Structural and Textural Dimensions of Territorial State Organisation

Abstract

In the framework of devolution, British scholars have inquired how to assess the constitutional position of the local government in the UK and its effect for improving or hindering autonomy and democracy at the local level. The purpose of this paper is to discuss the same topic from a more general and comparative perspective, combining conceptual tools of continental public law and political science. To do this the paper proposes to differentiate three concepts of territorial state configuration. These concepts are territorial structure, territorial texture, and vertical diffusion of state authority. This paper asserts that the confusion of these three dimensions has not only impaired an adequate assessment of the constitutional position of local governments, but that the structural dimension has been under-theorised, thus distorting measurements of subnational authority. Moreover, scholars have underestimated structural constraints upon changes in the local government position within the polity. Finally, an operationalisation of the new findings is offered by plotting selected countries in a Cartesian system comparing them from the perspective of territorial structure and vertical authority diffusion.

Key Words:

• Local government
• comparative federalism
• devolution

Acknowledgements

I want to thank the anonymous referees for their insightful comments on the first draft. For help in the editing process I want to thank Mariam Hamouda and Jennifer Cyr.

Notes

 1. Authority is defined in a formal sense to avoid confusing it with its sources (Marks et al. 2008, p. 114) and with power (Hutchcroft 2001, p. 26). Following Marks et al., formal authority is legitimate power exercised in relation to explicit rules usually written in constitutions and legislation.

 2. This definition comes close to the one worked out by Wheare, for whom a structure (as the federal one) must not fully command neither the constitution nor the government, but has to predominate (Wheare 1946, p. 16). For Marks et al., the term ‘structure’ has another meaning, since they refer to the fact that government is ‘far from uniform’ (Marks et al.2008, p. 112). For them government is ‘structured’ in many tiers, while in this paper structure relates to a framework quality of the state. Structure in this sense is based not on the fact that there is more than one territorial level within a polity, but on the fact that, because not all levels have the same status, different structures arise. It also deviates from seeing structure as a ‘process’ as Rodden proposed for defining ‘federalism’ (Rodden 2004, p. 489).

 3. Under state territorialisation the degree of own decisions on politics, public policy, financing and administration is understood, which are formally taken within a polity at a scale different from the national level. If we add to this territorialised authority those decisions taken by the national level through own outposts in the periphery, and those decisions at subnational units upon which the national level has both the power to reverse them and to influence the content of their actions, we come to the entire system of territorial state configuration.

 4. The merging of textural with diffusional factors is in fact usual in the works on measuring decentralisation. Thus, rulemaking modalities (texture) have been undifferentiatedly connected to the concrete range of issues (diffusion) that are affected by them (Bracanti 2006, p. 666, 667); or textural factors have been linked to the effective degree of financial allotment (Schneider 2003, p. 53).

 5. Other authors have tried to combine the political science approach with the public administration approach for dealing with decentralisation (Hutchcroft 2001, p. 26, Schneider 2003, p. 34). For limits of the legal and the political science approach relating to decentralisation see Hutchcroft (2001).

 6. It was on these grounds that Kelsen believed that a totally centralised state was impossible (Kelsen 1993, p. 164). In a comparable sense, Treisman asserts that the fact that polities have normally more than one territorial tier leads to conceiving them as mostly ‘compound’, even if decision making authority remains at the upper level (Treisman 2002, p. 4).

 7. I use the term statics because it is a useful analogy to that part of mechanics that deals with loads on physical systems in static equilibrium. Not all analytical disclosures of territorial tiers have to do with my concept of territorial statics. For instance, Treisman (2002, p. 6) correctly directed his attention to the quantity of tiers, and thus deviated from the common tendency to neglect the analysis of tiers as tiers. However, his effort did not lead to the measurement of structural features, but to the measurement of mostly diffusional aspects (decentralisation). His counting of tiers remains, though useful, ‘arithmetical’ in comparison to an algebraic counting, which reduces the quantities of tiers by considering the structural weights they carry.

