The Breton Model Between Convergence and Capacity

Abstract

Drawing upon mainly qualitative inquiry with political, associative and economic actors over a two-decade long period, the article seeks to provide answers to a key conundrum that challenges, in different ways, territorial politics scholars, as well as those working primarily on France. What are the conditions for a successful form of regional advocacy in a unitary state? The French region of Brittany has a specific mode of operation, one based on mixing identity and instrumental claims, and accessing a repertoire of responses that are not naturally open to other French regions. A related question follows logically from the first: Can a specific territorial model developed in one set of conditions adapt when circumstances change? The Breton case demonstrates limited evidence of endogenous change (a central tenet of discursive institutionalism), though it does admit a continuing capacity to filter external pressures in a way that makes sense to regional actors. Analytically, the article develops territorial political capacity as a part material, part constructed framework that can be used for comparing regions at a particular point in time, as well as for capturing the evolution over time of a specific region.

Extracto

A partir de principalmente entrevistas cualitativas con protagonistas políticos, asociativos y económicos durante un periodo de dos décadas, la finalidad de este artículo es ofrecer respuestas a un enigma clave que desafía, de maneras diferentes, a eruditos políticos territoriales, y los que en su trabajo se centran sobre todo en Francia. ¿Cuáles son las condiciones para conseguir un tipo de apoyo regional que tenga éxito en un Estado unitario? La región francesa de Bretaña tiene una modalidad específica de funcionamiento basada en combinar reivindicaciones identitarias e instrumentales, y acceder a un abanico de respuestas que normalmente no están abiertas a otras regiones francesas. La primera cuestión lleva de forma lógica a otra cuestión relacionada: ¿puede un modelo territorial determinado que se ha creado bajo una serie de condiciones adaptarse cuando cambian las circunstancias? El caso bretón demuestra poca evidencia de un cambio endógeno (un principio central de la institucionalidad discursiva), aunque admite una capacidad continua para filtrar las presiones externas de un modo acorde con los protagonistas regionales. Analíticamente, en el artículo se desarrolla la capacidad política territorial como una estructura parcialmente material y parcialmente construida que puede utilizarse para comparar regiones en un momento dado, y para captar la evolución con el paso del tiempo de una región específica.

摘要

本文主要是以质性研究，对政治、联合与经济行动者进行为期二十年之久的探问，企图藉此对一个关键难题提出解答，该难题以不同的方式，挑战了地域政治学者和以法国为主要研究对象的学者。在单一国家中，成功的区域倡议形式的条件为何？不列塔尼的法语区，则具有一种特殊的操作模式，该模式建立在混合身份认同与工具性的要求，以及取得并非自然对其他法国区域开放的全部回应的基础之上。一开始便随之而来的相关逻辑问题是：在一组条件中建构的特定地域模型，当面临境况改变时，是否能够进行调试？不列塔尼的案例，显示出内生改变的有限证据（论述制度主义的核心原则），但该案例却承认了以对区域行动者而言有道理的方式来筛选外在压力的持续能力。在分析方面，本文将地域政治能力发展成为部分是物质、部分是建构的框架，并可用来在特定的时间点上比较区域、以及捕捉特定区域随着时间的演化。

Résumé

Puisant principalement dans une enquête qualitative auprès des acteurs politiques, associatifs et économiques sur une période de vingt années, l'article cherche à fournir des réponses à un dilemme clé qui représente, de manière différente, un défi aux chercheurs dans le domaine de la politique territoriale, ainsi qu’à ceux qui poursuivent des recherches sur la France. Quelles sont les conditions préalables dans un État unitaire qui sont indispensables pour la création d'une forme réussie de plaidoyer régional? La région française de Bretagne a un mode particulier de fonctionnement, fondé sur une mixité de propositions quant à l'identité et de nature instrumentale, et ayant accès à un répertoire de réponses qui ne sont pas par nature à la portée des autres régions. Une question connexe découle logiquement de la première: est-ce qu'un modèle territorial spécifique mis au point dans certaines conditions peut s'adapter au moment où les conditions changent? Le cas de la Bretagne démontre que rares sont les preuves du changement endogène (un principe fondamental de l'institutionnalisme discursif), bien qu'il reconnaisse une capacité continue d’éliminer les pressions externes d'une manière qui est logique aux yeux des acteurs régionaux. Du point de vue analytique, l'article développe la notion de capacité politique territoriale comme un cadre, moitié matériel, moitié construit, dont on peut se servir pour comparer les régions à un moment donné, ainsi que pour capter dans le temps l’évolution d'une région particulière.

Keywords:

• devolution
• regions
• Europe
• territory
• new regionalism
• multi-level governance
• politics

INTRODUCTION

Drawing upon extensive qualitative inquiry with leading political, associative and economic actors in the French region of Brittany over a period of almost two decades, the article seeks to provide answers to a key conundrum that challenges, in different ways, territorial politics scholars, as well as those working primarily on France. What are the conditions for a successful form of regional advocacy in a unitary state? A related question follows logically from the first: Can a specific territorial model (the ‘Breton model’) developed in one set of conditions adapt when circumstances change? Brittany is arguably the best case in France to demonstrate successful forms of regional advocacy (as opposed to ethnic nationalism (the case in Corsica) or the dynamics of metropolitan governance (Lyon)). A sense of regional identity is a constant across the period, measured both by responses within the political and policy community and by survey evidence. This article draws upon interviews carried out by the authors at three distinct stages in the recent past: in 1995, from 2001 to 2002 and in 2013.¹ A representative survey was also commissioned in 2001, and Brittany formed part of the cross-national Citizenship after the Nation State (CANS) survey in 2009.² The article develops territorial political capacity as an operational concept for comparing regions and for capturing the evolution over time of a specific region. Though the level of analysis is territorially specific (the administrative region of Brittany), the approach is sufficiently generic to be used in other regions and is being explored in cognate surveys of Andalucía (Spain), Wales (UK) and Wallonia (Belgium).³ We now present territorial political capacity as a framework for analysis.

THE TERRITORIAL POLITICAL CAPACITY FRAMEWORK

Territorial governance can be defined as the set of formal and informal structures and processes that contribute to the governing of a given territory (Simoulin, 2013, pp. 3–25). Regions and cities, at the heart of reflections about contemporary territorial governance, produce the strongest territorial narratives (Brenner, 2004; Keating, 2008; Le Galès, 2012; Pasquier et al., 2013; Storper, 2013). Cities and regions are both affected by similar dynamic forces: economic globalisation, European integration and the reform of the state in a context of diminishing state capacity and resources (Keating, 1998; Hooghe and Marks, 2001; Pasquier, 2012). These macro parameters create new opportunities for some cities, regions and other territorial actors, but also impose new constraints and forms of political and economic uncertainty.

Different theoretical grids have been proposed to analyse the dynamics of territorial governance (see Pasquier et al., 2013, for an exhaustive literature review). Our preferred framework, territorial political capacity, is briefly presented in this article. Territorial political capacity has both a material and a more constructed dimension. The framework combines hard and soft variables: potential indicators of regional capacity sometimes require quantitative data and sometimes mobilise qualitative evidence. We select eight indicators of territorial political capacity which can each have a material, as well as a constructed dimension (though, as described, they lie more or less in one camp and are analysed in this way). These indicators are briefly presented in Table 1 and discussed more fully in the main body of the article.

