Expediency, Legitimacy, and the Rule of Law: A Systems Perspective on Civil/Criminal Procedural Hybrids

And that is exactly why the legitimacy of law is questioned time and again—acutely or hopelessly, out of frustration or anger, full of value-perspectives that are beside the point for law.

Niklas Luhmann (2004: 261)

No good society can be unprincipled; and no viable society can be principle-ridden…

Our democratic system of government exists in this Lincolnian tension between principle and expediency.

Alexander M. Bickel (1962: 64, 68)

Abstract

In recent years an increasing quantity of UK legislation has introduced blended or ‘hybridised’ procedures that blur the previously clear demarcation between civil and criminal legal processes, typically on the grounds of normatively-motivated political expediency. This paper provides a critical perspective on instances of procedural hybridisation in order to illustrate that, first, the reliance upon civil law measures to remedy criminal law infractions can raise human rights issues and, second, that such instrumental criminal justice strategies deliberately circumvent the enhanced procedural protections of the criminal law. By conceptualising the rule of law as a structural coupling between the political and legal systems, and due process rights as necessary and self-imposed limitations upon systemic operations, this paper employs a systems-theoretical approach to critique this balancing act between expediency and principle, and queries the circumstances under which legislation contravening the rule of law can be said to lack legitimacy.

Introduction

There has been a marked increase in recent years in the quantity of legislation passed by the UK Parliament that provides for hybridised procedural approaches to specific legal issues. By ‘hybridised’ procedures, we mean those blended processes in either civil or criminal law that rely upon mechanisms normally associated with the other type, or those that omit procedural dimensions normally required by their own sort, and thus blur the lines between the civil and the criminal.¹ Many examples can be cited of the use of civil processes to target criminal behaviour, for example, in relation to anti-social behaviour (e.g., Anti-Social Behaviour Orders (ASBOs) under the Crime and Disorder Act 1998, s. 1, recently replaced by the Injunction under Part 1 of the Anti-social Behaviour, Crime and Policing Act 2014²), domestic violence (e.g., Domestic Violence Protection Notices (DVPNs) and Domestic Violence Protection Orders (DVPOs) under the Crime and Security Act 2010, ss. 24–29), forced marriage (e.g., Forced Marriage Protection Orders, as inserted in the Family Law Act 1996 by the Forced Marriage (Civil Protection) Act 2007), sexual offences (e.g., Sexual Harm Prevention Orders and Sexual Risk Orders introduced under the Anti-social Behaviour, Crime and Policing Act 2014, replacing other civil orders under the Sexual Offences Act 2003), serious/organised crime (e.g., Serious Crime Prevention Orders under Part 1 of the Serious Crime Act 2007 and Civil Recovery Orders under Part 5 of the Proceeds of Crime Act (POCA) 2002), and terrorism (e.g., Control Orders introduced under the Prevention of Terrorism Act 2005, then abolished in 2012 and replaced by Terrorism Prevention and Investigation Measures (TPIMS)).

The reasons for this increasing procedural hybridisation are themselves context-specific and thus variable, but one common aspect across each of the illustrations listed above is a degree of normatively motivated political expediency. Control orders were initially conceived of as hybrid administrative measures with a starring role in the ‘War on Terror,’ for example, while the hybrid nature of DVPOs results from the policy goal of combating more effectively domestic violence against women and girls, their conception as expressly protective, not punitive. This value-based approach can also be seen in terms of civil recovery, the policy position behind which is that crime should not pay, and that assets arising from criminal activity ought to be forfeited. The inherent instrumentality behind the introduction of each of these hybrid measures is overt, although not in itself problematic. What is concerning, however, is the manner in which such legislative privileging of expediency over considerations of human rights and due process is becoming increasingly normalised.³

The aim of this paper is to draw attention to the rising frequency of such procedural hybridisation (see Ashworth and Zedner 2008: 29–31), and to demonstrate that hybrid orders illegitimately circumvent criminal law procedural protections.⁴ We employ a systems-theoretical approach to critique what, we argue, is a prioritisation of expediency over principle, and engage with the following fundamental question: can a legislative provision, properly passed according to the requirements and procedures of the enacting Parliament but which contravenes those higher legal principles comprising the rule of law, lack legitimacy? These issues are scrutinised in terms of the rule of law, which we conceptualise not only as a composite of legal standards, normative aspirations, and quality benchmarks but also as a structural coupling between the political and legal systems; we rely upon this insight to analyse the introduction of these hybrid orders and procedures. Our conclusion will be that, in spite of their undisputed legal validity, their effective circumvention of rule of law standards places them squarely in a position of questionable legitimacy. The first section of this paper will articulate what we understand by ‘legitimacy’ in this context, with specific discussion of this composite group of ‘rule of law’ standards in systems-theoretical terms, while the second will provide a comprehensive analysis of civil/criminal procedural hybridisation, and will present our case study of civil recovery. The third section will reintroduce the core question and argue the thesis that, in spite of their undisputed legal validity, such hybridised measures lack legitimacy because they exceed both the legal system’s self-imposed limitations and those resulting from its structural couplings⁵ with the political system. Our conclusion presents the increasing reliance on procedurally hybrid approaches as an over-emphasis upon expediency at the expense of principle.

It should be noted here that this paper employs a systems-theoretical perspective in leading its principal argument. This perspective provides a fresh insight into both ongoing debates on procedural hybrids (see, e.g., Bronitt and Donkin 2012; Zedner 2007) and legal-theoretical discussions of the rule of law. The advantages of a systems perspective on procedural hybridisation lie in how the theory’s emphasis on functional differentiation and the boundaries between systemic operations highlights issues often left unseen by conventional analyses. Systems theory draws clear dividing lines between the concepts at the heart of this analysis—validity and legitimacy, the legal and the political—and this clarity provides an invaluable foundation for critique. This study also contributes usefully to the further development of systems theory itself, as the issues raised by considering procedural hybrids test both its positivistic⁶ and descriptive nature, not least by presenting it with the obstacle of normativity. Indeed, it is with normativity that we will begin, for it is in terms of two specific normative dimensions that this analysis establishes its parameters relative to the competing concepts of expediency and principle. These normative dimensions can be articulated in terms of our selected case study, namely civil recovery under POCA 2002 Part 5. Civil recovery under POCA perfectly illustrates the contentious nature of civil/criminal hybrid procedures in that civil recovery allows the State to go after ‘criminal’ proceeds albeit by circumventing enhanced procedural protections of the criminal process.⁷

The first normative dimension to consider, as mentioned above, is the value-based motivation behind the adoption of procedural hybrids according to policies that can be cited as undoubtedly politically expedient in character. This was clearly reflected in the build up to POCA by then—Prime Minister Tony Blair, who in September 1999 stated that ‘we want to ensure that crime doesn’t pay. Seizing criminal assets deprives criminals and criminal organisations of their financial lifeblood’ (Performance and Innovation Unit 2000: 13). Civil recovery is presented as a key strategy in the fight against serious crime. Initiated as a result of perceived inadequacies of existing criminal processes in controlling high-level and high-value organised crime, civil recovery enables the seizure of ‘criminal’ proceeds in the absence of a criminal conviction and on a reduced standard of proof.

