A tragedy of errors ? Institutional dynamics and land tenure in Finnmark , Norway 1

Reindeer herding in Finnmark has been widely perceived during the last few decades as a perfect example of the tragedy of the commons. The present article claims that this discourse relies on flawed assumption regarding land tenure. Our historical analysis of the term ‘common’ in relation to resources in Finnmark shows the term to reflect a misunderstanding of local categories, practices, and concerns related to pastures, territories, and natural resources more generally. In this sense, it exposes a case of ‘mistaken identity’ between the formal legal conception of ‘commons’ and the customary rules and thinking of reindeer herders. We turn to different strands of critical institutionalism to analyse the processes of institutional change that have allowed these errors and misunderstandings to be formalised and naturalised in the current governance system. We show that a process of institutional bargaining between the Norwegian Parliament, the Sámi Parliament, and the International Labour Organisation has recently re-enforced an alien conception of a ‘commons’ to which ambiguous groups of people have equivocal rights. In parallel, a process of institutional layering of new regulatory actors and rules on top of existing ones has taken place. This regulatory ratcheting has resulted in the blurring of the authorities and jurisdictions intrinsic in the customary tenure system. Moreover, the new layers of regulations have actively overemphasized the Sámi customary obligation of sharing resources to legitimize 1 The “Results” section of the present article is developed from our book chapter “Er Finnmarksvidda en allmenning?” forthcoming in Norwegian in Benjaminsen et al. (2015). 2 Andrei Marin and Ivar Bjørklund the new, ambiguous, conception of commons. This process is explained as one of institutional bricolage based on naturalisation by analogy and authority processes that allow certain powerful actors to influence the production of institutional arrangements favourable to them. All three processes underline the negotiated, dynamic nature of institutional change. We propose this integrative analysis of institutional and general social dynamics is beneficial in studying commons as everyday practices affecting natural resource governance.


Introduction
Finnmark is the northernmost, largest and least populated county of Norway, where the indigenous Sámi population have been practicing mobile reindeer husbandry for hundreds of years.The governance of natural resources in this area has for a long time reflected complex interactions between local and central institutions, often with conflictual undertones.
A widespread idea that has influenced resource governance and public perception about reindeer herding in Finnmark is the 'tragedy of the commons'.This idea refers to Hardin's (1968) argument that whenever pasture resources are commonly owned herders will act in order to maximize profits at the expense of all the other herders.
In Finnmark, Hardin's scenario has been employed implicitly and explicitly in policymaking and governmental management, and produced institutions hinged on the general understanding that much of the territory in Finnmark is held under a common property regime.In other pastoralist systems, similar debates have taken place during the past decades.In reindeer pastoralist systems in Sweden, Finland and Russia pasture degradation due to overgrazing is still being debated, and there is an emerging understanding that these areas are in fact cultural landscapes created by the practice of pastoralism (Torp 1999;Forbes et al. 2006;Kryazhimskii et al. 2011).In other environments (e.g.Leach and Mearns 1996;Fratkin and Mearns 2003) these debates have led to detrimental institutional changes to local governance systems.
In this article, we investigate the proposition that large parts of Finnmark are a 'commons' (see Figure 1).We argue that this assumption is misleading and deriving from a misunderstanding with historical roots.Furthermore, we are concerned with explaining how and why, despite this obvious case of 'mistaken identity', the conception of 'commons' proposed by the Norwegian State has been naturalized and institutionalised in the current governance system.
We begin with a historical investigation of the emergence of the concept of 'commons' in relation to reindeer herding.We then turn to different theoretical strands of critical institutionalism to identify processes of institutional change that have had bearing on the emergence of the present resource governance system.We conclude with some insights relevant for more equitable policy arrangements.

Methods
We employ a case-study contextual analysis of the current governance of reindeer herding in Finnmark, with particular emphasis on institutional arrangements that influence land tenure.This qualitative approach relies chiefly on data derived from legal documents, maps, and official records documenting policy-making processes and interactions between national and international institutions.We rely on primary data we acquired through interviews and participant observations during the past 30 years to illustrate briefly how customary institutions are enacted and constantly reshaped on the ground.Our methodology focuses on chains of variables that affect the processes of institutional change -an approach common in historical institutionalism (e.g.Pierson and Skocpol 2002;Hall 2010).

