I. Introduction
The rule of law is often characterised as a political ideal to which every legal system, regardless of its substantive aims, must aspire and against which all legal systems may be judged.Footnote 1 This is particularly true of “formal” conceptions of the rule of law which are seen as speaking only to legal form and the methods of promulgation and application of legal rules.Footnote 2 Formal conceptions are often portrayed as somehow politically neutral. More precisely, they are said to constitute the threshold conditions for a valid legal system, while nevertheless remaining neutral as to the substantive ends that may be pursued by law.Footnote 3 This characterisation ensures that the rule of law can carry political weight to pressure certain regimes to treat their subjects better without being dismissed as a product of western or liberal-democratic values.
In contrast, substantive theories are seen as expanding upon the formal or procedural limitations contained within formal theories by establishing a number of fundamental rights which are said to be based on, or derived from, the values which underpin the rule of law, and which affect and limit the substantive aims of law.Footnote 4 Within the context of British constitutional law, the rule of law forms a core part of common law principle and is increasingly used in judicial review decisions to ground specific heads of review and to justify the operation of judicial review generally.Footnote 5 In particular, the concept incorporates both formal and substantive elements which the legislature is presumed not to contravene. As Lord Steyn notes, “[u]nless there is the clearest provision to the contrary, Parliament must be presumed not to legislate contrary to the rule of law. And the rule of law enforces minimum standards of fairness, both substantive and procedural”.Footnote 6
Often, substantive conceptions of the rule of law are also portrayed as intimately connected to justice, such that adherence to the rule of law guarantees the “goodness” of the legal system as a whole and of all individual laws which comprise it.Footnote 7 It is impossible to separate conceptions of the rule of law from underlying philosophical commitments regarding the relationship between law and morality. As Craig notes, it is unsurprising that proponents of formal conceptions of the rule of law tend to also embrace a legal positivist conception of legality and that proponents of substantive conceptions tend to embrace anti-positivism.Footnote 8 Nevertheless, it seems that there is some space between conceptions of the rule of law which are completely neutral regarding the content of law and the protection of certain fundamental rights and those which fully collapse legality into a particular conception of justice. The arguments advanced in this paper seek to problematise the neat distinction between formal and substantive conceptions of the rule of law: once we move beyond an impoverished conception which sees the rule of law as simply the faithful application of the commands of a political ruler,Footnote 9 we inevitably embrace an ideal which speaks to the content of law and requires the protection of certain fundamental rights, even if it does not guarantee the substantive justice of each and every law.
The paradigmatic example of a formal conception of the rule of law is that of Lon Fuller.Footnote 10 Fuller argued that legal systems, if they are to be truly classed as such, must generally contain laws which are (1) general, (2) open, (3) prospective, (4) sufficiently clear, (5) non-contradictory, (6) stable, (7) capable of being obeyed and, further, (8) there must be congruence between laws as enacted and as applied.Footnote 11 Conformity with these eight desiderata ensures that governance occurs in accordance with law as opposed to the arbitrary will of some tyrannical political ruler and generates what Fuller referred to as the “internal morality of law”.Footnote 12 While Fuller referred to these desiderata as constitutive of this inner morality of law and integral to the concept of law, later theorists have recast these requirements as pertaining to the “rule of law”, an ideal which may or may not be necessary for legal validity. This paper will focus primarily on Fuller's theory to show that even these apparently rather thin, formal conceptions of the rule of law contain elements which have traditionally been associated with a substantive account.
While Fuller saw the rule of law as an internal morality, substantiating a connection between law and justice, his explanation of how this obtained was often vague. In particular, he failed to clearly explain how adherence to the rule of law might affect the content of legal rules, a failure which led Hart to conclude that Fuller offered no “cogent argument in support of his claim that these principles are not neutral as between good and evil substantive aims”.Footnote 13 This paper seeks to provide such an argument by showing how fidelity to the rule of law has significant implications for the kinds of ends which might be pursued through law, even as it does not guarantee that all laws are substantively just. Furthermore, one must have some conception of the ends which might be pursued through law in order to determine what the appropriate means are for achieving those ends and what legal forms are best suited to do so.Footnote 14 In this sense, form and substance, and means and ends are intimately connected to one another and are in many ways mutually transformative.Footnote 15 Fuller's theory of the inner morality of law must be read in light of this conclusion.
Throughout this paper the precise nature of the “content” and “form” of law will be elucidated. The claim that formal theories such as Fuller's do not speak to the content of law will be challenged. Once it has been shown that Fuller's desiderata (and, indeed, any conception of the rule of law which places some limitations on the use of political power) does, in fact, speak to the content of law, I will then move on to examine the rights that are entailed by fidelity to these (allegedly) “formal” accounts. If the rule of law entails respect for fundamental rights, then those rights must amount to legal ends which contribute to the content of law in some capacity. This being the case, the rule of law, by focusing on both legal means and legal ends, cannot be easily described as formal or substantive.
In his most recent book, John Gardner also challenges the supposed formality of the rule of law by relying on exactly these arguments concerning legal content and legal rights.Footnote 16 This paper develops those claims, arguing that they create significant problems for Gardner's own account of the rule of law as “modal” – defined by legal means and not legal ends. The requirements of the rule of law are not means to an end; they serve to limit the range of ends which can legitimately be pursued by a lawmaker, restricting far more than they facilitate. I also suggest that Kristen Rundle, in her critique of Gardner, fails to appreciate the value that his contributions might have for those of us who wish to defend a moral conception of legality.