 8. The self-rule/shared-rule tenet remains obviously useful, but should not be seen as a sufficient tool. Neither structural weighing of tiers, nor the subnational system of statehood can be adequately addressed with this tenet. Moreover, whatever the concept of ruling, little is gained without clarifying the scope of the various ‘legislation’ modalities. Finally, this tenet is also not able to address the breaking of the monopoly of lawmaking authority. When shared-rule exists, breaking of the monopoly is not necessarily given, because subnational entities might be allowed to veto national legislation, without having own legislation. On the other hand, if self-rule is reduced to secondary legislation, there is no breaking of the monopoly either.

 9. For authors problematising this municipal status see Salmon (2000) and Steytler (2005). Elazar (1997) notices this, but leaves it without drawing consequences for the understanding of territorial structures.

10. See the ruling of the Spanish Constitutional Court STC 32/1982. One important sign of the encapsulation within the national level is the existence of a national law for municipalities, which would be unusual in federal countries where federated entities retain the regulation over their municipalities. However, it should be mentioned that in Spain, statutes of the intermediate entities can also co-legislate on municipal matters relativising the current national encapsulation of municipalities (Ortega 2006).

11. The principle of equipotency does not refer to the principle of vertical separation of powers, since a varying degree of separation can be existent without guaranteeing an equal structural status of levels. Neither does it necessarily refer to the concrete modalities of task allotment, since a high degree of tasks sharing can prevail with fully recognised equipotency.

12. The very term sphere was deliberately chosen to avoid the more hierarchical aspect of the common term ‘level’. See Devenish (1998, p. 105). In spite of being aware of the interesting proposal to differentiate the terms order, level, and sphere in addressing their varying degrees of formal status (Leuprecht & Lazar 2007, p. 4), I will use the term level or tier in this paper without referring to a hierarchical effect.

13. That in comparative terms regional levels may exert shared-rule, while local levels seldom do, is a fact that may differentiate the status of both tiers. However, this difference is not decisive, since regional entities may lack shared-rule, but still vertically segment the polity, and thus may be structurally more powerful than local entities. On the other hand, a local tier may exert shared-rule, without overcoming its encapsulated character.

14. The concept of texture shares theoretical concerns with other approaches. Treisman rightly distinguished decentralisation as a characteristic of a system, from the system itself (Treisman 2002, p. 3). He spoke of a ‘governmental system’ and its ‘parts’, including the judiciary, which has also to accomplish the ‘task of governing’. The concept of territorial texture goes further, because it claims not only that such a system may also exist at the subnational scale, but that its attributes encompass more than what Treisman calls ‘governmental bodies’ (legislative, executive and judiciary), and extends the factor ‘governmental bodies’ to entities like the Auditor General. Thus, the dimension of texture allows to conceptually grasp the quality of statehood at the subnational scale in a more comprehensive way. In a similar way Cameron and Falleti spoke of ‘constitutional features of the constituent subnational units’, or the ‘organisation of subnational powers’ (Cameron & Falleti 2005, p. 245, 258), and analysed the division of power at the subnational scale. Regrettably, they mix texture with structure, and therefore come to the hastened conclusion that the UK is a federation, because it meets their parameter of having division of power at the subnational level. Other scholars have pointed out that having subnational authority over matters such as citizenship or policy reinforces a statehood quality (Hooghe et al.2008a, p. 126). The idea of a reproduction of state features at the subnational scale appeared also in theories on the plurinational state. The point here is, the existence of ‘governmental networks and institutions which make…self government…a feasible option’ (Tierney 2004, p. 33), and the fact that the ‘capacity for self government makes sub-state national society comparable to the host state’ (Tierney 2004, p. 37, my italics).

