I. Introduction
Constitutions involve paradoxes which have many elements (see, for examples, the essays in Loughlin and Walker, Reference Loughlin and Walker2007). First, there is the temporal paradox that the nation or ‘people’ which purports to authorise the constitution has no clear identity prior to their designation as such by the constitution. So it is equally possible to say that the constitution brings the people, as a constituent power, into being, as to claim that the people, as a constituent power, establish or authorise the constitution. Second, there is the ‘self-binding’ paradox.Footnote 1 If the constituent power has the authority to establish the constitution, on what basis can that constituent power be restrained from re-exercising its authority, to create a different constitution? What can explain the loss of authority to alter a constitution by the fresh exercise of the constituent power? And linked to this second paradox is the location of power which, if self-binding is possible, is not expressed in the constitution, for there is a surplus of power. Does this surplus disappear on the creation of a constitution, or remain available to be referred to or drawn upon in another form? It is the last aspect of this paradox that we wish to explore in this paper, for it amounts to the closely related paradoxes of civil disobedience, and in particular the claim that it might be right to undertake acts of civil disobedience despite their apparent illegality or illegitimacy.
Within discourses of civil disobedience, the expression of political power through a legal constitution does not exhaust the resources of legitimate political action. Thus, the claim of a right of civil disobedience, a right to disobey laws enacted in accordance with the provisions of the constitution, remains something that is articulated as a moral and/or political right. In this paper, we want to use some of Luhmann's writings, and the version of systems theory which he developed extensively in his academic writings throughout his life (for a comprehensive statement, see his lectures originally given in 1991–1992 and now published as Luhmann, Reference Luhmann and Gilgen2013a), to consider the nature of this right. This is not a normative exercise. Our use of systems theory is not intended to establish the appropriate situations in which this right may be claimed (many others, of course, have attempted this normative exercise, a classic example being Arendt, Reference Arendt1973; for a recent example see Herstein, Reference Herstein2014, building on Herstein, Reference Herstein2012). Rather, we wish to explore the sociological existence of this right, taking a systems theory approach to this question. This exercise should inform our understanding of constitutions, and how they operate in the face of paradoxes, which, as paradoxes, are necessarily irresolvable even though the contexts within which they operate will, on occasion, make them more or less visible, or make their resolution appear more or less plausible.
II. Civil disobedience
Civil disobedience involves a claim that the resources of legitimate political action have not been exhausted by the constitution. Those who practise civil disobedience seek to distinguish themselves from both revolutionaries and criminals, via the justifications which they offer for their actions.Footnote 2 Revolution is a challenge to the present political order, which those who seek revolution regard as illegitimate. Revolution is a call for a new order, a new expression of constituted power.Footnote 3 Civil disobedience (to the extent that it does not participate in, or anticipate, revolution) seeks to communicate that it does not question the current constitution as constituted, and its allocation of competences. Rather, it seeks changes to a policy or law through behaviour which is illegal, without seeking changes to the institutional arrangements that have resulted in the policy or law in question. But (and here these paradoxes unfold) the claim of justification cannot but generate a communication whose meaning is revolutionary, for if there are normative resources which can legitimate illegal actions, the exercise of constituent power cannot be taken to have exhausted these resources in the formation of the constitution. In this sense, claims of civil disobedience on the part of citizens represent a mirror image of claims of acts of stateFootnote 4 on the part of government officials. Both involve the assertion that the legitimacy of political action cannot be exhausted by the establishment of a constitution and the legal norms created in accordance with that constitution.
The issue of whether the legal constitution can, or cannot, exhaust the legitimacy of political action has been debated within legal and political theory in the form of the opposing approaches of Hans Kelsen and Carl Schmitt (see particularly Lindahl, Reference Lindahl2007; Dyzenhaus, Reference Dyzenhaus1997). Kelsen took the view that power could only be constituted as legitimate through the legal form of constitutions.Footnote 5 The constitution identifies the constituent power which validates it. And the authority of that constituted power, which the constitution retrospectively validates, is provided by a norm which is simply presupposed to exist prior to the constitution. This grundnorm is a logical presupposition, and one whose content is limited to retrospectively authorising the historically first constitution (since the last revolution). For Kelsen, since no objective legitimate political power or action could be identified without the legal knowledge brought into existence by the constitution, there can be no legitimate exercise of political power that is not authorised via a pre-existing legal norm.Footnote 6 The strength of Kelsen's position lies in the difficulties of identifying a normative foundation for any particular configuration of constituted power. Natural law offers only an endlessly contested source for it. Democracy offers a grounding for expressing power once a democratic constitution has come into existence, but it cannot establish the constitution itself, as the polity or nation which is to exercise democratic power does not exist prior to the constitution. It is the constitution which retrospectively brings this polity into existence. The people are no less of a fiction or presupposition than the other aspects of the grundnorm – the norm which must logically pre-exist any claim to exercise constituent power in order to give that exercise the requisite normative meaning.
Schmitt, by contrast, insisted that constituted power exists prior to the constitution, and remains available to be drawn upon, particularly in moments of emergency. He identifies this power with the unity of a people who identify with each other to the extent that they are prepared to fight and die in defence of their common association (as friends) when faced with threats from those who might threaten that association (their enemies) (Schmitt, Reference Schmitt and Schwab2006; see also Mac Amhlaigh, Reference Mac Amhlaigh2013). Whilst such united peoples may have existed on some occasions of constitutional enactment, it is not plausible to claim that such commitments are present in all populations that have at one point or another adopted a constitution which formally identified them as a nation. Thus, even if one accepts that a ‘people’ have the normative right to establish a constitution for themselves, this would not explain the normative source of most constitutions. Moreover, and more central to the focus of this paper, even if there are situations where national peoples have existed as a clearly defined group in advance of the first constitution that refers to them as such, this still leaves us with the question of why those who did not form part of the nation at the date of the first constitution are bound by its terms.
