Notions of border in regionalism theory and praxis

This article provides a concise and critical overview of the notions of border within regionalism theory and praxis. It also discusses the last decade’s increase of border barriers in this globalized world and its relation to regionalism. We provide a historical approach to theoretical construction and consider implied ideas of border derived from theoretical and empirical developments of regionalism. This paper finds that old regionalism indicates a notion of border as separation; new regionalism implies a notion of border as filter. The current wave of comparative regionalism identifies two apparently contradictory processes (multilevel regionalisms and nationalisms) that actuate the formation of (cross) border regions embedded into tangled webs of global networks, which requires a globalist perspective of regionalist dynamics.


Introduction
Two questions guide this brief overview of (international) regionalism, both attempting to take to greater lengths the thinking of the theoretical and empirical relationship between regional processes and border: (1) How has regionalism understood borders and their role in regionalist dynamics in history? (2) How can comparative regionalism recognize borders in face of today's political recrudescence of (inter)national barriers? The objective of this article is to provide a concise outline of the notions of border within regionalism theory and praxis. It also aims at discussing the last decade's increase of border barriers in this globalized world and its relation to regionalism. Important to note the uncontemplated objective of providing exhaustive account of theoretical production, as neither is envisaged the analysis on the concepts of border. Rather, this paper only reflects upon the intrinsic notions of border within major regionalist processes and theories from 1945 to current days.
Regionalism(s) and border(s) are intertwined social constructs. As social ideas, they affect each other both in functional and in conceptual terms; as social institutions, they are integral parts of the same multiescalar networks. However, the production of theoretical analyses on this relationship remains incipient. As asserted by Vaughan-Williams (2009), borders are still a blindspot of international relations theory, let alone in regionalism theory. This article intends to fill this gap by connecting the established (theoretical and empirical) waves (or phases) of regionalism with the parallel development of the notions of border and its roles.
The first section of this article presents the theoretical and conceptual bases of this overview, in which we propose an understanding of regionalism and border and develop an analytical framework. The second and third parts are focused on the analysis of the first two waves of regionalism vis-à-vis the multiple roles of border in the historical development of both ideas/institutions. The fourth section elaborates on the current phase of regionalism, discussing the last decade's upsurge of border barriers. Final remarks conclude this paper.

Interrelating regionalism and notions of border
Regionalism is any (social, cultural, economic, or political) process occurring within, toward, or across a region. It is a space in which (public or private, state or non-state) players act and produce norms (Santander, 2012). Regionalism can also be considered a type of order (as configurations of power, for example) and, in this perspective, is closely connected to the formation of collective identities and the creation of borders and boundaries (Lapid, 2001). This reveals the relational facet of regionalism and borders as well as of the conceptual effort here undertaken. 1 Regionalist dynamics play thus a significant role in giving borders a political -hence practicalmeaning as such relations unfold. Regarding its influence on the role of border, regionalism functions in a nonexclusive dichotomy, surpassing the political boundaries delineated by international borders while preserving territorial limits and logics (Smouts andBadie, 2006 [1996]). Regionalism can either function as an escaping mechanism -as political processes designed to avoid state territoriality -or as a reinforcing tool to its limiting role. Both functions coexist in a dynamic space of continuous political interrelations.
To know which border function is being used, why, and for who depends on the historical contexts in which key actors of regionalism (be it the state, industrial/commercial sectors, civil society, etc.) are placed, as well as on the theoretical efforts to embrace such contextual reality. This indicates that theory and praxis are interrelated. The understanding of (geo)political contexts in which and through which thinkers of international relations formulated their concepts and notions is key in the process of grasping theoretical models and conceptual frameworks. Looking at theories in the light of historical contexts facilitates the comprehension of not only the construction of knowledge regarding regionalism and border, but particularly the way scholars understood these objects.