15. As Brancati asserts, where the subnational level administers decisions made at the higher level of government, there is no decentralisation, not even if legislatures were elected, since the condition for decentralisation is that decision making is made through legislation (Brancati 2006, p. 654).

16. For Schneider, for instance, the term ‘political’ in front of the word ‘decentralisation’ is at best expressed in electoral components (Schneider 2003, p. 40). Treisman chooses the more convincing interpretation and uses the term ‘political’, when it comes to assessing decision making decentralisation, and connects this issue with ‘legislation’ and the mechanisms of policy-making concurrency (Treisman 2002, p. 7), but regrettably without disclosing the many kinds of ‘legislation’. For Hutchcroft the term ‘political’ encompasses both aspects, electoral and decision making (Hutchcroft 2001, p. 34).

17. The control of appropriateness of decisions is my translation of the French term contrôle d'opportunité, which correlates very much to the German term of Zweckmäßigkeitskontrolle. According to French administrative law, control of legality checks the compatibility of a state action in relation to the prevailing juridical order, while the control of appropriateness makes a judgement in political, technical or even moral terms of that action (Dupuis et al.2007, p. 35).

18. This must not be the only way. For instance, German municipalities, which are only under control of legality in their self-governmental domain, are overseen not by the courts, but by state officials of the executive branch of the respective federated unit. Given the German development of its state bureaucracy this oversight can be deemed to be technical, and in so far not less objective than a control through courts, but this may change in countries with strongly politicised state administration.

19. So a ruling of the Spanish Constitutional Court STC 37/1981.

20. This more legal approach correlates to Kelsen's theory of ‘complete decentralization’ (Kelsen 1993, p. 178) as the case when no reversion of decisions (‘definitive’) and no policy preforming from above holds (‘independent’).

21. The general-purpose quality is pointed out in Hooghe et al. (2008a, p. 124) and Treisman (2002, p. 6).

22. See the ruling of the Spanish Constitutional Court STC 4/1991.

23. One good example for quality analysis of the second textural factor has been made by Treisman by differentiating ‘appointment decentralisation’ and ‘electoral decentralisation’ and showing their relationship to democracy (Treisman 2002, p. 11). For Treisman these are ‘decentralisation’ dimensions, while in this paper, not even full subnational elections and appointments would tell us whether decentralisation necessarily exists.

24. Hutchcroft has been one of the scholars that have rightly underscored the analysis of tiers as tiers (Hutchcroft 2001, p. 30). While he addresses this issue as a feature of decentralisation and not of structure, his analysis of different decentralisation regimens, which can be applied to only one level ‘at the expense’ of the other levels of the polity, comes close to my stratification theory derived from structural statics within the simple mode of state.

25. The assessment of the extent to which this distortion holds is not easy, since the higher score for France in this example might be due to the methodology of Hooghes et al. (2008a, p. 139) that values the quantity of regional levels within a country. Already the question whether French departments should be taken as regions as Hooghes et al. (2008c, p. 192) do or as local entities (I would follow those that assert this), entails enough material that could alter the data before discussing the distortion related to policy-making and rulemaking modalities.

26. For the differences of Scottish Law and Westminster Legislation in respect to their risk of being challenged in the courts, see Page and Batey (2002, p. 517). For more on the constitutional weakness of devolved regions, see Trench (2007, p. 12, 15).

27. Bailey and Elliot explain this special position out of the direct election of local authorities (Bailey & Elliot 2009, p. 441).

28. Notwithstanding the varying shape of local government performance in each devolved region, it seems that the improvements seem to stem more from an increased capacity of local governments to influence policy-making in their respective devolved centers (Jeffery 2006, p. 63), than from changes in their autonomy capacity or constitutional status.

29. The common term of ‘union state’ used for describing the UK's territorial development is compatible with both the simple and composite structure, because it refers, as Neil Walter asserted, to a ‘particular model of territorial pluralism’ (cited in Tierney 2004, p. 13), not to the territorial structure as defined here.