What does Luhmann's systems theory bring to these questions? The debate between Kelsen and Schmitt can be interpreted as an attempt to solve the paradox of constitutionalism by locating the origin of the constitution within one or other of these systems: law for Kelsen, politics for Schmidt (see Přibáň, Reference Přibáň2011). Both utilise arguments of logic (though Kelsen more obviously and consistently than Schmitt). But ‘the study of legal [and political] paradoxes is not simply a matter of logical or linguistic analysis … [Politics and] law [are] social system[s] … [which] means that the riddle of [their] parodoxicality cannot be solved by relying solely on the tools of logic’ (Perez, Reference Perez, Perez and Teubner2006, p. 4). One must instead consider how these paradoxes play themselves out in the practices of legal and political systems. That is the premise of this paper. We wish to consider how claims, that civil disobedience can be right, operate within the legal and political systems. As such, we will not here reproduce the manner in which this topic is typically discussed within philosophy, political philosophy, political ethics and jurisprudence. Authors who operate within these traditions seek to identify cogent, coherent and persuasive accounts of the right to civil disobedience. These are attempts, we believe, to defuse one of the paradoxes of constituted power logically – to find a description of constituted power that leaves a conceptual space for disobedience, without simultaneously taking a revolutionary position, which denies that currently constituted authority has legitimate authority. Instead, we wish to address the question: What space is there within society for the praxis of civil disobedience, sociologically?
III. Constitutions within systems theory – and their connection to civil disobedience
Luhmann's interpretation of constitutions within systems theory is that there are not one but several constitutions. Systems construct themselves and their environments, including their own realities through their own operations. Constitutions, whether they are attributed by a system's operations to themselves (as occurs within political and legal systems) or attributed to the operations of many other systemsFootnote 7 (as may, for example, occur within art), are constructed by and within systems. However, the two systems which are generally referred to when there are observations about constitutions are the political system and the legal system. Applying Luhmann's approach, what is generally referred to as the constitution is a separate construction of law and politics.Footnote 8 Each constructs its own version of the constitution. As such, its origins do not lie in either, any more than its continued existence lies in the one and not the other. What generates the reality that is observed from both within and from outside these two systems is the nature of the relationship between them. This is the relationship of, what Luhmann calls, ‘structural coupling’.
Structural coupling occurs when ‘a system presupposes certain features of its environment on an ongoing basis and relies on them structurally … the forms of a structural coupling reduce and so facilitate influences of the [system's] environment on the system’ (Luhmann, Reference Luhmann and Ziegert2004, p. 382, emphasis in original). What this means for the legal and political systems is that they rely on certain communications generated within the other system; communications from one have a second existence within the other. Luhmann calls this ‘second coding’. This refers to the fact that communications which within the political system are being linked together in order to apply its code ‘government/opposition’ are, within the legal system, being linked together in order to apply its code ‘legal/illegal’. The different connections of communications about constitutions within each of these social systems create different meanings, even where the syntax is identical. (So the ‘political’ version of the constitution will not have the same meaning as the ‘legal’ version of the constitution.) However, each system is relying on the other's communications ‘structurally’ because we do not have a one-off situation, but a recurrent one; each system is orientating itself (i.e. generating both more, and more complex, communications) in response to what they take to be the relatively stable and predictable communications being generated within the other.
Why should this structural coupling occur between two systems that are, according to systems theory, autopoietic, normatively closed, and self-reproducing? The reason lies in the productive consequences of the relationship for each system.Footnote 9 Law gains access to physical force, which whilst not the sole motivation for human beings to expend their energies seeking to trigger communications within the legal system, is certainly a major one. And politics gains a means to apply power technically, impersonally and with a precision that far exceeds the available alternatives – ideology, loyalty, tribalism, tradition, custom or self-interest. The attractions of this form of power have, in Western Europe, since the eleventh century, motivated political actors to seek access to legal forms, a process which has encouraged a transformation of political power from feudal relationships, where political power was personal and based on allegiance and tradition, to the form of power associated with the modern state – impersonal, and technical – the form of power that is generally described as the rule of law.Footnote 10
A second productive aspect of this relationship, and the one most relevant to the character of civil disobedience, is that each system is able to export the paradox of its own foundation to the other. The legal system has historically exported its foundation into morality, through natural law doctrines and, in England, through the attribution of the sources of law to the wider community. But with the development of posited forms of law (especially legislation, but also modern constitutions and Bills of Rights) the legal system has been able to ascribe its own central paradox (what it is that, on all occasions, makes the legal lawful) to the political system: the legal is legal/lawful because it is made in accordance with a constitution that has been created politically – law is constituted political power. But, as we discussed at the beginning of this paper, the political system faces a similar paradox: On what basis is the establishment of constitutions collectively binding? What resources can politics draw upon (democracy, tradition, charisma, tribalism, etc.) to establish the legitimacy of the constitution which does not at the same time provide a resource for the endless questioning of the legitimacy of the constitution? The export of this paradox from the political system takes the form: it is legitimate because it is legal, i.e. something that has occurred in accordance with the legal constitution. In exporting this paradox to the legal system, politics gains a stability that it could not create for itself. The basis of this stability is not found simply in the technical precision that legal systems offer to the expression of political power, for that feature of legality can lead to the impression that all political rights can find expression as legal ones. Indeed, it is a staple of the structural coupling of politics and law for the former to organise itself around processes and procedures that constantly translate political claims, including claims of right, into legal forms, as legal rights. However, what separates civil disobedience from any other kind of political claim of right is the inability of the legal system to recognise this claim as a legal right within its legal communications in any self-standing form.Footnote 11 Whatever the political system recognises as the basis for legitimate power, and whatever surplus remains once that political power has been constituted within a legal system, the legal system cannot give effect to that surplus by simply recognising a right to disobey law. What, within the political system, represents a constant possibility for the destabilising of politics and constituted power, cannot easily enter the legal system and its operations. Civil disobedience can remain as a claim, even a claim of right, within the political system, but cannot operate, in a straightforward way, as a right operating within the legal system. This has implications in turn for the political system in its relationship to the legal system. Those parts of the political system that have made use of law's ability to direct political power in highly technical forms – as mandates, powers, institutions, offices and responsibilities (which has occurred to the extent that the self-description of the political system now takes the form of ‘the state’)Footnote 12 are not open to communications that in a straightforward way acknowledge a right to disobey law.
