1 Introduction

Legal rules often conflict with each other. For our present purposes rules may be said to conflict if they are all applicable to one and the same case, while the legal consequences that would result from their application are incompatible.Footnote ¹ This happens within one legal system, as when theft is punished with at most four years’ incarceration, while theft during the nighttime is punished with at most six years,Footnote ² or when a municipality locally prohibits pubs to be open on Sundays, while that is allowed nationally. It also happens when the norms of a national legal system deviate from those created by an international organization, as when a Member State of the European Union taxes what is not taxable according to EU Law.Footnote ³ It happens when a contract violates rules of mandatory law, and also when state law collides with rules belonging to a particular religion such as the Shari’a or Jewish law.Footnote ⁴

With legal rules stemming from diverse sources, law has perhaps always been pluralist, with rule conflicts as a consequence. However, even if legal pluralism is an old phenomenon, it has gained in importance during the last decades of globalization. The resulting rule conflicts have also gained in prevalence, and it is important to pay special attention to the ways in which law can deal with them. The central question of this contribution is whether logic can be helpful in this connection. In his famous book Legal Traditions of the World, Patrick Glenn (Reference Glenn2014) already raised this question. After having described the main legal traditions, he addressed in the tenth chapter of his book the question how the different traditions could be reconciled, and in this connection he pointed to modern developments in logic which might make it possible to leave the traditional laws of consistency and of the excluded middle behind. This contribution is an attempt to explore these possibilities in a mainly informal way. The emphasis will in this connection be on the use of logical tools that are usually associated with the technical notion of non-monotonicity (Hage Reference Hage2005, p. 24).

At first sight, the logic of rule application is nothing else than an ordinary syllogistic argument (Alexy Reference Alexy1983, pp. 273–283; MacCormick Reference MacCormick1978, pp. 19–53). The facts of a case are subsumed under a general rule, and the conclusion that describes the legal consequence of the case follows deductively. For example, a rule reads that the major of a city has the power to issue emergency regulations. Boris is the mayor of Kropotkingrad. By subsuming this fact under the rule, the conclusion can be deduced that Boris is competent to issue emergency regulations.

An argument can only be deductively valid if it is logically impossible that all the premises of the argument are true, while the conclusion of the argument is false (e.g. Weinberger Reference Weinberger1989, p. 91). Because a conclusion of a rule-applying argument is deduced from the two premises, it can only be false if either one of the two premises is false. The falsity of the conclusion would imply that either the used rule is not valid, or the facts are different from what the second premise said.

Given this definition of deductive validity, a conflict of rules may lead to inconsistent legal consequences. Suppose that there is a rule that emergency regulations can only be issued in case of an emergency. If Boris issues an emergency regulation although there is no emergency, the power-conferring rule leads to the conclusion that the emergency regulation that Boris created is valid. The rule specifying the circumstances under which an emergency regulation can be issued, however, makes invalid the regulation by Boris. Going by deductive logic alone, the regulation is both valid and invalid, an unattractive outcome.

It will be obvious to most lawyers that the two rules should be combined. An emergency regulation can only be created in case of an emergency, and then the mayor of the city is empowered to do so. Obvious as this outcome may be, it can only be reached if the logic of rule application is not ordinary deductive logic. Lawyers often say that they handle such cases by means of ‘interpretation’, but that does not clarify much. We would like to have a more precise account of what goes on when legal subjects or legal decision makers must cope with conflicting rules. In the last few decades major developments have taken place with regard to the logic that can be used to model rule application (Schauer Reference Schauer1991; Gordon Reference Gordon1995; Verheij Reference Verheij1996; Hage Reference Hage1997, Reference Hage2005; Prakken Reference Prakken1997; Lodder Reference Lodder1999; Sartor Reference Sartor2005). The question that will be addressed in this paper is in what manner these modern logical theories can help lawyers to deal with conflicting rules.

2 The Roles of Logic

Before the real argument starts it is important to make clear what can be the role of logic in law. Lawyers sometimes have been a bit suspicious about this role. The suspicion can be recognized in the resistance of lawyers against ‘formalism’ (Hart Reference Hart2012, pp. 124–154) and ‘Begriffsjurisprudenz’ (Kaufmann and Hassemer Reference Kaufmann and Hassemer1977, pp. 92–95). Characteristic in this connection is the famous quotation from Oliver Wendell Holmes that ‘[t]he life of law has not been logic: it has been experience …’.Footnote ⁵

Most likely, this resistance can be traced to the fear that logic would dictate solutions for legal cases that are not desirable, but which are nevertheless taken because logic would ‘require’ them. Such a fear would be misgiven, because in the first instance, logic does nothing else than to indicate which conclusions follow from which premises. It cannot indicate that particular conclusions should be drawn; the most it can do is to show that a particular conclusion follows from certain premises. Moreover, deductive logic is as ‘empty’ as possible. Logical rules are formulated in such a way that the task of determining the outcome of an argument is as much as possible assigned to the premises of the argument. Deductive logic aims to do no more than to make explicit what was already implicit in the premises. This means that a legal decision maker who, for whatever reason, does not want to draw a particular conclusion that follows from his premises, can always decide to drop one or more of the premises rather than to draw the conclusion. Logic, in the first instance, informs us what follows from what, but it cannot force us to draw a specific substantial conclusion. Therefore, the fear of some that logic may require that legal decision makers take the ‘wrong’ decisions is not justified.

The limited role of logic also means that the hopes of some others, that logic may assist the legal decision maker in cutting difficult knots, is also misgiven. This still holds if the blunt instrument of deductive logic is replaced by the sharper edged instruments of modern logic. It is the task of logic – and this is more a programmatic starting point than an empirical observation – to be as ‘empty’ as possible. Logic should study which conclusions follow from which premises, but it should at the same time attempt to leave as much as possible of the factors contributing to some conclusion to be determined by premises that deal with the content of a domain. To phrase it by means of a slogan: if it can be a premise, do not turn it into logic.

However, there is also an additional function for logic, a function that has little to do with making inferences. Logic can help us in making analytical distinctions that are useful in formulating precise theories about law. We have already seen an example of this function of logic when we noticed that if the logic of rule application would be deductive logic, the application of conflicting rules to a set of cases may lead to incompatible legal consequences. Logic does not force us to adopt these incompatible consequences; it only makes clear that if we want to avoid them we had better look out for a different logic of rule application. Moreover, it can also help us to find alternatives for deduction that do more justice to good legal reasoning as it actually takes place. If the logic of rule application is treated as if it were a branch of deductive logic, it is likely that conflicting rules will lead to inconsistent conclusions for concrete cases. Through adoption of a non-deductive logic, which allows for exceptions to rules, the inconsistencies that seemingly result from conflicting rules can be avoided (see Sections 4.4 and 4.5). However, let me repeat the warning: do not expect from logic that it tell us how immanent conflicts of rules should be handled. The law determines which rule prevails if there is a conflict; logic can only show how a higher-level rule can solve the conflict. The content of this higher-level rule is still a legal matter.

3 Kinds of Conflicts

There are two ways in which rules conflict with each other. It is possible that one rule provides that something is the case, while another rule provides that something else is the case. For instance according to one rule, Blackacre is owned by Geraldine, while according to some other rule John has title to this ranch. Because rules impose facts upon the world – they make it the case that these facts obtain – such conflicts will be called conflicts of imposition. The conflicting rules impose incompatible facts upon the world, facts that cannot co-exist. On the assumption that the ranch Blackacre cannot have two owners, the fact that Geraldine owns the ranch is incompatible with the fact that John owns it.Footnote ⁶

The second kind of conflict has to do with rules which prescribe incompatible forms of behavior. Such a conflict occurs if a farmer is obligated to combat thistles, but is at the same time not permitted to destroy plants. Another example would be that a journalist is obligated to reveal the sources on which he based a controversial publication, while she promised her informant not to reveal his identity. Such conflicts may be called compliance conflicts.

Rules conflict with each other if their conditions for application allow a case to which the rules attach to involve incompatible legal consequences. Which legal consequences are incompatible may depend on other rules. Examples of conflicting rules are:

• - the rules that a bona fide buyer becomes the owner of a good that was transferred to him, and the rule that the transfer of a good by a non-owner does not pass ownership;

• - the rule that prescribes house owners to clean away the snow from the pavement before their houses and the rule that prohibits the use of snow blowers, if somebody can only clean the pavement with the use of a snow blower.

4 Rule Conflicts within a Legal System

To discover which logical tools are available to deal with conflicting rules that stem from different legal systems, it is useful to study first how rule conflicts are dealt with within single legal systems, because conflicts within a single system are the most frequent manifestation of rule conflicts. This immediately raises the question what a single legal system is.

5 Rule Conflicts between Distinct Normative Systems

Rule conflicts can exist between distinct legal systems and between legal and non-legal systems (Glenn Reference Glenn2014). Examples of conflicts between distinct legal systems would be conflicts between:

• - the national law of two countries, such as France and Argentina;

• - national law of a country and a personal (religion-based) legal system, such as a conflict between German and Canon law;

• - national and international law, as when a Polish rule violates the freedom of religion as protected by the European Convention on Human Rights.

The typical case of a conflict between law and a non-legal system is a conflict between law and morality.

The question that needs to be addressed in this section is how logic can help us in dealing with conflicting rules from separate normative systems.

6 Conclusion

The development of the internet, the rise of transnational law, the co-existence of different legal traditions and subtraditions, and globalization, all increase opportunities for conflicts between rules of different legal systems. One of the tasks of legal science is to deal with these possible conflicts and, as Glenn pointed out, the insights of modern logic are one place to look for assistance. Logic neither can nor should dictate a particular way of dealing with rule conflicts, but it can be of help by providing a conceptual framework that clearly defines when a rule conflict occurs and which tools are available to avoid these conflicts or to deal with them.

A conflict of rules occurs when it is possible that two or more rules attach incompatible legal consequences to a case. Within a single legal system, the possibility of rule conflicts is limited by adding scope conditions to the ordinary rule conditions. Mechanisms such as PIL extend this role of scope conditions to the relation between rules from different systems. Another limitation of the possibility for rule conflicts consists in the limitation of the powers of rule creating agents. Often it is impossible to create rules that conflict with already existing rules.

If there are actually conflicting rules, it becomes necessary to handle the conflict. By adopting a non-deductive logic for rule application it becomes possible to account for exceptions to rules. An exception to a rule leads to the result that a rule which is applicable in a case should nevertheless not be applied. Then the rule does not attach its legal consequences to the case. One reason for making an exception to a rule is that application of the rule would lead to legal consequences which are incompatible with the legal consequences of another applicable rule. An important insight in this connection is that inconsistencies because of conflicting rules are prevented, not by removing information, but by adding more information in order to handle the rule conflict (more is less).

If a conflict occurs between rules of different legal systems, there is logically speaking no need to deal with the conflict. It is logically well possible that an agent ought to do X from the point of view of one legal system, and ought to refrain from doing X from the point of view of another legal system. However, an agent who is subject to incompatible prescriptions from different legal systems is burdened with a dilemma. Legal system often come to the assistance of such individuals by referring to the rules of other systems, or by incorporating ‘foreign’ law.

1 A Model of Legal Pluralism

Legal pluralism emerged as a collection of approaches to understanding legal orders during the 1970s and 1980s.Footnote ³ There were two broad groups of ideas that marched under the pluralist banner. First, some writers developed a pluralist model through reflection on the ways in which different normative orders can interact. These accounts of pluralism started in the analysis of imperial legal systems, examining the ways these systems had accommodated, incorporated, and limited religious and local law.Footnote ⁴ Other scholars challenged the divide between ‘legal’ and ‘non-legal’ rules by drawing attention to the totality of rules that govern people’s lives: state law was not the only normative system that impacted on members of the state.Footnote ⁵ For these scholars, legal pluralism embodied the recognition that people were governed by different normative systems, systems which the legal orders of states often tried to accommodate or shape in some way. The second strand of legal pluralism – and one which is more directly relevant to the focus of this book – considered the potential for inconsistency, or contradiction, between rules, claiming that there can be contradictory ‘legal mechanisms’ applying to single factual situations.Footnote ⁶ These two strands sometimes come together: on occasion, it is the interaction of two normative systems that has generated the inconsistency in the rules that apply to individuals.

A significant challenge to the would-be pluralist is to show that their accounts of legal orders are distinctive. If it is the recognition that there can be interaction between different normative orders that is at the heart of pluralism, pluralism, as a distinct way of conceiving law, would appear unremarkable. Legal orders commonly recognise and even apply the rules of other systems,Footnote ⁷ and even the staunchest of legal positivists would acknowledge that other normative systems can bear on an individual and can have a practical impact as great as, or greater than, law.Footnote ⁸ Whilst the consequences of interaction between normative orders can be fascinating,Footnote ⁹ the recognition that they do interact is unexceptional. Similarly, the assertion that rules can place contradictory demands on those they address would be very widely endorsed. There is very broad, perhaps even universal, acceptance that the rules of different normative orders can impose contradictory demands on individuals. And, with the exceptions of Hans KelsenFootnote ¹⁰ and, perhaps, Ronald Dworkin,Footnote ¹¹ most would recognise that it is possible that rules within a single normative order can conflict; indeed, normally, it is one of the core tasks of the judge to bring such conflicts to an end.

Indeed, the challenge to those who want to talk of ‘contradiction’ or ‘inconsistency’ in the context of legal rules may be even more profound. Others in this volume debate the question, but the logical principle of contradiction is often taken to be confined only to propositions of fact; normative systems are commonly excluded from its reach.Footnote ¹² The factual claim that A exists and A does not exist violates the logical principle of non-contradiction; most people would think that one or other of these claims must be false. But on some understandings of morality, a person can find that they are subject to moral obligations that conflict; morality may require both action and inaction.Footnote ¹³ More strongly still, law is an artificial normative structure, a structure within which legislatures and courts can craft and impose obligations. Even if it is the case that the principle of non-contradiction applies in systems of ethics it does not follow from this that it necessarily applies in systems of law. Indeed, one of the foremost advocates of deontic logic, G. A. von Wright expressly excluded legal systems from its reach.Footnote ¹⁴ Though it may be desirable that rules within a legal order not contradict each other, this, according to von Wright, is not a logical truth about such orders.Footnote ¹⁵ Legal systems are artificial normative structures, and, as such, are not constrained to comply with all the dictates of logic – and the examples given in the remainder of this chapter will show how the institutions of law can, sometimes generate opposing rules. The challenge to those who talk of contradiction within and between legal orders is, then, to explain what this means in the context of law.

Bernard Williams tackled this problem by shifting normative statements into their descriptive equivalents, which might be termed compliance statements.Footnote ¹⁶ So, the normative statement ‘Albert must not kill Alberta’ is transposed to ‘Albert did not kill Alberta’, which is contrary to the rule ‘Albert must kill Alberta’ transposed to ‘Albert did kill Alberta.’ This device allows us to talk of contradictory rules without assuming the logical conception of non-contradiction can be directly applied to normative statements. The transposition also makes clear that the contradiction between rules lies not in their inherent truth or falsity, but in their subjects’ inability to fully comply with the rules: a claim that both had been fully complied with would necessarily be untrue. Whilst it is appropriate to judge assertions of the possibility of joint compliance as being true or false, it is not necessary to assume that the obligations themselves that generate these compliance statements are susceptible to attributions of truth-value. Consequently, the claim that contradiction can be found within and between legal orders is, or should be, less controversial than a general rejection of the logical principle of non-contradiction found in classical logic.Footnote ¹⁷

On both of these grounds – the possibility of multiple normative orders and the potential for inconsistency within a normative order – legal pluralism faces the challenge that its central theses might prove to be trite: if its central claims are very widely endorsed, it might not be all that valuable as a distinct way to look at legal orders.