Table 1. The territorial political capacity framework.

Political capacity	Constructed (Brittany)	Material	Evidence
Material
Institutional resources	Imagined territorial assembly	Legal competencies and finances	The legal status of the region
			The fiscal structure of sub-national budgets
Economic profile	Narrative of spatial isolation	Fiscal capacity, GDP per head, wealth of public and private partners	Official Statistics (INSEE, OECD)
Mixed
Inter-governmental relations (IGRs) and multi-level governance	Repertoire of strategies to press the Breton cause in multi-level arena	Instrumental use of identity in dealings with Paris; strategic Europeanisation	Case studies of specific examples of the Breton interest: the high-speed train; the refusal to merge with Pays de la Loire region, 2014
Political leadership	Pursuing the Breton interest by leadership	Long-term regional collective leadership	Role of first-rank leaders operating mainly in Paris/Brussels
Party system	Pursuing the Breton interest by party	A powerful regionalist party? Regional advocacy within existing parties?	Influence in state-wide parties (PS, UMP), little evidence of split voting
Territorial regimes	Pursuing the Breton interest by association	Region-centric networks?	Spatial planning, inter-governmental lobbying and EU regional policy processes
Constructed
Identity logics	Sense of belonging, identification of self	Comparison with other French and European regions	‘Moreno’ scale identifies a dual identity
Territorial narratives	Regional discursive ‘common good’.	Breton negotiations as a two-level game	The 2013 ‘red bonnet’ movement confirmed key features of the territorial action repertoire

The framework and its indicators are best understood in terms of a continuum rather than a set of hermetic positions. Mainly material indicators include institutions and institutional resources (What is the degree of self-rule?); and economic profiles (the economic well-being of a region and its ranking relative to other places). Mixed material and constructed indicators include styles of IGRs and multi-level governance (e.g. Do regions engage with strategic Europeanisation?), the party system (Can we identify a regionalisation of the party system or regional advocacy in central government?), political leadership (Is there a form of (regional) territorial political leadership?) and, finally, the operation of territorial regimes (Is there a consensus between political, economic and associative actors?). Mainly constructed indicators include territorial identity (Is the regional space underpinned by a (regional) territorial identity or not?); and territorial narratives (Is there a coherent shared view on the nature of the challenges facing the region, or a shared repertoire of responses?). The mix of material and constructed indicators is used to provide insights from one particular region, Brittany, that has been observed closely over two decades.

The general framework is used here to study one strong identity region over time, but it ought to have a more general usage. Comparing city governments, for example, also involves combining material and constructed variables (economic dynamism, formal legal competencies, fiscal autonomy and capacity, the party system, city visions and traditions, forms of urban political leadership, urban coalitions and narratives of the city's common good). In the case of city governments, or city regions, the balance between these variables is very likely to be different; stand-alone fiscal autonomy, for example, is much more likely to be a fundamental source of differentiation than in the case of one strong identity region. How this framework might be operationalised is now considered in relation to the case of the French region of Brittany.

THE BRETON MODEL BETWEEN CONVERGENCE AND CAPACITY

Brittany is one of France's most distinctive regions (Mcdonald, 1989; Ford, 1993; Le Galès, 1993; Le Bourdennec, 1996; Le Coadic, 1998; Pasquier, 2004; Gemie, 2007; Nicolas, 2012; Le Coadic, 2013). Though the symbols of statehood of this former Duchy have long been suppressed, the region retains many distinguishing characteristics. The Breton language is the European continent's only Celtic language. The spectacular growth of Breton cultural movements (music, dance, theatre and costume) is testament to a revival of Breton values and self-consciousness. A Breton political model is commonly identified as being based on cross-partisan consensus and regional advocacy, a post-war model of public action that was built upon inter-war failure of a more assertively nationalist movement. Breton perspectives across the political spectrum continue to be shaped by the spatial challenges of Brittany (at the far-western tip of the European continent), and claims for the legitimate delivery of public goods by the French state and by the European Union (EU). For several decades after the Second World War, a Breton economic model, based on a mode of economic production that is diffused across the Breton peninsula, was proposed as a source of emulation for other peripheral regions (Phlipponeau, 1993; Ollivro, 2005). We have argued elsewhere that Brittany scores highly in terms of institutional inter-connectivity and social and cultural capital, embodied by traditions of inter-communal co-operation, close informal relationships between regional politicians and representatives of the state field services, a vibrant associative life, strong electoral participation and robust social networks (Pasquier, 2004; Cole, 2006).

How useful is the territorial political capacity framework for understanding the case of Brittany? This section mainly considers the changing material dimension of the Breton model; in the next section, we explore whether the social construction of Brittany can provide a resource to tackle economic uncertainty. The core argument is developed, mainly inductively, around three phases of empirical data gathering, in 1995, 2001–2002 and 2013–2014. The largely positive framing of the Breton model is mainly based on the 1995 and 2001–2002 interviews; the decline of the Breton model is the main organising theme for the latest round of interviews in 2013–2014.

Institutional Resources

Institutional resources are the most obvious indicator of territorial political capacity. The most influential instrument in the past decade has been that of the Regional Authority Index (henceforth RAI), which draws on the work of Elazar (1991) and disaggregates regional authority into a set of dimensions for self-rule and shared rule (Hooghe et al., 2010, p. 6). Self-rule is operationalised as ‘the extent to which a regional government has an independent executive, the scope of its policy competencies, its capacity to tax and the extent to which it has an independent legislature’. Shared rule, the ‘capacity of a regional government to shape central decision making’, is disaggregated across four dimensions: law-making, executive control, fiscal control and constitutional reform.

A narrow reading of the RAI would not place Brittany very high in the European hierarchy, or, for that matter, any other French region. The RAI measures the self-rule axis in terms of an independent executive, the scope of its policy competencies, its capacity to tax and the extent to which it has an independent legislature. In the case of Brittany, the regional council does not approximate an independent executive (unlike, for example, the Welsh or Scottish governments); indeed, the region has no hierarchical authority over other levels of sub-national government, the 101 departments or the 36500 communes. The policy competencies of 22 French regions have traditionally been limited. If the regions gained new responsibilities in the 2003 decentralisation reform, the departments emerged with more staff, policy responsibilities and budgets. French regions have limited fiscal capacity, which we define in terms of having access to a wide tax base, some freedom in setting tax rates and borrowing capacity. Initially able to exercise a measure of fiscal autonomy, the French regions (including Brittany) now depend upon central government transfers or formulae for over 90% of their resources and are strictly limited in their ability to levy taxes.⁴ Finally, thus far Brittany has had nothing approximating an independent legislature. In the run-up to the decentralisation reform announced for June 2014, the Brittany regional council and leading Breton parliamentarians (J.-J. Urvoas) called for a Breton territorial assembly with secondary legislative powers, along the lines of the UK's devolved assemblies. The outcome of this demand was unknown at the time of writing.