IV. The inability of the legal system to incorporate the right to civil disobedience
This inability is not self-evident, or at least not self-evident to all. The German Basic Law seems to provide for a right of civil disobedience.Footnote 13 Joseph Raz has discussed the possibility of a legal right of civil disobedience and concluded that it is undesirable and raises difficult issues for adjudication, but not that it is impossible.Footnote 14 And whilst one can point to any number of reasons why members of the judiciary might be reluctant to recognise rights to disobey law (conservatism, identification with government, etc.), the claim we are making here is independent of the values of judges as human actors. It is an observation on the operations of law as a social system – that it cannot nowFootnote 15 recognise a right to disobey law.
To introduce this inability, we propose taking an iconic case which is the subject of a recent essay by Austin Sarat (Reference Sarat, Nobles and Schiff2014), on the need for less civility, and more dissent. The case in question is Walker v. Birmingham.Footnote 16 The Supreme Court heard an appeal against the sentence imposed on Martin Luther King and some of his followers in 1963, when they defied a court injunction granted by an Alabama circuit court judge prohibiting them from conducting demonstrations in Birmingham, Alabama. Sarat sets out a compelling argument that the injunction was unjust. Not only was the protest against segregation and racial discrimination, but the town council had made it clear that they would not grant permits for this kind of demonstration, and had sought an injunction in terms that were so wide that it threatened the right of free speech. The demonstrators had a clear position: the injunction was an unjust law, and should not be obeyed. Citing Aquinas, the demonstrators also gave their view that an unjust law is not actually a law, which, if accepted, would have meant that they were not disobeying the law. And lastly, on appeal, they argued that an unconstitutional provision was a nullity, with the result that their actions could not be considered to be breaches of the law. The Supreme Court upheld the convictions and sentences imposed (five days in jail) on the basis that the demonstrators had chosen to break the terms of the temporary injunction, rather than attempt an application to court to seek to have the injunction restricted or overturned. For Sarat, the Supreme Court decision supported those who expected the civil rights demonstrators to react to racists abusing legal procedures to prevent their protest by obeying the court order and seeking only legal means to challenge it. He calls this a pathological example of civility.
The issue, as presented to the court, was whether the behaviour of demonstrators in defying the injunction was legal, or illegal. If it was legal, then their actions may have been ones of dissent, but they would, retrospectively (to the court), not have been acts of civil disobedience – justified breaches of the law – but legal acts. If their actions were illegal, then they had no legal right to undertake them.Footnote 17 A right to disobey law was not offered to, or discussed by, the court. What the court did discuss was the status of temporary injunctions – orders made by a court in advance of a decision on the merits of the case. The injunction-issuing court could not assume a jurisdiction which it did not have, or make an order that was patently unconstitutional – both of these would be a nullity. But where the constitutionality of the law which supported the order had yet to be decided, and the legal arguments and evidence necessary to decide those questions had yet to be heard, did the injunction decide what was legal? The Supreme Court determined that the injunction established what was legal at the time. Whilst Sarat may be correct to describe the demands made by the legal system upon Martin Luther King and his followers as a pathological example of civility, the pathology is not so clearly that of the judicial actors as human beings. Faced with the binary choice of the actions of the demonstrators being either legal or illegal, and being only able to declare them legal by articulating an exception to temporary injunctions which went wider than an excess of jurisdictional power or the enforcement of a patently unconstitutional provision, the court decided to maintain the stability of the meaning of temporary injunctions, and not articulate an exception to them which could embrace the actions of the demonstrators. What can systems theory tell us about the choices made in this case?
Luhmann (Reference Luhmann1988) examined the nature of judicial choices in his essay ‘The Third Question: The Creative Use of Paradoxes in Law and Legal History’. Whilst there are situations within other systems where the relative rightness of a person's position can be considered (he gives the examples of teachers and therapists), a judge is faced with a binary opposition in deciding what is legal, or illegal. The only option open to a judge is to draw endless distinctions between situations – like cases must be treated alike. This process is taken to its greatest level of complexity in the courts, which he locates as at the centre of the legal system. For Luhmann, this process is productive for law – productive in the sense that it stimulates law to evolve by reaching ever greater levels of complexity. The construction of doctrine through the endless comparison of cases (prior decisions, and hypothetical examples that have relevance to the legal system) generates doctrine – definitions, rules, exceptions, exceptions to exceptions, new rules, etc.Footnote 18 For Luhmann, this process suppresses a central paradox, which is the lack of any foundation which establishes, under all conditions, how the legal binary code (legal/illegal) is to be applied. There are only the endless applications of the code, and the ever more complicated justifications of their application, which produces what he calls law's ‘conditional programmes’.
