I. Global constitutionalism

The concept ‘global constitutionalism’ is used in several only partially overlapping ways.^{Footnote 2} As part of a specification it is helpful to disentangle ‘constitution’, ‘constitutionalism’ and ‘constitutionalisation’.

The main functions of constitutions are to constitute institutions of governance: to establish and limit political authority.^{Footnote 3} Written and de facto constitutions typically do so by performing four tasks – though there are variations among domestic constitutions in this regard.^{Footnote 4} A constitution creates – ‘constitutes’ – new institutions or codifies existing institutions with specified legal competences. In particular, it establishes ways to create new legislation – thereby channelling competition and conflict. A constitution secondly curbs such powers, e.g. by principles of legality and the rule of law, by securing human rights and other legal constraints by virtue of being superior to ordinary legislation, or by dividing political authority among several institutions. A constitution may also channel the use of such powers further by indicating the goals to be pursued, often stated in a preamble. Fourthly, a constitution contains rules for constitutional change, so as to combine desired stability and robustness with flexibility to adjust it to new circumstances. There is no guarantee that all such constitutions are of value. Still, individuals may benefit in several ways from a domestic effective and normatively defensible constitution in this sense. It can safeguard citizens against risks and secures their interests, and contributes to shaping them by affecting the institutions that in turn shape their life plans, interests and expectations. A constitution allows individuals to use institutions to establish new patterns of coordinated behaviour, it can reduce fears of abuse of powers, and it can bolster legitimate expectations.

Whether a global constitution can serve similar tasks requires arguments, and the appropriate standards may differ – just as many of the standards vary among domestic constitutions. To further complicate the picture, even the features and concepts may be different in significant ways – e.g. concerning the contents of ‘due process’, ‘(international) human rights’, or ‘(international) rule of law standards’.^{Footnote 5} A further important difference between domestic and global constitutions is that the former is close to inescapable. Individuals living under a domestic constitution can exit any particular domestic constitutional order – but only at the price of leaving that jurisdiction. And they in practice must move to another jurisdiction which itself has a constitution – so living under a domestic constitution is in practice inescapable. In contrast, states are bound by much international law only by consent – though not customary international law, which binds new states regardless of such consent. State consent is thus a central though not exhaustive mode of ‘constitutional change’. Their exit from existing treaties may almost always be permitted as a matter of international law – and any set of states may consent to new treaties and institutions.

Constitutionalism is the idea that government can – and should – be legally limited,^{Footnote 6} to ensure or promote the various functions or values indicated – which still remain to be specified:^{Footnote 7} ‘the commitments of principle that underlie the tradition of democratic constitutionalism’.^{Footnote 8} By extension, global constitutionalism can thus be taken to hold such claims about the ‘global basic structure’, consisting of domestic, regional and international legal orders.

Note that this understanding of global constitutionalism is firmly placed within what we may think of as normative ‘non-ideal theory’. The appropriate normative standards for assessing actions and institutions may often be quite different for settings where those institutions are fully in compliance with the standards, and are generally complied with, compared to in circumstances where existing institutions are less than fully just, or where they are only partially complied with. The normative assessment of CM must thus be made in light of the imperfect present set of institutions. There is no claim that existing international public law meets normative criteria e.g. as expressed in theories of global justice, nor is there a claim that the values and principles can simply be mechanically transposed from the domestic to the regional and global legal orders. Thus this conception of global constitutionalism avoids Koskenniemi’s criticism that the vocabulary of constitutionalism assumes

The non-ideal setting of this enquiry is further underscored by considering talk of ‘constitutionalisation’ of international law or of international courts. Such claims often relate to one or more of the tasks listed above, or to the ‘constitutional values’ maintained by global constitutionalism. Thus constitutionalisation is said to occur when systems or regimes increasingly realise constitutional functions or codify constitutional processes – which they do not currently realise. These functions or processes might include predictability and protection against domination, or their compliance with the international rule of law,^{Footnote 10} or when there are steps to ensure more coherence or even integration of previously independent legal orders – e.g. by specifying hierarchies of legal norms by source (jus cogens) or sector such as human rights.^{Footnote 11} Some important contributions illuminate the complex interface between the levels of authority.^{Footnote 12} Others focus on particularly salient shifts in the sites of some of the functions, such as

Several aspects of such processes of constitutionalisation are relevant. Constitutionalisation may occur without empirical or normative claims concerning an end result of a full global constitution, and certainly not one of a kind familiar from many states’ constitutions. Talk of such constitutionalisation makes no claim that there is such an order already.^{Footnote 14} Nor is there any assumption that increased constitutionalisation is always a normative improvement. There is not a uniform or monotonic relationship between any institutional changes that increases one of the functions of constitutions, and particular normative standards. Thus some cases of constitutionalisation may be criticised by standards of constitutionalism – as has been the case concerning the alleged ‘partial’ constitutionalisation of WTO/GATT rules which is said to give these rules or the values they promote undue priority.^{Footnote 15} Finally, this sort of constitutionalisation is not committed to a particular grand architectural principle to secure unity, e.g. not necessarily ‘hierarchical priority to institutions representing general international law (especially the United Nations Charter)’.^{Footnote 16}

Empirical research on constitutionisation may map the extent of such processes with geographical or sectoral patterns, or uncover their causes and effects, if any. Normative research on constitionalisation includes reasoned assessment both of such increased promotion and compliance with the requisite functions and values, and of the processes of constitutionalisation themselves.