After the consolidation of the nation-state, principally after the end of the Second World War, three phases of regionalism presented distinct socio-economic and political events, and theoretical explanations that molded the understanding of the roles of both regionalism and border. The phases of old regionalism (1945-1990), new regionalism (1990-2000), and comparative regionalism (2000-today) are contrasting macro-scenarios. Diverse world orders, various forms of international organizations, and changing interests (Söderbaum, 2016) shaped the roles of regionalism and the roles of border within regional relations and international systems. Hence, the theoretical accounts on regionalist developments follow the historical background of each phase, provided the contextualization necessary.
There are significant differences regarding regionalist scholarship and notions of border within each period. Old regionalism was characterized by the predominance of rationalist epistemology as it refers to the period of consolidation of IR theory (realist and rationalist at its birth). It understood border under its separation attributes. New regionalism witnessed the emergence of reflectivist epistemology, which tended to conceive border under its filtering properties. Comparative regionalism is plural (or eclectic) in its core, able to navigate through rationalist-reflectivist epistemologies. It allows multiple understandings of border and is most suitable for apprehending the fast-paced shifts in the global scenario in which regionalism is embedded. However, the rising tensions regarding border policies nowadays require a more global view on the way comparative regionalism understands both regions and borders.

Old regionalism: borders as separation and control
Old regionalism encompasses the so-called classic regional integration theory  and their posterior revisions , all of which concentrated their focus on the European experience of regionalism, extending it to some sort of comparison to other experiences around the globe, with special attention to Latin America and Southeast Asia. The main question behind these theories focused on knowing how states could maintain peace and rebuild economies after the end of the Second World War. They tracked the historical developments of the following five decades, attempting to provide explanations for the creation of integrated, supranational entities in Europe (in all its ups and downs); for the highly cooperative settings that led to the signing of dozens of regional agreements in Latin America (and the demise of most of them); and for the informally built cooperation systems in Asia.
The Cold War, fueled by the growing tension between the East and the West, and the hegemonic power exercised by the United States (USA) and the Soviet Union onto Europe, Latin America, Asia and the rest of the world sponsored in great part the recrudescence of borders and the protection of nation-states. Among diverse routes taken by European countries to deal with the post-war socio-economic disaster in a sensitive political environment, the reconstruction of borderlines and the delimitation of space and territory found place in every national agenda (Newman, 2006). However, aware of the perils of nationalism, policymakers and scholars designed treaties and engendered agreements to surpass the territorial rigidity of nation-state. Alter and Steinberg (2007) show, for example, that the key role of the European Coal and Steel Community (ECSC), created in 1951, was assuring European countries that Germany would not again develop a dangerous, leading power propensity, and rebuilding Franco-German partnership, with minor impact in regional economy. ECSC also became a blueprint for subsequent agreements and a relevant forum for political discussions. In sum, the objectives of this organization and of later agreements were defined in terms of security through trade.
On the other hand, the foundation of the Latin American Free Trade Association (Lafta), nine years after ECSC, demonstrated it to be grounded on different objectives. Lafta aimed at creating a free trade area to enjoy the benefits of an integrated economy, increasing national markets and deepening intraregional commercial ties through the progressive liberalization of tariffs and duties -thus promoting conditions for a common market (Díaz, 2015). Although Lafta did not reach its objectives, what is important here is that (differently from Europe) regionalism was a means to solve macroeconomic issues in Latin American countries and achieve development. It also aimed at ensuring nation-building by doing so (Söderbaum, 2016). Similarly, the foundation of the Association of Southeast Asian Nations (Asean), in 1967, was propelled by the need to foster economic development and reduce regional conflict while not interfering in domestic affairs. In many ways successful, Asean managed to create a more stable regional environment -politically and economically -particularly in its first two decades of existence (Yue, 1996). In common, the three anecdotes demonstrate that despite recognizing the need of a cooperation that surpassed state borders, regional projects during this wave upheld them in its highest levels. The Europeans, by bypassing borders' separation functions to engender a calmer political environment; the Latin Americans, by attempting to integrate but refusing to lower commercial obstacles and defending national production; the Asians, by promoting regional security through the protection of territory.