This normative stance—that crime should not pay—is unlikely to prove particularly controversial, or to give rise to much political contestation. If someone has committed a criminal offence then that person should undoubtedly be denied the benefit of that offence. It is when focus shifts to the mechanics of implementation, however, that concerns arise (see Gledhill 2011: 81). This very issue is the second normative dimension of this analysis, namely the tension created by the juxtaposition of the realisation of the stated policy goals (see Performance and Innovation Unit 2000) with the requirements of due process, or, rather, the apparent conflict between the goals of controlling high-level, high-value criminal activity and ensuring the adequate observance of the alleged perpetrator’s civil and political rights (Ivory 2014). This paper submits that, in their effective bypassing of enhanced procedural protections, this hybrid measure is contrary to the rule of law and thus lacking in necessary legitimacy (on the point that the legislative remedy of civil recovery has gone too far in its attempt to remedy an existing inadequacy in the law, see Hendry and King 2015). Although this critique may appear prima facie to be a legal-theoretical one, it is important to note that this opens civil recovery up to challenge on the grounds that it violates due process rights that are inherent in the criminal process.⁸

Legitimacy and the Rule of Law

Before we proceed with this argument, a number of concepts require further explicit attention, not least that of legitimacy. Niklas Luhmann is not alone in despairing of the concept of legitimacy (2004: 261),⁹ the myriad uses and conceptions of which mean it can rightfully be considered an essentially contested concept (Gallie 1995), even if within certain fields of study there does exist a tentative consensus.¹⁰ A concept perhaps more familiar to politics than law, political legitimacy can be understood in a contractarian vein as the popular acceptance of authority, typically an established system of democratic government according to the twin Lockean principles of consent of the governed and majority rule. The contestation arises when we look beyond this basic conception and attempt further specification or refinement, such as considering whether political legitimacy is premised upon descriptive or normative grounds, or whether it has procedural or substantive requirements. Such arguments cannot be premised upon an objective foundation but rather cite contingent political values. Indeed, if we shift our viewpoint to adopt a more expressly legal perspective, then it becomes apparent that, while legitimacy is an important concept for law, the task of furnishing it with content is one that rests not with law, but with political and moral philosophy. The reason for this, as Weinberger (1999: 347, emphasis added) explains, is that:

Criteria of legitimacy are value-criteria which depend on philosophical and political opinions. The criteria may concern the content of the law—this is the position of natural law theorists—or they may postulate some forms of generation and/or some kind of acceptance of the law as criteria of legitimacy. […] Legitimacy is not an objective feature of valid law but a valuation based upon presupposed political convictions. […] Legality of legal processes in the broad sense of the dynamic theory of law is interpreted as a sign of legitimacy, but in fact the judgment about legitimacy is distinct from the proof of legality.

This distinction is important for the purposes of this analysis, as it draws attention to the way in which legitimacy is dependent on criteria that are markedly and essentially extra-legal. Indeed, while legal theoretical discussions of the content, source(s), pedigree, or reception of the law may inform (the selection of) such criteria, issues of legitimacy are wholly separate from those of legal validity. In other words, while it is certainly possible to establish legitimacy as a condition for or objective corollary of legality, this duty belongs not to law but rather to politics. And this takes us to the crux of the matter, which is that legitimacy can only be introduced to law through politics, a state of affairs that even in democratic situations gives rise to undeniable contingency. The next section submits that the vehicle of this introduction is the rule of law.

The Rule of Law

The rule of law is not a straightforward concept either to employ or to rely upon. As Tamanaha (2004: 4) observes, the rule of law ‘stands in the peculiar state of being the preeminent legitimating political ideal in the world today, without agreement upon precisely what it means’. The textbook definition at least provides a starting point.

[T]he rule of law concerns the relationship of the government to the law [… It] is both a legal rule and a political idea or principle of governance comprising values that should be reflected in the legal system and should be respected by those concerned in the making, development, interpretation and enforcement of the law. (Turpin and Tomkins 2007: 76)

Immediately apparent within this definition is the concept’s duality. The rule of law is simultaneously a legal rule and a value-laden political principle that is, importantly, determinative of all aspects of law’s operation. More than this, however, is the manner by which the rule of law embodies the law’s legitimacy, even although this legitimacy is wholly contingent on the context of the political circumstances and indeed values at hand. Legal legitimacy is thus a black-box concept—an empty vessel to be filled with animating (political) values. The benefit of this is evident: by keeping any and all consideration of values separate from the legal, it precludes a slide back into well-worn debates on law’s normativity or otherwise. While these foundational values can be both contingent and indeterminate, their existence cannot be disputed; furthermore, their institutionalisation is a vital dimension of their effectiveness. The rule of law in this regard ‘provides an institutional morality of rule-following, of rule formation, and of rule implementation and interpretation, which speaks not so much to the content of jurisprudence, as to the way in which they must be framed so as to sustain rather than subvert the moral basis of political and legal order’ (Dyzenhaus 2001: 498). Also notable here is that this more procedurally minded approach avoids drifting into normative matters by maintaining its focus on the conditions of legal legitimacy, as opposed to its content.

To be clear here, we are required to separate out what Paul Craig in his seminal article refers to as the formal and substantive conceptions of the rule of law:

Formal conceptions of the rule of law address the manner in which the law was promulgated (was it by a properly authorised person, in a properly authorised manner, etc.); the clarity of the ensuing norm (was it sufficiently clear to guide an individual’s conduct so as to enable a person to plan his or her life, etc.); and the temporal dimension of the enacted norm. (was it prospective or retrospective, etc.). Formal conceptions of the rule of law do not however seek to pass judgment upon the actual content of the law itself. They are not concerned with whether the law was in that sense a good or a bad law, provided that the formal precepts of the rule of law were themselves met. Those who espouse substantive conceptions of the rule of law seek to go beyond this. They accept that the rule of law has the formal attributes mentioned above, but they wish to take the doctrine further. Certain substantive rights are said to be based on, or derived from, the rule of law. The concept is used as the foundation for these rights, which are then used to distinguish between “good” laws, which comply with such rights, and “bad” laws which do not. (Craig 1997: 467)

In terms of this dichotomy, our analysis focuses explicitly upon the rule of law’s formal conception with specific reference to its procedural dimension; although formal conceptions of the rule of law can be premised upon abstract substantive moral considerations, it is nonetheless quite a leap to make the rule of law co-extant with specifically articulated substantive provisions (Craig 1997: 481–482, critiquing Allen 1993: 46). Whereas the rule of law is recognised as comprising a ‘whole set of normative aspirations and quality benchmarks’ (Nickel 2006: 169), the normative content of these benchmarks necessarily remains always-already undetermined. Although it cannot be denied that ‘the formal conception of the rule of law, and the desire to keep legal questions separate from broader uses of political theory in deciding what the context of the law is, fit naturally together’ (Craig 1997: 477), it is as a particular result of both this paper’s reliance upon systems theory, inherently positivist in its scope and ambit,¹¹ and its focus upon procedural considerations that this approach is adopted.