II. Constitutional pluralism

To identify some of the implications of contested multilateralism for global constitutionalism it is also helpful to consider some features of ‘Constitutional Pluralism’. In Neil MacCormick’s classic formulation, this term refers to the presence of a plurality of institutional normative orders, each with its own functioning constitution

Neil Walker has contributed to our understanding of how such constitutional pluralism puts the present global basic structure under strain: several legal authorities overlap in their jurisdictions, and are making constitutional claims over the same individual, events or activities.^{Footnote 18} Such competition challenges predictability, and may prevent protection against domination, which are central norms of global constitutionalism. Global constitutionalisation in these regards also seem unlikely. The competing authorities are in a state of nature among each other – whether it is in the sense specified by Hobbes, Locke, Rousseau or Kant remains a contested issue. The main point is that there are no mechanisms or authorities for adjudicating such conflicts, and none likely to emerge soon:

CM is a particular kind of conflict among competing authorities with features which sets CM apart from most other cases of constitutional pluralism. CM are conflicts which are consciously created, or at least envisioned, by coalitions of states or other actors who are dissatisfied with existing authorities. The conflict is not among bodies that address different issue areas yet overlap somewhat, which might be resolved by demarcation which removes conflicts or sets up ways to resolve conflicts among them e.g. by reference to a hierarchy. Rather, the conflicts are among bodies which make conflicting claims of authoritative governance of the same issue area. And some actors have established some such new bodies to improve on the current authorities’ contributions to global constitutionalisation. To the extent that the actors are establishing CM deliberately, some actors will have assessed the – uncertain – prospects of such strategies favourably and preferred them over attempts to modify the procedures or outcomes of existing authorities. The nature of these actors, their objectives, as well as these new authorities, has several implications for global constitututionalism.

III. The status of ‘extra-constitutional’ modes of constitutional change

In the international sphere, changes to international law would standardly occur by treaty changes or protocols requiring unanimity, by specification of treaty interpretations which may not require unanimity; or by complex decision rules securing both a requisite threshold of states and of population such as in the EU. In contrast, CM is an ‘extra-constitutional’ mode of law-making and even of ‘constitutional’ changes which ignores such procedures. The new authorities may be established outside specified procedures, by a sufficiently dissatisfied and powerful coalition regardless of the effects e.g. on various constitutional values or standards of global distributive justice. Indeed, CM can be an alternative to all the four functions of constitutions identified above: the creation, curbing, channelling and changing of institutions. On the other hand, cases of CM may also be improvements on existing authorities, e.g. establish authorities that are more accountable to more of the affected parties, with decision procedures that better secure rule of law standards, etc.^{Footnote 20}

How are we to assess this feature of CM? Surely, normatively speaking the existing mechanisms of ‘constitutional changes’ of the global order arguably leave much to be desired. States’ ability to veto treaties and treaty changes may on the one hand reduce or temper powerful states’ ability to use international law merely as new instruments of domination. On the other hand, states’ ability to block fair solutions to collective action problems may prevent normatively defensible improvements either from some states’ blatant refusal, or due to ‘battle of the sexes’ games: in some issue areas they should agree to some shared authority even when they disagree about which one is to be chosen.^{Footnote 21} Given this impasse, CM may thus provide a much needed alternative mode of existing ‘constitutional’ change. Indeed, a ‘club model’ of CM starting with some able and willing states may be a preferable way to establish a better regime than to wait for universal agreement to modify existing institutions.^{Footnote 22}

It remains to be determined where to place CM within a taxonomy of such ‘extra-constitutional’ modes of constitutional change. Might CM develop into a stable practice with predictable features, or will cases of CM only serve as emergency measures by highly dissatisfied and powerful actors? Will CM function as a supplement to the standard modes, or does general awareness of CM as a live option chill these modes or even replace them? How does CM compare to alternatives such as simple exit from a particular regime, particular incidents of secret or public noncompliance with authorities, or ‘international civil disobedience’? The latter might be understood as actors’ attempt to change authorities’ practices by illegal means whilst seeking to convey loyalty to the underlying ‘constitutional values’.^{Footnote 23} All of these might be regarded as supplements to or deviations from existing authorities, rather than downright alternatives to them as CM is intended. Actors who initiate CM and international civil disobedience have similar objectives in not simply seeking to be exempt from existing rules, but to change them. However, CM is not illegal, whilst international civil disobedience by definition is.