The notions of border in old regionalism theory had thus followed such regional political developments, which in general promoted the strengthening of nation-state after the Second World War, the protection of Latin American nation-building in face of the international economic environment, and the security and protectionist policies of Southeast Asia. The theories of federalism, (neo)realism, transactionalism, (neo)functionalism, intergovernmentalism, and institutionalism created explanations on how to build peace and prosperity with political and/or economic integration in Europe (Rosamond, 2000;Wiener and Diez, 2009), also endeavoring to grasp the attempts toward regional integration in Latin America (Haas and Schmitter, 1964;Aitken and Lowry, 1972) and the Asian cooperative systems (Leifer, 1978;Arndt and Garnaut, 1979).
However, these theories were neither concerned with borders, nor were they inclined toward theorizing the effects of regionalism on border functions. Witnessing the foundation of a myriad of regional institutions, such theories were first and foremost focused on explaining state-led regionalism, which made them focus on analyzing such organizations, their developments, and their capabilities and limitations as regional frameworks. Regarding ontological and epistemological similarities, such theories privileged the rational-thinking paradigm, formal (and top-down) approaches, the analytical centrality of the state and of its institutions (such as material power, sovereignty and territorial control mechanisms). Non-state actors were occasionally acknowledged either as integral parts of national units or as demand generators. Concerning their differences, old regionalism theories presented different emphases (economics, political alliances, trade, etc.) and concentrated on certain aspects of integration, such as political negotiations, institution-building, power, and others.
The roles of border, from the viewpoint of regional institutions, in Europe, Latin America, Asia or elsewhere were solely acknowledged vis-àvis the needs of nation-states, which were aware of the necessity to reinforce positions of security, trade, and economic development, applying regionalism to bestow upon their borders provisional, dissimilar statuses. For parties of the same cooperative/integrative institutions, borders would eventually become less separative (in some sectors); for parties outside of the regional bodies, borders would continue to be altogether reinforced. Thus, regionalism and border were complementary elements of state-led policies concerning (and within) a global political system. This goes hand in hand with a definition of 'regional integration' by Malamud (2011, p. 219): [regional integration] can be understood as an attempt to rebuild the eroded national borders to a higher level. Therefore, it can be interpreted as a protectionist maneuver by states that cannot guarantee their interests and objectives by themselves, and seek to do so in conjunction with others. 2 In old regionalism, there was no problematization of 'region' as a relevant concept, since the regional organization was understood to be its materialization. A spatial notion of region was therefore non-existent. Following the same line of thought, territorial contiguity was a prerequisite for region-building, that is, contiguity of national territories was the sole manner to form a region. Thus, both in theory and in praxis, regional external borders were thought to work as hermetic barriers to protect the regional space in a way states had been found unable. The continuation of the separation and control functions of regional internal borders have not, however, hindered their relativization according to specific regional necessities, albeit the simultaneous reinforcement of such closing functions in the (re)construction of nation-states across the globe continued in fast pace.

New regionalism: borders as filters
The phase of new regionalism (1990)(1991)(1992)(1993)(1994)(1995)(1996)(1997)(1998)(1999)(2000) refers principally to world order vicissitudes caused by globalization. Technological advances in communications and transportation changed the way countries, companies, and people related to each other. If in the former phase the nation-state was the channel for and the center of global societal processes (international in its core), in new regionalism, such exchanges were increasingly transnational, transcending national territories in ways not before experienced, increasingly slipping away from state's control. The transnationality of the driving forces of political, economic, and cultural dynamics due to the increase of global trade and transnational economic flows, the intensification in speed and density of communication networks, and the augmented agitation between and within diverse cultural collectivities raised the awareness of a global interconnectedness, which begot the sense of new opportunities and challenges in all levels of society (Eriksen, 2007). The end of the Cold War and the decline of bipolarity also altered the structure of the international system.
