘On the Legitimate Objectives of Criminalisation’ (2016) 

Criminal Law and Philosophy 367, 376.  
42 As noted by Matravers, ‘Duff on Hard Treatment’ in R Cruft, MH Kramer and 

MR Reiff, Crime, Punishment and Responsibility: The Jurisprudence of Antony 

Duff (OUP 2011) 68, 81.  
43 It should be noted that just because a theorist acknowledges that the criminal 

law is coercive does not mean that achieving that coercive effect need feature in 

the theorist’s justification for the criminal law. For example, Michael Moore 

acknowledges that the criminal law is coercive but does not see preventing 

wrongdoing by coercion as justifying its existence. Rather, he argues its raison 

d’être is the imposition of punishment on those who deserve it: culpable moral 

wrongdoers; see Moore, ‘Liberty’s Constraints on What Should be Made 

Criminal’ in Duff et al (eds), Criminalisation: The Political Morality of the 

Criminal Law (n 22) 182, 184. 
44 The expression “bloody minded and recalcitrant” is taken from Scott Anderson, 

‘Coercion’ in the Stanford Encyclopedia of Philosophy 

<https://plato.stanford.edu/entries/coercion/> accessed July 2017. 
45 As noted by Simester and von Hirsch in Crimes, Harms, and Wrongs (n 1) 118 

and ‘On the Legitimate Objectives of Criminalisation’ (n 41) 376. 



PUBLIC GOODS AND CRIMINALISATION 

 

86 

autonomous lives, that supplies a justification that engages directly with 

these facts.  

It is also worth remembering the overarching benefits of public goods 

noted under Section 2, because such benefits are a key reason why public 

goods are valuable and thereby worthy of protection. In particular, these 

overarching benefits give rise to what Raz has termed the quality of dual 

harmony, whereby protecting the valuable autonomy of the individual 

creates the common good, and, in return, the common good, brings 

additional benefits to the individual.46 Raz explains dual harmony as 

follows: 

 

…to the extent that the rightholder's interest is given extra weight 

for reasons of the common good, these reasons are not altogether 

detachable from considerations of the rightholder's own interest. 

The common good is the good of all, including the good of the 

rightholder. By serving the common good, the right also serves the 

interest of the rightholder in that common good. There is here 

what I have called elsewhere a dual harmony between the interest 

of the rightholder and the interest of other people which is served 

by his right. The right protects the common good by protecting his 

interest, and it protects his interest by protecting the common 

good.47 

 

An example of dual harmony was touched upon in Section 2, when the 

overarching benefits of the public good of the property regime were 

described. The systematic protection of the individual and shared goods of 

ownership and contract creates the public good of the property regime, the 

existence of which, not least through its capacity for wealth creation, 

benefits the individual in return. It is this dual harmony of public goods 

that is also an integral part of justifying the expensive, autonomy 

impacting form of coercion known as the criminal law.  

 

4. THE PGA AND DESIDERATUM 2 
 

Desideratum 2 (‘D2’) means D1’s evaluative framework includes a 

commitment to a coherent and defensible account of the criminal law as 

morally censorious.48 At the heart of D2 is the notion that the criminal law 

                                                      
46 For an explanation of dual harmony, see Raz’s Ethics in the Public Domain (n 

3) 53-55. 
47 Raz, ‘Rights and Politics’ (n 5) 39. 
48 Of course, if a theorist rejects a distinct account of the criminal law, D2 is an 

irrelevancy.  



THE DENNING LAW JOURNAL 

 

87 

condemns the behaviour it criminalises: a theory’s success in satisfying 

this desideratum will therefore depend on the coherence and defensibility 

of the condemnation it articulates. It also means the theory must not only 

condemn the wrong but also the wrongdoer: it should therefore have some 

conception of the moral culpability of those convicted of criminal wrongs. 

This means it should give an account of attribution. Attribution is 

concerned not with wrongdoing and justification, but rather with whether 

and how people are held to account for their (all things considered) 

wrongdoing.49 Amongst other things, it addresses the nature and reach of 

excusatory and exempting defences such as loss of control and insanity.50 

A conception of the criminal law as condemning both wrong and 

wrongdoer offers a distinctive account of such law in two ways.51 First, it 

seeks to distinguish the criminal law from those forms of legal regulation 

that are not censorious in nature, for example those associated with 

corrective justice, such as tort, or distributive justice, such as taxation. 

Second, and more crucially for the purposes of this article, it may be 

contrasted with non-distinctive accounts of the criminal law itself, which 

lack the element of moral censure.  

The difference between distinctive accounts and non-distinctive ones 

is one of degree. At one end of the spectrum are purely distinctive 

accounts, which propose no other goal for the criminal law than the moral 

                                                      
49 John Gardner describes these elements of the criminal law as addressing 

“whether and how we should count what people have done when we are judging 

them.” Gardner, ‘Criminal Law and the Uses of Theory: A Reply to Laing’ 

(1994) 14 OJLS 217, 220. See also Duff, ‘Harms and Wrongs’ (2001) 5 Buffalo 

Criminal Law Review 13, 19. 
50 Thus, a theory of criminalisation embraces all those elements that govern the 

possibility, and nature, of conviction. This article agrees with Simester and von 

Hirsch that the grounds of criminalisation need not perfectly match those of 

punishment: see Crimes, Harms, and Wrongs: On the Principles of 

Criminalisation (n 1) 8. See also JR Edwards and Simester, ‘Prevention with a 

Moral Voice’ in A du Bois Pedain and U Neuman (eds) Liberal Criminal Law 

Theory: Essays for Andreas von Hirsch (Hart Publishing 2016) 43, 47: “There is 

no reason to think that the legitimate aims of criminalisation and punishment 

must be identical.” 
51 For a summary of the nature of a distinctive account of the criminal law, see M 

Matravers, ‘Political Neutrality and Punishment’ (2013) 7 Criminal Law and 

Philosophy 217, 219-223. See also A Cornford ‘Rethinking the Wrongness 

Constraint on Criminalisation’ (2017) Law and Philosophy 1, 2-3 and P Pettit, 

‘Criminalization in Republican Theory’ in Duff, Farmer, Marshall, Renzo and 

Tadros (eds), Criminalization: The Political Morality of the Criminal Law (n 22) 

132, 135.  