How to categorise CM among forms non-compliance with commands by existing authorities is one important intellectual task. A second issue is to determine when CM is more likely to occur. It is costly and uncertain to establish a new authority in the hopes that it will bring improvement. We might therefore expect actors to first spend some resources to identify existing options, assess opportunities to change them, and give voice to their dissatisfaction before choosing exit and establishing a new CM.^{Footnote 24} To gather a coalition around a new CM may also require a sufficiently convincing account of the objectives sought, with a sketch of why a new authority with certain rules and procedures are likely to secure these objectives – more so than existing authorities. We might also expect to see some arguments for why these actors dismiss attempts to change existing authorities’ policies – e.g. that setting up rivals may be easier than revising existing institutions, in pursuit of some of the functions of constitutions listed above.

A third area of scholarly contributions is to ascertain when any CM or other forms of challenge to existing authorities are normatively defensible or reprehensible. There is no reason to believe that all existing authorities and civilly disobedient actors have normatively defensible objectives. The same holds for coalitions engaged in CM. To the contrary, CM would seem vulnerable to a charge that such attempts at establishing counter-authorities do not shape, constrain or structure contestation about objectives and procedures at all. Instead, insofar as a CM is successful in supplanting existing institutions, there is a serious risk that it is largely shaped to further the interests of the more powerful coalitions, rather than also heed the interests of other affected parties. But this risk should not be overstated – sometimes states may be sufficiently guided by concerns for the common good, for instance in pursuit of public goods.^{Footnote 25} Yet, just as with civil disobedience, it may be possible to lay out some conditions for when such actions are justifiable.^{Footnote 26} Three such conditions might be drawn from the civil disobedience literature, to illustrate some paths for future research. One condition may be that the objective is normatively justifiable – e.g. by standards of global distributive justice. There is much disagreement about such standards, and especially about permissible inequalities.^{Footnote 27} Leaving aside discussions about the normative relevance of such disagreement, it would appear easier to justify CM if the objective is to avoid contributing to gross injustices. Such objectives may enjoy greater consensus than more contentious regimes that distribute material resources. A hypothetical example of a CM that may satisfy such a condition might be efforts to establish an alternative to the Security Council with a mandate to authorise somewhat more humanitarian interventions.

A second condition might be that reforms of existing authorities have been attempted in good faith, at least if such efforts would have a chance of being effective. A third condition may be that the efforts to establish a CM do not threaten the stability of the international legal order as a whole. We move to consider that requirement in the next section.

IV. Challenges to stability

One of the values served by global constitutionalism is to stabilise expectations about other parties’ future actions, to reduce uncertainty and to allow longer-term planning. Even changes still deemed necessary to legislation or the ‘constitutional’ creation or modification of institutions are likewise regulated, often even requiring qualified majority vote and a cooling-off period if sufficiently drastic changes are contemplated. What are we to make of CM in light of this value? Four points in defence of CM against this charge may merit mention. Firstly, predictability is only one of several values promoted by global constitutionalism. There is no reason to elevate it over potentially conflicting values such as increasing equality or due process of law. Secondly, even if predictability is an important value, not all predictability is good: a better entrenched tyranny is not thereby normatively to be preferred over a less stable tyranny. Thus threats to the stability of a global order need not be regrettable – that will in part depend on the grounds of dissatisfaction by the coalition, and whether new candidate authorities contribute to a normatively better global order. Thirdly, even when predictability is a value, the international context is less susceptible to predictability, so any undermining of predictability by CM may be more marginal in the international context. Predictability of this sort may generally be more difficult to secure in multi-level political and legal orders. Thus many scholars note that federal political orders – and especially ‘asymmetric’ such arrangements – are more susceptible to ‘constitutional politics’ than are unitary states.^{Footnote 28} A fourth aspect is that CM might not actually lessen predictability in any substantial way. CM may not prevent stable expectations, but rather increase the opportunities for forum shopping and contestation for various actors. This may make predictions more complex, but not impossible about how norms will be specified, and what other actors will do. The upshot is that the critique that CM may challenge stability appears to be substantially overblown.

V. Piecemeal adjustments without a constitutional convention or moment

An optimistic interpretation of CM might argue that such new authorities provide better governance of the issue area at stake. Even when this is the case, CM challenges one of the values of a constitution, namely the benefits of ensuring more coherence or even integration of independent legal orders – to promote finality and predictability, and to harness the unintended and systemic effects of individual actions and of insulated authorities. ‘Tinkering’ by CM to improve on a part of the global basic structure runs the risk of creating highly problematic unintended consequences, lacunae etc – a risk which is even more worrisome than in the domestic setting, given that there is no institution which can be expected to address and alleviate such problems from a sufficiently all-encompassing perspective. This does appear to be a major challenge to CM. However, it should not be overdrawn: there is no reason to believe that the current cluster of international regimes and authorities operates in a coherent way with a high level of normatively desirable integrity. Rather, the complex interdependence of domestic, regional and international authorities is opaque, with impact very difficult to discern. Thus criticism against CM on this score must also be addressed to the present global basic structure – which has developed without the benefit of a global constitutional convention.