Figure 1: Current administrative division of Finnmark's territory between seasonal reindeer ranges (A) and customary division of winter ranges between siidas of Western Finnmark in the 1960s (according to Paine 1994) (B).
Andrei Marin and Ivar Bjørklund

Tracking the 'commons' in Finnmark
In Norwegian, the term 'commons' is usually translated by 'allmenning' (as in 'en allmenning '-a commons) in reference to physical land entities, "circumscribed by true property boundaries" (Sevatdal 1985, 1).The terms has three variations: 'statsallmenning' or State commons, 'bygdeallmenning' or parish commons, and 'realsameier' or farm commons (Grimstad and Sevatdal 2007).The oldest references to rights of commons go back to the Viking age, the so-called Gulating Law (1164) and Frostating Law (1220) which were formal, written formulations of the existing customary practices of local communities.
In Finnmark, the first documentations of such rights date to the 18th century.Prior to 1751, the inner part of Finnmark (most of its territory) was an area shared between Denmark-Norway, Sweden, and Russia.Interestingly, the terminology used to refer to the 'common land' did not use the term 'allmenning', but 'common' (Norwegian: 'felles').There are several maps from the time (e.g. the so-called Knoff map from 1749) showing demarcated areas denoted as "Norwegian and Russian Common District" or "Norwegian, Swedish, and Russian Common Lapps".'Common' in this context referred to areas common to several states (the States did the 'owning', not the local communities, or individuals2 ).In 1751, Denmark-Norway and Sweden signed an agreement regarding a final border between them (The Lapp Codicil) leading to the division of 'common' districts between the countries and to the express provision (in §2) that the conception of 'common' ('faelles') was to be abandoned.
Nevertheless, we have clear indications that at the time there existed welldefined territories to which small groups of Sámi users held some form of collective rights.The Lapp Codicil gives the names of particular territorial units ('lapmarker'), which were 'owned' or 'had' by certain groups and therefore prone to taxation.The unit of taxation during the 17th and 18th century were collective units such as 'hearth' (rök) and 'tax land' (skatteland) (Hansen 2009).Land register records in Finnmark from the period 1620-1770 also suggest exclusive, collective rights to use land and water resources connected to local groups (bygdelag) (Kristensen 2001).
The existence of socio-territorial units named siida that managed natural resources has been documented in written sources since around 1550.These are Swedish taxation lists for the Sámi living along the coast and the names are organized according to the different villages (byer) to which they belonged (Vorren 1978).Despite having clearly delimited borders, at least some of these territorial units had overlapping areas for 'common hunting', which could also be used by siidas hundreds of kilometres away (Vorren 1978, 3).This way of accepting common access to certain resources like game, was also reflected in the coastal Sámi's practice of defining some fish species as common prey (coalfish, herring), while other (cod, salmon) were considered private in terms of access to established fishing spots (Bjørklund 1991, 43).The above-mentioned documentation done in general by the Norwegian State thus implies that at least until the 18th century there were in Finnmark local rules of use, rights and obligations connected to resources, that simultaneously delimited exclusive access to resources and suspended this exclusivity in certain circumstances.
When reindeer pastoralism slowly became common among the Sami during the 18th century (Bjørklund 2013), siida became synonymous with -" a particular group of people who camp, work and migrate together and also the reindeer owned and herded by these people" (Pehrson 1964, 4).From the middle of the 19th century, an increasing number of farmers colonising the northern coastal areas of Troms and Finnmark came into conflict with herders using summer pastures on the coast.This situation led the Norwegian State to decide in 1883 through the 'Common Lapps Law' to formalize the locations in which specific herders had rights to use resources.Subsequently, the Lapp Commission of 1897 attempted to provide rules for this division between areas "where old traditional grazing can be proven, or not, in other words between legal and illegal grazing" (NOU 2001, 197).
Nevertheless, it was not until the 1933 Law on Reindeer Herding that this division was formalized in Finnmark.The law stipulated that (in §2) the county should be divided into parishes (reinsogn) and districts (reinbeitedistrikter), and that (in §4) the county governor can decide the borders between the spring, summer, autumn and winter grazing areas within each district and the grazing schedule for each area.The law also suggested to further divide the areas within each district between different groups (implicitly siidas).This suggestion got the support of some parish bailiffs, but the County Council opposed it as "inappropriate and impractical" (NOU 2001, 219).This situation illustrates the bargaining process involved in new institutional arrangements, as well as the tendency to recognize resource rights only at aggregate administrative levels and not in detail, an indication of the idea of a blanket term 'common' to cover resource rights.