In earlier work, I have argued that contradiction between legal rules becomes particularly interesting when it is found at the foundational level of legal orders. Sometimes, this inconsistency concerns the hierarchy of sources of law within a system and, as a consequence of this, the system may lack an institution that is legally empowered to resolve the inconsistency. There are multiple, inconsistent, rules of recognition that give priority to rival sources of law, and there are rival institutions that purport to have the final say about the content of the legal order. Consequently, pluralist legal orders contain a risk, which need not be realised, of constitutional crisis; of officials being compelled to choose between their loyalties to different public institutions.

The claim that legal orders can exist with multiple, unranked, rules of recognition is controversial. The rule of recognition is a central part of Herbert Hart’s account of law and – if it is to achieve the ends Hart set for it – such inconsistency must be rare. For Hart, the rule of recognition served to unite the rules of a legal system, providing a test by which other rules could be shown to form part of the legal order.Footnote ¹⁸ Each legal system therefore possessed its own unique rule of recognition,Footnote ¹⁹ and all the rules that could be identified through the application of the rule of recognition constituted a single legal system. The rule of recognition also provided an answer to the question of the continuity of legal systems, the means by which we can show that two different sets of legal rules are manifestations of the same legal system at different points in time.Footnote ²⁰ These sets of rules are aspects of a continuing legal system when the changes that occurred between the two sets of rules occurred in conformity to the rules of change identified by the rule of recognition. The existence of a single rule of recognition was necessary for the success of Hart’s answer to these central jurisprudential questions: it sought to explain how the boundaries of a momentary legal order could be drawn, and how a series of momentary legal orders can be identified as constituting a single legal order stretched over time.

Hart was not unaware of the problems that disputes over the rule of recognition caused his theory. In the context of revolutions and invasion he acknowledged the possibility of such a state of affairs: it was conceivable that two rival rules of recognition might operate within a territory, and yet only one legal system was in operation.Footnote ²¹ This was, though, a ‘substandard, abnormal case containing with it the threat that the legal system will dissolve.’Footnote ²² For Hart, such cases needed to be marginalised because if they were a common occurrence they would throw doubt on the success of the rule of recognition as the answer to the identity questions posed earlier.

Hart’s account of a legal system was non-pluralist, in the sense invoked in this chapter. The central case of a legal system contained a single rule of recognition, directing the law-applying institutions to the sources of law. Whilst Hart envisaged the possibility of pluralist systems, where there were inconsistent rules of recognition, he understood these as deviant cases. They constituted legal systems because of their similarity to the central case he presented, and were inherently unstable. If pluralist systems are more common, or more lasting, than Hart believed, the centrality of his non-pluralist account would have to be reassessed. Reflection on these, and other, problems with the rule of recognition has led Hart’s students away from the rule of recognition as the answer to the questions of the identity of legal systems: both Joseph Raz and John Finnis have advanced more flexible understandings of legal systems, which moved away from Hart’s rule-focused account and have made room for the possibility of pluralist legal orders.Footnote ²³

2 Examples of Pluralist Legal Orders

There are a number of reasons why it is valuable to identify pluralist legal orders. First, and building on arguments made in the previous paragraphs, such an exercise may help us answer general questions about the nature of law, questions of the type Hart sought to answer. If it can be shown that pluralist legal orders are comparatively common and, at least sometimes, are stable, it suggests that the general account of law provided by Hart requires modification.Footnote ²⁴ Second, the exercise may be of interest to those who study constitutions and the operation of states. A pluralist legal order may provide a distinctive form of constitutional ordering that might mitigate the corrosiveness of some profound political disagreements. Thirdly, it may prove of interest to those interested in the particular legal orders examined: comparing these systems with others experiences similar challenges may illuminate shared and significant features. The present chapter focuses on the second of these three tasks.

Three examples of potentially pluralist legal orders, or groups of legal orders, will be given in this section: the legal orders of the European Union, the legal order of Australia before 1986, and the current legal order of Hong Kong. As we shall see, in all of these cases the legal inconsistency that is characteristic of legal pluralism has, at its root, a political disagreement over the boundaries of the state. Here, legal pluralism is a consequence of the disputed relationship between a region – the Member States, Australia, and Hong Kong – and a larger constitutional entity – a putative European state, the British Empire, and China. This political disagreement then entails disagreement over the authority of the institutions of that larger entity over the region.

3 Conclusion

Ordinarily, inconsistency in law is a bad thing. Lon Fuller rightly claimed that the presence of inconsistent rules amounts to a standing criticism of the order that contains them, a blow against the rule of law.Footnote ⁶⁴ But there may be some situations where inconsistency is desirable; a way through which the law can accommodate competing political visions that are, on the surface, irreconcilable. This may be especially useful when the competing visions relate to the identity of the state. The three examples given in the last section show the different ways in which this model of legal pluralism may moderate the political problems raised by the emergence of new states by accommodating different visions of their constitutional futures.

In Europe, legal pluralism permits the Member States and the European Union to present themselves as states.Footnote ⁶⁵ The legal model favoured by the Court of Justice of the European Union is very similar to that of a federal state, with the treaties effectively dividing power between the European and national levels, and the Court of Justice acting as constitutional court, resolving questions over this division. The model favoured by many national courts, in contrast, is confederal, with the European Union enjoying a limited law-making power that is dependent on the agreement of, and support of, Member States.

In Australia, the shifting judicial view of the basis of the authority of the Australian constitutional tracked a shifting political consensus about Australia’s relationship with the United Kingdom. Whilst the judges lacked the legal power to effect a split from the old imperial legal order – their legal authority was grounded in that system – their constitutional position enabled them to open the possibility of a split. The inconsistent rules within the legal order reflected the older view of the Australian legal order within the imperial system and a newer view, which accorded the Australian system autonomy. Had the United Kingdom resisted calls for Australian autonomy, or if it had sought to start legislating for the territory, this new view would have allowed the courts to counter such actions, presenting the possibility that the judges would side with Australian institutions against the old imperial power.

Unlike Australia, there is no clear political consensus about the future of Hong Kong. The legal relationship between Hong Kong and China resembles, in some respects, the old legal relationship between Australia and the United Kingdom, but the political connection between the two is far tighter. And whilst the gradual constitutional separation of Australia and the United Kingdom was both anticipated and broadly accepted, Hong Kong’s constitutional future is far less clear. The inconsistency in Hong Kong’s legal order stands as a potential vision of the future of that territory, a future in which Hong Kong would enjoy practically full constitutional independence from China, an autonomy that would virtually amount to statehood. Whilst such a development is politically inconceivable today, it may become viable in the future – or, perhaps, presenting it as a possibility may help guard the more limited autonomy Hong Kong presently enjoys.

There are, then, at least three reasons why legal inconsistency of this type may sometimes be valuable.

First, this sort of inconsistency described here may amount to a political compromise; a tacit agreement to disagree.Footnote ⁶⁶ It allows supporters of European supremacy and supporters of national supremacy to both claim victory; conversely, and perhaps even more importantly, it avoids either constituency having to admit defeat. These inconsistent laws need not demand inconsistent action; the constitutional dilemma can remain unresolved, provided that each side exercises restraint. The emergence of pluralist legal systems with the European Union may provide a desirable compromise between the old models of sovereignty and constitutional supremacy, and the new claims to supremacy made by Europe. Similarly, in Australia the relationship between Australian law and imperial law was widely debated, whilst a split was inevitable the speed with which this occurred was debated: the judicial disagreement over the fundamental base of the Australian legal order tracked this political debate. Whilst the political debates in the European and Australian examples turn on finality, and cannot reach a compromise through the adoption of an agreed middle course, the pluralist model provides a compromise framework within which these inconsistent claims can co-exist in law. Provided that the practical conflict within this model remains potential, and actual disputes are avoided, this can provide a stable, even a long-lasting, form of settlement. The advantage of such a settlement is that it avoids unnecessary and potentially destructive conflict, and allows the protagonists to work together on beneficial projects where agreement exists.Footnote ⁶⁷

Secondly, the pluralist model can create flexibility in the law that may facilitate the emergence of a new constitutional settlement. The inconsistency in the European legal order may prove long-lasting, but it could be, over time, that the federal model wins out. The existence of inconsistency presents this model as a possibility within the constitutional order, an option that the system can move towards. This facilitative role for inconsistency can be seen even more clearly in the Australian example, where the beginnings of constitutional autonomy pre-date the legislative provisions that brought about this end.

Thirdly, these inconsistent rules, embodying competing supremacy claims, could provide a form of ‘constitutional self-defence’.Footnote ⁶⁸ A rule of constitutional self-defence empowers an institution to protect itself against other constitutional bodies. For instance, legislatures are given judicial powers in the area of privilege to stop the encroachment of the courts, and judges often run the administrative side of the court process. Sometimes these measures are more aggressive, giving one institution a weapon it can use against another: for instance, giving one legislature the power to strike down the acts of another legislative body. Competing claims to supremacy arm national and European courts with weapons that may help ensure mutual respect and restraint. If the potential conflicts caused by inconsistent rules of recognition were realised, with inconsistent rules addressed to individuals, all sides in the dispute would pay a price. Further, in the event of actual conflict, one side will, probably, emerge from the crisis as a victor: whilst it is unclear who will win, each side has an interest in avoiding the contest. The risks of actual conflict provide incentives on each party to strive towards harmonious interpretation of the law. It encourages the Court of Justice to interpret European law in a manner that will be palatable to national courts,Footnote ⁶⁹ and, at the same time, discourages national courts from blindly insisting on the primacy of national rules. In short, the competing supremacy claims may serve to create an atmosphere of co-operation between the courts, where each side has an incentive to strive to respect the position and tradition of the other. Similarly, in Hong Kong the possibility that the Hong Kong courts might reject a purported ruling of the Standing Committee of the National People’s Congress on the interpretation of the Basic Law might encourage this body to exercise restraint when engaging with Hong Kong.

1 Binary Logic and International Law of Self-determination

International self-determination law has long been acknowledged to be unhelpful to resolving secessionist disputes. It has become a truism to state that the law on self-determination is unclear, unhelpfully promising different things to states and to peoples. Emerging in the de-colonisation period as a legal norm to underwrite a move to colonial independence (somewhat after-the-fact), its post-colonial significance has been unclear. A set of binary oppositions underwrite this clash.Footnote ²

On one hand, the norm appears to promise self-government and nations, and on the other hand to promise to existing states that their territorial integrity will be preserved. Self-determination conflicts typically therefore see both sides to a conflict appeal to the norm whose application remains disputed. Outside the de-colonisation context the norm’s two pillars, respect for territorial integrity and a commitment to representative government for peoples, appear to clash. Rather than resolving violent self-determination disputes the norm stands accused of fuelling them.Footnote ³

In another binary distinction self-determination law rests on a notion of absolute indivisible sovereignty. This indivisible sovereignty is the prize for both the state and its non-state contenders in a tussle over whether the sovereignty of the unified status quo state, or the sought sovereignty of the secessionist people and entity will prevail. It appears that no zone of compromise is possible between two logically incompatible positions over the appropriate and legitimate demos, polis and territory of the state. Over the years, attempts to square the circle and render the norm’s two poles internally consistent has led to other binaries being reached for. For example, a binary division between a ‘people’ and a ‘national minority’ tries to distinguish between those who are entitled to ‘independent statehood’ and those who are not. The idea that there can only be one constituent people, and one ‘nation’ in the nation state, forces the logic that only the ‘whole’ peoples of a state are entitled to statehood, while non-majority ‘nations or nationalities’ are relegated to a status of ‘national minority’ and entitled to something less.

2 Multivalent Logic and Real Life

Even if these legal binaries were ever so absolute, increasingly none of them match up with reality in our contemporary world. Any idea that states contain only one ‘nation’ is difficult to maintain: states in which self-determination disputes exist are now often acknowledged to be pluri-national with different nationalities co-existing, and in cases where there are more than two nationalities, no one group may be in the overall majority.Footnote ⁴ We also increasingly accept the complexity of relationships among political, ethnic, racial and national identities. Attempts to distinguish between ‘peoples’ entitled to statehood and ‘nationalities’ entitled to something less – as something that can be determined by water-tight definitions – quickly run aground. It is almost impossible to create mutually delimited categories which draw a clear distinctions between these categories sufficient to deciding the different national entitlements that apply as a consequence. Indeed international law has had to bend to deal with indigenous identities by acknowledging indigenous peoples as ‘peoples’ even though they are not the majority within the state’s borders. Moreover, as Anderson so famously observed, all these identities involve a measure of self-identification and are ‘imagined’ as much as real.Footnote ⁵ We could go further and argue that in fact such identities fuse the binary of ‘imagined’ and ‘real’, as to imagine and act on is to make real – as those in former Yugoslavia found out in 1990 when the former socialist republic began dissolving (or its bits seceding – another of self-determination law’s binary choices) under the pressure of resurgent national rhetoric.

In summary, law’s binaries inadequately map onto more messy realities of identity claims that underlie secessionist disputes. These claims both appear absolute and exclusionary, but are also made complicated by people who understand themselves as neither and all of the identities on offer, and – even more complicated – who understand identity as having a fluid and relational dimension depending on what is being discussed or who it is being discussed with. The difficulty appears to be one of law not being able to adequately capture and appropriately frame categories which can map in a targeted way, to appropriate forms of participation in public decisions.Footnote ⁶

Law’s binaries, in particular its sovereign/not sovereign binary, also seem inapposite to our globalised world which places ever more limits on territorially-based sovereignty. Sovereignty no longer means what it did; even the most stable and monolithic of Western states accept that governance can take place out-with and across state borders. Regional and international organisations limit traditional state sovereignty in all sorts of ways. Post-conflict these organisations can even create explicit forms of ‘shared sovereignty’ post conflict, with interim international administrations.Footnote ⁷ Territorial borders can be understood as not necessarily territorially defined but conceptually movable depending on context (see US immigration desks in Shannon Ireland, or US military use of Guantanamo Bay in Cuba). Or borders can be conceptually made fuzzy in novel forms of alliance and cross-border arrangements, for example by giving state-like powers, including treaty-making powers, to sub-state entities. Border delimitation can be indefinitely postponed by providing for staged processes as the attempts to provide for arbitration in the Abeyi region on the border between Sudan and South Sudan, or the Brcko district of Bosnia Herzegovina illustrate – these last two borders having been left open by respective peace settlements.Footnote ⁸

Therefore, the contemporary legal world appears to hold out all sorts of institutional and territorial innovations, including ‘degrees’ of sovereignty, and porous borders, that are capable of breaking free from law’s binary logic because they enable breaking free from an ‘all-or-nothing’ form of territorially-based nation-state sovereignty.Footnote ⁹

3 The Move to Multivalent Logic in Self-determination Law

Law, of course, is traditionally taught in terms of binary logic that has only two values – yes and no (or true and false). Traditional logic traces its foundations back to Aristotelean reasoning that is bivalent. To some extent law’s commitment to bivalent logic is driven by its function as a tool of decision-making through legal adjudication. Legal adjudication has to come up with binary answers to legally framed questions. This pushes the art of effective advocacy into manoeuvring with respect to how the relevant question is decided on, framed and so answered.