PUBLIC GOODS AND CRIMINALISATION 

 

88 

condemnation of those who commit crimes.52 Such theories reject any 

deterrent and preventative role for the criminal law.53 At the other end of 

the spectrum, purely non-distinctive accounts reject concern with moral 

condemnation. Purely non-distinctive accounts are characterised by two 

aspirations: first, to conceive of the criminal wrong in as morally 

parsimonious fashion as possible, for example that behaviour is criminally 

wrongful simply because it is harmful or autonomy reducing; second, to 

see the criminal law is nothing more than a regulatory tool designed to 

reduce offending through the provision of prudential reasons to comply, in 

the form of fear of conviction and punishment.54 And some accounts find 

                                                      
52 The legal moralism of Moore and Duff are examples of purely distinctive 

accounts. For Moore, the purpose of the criminal law is to identify those who 

should receive ‘deserved’ punishment for moral wrongdoing: see his Placing 

Blame: A General Theory of the Criminal Law (OUP 1997) and ‘Liberty’s 

Constraints’ (n 43). For Duff, the criminal law should be exclusively concerned 

with the communication of censure, both to the public at large and the criminal 

convicted, when public, as opposed to private, wrongs have been committed: 

“[W]e should not see the criminal law as prohibiting the conduct that it defines as 

mala in se—as offering the citizens content-independent reasons to refrain from 

such conduct. We should see it instead as declaring such conduct to constitute a 

public wrong properly condemned by the community, for which the agent is 

answerable to the community through a criminal process.” See his Punishment, 

Communication and Community (OUP 2001) 64. This approach is a central pillar 

of Duff’s conception of the criminal law: see, e.g., Answering for Crime (n 27) 

84-93 and ‘Responsibility, Citizenship, and Criminal Law’ in Philosophical 

Foundations of Criminal Law (n 14) 125, 129. For an excellent summary of 

Duff’s philosophy, see M Thorburn, ‘Calling Antony Duff to Account’ (2015) 

Criminal Law and Philosophy 737. 
53 Duff’s rejection of deterrence flows from his commitment to the notion that the 

criminal law should display respect for persons, with the result that the provision 

of content independent reasons (the fear of sanction and punishment) is not a 

legitimate feature of the criminal law. Matravers has stated that Duff “… recoils 

at the slightest hint of deterrence”: see ‘Duff on Hard Treatment’ in The 

Jurisprudence of Antony Duff (n 42) 81. For an analysis of Duff’s approach, see 

Matravers, Justice and Punishment: The Rationale of Coercion (OUP 2000) Ch 9, 

‘The Moral Community, Justified Coercion, and Punishment’. See also Thorburn, 

ibid 746.  
54 A purely non-distinctive account is that of Barbara Wootton: see her Crime and 

the Criminal Law: Reflections of a Magistrate and Social Scientist (2nd revised 

edn, Steven & Sons 1981). For a critique, see HLA Hart ‘Review: Crime and the 

Criminal Law’ (1965) 74 Yale Law Journal 1325. See also Matravers and Arina 

Cocoru ‘Revisiting the Hart/Wootton Debate on Responsibility’ in Christopher 

Pullman (ed), Hart on Responsibility (Palgrave Macmillan 2014). 



THE DENNING LAW JOURNAL 

 

89 

themselves somewhere in between these two extremes, mixing distinctive 

and non-distinctive elements.55  

According to this article, the aspiration for a pure account is 

misplaced. Contra purely distinctive accounts, the notion that the criminal 

law has no preventative role, does not seek a forward-looking goal of 

reducing offending by impacting on the practical reason of potential 

offenders, ignores the fact that, as we saw in the previous section, the 

coercive power of the state must do more than merely seek the vindication 

of moral values: it must impact on the lives of citizens in a positive way. 

On the other hand, purely non-distinctive accounts fail to engage with the 

fact that the criminal law is concerned with the preservation of valuable 

autonomy, and therefore should speak a moral language of some kind; 

that, amongst the various functions of law, a morally-loaded regulatory 

tool has its place.56 In demonstrating how the PGA satisfies D2, this 

section will therefore show how it combines distinctive and non-

distinctive elements.  

In order to offer a defensible account of D2, the distinctive element of 

a theory of criminalization must have two components: first, it must 

possess a morally fine-grained approach to criminal wrongs; second, it 

must articulate a condemnation of those wrongs. In order to understand 

why a defensible account of D2 requires the first component, it is 

necessary to examine, and more importantly reject, a theory of 

criminalisation that eschews a fine-grained approach to criminal wrongs, 

the public law account of Malcolm Thorburn (the ‘PLA’).57  

The PLA’s justification of criminalisation embodies a constitutional 

conception of the state’s role with respect to the use of coercive and 

punitive force. Such an approach begins with a constitutional conception 

of the state, and then finds a role for the criminal law within that 

conception.58 It sees the state’s fundamental responsibility as securing 

                                                      
55 As Simester and von Hirsch argue, where criminalisation is concerned … “… 

[t]he truth is, we think, somewhere in between.” (n 1) 4. See also T Hornle, 

‘Theories of Criminalization’ (2016) Criminal Law and Philosophy 301, 302.    
56 The expression ‘a morally-loaded regulatory tool’ is taken from Simester and 

von Hirsch Crimes, Harms, and Wrongs: On the Principles of Criminalisation (n 

1) 11. 
57 Set out in various works, including ‘Punishment and Public Authority’ in P 

Asp, A Dubois-Pedain and M Ulvang (eds), Criminal Law and the Authority of 

the State (Bloomsbury 2017) 1; ‘Constitutionalism and the Limits of the Criminal 

Law’ in  Duff, Farmer, Marshall, Renzo, and  Tadros (eds), The Structures of 

Criminal Law (OUP 2011) 85; and ‘Criminal Law as Public Law’ in Duff and S 

Green (eds), Philosophical Foundations of Criminal Law (n 14) 21. 
58 ‘Constitutionalism and the Limits of the Criminal Law’ ibid 87-88. His 

argument is noted by Ashworth and L Zedner in ‘Punishment Paradigms and the 



PUBLIC GOODS AND CRIMINALISATION 

 

90 

each citizen’s equal freedom or autonomy, a responsibility that grounds its 

legitimacy.59 Thorburn explains: 

 

Unlike any private actor, the state claims to speak in the name of 

everyone’s claim of freedom equally. For this reason, the state is 

the unique instrumentality through which we may collectively 

ensure our freedom as independence. It speaks for us all together 

in setting down general laws that define the scope of everyone’s 

freedom in the same way, but it does not speak for anyone in 

particular—and so, in that way, the state’s actions are not to be 

confused with the partisan choices of some particular individuals. 