The 1978 Law on Reindeer Herding formalised and legally codified for the first time the idea of common districts and common pastures.In Chp. 1, §2 (3) of the Law it is stated that The National Reindeer Herding Board can "undertake division into districts of the traditional common spring, autumn and winter pastures in Finnmark, where utilization has been according to customary patterns.There can also be determined grazing times, weight limits for reindeer and/or maximum number of reindeer for the new districts" (emphasis added).Not much is made of the formulation "common" (felles) since it is never mentioned again in the text but the consequences of this provision have been rather significant.
It was the first time a legal text implied that any existing divisions between groups were superseded, indeed that the 'traditional' way of managing these territories was some sort of collective (felles) use that gave equal rights to individuals by virtue of their simple membership to an undefined collective.' Solem (1970) admiringly describes the way Sámi reindeer herders of the 1930s organized their resource use, emphasizing their respect for the customary rules governing this use.He underlines that individual siidas have clearly delimited areas for winter pastures (190) and that the inherited right to use the territory of each siida, is not connected to any person but to the group.Nevertheless, such arrangements do not entail a sort of private or group property to these pastures, or the right to 'keep others away', whereas the incoming herders, on their side, know they should not enter another group's areas unless forced by necessity (ibid).

How herders conceive and practise the 'commons
This seems to be a situation similar to parish commons elsewhere in Norway.There, ground can be held 'in common', whereas resources (e.g.pastures, hunting, fishing) can be held 'jointly'.'In common' entails individuals having rights that are fractions of the total commons, and the rights will devolve on the owner's descendants; whereas 'jointly' means individuals get rights by virtue of their belonging to the collective, the felles-and these rights will not devolve on her personal descendants, but rather on her co-owners (Berge 2007).It would appear that it is only in the sense of an individual siida's rights to pastures or other specific resources being held jointly that the parish commons concept can be applicable in Finnmark.
Nevertheless, reindeer herders seem to have a much more complex view of rights and obligations connected to resource management than the dichotomy outlined above allows for.Not being allowed to keep others away seems contingent on emergency situations that render borders 'permeable'.But, the obligation to share, practised as intra-and inter-community mutual aid and reciprocation, seems to be a clear cultural pattern reflected in land tenure institutions that emphasize flexibility and negotiation (Hågvar 1989;Bjørklund 1990).These two qualities are intrinsic to the pastoral siida institution.The siida refers to a group of reindeer owners who work and migrate together and to the herd of reindeer owned and herded by them.Because the herds differ in size throughout the year due to shifting pasture conditions, so does also the demand for herding tasks.Accordingly, the siida changes size and composition through the year, as the herders have to divide and regroup their herds.The siida is in other words "an alliance recruited through cognatic and affinal kinship relations based upon mutual herding strategies among its members" (Bjørklund 1990, 81).This principle of organization provides each herder with potential access to pasture and herding partners over a large area.Such dynamics presuppose both continuous negotiations and flexibility among the participants.Emergency situations (e.g.pastures rendered inaccessible due to snow or ice), which call for temporarily using a neighbouring siida's territory, are not codified in terms of formal rules, but rather open to negotiation and interpretation.Vorren (1962) describes how in the 1950s areas were used in common by several winter siidas routinely, not as a result of difficult circumstances.Yet, these areas seem to have been mainly adjacent to migration routes.In a recent interview (July 20123 ) with practicing herders, they described a similar use pattern in an area officially defined as part of the 'common' winter pastures: "This is a migration route.Many siidas use this area during migration [between winter and summer pastures].And there are also some [one siida] which are here all winter.(…) This is also a migration road for animals taken to slaughter".
Herders have always conceived their traditional siida area as clearly delimited.Paine has documented how in the 1960s winter siidas had clear borders that overlapped (see Paine 1994 and figure 1).Our interviews with siida leaders in 2002 confirmed this: "The traditional borders are stable throughout the winter.They don't fluctuate as long as there are good pastures; just if there are bad pastures one moves maybe to a place where a neighbour should have grazed (his animals)".