Justice is relevant to this process, but not because it generates a right to disobey laws. Luhmann calls justice the legal system's ‘formula for contingency’ (Reference Luhmann and Ziegert2004, chapter 5; discussed in Nobles and Schiff, Reference Nobles and Schiff2006, pp. 112–125). In making selections, identifying what is like what, and what can be distinguished, the legal system communicates its choices to itself in terms of justice. This description may seem like an empty formula. And any claim that it forms part of the legal system may seem to be contradicted on those occasions when judges admit that what the law requires is harsh, or even immoral, and yet uphold the law. But justice has a role in articulating the law which cannot be replaced by communications taken from other social systems, such as ‘national interest’ (politics) or efficiency (the economy). Whatever one feels may have influenced the decisions of judges, communications in these other forms have not been the mode of law's evolution. Instead, choices between indeterminate alternatives are articulated and justified through communications about the justice (especially in the form of like cases being treated alike) of a chosen outcome. Although justice in this form does not give determinate outcomes, it does provide law with its own communications for affecting its own choices. And thus it reduces the ability of other communications (political, economic, scientific, etc.) to operate at these moments. Justice within the legal system is an example of what Luhmann calls an ‘eigenvalue’. It is a value generated by law's operations. The communications which recognise law's contingency refer to something beyond the law which is to be identified by the routine manner in which law makes its selections – the most important of which is to treat like cases alike. Any suggestion that this value is fragile is belied by the difficulties of any legal operation being carried out using communications that deny that justice is a value, or assertion that like cases should not be treated alike. Justice thus has implications for a legal right of civil disobedience. Identifying an interpretation of a law as not just forms part of the process whereby alternative interpretations are chosen.Footnote 19 Because of this the difficulties created for the legal system by a legal right of disobedience are not limited to the creation of contradictory communications about a particular legal act (the illegal is legal). A right of civil disobedience based on the identification of what is unjust requires the legal system to abandon the use of many of its communications about justice to construct its selections between indeterminate choices.
Put these two aspects of a right to disobey law together, and one can articulate the future of any hypothetical legal right of civil disobedience in two ways. First, it would involve a large potential loss of redundancy,Footnote 20 in that in recognising a general moral or ethical basis for disobeying law one simply opens up previously settled areas of doctrine to reassessment. And second, in the very process of opening up the law it also energises the processes for its own removal. A loss of redundancy in circumstances where individuals will litigate in order to determine what is legal, or illegal, leads to the processes described by Luhmann in ‘The Third Question’, whereby an apparently open moral right of disobedience is replaced by doctrine establishing what is legal, and what is not.Footnote 21 Thus, and this applies to other situations where legal actors encounter possible losses of redundancy – as well as the pressure on the human actor represented by their awareness of the loss of an ability to decide what is legal (often referred to as the value of legal certainty) – there is also a compensating mechanism to restore redundancy where its loss is not sufficient to deter particular actors from articulating interpretations that will challenge it.
At this point let us consider again Austin Sarat's example of the pathological civility displayed by the Supreme Court in Walker v. Birmingham. What alternatives were open to the court? Responding to the demonstrators required the court to reconfigure the rules on temporary injunctions, allowing for the legality of disobeying them in circumstances where the injunction was not patently illegal and the factual and legal arguments which would establish the facts on the ground were yet to be established. This would not have been an interim injunction as then, and currently, understood. Such a change might lead to injunctions that operated only as a warning that if the substantive hearing decided that the merits of the action did not lie with the demonstrators, they would be punished. But as the Supreme Court made clear in its judgment, the loss of the ability to prohibit actions through temporary injunctions would not only prohibit potentially illegal civil rights marches, but potentially illegal Ku Klux Klan ones too: like cases must be treated alike. This retreat from the power to issue temporary injunctions would thus not only result in a loss of redundancy with regard to the availability of injunctions, but it could also be expected to stimulate further litigation as to which ‘political’, or other kinds of case, could no longer give rise to conventional temporary injunctions. Those in Birmingham, Alabama, who sought a right, based on injustice, to disobey the law, did not want to see such doctrinal developments as these. Indeed, as history subsequently has shown, a legal system that does not have the capacity to declare politically motivated resistance to court orders illegal would have had less power to enforce desegregation.
Is there any space within the legal system where a right of disobedience can operate? The answer to this is already familiar to us: within those parts of the legal system which the system identifies as involving discretion. Through prosecutors' discretions, the legal system can leave it to politics to decide whether to enforce laws that will be disobeyed for reasons of conscience. By having discretionary sentences rather than mandatory ones, it can leave it to judges to decide how to react to particular acts of disobedience.Footnote 22 But, as the next section will show, if these discretions are replaced by rules, such that acts of disobedience are acknowledged by the legal system as rights, then the law changes, and the disobedience, for which a right had been claimed, disappears too.
V. The examples of R (Purdy) v. DPP, and R (Nicklinson) v. Ministry of Justice Footnote 23
The legal background to these recent cases is the tension between a prosecutor's discretion and the substance of the criminal law. A prosecutor is not required to prosecute on every instance when a criminal offence is committed. At the other extreme, a prosecutor cannot be required to identify precise circumstances in which, despite the commission of a criminal offence, a prosecution will never follow. Both positions remove the element of discretion. With respect to the first position, this creates significant practical and political difficulties. The prosecution of all offences on all occasions when there is sufficient evidence to make out the offence removes the ability of prosecutors to set priorities on the use of their resources. It also prevents prosecutors from abstaining from prosecuting offences where this would have adverse (sometimes clearly political) consequences. The second position raises the issue of a de facto repeal or amendment of a criminal provision. If a prosecutor is not allowed to prosecute in specific circumstances, even though those circumstances constitute a breach of the criminal law, in what sense can it be said that those circumstances are in fact criminal? And, where the criminal law in question is a statute, on what authority within the UK legal system can this restriction on the prosecutor's right to prosecute be justified?