II. Three Key Distinctions
Since the seminal writings of A.V. Dicey, the rule of law has played an important role at the centre of the British constitution.Footnote 17 To Dicey, the rule of law entails “the absolute supremacy or predominance of regular law as opposed to the influence of arbitrary power, and excludes the existence of arbitrariness, of prerogative, or even of wide discretionary authority on the part of the government”.Footnote 18 At a minimum, this concept consists in the impartial enforcement of legal rules such that equal subjection of all to law obtains.Footnote 19 As Allan notes, “arbitrary conduct by powerful persons or groups, unconstrained by rules, represents the antithesis of law”.Footnote 20 Indeed, this formed a core aspect of Dicey's conception – that no man is above the law, and “every man, whatever be his rank or condition, is subject to the ordinary law of the realm and amenable to the jurisdiction of the ordinary tribunals”.Footnote 21 It may be the case that certain classes of persons (e.g. soldiers) are subject to legal duties that ordinary citizens are not and, as such, may be said to be subject to different laws. However, Dicey quite rightly stressed that this is not inconsistent with the rule of law so long as those persons, or any other legal officials, are not exempted from the application of ordinary law or the legal duties which are entailed by it.Footnote 22 No man may be above the law in the sense that all must follow it. However, Dicey seems to have envisaged an additional requirement, one which prevented the existence of legally enshrined exemptions to law for legal or political rulers.
The rule of law has never simply required the consistent application of whatever commands a political ruler happens to make. These commands must be in some sense legal and there must be some standard by which legal pronouncements can be distinguished from illegal or non-legal pronouncements, even if they come from a duly recognised political authority. What is more, the rule of law appears to demand a conceptual distinction between a mere political or governmental system and a legal system.Footnote 23 Beyond this, however, significant disagreement abounds concerning the precise requirements of the rule of law as well as their character as formal or substantive in nature. There are three key distinctions which could delineate formal from substantive conceptions of the rule of law:
(1) Formal conceptions do not affect the content of law while substantive conceptions do.
(2) Formal conceptions do not entail respect for certain rights of legal subjects while substantive conceptions do.
(3) Formal conceptions do not equate legality with justice while substantive conceptions do.
This paper argues that, once one moves beyond a conception of the rule of law as simply requiring the faithful and consistent application of the commands of a political ruler, these first two distinctions collapse inwards. The only salient distinction that exists between differing conceptions of the rule of law are those which envisage fidelity to the rule of law as guaranteeing the substantive justice of law, on the one hand, and those which, on the other hand, see the rule of law as entailing some commitment to basic requirements of justice and fairness while nevertheless leaving conceptual room for unjust laws which must be critiqued from an external standpoint. Ultimately, fidelity to the rule of law entails fidelity to “a corpus of basic principles and values, which together lend some stability and coherence to the legal order”, even if adherence does not guarantee that every legal rule is a just one.Footnote 24 Once these standards are incorporated into the rule of law, a neat distinction between formal and substantive conceptions becomes elusive.
III. Legal Form and Legal Content
The first, and perhaps most common, distinction between formal and substantive conceptions of the rule of law pertains to the effect that conformity with the rule of law has on the content of legal rules. Craig argues that formal conceptions of the rule of law do not address “the actual content of the law itself”.Footnote 25 Raz similarly argues that “rule of law principles are not about the content of the law, but about its mode of generation and application”.Footnote 26 Kramer sees the rule of law as a “divided phenomenon”, representing both content-independent existence conditions of a legal system and a political ideal that is to be aspired to should the legal ends pursued by that legal system be benign.Footnote 27 Both Raz and Kramer utilise or build upon Fuller's theory, and so the proceeding analysis will primarily focus on his eight desiderata.Footnote 28
What exactly does it mean to say that some theories of the rule of law do not address legal content? Specifically, how can legal content be distinguished from legal form? One starting point is to note that the form of something is naturally contrasted with its content. To use Gardner's example, “the content of a book is one thing (jokes, short stories, etc) and its form (hardback, e-book, etc) is another”.Footnote 29 This seems intuitively plausible. However, an initial cause for concern arises when one notes that this approach begins with the presumption that form and substance are neatly separable and conceptually distinct – one defined in opposition to the other. Indeed, this also presumes that our understanding of legal form and legal content can fit neatly with our understanding of form and substance in other contexts. Such an approach may cut off by definitional fiat any interpretation of legal form and legal content which sees them as mutually constitutive.
Nevertheless, if we begin by presuming that legal form and legal content can be easily contrasted, we immediately run into a broader, more foundational issue: on this definition, almost nothing in traditionally formal theories can be properly described as formal. Gardner makes this point with regard to Fuller's conception:
True, a law that goes unpromulgated (i.e. that is kept secret) need not have different content from its open counterpart. All else being equal, however, a law that it is impossible for people to obey needs to have its content changed if it is to become possible for people to obey it. Likewise, all else being equal, at least one of two mutually inconsistent laws needs to have its content changed if they are to be rendered consistent with each other. What is more, a retrospective law that regulates ϕing necessarily has different content from its prospective counterpart, in that it regulates ϕing in the past as well as ϕing in the future. Stability and generality in a law are also none other than stability and generality in that law's content; ie in what is regulated by it and what is not. A more general law regulating ϕing is one that regulates more cases of ϕing than its less general counterpart. A more stable law regulating ϕing is one that varies less over time, in respect of which cases of ϕing it regulates, than does its less stable counterpart. All of these Fullerian desiderata pass judgment on the content of the law, and indeed on nothing else.Footnote 30
While Gardner is correct in this assessment, his analysis needs significant qualification. In order for some of these desiderata to speak to the content and not the form of law, one must first distinguish between a particular law and the norm which it expresses. Often when we speak of the form of a specific law, we are not actually referencing the form of the law itself (its expression through legislation, case law etc.). Rather, we are referring to the form of the norm that is contained within that law.