Herders also have a keen sense of when an act represents 'trespassing' on somebody's resources.According to complex interpretations of the current context (of pasture resources, weather, herds, and labour force available) such acts can be judged merely as something 'baha' (bad) or as 'haeppat' (shameful) (Paine 1994, 83;Paine 2009, 120).This is confirmed by herders nowadays.One siida leader reflected in 2002 that "I have lost access to some of my winter range, and neighbours use it now; more than half of it.We have the least reindeer there so I guess it's ok (enough) for us for today (…)".Other herders (interviewed in 2012) reflected in reference to a particular area of the 'commons' particularly attractive due to snow-free patches of accessible vegetation: "This is spring-and winter-pasture for three siidas.We are right on the border area between them.They compete for barren (snow-free) patches".
Nevertheless, the impacts of the imposition of the alien concept of an indiscriminate 'common' are also evident in contemporary practice.The siida leader quote above reflected (2002) that he would not have allowed such a large portion of his siida territory to be used by others but that this was possible only because of the institutionalisation of the 'common' concept: "Yes, the customary borders are set.It was later, with the concept of 'common pasture' that the traditional borders were no longer important, then it was 'common', there was a possibility (to ignore them)".
Such details reinforce Solem's observation (1970, 191) that herders keep a highly cooperative attitude because it is practically important and because every individual's reputation is a powerful asset that allows them potential access to other siidas.Such negotiation is at least equally concerned with circumstantial access as it is with fixed rights to a piece of the landscape.
The herders' conception of appropriate resource governance thus underlines two major elements in apparent contradiction: well-defined exclusive rights and possibility to suspend and/or negotiate these rights.The well-defined rights are connected to specific siidas and reflect Tønnesen's position (1979, 312-313) that rather than speaking about one Finnmark commons (Finnmarksallmenningen), the region should more accurately be referred to as a collection of many commons (the siidas).The flexible and negotiated nature of resource management indicates the need to understand siidas also in political terms, as units whose main purposes are legitimate authority and jurisdiction rather than formalized exclusive property rights.This illustrates why treating three of Finnmark's four regions of seasonal pastures as three commons is historically inaccurate and misleading.In addition, it also prompts the more interesting question of what are the institutional dynamics that have allowed for the perpetuation of the use of a misguided conception of 'commons' and the related idea of a 'tragedy of the commons'.

Discussion: How institutional dynamics shape land tenure
Starting from the situation of 'mistaken identity' presented so far, this section illustrates three processes through which the current land tenure arrangements in Finnmark have come about, and the institutional dynamics that underlie them.These processes (institutional bargaining, layering, and bricolage) have previously been analysed in separate strands of critical institutionalism that have different, if complementary, analytical strengths and theoretical ambitions.They are usually employed to explain institutional dynamics at respectively the international, national or subnational level.An analysis that integrates them is fraught with ontological and epistemological tensions.The main ontological tension we see is between a rational-choice inclination of institutional bargaining and the sociological/cultural understandings espoused by institutional layering and institutional bricolage.On the epistemological side, the three processes have previously been analysed under different assumptions about power, agency, and models of institutional change. 4We nevertheless, choose this integrative approach as an empirically-driven research strategy and argue with Hall (2010) that these tensions are not insurmountable and that the integration is beneficial.

Institutional bargaining
Institutions for governing natural resources and the environment often rely on an 'interactive process involving the interplay of distinguishable forces' (Young 1994, 82) termed institutional bargaining.In Young's acceptance this process refers to the emergence of mutually accepted arrangements (ibid).'Institutional' in this sense refers equally to the production of new institutions as to the bargaining that takes place between already existing institutions.It is therefore not necessarily the institutions that do the bargaining, but the results of it are new institutions or major adjustments of old ones.
The 2005 Finnmark Land Act is an interesting example of institutional bargaining leading to the development of a formal land tenure institution.This process of institutional change originates in the so-called Alta Conflict, which began in the 1970s.The planned building of a hydro-electric power plant on the Alta river in Finnmark entailed the flooding of key pasture resources used by reindeer herds (Bjørklund and Brantenberg 1981).Local residents and Sámi herders protested for several years.The conflict ended with the construction of the dam and power plant but also with the establishment of the Sámi Rights Committee and the subsequent establishment in 1989 of the Sámi Parliament.The Sámi Rights Committee's work resulted in 1997 in a report, which the Government drew heavily on in their preparation of a new legal bill dealing with land rights in Finnmark.This bill for a new act was proposed to the Norwegian Parliament in April 2003.The reactions to the bill were overwhelmingly negative from the Sámi Parliament and most Sámi organizations.Some of the criticisms referred to the procedures leading to the proposed bill (Broderstad 2008), but more importantly, they protested against the fact that the land and resources in Finnmark were still to be under the ownership of the State -arguing that this was not in accordance with international human rights (Josefsen 2011).Young's (1994) analysis of bargaining processes underscores that transnational alliances between local and international interest groups can play an important role (104).This was also the case with the emergence of the Finnmark Land Act, where the International Labour Organisation (ILO) played a significant role.