Purdy followed an earlier case, that of R (Pretty) v. DPP,Footnote 24 which had involved a more straightforward modern attempt to argue that an unjust law is not, in fact, a law. In that case the appellant, who was suffering from an incurable illness, sought a declaration that Section 2(1) of the Suicide Act 1961, which makes it unlawful to assist someone to take their own life, is an infringement of her right to a private life under Article 8 of the European Convention on Human Rights and the Human Rights Act. She failed, with the House of Lords declining to make a declaration of incompatibility between the Convention and this criminal provision. However, on subsequent appeal to the European Court of Human Rights, the European Court acknowledged that her Article 8 rights might have been infringed if prosecution followed on all occasions when a Section 2(1) offence is committed. In Purdy, relying on Pretty, the appellant asked the House of Lords to order the DPP to issue guidance making it clear when prosecutions would occur against those who assisted others to commit suicide. The Court so ordered and the DPP produced a detailed specific guide, listing the factors indicating when prosecution would, and those indicating when it would not, be likely to occur. Nicklinson involved, among other things, a complaint that the guidance given was not clear enough. The DPP was ordered by the Court of Appeal to provide additional guidance regarding the involvement of health professionals in assisting persons to commit suicide.
All of those bringing the litigation in these cases were effectively seeking an immunity from prosecution under Section 2(1) on behalf of those who they might call upon to assist them in taking their own lives. All of the judges declined to give them this immunity, recognising that such an enforceable right would amount to a repeal of the statutory provision. The judges were aware of the constitutional implications of interpreting a prosecutor's discretion as a power to grant rights against enforcement, noting that this undermined parliamentary sovereignty and threatened the rule of law. The issue was discussed in terms of the right of those who might assist in suicides for reasons of compassion to know, with ‘sufficient certainty’, their risk of prosecution. But the relevance of these decisions to our discussion is not the constitutional question of whether the executive, or Parliament, has the right to make and amend law. Rather, it is the awareness, within the judgments, that recognising an enforceable right to disobey law will constitute an amendment to the law. Whilst the judges discussed entitlements to different degrees of certainty of the possibility of not being prosecuted, and even accepted that a decision not to follow the guidelines for irrational or arbitrary reasons might support the judicial review of a decision to grant consent to prosecution, none of them would go so far as to articulate circumstances in which those who act contrary to Section 2(1) would have an enforceable right not to be prosecuted. There is a common awareness that this extra step – the recognition of a right to disobey law – changes the law.Footnote 25
VI. Civil disobedience within the political system
Individuals can exhibit attitudes towards law that might be called ‘partial respect’ (see Raz, Reference Raz2009, chapter 13, ‘Respect for Law’), in which they treat some legal duties as reasons for action but not others. But, as our discussion of a legal right to disobey law shows, they have considerable difficulties in making legal communications in keeping with such attitudes. To make legal communications, individuals have to utilise the store of communications currently available within the legal system to carry out their desired legal operations. If the legal system does not offer communications which allow a claim to a right to disobey to be acknowledged, so that instead it constantly translates such communications into claims to reinterpret the law, or admissions that an action is illegal which excludes the possibility that it can also be legal, or requests that those charged with discretion over prosecution, sentencing and conviction exercise that discretion so as to avoid the claimant suffering penalties, then we can say that there is no legal right to disobey the law. And this restriction covers both those who administer its courts, and those who appear before them. Whilst one might expect at least some, if not all, judges to be familiar with moral and political theory, and thus to be quite capable of understanding claims that some legal duties are morally wrong,Footnote 26 there is no way for this ‘knowledge’ to be translated into a recognition that disobedience to law could be legally correct. What then of attempts to generate such rights within the political system?
Claims of civil disobedience in the political system are claims concerning the legitimacy of political action. A systems theory analysis of the political system can only start from the perspective that the political system generates legitimacy internally, basically through the stability and predictability of its own operations. Within the political system, legitimacy plays a role similar to that played by justice within the legal system: a formula for contingency.Footnote 27 In building its programmes for the making of collectively binding decisions, and when choosing its selections from alternatives (dealing with its internal contingencies), the political system orientates itself through using communications which represent its decisions to itself as legitimate (and thereby justified). Such representations engage communications that would be well known within the education system as those of moral and/or political philosophy. But if one appreciates the claim that politics, like law, is an autopoietic system, then this means that neither moral norms nor political ethics can directly bestow legitimacy on the political system. Rather, from the perspective that the political system constructs itself self-referentially, one is required to reflect on how selections of communications are made internally, within the political system. This does not amount to saying that ‘moral’ communications have only an ideological role within the political system, i.e. a kind of communication that is symbolic rather than ‘real’. Such communications enable the political system to generate its political programmes, and in doing so are a part of its operations. However, they may well appear ideological rather than ‘real’, because the embodiment of such values seems hardly to be represented fully by the political system, or much of it. One may find recognisable moral communications all around the political system, and one can observe the operations which such communications seem to construct, but that is very different from concluding that the political system as a whole is expressive of a sound moral scheme, or particular moral values. Even those political programmes that most appear to represent particular moral values, such as elections as symbolising representative democracy, are constrained within the system and its need to select from its internal contingencies. National periodic elections are an operation that anticipates and precipitates enormous numbers of other political communications. Indeed, for the political system as a whole the legitimacy of elections is much more than the value of democracy; without elections so many other political communications would have to be reconfigured. Thus, internal to the operations of political systems, what elections ‘legitimate’ is the provision of a platform for so many other communications which anticipate elections and follow from them, such as, in particular, the details of how government will be formed. These further political operations do not make unimportant ‘democratic’ communications, but rather allow those communications to ‘legitimate’ so many further political communications. Internal to the political system, an illegitimate election is one that fails to decide how the government, its offices, responsibilities, etc. will be organised. Any acknowledgement, as a critique, that elections involve serious democratic limitations of choice (How much democracy does one get from a single vote?) does not make those elections ‘less’ legitimate for the purposes of the vast range of further political operations.