The score sheet is meagre, but it is necessary to understand informal institutions to make full sense of the Breton model. Informal institutions have taken three main forms: a mode of close co-operation between regional and regional state actors; a region-centric mode of territorial governance and a close synergy between political, business and associative actors (see below). There is a long tradition of collaboration between state and regional elites going back to the early post-war period which remains present in the discourse of actors. In an interview in 2013, one key collaborator explained how one regional Prefect had been ‘socialised’ into the Breton way:

I have known three prefects. At the beginning, the second one was firmly of the belief that ‘the State decides alone’, but by the time he left he was a strong supporter of the Breton system of a dialogue between the State and the local authorities.⁵

This example could be repeated with several others drawn from earlier testimony.

There is, second, a mode of regional regulation, whereby the regional council has long adopted a leadership role among cities and departments in Brittany, a mode of regional governance that has no real equivalent elsewhere in France. This regional leadership role is well illustrated by the B16 process.⁶ Regional leadership has been facilitated by the absence of a powerful city mobilising metropolitan functions. Brittany's post-war economic revival was attributed by some political geographers to a polycentric model of spatial development (Phlipponeau, 1993; Ollivro, 2005). Even after continuous post-war growth, the regional capital Rennes remains only one of several medium-sized Breton towns. These towns, with their dense tissue of small and medium enterprises, include Redon, Vitré, St Brieuc, Lannion, Brest, Morlaix, Vannes, Quimper, Lorient and several smaller places. This balanced model of economic development has been challenged by economic globalisation and the impact of the post-2008 crisis, which has had a detrimental effect on towns such as Lannion (telecoms), Brest (defence), Lorient (agri-food) and even Rennes (the automobile industry). Globalisation provides opportunities for a few regions and cities able to incarnate the new modernity in international competition. But economic globalisation also creates losers; this process has gathered pace since the 2008 crisis and is reflected in some senses in our Breton case study.

Economic Profile

Economic dynamism and more generally economic well-being are fundamental concerns for understanding the political capacity of regions. The stronger regions of the European continent are also those that enjoy a significant level of economic growth (Baden-Württemberg, Bavaria, Basque Country, Flanders, Rhône-Alpes and northern Italy). In a globalised context ‘where the supply chains and knowledge networks are not locationally restricted’ (Amin, 2004, p. 36), certain authors have refused the notion of the region as a closed territorial unit. They have instead adopted a relational and permeable conception of the region in terms of networks and interdependencies (Allen et al., 1998; Amin, 2004). As Morgan (2007) rightly points out, however, regional political spaces continue to have borders in the sense that political representatives are obliged to conceive elections in territorial terms; and for this simple and prosaic reason alone the postmodern geographer should not lose sight of the political within territorial policies.

Is there a linkage between material and constructed capacity? There is certainly a material basis to regional identity: the most autonomist-minded European regions are usually those with a more developed sense of material capacity, richer regions such as Catalonia, Flanders, Bavaria, Lombardy and (in some respects) Scotland. Wealthy regions and ‘region-states’ have greater capacity to impose their interests than poor regions (Ohmae, 1993). Such regions often demand greater stand-alone fiscal autonomy, resent welfare transfers to poorer places and sometimes advocate enhanced territorial autonomy. These richer regions constitute an exceptional family of European regions. In our comparative framework, Brittany (along with the cases of Andalusia, Wales and Wallonia) forms part of another family of European regions that we label as ‘second-order strong identity’ regions. This family of regions share many characteristics. They are economically challenged yet have a distinctive and developed territorial identity. They each have traditions of social-democratic party control. They are traditionally pro-European regions, or at least regions benefiting from substantial EU investment. Though these regions exercise limited control over core macro-economic levers, and a fragile basis for enhanced material capacity, they sustain a coherent territorial narrative that mixes signifiers from the centre–periphery and the left–right axes.

The rapid demographic and economic expansion of Brittany during the post-war period was remarkable. From being the poorest French region in 1945, Brittany was in the second quartile of French regions in 2013 in terms of GDP per head (Insee, 2013). The post-war period was marked by heavy inward investment into the region, mainly from the French state, but also by foreign direct investment and the EU (Phlipponeau, 1993; Ollivro, 2005). Brittany enjoyed strong rates of economic growth, testament to the modernisation of its agricultural sector, external investment in high technology sectors such as telecommunications, defence and health and a specific pattern of multipolar spatial development. Brittany is today France's principal farm export region and a leader in the agri-food industry. In a material sense, however, the Breton miracle has turned sour since the early 2000s, though economic decline is felt more keenly in west and central than in the eastern Brittany. The 2013 round of interviews occurred at a time of economic crisis, especially severe in the farming and agri-food businesses. The key underpinnings of the Breton economic model were under threat. The advantages identified in earlier decades had, by 2013, been transformed into heavy economic constraints. State-led investment in the telecoms and automotive industries had concentrated industrial capacity, so much so that economic decline in these sectors shook the model to its foundations. The highly intensive Breton agricultural model was showing signs of diminishing economic returns at a very high environmental cost. The European attachment was called into question by a series of high-profile EU challenges to the underlying Breton economy. The ending of EU subsidies for the poultry industry in summer 2013 came as a rude awakening. In the opinion of one interlocutor, the Breton model had ‘failed to reinvent itself when it could’.⁷ The mass strikes and the direct action undertaken in Autumn 2013 were signs of a radical disenchantment.

The decline of the Breton model is in part a material reality, based on a weakening in the core economic markets of traditional post-war growth (farm products and agri-food business), a threatened natural environment (with a knock-on effect on tourism) and cut backs in production in the key automotive, telecommunications, IT and defence industrial sectors. Material decline ought to be relativised, however. Brittany remains in the second quartile of French regions in terms of GDP and other regions (Languedoc-Roussillon, for example) have fared far worse since the economic downturn. The decline is most apparent in terms of constructed image and capacity; the ability of the Breton model to deliver public goods has been called into question, as much by self-doubt as by the diminishing ability of central government to engage in fiscal transfers. The French state's ability to deliver side payments has been challenged by the rise of global markets (with stiff competition for French food producers from Brazilian or German firms), by EU regulation (in the area of public services), by diminishing French macro-economic capacity and by tougher budgetary rules.

The shift in Breton attitudes can be mapped by comparing the answers of the 2001–2002 and 2013 interviews. All correspondents were asked, in an open-ended question, what they considered to be the main policy problems facing Brittany in the years ahead. In the 2001–2002 round, three types of response came ahead of others: first, problems related to the environment; second, the need for more regional economic development and third the need for a strengthening of the powers for the regional council. Responses to the policy challenges question in 2013 suggested a weakening of the Breton model, and a sense of uncertainty about what might replace it. The crisis of the Breton model in general and the agricultural and food production sectors in particular was identified as the major challenge. There was a deep sense of pessimism about the future of intensive agriculture, inspired by the massive layoffs or plant closures in 2013 at food producers Doux, Tilly-Sabco and Gad. Potential responses varied from developing more quality, niche products, to assuming a more protectionist stance against foreign and European competition, to diversifying away from agriculture. A secondary challenge was identified in terms of environmental damage and its effects on tourism and marine biology. The Breton ‘economic miracle’, based in part on intensive agricultural methods, had endangered one of Brittany's greatest assets, its reputation for unspoiled natural products.