In 1990 Norway was the first country to ratify the ILO Convention 169 (The Indigenous and Tribal People Convention).The convention is legally binding for states that have ratified it, who have to ensure its implementation in good faith and its effectiveness (ILO 2007).The Finnmark Land Act established land rights for an indigenous group, and thus had to conform with the ILO 169 Convention by partly incorporating it and giving it priority in case of conflict (Smith 2011).
The Act was prepared based on interactive bargaining ('consultation') between the Sámi Parliament and the Standing Committee on Justice of the Norwegian Parliament.The first proposal of the bill (in 2003) was rejected unanimously by the Sámi Parliament whereas following the 'consultations' the vote was unanimously in favour of the new version of the bill (Smith 2011).The crux of the initial rejection was the fact that the bill left land and other resources' ownership in the hands of the State, despite transferring management authority to the parastatal Finnmark Estate (Finnmarkseiendommen).Such an arrangement (despite giving equal representation to the County Council and the Sámi Parliament in the board of the Finnmark Estate) in effect gave equal rights to land and resources to all Norwegian citizens, as well as all European Union residents (Josefsen 2011).This was indeed an extreme version of Finnmark as 'commons'.
The initial proposal was amended to give, in the final version of the bill, not only management authority but ownership rights to the Finnmark Estate.
In addition, the final text gives precedence to the land use rights of Finnmark residents (Sámi as well as non-Sámi) over the rights of other Norwegian citizens, and acknowledges that the rights of Finnmark's resident derive from customary or ancestral use.
This was unanimously approved by the Sámi Parliament in 2005.The 'consultation' between the Sámi Parliament, the Finnmark County Council, and the Norwegian Parliament's standing Committee on Justice represents the national side of the institutional bargaining process.The aim of the consultations seemed to be what Young and others conceive of as a typical aim of institutional bargaining, namely 'mutually acceptable arrangements'.The process of bargaining among self-interested parties aims at devising arrangements acceptable to as many actors as possible, that is it aims at achieving consensus rather than a majority through a form of integrative (or productive), rather than distributive (or positional) bargaining (Walton and McKersie 1965;Cross 1969;Young 1994).Despite the obvious consensus between the Sámi Parliament and the Norwegian Parliament, the process did not allow for (or aimed at) achieving consensus with the different interests of the larger Sámi community.The four rounds of consultations that took place between the two drafts of the bill were neither traditional hearings of grievances, nor negotiations over disputed issues and were not open to the public access (Josefsen 2011).In effect, this form of institutional bargaining achieved what Young (1994, 106) refers to as the development of winning coalitions by excluding parties deemed likely to dissent.In this case, one such party was the reindeer herding community, who has always argued for the need for more specific tenure arrangements, rather than the indiscriminate 'commons' form that the Finnmark Act perpetuates.
The minutes of the consultations between the Sámi Parliament and the Justice Committee illustrate that the ILO was an international third party in the bargaining process.The documentation of the consultation process (e.g.Arbeidsdokument 2, 2004;Arbeidsdokument 6, 2005) clearly illustrates that many of the arguments for reformulation of the initial proposal were made in reference to specific articles of the ILO 169 Convention.
Communication between ILO's Committee of Experts on the Application of Conventions and Recommendations (CEACR) and the Norwegian State at the time clearly states ILO's concerns.According to the ILO, the initial proposal (giving merely usufruct rights to local people, and not ownership) was in violation of Articles 14 and 15, protecting rights of ownership to land, of primacy to resource use, and the right to participate in and influence management decisions regarding such resources.The alternative solution ILO emphasized was an institutional regime that ensures the rights of locals (Sámi and the non-Sámi) and must 'be fair, and perceived as fair' (CEACR 2004).
Nevertheless, ILO's influence in the bargaining process has been limited.Although the text of the Act opens for the possibility of detailed identification of (user) rights to resources for siidas, it does not provide any rules that could be used by local communities to influence resource management in their areas (Ravna 2014).Indeed, although the process of investigating individual siidas' tenure rights has started, the process is still very slow (expected to last at least until 2025 -Sara 2011).In addition, legal principles to decide on particular claims to specific pastures or migration routes are not specified in the Act, and competence is clearly lacking in the current institutional arrangements (ibid).