However, the logic of statecraft and conflict-resolution is not binary but involves the tricky business of how to move parties from irreconcilable positions to exploring their mutual interests and finding some creative way of meeting them, by re-framing binary clashes over state identity, territory and sovereignty so as to address violent conflict that resolves around attempted ‘winner-takes-all’ sovereign solutions. The logic of conflict resolution must seek to attempt to find middle ground where there appears to be none.

There is, of course, a model of logic which speaks to this project, as centrally addressed by Glenn and this collection – multivalent logic, or ‘fuzzy logic’. Often attributed to Bart Kosko, fuzzy logic is the idea that instead of the Aristotelian A or not-A mode of thinking some things can be defined mathematically as including statements that are true to some degree and therefore having a value between 0 and 1. This more multivalent logic deals with the imprecision of language and the difficulty of capturing in words, concepts that are themselves relational and mutable. Multivalent logic also attempts to confront and negotiate between apparently incommensurate claims which in a sense pass each other by, because they speak to different truths or realities. As Patrick Glenn argued in one of the last pieces he wrote on law and logic, multivalent logic underlies and even underpins diverse legal and constitutional traditions, despite law’s ostensible commitment to bivalent logic, because it is necessary to holding different world and national views together:

In part because of our contemporary reconceptualisation of both identities and state sovereignty, our new world appears to hold possibilities for a more multivalent approach to self-determination law. In recent years, the more fluid concepts of sovereignty and a burgeoning peacebuilding architecture and technology have been argued to have produced a new approach to self-determination law that tries to move beyond polarised binary choices, and work more as a background facilitative norm.

A number of writers, drawing on slightly different data, illustrate the ways in which self-determination norm’s internal contradictions are now being understood to be being transcended in approaches which point to a more multivalent understanding of international law’s capabilities. To quickly review the key accounts: Klabbers, for example, reviewing post-colonisation decisions of the International Court of Justice in 2006, argued that the then scant jurisprudence illustrated a tendency to avoid direct pronouncements on self-determination or grants of secession in favour of endorsing a ‘right to take part’ or a right to negotiations.Footnote ¹¹ In essence, he saw a norm which was moving from a substantive outcome-focused norm, to a process oriented norm. He argued that the procedural right he saw evidence of, was bolstered by analogy to procedural theories of rights more generally. However, the extent of the normative shift he identified was unclear: it was unclear who had the right to be heard, and what that right entitled them to – that is, whether there was any requirement on the state to act differently as a result. In a sense, Klabbers’ account of the norm’s new trajectory could not fully evade questions of substance by being framed as a procedural norm.

Weller in 2009 similarly argued that new ways of resolving self-determination claims were being developed in practice in ways which were ‘dissolving’ the traditional distinction between internal and external self-determination.Footnote ¹² Analysing forty settlements he set out nine techniques by which self-determination disputes have been resolved through techniques such as constitutional recognition and accommodation. He too faced difficulties in his account of the norm’s shift, most notably while he notes that a sharp distinction between external and internal self-determination appear to be in the process of being dissolved, he viewed the developments as in essence a shift from external self-determination (for example secession), to internal self-determination (a right to more representative and inclusive government), and a political rather than a legal development.

A third perspective was provided in Kymlicka’s Multicultural Odysseys of Reference Kymlicka2007. Famous for having saved liberalism from incapacity to accommodate group rights with a concept of ‘multicultural liberalism’,Footnote ¹³ more than ten years on Kymlicka argued that the world had broken out in a rash of ‘multicultural liberal’ activity.Footnote ¹⁴ For Kymlicka, international law, international legal institutions and a wide variety of states in conflict had all turned to a practice of multicultural liberalism, as in liberal institutions re-worked to enable group accommodation necessary to resolving identity conflicts. However, many of his examples embraced neither liberalism nor cultural diversity but could more easily be analysed as rather flimsy conflict resolution devices between highly illiberal intolerant groups, coerced by international organisations and third party states into uneasy power-sharing arrangements and a commitment to elections. Kymlicka argued that both a practice and a legal requirement of multicultural liberalism was emerging. However, he found the international legal norms promoting multicultural liberalism were ‘ad hoc and unstable’ and difficult to roll out beyond the European context. He viewed international law in this area as at a cross-roads between general anti-discrimination law standards and more specific norms targeted at specific types of minority group – national minorities, indigenous peoples, racial minorities.

More recently, Tourme-Jouannet has argued that these types of development comprise a new ‘international law of recognition’ in which minority groups get forms of recognition through international legal provision for cultural identities, human rights, and reparations.Footnote ¹⁵ However, she too points to a potential ‘dark side’ of the new law of self-determination and unresolved questions of how international law will satisfy demands for recognition, or resolve questions of whose identities will be recognised, and how recognition of cultures and identities will be reconciled with other branches of law – notably human rights law.

All these writers, on their own and taken cumulatively, usefully set out the ways in which the practice and law of self-determination appears to be moving away from absolutist notions of peoples versus territory that fuelled rather than resolved violent conflict, to a creative ‘something else’. They agree that something new is going on in self-determination practice. They also agree that the ingredients of this something new in general terms include a move towards giving space to political negotiation aimed at group accommodation and new ways of achieving group accommodation through new power-maps for the state. They agree that this something new has implications for the law of self-determination and the international legal system more generally.

However, while each of the above accounts usefully attempts to re-frame the traditional binary logic of self-determination law, rather than moving beyond such logic, each commentator suggests merely an alternative binary logic that should prevail. For Weller, this new logic is one that moves from external to internal self-determination: new conceptions of self-determination work within notions of state sovereignty, by pushing forms of self-government that can be delivered within the framework of the new globalised state. For Klabbers the new logic is one that moves from substance to process: self-determination is reconceived as a process-oriented, rather than an outcome oriented norm. Similarly, for Tourme-Jouannet the new logic involves a move from international law as concerned only with state recognition to recognition also of sub-state groups. For Kymlicka the move is from forms of majoritarian liberalism to multicultural liberalism as enabling better accommodation of groups within state formations.

4 How to Be Both: Re-claiming the Excluded Middle

In contrast, I suggest that the new legalised practice is better understood as a normative shift towards a more multivalent approach to what the norm requires. The new normative approach, rather than involving a shift from one type of binary to another, involves an attempt to combine the different poles pointed to by the norm, and use creative institutionalisation to deliver elements of both. Rather than a new process right, a set of techniques of internal self-determination, or a right to multicultural liberalism, or recognition, I suggest that the phenomena described by Klabbers, Weller, Kymlicka and Tourme-Jouannet can best be understood as evidence of a new form of ‘hybrid self-determination’. This hybrid self-determination incorporates and addresses both internal and external self-determination claims in identity conflicts, but also operates to fashion a form of self-determination that is ambiguous as to whether processes of constitutional negotiation or substantive change in the nature and territory of the state, or a corporatist or liberal democratic state is on offer. In fact, I suggest that this understanding of current developments helps explain why these writers are all right in their quite different accounts of what is going on, and also helps explain why they all see the same ‘something new’ but understand it in quite different terms. A little like the story of the three blind men who each grasp a different part of the elephant – the tail, the leg and the trunk and believe its nature to be rope – like, tree – like or snake – like, when in fact an elephant combines all these attributes, so with the new self-determination each commentator looking from a different angle recognisably ‘sees’ the same thing, but grasps different bits of a norm which is best understood more holistically in terms of a new reach to multivalence rooted in an apparently impossible fuzzy logic.

The substance of the practice from this holistic point of view can be understood as four elements.

A procedural ‘right to be heard’. The first element of the new norm is indeed a procedural ‘right to be heard’ as identified by Klabbers and Tourme-Jouannet. The new practice of self-determination has involved states being prepared to negotiate with peoples or non-state actors. Increasingly, states have negotiated with their most fundamental dissenters – minorities, armed opposition groups, terrorists, or their representatives. Since 1990 negotiated settlements between states and non-state armed actors or political groupings have come to form one of the main ways of ending intra-state conflict. Leaving aside for now the legal status of a ‘right to be heard’, who have states negotiated with in practice? The answer appears to be: groups with long-standing connections to the territory who can coherently argue that they are structurally excluded from a state that has been defined in terms of just one of its pluri-national groups. The key question with regard to how to constitute the group that is to be heard, is to ask whether it has a coherent claim of being historical excluded ofrom the state’s foundation, in ways that traditional majoritarian democracy cannot remedy. The on-going group exclusion may happen because of ethnic voting patterns, the lack of a vote, or because of state ‘capture’ by powerful and elite groups.

The new practice of self-determination while responding to identity claims, is therefore not dependent on finding a globally agreed definition of what defines and unites the group in identity terms. The troubled question of who are the ‘people’s or ‘minorities’ is bypassed in favour of an approach that accepts the group’s self-definition, and responds to their capacity to destabilise the state and its constitutional order, politically, morally, and even militarily, by withdrawing their consent from any social contract.

The move to direct negotiations between state and sub-state groups has both a political and a legal basis. Politically, negotiations tend to emerge as a response to ‘mutually perceived hurting stalemates’ where both parties stand to gain more at the negotiating table than they can on the battlefield. Legally, in addition to the law cited by Klabbers, emerging international legal standards dealing with marginalised groups have begun to involve the participation of those groups in international law-making, so as to produce standards which then require effective participation as a substantive legal requirement of domestic constitutional processes.Footnote ¹⁶

The involvement of the marginalised in international law-making can in part be viewed as a response to the legitimacy deficits of international law-making as a purely interstate process. Those with an interest as beneficiaries of international instruments have been asserted to have a necessary place at the table of their negotiation: women in instruments affecting womenFootnote ¹⁷; victims in international criminal lawFootnote ¹⁸; indigenous peoples with regards to law on indigenous peoplesFootnote ¹⁹; minorities with regard to minorities.Footnote ²⁰ The resulting international legal instruments then require participation of the relevant group in relevant processes, including domestic processes as one of their key elements.Footnote ²¹ Indeed, the twinning of participation in international standard-setting to required participation of the excluded in domestic political decision-making is less an example of international regulation of the domestic constitution, and more an example of marginalised groups needing to impact on both international and domestic constitutional spheres together to impact on either.Footnote ²² Interestingly, in addition to standards addressing groups on the basis of their identity a number of legal standards now also address the need for inclusion and participation in peace processes and agreements and the institutions they establish, most notably UN Security Council Resolution 1325 (2000) requiring the involvement of women in peace negotiations.Footnote ²³

These participation standards point to the need for the exercise of international and domestic public power to be inclusive of marginalised groups who stand to permanently lose out in majoritarian systems. The standards operate to underwrite a right to participate as requiring creative revisions of representative democracy at the domestic level.

A right to elections. The second element of the new multivalent law of self-determination, is the commitment to a substantive right to elections understood in individual rights terms. Self-determination settlement terms invariably mention elections and aim for some form of liberal democracy, although they vary on matters of timing and transition. Again, the move to elections has an international legal basis: international standards now promote and enforce electoral representation.Footnote ²⁴ There are two distinct sources of law. The first is international human rights treaty provision for a right to free and fair elections along with the jurisprudence that has emanated from those treaty provisions. The second is the common standards found in the mandates of UN election monitoring missions.Footnote ²⁵ Fox argues that these two strands, as a matter of practice, abre converging to produce a new regime that goes some way towards a ‘right to democratic governance’ by providing detailed requirements for conducting representative democracy through periodic elections.Footnote ²⁶

A right to state redefinition and accommodation through fair participation in public institutions. The third element is a substantive right to mechanisms of fair participation through state redefinition. At the level of symbolism and rhetoric, self-determination settlements aim to provide a fundamental redefinition of the state, to enable the inclusion of the state’s dissenters and so respond to their allegations of state illegitimacy. The redefinition of the state’s nature and purpose is reflected both in constitutional language claiming that the state has changed, and in the articulation of new principles that will guide its new formation, such as inclusiveness or democratic renewal. These re-definitions defy any binary logic – they work in the sphere of the excluded middle. See, for example, the Belfast Agreement’s provision of a referendum both North and South of Ireland – across two jurisdictions subject to the consent of the Northern vote – as an exercise in constructive ambiguity between self-determination claims to united Ireland and claims to unity with the UK.Footnote ²⁷

‘Constructive ambiguity’ enables opposed sides to live with a common language. However, in reaching for this new language for how the state is to be understood, the apparently symbolic redefinition of the state does not remain symbolic. The very articulation of state redefinition has a performative power changing the nature of the state, as one which has a multivalent character as newly inclusive of even those who reject the state’s legitimacy and foundational narrative. To return to the Northern Irish example, the new Northern Irish state which is created is in some fundamental sense ‘bi-national’.

A right to a disaggregated states. Fourthly, at the level of institutional detail, self-determination settlements disaggregate power by re-conceptualising state governance and jurisdiction as being capable of being disaggregated into a wide variety of territorial, functional, and identity-based mechanisms and institutions, so that competing group demands for power can be accommodated. Territorial sub-division of government, consociational government, forms of personal jurisdiction, robust human rights machinery, and power-sharing in justice and security apparatus – used separately or even cumulatively – are some of the vehicles through which power can be disaggregated. The disaggregation of power addresses the state’s internal sovereignty. While state redefinition re-configures the conceptualisation of the relationship between state and people, disaggregating power involves re-configuring the state’s institutional formations, so as to give effect to the redefinition.

As with the first prongs of the new self-determination, this fourth prong is underwritten by international legal standards. A series of legal developments largely initiated as a response to intrastate conflicts, now underwrite a requirement for states to work to accommodate minorities that lose out in majoritarian conceptions of democracy, most notably convention and declaration.Footnote ²⁸ Minority rights standards were also promulgated as discrete responses to the dissolution of the USSR and former Yugoslavia, in an attempt to address the prospect of escalating intrastate conflicts in the states and entities that emerged from these dissolutions, in the form of Guidelines on recognition.Footnote ²⁹ These minority rights standards contemplate pro-active policies of recognition and accommodation as opposed to assimilation, emphasising throughout the ‘right to participate effectively’, with a positive duty on the state to facilitate this throughout public decision-making; similarly indigenous people’s rights standards talk of the need for free and informed prior consent to decisions affecting them.Footnote ³⁰ The standards, as discussed in detail in Kymlicka’s work, have indeed begun to suggest that states should move beyond representative democracy towards fair participation in accommodating national minorities and indigenous peoples.Footnote ³¹

A right to fuzzy sovereignty. The fifth element of the new self-determination is fuzzy sovereignty. While so far the elements of the new law appear to be consistent with a concept of ‘internal self-determination’, many new self-determination settlements tend to add elements of external self-determination to the settlement mix. Settlement terms often operate to dislocate power from the pre-existing demos of the territorially-defined state by making concepts of sovereignty ‘fuzzy’. Dislocated power addresses the state’s external sovereignty by attenuating it. In practice dislocating power involves ‘fuzzying-up’ the concept of sovereignty through two devices used separately or together: bi or pluri-nationalism – typically used in cases of ethnic conflict and secessionist demand, and international supervision – used in a range of conflicts.

A form of institutionalised pluri-nationalism can be achieved by acknowledging competing nationalisms as equally legitimate and creating institutional vehicles for governance that move beyond a traditional understanding of the state’s territorial limits. Mechanisms for achieving pluri-nationalism include: cross border governance; devolution of all powers and trappings of statehood to a sub-state entity; dual citizenship and provisions providing for ‘parity of esteem’ of different national groupings; and postponed referenda on secession which reconstitute the state as requiring on-going consent. These mechanisms can be combined in various matrices to produce arrangements that unravel the conventional links between sovereignty, territory and national citizenry, so as to produce a new set of relationships capable of accommodating national identities not just within but also across borders.

