The spark for this series of posts was a brief comment at the end of the Introduction to Hart’s Essays in Jurisprudence and Philosophy:
I hope that in what is a second exchange of friendly polemics between myself and Fuller … I have not been unfair in my criticisms of his conception of an inner morality of law; but I see now largely as the result of Professor Lyon’s [sic] essay on Formal Justice that an argument similar to mine against Fuller might be used to show that my claim made in [“Positivism and the separation of law and morals”] and repeated in my Concept of Law that a minimal form of justice is inherent in the very notion of a general legal rule applied according to its tenor to all its instances is similarly mistaken. I am not sure that it is so, but I am clear that my claim requires considerable modification.
As well as wanting to investigate the possible implications of this suggestion myself, I was intrigued to see what writers on Hart had made of it. A quick literature review – carried out by the unscientific method of searching for the name Hart and the phrase “requires considerable modification” on Google Scholar – brought back three papers, by Matthew Kramer, Leslie Green and John Gardner. (This is not a discussion of Hart’s approach to procedural justice more generally, which would have to range much more widely. To name only the most obvious omission, Gardner’s paper is a response to Nigel Simmonds’s book Law as a moral idea – which I read some time ago and, I’m afraid, have not re-read for this blog post.)
Here, then, are our three authors on Hart on Lyons, and on Hart’s later inclination to row back from the idea that “the very notion of a general legal rule applied according to its tenor to all its instances” embodied a form of justice.
Kramer, M. (1997), “Justice as constancy”, Law and Philosophy 16:561-80.
The focus of Kramer’s paper is on Lyons’s argument against classing formal regularity as procedural justice, and on Lyons’s challenge to Hart. I have argued that there is some equivocation in The Concept of Law as to whether procedural justice is merely a technical merit of a well-functioning system of law – just as killing without detection is a technical merit of a well-functioning poison – or a quality which deserves the name of justice, albeit without any necessary moral weight. Lyons argued for the first of these positions; Kramer argues persuasively for the second. Procedural regularity, Kramer argues, “ensures that official conduct in the administration of laws is no worse (and no better) than what is required by the substantive standards of fairness in the laws themselves”; “procedural justice ensures a minimum of substantive justice – a minimum that is also a maximum, to be sure”.
Lyons argued that Hart’s briefly sketched examples of procedural justice – treating like cases alike, following a rule, impartiality – were empty in the absence of substantive criteria of justice; he went on to argue, contra Hart, that the justice of rule-following was no such thing and that rule-infractions were no more ‘unjust’ than infractions of the rules of grammar. In this respect Kramer’s paper arrives at similar conclusions to my posts, by a slightly different route. Where I had argued that Hart’s idea of procedural justice is best understood as a combination of rule-following and impartiality (in cases where discretion is required), Kramer focuses on rule-following and dismisses impartiality; however, he dismisses it as a stand-alone virtue and does not discuss the relationship between impartiality and discretion. Similarly, with regard to the neutrality of laws and the comparison between criminal and grammatical offences, I wrote:
rules may be contradictory or incomprehensible without any injustice being done, just as rules can be followed or broken without moral implications. However, in social practices where distributive or allocative justice is at stake – where a single standard is being applied to multiple individuals, with potential consequences affecting them – procedural justice is necessarily engaged, and rule-writing and rule-following become questions of justice
(“Allocative justice” refers here to the assignment of legal statuses (e.g. ‘guilty’) and of criminal penalties, which are justly allocated when allocated to those who deserve them. The same argument holds if we think in terms of “corrective justice” instead.) Kramer wrote (in 1997):
a charge of procedural injustice can be perfectly apposite in connection with an infringement or a putative infringement of a linguistic norm – provided that the charge pertains to the groundlessness of someone’s condemnation (or exoneration) of the person who has allegedly infringed (or actually infringed) the norm. When we determine whether a procedural injustice has occurred in relation to such a norm, what matters is not the norm’s morally neutral status but the misdirectedness of someone’s judgment about someone else’s compliance or noncompliance with the norm’s requirements.
Kramer’s main argument turns on his reference to procedural justice as “a minimum that is also a maximum”. As such, he argues, it may be profoundly immoral, to the extent of enforcing the violation of moral obligations:
Although an official of course has a moral duty not to enforce a wicked mandate in situations where it is inapplicable, he does not have a moral duty – even a prima-facie moral duty – to enforce the mandate in situations where it indeed is applicable. He has a moral obligation not to go below the minimum of decency secured by procedural justice, but he also has a moral obligation not to treat that minimum as a maximum.