Systems Theory: Closure, Coding and Structural Coupling

Some of the main tenets of systems theory—that is, the theory of autopoietic social systems as developed by Luhmann (see, for example, Luhmann 1985, 1995, 2004, 2012, 2013) and elaborated upon extensively by Gunther Teubner (see 1993, 1998, 2006)—should be introduced and explained at this point. According to Luhmann, modernity marked the arrival of societal differentiation on the basis of function. In contrast to the segmentary and stratified pre-modern forms of social ordering, modern society is decentred in the form of multiple self-referential function systems, such as law, politics, the economy, religion, science, family, education, art, and so on. Such differentiation generates stability within the social sphere by creating ‘broad-ranging societal conditions of liberty, pluralism and autonomy, which are usually construed as the features and pre-conditions of political democracy’ (Thornhill 2006: 89).¹² Each social system has a different function: while the function of the legal system within society can be established as the stabilisation of normative expectations over time, including the production and maintenance of counterfactual expectations in the face of their repeated disappointment, the economic system regulates scarcity by meeting future needs for material goods (Beckert 2009: 212), and the political system provides the conditions for the exercise of power.¹³

The relation between function and environment is, as Andreas Philippopoulos-Mihalopoulos (2010: 68) points out, one of ‘complexity reduction.’ A system distinguishes itself from its environment, which comprises all other systems excluding itself, by drawing a distinction between those communications that are relevant to it with reference to its own binary coding. For example, while the unitary legal system communicates in terms of lawful/unlawful, politics relies upon the binaries of government/opposition and government/governed, science on the code true/false, and the economic system employs payment/non-payment.¹⁴ Meaning is, therefore, system-specific—indeed, it is a core feature of autopoietic communication that information does not cross from environment to system but is rather internally constructed by the system. Not only does this construction of a systemic boundary according to the system/environment distinction facilitate both the system’s autonomy and operational closure, it also enables the system to ignore those communications¹⁵ as irrelevant to its own operations, and thus to reduce complexity.

In Luhmann’s description of the operation of systems, communications are systemic elements that produce and reproduce themselves and their environment. Communication occurs ‘by splitting reality through a highly artificial distinction between utterance [the condition of self-reference] and information [external reference], both taken as contingent events within an ongoing process that recursively uses the results of previous steps and anticipates further ones’ (Luhmann 1992: 1424). Every system operates recursively with and in terms of communications that are meaningful to it—political communications, legal communications, and so on—thus maintaining its own self-reference. Through this distinction between the utterance and the information within a communication, meaning becomes the ‘simultaneous presentation… of actuality and possibility’ (Luhmann 2002: 83),¹⁶ which is to say that the system’s cognitive openness to its environment remains in accordance with the originary distinction of its binary code, the system-internal structure that determines connection and attribution (Luhmann 1992: 1428).

Despite their operational closure, autonomy, and self-reference, however, different systems can become ‘structurally coupled.’ Systems are structurally coupled when they presuppose certain features of their environment on an ongoing basis and rely upon these structurally (Luhmann 2004: 382). By way of illustration, we can point to the national constitution as a coupling between the systems of law and politics, taxation as structurally connecting the political and the economic systems, and both contracts and property as couplings between the economic and legal systems. Importantly, these couplings are neither interactions nor intersections but rather involve simultaneous, analogical coordination (Luhmann 1992: 1432) whereby these structures become ‘co-evolutionary without becoming common’ (Philippopoulos-Mihalopoulos 2010: 132). When systems are structurally coupled, mutual perturbations and irritations¹⁷ occur that influence systemic structural development—a system’s ‘cognitive openness’ facilitates the coordination of a system’s ongoing self-referential processes with those occurring in its environment (Luhmann 1987: 20).

It is as such a structural coupling between the legal and political systems that we conceptualise the rule of law. Indeed, a closer look at its role makes this more readily apparent because, as defined above, the rule of law is simultaneously ‘a legal rule and a political idea’, establishing standards that ought to be applied and maintained within both function systems (Turpin and Tomkins 2007: 76). Even under the formal definition as outlined by Craig (1997), the rule of law comprises law’s creation, promulgation, and application, including certain procedural standards and limitations¹⁸; as such it can be understood systems-theoretically as establishing a contingent connection between the two function systems of politics and law in the form of mutually constituted procedural thresholds and requirements.¹⁹ This conception of the rule of law is further informed by the distinction that systems theory draws between ideas of legitimacy and validity, or, rather, the ease with which it distinguishes the latter compared to the difficulties it experiences with the former. Their juxtaposition sets these issues in sharper relief.

Validity and Legitimacy in Systems Theory

Validity is the marker for the unity of the legal system and thus, as the symbol of the autopoiesis of its communications, the closest the legal system comes to a self-description of its operative function (Luhmann 2004: 122–123). In a passage reminiscent of HLA Hart’s positivist position regarding the justiciability of the rule of recognition, Luhmann (2004: 125) states that: ‘All law is valid law. Law which is not valid is not law. It follows that the rule that makes validity recognizable cannot be one of the valid rules. There cannot be any rule in the system that regulates the applicability/non-applicability of all the rules of the system’. It is in this manner that the legal system organises its own closure and functioning, premised upon the distinction between external reference and self-reference. Legal validity is the application of the coding lawful/unlawful to systemic operations and the means by which the legal system reproduces normative expectations with reference to normative expectations (Luhmann 1992: 1427). As a legal-systemic eigen-value—namely, a value constituted by the ‘recursive performance of the systems’ own operations [… which] cannot be used anywhere else’ (Luhmann 2004: 124)—validity itself is, importantly, without either intrinsic value (Philippopoulos-Mihalopoulos 2010: 83) or its own normativity (Hendry and King 2015: 407).