This sociological understanding of legitimacy as an internal formula for contingency for the political system is very different from other accounts that substantively link the legitimacy of legal and political systems to particular values. So whereas, for example, John Rawls's theory of justice attempts to justify civil disobedience in a ‘nearly just’ society (Reference Rawls1973, chapter VI), from a systems theory perspective a nearly just society is an internal construction of that society's political system. What communications allow a political system to construct its self-description of itself in these terms? If not simply ideology, such a self-description of its totality as a system is located within the communications of the system and its operations. Self-description and self-observation orientates selections made within the system. But having identified the values expressed within a political system's self-description, it is inadequate to expect that a critique of that system's operations, based on these values, will change any particular operations within the system. Rather, to change its self-description, a system needs to find alternative ways to communicate about itself, and ones which will allow its communications to continue. So, unlike Rawls's account of justified civil disobedience, or that of others (as a recent example, see Brownlee, Reference Brownlee2012), we ask what problems there are for a political system from communications which acknowledge rights to resist or disobey law.
The political system, like any other system, identifies which communications, and in consequence which choices, belong to itself. In modern societies it does this through the application of the code: government and opposition. (For a brief statement of the coding of the political system, see Luhmann, Reference Luhmann1989, chapter 13, ‘Politics’.) It is this communicative code which allows the political system to constitute what alternatives it is able to select (what parties, persons, policies, etc.). And it is not only its own choices that are created, but also the manner of their selection (via delegation, prerogatives, types of voting, etc.). Of course law is also engaged in these processes, namely those laws which the political system has selected as applying to itself. And the consequence of law's involvement is likely to be given more predictability having been enacted. Written constitutions, including Bills of Rights, are the most impressive example of this process. While the legal and political systems have become functionally ever more differentiated from each other, they nevertheless provide, through structural coupling, stable and consistent support for each other's operations.Footnote 28 In these circumstances, to contemplate a political right of civil disobedience we have to examine two sets of possibilities. First is the ability of the political system to recognise alternatives outside those which it has already constructed for itself applying the code government/opposition. Can there be an alternative to the alternatives expressed through political communication structured in terms of this code? Second, can the political system both legitimise itself using the legal system, and communicate about politics in terms of a right, for political reasons, to disobey law?
Let us begin by looking at the second of the two possibilities set out above. Legitimacy, as we have argued, is not something that is externally bestowed on the political system, but rather something that it generates through its own operations. Thus one needs to consider the political system's ability both to construct itself through many operations that are second-coded as legitimate from their reliance upon legal communications, and at the same time acknowledge a political right to disobey the law. Once again, from a systems theory perspective what is involved is ‘redundancy’.Footnote 29 If legality has come to form a significant part of the communications that bestow legitimacy on collectively binding decisions, then the political system cannot easily acknowledge the right to disobey the law with regard to such decisions without finding other communications which allow it to satisfy its continuing need to legitimise its own operations. This is certainly not a claim that legality always determines the legitimacy of political decisions, which would be to confuse the political system with the legal system. Rather, it is a claim that political decisions coded as legally valid provide a common platform for other legitimate political communications. Similarly, legitimate, collectively binding, political decisions form a platform for further political operations. There is, however, a crucial difference between legality as legal code, and legality as communication utilised to account for legitimacy in another system (in this instance, the political system) with its code. In the legal system there is no third value between legal and illegal. However, within the modern political system, self-observations (communications) about the application of its code (government/opposition) routinely draw on the third value of legality to account for its decisions. Nevertheless, decisions within the political system are not automatically legitimate when coded by the legal system as legal, but neither are they necessarily illegitimate even if coded by the legal system as illegal.Footnote 30 Significantly, the coding of these two systems does not necessarily coincide temporally; indeed, the ‘illegality’ of a political decision may take place months, or even years, after that decision, during which time many other political operations will have taken place on the basis of the legitimacy of that only later acknowledged ‘illegal’ operation. Coding political operations as illegal may well not be able to reverse many subsequent political operations based on them, nor can the legitimacy of political decisions be determined simply by their current status as legal or illegal.
Returning to the first possibility, one can accept that acts of civil disobedience, and even revolution,Footnote 31 may on occasion convey legitimacy upon operations within the political system (namely communications which provide the basis for further ‘legitimate’ collectively binding decisions). Indeed, that is the paradox with which this paper started – acts that had no objective legal or political basis have constituted political and legal orders. But acknowledging this possibility does not address the issue of whether and how civil disobedience can exist as a right within the political system, or rather, and returning to constitutional paradox, of how the values that are drawn upon to found a constitution are prevented from being reasserted after its creation. For systems theory's account of ‘rights’ in the political system, one needs to observe their self-referential construction (as an example, see Verschraegen, Reference Verschraegen2002). The political system constitutes itself by the application of its code, and by the self-observations which account for this application. To say that political rights exist, sociologically, is to observe how collectively binding decisions are communicated about as legitimate within the political system. Political rights are regularly exported to and reconstituted within the legal system as legal ones. Thus the right to freedom of speech can operate both as a political and a legal right, and its dual existence within the legal and political systems supports its existence within the other. Law constrains political action that contravenes freedom of speech, which gives some stability to its expression as a right within the political system. Within the legal system, the right to free speech represents a version of a political right that the legal system is able to process. In particular, constitutional rights, like free speech, enable the legal system to organise, through its communications, its version of the current political system. More broadly, from this systems theory perspective, constitutional rights bring the political system into the legal system to be constructed and organised there as particular interrelated sets of decisions, arguments and values.