In face of such deep changes, regionalism did not remain unaffected. It became a mid-level alternative of policymaking -neither global nor national (Söderbaum, 2003). Its multidimensional attributes, not limited to issues of economy, trade, and security, evinced a variety of institutional frameworks, and a more active participation of civil society and corporate actors (Mansfield and Milner, 1997). Dozens of organizations were created and hundreds of regionalist agreements and processes were launched by different actors with diverse social, political, and economic objectives. Non-state actors demanded inclusion in regionalist dynamics. Such a movement, called by many 'regionalization' (or informal regionalism), became an important driving force of regional processes. The state attempted to protect its territorial sovereignty in face of growing fluxes, transferring power "upwards, downwards, and sideways" (Perkmann and Sum, 2002, p. 4), thus territorializing policy. Subnational and transnational units of governance emerged. Once ignored or neglected as a secondary fixture of the interstate system (Williams, 2003), borders became an asset to expand markets and acquire a relevant role in an increasingly global economy. 3 Border areas were rapidly transformed into strategic poles in a public-private rush to attract transnational capital. The permeability of borders, highlighted in the mobility of capital, industry, information, and people, made scholars and policymakers even question their relevance in economic terms within this new world order (Ohmae, 1995).
The heterogeneity of new regionalism theories was depicted by the difficulty in finding a common ground in the definition and problematization of 'new regionalism' as basic concept. In common, only a vague idea that a different phase had started and the attempt to grasp it (Dabène, 2009). The proliferation of theories and approaches ranged from varieties of mainstream (i.e. rationalist) theories (neoliberal perspectives of institutionalism and trade, neorealist accounts), multilevel governance approaches, constructivist, reflectivist and critical perspectives as well as the so-called new regionalism approaches. The emergence of constructivist accounts had significant impact in challenging the conventions based on rationalism whereas reflectivist approaches shed light on the 'for whom' question and on the problem of the inconspicuous purpose of regionalist dynamics (Söderbaum, 2016).
Such a proficient theoretical production was key for a more detailed analysis on border.
Although not yet linking directly regionalism and border, new regionalism theories became aware of the globalization effects on macro and microlevels (of social reality, and as levels of analysis) as well as the empirical connections that macro-regions and microregions shared. Söderbaum (2005) highlighted the need to transcend the nation-state as the central spatial category, moving toward a more multiescalar perspective.
Regional space and scales of regionalism were then at the core of such a discussion. With such levels already recognized as relevant analytical elements, De Lombaerde (2010) developed a proposition to link the academic communities that tend to focus on macro or microregions, thus suggesting new research paths and a clearer understanding of global relations. The World Order Approach analyses inspired on Coxian thought by Gamble and Payne (1996), Hook and Kearns (1999), and Breslin and Hook (2002) acknowledged the existence of regions of center and periphery, and problematized microregions as well as their borderlines. Borders were being used to assure the inclusion of certain microregions in the global economic system by becoming the focus of a productive restructuration of the world. What these theories identified was the filtering character of borders -opened to capital, products, and information; closed for individuals of peripheral regions. Not only understood as a point of shock and a boundary, but as an area of transition between subnational dynamics (Leloup and Stoffel, 2001), the global insertion of borders accentuated their function as a multidimensional articulation structure.
The relationship between new regionalism and borders, in theory and in practice, is that of partial opposition. On the one hand, the impetus of global(izing) flows ran against the continuance of borders to the extent numerous scholars predicted their extinction. On the other hand, borders were key factors for the exploitation of relative positioning of certain territories within the expanding transnational system. The main progress of this wave can be identified as the acknowledgement of borders and border regions as germane categories for regionalism theorization as well as in the empirical advance of regionalist dynamics. The persistence in relegating border as a significant concept to other academic communities than regionalism scholars (e.g. political geographers, sociologists, and anthropologists) however had placed it -still -in inferior position in new regionalism theory, overshadowing its capabilities of explaining changes in border policies and systems within regional environments.

