The above argument may be valid in a strict sense, but it also possible to take a less absolute and more progressive view of the process of integration. Consider the Haas quote above with the early emphasis on the word process. If we understand integration in the sense of process rather than as a finished state, there are many more opportunities to observe it, and history provides a useful vantage point. The drivers of regional integration therefore may be seen as that collective of factors that, historically and at present, may contribute to processes of integration. In fact, the above list of what integration is not includes many factors that some would consider to be drivers of integration. For example, if efforts at conflict mitigation or resolution by a regional organization lead to the adoption of the norm of humanitarian intervention, and this is then accepted by member states as it has been in the case of the African Union (see Constitutive Act, paras. 138-139) and subsequently applied to other conflicts, then we might say that some level of integration of decision-making and policy coordination (in respect of intervention) had taken place at the political level. This would be true even if the norm were imperfectly, or inconstantly, applied. The same will hold true for decisions regarding the removal of barriers to trade (as in the long-standing proposals for a Greater Arab free trade area, still only partially achieved) or human rights legislation, as in the case of ASEAN’s adoption of a charter in 2008. If such decisions are charter-or treaty-based, or if they come to be increasingly regarded as binding by members, we may say that some sharing of authority, even transfer of sovereignty, with an international institution has taken place. Policy integration and coordination thus occupy a very wide spectrum of informal and more formal processes. This discussion may be usefully likened to one regarding the interpretation and application of ‘hard’ versus ‘soft’ law. In an analysis of state preferences regarding the uses of hard and soft law, Abbott and Snidal consider situations in which states might choose hard law to order their relations, noting that such a choice ‘restricts actors’ behaviour and even their sovereignty’ (Abbott and Snidal 2000: 421-422). Hard law, in this case may be likened to formal regionalism. But if hard law is harder and more restrictive, soft law nonetheless may also shape preferences and even lead to the adoption of harder laws. The same clearly occurs with regionalism, where the terms hard and soft, or alternatively formal and informal are frequently used (Fawcett 2013). Illustrating the case of early regionalisms in Pacific Asia, Foot has shown how ‘soft regionalism’ emanating from a sense of ‘regional awareness’ or belonging has become more embedded in consensus based policies (Foot 1995: 229). Indeed, Pacific Asia provides a good example of how soft regionalism has, over time, yielded harder regionalism, as the charter of the ASEAN shows (Prantl 2013: 6; Wong 2015). States in regional organizations are constantly making and remaking decisions about desired degrees of legalization. And, contrary to critical analysis, restrictions on

state autonomy are in fact widespread and growing, not only in wellestablished international institutions like the EU, but also less evenly and to different extents in ASEAN, the Mercado Común del Sur – Southern Cone Common Market (Mercosur) or the African Union. While scholars of integration, particularly those influenced by European models, may insist on more rigorous definitions, some conceptual relaxing and a focus on regional integration as a process, one that is necessarily punctuated and prolonged (like processes of democratization) is desirable. This permits a discussion of a wide variety of regional activity that can be clustered under the more general heading of regionalism, understood here as the formation of regionally-based groupings for the purposes of policy coordination (Fawcett 2004: Van Langenhove 2011: 48-49). The intention then is not to demonstrate whether, or how far, regional integration has succeeded or failed according to a rigid set of criteria, but rather to observe some of the conditions and drivers that have contributed to the start-up and development of the integrative potential of regional organizations. In using a historically informed approach, it is hoped not only to observe patterns of regional integration over time, but to add further explanatory power to debates about the drivers of regionalism and to demonstrate how such an approach can benefit comparative study, allowing us to elaborate further on the ‘problem of comparison in comparative regionalism’ (Lombaerde et al. 2010). Without denying the importance of the statistical apparatus that many scholars favour, one particular problem of the problem of comparison is precisely that it is often ahistorical and deprived of context, meaning there is not a viable basis for comparison across cases and across time. For example, without understanding the local conditions and particular constraints and opportunities that attended early efforts at regional cooperation in the Middle East or Southeast Asia – for example state fragility, domestic contestation and external intervention – it makes no sense mechanically to compare statistical indicators of integration. Hence nearly all efforts to compare nonEuropean with European or North Atlantic regionalisms – whether in the areas of trade or security – disappoint for their adherence to fixed criteria and expectations (Robson 1993). It is noteworthy in this context that newer literature on regionalism precisely rejects the notion of dominant or exemplary models, but advocates the possibility of alternative models or practices, particularly where newer actors, like China for example, are involved (Kavalaski 2009). Scholars of non-European regionalisms have become much more active in pointing to the different pathways and explanations for the development of regional institutions not only in Asia, but also Africa, Latin America or the Middle East (Journal of European Integration 2010). History, again, is particularly relevant to this argument because any alternative model will inevitably reflect the different experiences of any given region. In discussing the Southeast Asian case, Mark Beeson writes how the ‘history of place’ is a necessary accompaniment to understanding