Where procedural justice sets a minimum (preventing unrestrained official harshness), it is moral; where it sets a maximum (preventing unrestrained official benevolence) it is immoral. Since it always does both of these things, it must be considered as, in itself, neutral – neither moral nor immoral. This is a powerful argument, but focuses – as had Lyons – on the justice of the outcomes produced by procedurally just or unjust processes. As I posted earlier, I think it is sustainable that an inconsistent and arbitrary application of the law carries injustice in itself, by singling out individuals for disrespectful treatment on grounds which are not open to them to understand or challenge. There is no difference in this sense between procedural injustice which produces unlawfully harsh outcomes and procedurally unjust favourable treatment: in both cases, injustice is done to the population as a whole, none of whom can have any confidence that the law will address them as individuals responsible for steering their own conduct.
Green, L. (2010), “The germ of justice”, Oxford Legal Research Paper 60/2010.
Green takes his title from one of Hart’s remarks about procedural justice: “though the most odious laws may be justly applied, we have, in the bare notion of applying a general rule of law, the germ at least of justice”. This ‘germ of justice’ argument stands alongside Hart’s denial of any necessary moral content to the law – “it is in no sense a necessary truth that laws reproduce or satisfy certain demands of morality, though in fact they have often done so” – and may seem to contradict it, particularly if Hart is held to his earlier formulation: “there is, in the very notion of law consisting of general rules, something which prevents us from treating it as if morally it is utterly neutral, without any necessary contact with moral principles”.
Green asks: if the law (as a system of general rules) offers ‘the germ of justice’ and as such has some necessary contact with moral principles (the ‘as such’ here is the connection broken by Kramer’s argument), how can this be reconciled with Hart’s “separation of law and morals” – now commonly referred to as the ‘no necessary connection’ thesis? To be more precise, Green asks two questions: whether the ‘germ of justice’ thesis can be held consistently with legal positivism, and how (if so held) it could be true. An answer to the first question is to point out that the formulation ‘no necessary connection’ may be altogether too broad: to say that law does not necessarily “reproduce or satisfy certain demands of morality” is not to deny a number of other possible claims about law and morality, which a positivist might feel able to concede (e.g. “law necessarily deals with moral matters; law necessarily makes moral claims on its subjects, law is necessarily apt for appraisal by moral standards”). It may then be that a purely procedural morality of law – summed up in a statement such as “a functioning legal system is morally superior to a poorly-functioning system” – can similarly be entertained alongside a commitment to legal positivism (although Kramer, for one, would not accept this). Another answer is to stress that the ‘germ of justice’ is consistent with key principles of Hart’s legal positivism – specifically, what Green terms the ‘social thesis’ and the ‘fallibility thesis’. In other words, we can hold that the application of the law is always (in some, unspecified and perhaps very minimal, sense) a moral endeavour while also holding that the law is a social construction on which morality has no prior claims, and that actually existing legal systems may be highly fallible in terms of the justice they deliver. The ‘germ of justice’ in this latter formulation is reminiscent of Hart’s half-concession towards Fuller (“if this is what the necessary connection of law and morality means, we may accept it. It is unfortunately compatible with very great iniquity.”); it is a germ that makes no justice.
Green’s answer to his second question covers similar ground to Kramer’s response to Lyons, as well as the conclusion of my post on the purpose of law. Green notes the strong association between norms of allocation and the law, on one hand, and justice on the other: “perhaps we can say that every legal system contains allocation norms, that they are among the most important norms in the system, and that it is among these norms we find norms of justice”. Pursuing this line of thinking, Green notes that “Law is not just any old bunch of social rules”: “if there were general virtues of rule-application, and even if these had something to do with justice, then they would turn up promiscuously whenever and wherever rules are being applied. … If it were simply the ruly character of law that produced the germ of justice, then justice would have no more intimate association with law than it has with grammar or football.” Rather, the function of the legal system is to equip society with “institutions that can identify and authenticate the rules of the system, and that can render binding determinations in any disputes about them”. Hence the connection to allocation norms: “[a]uthoritative adjudication is in its essence an allocative enterprise, for its core function is to settle who is to get what, and on what grounds.” Perhaps, then, the establishment of a legal system (founded on the basic principles of procedural justice) represents the ‘germ of justice’ because it is the fundamental precondition of subjecting society to the rule of law. This remains compatible with the ‘social’ and ‘fallibility’ theses, and with a narrow reading of the ‘separability’ thesis:
To say that a legal system must of its nature settle disputes about certain kinds of concerns, and that these are moral concerns, is not to say that it must do so well or even tolerably. And as Hart insisted time and again, to say that every legal system must deliver the goods to some people does not show that it must deliver the goods to everyone, and that one of the characteristic ways that law can fail, consistently with satisfying the minimum content, is by not delivering them justly. But if we have law-producing and law-applying organs, then we do at least have social machinery by which justice could be made effective
This is an interesting and cogent argument, whose strengths (at least from my point of view) include the fact that I was thinking along similar lines before I read it; great minds, etc. Its key weakness is that – as Green concedes – the route it takes from premise to conclusion does not track Hart’s own thinking; indeed, the rule of law was a topic which Hart seems to have largely avoided, possibly mistrusting its potential for adverse judgments of ‘unlawlike’ legal systems.