… In short, we act together with others through the instrumentality 

of the state in order to secure for all of us the conditions of 

freedom as independence.60 

 

In turn, the criminal law addresses violations of such individual freedom 

by others, that is to say it is concerned with identifying those who 

deliberately impose their preferences on others, thereby suppressing their 

freedom: 

 

The ground of the liberal constitutional state’s legitimacy is the 

simple fact that it—and it alone—can provide the conditions of 

freedom for all. On this account, the role of the criminal law is to 

identify when individuals are attempting to supplant the law’s 

rules with their own preferred arrangements and to regulate the 

use of state power to resist such attempts.61 

 

As such, the criminal law is conceived as maintaining, through its 

processes of condemnation and punishment, the equal freedom of all 

citizens. It is the characteristic of being a violation of the constitutional 

commitment to equal freedom that justifies the criminalisation of 

behaviour. Again, Thorburn explains: 

 

… the criminal law’s concern is with someone’s efforts to 

undermine the whole system of equal freedom itself. … Criminal 

wrongs are those that demonstrate a willingness on the part of the 

                                                                                                                         
Role of the Preventative State’ in Simester, du Bois Pedain and Neuman (eds) 

Liberal Criminal Law Theory: Essays for Andreas von Hirsch (n 50) 3, 7-8.  
59 Thorburn also refers to this equal freedom as ‘jurisdiction’: ‘Criminal Law as 

Public Law’ (n 57) especially 31. 
60 Thorburn, ‘Constitutionalism and the Limits of the Criminal Law’ (n 57) 98. 
61 ‘Constitutionalism and the Limits of the Criminal Law’ (n 57) 88. 



THE DENNING LAW JOURNAL 

 

91 

offender to displace the legal rules themselves—they are 

concerned not merely with an injury to some specific rights claim, 

but to the very idea of living together under law rather than subject 

to the wishes of specific individuals.62 

 

The above quotes reveal that the PLA, like the PGA, offers a theory of 

the criminal law embedded in, and thus justified by, a political theory of 

the state.63 As a justification for the use of the state’s coercive power, 

there is no denying that the PLA’s political approach has much to offer 

the impartial liberal. This is because its conception of the criminal wrong 

as the violation of equal freedom is embedded in what those of a Rawlsian 

inclination would consider a political value, rather than a comprehensive 

doctrine. This is then married to a Rawlsian justification of the criminal 

law’s (coercive) power, the concrete realisation of that political value. At 

the same time, Thorburn claims to offer a distinctive account of the 

criminal law, and thereby respect D2: Thorburn states: “Criminal justice is 

not just a policy instrument for sharing the costs of bringing about a social 

good; rather it is an instrument for identifying wrongdoers and censuring 

them as such.”64  

Such an approach is distinctive to the extent that the failure to treat 

your fellow citizen as free and equal can attract a moral condemnation of 

a certain kind. As pointed out by Matravers: “To violate the demands of 

freedom and equality is, for the liberal, to do a substantive moral as well 

as political wrong.”65  However, the violation of equality is too generic a 

conception of the criminal wrong to do justice to the various mala in se 

that populate the criminal calendar; in order to understand why, it is 

helpful again to quote Thorburn: 

 

… what makes all […] conduct wrongful for the purposes of the 

criminal law is that the offender has intentionally undermined the 

possibility of interacting with others as free choosers who are 

entitled to live under the terms of interaction set out by the law. 

He has done so by treating that person as a mere object who may 

                                                      
62 ‘Constitutionalism and the Limits of the Criminal Law" (n 57) 100.  
63 Thorburn, ‘Criminal Law as Public Law’ (n 57) 24:  “… I propose a different 

way of justifying the practices of the criminal justice system—a position I call a 

‘public law account’ of criminal justice. I call it a ‘public law’ account because it 

conceives of the operations of the criminal justice system, insofar as they are 

legitimate, as concerned with the basic question of public law: when the use of 

state power is legitimate.” 
64 Thorburn, ‘Constitutionalism and the Limits of the Criminal Law’ (n 57) 97. 
65 Matravers, ‘Political Neutrality and Punishment’ (n 51) 221.  



PUBLIC GOODS AND CRIMINALISATION 

 

92 

be dealt with in whatever way he wishes. The wrong of rape—and 

of murder, assault, etc—is precisely the objectification of one 

person by another, but that objectification is of concern to the state 

because it is the state’s job to ensure the survival of the system 

that makes it possible for us all to interact on terms that preserve 

the status of us all as free and equal moral agents.66 

 

It is the fact that Thorburn equates the wrongs of murder, assault and rape 

that is precisely the weakness of his theory. Though these criminal wrongs 

share the moral failure articulated by Thorburn (and Matravers), their 

moral ‘wrongness’ is far richer and more varied than the mere violation of 

equality: murder entails the complete extinction of another person, rape 

involves a demeaning and arguably horrific violation of sexual integrity,67 

assault is a violation of bodily integrity. Other criminal wrongs implicate 

further (in some cases central) human interests, for instance burglary is an 

invasion of the legally constructed but also morally meaningful interests 

in property and privacy. Because the PLA is blind to these differences, it 

has a flat and hence distorted sense of wrong done to the victim where 

these crimes are concerned. It is now appropriate to recall Duff’s 

injunction above, that where crimes against individuals are concerned, “… 

what is central to the criminal wrongfulness of his action … is the wrong 

done to his victim.”68 The PLA, with its exclusive concern with equality, 

is insufficiently fine-grained to acknowledge that wrongfulness properly.  

It is worth noting that legal moralism, in all its forms, has no difficulty 

satisfying D2. This is because the fundamental rationale of all forms of 

legal moralism is the notion that a core, though not necessarily exclusive, 

aim of the criminal law should be the condemnation of criminal wrongs in 

their capacity as moral wrongs. The differences between the different 

forms of moralism concern which parts of morality they see as relevant to 

the criminalisation decision and the principles that limit the reach of their 

moralism.69 But what unites them is highly nuanced and developed 

concern with the moral nature of criminal wrongs, especially against 

individuals, and commitment to the notion that a deeper understanding of 

those moral wrongs represents a deeper understanding of how, and why, 

the criminal law condemns them. Thorburn might respond that this 

                                                      
66 Thorburn, ‘Constitutionalism and the Limits of the Criminal Law’ (n 57) 102. 
67 Stanton-Ife, ‘Horrific Crime’ in Duff (eds), The Boundaries of the Criminal 

Law (OUP 2010) 138.  
68 Answering for Crime (n 23).  
69 Where those limits are concerned, for Gardner, it is harm; for Duff, it is the 

quality of publicness; finally, for Moore, it is certain restraining principles: see 

their works referred in this article.  