Comparative regionalism: borders as components of multilayered systems
Globalization became unescapable as well as the advent of numerous actors and phenomena of global scale. Terrorism and non-traditional warfare on terror; emerging powers and formation of new politico-economic poles; rise of interventions in conflicts of third parties and 'responsibility to protect' (R2P); intensification of conflict in fragile states and negative consequences on global economy and on national stability; recurrent economic crises and financial breakdowns; increase of social and economic inequality. The list goes on. This new world order is progressively more fragile and volatile, as is the state and the conventional mechanisms used to respond to internal and external demands (Sørensen, 2016). The on-going wave of comparative regionalism (2000-today) involves the necessity of dealing with these issues, which require multilayered governance, and which involve agency of multiple actors in the global scene (De Lombaerde and Söderbaum, 2013).
Many authors (Fawn, 2009;Burki, 2011;Strange, 2014;Söderbaum, 2016;Vivares, 2018) concur that in face of such intricate matters, state and non-state actors have chosen to use regionalism as a preponderant means to assure their own existence and/or to take advantage of new possibilities that the global systems provide. The centrality of regions in offering solutions to the contradictions of this century's global relations -(inter)nationalism and globalism, states and markets, security and insecurity -is the core of comparative regionalism theory (Katzenstein, 2005). The progressive integration among European Union (EU) members and its enlargement toward the East in the early 2000s (Moravcsik and Vachudova, 2003), the Latin American search for regional autonomy (in a post-hegemonic world order) and the strengthening of South American institutions for dealing with matters of regional politics (Riggirozzi and Tussie, 2012), and the creation of multiple forms of interregionalism (Baert et al., 2014;Mattheis and Litsegård, 2017) all seemed to corroborate the reinforcement of regionalism's significance in world affairs.
Regarding theoretical grounds, approaches drawn from sociologically inspired accounts, now much more developed as alternative ways to explain regionalism (such as reflectivism, and critical theories) have provided a more complex, comprehensive view on the 'whys and hows' of region-formation (Acharya, 2012), opening pathways toward plural (or eclectic) methodologies and theories (Söderbaum, 2016) -although it remains unclear how to apply an eclectic epistemology. The fact is, mainstream theories of previous waves now share room with novel understandings of regionalist dynamics, designed to cope with the rapid changes in global relations, as traditional views did not manage to explain them. The multiple forms of region established in this wave has allowed the comparison between organizations, informal regions, subregions, subnational regions, states, etc. in economic, social, cultural, political, and symbolic terms (Van Langenhove, 2012).
Borders, now studied in most diversified ways, have slowly made their route into comparative regionalism theories. They have started to be scrutinized as microprocesses, informal (regionalization) dynamics, or simply as relevant social processes of regionalism. Perhaps the main advance has been to identify border in its region-building capacities within a multilevel governance environment. Among numerous concepts created, 'borderland' reflects such attempts to understand borders within regional systems (Matiuzzi de Souza, 2017). Nevertheless, borders persist as a factor yet to be explored in comparative studies of regionalism, particularly in macro-analysis. There remains a lack of research linking borders to regional policymaking and other regionalist dynamics in a comparative (not isolated) way, with few exceptions made in the study of mobility across regional organizations (Arcarazo and Geddes, 2014;Gülzau et al., 2016).
Global occurrences of the last decade were yet to intricate the already complex interconnected global system in which regions are (trans)formed. Among them, (1) the economic and financial crises that impacted the world in the last decade (American banking catastrophe (2007), European debt crisis (2008), global economic recession (2009) and the subsequent, worldwide political instability of both domestic politics and international relations spread over the continents; (2) the rising of conservative leaders throughout the world and their domestic-oriented policies, such as the Trump administration in the USA; (3) the decline of American hegemony in global politics and the concomitant emergence of China as robust player; and finally (4) the recrudescence of borders as important boundaries to 'protect' national territories, sovereignty, and even living standards, such as the rise of border control within the Schengen Area (currently reintroduced in France, Austria, Germany, Denmark, Sweden, and Norway) to contain the free flow of undesirable immigrants, and the strong lobby for the building of a physical wall in the United States-Mexico border.