Gardner, J. (2010), “Hart on legality, justice, and morality”, Oxford Legal Research Paper 44/2010.
After some fairly knockabout remarks aimed at Nigel Simmonds, Gardner focuses on Hart’s brief and often equivocal remarks about procedural justice (and about Fuller’s ‘inner morality of law’). He concedes that Hart can be interpreted as expressing outright scepticism about the concepts in question (the alternative being acceptance with limited enthusiasm). If a sceptical reading is preferred, Gardner argues, the question is where – in his formulation – the chain is to be broken: is Hart supposed to have believed that law had no necessary connection with the ideal of legality; that the ideal of legality had no necessary connection with justice; or that justice had no necessary connection with morality? In an illuminating argument, Gardner argues that Hart’s conception of law as a system of rules – “general standards of conduct communicated to classes of persons, who are then expected to understand and conform to the rules” – brings with it an assumption that laws are intended to be followed, and not simply cited by officials as a justification for coercive behaviour modification. It follows logically – whether or not Hart himself would have wished to draw this inference – that, to the extent that more laws within a system are being obeyed intelligently and willingly, that system is more lawlike. In Gardner’s words,
the ideal of legality or the rule of law is an ideal for law because there is a conceptually necessary feature of a legal system, namely that it is a system of rules, which entails that it has a proper way of functioning as a legal system, namely by guiding or (as we also put it) by ruling those who are subject to it.
As for the relationship between legality and justice, Gardner argues that the principles of legality are principles of justice: they guide free and rational individuals in such a way as to make it possible to reach a satisfactory adjustment of claims between multiple individuals. At this point Gardner endorses Lyons’s critique of formal justice (more or less in passing), dismissing any connection between justice and the law in and of itself. As with Lyons himself, the argument here seems to rest on a broader and more expansive definition of ‘justice’ than is usually implied in the phrase ‘procedural justice’. The question of whether, if an unjust society began to follow its rules more uniformly, there would be any gain in justice is not directly addressed.
Gardner sews up the relationship between legality and justice fairly tightly, but in an argument which I found difficult to paraphrase and not much easier to get to grips with. On the relationship between justice and morality he is more inconclusive and to my mind more interesting. He canvasses three readings of Hart on morality and the law, and the ‘inner morality of law’ in particular. One is that Kramer’s reading is correct – Hart believed in justice as a (more than technical) virtue of functioning legal systems, but not as a moral virtue. This reading is well supported by Hart’s own words; Gardner seems to reject it because he finds it too outlandish. A second is that Hart’s lukewarm endorsement of Fuller’s ‘inner morality of law’ should be taken at face value: Hart believed that Fuller had indeed identified a morality of law, but not a complete set of moral criteria for judging the law (which would require attention to the substantive aims of the law). This is weakly supported by Hart’s writing and – perhaps more importantly – would have represented a serious misreading of Fuller, whose ambitions for his ‘inner morality’ were less ambitious than this. The third reading, to which Gardner inclines, is itself inconclusive; perhaps, Gardner argues, the key is that ‘morality’ was a difficult concept for Hart, in whose presence his thinking became more than usually tentative and disconnected.
Reviewing the three papers, Kramer’s critique of Lyons is excellent, and his position on procedural justice – that it is a virtue but should not be seen as a moral virtue – seems authentically Hartian; I was not convinced by the argument by which he supported it, though. Green fills out the legal positivist background very usefully, as well as alighting on a potential connection – not necessarily one Hart had in mind – between a procedural morality of law and the minimum content of natural law. Gardner’s paper does an excellent job of presenting Hart as a liberal idealist, albeit one with a neuralgic reaction to the word ‘morality’; the argument is very much in line with my own thinking about the law, but as an interpretation of Hart I found it less persuasive.
Hart was clearly resistant to any argument which would suggest a necessary connection between morality and either the structure or the content of law; the evident appeal of Lyons’s formalism to Hart may have derived from its undertaking to sever morality from formal justice. If Hart had reworked his position in CoL to take account of Lyons’s argument, however, I think he might have found it came at too high a price; short of a Kelsenian late-career volte-face, I can’t imagine that the changes would have been any more than cosmetic.