Again, international legal standards relating to minorities and indigenous peoples have moved in the direction of underwriting fuzzy sovereignty. As well as underwriting inter-group contacts within the territory, these standards refer to the right of ethnic groups to maintain cross-border contacts with ethnic counterparts in other jurisdictions, and even provide for states to sign bilateral agreements to this effect.Footnote ³² This provision parallels and underwrites self-determination settlement attempts to sever notions of ‘nationhood’ from territorially-based notions of ‘statehood’. Soft law standards have gone further and attempted to suggest a right to specific institutional arrangements to ensure ‘effective participation’ – such as informal and formal types of power-sharing, or autonomy regime.Footnote ³³

A second quite different way to dislocate power is through the concept of international supervision. International supervision conditions full sovereignty on the building of political and legal institutions that will disaggregate power, to ensure that all groups are accommodated. In the interim period (often left indefinite) international actors are given extensive roles in domestic political and legal institutions, with a view to enforcing and developing these institutions and with them, the peace agreement. International involvement can range from full international territorial administration (Bosnia), to international actors playing roles on peace agreement implementation bodies (e.g. South African Brian Currin implementing prisoner release provisions in Northern Ireland).

Again, normative revision has underwritten these novel forms of international-domestic hybridity. Full forms of international supervision can be authorised by UN Security Council, but the internal constitutions of international organisations also permit diverse mechanisms of intervention. There are moves to articulate at the international constitutional level a more fundamental revision of state sovereignty in the concept of ‘responsibility to protect’ which attempts to link sovereignty to a duty to protect citizenry, endorsing international intervention when that duty is breached.Footnote ³⁴ Again, these legal developments have been produced in response to situations of conflict, and underwrite the types of settlement terms understood to be required to end these conflicts.

5 How Hybrid Self-determination’s Multivalent Approach Works

Under binary logic, the excluded middle cannot exist; it is logically excluded by framing the only possibilities as being ‘1’ or ‘0’. In contrast, these new self-determination approaches attempt to find a middle way, in an attempt to transcend very real conflict impasses. The descriptions of Klabbers, Kymlicka, Tourme-Jouannet, and Weller are useful in reconceptualising the current directions of international self-determination law and capture that indeed ‘something new is going on’. However, as suggested earlier, they each point to a new (and different) way of understanding the binary in question which prompts moving to a different poll: from external to internal self-determination, from substance to process; or from majoritarianism to more group-oriented approach.

In contrast, I suggest that we better understand the ‘something new’ of self-determination law as incorporating all of these binaries rather than shifting between them, involving a move to a more ambiguous multivalent formulation. Indeed, understanding the multilayered approach to how self-determination is being reworked helps explain why each of the commentators discussed sees something new and similar at play, and yet see it in such different ways. This multivalent perspective understands the norm to function not as an imperative-requiring norm, but as a shaping-programmatic norm, which attempts to enable the re-creation of the excluded middle capable of transcend ing the binary division. The re-creation of the formerly excluded middle enables both a conceptual space and a functioning real political space in which the only possibility for reconstructing polities and relationships can be attempted.

In other words: this is a shift from binary logic altogether towards multivalent or fuzzy logic (to embrace that term as non-pejorative). The form of self-determination that results is multivalent across all of the binaries identified above. Interestingly, this multivalence can be argued not to be a new fashioning of the norm, but rather to be a new acknowledgement of the multivalence that generated a somewhat normatively incoherent and ambiguous norm in the first place. Legal articulations of the norm can be argued to capture a variety of different elements of what is understood to be important to sovereign statehood, when conceived of as itself a multivalent relational concept. The ‘incompatible pillars’ of self-determination law’s legal formulation as hybrid between a commitment to territorial sovereignty and representative government which have underwritten so much conflict, when viewed from a more multivalent perspective can be understood as capturing the need for a unified demos, polis, and territory to be closely linked to representative government if the state is to be effective, rather than as incoherent. As Musgrave argues, the norm always been understood to underwrite different ideas as regards ‘the self’ protected by the self-determination norm among different regional blocks of states; often the key resolutions which have documented the norm have attempted to capture these quite different understandings in one definition.Footnote ³⁵ Thus the norm can be understood to be multivalent in its foundation: a norm whose conceptions of legitimate statehood have different emphasis in different parts of the globe. Rather than a departure therefore, we can understand the new law of hybrid self-determination as in a sense embracing and creatively working with the norm’s inherent multivalence to give it a clearer practical basis. How then does the law work as a multivalent norm?

Hybrid between internal and external self-determination. First, the new norm is multivalent between external and internal self-determination both of which historic legal and political statements of self-determination valorise.Footnote ³⁶ Hybrid self-determination in a sense reconciles irreconcilable claims to internal and external self-determination, both of which were inherent in the norm’s formulations. Traditionally, self-determination as a legal concept was viewed in terms of external self-determination – the right to change the status of the state from colonial to independent, or from one state to two or more.Footnote ³⁷ Co-existing much more nebulously has been a concept of ‘internal self-determination’ – a concept that focuses on ‘the relationship between a people and ‘its own’ state or government, rather than state formation itself.Footnote ³⁸ The two concepts are arguably both present in legal articulations of the norm – and of course de-colonisation aimed to deliver both external and internal self-determination simultaneously.Footnote ³⁹ From this view, the norm itself is hybrid claiming to link the state’s external legitimacy to its internal legitimacy, as reinforced by the persistent idea that denial of internal self-determination in extreme cases involving wholesale and brutal denial of human rights, may give rise to a right to external self-determination.Footnote ⁴⁰

The difficulty with delivering on both dimensions of the norm has been that the two concepts are conceived of as leading to quite different remedies: statehood in the case of external self-determination, and ‘more representative government’ in the case of internal self-determination.

The new law and language of hybrid self-determination points to a deliberate incorporation of both internal and external self-determination language and mechanisms. This language aims to reconcile the twin pillars of self-determination law, namely the commitment to representative government and the commitment to existing territorial boundaries. The commitments to a unitary existing state and the possibility of explaining institutional arrangements as forms of federalism, devolution, regionalism and autonomy which are consistent with unitary sovereignty point to an ‘internal’ solution. However, settlement terms frequently establish an institutional fabric and language that dislocate power by deliberately blurring the question of where sovereignty lies. In transition at least, the arrangements cannot easily be analysed as consistent with statehood as exclusive control over a fixed territory and a people with a common understanding of their national allegiance.Footnote ⁴¹ Self-determination settlements point to a state which claims to leave its territorial and sovereign basis somewhat ambiguous. However, the arrangements are often also designed so that they can be analysed and presented as in conformance with the concept of territorial integrity. In remaining deliberately ambiguous as to whether internal or external self-determination is being delivered, hybrid self-determination simultaneously delivers both and therefore neither one alone. The old state both continues to exist and is replaced. Conflict resolution and legal norm are brought together.

Hybrid between offering process or substance. Hybrid self-determination is also hybrid as between notions of process and substance, or put another way – multivalent in not committing to either a purely procedural or a purely substantive norm, but rather to an approach in which process and substance are closely interweaved. In one sense it offers self-determination as pure process: a right to negotiate. Again there is a conflict resolution imperative: often the only common endeavour that can ground agreement is the agreement to disagree or to talk about talking. The only matter on which there is agreement is the concept of groups participating in decisions affecting them. The move to conceptualising self-determination as process appears to enable the difficulty of what self-determination remedy should pertain to be transcended. Rather than a move to a particular type of state configuration, what is on offer is negotiation.

Yet in its very articulation as process, hybrid self-determination becomes substantive; it requires the state to have a commitment to on-going negotiation of who it represents and includes. A substantive conception of the state is required that understands it as capable of leaving its relationship to questions of identity and participation open. In other words, the state is ‘re-founded’ as more explicitly multivalent and pluri-national, and is understood to involve consent as to developing processes of on-going negotiation rather than consent to a particular static political settlement. The commitment to ongoing dialogue requires that there be something to conduct dialogue about – and this requires the state’s internal configuration to remain up for grabs. Very tangibly, the commitment to process dictates a particular substance to institutional provision in the form of concepts such as power-sharing and future constitutional revision. These concepts are explicitly designed around the need to leave open fundamental contests as to the nature and ownership of the state.

Indeed, logically, a right to negotiate cannot end with the production of broad settlement terms, but by implication must continue into all future political decision-making. The democratic vehicle therefore becomes re-conceived as a set of institutionalised mechanisms for the effective participation in the totality of governance of previously excluded groups. Power-sharing governments, complicated federal and confederal arrangements, carefully balanced police, armies, judiciary, and civil service, postponed referenda on independence and autonomy, and even the machinery of domestic human rights protection together claim to offer the possibility of ongoing mechanisms of ‘fair and effective participation’.Footnote ⁴²

Hybrid between individualist representation and corporate participation. More controversially, hybrid self-determination is hybrid as between notions of representative and participative democracy, or – more problematically – between democratic and corporate forms of power. Representative democracy refers to the concept of democracy through periodic free and fair elections. Participatory democracy is a concept, variously described, that attempts to ‘reinvent’, ‘engender’ or establish ‘deep’, ‘new’, ‘strong’, or ‘cosmopolitan’ democracy, premised on increased citizen participation, in particular through a ‘re-visioning’ of the role of civil society in a renewed democratic polity. Finding examples in Eastern Europe and Central America, but also in contemporary Western democracies, political theorists have argued with increasing force that civil society should be conceptualised within a theory of democracy as having a role in Habermasian deliberation.Footnote ⁴³ This role addresses a democratic deficit occurring in transitions to democracy, but also in Western Democratic states, where participation through periodic elections no longer seems to accurately account for how citizens actually seek to participate, or to provide an adequate theoretical account of how public decision-making should be connected to the citizenry.

Like self-determination the concept of democracy can in fact be analysed as itself ‘hybrid’ between concepts of representation and participation. ‘Participative democracy’ contemplates a relationship with representative democracy, while ‘almost every advocate of majority decision is committed to some form of deliberative democracy’.Footnote ⁴⁴ However, the concepts of representation and participation locate the legitimacy of political decision-making in different places: representative democracy views legitimacy as rooted in the fact of the decision-maker being elected, while participative democracy locates legitimacy in the quality of the deliberation, the inclusiveness of the participation, or both.

While self-determination settlements invariably point a direction to elections, as we have seen they couple this with an attempt to secure broader and deeper ongoing participation in governance. The move in identity conflicts from straightforward majoritarianism to consociationalism itself endorses representative electoral democracy while also responding to its limits by offering groups guarantees of participation as a group. Power-sharing, proportional representation, and mutual vetoes replace ‘pure’ representative democracy, and so respond to the idea that majoritarian voting systems do not work in situations of conflict, where they stand to exacerbate rather than channel political divisions. The popularity of consociational-type arrangements lies in their ability to give effect to both representative and participative democracy. Consociationalism, understood as ‘responsible realism’Footnote ⁴⁵ aims to retain a commitment to representative democracy while recognising the need for group accommodation. Consociationalism’s power-sharing devices, often thought of as a tool for ethnic conflict, now feature heavily as a conflict resolution tool in all types of intra-state conflict.Footnote ⁴⁶ Principled and responsible – consociationalism recognises the legitimacy of representative democracy and individual equality rights. Realist it provides a mechanism that recognises that group identities and allegiances impact decisively on voter preferences. Rather than wishing this away, consociationalism offers a mechanism that recognises the group, but in a framework requiring elections and individual equality rights.Footnote ⁴⁷ In the move from majoritarian voting mechanisms, consociationalism requires a notion of ‘effective participation’ rather than numerical ‘representation’ as the better measure of democratic legitimacy.

Yet, lurking between even the most liberal of consociational of formulae, is a notion of power as linked to ‘corporate’ as in ‘group’ interests, rather than as linked to rule-based individualism. This is the ‘realist’ part of consociationalism. As with the other hybrid dimensions of the new self-determination, this form of hybridity has a conflict resolution dividend. The revision of the concept of democratic legitimacy to include group participation attempts to enable the renegotiation and transition of politico-military elites from private to public exercise of power, by giving groups the chance to buy-in to power structures. In short, it enables the elite pact which is necessary to any resolution of the conflict. Sometimes the group participation is little more than an illiberal corporate deal. However, sometimes the corporatist dimension is broadened to a range of actors, including civil society actors, articulated in terms of a concept of participation as buying time for elections to be held, while responding to the defects of representative democracy. With representative democracy impossible in the short-term, and the relationship between elections to public decision-making potentially problematic even in the long term, participative democracy becomes attractive as a way to legitimate governance that is otherwise free floating of a connection with the governed.

However, there is also a normative dividend. As elections and participation vie with each other in terms of which will deliver legitimacy and stability, emphasising both can avoid the deficits of each one. The settlement’s legitimacy itself can be rooted in a notion of ‘will of the people’ that at once acknowledges the primacy of elections while offering fair and effective participation for minority groups. This hybridity does not just create an on-going working polity but creates an ambiguous, multiple overlaid authorship for the constitutional foundations of the political settlement itself: another form of multivalence.

6 Maintaining the Excluded Middle: Multivalent between ‘war and peace’?

As Glenn has pointed out, states have always been more cosmopolitan than international legal accounts of nation statehood, or constitutional law accounts of the state’s foundation have credited (Glenn Reference Glenn2013). Indeed, in a sense his work ‘reclaims’ the multivalent nature of statehood itself. His perspective is useful because it helps to locate the new law of hybrid self-determination, and the fluid concept of statehood that it promotes, as less ‘new’ than it might otherwise seem. However, a critical question remains in contemporary efforts to construct or reconstruct states destabilised by violent political conflict, as to who and how the state is to be created and sustained as multivalent – and whether this is possible at all. Do binary narratives have such a powerful hold on the political imagination of state-craft, that any other way of doing business is almost impossible?

There is a logic of conflict resolution which pushes political ambiguities – actual or linguistic: it is often vital to achieving an elite pact to tell everyone that ‘they have won’ the war in order to create and sustain some sort of compromise political outcomes. However, in creating an ambiguous form of statehood and democratic polity, a fundamental question remains as to whether the excluded middle is a space of radical potential in which identities and binary ‘truths’ about the state can be re-fashioned; or disguises a dishonest victory in which one side really ‘wins’ its preferred default position but with the other side given some linguistic concessions. If neither victory nor transformation can be achieved, the excluded middle may still self-maintain as some sort of dead in-between détente space: a frozen conflict zone of ‘no war, no peace’, because while there is no common commitment to a shared state at all neither is there on-going capacity for victory in war.

As a response to this question I argue that it is only by understanding the multivalent logic of attempted new solutions that we are able to understand that all three outcomes are possible and that progressive improved outcomes are not inevitable but require ongoing commitment. The very construction of the excluded middle requires that the choice as to the future is left open, the only alternatives being reversion to one of the default binary positions of the conflict – but this leaves the possibility of reversions open and sometimes reversion to more exclusive conceptions of the state is achieved.

The critical project for intervention in support of the settlement therefore, rather than working to institute a vision of a more traditional state, is to work to build the radical transformative potential of the excluded middle by ensuring that its contours are shaped by a project of agency rather than the balance of power dynamic between those at the heart of the conflict. Often the only people with a commitment to using the space of the excluded middle in a progressive and transformative way are what Bouventuras de Sousa Santos calls sub-alterns – that is non-elite power-holders such as activists and international actors.Footnote ⁴⁸ However, often these groups are relatively disempowered with respect to implementing transitions, vis-à-vis elites empowered by political settlements founded centrally on elite pacts. It is important to understand that the excluded middle involves a space of on-going contestation between those who continue to push that the new settlement defaults back to one of the conflict’s binary political poles, and those who seek a transformative space at the centre. Unless the space of the excluded middle is valued and built as a space of transformation it is unlikely to sustain and evolve ‘naturally’. If the balance of power holds, the space may sustain, but as a dead zone of détente. If it does not, conflict is likely to return.