As highlighted by Scott (2006, p. 3) "[…] questions of interstate relationships and political community are about borders and their wider significance". The increase of border securitization and control (and in some cases militarization) by the state has been one of the biggest challenges of regionalism. It can be considered a state response to such a chaotic and crisis-plentiful phase. Dupeyron (2017) reminds that the refugees and migration flows into the EU have split the European border policies with the restoration of control points and routines, as well as the referendum that confirmed the United Kingdom (UK) exit of the EU has rebuilt British borders to European citizens. The priority given to security issues for the USA-Canada borders has created a 'biometric border' and a hierarchical (instead of cooperative) field of cross-border relations. Concerning USA-Mexico borders, the boosting economic and cultural region they form has not been enough to prevent USA to invest in further control mechanisms, regardless of constructing new walls.
Evidence shows that the actions of players not concerned with regional policies (or acting without regionalist objectives) can affect how a regional organization will close or open their borders. It can also influence how states react within a regional organization. Illustrative examples may be: the fracture of subprime market of the US banking sector produced a long, global financial crisis that was later translated into political turmoil in many EU countries that now support disintegrative movements, which in turn pressure national politicians to leave the EU and close borders; the advance of Chinese 'appetite' for primary goods in Latin America reduced Mercosur's capacity to strengthen its industry, affecting how its members would negotiate internal trade tariffs (more protectionist); borderlands became very attractive to market actors due to special taxation in one country, yet they can be militarized due to increase of migration flows caused by internal conflicts in third countries.
The list of examples could go on and on. What is demonstrated is that the regionalism-border relationship today is that of a multilevel global connectedness due to the multiescalar character of border dynamics and the result of 'intermestic' policies vis-à-vis regional and global junctures. Understanding regionalism as a mid-level alternative thus does not suffice for acknowledging the influencing factors on regionalist phenomena and the relationship between regions and borders. Quite the opposite. Recognizing that the world is organized in tangled webs of networks that cut through and across all geographical levels (Dicken, 2015) allows us to identify regionalism and border as spatial components through which actors interact in a much more complex, multilayered, interconnected system of manifold networks. It permits us to understand how the concentration of economic activity and/or of political power in certain areas do influence the way regionalism and borders are used by nation-states and regional organizations. It joins together regionalism and borders as integral parts of a single complex macrosystem.

Final remarks
The notions of border have changed along with the theoretical accounts on regionalism. If in the first phase regional organizations were at the core of regionalism analyses and borders were understood as state's fixtures, the second phase showed a more complex understanding of border functions vis-à-vis the rapidly emergence of global interconnections, shifting its attention to many types of regions. Therefore, old regionalism replicated the notions of border originated in the studies on state and interstate relations; on the other hand, new regionalism developed a double notion of border, as an articulative component able to filter flows exchanges -opened to some but closed to others. One substantial difference is the acknowledgement of borders and regions as responses to the increasing globalization in the latter, while in the former the international system served as mold to comprehend both regions and borders.
The wave of comparative regionalism is unlike both previous phases to the extent that it is witnessing the strengthening of apparently contradictory processes: (sub, inter, and intra)regional movements, organizations, competitions, and agreements -in macro or micro-realities, from local to transnational levels -are still gaining ground within global relations; the re-emergence of border barriers and nationalisms -also within regional organizations, but not exclusively; and the rise of many global issues and the following demand for a global governance. These processes actuate the formation of (cross) border-regions, concomitantly opened and closed, local and global, national and regional. Thus, we identify that the wave of comparative regionalism has been emphasizing the so-called multilayered interconnections of borders into the tangled webs of global networks. This means that it will be conceivably much more complicated to understand and predict the roles of border (and of regions, for that matter) by applying a regional, analytical framework that does not include a globalist perspective of such dynamics.