The use of the term “law” here is admittedly quite ambiguous. For this paper, it is useful to distinguish between the following: (1) “law” as in a legal system; (2) “laws” as in the individual statutes or common law rules that exist within a legal system; and (3) the “laws of a case” as in the particularised propositions of law that decide individual cases by reference to the general laws which comprise a legal system.Footnote 31 When describing the form of law, Gardner seems to be focusing on the second of these meanings – the individual laws that exist within a legal system. Often, commentators referencing the formality of the rule of law seem to be referencing something else entirely, focusing more on the form of legal norms which are contained within those statutes or common law principles.Footnote 32
Thus, if we rely on a distinction between “laws” and “legal norms” when we look to legislation that retroactively prohibits the wearing of face coverings, we could say that the form of the law is legislative in nature, while the legal norm contained within or derived from this law has its own form and substantive content. The substantive content of the norm in question could amount to the prohibition on the wearing of face coverings whereas its form would, in part, speak to its temporal dimension. Nevertheless, it is crucial to recognise that the content of a law contains both the substance and the form of a particular norm; it includes a legal norm in its entirety. As such, while Gardner is correct when he says that the rule of law “has no general guidance to give regarding law's form”, it does seem to have some guidance to give regarding the form of the norms that form the content of particular laws.Footnote 33
At this point it may be possible to reformulate this claim to distinguish between theories of the rule of law which place limitations solely on the form of legal norms and those which place limitations on the content of these norms. This new formulation would necessarily accept that the rule of law, even on a “formal” conception, affects legal content. However, a conceptual distinction might remain which would allow for the categorisation of certain theories into the formal camp and others into the substantive camp. On this reading, formal theories of the rule of law do not affect the substance of legal norms, only their form. To use the example above, they might have something to say regarding the temporal dimension of a particular legal norm, or the scope of the norm, but nothing to say about the prohibition of face coverings, or the substance of any other legal norm. As such, we might reformulate Craig's initial claim to the following: formal theories of the rule of law do not address the actual substance of legal norms, even if they affect the form of those norms and thus affect the content of law.
While this reformulation is a better representation of an allegedly “formal” conception of the rule of law, it concedes that formal conceptions are not actually content-neutral and, crucially, it does not apply to Fuller's theory. His desiderata do, in fact, affect the content of legal norms. To draw out this point, consider the requirement that laws must be capable of being obeyed. If an enactment were to violate this criterion, we might have an issue with either the form of the norm, the content of the norm, or both the form and the content of the norm. An obvious example of a breach which can only be rectified by changing the content of a legal norm would be a requirement that people grow wings and fly to a designated place five years prior to the enactment of this statute. As absurd as this law might be, we can clearly see that this falls foul of the rule of law due to both the form and the content of the legal norm in question. On the one hand, the form of the norm is implicated because the temporal dimension of the norm makes it impossible to be obeyed. On the other hand, the content of the norm is implicated because it requires people to do something that they cannot possibly do.
Another example of Fuller's desiderata affecting the substantive content of legal norms is the requirement of non-contradiction. If two laws were enacted, one containing a norm which mandated the wearing of face coverings, while the other contained a norm which prohibited the wearing of face coverings, the only way to resolve such a conflict would be to change the content of one of those norms. Whether these desiderata speak to the content of the norm or to its form will depend on the context. What is certain however, is that these criteria will place limitations on the content of legal norms in some contexts. As such, we cannot say that Fuller's conception of the rule of law only manifests restrictions on the form of legal norms. His theory – the prime example of a formal understanding of the rule of law – affects both the content of law and, more narrowly, the content of the norms expressed through law.
In this sense, Fuller's theory is far more “substantive” than he seems to have recognised. It is one thing to say, as Rundle does, that Fuller's theory focuses on the form of law but, due to the mutually transformative nature of means and ends, that this has implications for legal content; it is another thing entirely to say that this theory doesn't say much about form at all and in fact focuses far more on legal content than we have hitherto recognised.Footnote 34 In particular, Rundle challenges the lack of nuance within most positivist descriptions of Fuller's theory, emphasising that the formality (as well as the modality) of law has implications for the content of law and, consequently, that “Forms Liberate”.Footnote 35 As such, her criticism of Gardner is focused on his failure to account for the transformative nature of legal form and his reliance on an instrumentalist conception of the rule of law. These criticisms ring true. However, there is more to be said in favour of Gardner's analysis than Rundle gives him credit for: the claim that the rule of law is not in fact content-independent has significant implications for any description of Fuller's theory as “formal”, or “procedural”, or even, as we will explore below, “modal”. In particular, if we accept that the rule of law focuses far more on legal content than it does on legal form, we must concede that Fuller may have misdescribed his own theory and that attempting to reclaim Fuller on Fuller's own terms (as Rundle does) may hamper the development of a theory which builds upon and develops beyond the foundations that he lay down.
In addition to Fuller's, other allegedly “formal” conceptions of the rule of law appear, on closer inspection, to affect the content of law as well. For example, Dicey's conception precludes legally enshrined exemptions for legal officials and consequently requires generality in aim. Laws must be impersonal, applied only to general classes of persons, and contain no proper names, – in contrast with blatant arbitrariness or caprice.Footnote 36 On this conception, lawmakers are prohibited from enacting a law, the substantive content of which specifically targets an identified individual or which places legal or political rulers outside the reach of legality.Footnote 37 Further, it would not be open to lawmakers to enact legislation which granted unlimited discretionary powers to government officials or which placed their decisions outside the bounds of judicial review.Footnote 38 Crucially, the issue here would be with the content of such laws, not their form or the procedure by which they came to be enacted.
Lawmakers, acting with fidelity to the rule of law, are evidently constrained in terms of the content of their enactments, even if this constraint is voluntary such that a principle of parliamentary sovereignty might take precedence over the rule of law in some circumstances.Footnote 39 However, this seems unlikely to occur within the contemporary UK context. As Lord Hope observed in Jackson, “Parliamentary Sovereignty is no longer, if it ever was, absolute”.Footnote 40 Importantly, the reason for this is that the “rule of law enforced by the courts is the ultimate controlling factor on which our constitution is based”.Footnote 41
Thus, if the hallmark of a substantive conception of the rule of law is that it affects the content of law, then we must conclude that both Fuller and Dicey expounded substantive conceptions. Indeed, it would be difficult to find any theory of the rule of law that did not affect legal content in some manner. Even Hayek's account, summarised as the requirement that “government in all its actions is bound by rules fixed and announced beforehand”, would necessarily preclude the existence of laws with content that delegated unlimited authority to a particular government minister to disregard previously announced rules.Footnote 42 Even as thin an account as this must necessarily speak to legal content if it is to distinguish legality from brute political force.
IV. Fundamental Rights
Substantive theories of the rule of law are portrayed as unique in their recognition of the existence of certain rights. These rights are said to be derived from the values which underpin the rule of law. According to this analysis, once one has accepted the existence of these rights, one can then distinguish between “good” laws, which respect these rights, and “bad” laws, which do not.Footnote 43 We have already seen how the rule of law speaks to legal content; we will now explore how it also mandates that certain fundamental rights form a necessary part of the content of law in some capacity and consequently how those rights limit the kinds of ends which might legitimately be pursued within a legal system.