THE DENNING LAW JOURNAL 

 

93 

concern with moral wrongs is a weakness of legal moralism, not a 

strength. My concern here is not whether he is wrong or right, but simply 

that anyone who takes D2 seriously cannot support the PLA, given its flat 

and monochrome conception of criminal wrongs.  

The PGA, however, shares legal moralism’s concern with the fine-

grained moral wrong against the individual. This articulacy flows from the 

fact that, as demonstrated under D1, all public goods have moral 

character, a moral character derived from their concern with the 

preservation of valuable autonomy. This concern enables a fine-grained 

approach to such male in se wrongs as murder and rape. It was also 

demonstrated under D1 how this concern with valuable autonomy 

articulates the moral wrongs that occur in the context of public crimes and 

explains, in a nuanced fashion, the criminal law’s concern with unjustified 

risk where its conception of recklessness is concerned. As such, the PGA, 

unlike the PLA, is sufficiently fine grained to articulate the moral nature 

of the various wrongs that populate the criminal calendar. 

However, does the notion of maintaining a public good commit the 

PGA to a theory of the criminal law as a pure deterrent, offering only 

content independent (prudential) reasons, in the form of the fear of 

conviction and punishment, to observe its prohibitions? If so, despite often 

targeting moral wrongs as explained under D1, the PGA will still fail D2, 

as there will be no element of condemnation, simply the brute fact of 

coercion. This would mean the PGA would not articulate a condemnation 

of the moral wrongs it criminalises. 

The PGA does not adopt such a reductive coercive conception of the 

criminal law, but instead integrates its concern with promoting public 

goods with the condemnation of the wrongs it criminalises. It does this by 

offering a dual ‘moral-practical’ account of the criminal law, an account 

that enables it to combine the distinctive and non-distinctive elements 

that, it was argued above, all theories of criminalization should possess.70 

The article will now explain how. 

Where condemnation is concerned, the PGA conceives of the criminal 

law as an official articulation of the moral reasons against the criminal 

                                                      
70 Hence Simester and von Hirsch’s phrase to describe the criminal law: a 

‘morally-loaded regulatory tool’ (n 56). As such it offers a hybrid (or mixed) 

account of the justification of criminalisation. For summaries of hybrid accounts, 

see Ashworth and Zedner, ‘Punishment Paradigms and the Role of the 

Preventative State’ (n 58) 3, 5, Thorburn, ‘Punishment and Public Authority’ (n 

57) 15 and A Spena, ‘Harmless Rapes: A False Problem for the Harm Principle’ 

(2010) Diritto & Questione Publicche 497, 506-507. For a defence of a mixed 

theory of criminalisation, as opposed to punishment, see Edwards and Simester, 

‘Prevention with a Moral Voice’ (n 50). 



PUBLIC GOODS AND CRIMINALISATION 

 

94 

behaviour: consequently, conviction condemns those who were not 

persuaded by those moral reasons. This approach also accounts for the 

rules and principles of attribution, as such condemnation makes little 

sense without a concern with the questions of moral responsibility and 

culpability. 

According to the PGA, however, the criminal law is also meant to 

impact on practical reason, with a view to preventing criminal behaviour. 

According to the PGA, it does this, first, by officially highlighting, and 

hence communicating, the moral wrongfulness, the mala, of the wrongs it 

criminalises. However, this moral appeal is reinforced by a prudential 

disincentive, in the form of the fear of conviction and sanction, for those 

who may not hear, or hear as loudly as they should, the criminal law’s 

concern with the underlying first order reasons. The liberal view that the 

law should not concern itself with the moral character of citizens is 

therefore accorded some weight within this conception: if you comply 

with the requirements of the criminal law, the criminal law has no interest 

in why you have done so. The criminal law should be happy for us to 

comply for any reason, including the prudential reasons it supplies, 

because that is all that is required for creation and maintenance of public 

goods.  

However, the criminal law does more than highlight, and supplement 

with prudential reasons, the moral reasons to not commit the wrongs it 

criminalises. It also offers moral and prudential reasons to exclude the 

reasons in favour of criminal wrongdoing.71 This is because, according to 

the PGA, excluding those reasons is necessary to ensure the behavioural 

consistency necessary to sustain public goods.72  This reason generates a 

moral obligation to pre-empt, namely that so doing provides the 

advantages made available to all by the creation and maintenance of 

                                                      
71 As such, it acts as an exclusionary reason, following Raz’s conception of 

authority: see The Morality of Freedom (n 2) and Ethics in the Public Domain (n 

3). See also ‘The Problem of Authority: Revisiting the Service Conception’ 

(2006) 90 Minnesota Law Review 1003, 1022, where Raz states that “... 

exclusionary reasons do not, of course, exclude relying on reasons for behaving in 

the same way as the directive requires … [but] … must … override our 

inclination to follow reasons on the losing side of the argument. Hence the 

preemption excludes only reasons that conflict with the authority’s directive.” 

This must be correct: the criminal law should be entirely happy for us to refrain 

from acting upon the moral reasons against, say, murder or rape. But, contra Duff 

(see n 52), it should also be satisfied if we observe its prohibitions for prudential 

reasons alone.  
72 This exclusionary power is subject to a limited number of exceptions, when the 

criminal law deems that acting on certain contrary reasons justifies committing 

the criminal wrong. The principles governing self-defence are an example.  



THE DENNING LAW JOURNAL 

 

95 

public goods, not least the highly valuable overarching benefits of public 

goods. But this is, once again, supplemented by a prudential incentive to 

exclude, the fear of conviction and sanction.  