However, the difficulty of valuing the space of the excluded middle – even for those most committed to it – is profound: the excluded middle’s multivalent attributes make it difficult even for those who most fought for it, to love and support it. The difficulties can be understood in terms of the continuing power of binary logic of both law and politics.

As regards law: the space of the excluded middle remains under pressure from binary logic of international law and its organising concepts. Notably the concept of sovereignty retains a firm hold: despite globalisation states still exist with a considerable authority and the concept of sovereignty as a totalising rather than a relational concept remains powerful. The death of the sovereign state has often been greatly exaggerated and in moments of crises it shows remarkable capacity to reassert itself. Weak or insecure states often assert their own sovereignty in particularly monolithic ways internally and externally as a form of political posturing and elite protection. Also, the dead hand of Westphalia has a powerful grip because we have little other known way of doing things. How does one really create a state that is not committed to its own territorial boundaries and whose self-definition leaves its future existence open and contingent? The language of law can be used to create this texture to the political settlement as we have seen, for example in the bi-national language of the Belfast or Good Friday Agreement in Northern Ireland. However, what does this language mean for how we create and work out new political and legal institutions? For example, what does a commitment to equality of aspiration to statehood and bi-national citizenship mean for who can join the civil service, the police or other security agencies – how loyal do they need to be and what are the implications of bi-nationalism for the state’s monopoly on the use of legitimate force? What does bi-nationalism mean – say for when the (British) Union flag is flown over buildings in Northern Ireland that are still jurisdictionally in the UK, and whether the Irish flag is ever to be so allowed? Or what does ‘parity of esteem’ between different national groupings with different perspectives on the conflict mean for how we understand who were victims and who were perpetrators when we talk of the past? All these issues have bedevilled the implementation of the agreement in Northern Ireland and remain on a peace agreement ‘to-do’ list and subject to on-going negotiation almost fifteen years later.

Law itself can also at times play an unconstructive role with regard to the space, notably where its commitment to binary logic is strongest: in courts. Legal adjudication persists and even abounds in post-conflict jurisdictions, and plays a heightened political role given the constructivist on-going reform of legal and political institutions. As a result, often courts find ways of adjudicating on peace agreements and hybrid self-determination itself. So courts, for example, have been asked to bring peace agreements to an end or gut their central power-sharing provisions by striking them down in terms of constitutions, legislation or international human rights law or international criminal law.Footnote ⁴⁹ Asking binary-framed questions of courts invites them to give binary yes or no answers that de-construct the multivalence at the heart of the agreement: does the new fuzzy sovereignty arrangement violate the Constitution’s commitment to ‘unity’; does the power-sharing mechanism violate individual electoral rights? So while the legal language of peace agreement often is perfectly able to create and hold the excluded middle, court decisions made according to binary logic risk unravelling this space by unwinding the bargains that created and sustain it.

The second difficulty of sustaining and using the excluded middle as a space of transformation of the political settlement is political. The political order of the excluded middle is not accepted as anything more than a transitional space by the main parties to the conflict. Most of these parties aim to transit back to their default ‘win’ position and tend to have entered peace negotiations in a hope of that getting them more than they want at the negotiating table, than the war trajectory they were on. Elites seldom enter negotiations because they have somehow ‘converted’ to the notion of peace. However, the excluded middle is also an uncomfortable space for normative actors – internal and external – who include those who most politically support a move from the binary certainties of conflict. The political orders created in the excluded middle are in a deep sense hybrid between normatively attractive and normatively unattractive actors and positions. These are indeed zones of ‘no war, no peace’, and often operate as hybrid political orders that are hybrid between normatively principled liberal democratic institutions and informal power arrangements between elites that often shape and limit how these institutions work and even bypass them.

Indeed, some of the most interesting work on post-conflict peacebuilding has independently of hybrid self-determination analysis, reached to a notion of hybridity in trying to understand the orders that emerge and the normative ambivalence that they create. So Boege, et al, and Mac Ginty, all point to ‘hybrid political orders’ rather than ‘liberal political orders’ as the almost inevitable outcome of trying to marry liberal democratic structures with forms of corporate accommodation of power blocks.Footnote ⁵⁰

Here a biological analogy rather than a mathematical one, has been reached to, to similarly capture both a creative innovative attempt to ‘be both’ a liberal solution and a corporate power accommodation, and therefore ‘neither’. The metaphor of hybridity also speaks to the dual possibilities of a brave new way of being, and of reversion to either of the mutant parents. In essence the biological metaphor like that of the multivalent or fuzzy logic metaphor, attempts to capture the complexity of political orders that emerge from the attempt to ‘be both’ in a way that speaks both to the creative fluidity of the order, but also its darker unresolved nature which are generated by the inherent tensions that multivalent value systems incorporated into the heart of the new order.

I suggest that situating hybrid political order analysis in a larger theory of multivalent self-determination logic helps to explain how these hybrid orders emerge, not as accidental compromised by-product of liberal peacebuilding failures, but rather as a part and parcel of the new fuzzy logic of hybrid self-determination techniques. The difficulty is, that having understood fuzzy logic and the reconstruction of the excluded middle as an important project for stopping the war, internal and external interveners find the space very unsatisfactory in terms of the types of transformation they would like to see the space enabling. The structure of the settlements means that each group engaging with the new political settlement – including international actors – often engages in an attempt to return to the binary pole they are normatively committed to. For the parties, this will typically be their exclusive version of the future, and a move to a limited access order in which they are the dominant (and dominating) group. For international actors it will typically involve attempting to move to a majoritarian form of liberal democracy – or open access order - in which the corporate dimensions of the ‘deal’ and indeed its ‘fuzzy statehood’ are pushed to evolve into something more recognisable in terms of Western liberal statehood.

Peacebuilding is strewn with the dilemmas for international actors that ensue, most of them focused on the question of how to sequence and the transition from fuzzy deal to liberal statehood. When should international actors permit amnesty and when insist on accountability? The first appears essential to the elite pact, and the second to any long-term rule of law. Should international peacebuilders insist on elections or implement an ‘institution-building first’ approach? Either choice risks stabilising an exclusive elite pact, without any real transformative agenda or possibilities, and either choice is controversial in terms of the normative commitments to human rights and democracy of international organisations. These dilemmas are profound because failing to sustain the elite pact means that the liberal peace will never be reached, while sustaining the elite pact also appears to institutionalise the very forms of exclusion and division which require to be transcended if a truly peaceful society is to be achieved.

Whether the peace process succeeds or fails in practice often depends on whether the space of the excluded middle can be sustained or not. However, with no-one particularly committed to it, the excluded middle becomes sustained not by the agency of any of the actors, but by the balance of power between them as they attempt to pull back to their preferred default poles. As noted, the excluded middle often self-sustains but less as creative space and more as a space of ‘frozen conflict’, because the balance of power coupled with war weariness may continue to promote it as a space of détente. Those committed to ‘good enough governance’, can perhaps contemplate this as a form of success, however the excluded middle as a space of frozen détente is not an easy one in which to peacebuild or deliver good government because it is characterised by instability and constant threat of a return to violence. As Barnett, Fang and Zürcher argue, in practice international peacebuilders often remain committed to their normative ideals which encourage them to undo the excluded middle of hybrid political and legal arrangements as they attempt to promote a more traditional vision of liberal peace, while recognising that they must ‘trade’ with local political actors operating within local power structures in ways that frustrate liberal peacemaking ambitions. Often this is the only space in which peacebuilding is possible.Footnote ⁵¹ As a result, peacebuilders tend to engage in ‘compromised peace building’ in which an implicit ‘peacebuilders contract’ sees peacebuilders in essence transacting with local actors to have norms articulated by local actors, in return for working with what are contrasting local structures and incentives that pull in quite different directions from international norms.

7 Understanding How to Assist Political Settlement in Divided Societies: From Theory to Practice

What then are the policy implications of hybrid self-determination for peacebuilding? Or is this merely an intense theoretical discussion as to the adequacy of mathematical and biological analogies? I suggest that the policy implications are profound, although clear policy ‘take homes’ remain elusive. At its heart, the appeal to understand the conflict resolution project of constructing and maintaining the excluded middle is an appeal to understand and grapple with the hybrid compromised space in which conflict transformation takes place, rather than wish it to be a space with some ‘better nature’ than it has.

At the outset, it is important to recognise that wars only end in three ways and that this is what produces the need for the excluded middle to be created. The first way to end a war is to let the parties fight to conclusion and hope that the balance of power outcome that sees one party prevailing sustains a stable future. The second way to end a war is for outside actors intervene militarily to change the balance of power by assisting one party to win in the hope that the best party wins and that this assists a more sustainable future. The third approach, is for the parties with or without international assistance, to negotiate a messy moral, political and legal compromise capable of ending the fighting by creating new political and legal institutions in which to share power and to attempt to work out their differences less violently. The first two options involve permitting, or assisting one of the poles of the binary logic of war to win out. Neither of these options have a good record with regard to liberal democratic outcomes or a just peace. Moreover, while ‘giving war a chance’ has advocates, the contemporary new ‘global marketplace’ of conflict and external intervention indicates that rather than bringing clean victories leaving wars to fester may risk more complex regional and even international conflicts.Footnote ⁵² From a normative point of view, the most powerful rather than the most just and democratic, tend to win, but in fact many of the conflicts in question prove intractable for both sides. The third approach involves trying to re-create the excluded middle. For all the mess and normative compromises of the excluded middle, this last solution is usually more acceptable to those who believe in justice and peace, and has a better evidence-basis for success, than either of the two binary-focused alternatives.

However, sustaining the excluded middle is an immensely political project, which involves international interveners being much more self-conscious from the outset about the nature of the project and the political compromises (for themselves and others) that they are caught up in. There currently appears to be some appetite for a more self-consciously political approach among a varied range of international actors. International interveners, from aid agencies to peacebuilding institutions are all expressing a strong sense of disillusionment with post-1990 practices of assistance to fragile and conflict-affected states based on understanding that they have failed to adequately engage with both their own internal political constraints, and the local political power realities in the contexts in which they intervene. This disillusionment is often expressed as a sense of being ‘outwitted’ by local actors and politics that they have field to sufficiently understand or anticipate. The institutions of good government that international interveners work to create often turn into something quite different, even when models of good practice are followed. Worse still, their own analysis and political will appear inadequate to an effective response. Rather than a failure of ‘lessons learned’, policy documents exhibit a sense that the real problem is more fundamental: they do not know what lessons to take from past failures and are not convinced that doing more of the same a bit harder will do the trick, and so they search for new analysis that will enable them to find a new approach altogether.

I suggest that if new ways of engaging with context-specific political realities are to be found, it is important for policy-makers to understand both the potential and the limitations of the excluded middle as a space for transformation which is deeply compromised from the outset. From this perspective, practical projects of intervention would seek to:

1. (a) Understand the ways in which peace settlements establish an excluded middle involving compromised institutional provision, which will remain under pressure to revert to one of the conflict’s binary poles and design strategies which anticipate how international intervention may be likely to tilt the deal in unhelpful ways

2. (b) Understand the ways in which political bargaining by the main contenders for power will involve the strategic use of normative arguments, for example appeals to elections, and human rights, and the institutions and language of democracy, to engage in forms of reversion to exclusive forms of governance. International actors need to consider how democracy promotion may play out in a world in which democratic change is not the key goal of elites nor what they understand to be at the heart of the deal they have signed up to

3. (c) Understand that sustaining the space also risks freezing the conflict rather than transforming it, and evaluate the types of process of development and re-negotiation that can assist in keeping the space for transformation open and forward moving

4. (d) Analyse and evaluate the mechanisms, institutions and constituencies which have transformative potential and ensure that they are supported post-settlement

5. (e) Analyse and evaluate the limitations on transformation, imposed by the need to keep powerful actors on board

6. (f) Build capacity to be flexible and opportunistic as regards opportunities with transformative potential

7. (g) Evaluate and be realistic as to how supporting transformation involves political will and be aware that interventions are themselves highly political. Interventions need to be made with sufficient political will and capacity to engage in forms of power reallocation that will be resisted

Admittedly how to do any of these things this needs further thought and differentiated policies by different interveners. But I suggest that moving forward requires the modalities of post-settlement engagement to be re-thought. The key shift required is to move from a project that is less goal-oriented in terms of particular political outcomes, to one that is understood more as a project of looking for and supporting open political spaces of deliberation by pushing back against any return to conflict default positions. Such an approach requires a more sophisticated analysis of the ways in which local elites use the very language and institutional apparatus of the peace settlement to achieve their old conflict goals.

As an alternative, it may of course also be possible to muddle through in the fog of détente with only a half belief in what one is doing – provided that the balance of power between warring factors sustains. As Barnett, Fang and Zürcher suggest, this may not be a completely bad solution: a very imperfect peace may still be eminently more desirable than a very perfect war. Sometimes muddling through is better than the alternatives. However, I suggest that even muddling through could benefit from increased self-awareness of international actors as to the nature and importance of the excluded middle space of conflict resolution and the imperative to understand both the importance of multivalent thinking, and its peculiar state-building dilemmas. At least such thinking would avoid navigating post-conflict terrains with a degree of bad faith rooted in the inability to reconcile normative desires to practical political possibilities. This is a useful second best approach, unsatisfactory as it sounds and much as I would prefer something a little more ambitious to be attempted.

1 Legal Orders and Legal Differences

Through most of Western legal history, differing laws were not seen as in conflict. Medieval legal orders were harmonized in their diverse operations on the same territory through multiple interpretive devices.Footnote ³ The notion of a common law was vital to this process and the (multiple) common laws of Europe all functioned in the same manner, acting as supplemental or relational laws, yielding to all manner of local laws and local particularities.Footnote ⁴ They were multivalent in contemplating the operation both of their own norms and those of particular, differing, laws within the same territory. The common laws, in yielding, were non-monotonic in character. With the expansion of European law overseas, this essential character of the European common laws (most notably English, French, Dutch and Spanish) was retained and magnified. Multivalence and non-monotonicity have therefore been the major logical devices of western legal traditions until approximately the nineteenth century.

In some cases, this multivalence continued to prevail even through the nineteenth century process of nationalisation of law. This was the case for Equity in England. The maxims of Equity (‘Equity follows the law’; ‘Equity supplements but does not contradict the common law’) were multivalent, interpretive devices for assuring the ongoing co-existence of the common law and Equity as distinct legal orders. The same can be said today. The trust itself may be seen as a multivalent legal instrument: the binary choice between owner and non-owner, accepted by civil law, canon law and common law courts was rejected by English Chancery judges and the intermediate notion of equitable ownership thus developed.Footnote ⁵ This raises the interesting question as to what then happens once an included middle – a formerly inconceivable middle ground between some proposition and its negation – has been formalized.

Legal differences were the object of major efforts of elimination within national states over the course of the nineteenth and twentieth centuries. By mid-twentieth century, books on legal ‘antinomies’ in the philosophy of law found few actual examples in domestic law.Footnote ⁶ They were seen as resolved by well-established methods of application of superior norms (for example, constitutional norms), or choice of the law later in time, as general techniques of interpretation. Conflicts were recognized between national legal systems, however, and by mid-nineteenth century an entire, rule-based discipline of the ‘conflict of laws’ had emerged.