the contours of regional cooperation (Beeson 2009: 5). Extending this line of thought, I argue that this history of place is essential to understanding all regionalisms. However, in taking a historical approach, it is also important not to become too entrapped by the arguments of historical institutionalism, whose attention to institutional starting points and pathways can be over-determined and allow little space for change. While such arguments may seem compelling, particularly in refuting strict rationalchoice perspectives, it is important to remember that however much ‘institutions constrain and refract politics . . . they are never the sole cause of outcomes’ (Thelen and Steinmo 1992: 3). In the analysis that follows, while acknowledging the stickiness of institutional norms and practices over time, I also highlight how different drivers of regionalism explain not only institutional start up but also change. Although the focus of this volume is primarily to draw comparisons between Europe and Asia, I seek to broaden the discussion to a wider set of comparative cases to illustrate the argument. I do not propose to offer any detailed analysis of the European case, but rather to use it as a point of reference. Though Europe casts a long shadow on regionalism – with the EU in particular influencing institutional design and activity around the world – and European (and non-European) specialists continuing to focus on Europe as model and flagship, there has been increasing interest in and recognition of non-European regionalisms and the experiences and lessons they offer, demonstrating that there is far more to regionalism than Europe (Telo 2007). This is in contrast to early analyses of comparative regionalisms whose focus was heavily, almost exclusively, European. The relevance and possibilities of non-European regionalisms has been brought further into focus in a world of new and rising powers and at a time when Europe itself is struggling with the consequences of expansion amid a widespread and protracted economic crisis whose proportions threaten to weaken some of its own instruments. ‘Southern’ or developing country regionalisms have not fallen victim to the effects of global crisis, at least not to the same degree: their levels of integration were shallower; their projects less ambitious. Indeed they have arguably benefited or learned from them: both in Latin America and East Asia regional institutions were strengthened by the perceived failure of multilateral institutions to support them through earlier financial crises in the 1990s (Higgott and Phillips 2000: 375). In another sphere, that of security, African institutions were similarly strengthened by the evident inability of the United Nations (UN) to deal with a series of acute crises following the end of the Cold War, inspiring the motto ‘African solutions to African problems’ (Olonisakin and Ero 2003: 233-234). Indeed, as discussed below, attempts to complement or even substitute for the UN and other multilateral institutions constitute important drivers or motives of regionalism. My purpose here is not to focus on crisis as a driver of regionalism (Fioramonti 2012; Gillespie 2015;

Ryan 2015), though I acknowledge its importance and believe that crisis needs to be integrated into any historically grounded perspective. Indeed historical institutionalists would argue that crises are important in providing critical junctures and therefore in explaining change in regional histories (Pierson 2004: 135). Consider the Second World War and the case of Europe. The war and its consequences, giving rise to a strong desire for an institutionally embedded peace, were seen to provide an important kickstart to the development of what became the EU. The shock and decentralizing effects of the ending of the Cold War similarly were seen to unleash a new regional wave, not only in Europe, but globally (Fawcett and Hurrell 1995: 1-3; Mansfield and Milner 1999). More recently, one might argue, the crisis posed by Arab Spring events has prompted a serious reconsideration of the workings of Arab institutions like the League of Arab States and the Gulf Cooperation Council (GCC), one that was reflected in decision-making regarding intervention and regime change in Libya, Yemen and also Syria. Or that ASEAN’s turn to humanitarian assistance has been prompted in part by a series of natural disasters in its region, leading, in 2011, to the launch of a Humanitarian Assistance Centre in Jakarta (website online, available at: www.ahacentre.org). Rather than studying crises directly, the purpose is to explore some alternative and common drivers, particularly of Southern regionalisms, over time and in doing so to speculate about their diverse roles and meanings in an emerging multilateral order. Within a historically informed framework, the focus is on three particular drivers, all of which help to inform our understanding of distinct regional histories and provide some bases for comparison:

1 Ideas and beliefs about regions and regionalism. These have been highly formative in influencing the start-up and determining the progress of regionalism not only in Europe, but also, for example in Latin America, Africa or the Middle East, where collective identities and pan-ideas, partly born of colonial experience, have acted both negatively and positively on integration processes. Ideas, and the individuals and groups who promote them – state leaders as well as intellectuals – act as a guide and constraint, setting parameters for what is possible and acceptable. Different regionalisms have distinctive intellectual histories that merit independent exploration.