To summarise, the PGA offers a dual ‘moral-practical’ account of the 

criminal law. For those who already accord the proper moral weight to the 

criminal law’s underlying moral concerns, its existence is, in fact if not in 

design, superfluous, though it serves as a vital indication that the state 

shares concerns with certain values and goals. But the criminal law is 

bilingual: to those who do not feel the rational pull of the underlying 

reasons, or does not feel that pull as strongly as they ought to, it offers 

content-independent prudential reasons to comply with its directives and 

exclude countervailing concerns.  

However, a key challenge raised by Duff is the notion that deterrence, 

the supply of prudential reasons in the form of the fear of prosecution, 

conviction and punishment, bypasses the moral agency of the person 

coerced. As a result, its use does not treat the person with the respect he 

deserves, but rather as an entity to be manipulated by the fear of 

conviction and punishment. Duff believes this problem applies even when 

prudential reasons are combined with, or limited by concern with, the 

moral status of the agent.73 There are two elements to this criticism. First, 

that the supply of prudential reasons disrespects the moral agency of the 

individual coerced; second, that it treats the agent as a means to certain 

ends, those of individual and general deterrence.  

These two elements constitute powerful objections to prudential 

reasons and reasons of space prevent this article from doing them justice. 

The second is the most powerful, and this article will have something to 

say about it in the next section. The first is addressed by the fact that the 

PGA does not conceive of the criminal law as an exercise in pure 

manipulation, but as a dual moral-practical form of persuasion. Moral 

agents acknowledge the role of prudential reason in practical reasoning 

and organisation, because of moral fallibility.74 But importantly too, those 

prudential reasons, and the exclusionary effect of criminal prohibitions, 

are offered in the name of the public goods: as such, desistance is 

demanded not without appeal to a moral reason of a kind and not without 

articulating a benefit to the person coerced. The prudential disincentive is 

therefore offered in the name of maintaining the valuable autonomy of all 

citizens, and so engages meaningfully with the moral agency of persons. 

                                                      
73 See Matravers, Justice and Punishment (n 53) 264. It is for this reason that 

Duff suggests conceiving of the criminal law completely differently, as an 

exercise in the communication of censure: (n 52).  
74 Matravers makes this very point: see The Jurisprudence of Antony Duff (n 42) 

82. 



PUBLIC GOODS AND CRIMINALISATION 

 

96 

To express the idea another way, citizens engage with prudential reasons 

as part of an overall scheme of which they are the direct beneficiaries.  

 

5. THE PGA AND DESIDERATUM 3 
 

Desideratum 3 (‘D3’) acknowledges that criminalisation involves the 

exercise of power by the state and, therefore, that any theory of 

criminalisation should demonstrate a coherent understanding of how its 

conception of the criminal law coheres with a theoretical account of the 

legitimacy of that power.75 Some theories of criminalisation believe this 

relationship begins with, and is governed by, the political. This means the 

criminal law is justified in entirely political terms: consequently, the 

approach of such theories to D1 emerges from their conception of the 

state’s nature and role. The PLA of Thorburn, described in the previous 

section, is an example. By way of contrast, the retributivist approach of 

Michael Moore begins life independently of political concerns, by 

offering an account of behaviour worthy of retributive punishment. It then 

only requires that the task of articulating that behavior, and punishing for 

its commission, is legitimately entrusted to the state, whatever form the 

state takes.76 As explained in the previous section, the PGA, like the PLA, 

adopts the former approach, with the result that its understanding of D1 

emerges from, and is governed by, a political concern with the promotion 

of public goods. This section will explore in greater detail the implications 

of this approach.  

                                                      
75 Duff, Farmer, Marshall, Renzo and Tadros (eds), ‘Introduction: Towards a 

Theory of Criminalization’ in Criminalization: The Political Morality of the 

Criminal Law (n 22) 1, 5: “A theory of criminalization must … include or depend 

on a political theory of state and society: it must be a theory of the role that 

criminal law should play within a particular kind of polity.” See also Duff, 

‘Criminal Law Theories’ in the Stanford Encyclopedia of Philosophy 

<https://plato.stanford.edu/entries/criminal-law/> accessed July 2017: 

“Philosophical theories of criminal law, whether analytical or normative, cannot 

subsist in isolation. For one thing, they cannot be wholly separate from other 

branches of philosophy. They must draw, most obviously, on political 

philosophy, since they must depend on some conception of the proper aims of the 

state and of the proper relationship between a state and its citizens.” 
76 Placing Blame (n 52). It is worth noting that Moore’s legal moralism includes a 

number of principles that limit its concern with moral wrongs, most notably the 

presumption in favour of “the standing case for liberty” and epistemic modesty on 

behalf of legislators: see generally ‘Liberty’s Constraints’ (n 43). See also 

Placing Blame (n 52) 75-80 and ‘A Tale of Two Theories’ (2009) 28 Criminal 

Justice Ethics 27, 32-33. 



THE DENNING LAW JOURNAL 

 

97 

Property offences and the public good of the property regime, as 

explained by Raz, illustrate the nature of this fundamentally political 

approach of the PGA: 

 

Every person has … an interest [in the protection of property 

rights by the criminal law] inasmuch as (1) every person may 

become a property owner; and (2) every person benefits from the 

fact that property rights are secure. These benefits take many 

forms. They are not easy to specify exhaustively. They come close 

to being the interest that all people have in living in a civil society. 

My right in my property is based on my interest in having that 

property. But the weight given to my interest, the degree of 

protection it deserves, and the form that protection should take is 

morally determined by considerations which transcend concern for 

my interest in itself. They reflect the interest of other people in the 

common good of respect for property.77 

 

Raz here is referring to a notion that lies at the heart of the PGA and its 

approach to D3. This is the notion that the intrinsic value of public goods 

outlined at the outset of this article, the fact that they exist to benefit all, 

and create benefits that transcend the sum of individual and shared goods 

they embrace, should inform and limit the definition of criminal wrongs. 

The full implications of this observation can be drawn out by returning to 

the distinction between mala in se and mala prohibita crimes outlined in 

Section 3.  

It was argued that the difference between these two types of crime is 

one of degree rather than one of kind. What this means is that, even with 

those crimes traditionally viewed as mala in se, there is an element of 

prohibita in the criminal law’s definition of the wrong. According to the 

PGA, this is largely a consequence of the criminal law’s political role as 

provider of public goods, and its concern with the overarching benefits of 

such goods. The explanatory and evaluative power of this approach can be 

illustrated with the following examples: crimes targeting driving with an 

excessive blood-alcohol concentration; the debate surrounding deceptions 

as to HIV-positive status within the crime of rape; finally, the reach of the 

defence of duress. These will be addressed in order. 