The above dichotomy is premised on the presumption that supposedly “formal” theories such as Fuller's or Dicey's do not imply or entail the protection of fundamental rights. This assessment seems intuitively plausible. It is certainly true that the desiderata that Fuller sets out are not framed as rights that legal subjects might have. They are presented as requirements that lawmakers must conform to if they wish to establish a legal system and as virtues that a legal system must aspire to in order to achieve some degree of excellence in legality.Footnote 44 This being the case, it would be impossible for a legal system to live up to this ideal without establishing some protections for citizens against exactly the kinds of enactments which might threaten the rule of law. As Gardner notes:
Even though there is no right that there be no secret laws, probably there is a right, derived from the openness desideratum, not to be prosecuted, convicted, or punished under a secret law. And even though there is no right that there be no non-general laws, such as Bills of Attainder, almost certainly there is a right not to be on the receiving end of one.Footnote 45
This is also true of other “formal” conceptions of the rule of law. Dicey's requirement of generality would entail a legal right not to be subject to a Bill of Attainder or to a secret law and may also include a right to challenge the legality of one's detention and to receive the remedy of habeas corpus.Footnote 46 There also appears to be a number of other rights entailed by some formal conceptions of the rule of law including a right of access to courts, a right to have adverse administrative decisions subject to judicial review,Footnote 47 and, following Hayek, a right not to be subject to governmental action which is not authorised in advance by sufficiently precise and determinate rules.Footnote 48 This is not simply a requirement that legislatures must conform to. As Lord Hope stressed in Jackson, the judiciary is also under a (legal) duty to adhere to the rule of law in its adjudication of legal disputes.Footnote 49 We must therefore conclude that Craig is wrong to say that formal conceptions such as Dicey's do not speak “to the content of the laws which an individual will have to face when taken before the courts”.Footnote 50 If a legal system were to conform to the rule of law, specific limitations would be placed on the content of law which would entail rights for citizens not to be subject to enactments, policies or administrative decisions which conflicted with the rule of law.
It is in this sense that these rights are “fundamental”. They are fundamental because that they are grounded within the very foundations of a system of law.Footnote 51 A legal system cannot be said to be in conformity with the rule of law if breaches of the requirements of the rule of law are not precluded by the establishment of fundamental rights. This point is reinforced by the fact that, if the rule of law did not guarantee certain protections to citizens against the abuse of governmental power, it would be of little value and could scarcely be seen as a political ideal at all. This is true even if these rights are not absolute and could therefore be legitimately infringed in certain contexts. It seems that in order for the rule of law to generate any normative force, regardless of whether it is a mere political ideal or an aspect of the existence conditions of a legal system, it must establish some protections. The real issue for our purposes concerns the nature of the protections offered.
Once again, we must accept that a seemingly clear distinction between formal and substantive conceptions of the rule of law becomes elusive. On closer inspection, fidelity to the rule of law evidently does entail respect for certain rights of legal subjects. This being said, there might still be a conceptual distinction at play here concerning the nature of the rights which are entailed by fidelity to the rule of law. It is here where the distinction between form and substance re-emerges. Within this context, formal rights are more often described as procedural, whereas substantive rights seem to have retained that marker even though the connotation of substantive has changed quite dramatically.Footnote 52
Gardner, resisting sharp-line distinctions, notes that terms like “formal” and “substantive” are notoriously unhelpful: their meaning is almost exclusively context-dependent.Footnote 53 Nevertheless, it may be possible to shed some light on this dichotomy. One might assume that a procedural right speaks only to legal means: the methods used to reach particular legal conclusions or to generate particular legal rules. Indeed, this reading would conform with how the term “procedural” is ordinarily used, as well as with how it is used when speaking of procedural justice.Footnote 54 In contrast, substantive rights are portrayed as establishing the content of the legal conclusion or of the legal rule itself – legal ends.
However, once we begin to explore how particular claims might be tested within a legal system, it becomes more and more difficult to clearly delineate between procedure and substance in this context. Again, insight can be drawn from Gardner's analysis. He argues that often the substantive content of a legal case will be focused entirely on procedural grounds.Footnote 55 He uses the example of a lawsuit brought against the police for unlawful imprisonment. Here the substantive content of the case concerns procedural impropriety. If the police failed to follow the correct procedures of arrest, detention and questioning of a suspect, then a right has been breached:
the rules governing these police procedures are procedural rules, and yet they are also substantive rules relative to certain other procedural rules (viz. the ones governing how to sue for false imprisonment) … . The substantive question for trial, indeed, is whether there were procedural violations in other proceedings. So the rules (and rights) at issue in the case are simultaneously procedural and substantive.Footnote 56
The conclusion that there is no clear-cut distinction between procedure and substance is of crucial importance and is particularly evident when focus shifts to analysis of supposedly formal conceptions such as Fuller's theory. While one of his desiderata, the requirement of congruence between legal rules as announced and their actual administration, certainly speaks to procedures of rule application, it also speaks to substantive questions relating to correct interpretation, implying a right to have the law interpreted and applied in a manner which respects and maintains the integrity of the legal system as a whole. Fuller notes that this congruence between a legal rule and its application “may be destroyed or impaired in a great variety of ways: mistaken interpretation, inaccessibility of the law, lack of insight into what is required to maintain the integrity of a legal system, bribery, prejudice, indifference, stupidity, and the drive toward personal power”.Footnote 57
While there are certainly procedural elements to this desideratum, Fuller is sketching a rich and nuanced account of adjudication which argues subtly against any clear distinction between procedure and substance.Footnote 58 For Fuller, adjudication is a form of social ordering which is “institutionally committed to a reasoned decision, to a decision based on principle”.Footnote 59 On this view, adjudication is not simply the application of legal standards to particular cases: it also embodies a relationship of reciprocity between lawmaker and legal subject which “gives formal and institutional expression to the influence of reasoned argument in human affairs”.Footnote 60 Interestingly, Fuller also argued that the institutional framework of adjudication tends to convert these arguments into claims of right or accusations of fault.Footnote 61 With this in mind, it is easy to see how a conception of adjudication as institutionally committed to the enforcement of legal rights might imply that the rule of law necessitates respect for the legal subject as a rights-bearing entity. Indeed, any legislation which attempted, directly or indirectly, to erode the right to a fair trial would be in fundamental conflict with the rule of law.Footnote 62 Crucially, such a right does not simply pertain to the faithful application of existing legal rules whatever be their content: it mandates that particular interpretations of rules (those authorising the abrogation of a fair trial) be disregarded as inconsistent with principles of legality. Within the UK, the right to a fair trial is “fundamental and absolute”.Footnote 63 It is difficult to see how respect for these rights does not affect the substantive outcomes of particular cases to which they pertain to.