2 Conflict of Laws as a Modern, Statist and Binary Construction

The expression ‘conflict of laws’ was invented in the Netherlands by Ulrich Huber in the late seventeenth century. Huber was also the father of the modern notion of the territorial application of state laws. State law thus in principle precluded the operation of other laws on state territory, and no law of another state could be applied within a given state. Huber was an important figure in the obtaining of Dutch independence from Spanish rule, and the assertion of the territorial supremacy of Dutch law on Dutch territory was an important feature of the political and military struggle for Dutch independence. Huber’s teaching was accepted by Joseph Story in the United States, and Story in turn influenced Friedrich Carl von Savigny in Europe in the mid-nineteenth century. Foreign law could thus be applied only as a gesture of ‘comity’ by the state of application.

Differences between national laws were thus conceptualized as ‘conflicts’ as an important feature in the construction of modern states from the seventeenth century. The construction of a ‘conflict of laws’ followed recognizable laws of classical logic. The law of identity (A is A) reinforced the separate existence and autonomy of each state, defined as a corporate person operative within a geographically limited territory. The law of non-contradiction (not [A and not-A]) dictated both the internal consistency of national laws and the impossibility of application of two national laws to a single, transnational problem.Footnote ⁷ Binary choice was necessary between conflicting laws. The law of the excluded middle (A or [not-A]) prevented any included middle between the laws seen as in conflict. It was legally impossible to admit of any solution other than one or the other of the two conflicting state laws and legally impossible to admit of any compromise between the two.

A Savignian-inspired discipline of rule-based choice of law (‘the conflict of laws’) was thereby generated. The rules in question were those of ‘conflicts justice’ as opposed to ‘material justice’ and their sole function was to allocate given relations to particular national legal systems for their regulation. Tort liability was thus to be governed by the law of the place of the tort; rights in real property by the law of the situs of the property; the form of marriage by the place of celebration of the marriage. All legal space was thereby nationalized, and the relations of national legal orders were defined as inherently conflictual in nature. The general conclusion was compatible with Hegelian notions of the ultimate destiny of the state being defined in terms of belligerent relations with other states. This position is the positive law of jurisdictions such as Germany, Switzerland and Italy, in which conflict is presumed, such that even in the absence of any allegation by the parties that there is any difference of national laws which affects the resolution of their case, the judge must on his or her own initiative apply choice of law rules and decide which law is applicable to the case.

Conflictual perspectives on the relations of legal orders are today, however, in decline, as different and potentially contradictory legal orders multiply. Many written constitutions contain rights seen as conflicting; many constitutions of deeply divided societies are unable to resolve differences, and leave their reconciliation to ongoing, casuistic techniques. The EU is characterized not by hierarchy but by ‘related hierarchies’. Religious legal orders are recognized by many states that guarantee freedom of religion. Multivalent or paraconsistent forms of logic are therefore emerging.

3 Multivalent Logic and the Conciliation of Laws

The multiplicity of contemporary legal orders is derived from the irrefutable claims of legitimacy that they make. Legal unity is thought of no longer as that which prevails within a state, but that which justifies recognition of a particular legal regime. The legitimacy of the EU cannot be denied by a national judge. No constitutional guarantee of a given right can be rejected by a national judge because of its inconsistency with another right. The discipline of the conflict of laws is now becoming subject to multivalent forms of analysis. A good argument can be made, moreover, that this was the case even throughout the nineteenth and twentieth centuries, although bivalence largely prevailed.

The constant feature of multivalence in the conflict of laws is found in the notion of public order. No choice of law rule was to be automatically applied, since the law it pointed to could be found to be objectionable in terms of the public order of the forum. Conflicts rules were ‘blind’ and the public order reservation was therefore a hedge against material solutions that were highly incompatible with the law of the forum. The public order exception, however, is an indication of multivalence. Both the governing law and the law of the forum are potentially applicable. The rules are non-monotonic and some form of reasoning is necessary in order to decide on their application or not. An included middle is opened. The French Court of Cassation has recently decided that whether an Islamic talaq divorce (given exclusively by the husband) should be recognized depends not on its compatibility with the constitutional principle of the equality of the sexes, but on whether the divorce makes adequate financial provision for the spouse.Footnote ⁸ Material justice emerges in the process of deciding on the application of an exception to the rule.

Further exceptions to the application of choice of law rules are now being recognized, all indicative of a larger reasoning process and all tolerant of multiple legal orders and their solutions. Such reasoning is now codified in the Rome I Regulation on The Law Applicable to Contractual Obligations, providing that ‘mandatory’ rules of a third state, not the forum state or the state whose law is in principle the governing law, may be applied.Footnote ⁹ Three laws are potentially applicable. A further process has also become evident of material solutions justifying a particular choice of law. The law to be applied is thus, for example, not the most geographically proximate law, but the law which best assures the interests of a child, or maintains the validity of a marriage, or assures the legitimation of a child.

In all of these latter cases, an included middle is being found which controls the resolution of the case in one manner or another. The included middle is also recognizable when arbitrators decide as amiable compositeurs and apply non-state law to the resolution of contractual disputes. The possibility of choice of non-state law in contractual matters was debated seriously in recent reforms of EU law and is presently before the Hague Conference on Private International Law.Footnote ¹⁰

The move away from formal rules of conflict of laws is most evident in the United States, where rules have been abandoned entirely in some states, in favour of an interpretive process which may involve examination of the policies of both of the interested states (‘comparative impairment’) or even involve choice of the ‘better law’. Professor von Mehren of Harvard Law School has advanced the proposal that in a simple dog-bite case, where a dog crosses the border between a strict liability state and a ‘one-bite allowed’ state, and bites someone, that person should recover (though depending on the precise circumstances of the case) approximately half of the damages they would receive in the state of strict liability.Footnote ¹¹ The contradictory laws of both states are both potentially controlling; neither is necessarily controlling and no binary choice between them is necessary; an included middle is opened which allows the appropriate solution according to the circumstances of the case. This type of solution was well known in the Middle Ages;Footnote ¹² it appears likely to re-emerge again.

4 Conclusion

The decline of the notion of conflict of laws, as evidenced above, appears to indicate two things.

1. (i) Laws do not conflict, as the editors of Dicey and Morris once concluded. Multivalent or paraconsistent reasoning has the potential not only of resolving conflicts, but of eliminating them. It is a conciliatory form of legal reasoning.

2. (ii) Multivalent legal reasoning is not inconsistent with binary legal reasoning or, put differently, multivalent legal reasoning includes binary legal reasoning. Given the current state of the discipline of ‘conflict of laws’, it is possible but not always necessary to follow a choice of law rule, Multivalent logic does not require a binary choice between itself and binary logic.

1 Introduction: Law and Logic

In his classic Legal Traditions of the World,Footnote ¹ Patrick Glenn documents and explores the fact that the world contains many different legal traditions, often inconsistent with each other; indeed, even a single tradition can contain different sub-traditions that may be inconsistent with each other. Moreover, these traditions may interact with and inform each other in complex ways. In chapter 10, he raises the question of how to view this matter from the perspective of formal logic. The point of this essay is to address the question.

2 Many-Valued and Modal Logic

In a section of chapter 10, Bivalence and Multivalence, Glenn suggests that many-valued logic, and particularly fuzzy logic, may provide what is required. I think that this is the wrong machinery for the job. Let me explain why, before I explain what I take to be the right machinery.

In standard logic, there are just two truth values, true and false. In many-valued logics, there are more than two. Thus, in fuzzy logics, there is a continuum of truth values, all the real numbers in the interval between 0 (completely false) and 1 (completely true). Moreover (and this is crucial), the truth value of a complex sentence, such as a conjunction (∧) or a negation (¬), is determined completely by the truth values of its parts. Thus, for example, if we write the value of the statement A$A$ as |A|$\left|A\right|$, in standard fuzzy logic, |¬A| = 1 − |A|$\left|\neg A\right|=1-\left|A\right|$, and |A ∧ B| = Min(|A|, |B|)$\left|A\wedge B\right|=\mathit{Min}\left(\left|A\right|,\left|B\right|\right)$.Footnote ²

If this machinery is to be applied to the situation concerning different legal traditions, it must be explained what each of the plurality of truth values means in this context. Nothing seems to be appropriate. The only things that suggest themselves are to interpret some value as both true and false, or – perhaps Glenn’s favourite – half true/false (0.5 in fuzzy logic). These understandings misdescribe the situation, however. Calling something half true/false – call this value i$i$ – is the wrong way to characterize a claim, A$A$, over which different legal traditions disagree. A$A$ is, in fact, wholly true/false: it is wholly true according to one tradition, and wholly false according to another. Even to say that it is both true and false is to misdescribe the situation, because it is true according to one tradition and false according to another. The qualifications are important.

To see this, just consider the case of a conjunction, A ∧ B$A\wedge B$. In every standard many-valued logic, if |A| = |B| = i$\left|A\right|=\left|B\right|=i$ then |A ∧ B| = i$\left|A\wedge B\right|=i$. Now suppose that A$A$ holds in one tradition, and ¬A$\neg A$ holds in another. Then they have the same status, and so truth value. If conjunction behaves in the way described, then A ∧ ¬$A\wedge \neg$ A will have the same truth value as A$A$ and ¬A$\neg A$. But characteristically, A ∧ ¬ A$A\wedge \neg A$ will not have the same status: it will be rejected by both traditions.Footnote ³

To do justice to the phenomenon in question, we need to make sense of the thought that the status of a claim is relative to a tradition. The obvious machinery to apply here, to one trained in the contemporary techniques of non-classical logic, is not many-valued logic, but modal logic.

In a standard modal logic, an interpretation for the language is a collection, W$W$, of things normally called “possible worlds.”Footnote ⁴ Statements at each world are two-valued, the values being true and false. But the value of a sentence may change from world to world. To compute the value of a conjunction or negation at a world, the standard rules of classical logic apply. Thus, A ∧ B$A\wedge B$ is true at a world if both A$A$ and B$B$ are true at that world; and ¬A$\neg A$ is true at a world if A$A$ is false there.Footnote ⁵ And what is true at a world will be closed under an appropriate notion of logical consequence (classical, in the standard case). In other words, if all of A₁ , … , A_n$A_1,...,A_n$ hold at a world, and {A₁, … , A_n} ⊢ B$\left\{A_1,\dots ,A_n\right\}⊢B$ (B$B$ follows from A₁ , … , A_n$A_1,...,A_n$), then B$B$ holds at the world. Indeed, the standard definition of validity normally given for modal logics is that an argument is valid just if it preserves truth at any possible world (in all interpretations).

Though possible worlds are usually given a metaphysical interpretation, they may be given many kinds of interpretation. It is common, for example, to think of them as bodies of information. In the present context, it is natural to think of the worlds as legal traditions. What holds (is true) at a world is what holds according to that tradition. A claim can then be true at one world, and false at another. And the fact that the content of each world is closed under the appropriate consequence relation, is just a way of representing the fact that reasoning plays an important role in legal traditions: given a tradition, people use things that hold in it to infer other things that hold. Thus, one can think of the content of each world as all the matters of law in some tradition, plus all the matters of fact, plus whatever follows from them by acceptable reasoning. The picture, then, is this, where the outer box contains all the “worlds,” and the inner boxes are the different traditions (three in this case):

Let me make it clear that I am not denying the appropriateness of the use of many-valued logic in an analysis of legal reasoning. Indeed, in the next section I will argue that a paraconsistent logic may well be required; and some paraconsistent logics are many-valued logics.Footnote ⁶ My claim that a modal logic is required, rather than a many-valued one, is with respect to the situation in which one is required to deal with multiple traditions.

3 Paraconsistent Logic

So much for the basic idea. There is, of course, much more to be said. At this point, I need to introduce the notion of paraconsistency. Consider the logical principle that any contradiction implies anything: {A, ¬A} ⊢ B$\left\{A,\neg A\right\}⊢B$ (for all A$A$ and B$B$). This goes by the medieval name of ex falso quodlibet, and the more colourful contemporary name of Explosion. It says that once a body of information contains a contradiction, everything follows from it. The principle of inference is valid in standard logics, including standard modal logics (just because there is no world at which contradictory statements hold). Logics in which the inference fails are called paraconsistent logics. Explosion hardly looks plausible for many applications of logic. In particular, many legal traditions seem to contain contradictions, which do not “explode” them. It would appear, then, that legal reasoning should require a paraconsistent logic.

The matter is a sensitive one.Footnote ⁷ It is of course to be expected that the legal principles of any one tradition will contain prima facie contradictions. And usually, legal systems will have principles which resolve some of these. Thus, something enshrined in constitutional law will trump something inconsistent with it in statute law. And the principle of lex posterior tells us that a later law will trump something inconsistent with it in an earlier law.Footnote ⁸

But all of this notwithstanding, there are going to be cases of genuine inconsistency. Actual cases at law are always going to be messy and contentious; so let me give a toy example. Suppose that a duly authorized statute contains the following clauses:

1. 1. All property-holders shall have the right to vote.

2. 2. No woman shall vote.

We may suppose that at the time when the statutes were authorized, the thought that a woman might hold property was just unthinkable; and maybe that there are other clauses in the statute determining which male non-property-holders may vote.

In due course, we may also suppose, as enlightenment creeps over the society, women do come to hold property; and at some point, a woman – call her Jan – eventually fronts up at a polling booth demanding to vote. Jan may and may not vote.Footnote ⁹

Of course, if and when this happened, the law would, in due course, be changed, either by a judge making a ruling, or by new legislation. The law is a guide for action, and contradictions frustrate this. But, until this is done, the law (plus the contingent circumstances) is inconsistent. And until the law is revised, there being no general principle of the kind just noted to resolve the contradiction in this case, both clauses are operative, as may be shown by the fact that each may be appealed to independently in other cases. However, no one would go into court and use this contradiction to argue that their cat has a right to vote. That would just be silly. In other words, Explosion is not a valid principle of legal argumentation in this context.

What this shows is that, in general, the logic operative in many, perhaps most, legal traditions, and under which the information in each of the “worlds” is closed, must be a paraconsistent one. Standard modal logic is not of this kind, but it is easy to construct modal logics where the logic is paraconsistent. The logic of each world may be a simple many-valued paraconsistent logic, for example.Footnote ¹⁰

Paraconsistent logic is also relevant in another way. There is a standard mechanism for generating a paraconsistent logic out of a world-semantics. Given an interpretation, we simply define truth simpliciter (not truth at a world) as truth at some world. Explosion then fails for this logic (even if the logic of each world is explosive), since A$A$ may be true at one world, ¬A$\neg A$ at another, and B$B$ at no world. This procedure gives a discussive logic.Footnote ¹¹ The main feature of a discussive logics is that the principle of Adjunction, {A, B} ⊢ A ∧ B$\left\{A,B\right\}⊢A\wedge B$, fails (since A$A$ and B$B$ may hold at different worlds, and their conjunction at neither). This is a natural way of viewing the legal case. Truth (simplicter) is holding in some legal tradition or other; and one should not expect Adjunction to hold for truth simpliciter. The legal cosmos is just a complex, many-faceted, place.Footnote ¹²

4 Interactions between Legal Systems: Examples

So far so good. The problem is that it’s not that simple. The only real problem on the horizon so far is determining which “world” – jurisdiction – we are in. There will doubtless be occasions when this is debatable, but usually it will be clear enough. The real trouble is that each world is not an island. The different systems and traditions interact, importing information from other traditions, even ideas that sometimes contradict the home tradition. Let me give three examples of this. I will then discuss how matters are to be handled logically.Footnote ¹³

Example 1: A court can import a ruling from outside its jurisdiction. To take an historical example, the secular law might provide that inheritance go to the first “legitimate” offspring of a person. But, in certain cases, a civil court might allow what counts as legitimacy to be determined, not by civil law, but by canon law.Footnote ¹⁴

Example 2: A court can import a precedent from another jurisdiction.Footnote ¹⁵ For example, in Australia, a Federal court may sometimes appeal to a precedent in State law (maybe not even the State where the case is being heard).