There is no simple correlation between economic resources and territorial political capacity. If richer regions can develop more autonomous strategies, such responses do not always occur. The less economically advanced regions in our sample are able to mobilise on the left–right and centre–periphery cleavages and use territorial identity to support resource redistribution claims. Both cases represent a form of territorial political capacity construction. This observation strengthens the analytical point that comparison should occur on the basis of ‘family resemblances’ (Collier and Mahon, 1993)—comparing in categories such as cities, first- and second-order strong identity regions, rural zones—rather than in terms of context-blind variables.

Modes of IGRs and Multi-level Governance

In the RAI, shared rule is described as the ‘capacity of a regional government to shape central decision making’ and disaggregated across four dimensions: law-making, executive control, fiscal control and constitutional reform (Hooghe et al., 2010, p. 115). The French regions would appear weak on most of these criteria: there is no shared law-making via a second chamber representing state interests (as in the case of Germany or some other federal states). There are no formal mechanisms of executive control. There is no share in determining national levels of taxes (as in the Basque Country and Navarre in Spain) and no formal input into constitutional reform (though the agreement of the second chamber, the Senate, is in practice required for any constitutional reform).

Focusing on formal variables does not capture actual processes of inter-governmental lobbying that, by their nature, are less open to formal observation and measurement than the RAI suggests.

Lobbying the French state (and the EU) for increased resources to rescue Brittany from its isolated geographical position has formed a core feature of the Breton repertoire for several decades. Several interviewees in 1995 and 2001 spoke of a Breton ‘network’ in Paris, actively defending Breton interests in the main ministries; by 2013, Bretons were running many of these ministries (Roudaut, 2013). From 1950 onwards, Breton actors co-operated closely in the CELIB—Comité de d’étude et de liaison des intérêts bretons—which brought together politicians from all parties, along with professional and economic interests to promote the interests of the region. Under the impetus of CELIB, Brittany was the first French region to publish a regional plan (in 1953), calling for industrialisation and improved transport facilities (Martray, 1983). The lobbying activity was crowned with success, as the French state poured massive resources into Brittany from the late 1950s onwards. Interviews with surviving actors and published historical accounts demonstrate that instrumentalist ends (bringing a backward region into national productivity) coexisted within the CELIB with a high degree of regional consciousness, a desire for powerful regional political institutions and a willingness to consider direct action to further Breton ends.⁸ The sometimes instrumental use of identity is a constant of the Breton policy style, reaffirmed regularly in the past 50 years.

The second core feature of the Breton (interactional) model is a strategic engagement with the EU. Interviewees across the three periods stressed the fundamentally pro-European sentiment existing within Brittany, with Europe part of the Breton ‘DNA’.⁹ In the immediate post-war period, Breton politicians were quick to recognise the opportunities implicit in European integration for a spatially peripheral, economically disadvantaged and politically contested region.¹⁰ Catholicism remained powerful and influential in Brittany at least until the 1960s (Rohou, 2012). In part, the region's pro-European engagement was a general consequence of the modernisation agenda of the Catholic Church and, more specifically, the political influence of Christian democracy. The Christian democratic, pro-European legacy has a long pedigree and its values are still daily propagated by Ouest France, the leading local newspaper. Along with Alsace and Lorraine, Brittany was the strongest pro-Maastricht region in France, with these three regions securing the victory of the Yes vote in the 1992 referendum. Brittany was one of the only regions to vote Yes in the 2005 constitutional treaty referendum. The region has been a key French beneficiary of EU structural funds and the Common Agricultural Policy.

The territorial political capacity framework also contributes to debates about Europeanisation. Carter and Pasquier (2010) identify strategic Europeanisation as being one response to European integration that can empower regions in terms of their own identity, strategy and networks. Strategic Europeanisation emphasises regions as actors, capable of refining their own EU strategies. In contrast to strategic Europeanisation, Carter and Pasquier (2010) identify normative Europeanisation as the process whereby domestic laws and regulations are adapted to conform to EU directives and rules. Such top-down Europeanisation is typically analysed in terms of reception, of adaptation, of goodness of fit, or conforming to European templates and best practice.

The practice of strategic Europeanisation has strengthened internal cohesion across Brittany. In 2005, the regional president Le Drian created a ‘European conference’, which brings together the four departments and the main cities of Brittany to define common positions around European issues. Within the conference a consensus soon emerged that any European funds should be used to co-finance the high-speed rail line from Brest to Paris. Via their office in Brussels (rebaptised the ‘Breton Embassy’ from 2006), and a certain amount of informal horse-trading in Paris and Brussels, the Breton representatives worked together, with the support of the French government, to achieve advances in two areas: that rail transport would be included in the ‘competitiveness’ operational programmes; and, most importantly, that the calculation of ‘Lisbon’ earmarked funds be carried out at national rather than regional level. This political agreement that Le Drian managed to obtain from Danita Hübner, Commissioner for Regional Affairs, went against the advice of the Directorate-General for Regional and Urban Policy.¹¹ The agreement enabled Brittany to receive 100 million euros from the European Regional Development Fund (ERDF) to co-fund the high-speed rail line, nearly a third of its ERDF envelope in 2007–2013. This example epitomises a proactive regional strategy that has engaged with Europe over several decades. The established framework of interaction with the Commission, the unanimity that regional actors demonstrate concerning the major issues of regional development and the perception of European civil servants that there are specific Breton interests are the defining elements in ensuring Breton representatives a certain influence over particular European decisions.

Has (Breton) territorial capacity been undermined by a shift from strategic to normative Europeanisation? One of the core novelties of the 2013 round of interviews related to disillusion with the EU as an institution (though not with the European ideal). The ‘neo-liberal’ European model of the Barroso Commission was contrasted unfavourably, in one interview with a UMP (Union pour une majorité populaire) parliamentarian, with the traditional Breton model of partnership, co-operation and support for public services.¹² The EU was associated with the downturn in Brittany's fortunes (though certainly not as the principal cause). The drying up of EU structural funds had removed a source of regional investment. The sense of relative deprivation was aggravated by the EU Commission placing obstacles in the way of direct aids from the French government. Above all the eco-taxe, the symbol of the Breton revolt of 2013, was implemented by French governments, but justified on the basis of an EU directive. The immediate fate of Brittany's poultry producers—Gad, Tilly-Sabco, Doux—had been made worse by the end of EU restitutions in July 2013.¹³ Several interviewees blamed the EU for allowing German and Brazilian firms (accused of social dumping and producing food and poultry below cost) unfair access to Brittany's traditional markets. This new tension between strategic and normative Europeanisation has weakened somewhat the European vision of Breton elites, but most criticism is directed against the French central state and its incapacity to reform the economy. In May 2014, the charismatic leader of the red bonnet movement, Christian Troadec, was also the head of a list for the European elections defending the federalist project of a Europe of regions.