Legitimacy, by contrast, is a far less straightforward notion for the legal system to accommodate. Indeed, Luhmann’s core explanation of legal legitimacy is remarkably arcane: he describes the structure of the legitimacy of law as ‘a mixture of the cognitive/normative expectation of normative expectation of cognitive expectation of normative expectation’ (Luhmann 1985: 204). His emphasis is on the unavoidable contingency of legal validity—in light of the lack of a foundational moment, positivist source, or transcendent value upon which to rely, the law must resort to ‘legitimising itself through the concept of validity’ (Philippopoulos-Mihalopoulos 2010: 87). Yet the law’s very positivisation means that, for the legal system, validity is nothing other than the re-production and re-entry²⁰ (Teubner 2009: 11; Luhmann 2004: 105) of law’s difference into the system, just as norms are nothing other than system-internal creations that provide it with necessary decisional criteria. Systems theory thus appears to confirm what we already suspected, namely that establishing legal legitimacy requires reliance upon a value relation that is necessarily external to the legal system (Weinberger 1999: 347).

Most interesting, however, is Luhmann’s rejection of the classical idea that legitimacy is the objective corollary of legality. Not only does he dismiss this position but he also argues that legitimacy is, instead, the formula of contingency for politics. As Thornhill (2006: 83) explains, ‘the legitimate political system is a political system which has woven a convincing web of legitimacy out of its own, utterly contingent, operations’, and thus has no need to rely on the legal for any constitutive input. The difference between the legal and political systems here is striking: whereas the legal system is hamstrung by its own inherent lack, the political system in effect behaves as if it were legitimate, creating system-internal communications ex nihilo and ordering them in a manner both predictable and coherent. In this manner, it is subsequently able to rely upon such self-reference to the extent that these systemic operations become externally meaningful (Thornhill 2006: 82–83, 95), for example: the requirement of a specific margin for an election victory, or the requirement for a referendum to underpin constitutional amendment. The conditions under which political systems can claim legitimacy are hugely variable and, moreover, arguably contextual as well as contingent. Similarly notable here is the difference between Luhmann’s conception of political legitimacy and the Lockean form cited earlier: although government is legitimate for Luhmann when its policies and legislation are accepted as such by its citizens, there is no objective or substantive ‘content’ requirement for either the behaviour of that government or the substance of its policies that determines its legitimacy in this regard. To the extent that political legitimacy is dependent upon meeting conditions or attaining thresholds, these exist only insofar as they are established by the political system as conditions upon its own operations.

And this is the rub. Since there are no pre-existing foundations upon which politics can base its decisions, with the effect that it is required to conceive and construct ‘the contingent parameters of its own legitimacy and so establish on its own the legal order which supports its decisions’ (Thornhill 2006: 95), any substantive content or procedural requirements operating as restrictions upon the operations of the political system must have been, inescapably in this regard, deliberately selected as systemic self-limitations. Such restraints are curbs upon the potential excesses not only of the political system, of course, but also the legal system, which, by virtue of their structural coupling through the rule of law, finds its operations and programmes subject to its own system-internal understanding of these restrictions. This situation manifests within the legal system in the form of conditional programming that regulates those procedures to which the system must adhere, both at the level of passing legislation (see, e.g., Raz (1979) on the formal conception of the rule of law and Fuller (1969) on a procedural understanding of law) and in terms of the implementation of specific legislative provisions (see, e.g., Tyler 2006b). The function systems are held in tension by their structural coupling—the rule of law—and as such are each restricted by their own respective reconstructions of these limitations. Indeed, Lord Woolf (1995: 68) conveyed the gist of this notion neatly when he stated: ‘As both Parliaments and the courts derive their authority from the rule of law so both are subject to it and cannot act in a manner which involves its repudiation.’ Furthermore, these self-imposed conditions of political legitimacy provide the necessary value-criteria upon which the legal system’s operative validity unwittingly and yet fundamentally depends. Systems theory’s positivist heritage shines through once more in the form of this ‘norm pyramid,’²¹ where the validity of a norm is premised upon the validity of a higher norm in the hierarchy. Teubner (1997: 768) draws further attention to this issue in his own discussion of law’s hierarchy of rules: the lower rules are legitimated by higher ones, the highest of all being the nation state constitution, which establishes ‘democratic political legislation as the ultimate legitimation of legal validity’.

This returns us to our core investigation, namely, whether legislative provisions properly enacted can ever lack legitimacy. We submit that this can be the case where such provisions fail to comply with the rule of law, understood in the formal sense of the term, and argue this position employing the example of civil/criminal procedural hybrids in general, and the case study of civil recovery under POCA 2002 in particular. The character of and motivations behind civil/criminal procedural hybrids will be the focus of the next section.

Civil/Criminal Procedural Hybrids

Many explanations for the adoption of hybridised processes to tackle essentially criminal behaviour have been advanced (for discussion in relation to ASBOs, see Ashworth and Zedner 2014: 78ff). First, it has been suggested that the criminal law alone is inadequate in tackling certain forms of harm. A second, related, explanation is that there can be evidential benefits in resorting to civil processes, for example, the avoidance of the strict application of the hearsay rule in criminal proceedings.²² Ashworth and Zedner (2014: 80) draw attention to a third, more political, account: they note that many countries ‘have seen a strain of penal populism in government statements and in legislative initiatives that are often presented as measures of public protection’ (see Dzur 2012: 116). A fourth explanation concerns policy transfer: civil measures have been used both in other jurisdictions and in other areas of domestic law (Ashworth and Zedner 2014: 81–82). Uniting all of this reasoning, however, is a sense of expediency—increasingly more common both within and outwith the UK, procedural hybrids are being first introduced and then employed with evident instrumentality. This paper not only queries the general legitimacy of such measures but uses the case study of civil recovery to illustrate their challenge to those rule of law standards already discussed.²³

The controversial civil recovery powers under Part 5 of POCA stemmed from concern that those engaged in organised criminal activity (so-called ‘Mr Bigs’) were beyond the reach of the criminal law; it was thought that the conventional approach of investigation, prosecution, conviction, punishment was not working. The ‘solution’ was this civil approach to seizing ‘criminal’ property without the need for criminal conviction and on the civil standard of proof—the balance of probabilities. There are many arguments advanced in favour of using a non-conviction based approach to seizing assets, including (1) to take the profit out of crime; (2) to disgorge property acquired through criminal acts; (3) to act as a deterrent; (4) to deprive wrongdoers of financial resources for future criminal activity, thereby acting as a form of prevention; (5) to disrupt criminal organisations; (6) to enable property to be restored to victims of crime; (7) to protect the community and to demonstrate that law enforcement is making efforts to tackle crime; and (8) to encourage cooperation between different law enforcement agencies (primarily through ‘equitable sharing’ or ‘incentivisation’) (see Cassella 2013: 99). Proponents of civil recovery contend that it is a civil process and, as such, does not require criminal procedural protections, that civil recovery operates in rem (against the property) rather than in personam (against the individual). Proponents also claim that civil recovery represents a necessary, proportionate response to growth in organised crime. It is often claimed that civil recovery abides by all human rights norms and, furthermore, that there are in-built safeguards to offer protection against abuse (see, for example, Cassella 2013, 2008; Cassidy 2009; Simser 2009; Kennedy 2005).²⁴ As we will argue, the courts have been overly acquiescent in accepting such claims.