There are, however, some potential drawbacks from this ‘cosy’ relationship. Some writers regret the reconstitution of political rights as legal rights, regarding the stabilising and concretisation resulting from this process as an undesirable stultification of the political process (see Christodoulidis, Reference Christodoulidis1998). Others consider any increase in the ability of unelected judges to influence the political system, through the interpretation of rights in general and, constitutional rights (which can trump legislation) in particular, as unacceptable.Footnote 32 However, the current general approach to rights within the political system appears to be to realise political rights through enactment as legal ones. And, in doing so, legality's role in establishing political legitimacy can be explained in terms of its ability to enable the political system to develop in complexity. Political power grows when it can be expressed in impersonal forms – especially through legal rules. To understand that evolution is neither to claim that political actors will never act illegally, nor that illegal actions of political actors will never be condoned by courts. It is only to claim that illegal actions will never be recognised by the legal system as illegal and, at the same time, legally right. The legal system can reconfigure itself in response to illegal actions by those with political power, but any such reconfiguration has implications for the legal system and, in turn, the political system.Footnote 33 The inability of the legal system to find illegality to be legal has implications for actors who occupy positions of political power. Being only able to respond, at best, by finding the illegal to be legal after all, the legal system never endorses a claim by the political system that illegality is justified; that, in other words, a political right to disobey law can exist as a legal right. Exporting a political right of disobedience to law involves difficult reconfiguration, and significant problems.
Within the political system, political actors may participate in civil disobedience campaigns, commit specific illegal acts, and claim their ‘right’ to do so, while at the same time they may find themselves executing decisions, participating in the formal political process and utilising communications whose meaning includes the affirmation that legality provides the parameters of legitimacy of political action, and that illegality is wrong. In this way, legality within the political system is similar to the justice of ‘treating like cases alike’ within the legal system. Legitimacy is an eigenvalue, since the political system constructs itself through operations that constantly communicate that collectively binding decisions are legitimate. And legality has become one of the stable, routine ways to account for the construction of these operations. As such, just as the legal system offers no way to execute legal decisions that does not include the meaning that like cases should be treated alike, so too the political system has no way to execute vast numbers of its own decisions without using communications that assert that legality is a significant basis for legitimacy. And in this way the political system generates a value, legitimacy, just by doing what it does. Thus, in utilising legal forms, political orders constantly reaffirm legality as a political value, and illegality as illegitimate.
If one had to give a sociological answer currently to account for the nature of any right of civil disobedience, one would say that that such a right has no obvious place in the formal political system, because it cannot take a legal form, and as such cannot easily operate within a system whose routine operations are executed on the basis that legality establishes the parameters of what is politically legitimate. With the political system's intense dependence on legality, it becomes very difficult to incorporate communications which claim a right to disobey the law. That said, perhaps not all of the political system necessarily operates in this way. Indeed, Luhmann has drawn a distinction in the political system between official power structure and more informal communications. What Luhmann points towards is the routine manner in which power is carried out through the political system using communications which rely on legality to establish legitimacy, and other political communications that challenge the results of formal political decisions and procedures. Does civil disobedience provide one example of such ‘political’ activity?
VII. Social movements and civil disobedience
‘[P]rotest differs from the format of political opposition within a constitutionally ordered democracy. In the latter, opposition is from the start part of the political system. This is exemplified by the fact that it must be ready to take over the government, or to participate in its running, and this has a disciplining effect. One may well engage in rhetoric and electioneering tactics … but ultimately one needs to be prepared to represent one's opinions and carry out policies as a government. Those who are protesting refer to ethical principles; and if one has an ethic, the question whether one is in the majority or the minority is of secondary importance. In all these respects, protest does not have to make any concessions. It purports to represent society against its political system. Thus it is not wrong to see the reason for the emergence of new style protest movements within the differentiation, and relative lack of vitality of the political system. The constitution serves to constrain the political system to itself. Protest movements consider this a provocation to provoke.’ (Luhmann, Reference Luhmann1982, p. 206, our translation)
Individuals protestingFootnote 34 often do so within social movements.Footnote 35 What is characteristic of the communications of such social movements is that they propose alternatives, but of a particular kind. These alternatives distinguish themselves from the status quo in being both alternative to the present, and the way in which that present has been generated. In adopting civil disobedience, social movements need to distinguish their form of protest – an act of disobedience (often referred to as ‘direct action’) – and their reasons for protest. They need to communicate that the political system in particular has selected their issue in a manner which they find unacceptable. Thus civil disobedience is both direct protest about a particular issue and protest against the political system in particular and, on occasion, other systems. Of course the political system is itself built on alternatives, those principally directed toward government and opposition; however, social protest movements (when not espousing revolutionary goals) express a paradoxical claim: they wish for alternatives to those current alternatives adopted by oppositions within the political system.
Civil disobedience protest involves more than disagreement, more than the coding of government and opposition. What it communicates is resistance. Such communication encourages individuals to join protests, and at the same time encourages subsystems (such as politics, law, mass media or the economy) to respond. In communicating resistance to the political system, protest movements tend to rely heavily on the mass media and their construction of ‘the public’.Footnote 36 The political system may well respond by constructing its version of ‘what the public want’, ‘what the public understand’, ‘getting our message across’, etc. In this sense the mass media clearly have effects upon the political system, though often unpredictable ones despite the attempts to ‘influence’ each other. In turn, the mass media often influences the nature of protest, in which events are created specifically for their newsworthiness, using drama, large numbers, dangerous exploits, etc. But the primary communication involves resistance and conflict.