The Party System and the Breton Interest

One dimension of (regional) territorial capacity lies in the party system. We identify two distinct configurations of regionalisation: first, the emergence of territory-specific (ethno-territorial) parties as the key parties in regional elections. The politics of ethno-territorial mobilisation reflects itself in sub-state political institutions, distinctive party systems, language rights movements and cultural traditions (De Winter and Tursan, 1998; Mcewan and Moreno, 2005; Moreno, 2007; Kernalegenn and Pasquier, 2013). There is a growing literature on the impact of split voting in ‘post-sovereign’ states. Jones and Scully (2006), for example, demonstrate the differential between voting for minority nationalist parties (Plaid Cymru, SNP) in devolved and UK-wide elections, the regionalists performing much better in the territory-specific elections than at the UK level where the traditionally unionist Labour Party continues to dominate. A rather different indicator identifies the weight of regional/nationalist lobbies within the national parties or national parliamentary institutions (in order to defend a territorial interest at the higher level). The case is best illustrated in the case of Spain, where the leading party in the lower chamber must often negotiate with pivotal parties (in general Basque, Catalan, Galician and Canary Island nationalist parties) whose votes are required to help it to pass government bills (Orte and Wilson, 2009). Which, if either, of these two configurations describe the Breton case?

That Breton politicians overwhelmingly share a ‘centrist’ political culture was affirmed often in interviews (especially those of 1995 and 2001). The political consensus emphasised the legacy of Christian democracy from 1945 onwards, the development of a Breton version of social democracy (from 1977) which shared key pro-European and social partnership beliefs with its predecessor, and a mode of political co-operation over and above party lines. We should sound a note of caution. The Breton model of consensus politics, built upon the Christian-democratic tradition and its legacy, was never hegemonic and its existence is questionable today. The model excluded the regionalist and autonomist movement (see below); and it did not prevent acrimonious relationships between Socialists and the Christian Democrats, especially during the 1990s.¹⁴ By 2013, the Breton model of cross-partisan consensus was wearing thin. The domination of the Socialist Party had become, by mid-2013, a withering hegemony, challenged by factional divisions within the governing PS (Parti socialiste), deteriorating relations with the Greens and the general unpopularity of the Hollande–Ayrault administration and disappointment with the Bretons therein.¹⁵ The centrist consensus that had sustained the Breton model appeared challenged by a more sharply divided political cleavage structure, exemplified by the red bonnet protests of Autumn 2013.¹⁶ By 2014, the Breton model was also threatened by the good performance of the National Front (FN) in the municipal and European elections of 2014.¹⁷ In 2012, already, the FN presidential candidate Marine Le Pen obtained scores reaching 15–20% in some semi-urban spaces and isolated small towns. By 2014, Brittany was no longer off-bounds for the FN.

Brittany is one of the few places that might be able to support an ethno-territorial party. In the Rokkan and Urwin (1982) schema, Brittany, combining economic underdevelopment with a strong cultural identity, ought to be a prime contender for a region that develops a strong regionalist or nationalist movement. Breton politicians, for sound instrumental and political reasons, have in the main chosen to operate through state-wide parties. The political opportunity structure in the French Fifth Republic is so resolutely Paris-centric that Breton forces have adapted accordingly. There remains a vibrant cultural movement, however, that sustains the territorial identity (and direct action repertoire) upon which Breton elites draw when constructing the Breton interest. There is a great deal of ambiguity in the interplay between the political and the cultural movements. In one interpretation: ‘many very well known Breton cultural activists are also openly nationalist. But they wear two hats and are careful, for funding reasons, not to confuse political and cultural demands’.¹⁸ This ambivalence was illustrated in interviews by the case of the fast-speed train and the proposed airport at Notre Dame des Landes. Most interlocutors favoured extending the fast-speed train to Brest, in the far west of Brittany (though opinions were more sharply divided on whether the Notre Dame des Landes Airport should go ahead). De Rohan (President of the Brittany region from 1998 to 2004) and Le Drian (President 2004–2012) combined their efforts to plead Brittany's case in Paris, on this and several other dossiers. A minority of cultural activists was opposed to both, however, in the name of defending Breton identity.

The presence or not of a powerful regionalist party, the attitudes adopted by national parties to issues of territorial identity, the rewards to be gained by emphasising one or another level of governance: all of these are in part shaped by the party linkage function. The party system performed an important role in representing the Breton interest, usually through the main national parties, but also in traditions of cross-partisan collaboration in the higher interests of Brittany. This collective endeavour explains why more robust forms of political leadership have not generally been necessary to defend the Breton interest.

Collective Political Leadership

Political leadership is a neglected variable for understanding territorial dynamics. Rather like supranational or national forms, studying territorial political leadership logically involves three main levels of analysis: the individual, the positional and the environmental (Lagroye, 1997; Cole, 2012). The individual level can refer to the mobilising qualities of particular individuals. Media attention usually focuses on Livingstone or Johnson as successive mayors of London or powerful mayors of French cities such as Collomb in Lyons, or Delanöe in France. The positional level refers rather to the capacity of particular institutional offices and party configurations to sustain clearly identifiable forms (urban or regional) and styles (executive-centred or collective) of leadership. Environmental constraints and opportunities present a third level of analysis. The key variable here relates to the hierarchy and importance of the entity embodied (the mayor of London or Paris, or the minister-president of a powerful German land is usually a significant national player as well). In the case of Spanish autonomous communities or devolved governments in the UK (Salmond in Scotland, Mas in Catalonia in particular), this level also involves consideration of the sometimes highly politicised and conflictual forms of inter-governmental relationships.

In Brittany, collective regional leadership has a long pedigree. In the traditional post-war model, defending the Breton interest occurred either via a parliamentary group (in the Fourth Republic) or through Bretons operating within the leading parties, especially the Socialist Party since the 1980s. Our 2001 interlocutors in Brittany were virtually unanimous in praising the capacity of Breton actors to join forces to promote their common interests and to defend Brittany against attacks from the outside world. The leader of the Breton Socialists, later President of the Regional Council and Defense Minister, expressed it thus in 2001: ‘In Brittany, we all believe in the merits of the Breton compromise. Left and right are perfectly capable of coming together and agreeing on the big questions that concern the region.’ The (then) Gaullist President of the Brittany region shared this analysis.¹⁹ From René Pleven (the Breton PM in the Fourth Republic committed to European integration) to Jean Yves Le Drian (the influential Defense Minister under President Hollande, 2012–) the Breton interest has been supported by influential Bretons operating in Paris and Brussels, rather than by powerful regional leaders, per se. The example of Le Drian is instructive: the Defense Minister (whose ministerial office has a section devoted to following Breton affairs²⁰) informally doubles up as Minister for Brittany in the Valls government. The paradox of political leadership is that Bretons have punched above their weight in Paris and Brussels (where power lies) but that this modus operandi recalls the current weakness of the Regional Council (far from being a powerful territorial assembly).

As elsewhere in France, there is some evidence of competition from big city mayors (traditionally from Edmond Hervé, the former mayor of Rennes (1977–2008)) but there is less evidence of a visible metropolitan leadership than in many other French regions. Neither of the two Breton métropoles—Brest and Rennes—can claim to be in the first rank of French cities.

Territorial Regimes

Capacity does not limit itself to institutional resources or leadership potentialities; it also involves styles of co-operation between actors in well-defined policy universes. As practices and norms are produced by interaction, networks are key to understanding territorial dynamics. Postmodern regional geographers seek to challenge an outmoded image of regions as cut-off and replace it with a vision of regions as a series of relationships, of networked regions at the intersection of global connections. Although this image of interconnected cities and regions finds support within the literature (Paasi, 2001; Scott and Storper, 2003; Le Galès, 2012), this approach has been criticised for the tendency to expunge the notion of territory from regional studies. As Amin (2004, p. 36) observes, ‘the sum is cities and regions without prescribed or proscribed boundaries which translates into regional territories without rules to define them’.