The public good of a safe road traffic system, in its capacity as a 

framing good, plays a key role in providing many private and shared 

goods. Behaviour that poses an unacceptable risk to the safety of this 

system, such as excessive speed, driving without due care and attention 

and driving whilst intoxicated, is therefore a candidate for criminalisation. 

                                                      
77 Raz, ‘Rights and Politics’ (n 5) 33. 



PUBLIC GOODS AND CRIMINALISATION 

 

98 

However, where road safety is concerned, the exact contours of any 

offences created, for example the speed limit on any given road, the 

standard of care for driving, and the legal level of blood-alcohol 

concentration, are not, and cannot be, articulated purely in terms of free 

standing morality, for example that it is wrong to put the lives of others at 

risk. The contours of these offences must also depend upon the overall 

objectives and overarching benefits of road use, as well as the costs of 

reducing risk and the collateral effect of criminalisation on other public 

goods. It is these factors that enable the relevant crimes to take their final 

form, and constitute the element of prohibita where such crimes are 

concerned. 

Where legal levels of blood-alcohol concentration are concerned, this 

is illustrated by Anthony Bottom’s analysis of Sir Peter North’s proposal 

to lower the blood alcohol limit for the crime of driving with a blood-

alcohol concentration above 80mg/100ml.78 As Bottoms points out, this 

proposal was rejected because it was decided that doing so would have 

very little impact on casualties, divert police resources away from the 

more serious cases, and even have a negative effect on commerce. This 

conclusion was reached despite a high degree of consensus that driving 

with any alcohol in the system is mala in se, that is to say morally wrong. 

It is the concern with public goods that explains why, nevertheless, the 

level was not lowered. 

As pointed out in Section 3, whether the wrong of rape embraces all, 

or only some, frauds in the inducement leading to sexual intercourse is a 

controversy within morality.79 Nevertheless, in order for the public good 

of sexual integrity to materialise, the criminal law must settle on which 

fraudulent inducements to include in its definition of the criminal wrong.  

According to the PGA, such a decision should be informed, in part, by the 

imperatives involved in maintaining the public goods affected by the 

decision to criminalise this or that fraud in the inducement. And so, 

whether misleading a sexual partner about HIV positive status, a fraud in 

the inducement, should negate consent within the context of sexual 

intercourse cannot be settled exclusively by moral concerns surrounding 

deceit; it should also engage with those moral and practical concerns 

raised by the maintenance of the public goods of public health and non-

discrimination.80  

                                                      
78 A Bottoms, ‘Civil Peace and Criminalization’ in Duff et al, Criminalisation: 

The Political Morality of the Criminal Law (n 22) 232, 261-264. 
79 See (n 20) and (n 21) and accompanying text. 
80 Matthew Weait has written extensively on this question: see, for example, 

Intimacy and Responsibility: The Criminalisation of HIV Transmission 

(Routledge Cavendish 2007). Gillick v West Norfolk and Wisbech Health 



THE DENNING LAW JOURNAL 

 

99 

The above has hopefully demonstrated how a concern with public 

goods has direct implications for the definition and reach of criminal 

wrongdoing. However, concern with the intrinsic value of public goods 

also has implications for the criminal law’s rules and principles of 

attribution. The relationship between the defence of duress in English law 

and the public peace can be used to illustrate this.  

The maintenance of the public peace demands a measure of consistent 

protection of physical and psychic autonomy, through both the deterrent 

effect of a police presence and the threat, and fact, of prosecution. The 

maintenance of this general climate of security for the benefit of all, and 

the overarching benefits it supplies, it is suggested, explain why the limits 

of the defence of duress should not be decided purely in terms of the 

moral culpability of the accused.  

To illustrate, in the UK, the defence of duress is denied to those who 

knowingly or negligently expose themselves to threats of violence.81 If 

duress is conceived as an excuse, denying the defence to defendants on 

such grounds is defensible, as, arguably, they bear a measure of blame for 

subjecting themselves to the risk of the threat of violence.82 But the 

restriction also flows, at least in significant part, from a concern with 

maintaining the public peace. Lord Simon of Glaisdale acknowledged this 

interaction between culpability and maintaining the public peace where 

duress is concerned in DPP for Northern Ireland v Lynch:83 

 

A sane system of criminal justice does not permit a subject to set 

up a countervailing system of sanctions or by terrorism to confer 

criminal immunity on his gang. A humane system of criminal 

justice does not exact retribution from those who infringe the 

substantive provisions of its code under stresses greater than 

ordinary human nature can bear, nor attempt, by making an 

example of them, to deter those who in the nature of things are 

beyond deterrent. A sane and humane system of criminal justice is 

sufficiently flexible to reconcile such considerations, and to allow 

for all their infinite degrees of interaction. I have ventured to 

                                                                                                                         
Authority [1986] AC 112 is arguably an example of where concerns with the 

public good of public health helped generate a decision not to criminalise 

behaviour. In that case, a decision not to criminalise, under certain circumstances, 

the provision, by doctors, of contraceptive advice to those under 16.  
81 Hasan [2005] UKHL 22. 
82 See Lord Bingham, ibid [38]. 
83 [1975] AC 653. 



PUBLIC GOODS AND CRIMINALISATION 

 

100 

suggest that our own system of criminal justice is capable of such 

sanity and humanity…84 

 

In the same vein, the Law Commission, in an exploration of the extension 

of the defence to murder, noted concerns expressed by the English 

judiciary that “… the members of a criminal gang might be capable, not 

only individually, but in collusion, of concocting a false defence of 

duress.”85  

Duff’s concern, mentioned under D2, about using persons as a means 

to an end is most pertinent here. The fact that the reach of the defence of 

duress is not exclusively concerned with the moral culpability of the 

accused, but rather the goal of maintaining the public peace, means that, 

on occasion, a conviction involves, to a certain degree, sacrificing the 

individual to that goal.  As a result, he is (partly) treated as a means to an 

end. Perhaps the only response to this is to accept that this is an inevitable 

feature of the criminal law, and that all that can be done is, as Lord Simon 

suggests, to balance the concern with the public peace with the concern 

with culpability. Perhaps some of this negative impact of the criminal 

law’s concern with public goods can be addressed in the context of 

sentencing. However, any theory that attempts to purify the criminal law 

of this concern with public goods is likely to result in serious descriptive 

failures, undermine preventative goals and be overly idealistic, but the 

defence of these claims must be left to another occasion.  