Civil disobedience represents resistance and conflict, with disagreement being communicated as protest. Here, to succeed, the acts of protesters have to be understood as potential breaches of the law. It follows that some contradiction may arise. If civil disobedience protest is a claim that acts of law-breaking are justified, to achieve that claim others, and especially other subsystems of communication, must not in principle accept that claim. If the legal system did so, concluding that the protest was not per se illegal, then it would not amount to civil disobedience. Perhaps those undertaking acts of civil disobedience will use lawyers and attempt to avoid punishment, but at the moment of their protest they need to be confident that the legal system (and political system) will determine that they have acted illegally (or illegitimately) – without this it will be difficult for them to communicate their resistance. In the relation between the legal and political system some space for civil disobedience is organised through the creation of discretion. And here the legal system may ‘export’ the decision whether to prosecute or not to the political system. But even so, to communicate resistance, those who commit acts of civil disobedience need actors within the state machinery of official behaviour not immediately to accept the protesters' claims. How can protestors communicate resistance by choosing to breach laws known not to be enforced? Deliberate breaches of the law communicate resistance as long as there is some expectation of enforcement.Footnote 37 Thus, civil disobedience protest would be seriously undermined if the subsystems of politics or law adopted civil disobedience communications within themselves. For law, the consequence would be that the act of disobedience would cease to be illegal, and thus not ‘disobedient’. For politics, prosecutions would not be legitimate. Civil disobedience, as a form of protest, would thereby be seriously undermined.
This is not an argument that civil disobedience has no effect on the political system. Clearly we have many examples of notorious civil disobedience campaigns contributing to political change (see the essays in Roberts and Garton Ash, Reference Roberts and Garton Ash2009). But such protest, part of what Luhmann identifies as an informal countervailing system of communication, cannot simply be incorporated into those routine operations by which the political system constructs legitimacy for the vast variety of technical decisions associated with modern states. The ethical arguments on which civil disobedience is often based do not provide an alternative basis for legitimating those decisions. Thus the alternative that such protest offers, an alternative to the alternatives constructed within the political system as opposition to government, remains something to which the political system cannot easily respond. Again, there is a difference here between the legal and political systems. Since legitimacy is not the same as legality, there is scope for the political system to operate an administrative system that does not enforce legal regulations, but still upholds the normative expectations that they, at least symbolically, represent. However, systems theory accounts invite us to consider what selections are open to a system at any point; it does not claim that a system determines how its selections are made without recourse to its environment. System contingencies are readily triggered from outside those systems. Thus, one way in which civil disobedience protest as communication operates is as pressure on officials. Officials have decisions to make (such as whether to prosecute or not), and may well have regard to civil disobedience protest in making their decisions. But what they cannot do is justify such decisions by adopting the same communications as the protesters. It may be legitimate in many situations for a prosecutor to decide whether to prosecute or not. Legality establishes that discretion and at the same time the terms of its exercise, such as whether it is ‘in the public interest’. It follows that decisions not to prosecute can be legitimate, as they are legal, without having to recognise, let alone incorporate, the ‘ethical’ arguments or tactics of protesters. Civil disobedience may well affect any particular decision, but does so without disturbing the basis for legitimacy, namely legality, as it operates within the political system.
The other places in which civil disobedience protest can be readily incorporated into the political system are in that system's construction of ‘the public’, and its reliance on the mass media for this construction (structural coupling between the political system and the mass media). Such protest raises questions for politics, in terms of ‘the public's' likely reaction to it. The fact that the mass media rely on the routine operations of the political system for daily news does not prevent it from giving positive accounts of demands which are incompatible with those routine operations. This allows those engaged in the political system to consider whether or not to accept what is being demanded by those committing acts of civil disobedience without having to adopt the communications of civil disobedience as such; the latter being incompatible with the operations by which reforms will be put into effect: legality.
VIII. Conclusion
Many philosophical discussions of civil disobedience recognise that the claim of right involved is difficult to formulate without contradiction. However, through careful analysis these discussions attempt to reduce the level of any contradiction in order to construct a meaningful and coherent set of related propositions. This sociological account probably does the opposite. It recognises that paradox is implicit in constitutions, and some of their related claims of rights. However, this does not applaud incoherence or contradiction for its own sake, rather it demonstrates, or tries to demonstrate, how paradox is a driver of evolution within systems.
The same claim, a right to disobey law as a form of protest, operates within each system as a trigger for that system's evolution. In the case of the legal system, the main implication is the opportunity it gives for law's eigenvalue of justice (treating like cases alike) to be explored with a view to better interpretation of what the law actually is. It precipitates a change in the interpretation of the substance or scope of particular legal norms. In the political system it activates administrative practices associated with discretionary powers, in order to reconstitute political legitimacy. In the case of the subsystem of the political system, that of public engagement through protest, it anticipates an alternative to the alternative – a new form of politics which can only remain as an alternative, despite the reconfiguration that may ensue.
None of these systems can articulate a right to civil disobedience in itself, but only as a self-referring operation. As a sociological account of the communications that represent what many call civil disobedience, this paper demonstrates how what others might think of as the ‘moral character’ of such claims of right can only operate systematically in different systems, with different, but not insignificant, effects. Further, it has attempted to show how one of the paradoxes of constitutionalism, as represented in the practices of civil disobedience, engenders societal evolution within differentiated social systems.