The case of institutional bargaining illustrated herein should nevertheless not be understood only in terms of the three main institutions involved in the bargaining (the Norwegian Parliament, the Sámi Parliament and the ILO).The general socio-political context in which this process of institutional change took place holds much explanatory power.As Josefsen argues (2011), this process was facilitated by the results of the 2001 election, which lead to a centre-right minority government and placed the Labour Party in opposition.This allowed Labour to bring about change by aligning with the interests of the representatives of the Labour Party in the Sámi Parliament.An additional element of the process was likely a political culture of Labour in support of the principles of the International Labour Organisation.
The importance of the larger context is therefore paramount to understanding institutional bargaining, something Young (1994, 104, 106) emphasizes as 'linkages' between bargaining around the issues at hand and other events occurring in the socioeconomic and political environment.This suggests that institutional bargaining can be better understood by exploring the complex social, political and cultural institutional contexts in which it takes place.This analysis, we argue, should nevertheless not overemphasize the structural nature of institutions (culturally ascribed positions that allow particular conducts), but pay equal attention to agency (personal influence) in political processes (e.g.Corson et al. 2014).

Institutional layering
Another process relevant to the analysis presented herein is institutional layering.Thelen (2003) and others (Streeck and Thelen 2005; Mahoney and Thelen 2010) conceive institutional layering as a process of subtle and gradual, slow and piecemeal institutional change by which new elements are attached to existing institutions resulting in the modification of "the ways they structure behaviour" (Mahoney and Thelen 2010, 16).This explanation is proposed as an important counterpoint to theories conceiving institutional change in terms of long-term equilibrium punctuated by short periods of disruption and reform due to shocks exogenous to the system (e.g.path-dependence, or rational-choice models).Institutional layering produces change in existing institutions by the partial renegotiation of some of their elements, while leaving others in place, or through the "active sponsorship of amendments, additions, or revisions" (Streeck and Thelen 2005, 24).The addition of new layers of actors or regulations to existing institutions often leads to the blurring of authorities and jurisdictions (van der Heijden 2011).
Institutional theorists have recognised this dynamic in reference to resource rights for some time.Bromley (1985, 790), for instance, describes 'institutional dissonance' as the result of layering of incongruent institutional structures (rules) and interests.
Similarly, Ostrom (2000, 337) proposes, that when resources previously controlled by local people are nationalised, local users' institutional arrangements lose their legal standing, turning for instance "a de facto common property regime enforced by local users (…) to a de jure government-property regime, but reverted to a de facto open-access regime" (original emphases).Nevertheless, these earlier models of change fail to explicate the process through which local institutions lose 'legal standing' and local people shift loyalty.Secondly, these approaches say little about the importance of institutional legitimacy (or 'social sanction' in Bromley's terms), implicitly dismissing it as inconsequential by the focus on rational choice arguments (e.g. in the situation presented by Ostrom change was explained as a result of state control being ineffective due to the lack of funds and personnel).By contrast, institutional layering explains similar changes by proposing that whenever new rules are introduced on top of, or alongside, old ones they can alter the logic of the original institution or compromise its stable reproduction (Mahoney and Thelen 2010).
In Finnmark, herders nowadays have to engage with an impressive array of new formal bureaucratic institutions that come on top of customary ones.They have to engage with municipalities and/or county councils in relation to migration routes (reindeer herds crossing roads may require roads to be closed temporarily), migration schedules, 5 or interact with municipalities to settle conflicts due to damage reindeer may do to the property of local settled population.Their housing situation and local whereabouts are a matter for at least two different municipalities, as they have one house in the inland and one on the summer pasture by the coast.They will have to relate to the 'Reindeer Police' -a special unit of the police force with responsibility of protecting reindeer herding interests, but de facto also the interests of local farmers.Issues pertaining to animal welfare are formalised in national legislation and are supervised by national or regional health boards (Dyrehelsetilsynet) and animal welfare tribunals (dyrevelferdsnemnda).These institutions have attributions that herders have to take into consideration when transporting, marking or slaughtering their animals.They have to submit to the growing influence of the Food Safety Authority (Mattilsynet) enforcing numerous regulations for hygiene and animal health that have a strong bearing upon the daily husbandry and herding of the reindeer owners.Furthermore, they must comply with the County Governor (Fylkesmann), who regulates all inland fishing and hunting, the latter becoming of great importance, as predators today are a great problem for the reindeer herders.Some analysts (Thelen 2003;Streeck and Thelen 2005) suggest that such layering of rules and actors often effect change toward a hegemony of the more formal institutions.Others have demonstrated that, particularly in the case of land tenure institutions, this shift toward the formal bureaucratic forms is facilitated when locals solicit formal land deeds, conferring formal institutions the source of legitimacy needed to replace the informal conceptions of land rights or tenure (Sikor and Lund 2009).In Finnmark too, it appears that the formalization of the large scale, unspecified 'common' is gradually replacing the logic of Sámi tenure institutions.This is also how we can interpret herders' anxiety with regard to the new Finnmark Land Act's influence on the land rights.They fear that the act will undermine their customary management practices and ideas of justice.