Furthermore, leaving this aspect of his theory aside, there is not much else contained within Fuller's desiderata that could be described as procedural or means-focused. The rest of his criteria do not establish rights concerning the procedures of rule application; they focus solely on rights pertaining to legal content and the character of the legal system as a whole, just as Dicey's conception entails substantive rights which would dictate the conclusion of particular cases. For example, fidelity to his conception of the rule of law would lead a court to conclude that a Bill of Attainder lacks legal authority or that legal subjects have a right not to be punished or made to suffer except by reference to pre-existing legal rules.Footnote 64 While, at first glance, these may appear to be procedural rights, Gardner is correct that they evidently take on substantive force once they become operative to guide the outcomes of particular cases.
Even though Fuller clearly envisaged a deep connection between his desiderata and a moral conception of law, his theory did not explicitly develop a connection between the inner morality of law and the protection of the fundamental rights that Gardner has identified. The notion that Fuller's theory is not straightforwardly a “formal” one is not new. Fuller himself and many theorists sympathetic to his analysis have stressed that the moral dimensions of his theory complicate this designation.Footnote 65 This is so even if positivists who have taken up his desiderata within their own work would seek to eschew such connections.Footnote 66 Nevertheless, Gardner's observations relating to the implicit connection between fidelity to these desiderata and respect for certain fundamental rights have the potential to expand upon Fuller's initial theorisation in ways that Gardner himself may not have realised. In particular, if respect for these rights is implicit within the rule of law, this has significant implications for Gardner's own theory and his characterisation of the rule of law as modal, to say nothing of the implications for positivist jurisprudence in general.
It is unclear how Fuller himself would see such a development of his theory. Indeed, Rundle is correct to stress that Fuller's arguments were focused primarily (at least as he presented them) on form, not substance.Footnote 67 Nevertheless, one could expand beyond Fuller and Rundle and embrace an account of the rule of law which emphasises necessary connections between the existence of a legal system and the existence of fundamental rights held by legal subjects. This is significant, particularly when one recalls that Fuller's desiderata are commonly recognised as the bare minimum conditions for the existence of a legal system. It is one thing to have a rights-based account of “thick”, “substantive” conceptions of the rule of law; it is another thing entirely to recognise that even the most impoverished, thin account of the rule of law implies respect for fundamental rights. If this is so, then the very concept of a legal subject becomes rights-laden: to be a legal subject is to be a rights-bearing entity.
V. Means, Ends and Modality
While this paper has offered some challenges to characterisations of the rule of law as formal or procedural, there may be some distinction that could be used to differentiate theories like Fuller's or Dicey's from theories which are similar to that of Ronald Dworkin or T.R.S. Allan.Footnote 68 This appears to be the driving force behind Gardner's conclusion that the rule of law is not formal but modal, in that it is distinguished by its means and not its ends.Footnote 69 Here, modality is defined by a focus on the means by which legal ends are pursued. Functional theories, in contrast, are defined by a focus on the attainment of specific legal ends. Indeed, Fuller often refers to the inner morality of law as procedural or institutional, indicating that he may agree with Gardner's assessment of the rule of law as modal in that it is defined by its means and not its ends.
However, such a reading of Fuller would present an impoverished account of his writing on this topic. In fact, Fuller was adamant in his conviction to problematise the means-ends distinction.Footnote 70 To him, legal ends are constituted always by reference to legal means and vice versa. To explain this, he offers the analogy of architecture. Architecture, according to Fuller, is “not an abstract science but a practical art”.Footnote 71 By this he means that it exists for the satisfaction of certain human ends, namely beauty and utility. The means of achieving these ends are materials such as cement, lumber and steel, as well as the skill necessary to assemble them into the product desired. Fuller quite rightly emphasises that these means are in many ways subservient to the ends of utility and beauty. Thus, it could be argued that one can only begin to define architecture once there is a clear understanding of the ends of architecture: “the study of architecture must begin with ends, with a definition of utility and beauty, for it is only when these ends have been clarified that it is possible to deal intelligently with means, or even to know what means are relevant to the objects of architecture”.Footnote 72
This does not mean that there must be a primacy of ends over means – that architectural means take their entire character from the ends to which they are subservient. Fuller stresses that the relationship between means and ends is not “a one-way affair”,Footnote 73 noting that “[i]n all areas, from the most trivial to the most exalted, the mind is compelled to sharpen its judgement by narrowing its range. Some limitation of means, imposed by circumstances or voluntarily accepted, is essential for an intelligent definition of the ends sought”.Footnote 74 Thus, once we have a goal in mind, it is imperative to account for the means available to us to achieve that goal. Conversely, Fuller argued that one cannot resolve problems associated with an instrumentalist conception of law simply by adopting the polar opposite position of focusing exclusively on means and ignoring ends. For Fuller, some cursory conception of ends is necessary to define the range of means which are worthy of consideration for achieving those ends.Footnote 75
Gardner's modal/functional distinction presumes that formal theories of the rule of law are distinguished by legal means and not legal ends. However, if we recognise the point made by Gardner himself that the rights entailed by fidelity to the rule of law are often both means-focused and ends-focused, then we must conclude that the rule of law is never distinguished solely by a focus on either means or ends. Many aspects of Fuller's or Dicey's theories are unobtainable without their entailed rights becoming legal ends in some capacity, whether in the form of legislation or adjudication, or, more broadly, the foundational principles of the common law. The rights Gardner identifies as entailed by the rule of law must be legal rights, contained within the content of law if they are to have legal force. A right not to be subject to a Bill of Attainder is a substantive right – a legal end in-and-of-itself, even if it might also be categorised as a means of achieving some other end.