Our approach is focused mainly on within-territory interactions. Which levels of sub-national authority are at the centre of territorial interactions? Are local and regional authorities able to federate interactions within territorial policy communities? Or do powerful professional interests (business and trade unions in particular) look beyond the sub-national level? Classic studies of urban regimes—first in the USA (Stone, 1989), later in Europe (Stoker and Mossberger, 1994; John and Cole, 1998a)—emphasised long-term coalitions between political and economic decision-makers as being a fundamental part of urban governance. More formal network analysis (John and Cole, 1995; John and Cole 1998b) has been used to demonstrate variation between cities; while narrative accounts have reached similar conclusions in relation to strong regions (Barone, 2011). What is the appropriate level in the case of Brittany (and how does this compare with other French regions)?

Political capacity is, at least in part, a process of mediation in which elites and social groups produce a vision of the world that allows them at once to structure relations among themselves and to define the ‘interests’ that they are pursuing collectively. In Brittany, the CELIB (see above) provided well-documented evidence of a robust territorial regime, which stabilised relations between political, economic and cultural actors from the 1950s onwards (Pasquier, 2004). This regional coalition negotiated a regional development project and forced the state to support it. Sectoral committees within the CELIB brought together all relevant regional stakeholders: farm leaders, university researchers, business leaders, higher civil servants and other relevant experts were mobilised to draw up a new model of the regional economy. Not only did the CELIB engage the state in a dialogue concerning the necessity of increasing public-sector investment in Brittany, but it also proposed alternatives to the solutions put forward by the national administration and then ensured the diffusion of its ideas through publications, aimed both at key decision-makers and at the general public. Today's leaders have received the model as an inheritance from the preceding generation. In their hands, it has been adapted to contemporary conditions but the dialogue between political, economic and cultural elites carries on. On the economic front, a new business association ‘Made in Brittany’ was created in 1995 to enhance Breton products and goods. Two decades later, it represents more than 300 firms and 60,000 employees in Brittany and has become a key interlocutor for regional political leaders. This economic association adopts a broad Breton regional advocacy position, supporting proposals for a single regional assembly and for the reunification of historic Brittany. From this point of view, the territorial regime of Brittany is stronger than ever.

THE SOCIAL CONSTRUCTION OF BRITTANY

Identity Logics

Interpreting territorial governance from an identity-based perspective implies looking to regions, or stateless nations as historic, cultural and political entities and identifying territorial networks or ethno-territorial parties sharing a common identity (De Winter and Tursan, 1998; Guibernau, 1999; Gagnon and Tully, 2001; Gagnon et al., 2006; Moreno, 2007). Interpreting the region as a subject assumes understanding territories as being more than individual units of analysis; rather as communities with shared memories, common inter-subjective meanings and a recognisable territorial action repertoire.

Though sometimes considered as the archetype of the unitary state, modern France also contains several territories with highly distinctive identities (of which Brittany, Alsace, the Basque country, French Catalonia, Corsica and Savoy are the most obvious examples). As expressed in an interview carried out by the authors with a regional actor in July 2013, the problem with Brittany ‘lies in the 21 other French regions’. In October–November 2013, the social movement of the ‘red bonnets’ illustrated the singularity of the Breton model of collective action. Using identity symbols such the Breton flag (Gwen ha du) and the red bonnet (the symbol of the Breton revolt at the end of seventeenth century against a new tax imposed by the King Louis XIV), this regional coalition composed of different groups (business owners, employees, cultural associations and regionalist leaders) forced the central government to stop the implementation of a new environmental tax and to pledge to negotiate a new economic model for Brittany.²¹ This example confirms survey evidence providing some support for this framing of the Breton region as distinctive. Dargent (2001) revealed that Brittany was the only mainland French region with a sense of regional identity as strong as that of its national belonging (i.e. French). Cole and Evans (2007) concluded, on the basis of survey evidence carried out in 2001 using the Moreno scale, that Bretons combined a strong sense of regional (Breton) and national (French) identity, a finding confirmed by the 2009 CANS survey, according to which 75% of the Bretons interviewed declared themselves, to varying degrees, to feel at least as Breton as French (Hendersen et al., 2014).²²

Territorial identity can matter more than budget-maximising or abstract scalar calculations. This preference was demonstrated in the June 2014 horse-trading over the new regional map, when former regional President (and incumbent Defense Minister) Le Drian preferred to maintain the administrative region of Brittany as it currently stands, rather than accept a larger region including all of the Pays de la Loire region. Though the current administrative region is composed of four departments (Côtes d'Armor, Finistère, Ille et Vilaine and Morbihan), the historical boundaries of Brittany included a fifth department, the Loire-Atlantique, in which is located the city of Nantes, historical capital of the Duchy of Brittany. In 1955–1956, when the map of the French regions was drawn up, Nantes had been separated from Brittany to act as the capital of a new region, the Pays de la Loire. From this period onwards, there has been a traditional tension between Breton political and cultural elites seeking a reunification of historical Brittany and those preferring a larger region including all of the Pays de la Loire region. The tension between these two coalitions was revived in 2014. Caught in the crossfire between Le Drian (and most Breton elites, backed by public opinion) in favour of the reunification of historic Brittany, and former Prime Minister, Jean-Marc Ayrault (who, former of mayor of Nantes, also resisted the five department solution and proposed the fusion between Brittany and Pays de Loire), President Hollande chose to maintain Brittany and Pays de la Loire as separate regions with their existing boundaries. Breton political elites preferred not to dilute their region into either a vast Western France or a merged Brittany–Pays de la Loire region, even if this meant that the coveted Nantes would remain outside.

Territorial Action Repertoires?

The territorial répertoire, which centres on the practices and discursive registers adopted by actors over time, completes the presentation of our framework. Tilly's (1984, pp. 89–108) concept of the collective action répertoire was initially developed to uncover historical regularities in forms of social protest. It has been usefully adapted by writers such as Keating (1998) to studying the processes of social and institutional mobilisation in Western European regions. Practices form the analytical core of the territorial répertoire. Are there specific practices and modes of operation that distinguish Bretons over time? Are there modes of routine behaviour (the combination of direct action and lobbying, for example) that have proved effective in the past and that are a natural response to crisis situations? Following practices come discourses. Schmidt’s (2010) theory of discursive institutionalism might also have heuristic value in understanding territorial regimes: how actors communicate and share ideas is critical for understanding continuities and changes in policy universes.

Pasquier (2004) summarises the Breton territorial repertoire in the following terms: the regional space is a legitimate one, not an artificial construct; actors share the common referents of Brittany as a backward region that needs to be modernised and deserves special treatment; actors seek to build regional consensus, over and above departmental and local rivalries; the Breton policy community seeks a collaborative mode of governance and to build a measure of trust in relations with central government (but using direct action tactics to back this up). There are shared views on the regional interest and a shared belief in negotiating with the French state as an equal partner. The sense of regional identity goes much deeper than the electoral performance or otherwise of an explicitly regionalist party. There was very strong support in interviews at all three periods for strengthening the regional political institution and for vesting the Brittany regional council with a leadership role in relation to other sub-national authorities. There was widespread support, among almost everybody interviewed in 2001–2002 (irrespective of party), for enhanced regional powers in the spheres of education, culture, historic monuments, transport, training, regional languages and, to a lesser degree, health. If anything, there was an even stronger consensus in favour of enhanced regional powers in 2013, coupled with a sense of disillusion with the inability of Breton politicians to influence the incumbent Ayrault government in favour of more regionalisation.