The adoption of civil recovery in the UK followed in the footsteps of other jurisdictions that already had similar powers (see Kennedy 2006), but resort to civil law tools to tackle criminal activity was not a new phenomenon in the UK—as outlined at the outset of this article, the UK government has a long history of using civil tools in this manner. Zedner (2009: 81) discusses civil preventive measures which ‘circumvent the protections of the criminal process by operating in parallel systems of questionable justice: according to the less exacting requirements of the civil process or enforced via hybrid systems in which breach of civil orders result in criminal sanctions’. Although her discussion concerned civil orders such as ASBOs, control orders, and serious crime prevention orders,²⁵ a similar complaint can be made in relation to civil recovery under POCA. While there is no threat of imprisonment, civil recovery orders permit deprivation of property in civil proceedings on the grounds that that property constitutes proceeds of crime—another example of a ‘parallel system […] of questionable justice’ to circumvent criminal procedural protections.

Civil Recovery and POCA 2002

POCA 2002 was introduced specifically to remedy perceived failures of existing criminal justice processes in tackling serious and organised crime. Most problematic in this regard is the supposedly hierarchical structure²⁶ of such criminal activity, where:

[M]any major criminal figures have become untouchable by criminal prosecution. They organise or finance the criminal activity of others and profit from the results, but remain remote from the commission of particular crimes. That often makes it impossible for law enforcement authorities to build a case against them. (Proceeds of Crime Bill, HC Deb, 30 October 2001, vol. 373, c. 760, per Minister Denham)

It is this remoteness from the ‘coal-face’ that causes most difficulties for conventional law enforcement. Where organised crime syndicates do operate hierarchically, the main beneficiaries (i.e., the ‘organisers’) of their activities are said to be insulated against successful police investigation, prosecution, and conviction. What is highlighted here are the perceived inadequacies of conventional criminal justice tools in coping with hierarchically organised criminal activity, and the extent of the problems allegedly²⁷ faced by law enforcement. Indeed, when one considers that it is the primary function of the legal system to ‘establish and stabilize societal expectations through the handling of disappointment’ (Philippopoulos-Mihalopoulos 2010: 71; Luhmann 1985, 2004) and thus to provide constancy within society, this supposed ‘failure’ of the criminal law has the same effect as a genuine failure. What, then, is the legal system to do in such a situation? The options available are made clearer by considering further the idea of legal-systemic expectations.

The legal system relies upon expectations as a means of controlling normativity, reducing complexity, and eliminating contingency. Expectations are generated on the basis of norms that, even in the event of their disappointment, remain unaffected—in this manner, existing normative expectations are stabilised on a counterfactual basis. This is a particularly familiar situation within the field of criminal law where, for example, criminal behaviour ought to be subject to legal sanction, wrongdoers ought to be punished, crime should not pay, and so on. In spite of their repeated disappointment, these normative expectations stay stable, and illegal conduct remains exactly that. Although necessary in terms of reducing complexity by limiting the number of possible selections generated by conditions of uncertainty, the effect of this systemic normative closure is that disappointments do not lead to the legal system learning from its previous operations. For systemic learning to occur, the legal system is instead reliant on its cognitive expectations, which is to say, on its openness to changing factual conditions within society. It is through this cognitive openness that the law is able to adapt—the legal system’s programming adjusts both to deal with changes in its environment and to become more successful in the realisation of its primary function. Such programming is guidance for the operation of the system’s binary coding; it stipulates the conditions under which the coding lawful/unlawful can be applied, and is the means by which the legal system modifies itself to recognise that, for example, something that was lawful is now unlawful, or vice versa.

In this regard, the introduction of these hybrid civil recovery powers can be understood as an endeavour to stabilise the normative expectation that the legal system will counteract and ‘punish’²⁸ criminal and thus illegal activity within society. At the same time, however, and through the new procedures introduced, the legislation adjusts the system-internal programming that guides the lawful/unlawful distinction. This situation results directly from the criminal law’s perceived inadequacy to deal with the particular challenges of serious and organised crime: the importance of this attrition in terms of the law’s self-regulation cannot be overstated. This alleged failure of law, this apparently continuing disappointment of its normative expectation over time, was what gave rise to these wholesale procedural changes. In its adoption of such a hybrid approach, we see the supposed failure of law being concretised through procedure (Hendry and King 2015).

The next section scrutinises the jurisprudence of civil recovery and analyses the effects of these in rem measures. At this juncture, however, it is worth emphasising the following: as a criminal justice strategy, civil recovery is overtly normative in its stance that criminal activity should not pay, is instrumentally engaged in undermining the profit incentive of organised criminal behaviour, and—by using civil rather than criminal procedures—deliberately prioritises this motive over compliance with procedural safeguards inherent to the criminal process in the UK.

Civil Recovery in the Courts

As the primary legislation providing for asset recovery, POCA makes provision for both post-conviction confiscation of assets (POCA, Parts 2–4) and non-conviction based confiscation of the proceeds of unlawful conduct, labelled ‘civil recovery’ (Part 5).²⁹ Despite our focus resting upon the latter, distinguishing the two forms is vital: while actions under civil recovery (POCA, Part 5) do not require a criminal conviction, it is only subsequent to a conviction that a confiscation order (POCA, Parts 2–4) can be made. With such a post-conviction confiscation order, all of the enhanced procedural protections of the criminal process apply at the criminal trial, including the presumption of innocence, the standard of proof beyond reasonable doubt, and exclusionary rules of evidence, for example, the rule against hearsay. It is only post-conviction, at the confiscation hearing, that the standard of proof employed is the civil one—the balance of probabilities—and there the rules of evidence are akin to those in a sentencing hearing (Alldridge 2014a: 173–174). By contrast, a civil recovery order can be granted even in the absence of criminal conviction, and ‘whether or not any proceedings have been brought for an offence in connection with the property’ (POCA, s. 240(2)), including instances where defendants have been acquitted in criminal proceedings (Taher 2006)³⁰ or even where a conviction has been quashed (Olden 2010). The civil rules of evidence apply, meaning that both character and hearsay evidence can be admitted, as can evidence obtained by improper means (Olden 2010), and the standard of proof is on a balance of probabilities (POCA, s. 241(3)). Finally, a confiscation order operates in personam, while a civil recovery order operates in rem.