Even the most “substantive” theories of the rule of law always look to both legal means and legal ends, informed by concerns for procedural justice, correct interpretation, and faithful application of law as much as they stress the protection of rights to freedom of speech, habeas corpus or freedom of assembly.Footnote 76 Indeed, many of the rights entailed by fidelity to the rule of law cannot be neatly classified as modal or functional, means-focused or ends-focused. They are simultaneously modal and functional. As Rundle notes, Fuller's theory focuses on how legal means and legal ends are mutually transformative.Footnote 77
Gardner offers greater nuance to the means/ends distinction but continues to uphold it through the introduction of an intervening argument which is premised on a characterisation of the fundamental rights entailed by fidelity to the rule of law as subsidiary ends of law.Footnote 78 Gardner did not premise the modality of law on some crude opposition between means and ends: he does not see the requirements of the rule of law as means to an end in their entirety. Rather, being subsidiary ends, they do a significant amount of “end work” even as they act as means to some further, ultimate end.Footnote 79 To him, they are primarily instrumental, pursued “with an eye to whether [they] will help us to reach the further ends to which [they are] subsidiary”.Footnote 80 Fuller never explicitly identified any connection between fidelity to law and respect for fundamental rights. Nevertheless, he firmly rejected an instrumentalist conception of law.Footnote 81 Fidelity to the rule of law ensures that a legal system cannot act as if ends justify means or vice versa; legal ends are always constituted in part by the means used to achieve those ends and legal means are intelligible only when we have some conception of what ends they are to be put to. Indeed, this is the central thrust of Rundle's response to Gardner: he has failed to account for how the rule of law renders legal means and legal ends mutually transformative and consequently relies too heavily on an instrumentalist conception of law.Footnote 82
However, there is a much more foundational problem with Gardner's description of the requirements of the rule of law as means to an end: many of the requirements of the rule of law cannot reasonably be seen as means of achieving further ends at all. While we might describe rights of access to courts or to have laws applied according to their terms as means of achieving certain yet to be defined ends, other desiderata can be seen only as ends in-and-of themselves and operate specifically to limit the range of possibilities open to a lawmaker. They are means only of protecting the rights which are entailed by them, if they can be seen as means at all. These aspects of the rule of law operate to limit the use of political power: to prevent the existence of Bills of Attainder, of unjustified retrospective laws, and of other legal ends which come into conflict with the rule of law such as the attempt to exempt legal or political rulers from legal obligation or to place administrative decisions beyond the reach of judicial review. Kramer is therefore mistaken to suggest that “[w]hether … the objectives pursued by officials are products of moral concern or of exploitative selfishness, the officials can most effectively achieve their ends through the operations of a legal system”.Footnote 83 Some ends are most definitely incompatible with the rule of law and, consequently, conformity with the rule of law cannot be the most effective means of pursuing those ends.
The requirement of non-contradiction is not a means of achieving any specific legal end other than the protection of a right not to be subject to contradictory legal rules. Even then, it seems odd to conceive of this requirement as a means of achieving this end; it is better seen as a limitation which implies and entails a legal right rather than a method of achieving any particular goal. The requirement of prospectivity is a restriction on the ends which can legitimately be pursued within a legal system; it prevents a lawmaker from pursuing the end of punishing legal subjects for conduct which was not illegal at the time of action. This desideratum is a means to an end only insofar as it sets out the threshold condition for recognition of a legitimate end of law; it is a “means” in the same way that a requirement of consideration is a “means” of creating a contract – that is, it is not a means at all. Consideration itself may be necessary for the creation of a contract and, as such, its attainment could be seen as a means of doing so. However, the requirement of consideration as a constitutive element of a valid contract is not a means of creating a contract: it is a categorical limitation which places certain kinds of agreements outside of the class of valid contracts.
Many of the desiderata of the rule of law and the fundamental rights which are entailed by them operate in much the same way through the removal of certain legal ends from the category of rule of law compliant laws. Law may be a means to an end, but the requirements of legality are not. If the requirement of prospectivity is pursued with an eye to some further end, that end is constituted in part by reference to this requirement. This is not a means to an end: it is a foundational aspect of the end itself. A lawmaker attempting to prohibit and punish the wearing of face coverings in the future and one attempting to prohibit and punish the wearing of face coverings in both the past and the future are seeking to achieve two different ends, one of which is compliant with the rule of law and the other is not.Footnote 84 Indeed, this is a point made by Gardner himself when arguing that retrospective laws necessarily have different content from their prospective counterparts.Footnote 85 If they have different content, they also pursue different legal ends.