Our longitudinal investigation allows us to break down the Breton territorial action repertoire into core themes, selected in terms of frequency of response over the three periods to two linked questions. The first of these related to the qualities of the region itself and the mode of operation of relationships therein:²³ this question was designed as an inductive means of uncovering regularities, both synchronically (by comparing three distinct groups of actors within Brittany) and diachronically (comparing responses to similar questions over time). A second question asked in 2001–2002 and 2013 looked to the future, namely ‘what are the main challenges that decision-makers in Brittany face over the next ten years’.²⁴ The first question was designed to elicit perspectives on the constants and contingent variables of the Breton territorial action repertoire; the second question was intended to provide a cross-sectional overview of perceived priorities.

In Brittany We Do Things Differently

The statement obtaining the highest score was: ‘In Brittany we do things differently’.²⁵ Interviewees across the three periods referred to a Brittany effect, particularly in relation to the outside world. Though conflicts within Brittany could be fierce, there was a common front presented to the outside, whether the French state, the EU or other regions and countries. Some interviewees referred to Brittany's past as an independent Duchy, a degree of autonomy they equated with pre-unification Scotland.²⁶ Politicians from all parties, including the UMP, have a tradition of joining forces and defending the Breton interest in Paris (and Brussels). Some evidence of this was provided in interviews, notably concerning the common cause of all Breton parties to obtain central government investment in the high-speed train (LGV project) or in relation to the role of Breton language teaching as part of the normal school curriculum.²⁷

The Twin Faces of the Breton Negotiating Model

Strong linkages with Paris and Brussels are combined with a willingness to resort to direct action tactics to further Breton objectives. Direct action strategies are an established part of the collective action repertoire, sensu stricto, which mobilises symbols of Breton identity—the Breton flag, music, occasionally the language—in order to defend narrow economic interests. Direct action forms part of a longer term collective action memory. In the protests against the eco-taxe of Autumn 2013, for example, striking workers and farmers wore red bonnets, an allusion to the anti-tax protests of the seventeenth century. Two decades earlier, rioting farmers and fisherman caused the burning of the Breton Parliament in 1994. Road blockages and rail sabotage were adopted to further the Breton cause in the 1960s and 1970s. The ability to resort to direct action demonstrates one of the twin faces of the Breton repertoire that does not sit easily with other more negotiated representations, and notably that the Bretons are politically centrist. In Tilly's (1984) model of the collective action repertoire, the dynamics of social protest are rooted in a historically long time frame. The Breton collective action register drew as much from the eighteenth-century repertoire in pre-revolutionary France (the preservation of the historic rights of specific territorial communities (refusal to pay tolls, for example), the defence of local communities or corps (defence of the language), and adopting a common cause with employers to negotiate with central authority (with Breton employers such as Armor Lux or Tilly Sabco joining in the social movement)), as it did from a more ‘modern’ nineteenth-century repertoire (the practice of direct action exercised on public authorities). While negotiations stalled in Autumn 2013, direct action produced immediate pay-offs: Brittany was the object of a full Council of Ministers meeting (in October 2013) and a far-reaching plan in December 2013 (in the pure tradition of the CELIB).

CONCLUSION

In the introduction, the core research question was presented as: What are the conditions for a successful form of regional advocacy in the model of a unitary state? Regional advocacy in Brittany has been effective insofar as it has combined heightened regional consciousness with an awareness of the most effective modes of operation within the political opportunity structure of the French polity. Brittany has a specific mode of operation, one based on mixing identity and instrumental claims, and accessing a repertoire of responses that are not naturally open to other French regions. Identity markers are very present in dealings with Paris, in claims made upon public resources, in the demand for differential treatment and resentment of ‘Jacobin’ logic of centralisation. The favoured mode of operation is an elite-level one, but it can also draw upon a social movement that uses the symbols of regional identity and cultural distinctiveness to strengthen its case. Territorial distinctiveness does not express itself by a powerful ethno-territorial party, which would be counterproductive. One Breton autonomist interviewed in 2013 estimated the audience for the main autonomist party, the Democratic Breton Union (UDB) at 3–5%.²⁸ Though the political opportunity structure is not conducive to political regionalism, in the case of Brittany a powerful cultural movement provides an influential milieu for both regionalist and French parties (Storper, 1995). No other French region (qua region) is able to combine regionality and regional advocacy in this way. However, the broader context remains one of limited decentralisation in a unitary state: the French regions have modest legal powers or financial capacities, and (at the time of writing) these have not been substantially strengthened by recent decentralisation reforms of 2004, 2010 or 2013–2014. Hence, the continuing reliance on a specific model of regional advocacy, however imperfect, has demonstrated its capacity to produce public goods.

These observations lead directly to the second question: Can a specific territorial model (the ‘Breton model’) developed in one set of conditions adapt when circumstances change? The method adopted throughout the article is inspired in part by discussions over Schmidt's discursive institutionalism, which understands institutions as sets of understandings, ways of knowing, practices and routines that are embedded within recognisable structures. In the specific case of Brittany, the territorial action repertoire might be interpreted as a form of discursive institutionalism, a mix of beliefs, ways of operating, routines and practices that make sense over time. For Schmidt (2008, p. 303) what ‘actors think and say’ is key to understanding change. As presented here, however, the Breton territorial model of 2013 had a rather defensive character. Departing from one of the central tenets of discursive institutionalism, our survey of Brittany casts some doubt on the endogenous character of change. The core drivers of change lie beyond the discourse and involve changing material realities. External factors include the impact of economic crisis on a model relying heavily on external investment, changes to the farm subsidy regime, a certain disillusion with the EU and the bureaucratisation of scarce resources. Internal factors include the diminishing returns associated with three decades of PS hegemony, the economic crisis and its spillover in rural and semi-urban France and the lessening fiscal capacity of the French state. These heavy variables underpin the evolution of the Breton model. The Breton case demonstrates limited evidence of endogenous change (a central tenet of discursive institutionalism), though it does admit a continuing capacity to filter external pressures in a way that makes sense to regional actors. The evidence supports another established hypothesis, drawn from Bourdieu, whereby actors interpret crisis in terms of their existing mental maps or habitus (Dobry, 2009).

Above all, the article has developed territorial political capacity as a hybrid framework for understanding and comparing cases of territorial governance across time and space. In the specific case of Brittany, Table 2 outlines the direction of change from 1995 to 2014, corresponding to the long fieldwork period.

Table 2. Breton political capacity 1995–2014: the direction of change.