Part 5 of POCA enables the ‘enforcement authority’ to recover, in civil proceedings, property that is, or represents, property obtained through unlawful conduct (POCA, s. 240).³¹ Given the perceived inadequacies of existing criminal legal processes and the attendant difficulties in securing criminal conviction, post-conviction confiscation was often impossible—the result of this was that the benefits of non-conviction-based approaches in this purely instrumental regard were increasingly proclaimed. Nevertheless, and although these processes have made strides in terms of efficacy, it is hard not to rue their introduction as a privileging of instrumentality over considerations of due process, not least because there are manifest reasons for the enhanced procedural safeguards inherent in the criminal process. As Roberts and Zuckerman (2010: 247) articulate in relation to the heightened standard of proof, ‘[t]his asymmetrical standard is not a natural or inevitable incident of allocating probative burdens; it is, rather, an additional commitment, over and above requiring the prosecution to prove guilt, to the presumption of innocence and its animating liberal philosophy of respect for persons’. By requiring this elevated standard of proof, the criminal process establishes itself as innately risk-averse,³² with a wrongful conviction being perceived as far worse than a wrongful acquittal.³³ Criminal law’s ‘asymmetric standard’ can be considered, therefore, as a prophylactic measure against an erroneous trial outcome (i.e., a false conviction) or, in systems-theoretical language, a ‘bulwark’ against systemic excess.³⁴ We consider this, first, concerning the ‘civil’ nature of the proceedings under Part 5 of POCA and, second, in relation to those due process concerns that arise as a result of this kind of hybridised procedure.

‘Civil’ and ‘criminal’ processes have traditionally been distinguished on the basis of a variety of factors, including inter alia: subjective/objective culpability, harm, the role of a prosecution authority, the extent of investigatory powers available to the State, evidential rules, punishment, and stigma (see Hall 1943; Ashworth 2000). Far from being a mere semantic distinction, in practice it means that while ‘civil’ processes operate under lesser burdens, processes designated as ‘criminal’ attract for the accused enhanced procedural protections as additional commitments. Civil recovery, as the name suggests, purports to be a civil process and, importantly, has been held not to be of a criminal character³⁵ in spite of its ‘potential for use as an uneasy and unsatisfactory substitute for the criminal process’ (Rees et al. 2011: para. 6.13). Indeed, the courts have been consistent on this position. Perhaps the leading case on this issue is Walsh where, on considering whether such proceedings are criminal for the purposes of Article 6 of the ECHR,³⁶ both Coghlin J (Re the Director of the Assets Recovery Agency, 2004) and the Court of Appeal (Walsh 2005) concluded that civil recovery proceedings under Part 5 of POCA should be classified as civil proceedings.³⁷ The three criteria established by the Strasbourg Court in Engel (1979–80) were: the domestic classification of the proceedings at issue, the nature of the offence in question, and the nature and severity of the penalty that may be imposed. Applying these criteria, it was held that:

[A]ll the available indicators point strongly to this case being classified in the national law as a form of civil proceeding. The appellant is not charged with a crime. Although it must be shown that he was guilty of unlawful conduct in the sense that he has acted contrary to the criminal law, this is not for the purpose of making him amenable as he would be if he had been convicted of crime. He is not liable to imprisonment or fine if the recovery action succeeds. There is no indictment and no verdict. The primary purpose of the legislation is restitutionary rather than penal. (Walsh 2005: para. 27 (per Kerr LCJ delivering the judgment of the court))

Similar considerations applied in relation to the second Engel criterion, with Kerr LCJ stating:

The allegation made against the appellant does not impute guilt of a specific offence; the proceedings do not seek to impose a penalty other than the recovery of assets acquired through criminal conduct; and they are initiated by the director of an agency, which, although it is a public authority, has no prosecutorial function or competence. (ibid: para. 29)

The final criterion was also dealt with rather dismissively, with civil recovery being described as a preventative measure: ‘After all, the person who is required to yield up the assets does no more than return what he obtained illegally.’ (ibid: para. 38) Kerr LCJ also dismissed the argument that, viewed cumulatively, the Engel criteria ought to result in civil recovery being held as criminal in character:

The essence of Art 6 in its criminal dimension is the charging of a person with a criminal offence for the purpose of securing a conviction with a view to exposing that person to criminal sanction. These proceedings are obviously and significantly different from that type of application. They are not directed towards him in the sense that they seek to inflict punishment beyond the recovery of assets that do not lawfully belong to him. As such, while they will obviously have an impact on the appellant, these are predominantly proceedings in rem. They are designed to recover the proceeds of crime, rather than to establish, in the context of criminal proceedings, guilt of specific offences. The cumulative effect of the application of the tests in Engel is to identify these clearly as civil proceedings. (ibid: para. 41)

The robustness of this judicial position is disconcerting, however, not least because, in spite of civil recovery’s adoption of civil processes, it is still evidently concerned with allegations of criminal wrongdoing.³⁸ We say ‘evidently’ because, revisiting the underlying policy, the stated aim in introducing civil recovery was for it to be a more effective tool in dealing with ‘criminal Mr Bigs’ (Proceeds of Crime Bill, HC Deb, 30 October 2001, vol. 373, c. 803, per Mr McCabe). That there was a clear punitive purpose underpinning this draconian power is simply undeniable.³⁹ In applying the Engel criteria to civil recovery under Part 5 of POCA, the UK courts appear to have failed to recognise what civil recovery actually is, which is to say, a wholly instrumental approach conceived to employ civil law processes in pursuing criminal law objectives, and to circumvent those procedural safeguards that inhere in a criminal trial. This is deeply unsatisfactory on two counts. First, the courts over-rely upon the first limb of the Engel test, itself not decisive—however much the legislative intent was to design a civil procedure, this alone is insufficient (Özturk 1984; also Trechsel 2005: 18). We must instead consider what was actually created: intention does not dictate substance (King 2012: 347). In their overly deferential stance regarding the legislative label ‘civil,’ the courts have clearly failed to ensure the adequate protection of individual rights. Second, it is the duty of the courts to scrutinise the behaviour of Parliament and prevent the excessive exercise of political power. We submit that the application of a civil label to a criminal procedure, and the resultant erosion of due process rights and standards that would apply in the event that ‘civil’ recovery were to be properly regarded as a criminal procedure, constitutes such an excessive exercise of power. Furthermore, this can be framed specifically in terms of the rule of law: as Lord Woolf (1995: 68) stated, ‘ultimately there are limits on the supremacy of Parliament which it is the courts’ inalienable responsibility to identify and uphold’. Due process rights safeguard against systemic excess, and their maintenance and application is the task of the courts; this critique is elaborated upon in the final section.