The requirement of prospectivity, through the pre-emptive preclusion of certain objectives from the class of legitimate ends which can be achieved through law, is an example of the form of a legal norm limiting and defining the end that is to be pursued in much the same way that the tools of architecture limit and define the products that can be produced. Gardner argues that “to say that law is a modal as opposed to functional kind is merely to say that law is not distinguished by its functions – by the purposes it is capable of serving. It is distinguished rather by the distinctive means that it provides for serving whatever ends it serves”.Footnote 86 However, the rule of law clearly does speak to the purposes that it will allow law to be put to. It will not recognise ends such as the retroactive or particularised punishment of legal subjects as legitimate ends of law. The requirements of prospectively and generality serve to cut those ends off at the pass. No matter how much a lawmaker wished to intimidate a political rival through the creation of a law which targeted them by name, that end is not one which would be permissible within a rule of law compliant legal system. It seems, following on from the above, that we can conclude that the rule of law is not distinguished by the means that it provides for serving whatever ends may be desired. Rather, most rule of law principles focus on limiting the range of possible ends that might be open to a lawmaker or governmental official. In light of this conclusion, the following claim advanced by Gardner simply cannot be correct:
If law were a functional kind, distinguished by the purposes it is capable of serving, then the list of desiderata for law's internal morality would also naturally pass judgment on law's ends. But law is not a functional kind, and the evaluation of its ends is therefore left to ordinary, unspecialized, external morality, the same morality that binds us all.Footnote 87
The rule of law evidently does pass judgment on legal ends. Fidelity to the rule of law might not necessarily guarantee the substantive justice of those ends, but it would be wrong to conclude that a rule of law compliant legal system could pursue any ends at all.Footnote 88 Indeed, Fuller recognised that the internal morality of law “affects and limits the substantive aims that can be achieved through law”.Footnote 89 Formal theories are indeed distinguished, at least in part, by the purposes they render a legal system capable of serving. Thus, while compliance with the rule of law may ostensibly lead to the efficient attainment of a wide variety of legal ends, Fuller rightly stressed this does not mean that “any substantive aim might be adopted without compromise of legality”.Footnote 90
VI. Functionalism and Modality
Does the above analysis lead us to the conclusion that the rule of law is, in fact, of a functional kind? For the rule of law to be functional, it must do more than simply affect, limit or stipulate legal ends; it must also identify the overarching purpose of law. Functionalist jurisprudence concerns itself with identifying the functions that legal systems serve, usually with an eye to assessing the connection that may exist between those functions and some conception of morality or justice.Footnote 91 As such, it attempts to define the concept of a legal system by reference to its function or purpose as opposed to its structures.Footnote 92 Was Fuller a functionalist? Early in his career he likened a legal system to a steam engine in that both were functional kinds. This meant that, for Fuller, one could not clearly delineate between questions seeking to determine if X was a steam engine and those seeking to determine if X was a good steam engine, as to be a steam engine at all is to perform the function of steam-engines to some degree.Footnote 93 Furthermore, Fuller regularly stated that the function of a legal system was to subject human conduct to the governance of general rules.Footnote 94 If his argument was that the function of law is to guide human conduct through the use of general rules and that it was this that defined law and grounded the connection between law and morality, then it seems plausible that we could conclude that Fuller was indeed a functionalist as Moore claims.Footnote 95 Of course, if this is the case, then the well-known critique that Fuller's inner morality of law amounts to nothing more than instrumental efficiency is correct.Footnote 96
However, it seems that Fuller, in a quite characteristic manner, referred to the function of law as being concerned with the guidance of human conduct in accordance with general rules without intending to commit himself to the functionalist approach of defining law by the functions that it serves and assessing its morality in light of those functions. Fuller argued that law was functional in the sense that it has functions which can be achieved by subjecting human conduct to the governance of general rules. Gardner is quite right to say that Fuller did not conceive of this subjection as the purpose of law; rather it is a means of achieving the ends that law might be put to.Footnote 97 Fuller himself noted that legal systems can serve “objectives … of the most diverse nature”.Footnote 98
It seems that Fuller's account, despite him describing law as functional, does not fall into the functionalist camp, distinguished by an approach of defining law by reference to its functions. However, this does not necessarily mean that his theory is modal either in that it defines law by reference to legal means. It is true that Fuller saw the rule of law as concerned with ensuring that human conduct could be guided in accordance with general rules. Fuller's theory is certainly premised on the belief that the subjection of human conduct to the governance of general rules is a means of achieving certain legal ends. However, Fuller's work on means and ends attempted to shift focus away from functionalist approaches which identify ends in abstraction from the means used to achieve those ends.Footnote 99
The crux of his argument against functionalism can be seen when we examine his critique of utilitarianism and the analogy to architecture mentioned above. Fuller adamantly resisted the claim that rules of action “must take their whole character and colour from the end to which they are sub-servient” and stressed that no end can remain unaffected by the means through which it is pursued.Footnote 100 As Rundle notes, this is clearly not an instrumentalist or “modal” understanding of legal form – such an interpretation would be antithetical to the whole tenor of Fuller's work.Footnote 101 Note, however, that while Rundle challenges Gardner's analysis, she remains wedded to an interpretation of Fuller which is almost exclusively focused on the “formality” of law and fails to lend appropriate weight to the aspects of Fuller's theory which do not focus on legal form and instead focus on the integrity of the legal system as a whole. Her approach emphasises how legal form and legal ends are mutually transformative, but this is only one aspect of Fuller's inner morality of law and Rundle at times places too much emphasis on this point to the exclusion of other, equally important aspects of his work. In addition, Rundle's commitment to reclaiming Fuller on Fuller's own terms amounts to “a collection of reminders about what the commitments of Fuller's jurisprudence actually are”.Footnote 102 While this is undoubtedly an important and valuable task, the critique ultimately fails to acknowledge how Gardner's contribution might generate considerable scope to expand upon Fuller's analysis through an increased emphasis on the rights which are implicit within a Fullerian account of the rule of law. This lens could arguably be used to show that any conception of the rule of law will necessarily imply respect for the rights of legal subjects qua legal subjects.
Fuller's theory actually says very little about means; it says quite a lot about the form of legal norms, the ways in which restrictions on form affect and limit the ends which might legitimately be pursued through law, and the internal coherence and integrity of a legal system as a whole, but very little about legal means. Furthermore, even if the rule of law were primarily seen as a means to an end, this does not mean that it would not pass judgment on legal ends. A hammer is a very efficient tool for some purposes and completely useless for others. Inherent in the instrumentalist conception of the law lies a commitment to recognising that the rule of law does in fact stipulate certain ends as appropriate legal ends and certain ends as inappropriate legal ends.Footnote 103 Legal form may simply be unsuited to or incapable of guaranteeing the attainment of certain ends.
A system of law is never neutral with regards to the ends that can be pursued through it; many ends can never be attained through the use of law. Legality limits the range of possible ends that can be classed as legal ends. Even a purely instrumentalist conception of law must recognise that the very instrumentality of law ensures that certain ends or goals are simply incompatible with that instrument. The judgment of those ends is not necessarily left to “external” standards; an instrumentalist conception of law may reject internal standards grounded in morality, but it necessitates standards grounded in efficacy. Such an approach would evidently reject necessary connections between law and morality, but it cannot hold fast to the contention that legal ends are judged solely and exclusively by external standards.