Example 3: In contract law, a court may choose to enforce the terms of a contract according to the laws of another jurisdiction.Footnote ¹⁶ For example, suppose that a$a$ and b$b$, in countries A$A$ and B$B$, make a contract concerning matters that will transpire in some other place, C$C$, it being made clear that the laws of C$C$ shall govern their contract. a$a$ may then choose to sue b$b$ for breach of contract in A$A$. The court may find against b$b$ on the ground that it was a breach of contract according to C$C$’s laws, even though it is not a breach of contract according to A$A$’s. Aspects of C$C$’s laws not relevant to laws of contract, such as laws of evidence, need not be accepted, though.

5 Chunk and Permeate: The General Framework

In this section I will explain a general logical framework for handling such matters. In the next, we will see how it may be applied to our three examples. The framework is called “Chunk and Permeate,” and was originally developed for cases in science where mutually inconsistent information is appealed to in a single application.Footnote ¹⁷

We suppose that our information is chunked. Information is then allowed to flow between chunks. Thus, information may permeate from a source chunk into a target chunk. The information that permeates may then be used as part of the information available for reasoning in the target chunk. It is crucial, however, that not all the information forthcoming in a source chunk be allowed to flow into the target chunk, or this may have untoward consequences; most notably, it may be inconsistent with something else already present in the target chunk. To prevent this, a filter is applied, letting through only information of a certain, predetermined, kind. A simple picture is something like this:

Chunk 1, we may suppose, is the “output chunk,” that is, the place where we look for our final conclusions. The structure is called a Chunk and Permeate structure. The general mechanism of such a structure is more complex than I have so far explained, but the extra complexities are not relevant for our purposes.Footnote ¹⁸

6 Application

Let us now turn to its application to our examples of Conflict of Law. In general, how to chunk information in an appropriate way is a non-trivial matter. But in the present case, it is straightforward. The chunking is already provided by the “worlds,” that is, the different legal traditions. We need, in addition, an output chunk (let us call this “Chunk 1” in what follows), which we suppose to contain all matters of fact. Each of the other chunks contains the matters of law pertaining to a relevant legal tradition. In principle, there can be an arbitrary number of these, but for our examples, two will suffice (Chunks 2 and 3). Given the appropriate filters, legal information is allowed to flow into Chunk 1, where it is applied to the factual information already present there, to determine a final judgment. Let us see how this mechanism is applied in our three examples.Footnote ¹⁹

Example 1: Here, Chunk 2 may be taken to contain all the legal information of the jurisdiction of the court. Filter 2 may let through everything except those laws relating to legitimacy. Chunk 3 may be taken to contain all the principles of Canon Law. Filter 3, however, lets through only matters pertinent to judgments of legitimacy. One would not want other things from Chunk 3 to flow into Chunk 1, simply because one would not want them to be operative. When, in Chunk 1, the judgments about inheritance from Chunk 2 and the principles of legitimacy from Chunk 3 are applied to the facts in Chunk 1, an appropriate judgment is drawn.

Example 2: Here, Chunk 2 may be taken to be the legal information relevant to Federal Law. Filter 2 may, this time, let everything through.Footnote ²⁰ Chunk 3 may be taken to contain matters from the State law of the relevant state, including judgments based on precedent. Filter 3 lets through only the judgement of precedent relevant to the present case. Clearly, letting through other matters is very likely to conflict with matters of Federal law in this case. In Chunk 1, the judgment of precedent may then be appealed to in applying the Federal law to the matters of fact present.

Example 3: Here, Chunk 2 may be taken to contain the laws of jurisdiction A$A$. In this case, the filter allows through only those things not relevant to the laws of contract (e.g. laws relating to permissible evidence). Chunk 3 may be taken to contain all the laws of jurisdiction C$C$. Filter 3, however, lets through only those things relevant to laws of contract. When all the available information from Chunks 2 and 3 is then applied to information in Chunk 1, the appropriate judgment is forthcoming.

So much for the examples. There has to be a bit more to the general story than this. Something must determine what the relevant chunks and filters are. (The whole Chunk and Permeate structure which is operative must be determined by something.) Presumably, this is itself the jurisdiction we are actually in (Chunk 2, in our examples).Footnote ²¹

7 Conclusion: Logic and Law

An important part of the law is reasoning.Footnote ²² It is the job of the discipline of logic to theorize reasoning. This does not mean that lawyers have to take notice of the discipline of logic (though occasionally this may help): they just have to know how to reason well. Similarly, it is the job of the discipline of linguistics to theorize language. People whose job it is to speak and write, do not, however, have to take notice of this (though occasionally this may help): they just have to know how to speak and write well.

Logical theory in the history of Western philosophy has been dominated by the notion of deductive reasoning.Footnote ²³ Most notably, in the last 100 years, it has been dominated by reasoning in mathematics. The tools of “Classical Logic” (or more aptly, “Frege-Russell Logic,” after its inventors), were developed with this in mind. Though many have sought to impose a hegemony of classical logic, this is now coming to appear an unduly imperialistic attitude. In particular, the many different techniques of non-classical logic developed in the last 50 years, such as those of modal logic, many-valued (including fuzzy) logic, paraconsistent logic, can be seen as providing a more versatile and robust armoury of devices for analyzing reasoning in other areas (and maybe in some bits of mathematics too). The area that has been our concern in this essay is legal reasoning, especially in the context of the multitude of legal traditions and their interactions. This, it seems, is such an arena.Footnote ²⁴

Appendix: Beall and Release

In “Do Inconsistent Laws Deliver Gluts?” this volume, pp. 199–207, Jc Beall takes issue with my claim of Section 3, that the law can deliver genuine inconsistencies, that is, dialetheias, that is, true things of the form A$A$ and ¬A$\neg A$.

His point is essentially this. A claim of law, A$A$, such as “one must drive on the left hand side of the road” or “people under 18 may not vote,” is always relative to a jurisdiction. So, when properly understood, it is to be understood as of the form: According to the law A$A$. Beall writes this as OA$\mathit{OA}$. Actually, it is better to make the jurisdiction explicit. So let JA$\mathit{JA}$ be “According to jurisdiction j$j$, A$A$.” So far so good. Next, Beall observes, J$J$ does “Release”; that is, one cannot infer A$A$ from JA$\mathit{JA}$; and such an inference is required to deduce a contradiction. Again, so far so good. Dialetheias, infers Beall, do not arise.

Too fast. J$J$ certainly does not release in general. If I am driving in the US, and j$j$ is Australian jurisdiction, I cannot infer from J$J$ (I must drive on the left) to “I must drive on the left.” But one can release if one is in the jurisdiction in question. Merely consider, for a moment, the following dialogue between a shopkeeper (s$s$) and a customer (c$c$).

• (c$c$) A packet of cigarettes, please.

• (s$s$) How old are you?

• (c$c$) Sixteen.

• (s$s$) I’m sorry, you can’t have it.

• (c$c$) Why not?

• (s$s$) It’s the law. The law says that if you are under eighteen, I’m not allowed to sell you cigarettes. So I can’t give them to you.

The shopkeeper has exactly used “Release.” He knows the legal jurisdiction he is in, and knows its law. He “releases” to the conclusion that if a person is under 18, a shopkeeper may not sell cigarettes to that person. He joins this to the premise the 16 is less than 18, to conclude that he is not allowed to sell to c$c$.

In exactly, the same way, in the example of Section 3, let us suppose that the legal jurisdiction in which Jan finds herself is j$j$. Then J$J$(All property holders have the right to vote) and J$J$ (No women has the right to vote.) The person in control of the polling booth is, again, well aware that they are in jurisdiction j$j$. They “release” to infer that all property holders have the right to vote, and that no women have the right to vote. A few more simple inferences shows them that they are in a bind.

It might be suggested that the whole inference process is under the scope of the J$J$ operator, so any ultimate conclusion is of the form JA$\mathit{JA}$, which is not a contradiction. Thus, to illustrate, consider the shop-keeper example again: s$s$ knows that J$J$ (one must not sell cigarettes to people under 18) and J$J$ (c is under 18) and concludes that J$J$ (I should not sell cigarettes to c$c$). This is hardly a correct way to reconstruct the reasoning. The age of c$c$ is a matter of fact, not a matter of law. It is simply about the time elapsed since c$c$’s birth. One cannot, therefore move from “c$c$ is under 18” to J$J$ (c$c$ is under 18).Footnote ²⁵

And in any case, the move is of no avail. Even if the conclusion of the explicit reasoning is of the form JA$\mathit{JA}$, if one is in jurisdiction j$j$, one must still apply “release.” The law is action-guiding, and release needs to be applied to infer what to do. So imagine that s$s$ does, in fact, sell cigarettes to c$c$, at which point a police woman, p$p$, enters the shop. The following dialogue ensues:

• (p$p$) Did you just sell cigarettes to this kid?

• (s$s$) Yes.

• (p)$p)$ You shouldn’t have done that.

• (s$s$) Why?

• (p$p$) Because the kid is under 18, and the law forbids selling cigarettes to minors.

• (s$s$) Yes, I know the law forbids selling cigarettes to minors.

• (p)$p)$ So you shouldn’t have done it.

• (s$s$) That doesn’t follow.

The police woman is not impressed

Beall buttresses his argument by an extra ad hominem argument. If the law delivers gluts (things that are both true and false), it also delivers gaps (things that are neither true nor false).

His example is where the law in jurisdiction, j$j$, says that persons in category C(x)$C\left(x\right)$ must do such and such, A(x)$A\left(x\right)$; and some person, a$a$, is such that whether or not C(a)$C\left(a\right)$ is indeterminate (maybe because the matter is vague). Then, says Beall, neither A(a)$A\left(a\right)$ nor ¬A(a)$\neg A\left(a\right)$ is true. No. By Beall’s own account, what holds is that ¬JA(a)$\neg \mathit{JA}\left(a\right)$ and ¬J ¬ A(a)$\neg J\neg A\left(a\right)$. This is compatible with J(A(a) ∨ ¬ A(a))$J\left(A\left(a\right)\vee \neg A\left(a\right)\right)$, and so, given that we are in jurisdiction j$j$, to A(a) ∨ ¬ A(a)$A\left(a\right)\vee \neg A\left(a\right)$.

1 Introduction

In arguing for the relevance of non-classical logics to legal analysis, Patrick Glenn often pointed out that until the rise of the nation-state, it was quite common to have multiple legal orders operating in the same territory.Footnote ¹ This was certainly true in the early days of English law. To get a sense of the multiple overlapping and interlocking jurisdictions and systems of norms, it is only necessary to leaf through the pages of Sir John Baker’s detailed study of the period 1483–1558.Footnote ² Part III on ‘The Courts’ runs to over 200 pages. Almost half of this deals with the three royal courts of common law, and the Court of Chancery that administered English Equity; and it is the logical interaction between the common law and Equity that will be the focus of this study.Footnote ³ But there were many other courts and normative systems. There was the Star Chamber and the Court of Requests, which like the Chancery derived their authority from the King’s Council. There was the High Court of Admiralty and the Court of the Constable and Marshal, which applied respectively the law of the sea and martial law. There were ecclesiastical courts, which applied the law of the church to issues such as the proof of wills, and to sinful conduct like adultery and simony. In London, there was a Court of Hustings, a Mayor’s Court, Sheriffs’ Courts, and others. Outside of London too, there were innumerable local courts, such as the courts of the counties palatine of Chester and Durham, which were outside the jurisdiction of the royal courts of common law, and county courts in every county for local disputes. Many towns and cities had local courts, including staple courts for merchants, which applied the law merchant. Every fair and market had its own court of piepowder to resolve issues arising within the fair or market, with cases often resolved within hours.Footnote ⁴ And in principle every manor had its own local court; just as the king was obliged to do justice between and among all of his subjects, each lord was obliged to do justice between and among those who owed him fealty. These courts applied the local law of the manor, which might in principle differ from the law that was common to all the realm – the common law.Footnote ⁵ Taking all these local courts together, one author estimated that in the early seventeenth century, there were 12,000–15,000 courts in England and Wales.Footnote ⁶

2 Common Law and Equity

It is not surprising that there were often difficulties about which court had jurisdiction over a cause or an issue within a cause.Footnote ⁷ The principles governing these questions could be complex, not least because they were subject to many exceptions.Footnote ⁸ If we consider the relationship between the royal courts of common law and the Court of Chancery, it must first be said that this relationship went through many phases over a long history. At its origin, English Equity was not understood by anyone to be a system of legal norms. It began in the twelfth or thirteenth century when petitioners sought the intervention of the Crown in cases of perceived individual injustice, and the Chancellor was delegated to hear these petitions. The decisions were individual and based on considerations of conscience. They were not law in the sense that they were understood to be particular interventions that did not reflect or create general norms; they bound no one but the parties. Over time, however, the Chancellor’s rulings became more predictable and evolved into a system of rules that were clearly legal in nature; the Court of Chancery, which had been a court in the same sense that a king holds court, became a court administering a system of law. When precisely this change occurred is impossible to say, but it was during the Tudor period that the basis of the Chancellor’s intervention shifted from ‘conscience’ to ‘Equity’.Footnote ⁹

In some ways, however, Equity never lost touch with its roots as an extraordinary corrective. It never became a complete system of law: for example, although Equity has its own remedies for breach of contract, it does not have its own rules as to how contracts are created.Footnote ¹⁰ Every rule of Equity originated in an attempt to correct a defect of the common law, but the converse was not true. A much-used example of the Chancellor’s intervention is the case of the debtor who gave his creditor a deed (a document bearing a formal seal) as evidence of the debt, and who paid the debt but neglected to recover or destroy the deed. If the creditor now sued at common law, the deed was effectively incontrovertible evidence of the debt. In the application of what we might now describe as rules of evidence, a deed could not be contradicted by oral evidence; and in any event, the parties to an action at common law were not competent to give evidence. If the common law case were to proceed to judgment, the debtor would lose and would have to pay the debt a second time. To prevent this, the debtor would commence a proceeding in Chancery, where the rules of evidence were different. The Chancellor would investigate what had happened (including via the sworn testimony of the parties), and if he were convinced that the debtor had paid, he would order the creditor to hand over the deed and to cease the proceeding at law. In understanding the relationship between the common law and Equity, it is crucial to understand that the Chancellor did not make any order that was intended to bind, override, or contradict the common law courts. He had no jurisdiction to do that. He made orders against the parties.