 

6. THE PGA AND DESIDERATUM 4 
 

The above discussion reveals that PGA articulates the reach of the 

criminal law in ways that flow from its concern with the maintenance of 

public goods. In this final section, the article will explore Desideratum 4 

(‘D4’), that is, whether that concern distils criminal from non-criminal 

behaviour correctly. There are two facets to this question. First, D4 

requires the articulation of principled limits to the reach of the criminal 

law. Second, respecting D4 means a theory of criminalisation should not 

exclude from criminalisation behaviour that ought to be included. This 

section will explore these two facets through an examination of two 

examples that test the plausibility of theories of criminalisation where they 

are concerned. It will begin with the question of principled limits. 

The PGA has two features that ensure principled limits to the criminal 

law: first, the fact that criminalisation must promote valuable autonomy; 

                                                      
84 Ibid 696. 
85 Law Commission, Legislating the Criminal Code: Offences against the Person 

and General Principles (Law Com No 218 Cm2370, 1993) [33.2]. 



THE DENNING LAW JOURNAL 

 

101 

second, the characteristic of public goods as non-excludable and non-

rivalrous, meaning the autonomy they provide must be available to all 

citizens equally. 86 This can be illustrated by contrasting offences against 

the person with the notion of criminalising homosexual sex. 

The ban on interpersonal violence provided by the various offences 

against the person in the criminal calendar enables each citizen to exercise 

the valuable autonomy that flows from freedom from physical and psychic 

violence. This is achieved by criminalising, and hence systematically 

prohibiting, behaviour that prevents or hinders the exercise of that 

valuable autonomy; in other words, providing protection to all from 

violence helps create the non-excludable, and non-competitive good, 

known as the public peace. By way of contrast, banning homosexual sex 

does not augment the autonomy of citizens in any way. This is because 

such a ban does not target behaviour by any given citizen that impinges on 

the exercise of sexual integrity by another. Citizens do have an interest in 

their sexual autonomy and the protection of their sexual choices and 

inclinations, whatever they may be, and so behaviour that infringes on that 

sexual autonomy, such as sexual assault and rape, are legitimately 

criminalised according to the PGA. By way of contrast, prohibitions that 

seek to prevent citizens from exercising that valuable autonomy are 

illegitimate according to the PGA.  According to the PGA, criminal 

coercion is used only to promote valuable autonomy, not suppress 

valuable autonomy, in order to create a non-excludable good.  

Some might suggest that homosexual sex is not a form of valuable 

autonomy, on grounds of immorality.  However, even if this premise were 

true (a position rejected by this article), the PGA’s exclusive interest in 

promoting valuable autonomy, as opposed to suppressing valueless 

autonomy, would mean that any supposed immorality of such behaviour is 

irrelevant to its criminalisation according to the PGA. The PGA does not 

permit the criminalisation of valueless autonomy merely by dint of it 

being valueless: that behaviour must negate the valuable autonomy of 

others, directly or indirectly, thereby undermining a public good. 

Homosexual sex does not do that. 

Attention can now turn to the second facet, which is whether the PGA 

excludes from criminalisation behaviour that ought to be criminalised. Its 

plausibility here can be interrogated by examining its approach to the 

criminalisation of non-consensual medical treatment that heals and, 

therefore, arguably does no harm to that patient. This is the case of 

wrongdoing that is beneficial to the victim.  

                                                      
86 This does allow for some exclusions of a non-arbitrary kind, such as children 

from the shared good of contract or the individual good of voting, see text of (n 

4). 



PUBLIC GOODS AND CRIMINALISATION 

 

102 

Imagine a doctor who ignores the refusal of a patient for a blood 

transfusion, saving the patient’s life. Given the non-consensual invasion 

of bodily autonomy, this example is intuitively one of criminal activity, in 

the form of an offence against the person. It is used to challenge those 

theorists who conceive of harm, in the form of forward-looking losses of 

autonomy, as a necessary condition of criminalisation.87 Given the 

forward-looking autonomy enhancing, as opposed to negating, effect of 

such treatment, it would seem the treatment is harmless, and therefore 

cannot be criminalised according to those theorists. However, whilst this 

is a powerful challenge to such theorists, it does not undermine the PGA. 

First, the fact that the treatment heals does not mean the autonomy 

exercised in refusing is valueless: it may be the product, say, of religious 

conviction, and the ability to put into practice one’s religious beliefs is 

valuable activity. So, the criminalisation of such treatment still promotes 

valuable autonomy. But, for the sake of argument, let us imagine that the 

refusal of treatment is worthless, for example based on racial prejudice 

against the treating doctor. If the PGA requires that criminal prohibitions 

should only target behaviour that negates the exercise of valuable 

autonomy in others, it seems that such a prohibition is not justified by its 

lights, as it only protects the exercise of valueless autonomy by the person 

refusing treatment. This is a counter-intuitive result and, if required by the 

PGA, would undermine its plausibility.  

The PGA’s approach to this question takes a more practical turn at this 

point. Offences against the person increase the valuable autonomy of 

persons, but in offering blanket protection, doubtless valueless autonomy 

is protected by them as well. However, the PGA considers it impractical 

to allow or encourage citizens to decide for themselves when their victim 

is exercising valuable or valueless autonomy, either empirically (have 

they assessed the situation correctly?) or as a matter of moral judgment (is 

the autonomy indeed valueless?). It is far safer for the overall promotion 

of valuable autonomy, including the valuable autonomy protected by the 

public peace, that the criminal law does not allow citizens to so 

                                                      
87 Those who subscribe to the forward-looking conception of harm as a necessary 

condition of criminalisation include Gardner: see Offences and Defences (n 1): 

the notion that “… life-prospects are being affected adversely …” 244; Simester 

and von Hirsch Crimes Harms and Wrongs, Ch 3 ‘Crossing the Harm Threshold’ 

(n 1) and ‘On the Legitimate Objectives of Criminalisation’ (n 41) 378. Raz also 

subscribes to a forward-looking conception of harm: see The Morality of 

Freedom (n 2) 413-414 and 416. This might be labelled the ‘prospect harm’ 

conception, to use Stanton-Ife’s term when summarising the position of such 

theorists: see his ‘Horrific Crime’ in Duff, Farmer, Marshall, Renzo and Tadros 

(eds), The Boundaries of the Criminal Law (n 67) 129, 159.    