Nevertheless, it must be stressed that the rule of law is not instrumental: it is not exclusively a means to an end. The rule of law is always a theory which weaves a path through form and function, resisting any sharp distinction between either and emphasising their mutually transformative nature. Maintaining a neat dichotomy between formal and substantive, or modal and functional theories of the rule of law only serves to prop up a distinction which would be better abandoned in favour of a more nuanced account which recognises that these concepts are so intimately connected that they make poor foundations for conceptual or theoretical distinctions. Once we recognise that the rule of law renders these concepts unintelligible without reference to each other, it becomes clear that any distinction between formal and substantive, or modal and functional theories is unsustainable.
VII. Legality and Justice
Returning to the three key distinctions between formal and substantive conceptions of the rule of law, all that remains is a distinction between conceptions of legality which guarantee the justice of all laws and those which can account for some unjust laws. Indeed, this appears to be the real distinction relied on by Craig in his analysis. In rejecting an interpretation of Dicey's theory as substantive, he argues that “Dicey was under no illusion that all English laws were substantively just”.Footnote 104 In particular, Craig stresses that Raz and Dworkin agree that “the adoption of a fully substantive conception of the rule of law has the consequence of robbing the concept of any function which is independent of the theory of justice which imbues such an account of law”.Footnote 105
If this is all that distinguishes some conceptions of the rule of law from others, then it makes little sense to conceive of Fuller's or Dicey's conceptions as in any way “formal”. There is evidently a commitment to an ideal of good law-making which limits the content of legal rules and entails respect for the fundamental rights of the legal subject. What remains is a jurisprudential distinction concerning the nature of law, a distinction which is broadly anti-positivist insofar as one accepts a necessary connection between law and the rule of law. On the one hand, there is an interpretative conception of legality which “does not distinguish … between the rule of law and substantive justice”.Footnote 106 On the other, there is a concept of legality which leaves room for rule of law compliant laws which are nevertheless unjust. Indeed, adherence to the rule of law as Fuller or Dicey envisaged it does appear to leave room for substantively unjust laws even if it might guarantee some basic degree of justice and fairness. Returning to the words of Lord Steyn, the rule of law, on this conception, “enforces minimum standards of fairness, both substantive and procedural”.Footnote 107 One can embrace a conception of the rule of law which is intimately connected to standards of justice without fully committing to a Dworkinian account which necessarily guarantees the “goodness” of all laws. In a similar vein, honesty can be a virtue, even if being honest might be compatible with great cruelty.Footnote 108 The fact that being honest does not guarantee the substantive goodness of one's conduct does not invalidate its status as a virtue.
It seems that Craig's analysis cannot accommodate such a conception, however. His idea of what makes a theory “substantive” is intimately connected to this particular jurisprudential commitment; as if a conception of the rule of law which affects legal content and respects legal rights necessarily also embraces this broader natural law position that law, properly understood, can only be morally good law. Fuller would clearly reject such a contention and would insist that legality contains its own moral standards even if it remains neutral across a vast array of legal ends.Footnote 109 There is room for external moral standards on this conception, even if the principles of legality demand adherence to some basic conceptions of justice which arise internally.
Raz argues that “[i]f the rule of law is the rule of good law then to explain its nature is to propound a complete social philosophy. But if so the term lacks any useful function”.Footnote 110 I suggest that, if the rule of law is a political ideal, then it must be the rule of good law, but that this does not necessarily require a commitment to the claim that legality and justice collapse into one another. Many conceptions of the rule of law entail respect for fundamental rights without resulting in this much broader commitment to a complete social philosophy.Footnote 111 A rule of law compliant legal system may leave open the choice between differing conceptions of justice even if it must commit to respecting the legal subject as a moral entity with certain minimum rights to the protection of law.
If, as Raz and Gardner argue, the rule of law is a political ideal, then it is by definition dependent on notions of good law (that which conforms to the ideal) and bad law (that which fails to conform to the ideal). Raz resists recognising a connection between the rule of law and theories of justice because he thinks it will rob the rule of law of any independent value. However, the rule of law can be of no value unless it is a commitment to some notion of good law, even if that good law could be compatible with great inequity.Footnote 112 As Allan observes, Raz is torn between two paths: the rule of law as a mere tool and the rule of law as a safeguard. These are ultimately contradictory perspectives.Footnote 113
VIII. Conclusion
The rule of law necessitates analysis of the products and practices of legal officials and an evaluation of those products and practices to see if they are in conformity with this political ideal. The rule of law, even on the thinnest account, is a nuanced confluence of divergent principles which manifest within law subtle pressure towards integrity as a cohesive whole.Footnote 114
In light of these conclusions, it seems quite improbable that we can maintain a neat distinction between formal and substantive conceptions of the rule of law. A rule of law compliant legal system is one which manifestly protects the fundamental rights of legal subjects and which precludes certain legal ends from being classed as examples of the law's legitimate functions. Inherent within the concept of a legal subject lies the rights entailed by fidelity to the rule of law. Inherent within the concept of a legal system lies restrictions on governmental power and duties to protect and respect the rights of legal subjects. This ensures that the rule of law cannot be portrayed as content-neutral, a mere tool to be put to whatever purposes a political ruler might wish. Fidelity to the rule of law may not guarantee that all products of that system are benign and morally acceptable, but at the very minimum, it ensures that some rights are protected, and some restrictions are placed on legal content and legal ends. The rule of law, on this conception, is always simultaneously concerned with the bare minimum threshold conditions of legality as well as aspirational ideals which serve as “distinct standards by which excellence in legality may be tested”.Footnote 115
Attempts to distinguish between formal and substantive or modal and functional conceptions of the rule of law inevitably fail. These dichotomies have shifting baselines, are completely context-dependent and even the multitude of meanings that could be attributed to those terms do not reflect the nature of the theories they purport to explain. The form/substance dichotomy, and its re-imagining in the modal/functional dichotomy, offer us nothing more than confusion and obfuscation. If Gardner is correct that the rule of law speaks to legal content and implies the existence of fundamental rights, then his characterisation of it as modal and instrumental is unsustainable.