The second and third Engel criteria provide further avenues for critique. Kerr LCJ is correct to say in terms of the second criterion that civil recovery proceedings do ‘not impute guilt of a specific offence’ (Walsh 2005: para. 29). They do, however, allow the State to make general allegations of criminality,⁴⁰ a lack of specificity that simply would not suffice in criminal proceedings. As for the third and final limb, and in spite of the emphasis resting on the ‘proceeds of crime,’ while civil recovery is presented as recouping ill-gotten gains, in essence it operates as a sanction on criminal wrongdoing. Focusing merely on the return of illegal gains without recognising the full implications of the civil recovery order is trite—an individual subject to such an order will not only suffer deprivation of his/her assets but the label and stigma of being a ‘criminal,’ if not in law, then at least in the eyes of his/her fellow citizens.⁴¹ To suggest that under such circumstances the individual is not being punished, that there is no imputation of criminal liability, and that in rem civil recovery proceedings concentrate solely on property, is to be entirely disingenuous about what civil recovery entails.⁴²

The Utility of a Systems-Theoretical Perspective

While the designation of the civil recovery hybrid procedure as either criminal or civil is, as we have argued, a problematic one for the UK courts, systems theory encounters no such problems; on the contrary, it accommodates this distinction with ease. This can be credited to the manner by which the legal system’s binary code ascertains the relevance or otherwise of societal communications and distinguishes them system-internally. The civil/criminal dichotomy is encompassed by the systemic code lawful/unlawful, and thus is included within the unity of the legal system—it is only once the communication has been deemed of relevance to the legal system that further systemic programmes are required to distinguish between these values. As Luhmann explains: ‘Since the values legal and illegal are not in themselves criteria for the decision between legal and illegal, there must be further points of view that indicate whether or not and how the values of a code are to be allocated, rightly or wrongly’ (Luhmann 2004: 192). Programmes, as system-internal semantic elements, are the ‘vehicles of connection between the code and the world-at-large’ (Philippopoulos-Mihalopoulos 2010: 74), and operate by linking self-reference with external reference.

It is within such programmes that rules of procedure are situated in the legal system. Their charge in this regard is the recursive application of norms to the application of norms (Luhmann 2004: 158), with all of these operations being undertaken to realise the legal system’s primary function, that is, the stabilisation of normative expectations over time. This can be articulated in terms of our case study: when existing strategies of criminal law enforcement were deemed unfit for the purpose of tackling the complexities of organised crime, the introduction of civil recovery via POCA 2002 was a direct reaction to the disappointment of the normative expectation that crime should not pay. That is not to say that the autopoietic legal system engages at any point with issues of either the moral content of the law or the policy motivations for legislative change—there is no consideration, for example, of the crime control ideology underpinning ‘follow-the-money’ approaches, or the overt political and normative instrumentality of civil recovery, indeed there could not be. No, where civil recovery has brought about a restructuring of legal systemic programming and, in turn, a reorientation of its rules of criminal and civil procedure, the system itself is only aware of these as system-internal reactions to the repeated disappointment of its normative expectations. Moreover, by virtue of its inclusion within the unity of the legal system, civil recovery accords to the systemic eigen-value of validity.

It is, however, not about the legal validity of a practice that sees a civil standard of proof employed within a criminal mechanism that we have reservations. Rather, we query the legitimacy of civil recovery with specific reference to those procedural standards contained within the rule of law that, we argue, are being eroded by the disastrous combination of legislative excess in implementing POCA and the subsequent judicial failure to protect individual due process rights in the courts.

Procedural hybridity and due process

The statement concluding the previous section is a robust one and, as such, requires both explanation and justification. The final section of this paper argues —in relation to the idea of the (formally conceived) rule of law as structural coupling—that due process rights comprise those minimal procedural standards that politics and law are required to uphold, and that these contingent values serve to connect legal and political legitimacy. In establishing this, we examine the attrition by civil/criminal hybridised measures of those enhanced procedural protections normally applicable at a criminal trial, and consider systemic self-limitations with reference to the concept of legitimacy.

Due Process Rights

Where an individual is charged with a criminal offence, enhanced procedural protections come into play. Although some of these entitlements span both criminal and civil proceedings,⁴³ certain supplementary protections are only applicable where a person is charged with a criminal offence.⁴⁴ The objective of these additional safeguards is the provision of ‘fundamental guarantees against arbitrary State conduct and [the] potential misuse of its authority, an authority that is considerable when the public censure of conviction and State punishment are at stake’ (Ashworth and Redmayne 2010: 403–404). Duff et al. (2007: 5) point out: ‘Defendants have various rights which must be protected, partly in order to ensure that verdicts are accurate, but also to ensure that accurate verdicts are sought with a proper degree of respect for the defendant as a citizen.’ These twin motivations appear conspicuously aware of the power asymmetries at play when an individual citizen is exposed to the might of the State’s criminal justice apparatus, and of the potential for the State to act arbitrarily if not subject to restriction. It is worth noting here that these restrictions upon State power are ones it has imposed upon itself through, for example, becoming a signatory to the ECHR, or upholding its rule of law.

This reasoning holds similarly true under a systems analysis, where rights are conceived as the means by which society guards against the dangers of de-differentiation⁴⁵ and systemic excess. As Philippopoulos-Mihalopoulos (2010: 154) explains, ‘rights are both the result and the precondition of political systemic formation and generally social differentiation. They consolidate the limits between systems while at the same time are guaranteed by the existence of those very limits’. By conceptualising procedural rights as self-imposed restraints upon systemic operations in this manner, it becomes easier to perceive the problem generated by their circumvention through civil recovery measures. Indeed, the very real concerns that the purportedly ‘civil’ nature of these proceedings raises for individual due process and procedural rights can be illustrated by the following example.

In criminal proceedings, the prosecution must establish guilt to the criminal standard of proof beyond reasonable doubt (there is extensive literature on the underlying rationale: recent examples in this journal include Ferzan 2014; Lippke 2014; Stewart 2014 Stuckenberg 2014; Tadros 2014; Weigend 2014). To return here to our civil recovery case study, in proceedings under Part 5 of POCA, it is the civil standard—the balance of probabilities—that applies for assets to be confiscated. Indeed, the legislation is explicit in this regard:

The court or sheriff must decide on a balance of probabilities whether it is proved:

• (a) That any matters alleged to constitute unlawful conduct have occurred, or

• (b) That any person intended to use any cash in unlawful conduct.⁴⁶

In Jia Jin He and Dan Dan Chen (2004: para. 66) Collins J stated:

As a general rule, no doubt, criminal conduct may be regarded as less probable than non-criminal conduct. But where there is evidence from which a court can be satisfied that it is more probable than not that criminal conduct has been involved, it does not seem to me that that is something that is so improbable as to require a gloss on the standard of proof. However, I recognise, and it is no doubt right, that since it is necessary to establish that there has been criminal conduct in the obtaining of the property, the court should look for cogent evidence before deciding that the balance of probabilities has been met. But I have no doubt that Parliament deliberately referred to the balance of probabilities, and that the court should not place a gloss upon it, so as to require that the standard approaches that appropriate in a criminal case. Apart from anything else, if that were necessary, the effectiveness of, in particular, Part 5 of the Act would be to a considerable extent removed. Since it is clear that Parliament intended that it should be used, even if criminal proceedings could not be successfully instituted, it is plain that Parliament deliberately imposed a lower standard of proof as the standard appropriate for these proceedings.