THE DENNING LAW JOURNAL 

 

103 

discriminate, as the potential for error by those persons is very great 

indeed.  

Theorists who criticise the harm principle for failing to account for the 

criminalising of the doctor’s behaviour argue that when wrongdoing 

consists in the violation of another’s rights to personal autonomy, that 

wrongfulness alone is enough to justify criminalisation, and so the case 

should be accounted for on those grounds.88 Such an approach risks 

missing the point that the harm principle addresses a central concern of 

criminalisation: the need to justify the fact that the criminal law is a 

creature of the state.89 All theories of criminalisation should therefore 

offer such a justification: it is not enough simply to declare violations of 

autonomy worthy of criminalisation upon the basis of wrongfulness alone. 

John Gardner has addressed this need in the context of the wrong of rape: 

 

… the would-be rapist is a would-be wrongdoer. This already 

picks him out as a suitable person to be threatened with 

punishment (coerced). It is not the job of the harm principle to 

pick him out again. The job of the harm principle is to regulate the 

wider purposes of the law that does the threatening. This law, and 

indeed every coercive law, must have and fulfil a harm-prevention 

purpose. The prevention of offence, distress, pain, vice, or indeed 

further wrongdoing is not sufficient warrant for coercion by law 

unless by such coercion the law also prevents harm.90 

 

The question boils down to finding the correct principle to justify state 

intervention. For Gardner and others, it is harm; for Thorburn, it is the 

state’s responsibility to secure the equal freedom of all citizens; for Duff, 

it is the vindication of moral values that concern the polity as a whole;91 

                                                      
88 H Stewart, ‘The Limits of the Harm Principle’ (2010) Criminal Law and 

Philosophy 17, 33, where Stewart calls such violations ‘juridical wrongs’. He is 

not the only theorist to suggest that the presence of juridical wrongs is sufficient 

to justify criminalisation: see Stanton-Ife, ‘Horrific Crime’ ibid 161. 

Alternatively, fidelity to the harm principle where juridical wrongs are concerned 

might be maintained by conceiving of the wrong itself as also a harm, though not 

of the ‘prospect’ kind: see Stanton-Ife, ‘Horrific Crime’ ibid 159-162 and A 

Spena, ‘Harmless Rapes: A False Problem for the Harm Principle’ (2010) Diritto 

& Questione Publicche 497, especially 513 onwards.  
89 As pointed out by JG Murphy in ‘Retributivism, Moral Education and the 

Liberal State’ (1985) 4 Criminal Justice Ethics 3, 4. 
90 Gardner, Offences and Defences (n 1) 243. 
91 It should be noted that Duff’s normative vision of the criminal law does not see 

it as a coercive practice. 



PUBLIC GOODS AND CRIMINALISATION 

 

104 

and for Moore, it is the state’s duty to impose just retribution. For the 

PGA, as this article has argued, it is the maintenance of public goods.92  

The nature of the PGA is further explicated by offering some brief 

observations on how the PGA might approach the issues concerning sado-

masochism raised by the decision in R v Brown.93 The accused in that case 

consented to the activities in question, and so were exercising their sexual 

autonomy. As a result, the criminalisation of their behavior would not 

appear justified under the PGA, since it arguably targets activity on the 

basis of its supposed immorality, as opposed to targeting behaviour that 

reduces the exercise of valuable autonomy in others. Can the 

criminalisation of the behaviour in Brown nevertheless be justified under 

the PGA? It is suggested that if the criminalisation of such behaviour is to 

be so justified, it should be in the form of public crime. If its 

criminalisation promotes a public good, say a general commitment to non-

cruelty or the maintenance of public health and so, directly or indirectly, 

increases the valuable autonomy of individuals, then that arguably 

presents a (prima facie) case for so doing. But such an objective must be 

carefully balanced, in an informed way, against the value of protecting 

and promoting sexual integrity, both where the individual and the public 

good are concerned. It is only after such an analysis that, according to the 

PGA, the decision to criminalise can be reached. Much depends, it may be 

supposed, on the level of violence intended by the participants. 

Finally, it is suggested that a concern with public goods allows us to 

understand why certain moral wrongs conventionally seen as beyond the 

reach of criminalisation, such as adultery and lying to friends, are 

legitimately seen that way. It is only when moral wrongdoing impacts on 

a public good that it is potentially worthy of criminalisation. This is why 

romantic infidelity and betrayals within friendship are not worthy of 

criminalization under the PGA, because their occurrence does not impact 

on any public good. Indeed, there is a public good in the general 

availability to form and manage our own romantic and other kinds of 

                                                      
92 For Raz, the impact on valuable autonomy constitutes harm, of the prospect 

kind (see n 87): see The Morality of Freedom (n 2) 417 and 426. The PGA 

therefore has much in common with those theorists who support the harm 

principle as a necessary condition of criminalisation. However, though this claim 

cannot be defended here, the PGA as this article conceives believes that valuable 

autonomy and public goods can be undermined other than through prospect harm. 

Also, in contrast to harm moralists such as Gardner, the PGA fuses its concern 

with moral wrongs with its political conception of the criminal law, as opposed to 

seeing the prevention of harm as an independent political concern of the criminal 

law once a (non-political) moral wrong has been established.   
93 [1994] 1 AC 212. 



THE DENNING LAW JOURNAL 

 

105 

relationships. If the criminal law was to threaten conviction for the various 

forms of betrayal that can occur in such relationships, it would often 

remove the element of sincerity that is key to the existence of the shared 

goods of marriage and friendship and the public goods of which they are 

part.94  

 

7. CONCLUSION 
 

As stated at the outset of this article, for a theory of criminalisation to 

draw on the notion of public goods is not new. The fundamental aim of 

this article was therefore to explore and defend, in greater depth than has 

occurred previously, the implications of a theory of criminalisation 

embedded in the notion of public goods. The use of the four desiderata 

was designed to tease out the merits of such a theory, in the most salient 

and explicit way possible. All of this has been done with the hope of 

stimulating further analysis of what this article believes is an extremely 

promising theory of criminalisation. 

 

 

                                                      
94 For a similar point, see Horder, Ashworth’s Principles of Criminal Law, (n 1) 

54.