Indicator	Comments	Direction of change 1995–2014
Institutional resources	Low score comparatively, but robust informal institutions	Gradual strengthening of the regional level domestically (+)
Economic profile	Region in slow decline in national rankings	Breton economic model challenged by some aspects of economic globalisation (−)
IGRs and multi-level governance	IGR model stronger than ever. Strategic Europeanisation challenged by economic crisis	Shift from strategic to normative Europeanisation?(−)
Political leadership	Collective, regional and removed	Stable, based on pursuing the Breton interest by first-rank politicians in Paris (=)
Party system	More sharply divided cleavage structure	Capacity for pursuing the Breton interest remains powerful (=)
Territorial regimes	Region-centric networks	Stable (=)
Identity logics:	Regional identity makes sense culturally, economically and institutionally	Stable (=)
		Even instrumental drivers (e.g. the regional map) are seen through lens of regional identity
Territorial narratives and modes of regulation	Repertoire of lobbying and direct action	Renewed vigour (+)

Of our eight indicators, four are broadly stable (regional collective political leadership, a party system where regional advocacy occurs through state-wide parties, region-centric networks and regional identity logics), two are in progression (the force of regional institutions and territorial narratives), while two appear in retreat (Breton economic capacity (especially in western Brittany) and the weakening of strategic Europeanisation). If the architecture of the Breton model remains recognisably robust, the weakening of the material capacity indicators contains more generic lessons for Europe's regions, especially that family that we label as second-order strong identity regions. In the context of economic crisis—and in contrast to the family of richer, ‘first-order’ strong identity regions such as Flanders and Catalonia—regions such as Brittany (or Andalusia, Wales and Wallonia) use territorial identity claims above all to promote solidarity and fiscal transfers. In these regions, the pursuit of socio-economic interests and preservation of existing welfare traditions and financial transfers place boundaries on using territorial identity to support more autonomous forms of governance. As a more general point, frameworks for understanding territorial governance need to accommodate family resemblances (Collier and Mahon, 1993) and operationalise ‘most similar’ approaches that allow for comparison in categories such as global cities, city-regions, first- and second-order strong identity regions, and peri-urban and rural zones. The territorial political capacity framework proposes a heuristically useful theoretical mix that facilitates the understanding of how actors construct their political actions over time, how accepted ways of operating influence forms of collective action and how actors can continue to be mobilised by existing discourses and practices, even when circumstances are changing.

There are strict limits to any territorial capacity, even for the strongest European regions that are faced with pressures of economic convergence, enhanced central steering and global economic competition. Within these strictly bounded limits, Bretons, via their relationships and modes of operation, have mediated change. The events of Autumn 2013—the crisis, the direct action and the temporary resolution by a promise of central state investment—recalled how effective the instrumental use of identity politics could be in the short term. Bretons revolted and Breton achieved short-term gains. The social movement of Autumn 2013 confirmed key features of the territorial action repertoire, most specifically the efficacy of direct relationships with Paris, the instrumental use of identity, the spectre of disorder, the mobilisation of past symbols of Brittany's specificity, the united front of (some) Breton employers²⁹ and employees in defence of their collective territorial goods. The Breton actors interviewed in 2013 were deeply conscious, however, of the limits of existing modes of collective action and the need to reinvent a new model.

Acknowledgements

This work was supported by the Leverhulme Trust under Grants IN-2012-109 and Rf+G/10711; and by the Economic and Social Research Council under Grants L219252007 and L311253047. The authors would like to thank the Leverhulme Trust and the ESRC for their generous support. Alistair Cole would also like to thank the Collegium de Lyon for providing an excellent ambience for his research leave in 2014.

Notes

1. The analysis for this article is mainly based on 87 semi-structured interviews carried out with political, associative and economic actors in Brittany at three stages in the past two decades: in 1995 (28 interviews), 2001–2002 (30 interviews) and in 2013 (29 interviews). In each of the periods of empirical investigation, three sub-groups were identified: party political actors (the main party families), public administration actors (generally working for the region, the regional state or public agencies) and the policy community (especially representatives of employers, trade unions, business groups and cultural associations). The interviews were taped and transcribed.

2. The CANS 2009 survey interviewed samples of 900 people in 14 regions (Hendersen et al., 2014). The earlier survey (2001) involved a representative sample of 1000 and was funded as part of ESRC project L219252007.

3. The empirical data forms part of the Leverhulme Trust's International Network on ‘Territorial Governance in Western Europe: between Convergence and Capacity’ (IN-2012-109).

4. Interview with the vice president in charge of budgetary affairs, August 2013, an account confirmed in other interviews and formal statistics.

5. Interviews, Regional Prefect, 2010 and Regional General Affairs Secretariat, 2013.

6. B (for Brittany)16 is the generic name given to the regular sectoral meetings between the region, the departments, the main cities and other stakeholders. The precise format varies according to sector. One interviewee described the process of a B16 in youth policy in the Brittany region:

The B16 brings together the main actors involved in youth policy in Brittany, from the region, the main department and the cities, notably. Through the B.16 we share a common methodology for dealing with youth issues–unique in France, though the region does not have formal competence in issues of youth policy.

7. Interview, 2013.

8. The founder of the CELIB, Joseph Martray, interviewed in 2002, emphasised the combination of lobbying pressures with direct action tactics, such as blocking trains, to further the Breton cause.

9. Interview, 2001.

10. Brittany was politically contested insofar as some Breton autonomists had sided with Germany during the war. A small brigade fought alongside the Germans.

11. Interview at the Directorate-General for Regional and Urban Policy, France section, December 2008.

12. Interview, September 2013.

13. The EU announced, in July 2013, that it was withdrawing its export subsidies for poultry producers with immediate effect.

14. One well-placed interlocutor, interviewed in 1995, affirmed: ‘Méhaignerie and Hervé are so unlike that they cannot understand each other. They dislike each other personally and disagree politically.’

15. Interview, 2013.

16. The UMP initially refused to sit around the table with other Breton politicians.

17. In the 2014 European election, the FN list came a close second behind the UMP—but well ahead of the Socialist list. Earlier in 2014, the FN had elected its first municipal councillors in Brittany.

18. Interview, 2001.

19. Interviews, 2001 (Le Drian) and 2002 (de Rohan).

20. Interview, Pierrick Massiot's cabinet, 2013.

21. The movement organised large and mediatised events, such as that of 2 November 2013 in Quimper, where 25,000 ‘red bonnet’ protesters urged the cancellation of the new tax and the granting of more regional power for Brittany.

22. In the Breton survey, 25% expressed either an exclusive (2%) or a predominant (23%) Breton identity, while 50% declared themselves to be ‘as Breton as French’. In Alsace, 18% were exclusively or predominantly regional (42% equally national and regional), while in Ile-de-France only 8% felt exclusively or predominantly regional (30% equally national and regional).

23. The precise questions asked were: What are the main traits of the Breton model? (2001, 2013). Is there a Breton model? Has it remained constant over time? (1995). What is the specificity of the Brittany region? (1995).

24. This question was asked in each of the semi-structured interviews that formed part of our survey.

25. Interview, 2001.

26. Interview (Ronan le Prohon, UDB), 2001.

27. Providing regular scheduled hours for Breton classes was hailed as a major advance in several interviews.

28. Interview, UDB, August 2013.

29. A note of scepticism was sounded in some of the 2013 interviews, especially in relation to Doux, whose owner had reputedly invested EU funds in new factories in Brazil that were now competing with Breton producers.