Civil recovery measures, as Kennedy (2006: 139) notes, were devised ‘to ensure that respondents cannot argue that the central issue (whether the property is criminal proceeds) ought properly to be proved to the criminal standard of proof’; here the instrumentality of the legislation is strikingly apparent. Kerr LCJ further emphasised this in Walsh (2005: para. 33), stating: ‘If recovery proceedings could only be taken on proof beyond reasonable doubt that the person from whom recovery was sought had benefited from crime, the efficacy of the system would be substantially compromised’.

Yet that is the crux of our criticism. Where the State is alleging that an individual has benefited from criminal activity, it is difficult for a lower burden of proof than the criminal standard of beyond reasonable doubt even to be countenanced.⁴⁷ Nevertheless, countenanced it is, and more: through Part 5 of POCA, efficiency is privileged over certainty and due process safeguards are abandoned in the service of the legislative intent to ensure that crime does not pay. Indeed, the lowering of the standard of proof makes it significantly more straightforward to prove matters of both fact and law: ‘[e]ven a modest degree of civil content introduced into the strategy, the trading of the criminal standard for the civil standard of proof in the confiscation process, facilitates the task of realizing an attack on the financial elements of crime’ (Gallant 2005: 19). More troubling still is the way in which a person who has been subject to an unsuccessful criminal prosecution can nonetheless be subject to subsequent proceedings under Part 5 of POCA⁴⁸ (Taher 2006),⁴⁹ or how civil recovery proceedings can continue even after a criminal prosecution has been discontinued. This was the case in Jia Jin He and Dan Dan Che (2004: para. 67), where in spite of the decision not to prosecute—‘no doubt because it was considered that [criminal proceedings] would not succeed’—civil recovery proceedings were still permitted. Part 5 proceedings have even been successful in circumstances where the criminal case was stayed as an abuse of process (Hymans 2011). This jurisprudence highlights the way in which civil recovery, where criminal prosecution is either impossible or has been unsuccessful, allows the State to take a second bite at the cherry. To be clear on this point: under such proceedings the parties are the same (i.e., the State against the individual), the allegations will often concern the same conduct, and the evidence can even be the same as that relied upon in the unsuccessful criminal prosecution—the only salient difference is the reduced standard of proof. These observations combine to indicate that civil recovery is an express mechanism for State circumvention of those enhanced procedural protections inherent to the criminal process.

The most bewildering aspect of this state of affairs, however, has been the willingness of the courts to accept unchallenged the legislative label ‘civil’ for such proceedings, particularly in light of the repercussions for due process and individual rights. Indeed, considering that it is arguably the role of the courts to regulate the exercise of state power with a view to ensuring its compliance with the rule of law (O’Cinneide 2009), this raises the question: why, then, such an apparent reluctance to do so in terms of civil recovery? Moreover, is this likely to be the case for all such procedural hybrids?

Procedural Rights and the Rule of Law

In undertaking this analysis, we are mindful that our discussions of rights may intimate that our conception of the rule of law has shifted from a formal to a more substantive one. This is not so, and neither is it our intention to conflate the rule of law with the substance of particular rules. That said, even the most formal and neutral conception of the rule of law allows for the inclusion of legal-cultural dimensions: indeed, Raz has stated that ‘[i]n insisting on the integration of legislation and other current measures with legal traditions enshrined in doctrine, the rule of law respects those civil rights which are part of the backbone of legal culture, part of its fundamental traditions’ (Raz 1994: 360). There are three important points to take from this statement. First, the rule of law is undeniably contingent upon its particular legal-cultural context. Second, in a marginally ‘thicker’ conception (Bingham 2011: 119) than perhaps usually expected from Raz, due process civil rights fall squarely within the ambit of the rule of law, arguably due both to their procedural character and their noted significance within the legal culture of the UK. In this regard, the third and final point from Raz’s statement has particular salience, and that is his insistence that not only is the rule of law so understood confined to democratic societies but that such due process rights are in fact a condition of political democracy.⁵⁰ He argues that ‘the law inevitably creates a great deal of arbitrary power [and] the rule of law is designed to minimize the danger created by the law’ (Raz 1977: 206): within a democratic society, the rule of law operates as a control on governmental power, intended to safeguard against its excessive use or abuse. Not only are there clear similarities between this position and the systems-theoretical understanding of rights as systemic self-limitations, but this privileging of political democracy is also evident in Luhmann, who considers it to be a vital condition for the functional differentiation of modern society. Indeed, as Thornhill (2006: 89) explains, ‘a political system is unlikely to make its contingency plausible (legitimate) if it fails to reflect and respond to the plural and differentiated reality of democratic societies’. Democracy can thus be cited as a condition of political legitimacy as much for a systems-theoretical understanding as it is under a classically social-contractarian one.

What of legal legitimacy, however? Although we have presented it as a problematic notion for a systems-theoretical construction, it is here elucidated through the existence and operation of the structural coupling of the rule of law, a structural coupling that operates to peg the legal system to those standards established by the political system. By means of this coordinated structural link, political systemic limitations—namely, due process rights—are extended to the legal system, wherein they embody value criteria within the ‘black box’ of legal legitimacy. Although such due process rights are ‘a pivotal part of the arsenal of values in the name of which the political system legitimises itself’ (Philippopoulos-Mihalopoulos 2010: 154), such rights act as a shield against the structural violence created by unrestricted functional (de)differentiation, and this is true for all function systems, not just the political. It is through a combination of such rights as systemic limitations and the rule of law as a structural coupling between the political and legal systems that we can understand legal legitimacy. Not sufficient simply to be legally valid, legal legitimacy requires that the operations of the legal system are in accordance both with those minimal limitations imposed upon it by the procedural rights that it itself has recognised as legally valid, and in terms of the political values that serve to furnish the content of the rule of law and structural coupling between the political and legal systems.

Our exemplification of this conundrum has relied upon the controversial provisions of POCA 2002, and so will our conclusion. We asserted from the outset that hybridised procedures in general, and civil recovery in particular, lack legal legitimacy, and we maintain this position on two counts. POCA itself constitutes overt legislative overstepping in its evasion of enhanced procedural rights protections, a situation then compounded by the failure of the courts both in identifying this excess and in taking steps to prevent or mitigate its effects. Such failures cannot but lead to the following conclusions: that civil recovery provisions are lacking in necessary legitimacy—a situation that places the continued reliance upon such hybridised procedures into far sharper relief—and that the instrumental and expedient circumvention of those procedural safeguards that usually inhere in a criminal trial should be recognised as a very real concern in terms of civil liberties within the UK.