When Equity came to be viewed as a system of law, it was perhaps natural that conflicts with the royal courts of common law would arise. When there was an issue as to jurisdiction between a royal court of common law and an inferior court, such as a local court or an ecclesiastical court, it was the royal courts of common law that held the whip hand; they made the final decisions about jurisdiction, and could forbid the other courts from proceeding. As between the royal courts of common law and Equity, however, there was no clear hierarchy. It was possible to sue at common law for unlawful vexation in Chancery.Footnote ¹¹ More often, the pressure was in the opposite direction. Since Chancery’s original function was to address the failings of the common law, it must necessarily issue orders that, for example, forbade someone from enforcing a common law right, as in the example of the deed that was given in the last paragraph. Those who defied such orders could be imprisoned until they complied.Footnote ¹² There is a temptation to say that the Equity rule prevailed over the common law rule; and this is a norm that ended up in legislative form, when the two court systems were combined into one.Footnote ¹³ But of the English provision, Maitland said to his students that while ‘it may well seem to you that those are very important words’, nonetheless, thirty years after its enactment, ‘it has been practically without effect’.Footnote ¹⁴

Patrick Glenn might have said that this legislation was based on a mistakenly binary view that one system must ‘win’ over the other.Footnote ¹⁵ It is certainly true that such a binary view is misleading, when the matter is viewed in historical perspective. As the old maxim has it, ‘equity supplements but does not contradict the common law’.Footnote ¹⁶ This statement is, at the same time, deeply true and fundamentally false. It is false inasmuch as the whole point of Equity was to contradict the common law; it is true inasmuch as the contradiction had always to be indirect in order for the creation and growth of Equity to be something other than a kind of civil war. That is what Maitland meant when he said the provision was without effect: it addresses itself to conflicts, and properly understood, common law and Equity do not conflict.Footnote ¹⁷

It was the maxim’s falseness that gave rise to tensions in different periods.Footnote ¹⁸ But these tensions were usually linked to strong personalities on one or both sides. The systems could operate in harmony. It was the truth encapsulated in the maxim that allowed the two systems to co-habit in peace for most of several centuries. Indeed, in some ways they cooperated.Footnote ¹⁹ A new phase began in the 1870s, via the Judicature Acts.Footnote ²⁰ These instituted a reform of the court structure that combined the previously separate courts (three royal courts of common law and one of Equity, as well as the Court of Probate, the Court for Divorce and Matrimonial Causes, and the High Court of Admiralty) into one court (with multiple divisions). The reform was procedural inasmuch as it was not understood to change the rights of citizens or the results of disputes; it was designed to simplify the ways in which rights were enforced and disputes were resolved. The effect of this reform is a matter of some ongoing controversy, in relation to whether it should, in the end, be understood to have changed the substance of the law and not just its procedures.Footnote ²¹

3 Chunk and Permeate

Patrick Glenn was interested in whether the relationship between common law and Equity could be understood in terms of non-classical logic. In this section, I ask whether that relationship can be captured by the non-classical logic called ‘chunk and permeate’, which is explained by Graham Priest in his contribution to this volume.Footnote ²² Described abstractly, in chunk and permeate there are multiple chunks, each of which contains information. The information in one chunk may be inconsistent with that in another. This is why they need to be seen as separate chunks. But if a filter is applied, some information may be allowed to pass from one chunk into another in a way that does not create an inconsistency and allows a useful result.

Applied to multiple legal orders, we might say that each legal order is a chunk. Certain rules apply within each chunk, but the rules in one chunk are not the same as the rules in other chunks and to this extent there are inconsistencies. In some situations, however, the rules of one chunk are allowed to permeate into another chunk to provide information that assists legal reasoning in that chunk. The role of the filter that controls permeation is crucial, because if everything were allowed in, the receiving chunk would be simply applying the legal order of the sending chunk, and would not be logically (or, in this context, normatively) distinct. One of Priest’s examples (Example 3) is from the subject called private international law.Footnote ²³ Parties to a contract may choose that any dispute about the contract will be governed by the law of jurisdiction C, and speaking generally, this will be effective. A court in another jurisdiction, A, will do its best to apply this (foreign) law of contract. But the court in A is also applying its own law: for example, it applies its own law of procedure.Footnote ²⁴ Moreover, it is not all of the substantive law of C that permeates; only the law of contract. And indeed we might say that this happens only because the law of A contemplates it. It is the private international law of A that provides that effect shall be given to the choice of a foreign law. It permits the permeation and it creates and defines the filter.

One of the challenges of applying chunk and permeate to multiple legal orders is that it is not always clear what are the chunks, and what is the information. In his discussion, Priest supposes an output chunk that contains all matters of fact. The other chunks contain rules of multiple legal orders. The filters control which rules pass from each chunk into the output chunk. In the output chunk, the rules are applied to the facts and conclusions are drawn. This is not exactly how a jurist would be likely to envisage the example discussed in the preceding paragraph.Footnote ²⁵ The jurist would have some trouble in seeing the decision-making process of the court in A as accepting information (rules) from another chunk which contains the laws of A. The more natural view is that the court in A finds the facts and applies the law of A. It so happens that the (private international) law of A provides that in general, parties may choose the law applicable to their contract, and this is why the court applies the contract law of C rather than the contract law of A. The court in A understands itself to be doing nothing but applying the laws of A, even when exceptionally those laws dictate that the laws of C should be used for some issues. This seems more like a two-chunk structure (as envisaged by Priest as a possible understanding of his Example 2). One chunk thus includes both facts and rules; some rules are allowed to permeate from another chunk.

Because the contract law of A is not applicable to the resolution of the dispute, one could say that it is necessary to see three chunks. Thus the court of A, in resolving the dispute, applies some but not all of the laws of A (leaving out A’s contract law); so the decision-making chunk should be seen as notionally separate from the laws-of-A chunk (as well as from the laws-of-C chunk). In this view, it is still possible to see the filters as constituted by the ‘choice of law’ rules of the private international law of A. Choice of law rules are, in a sense, meta-rules or rules about rules.Footnote ²⁶ They do not resolve the case directly, but rather they tell the judge which legal order’s rules apply to an issue. Examples of such rules are: (i) if the issue is about an allegedly non-contractual wrongful act (like negligent harm to property), the governing law is the law of the place where the act occurred; (ii) if the issue is as to the formal validity of a will,Footnote ²⁷ the governing law is the law of the place where the will was executed; (iii) if the issue is as to the consequence of a breach of contract, the governing law is the law chosen by the partiesFootnote ²⁸; and so on.

In Priest’s articulation of the notional three-chunk view, the facts are in the decision-making, output chunk. One difficulty with this is that the filters themselves are shaped by the facts, as the examples in the previous paragraph show. Putting the facts inside a chunk (and so apart from the filters) thus seems inappropriate. Perhaps the facts need to be seen as common to all chunks, and indeed to the filters, thus lying in the background of the whole structure of chunks and filters. That structure, then, addresses itself more to the application of legal rules to the facts.Footnote ²⁹

Assuming that the facts do not belong to any particular chunk, another complication is that the ‘information’ in the chunks may not only be legal rules but also legal results. Assume that person a has successfully sued person b in the courts of jurisdiction A, and obtained a judgment ordering b to pay a $10,000. If b does not live in A and has no assets there, this judgment might not be very useful. Assume that b lives in B and has assets there. The order of the court in A is not directly enforceable in B by the legal officials of B; in B, it is an order of a foreign court. It would be possible for a to start a new lawsuit against b in B, but this would require a range of risks and expenses that a has already incurred in A. Thus a might show to the court in B the judgment of the court in A and ask the court in B to ‘recognize’ it. If the court in B does so, it will make its own order that b pay a $10,000; this order will be a new judgment, directly enforceable in B via the legal officials of B. The subject of the recognition and enforcement of foreign judgments is also considered part of private international law. This part, however, does not involve choice of law rules but rather rules and principles that are based on comity but that also aim to protect citizens against unfair foreign judgments.Footnote ³⁰ In this example, the ‘information’ passing from one chunk to another is neither a rule nor a fact, but a judgment, which is the result of the application of (foreign) rules to facts. Since a judgment as such is particular to the parties, it is rather like a fact for the analysis: not being a general norm, it can be seen as outside all the chunks and filters.Footnote ³¹ It is something important that has happened in the past. In this example, the rules governing which foreign judgments can or must be recognized are the relevant filters.

If all of this be right, we can adopt Priest’s three-chunk implementation for these examples from private international law (choice of law, and recognition and enforcement of a foreign judgment). The court’s decision-making role is seen as belonging to a separate chunk from the norms of its own legal system; this is because some of those norms may end up being inapplicable in that process. Perhaps modifying Priest’s implementation slightly, the facts (possibly including outputs from previous legal processes) do not belong to any chunk but lie in the background of all the chunks and filters. The chunks and filters contain general norms. The filters, in these examples, are the rules of private international law (of the legal system of the deciding court). If the legal systems of more than two jurisdictions were applicable, there would be more than three chunks.

4 Common Law and Equity as Chunk and Permeate

Applying all of this to the relationship between common law and Equity is interesting, but difficult. It is difficult partly because, as we have seen, that relationship evolved over the centuries.

We start with the early days of Equity, when it is evolving into a system of legal norms. As was mentioned above, the common law was a complete system of law in the sense that it was apt to give an answer to any legal question.Footnote ³² It is generally assumed that the courts of common law did not pay any attention to rules and principles of Equity. From the perspective of the common law courts, therefore, there was no plurality of legal orders in relation to Equity.Footnote ³³

From the perspective of Equity, it was rather more like the choice-of-law example from private international law. How much of the common law, and how much Equity, should be permitted to permeate into the Chancellor’s decision (the decision-making chunk)? At least some common law had to permeate, because Equity was not a complete system of law. To take two examples, Equity needed to know who was the owner of property ‘at law’ before it could, or might, subject that person to some order or obligation regarding that property. Again, as was mentioned earlier, Equity did not have its own system of power-conferring rules for contract formation; for the Chancellor to know whether there was a contract or not, he had to apply the common law rules.Footnote ³⁴ The rules of the common law were the starting point of the parties’ juridical relationship.

It was not in every case, however, that any of the rules of Equity permeated into the Chancellor’s decision. Let us start from a historical truth: originally, every petition to the Chancellor had to recite that the petitioner could not obtain justice from the courts of common law.Footnote ³⁵ Part of the story of the origin of Equity was that it arose to address such injustices. But the other side of the coin is that if the Chancellor thought that the common law courts were able adequately to deal with the issue, he would not intervene. Right up until the nineetenth century fusion of the courts, a plea that could be made in response to any claim for Equitable relief was ‘want of equity’.Footnote ³⁶ This was an argument that the plaintiff was not entitled to what he was claiming, even if the facts he alleged were true; and one reason this might be the case was that he had an adequate remedy at common law. If, in such a case, the conclusion was that there was no need or room for the Chancellor’s intervention, the plaintiff’s claim would be dismissed (without prejudice to his or her rights at common law).

In the many cases where the Chancellor did intervene, the grounds for intervention were various. In modern terms, we might say they were a mix of procedural and substantive reasons.Footnote ³⁷ Applied historically, however, this distinction has to be treated with some care, because it can easily be anachronistic. Consider the example mentioned earlier, of the debtor who paid his debt but failed to destroy the deed that was the evidence of the debt. The creditor sues at common law. The defendant cannot tell his story in the court of common law, for what we could call procedural reasons: he cannot give evidence himself (because no party is allowed to give evidence at common law); nor can anyone else give oral evidence that is inconsistent with the deed. The common law courts were aware of the risks of individual injustice created by their rules, but in some cases, preferred to protect other values. Thus, in this kind of case, the importance of preserving the sanctity of deeds, and of avoiding the risk of perjury by conflicted witnesses, were considered paramount.Footnote ³⁸ As we have seen, the Chancellor had different rules for getting at the truth. The Chancellor would not deny the legal validity of the deed; but the creditor would be told that even though he had a legal right to be paid again, it would be unconscionable for him to insist on it. In terms of chunk and permeate, we might say that in this case, the filter lets in the rules about the validity of the deed; but it also let in some Equity, that forbade this particular creditor from enforcing his (valid) deed. This was how Equity could be said to supplement the law without directly contradicting it.

During the nineteenth century, it was popular to organize Equitable jurisdiction according to three heads: exclusive, concurrent, and auxiliary.Footnote ³⁹ The auxiliary jurisdiction was said to be activated where a party engaged in litigation at common law had need of some procedural device that was only available in Chancery, such as ‘discovery of documents’.Footnote ⁴⁰ The concurrent jurisdiction operated where the plaintiff had some right at common law but sought to invoke Equity’s intervention in order to obtain a desired remedy unavailable at common law. For example, the remedy for breach of contract at common law was compensatory damages. The courts of common law did not actually order people to perform their contracts, unless the performance involved merely the payment of a sum of money.Footnote ⁴¹ A plaintiff could go to Equity to seek a decree of specific performance, by which the defendant would be ordered to perform the contract and not just to pay damages. Finally, the exclusive jurisdiction referred to those legal institutions that were the entire creation of Equity. The core illustration is the law of trusts. In a trust, property is held by one person (the trustee) for the benefit of another (the beneficiary). The trustee is usually the owner of the property according to the common law, but is obliged to ensure that the benefits of the property (through its enjoyment or investment) accrue to the beneficiary.Footnote ⁴² The rights of the beneficiary exist only in Equity. In this sense, it is clear that the beneficiary has no adequate remedy at common law, because he has no rights at common law.

Already in Maitland’s time, in the early twentieth century, this classification was said to be obsolete as a result of the fusion of the courts.Footnote ⁴³ Certainly the auxiliary jurisdiction makes no sense when procedural law has been unified. There remains, however, a distinction between the other two that is arguably still relevant; this distinction might be described as the threshold question for Equitable intervention. When a plaintiff seeks a decree of specific performance, he must in principle show that monetary damages would be inadequate.Footnote ⁴⁴ If he cannot, he will not get the decree. In this sense, and for other reasons, the power to grant such a decree is said to be discretionary. But the exclusive jurisdiction is different. A trust beneficiary could not be required to show that monetary common law damages are inadequate, because speaking generally, such a beneficiary has no rights at common law. And it has never been suggested (or at least not since the fifteenth century) that when the beneficiary wishes to enforce a trust, this is a matter of discretion for the judge. Trusts are always specifically enforceable. One way to make sense of this difference is that in the concurrent jurisdiction, the plaintiff has a non-discretionary right to compensatory damages, coming from the common law. Thus, if she gets nothing from Equity, she will still be fully compensated so far as this is possible with money. In the exclusive jurisdiction, by contrast, if the plaintiff gets nothing from Equity, she gets nothing at all, because she has no rights except Equitable rights.Footnote ⁴⁵

It might be possible to interpret this difference as a difference in the filters of a chunk and permeate structure. The ‘adequacy’ threshold does not apply in the exclusive jurisdiction but it does apply in the concurrent jurisdiction. If it applies but is not satisfied, Equity does not change the outcome. Alternatively, if applies and it is satisfied, Equity may make an order which could not have been obtained from a common law court, such as a decree of specific performance. The modern court that applies both common law and Equity could be understood as represented by Priest’s three-chunk structure, with Chunks 1 and 2 being the common law and Equity, and Chunk 3 being the decision-making forum. The filters that control how much Equity gets into Chunk 3 can, again, be understood to be different as between the exclusive jurisdiction and the concurrent jurisdiction.

5 Conclusion

Earlier I suggested that the key to understanding the relationship between the common law and equity lies in the paradoxical maxim that ‘equity supplements but does not contradict the common law’. So far from being only of historical interest, in my view both the truth and the falsity of this statement are crucial in comprehending the modern relationship between the two sets of norms.Footnote ⁴⁶ In this sense, chunk and permeate may offer a helpful way of formalizing the relationship between the common law and Equity, since it provides a way to understand how bodies of information that are inconsistent with one another can be integrated in a manner that avoids inconsistencies.