Although it is clear by now that the Finnmark Act is the result of a political compromise (Hernes and Oskal 2008;Ravna 2014) based on bargaining, it also illustrates how the addition of regulations and actors can lead to changes in meaning.The Act strengthens the alien conception of a 'commons' to which ambiguous groups of people have equivocal rights.This comes on top of customary rules that regulate migration routes in specified intervals and certain resting areas for the herd for limited periods.These rules make it possible to avoid both the mixing of herds, the degradation of pastures by trampling, and give access to alternative pastures if your own were inaccessible due to climate conditions (Bjørklund and Brantenberg 1981;Marin 2006).They place responsibility on the herders to avoid a harmful impact upon the ranges of another herder.Failure to do so is always regarded as a conscious act, but it is interpreted in a complex web of social and environmental circumstances, leading to a chain of strategic decisions to counteract the cause (meetings among herders, change of migration schedules, retaliation through raiding, etc.) (Paine 1994).
The gradual changes of meaning as a result of layering has also been proposed in relation to property rights institutions (Mahoney and Thelen 2010).Some propose that such changes derive from the intrinsic ambiguity of institutions, allowing the meaning of institutions to always be open to interpretation, and 'what an institution makes possible or not is continuously redefined by what might be called interpretive entrepreneurs' (Streeck and Yakamura 2003, 40).This view is similar to that of Streeck and Thelen (2005) who propose that institutions represent strategic behaviour, rather than the shared cognitive templates that Douglas (1987) advocates.Still, they disagree with the voluntaristic (design-centred) view of institutions and maintain that institutions are relations -of authority, obligation and enforcement -whose stability depends on publicly derived legitimacy.This conception of institutions underlines the social processes by which institutions are translated into behaviour, which are ultimately struggles over meaning (rather than either struggles over getting the right design, or waiting for 'shared cognitive scripts' to change) (Streeck and Thelen 2005, 11).
In Norway, the layering through the 'regulatory ratchet' (van der Heijden 2011, 14) of more actors and more regulations, can be traced back to the Reindeer Act of 1978.This law -together with an encompassing subsidy system -must be understood within the context of the prevailing social democratic ideology and governmental efforts to include the reindeer herding Sámi in the Norwegian welfare state.Previously, the pastoral production system had been rather autonomous, governmental institutions controlled neither recruitment nor herding or husbandry strategies.The 1978 policy was based upon economic and biological ideas of how to make the meat production more effective and thus increase the income for the reindeer herding Sámi.To be able to do this, a wide array of new regulations were introduced.The outcome could be described as the final political and economic integration of the Sámi pastoral society (Bjørklund 2004).A hierarchical administrative system was introduced and the herders now found themselves within a tight corporative system defining their main interests as being meat producers.From now on, the State would limit the recruitment to reindeer herding by means of a mandatory license.The decision was taken away from the reindeer herding society along with their right to decide how the ranges should be used and the herd be managed.
Thus, the traditional logic of resource management is in the process of being redefined and the State has partly succeeded in introducing their new conception of 'commons' as a legitimate institution.Still, this situation of layering does not explain how and why the conception of land rights can be changed.The explanation we propose is that the new actors are not equal and that institutions in general provide a certain amount of 'play' for institutional change in the process of normal, everyday implementation and enactment (Streeck and Thelen 2005).This can be better explained in terms of institutional bricolage.

Institutional bricolage
The concept of institutional bricolage proposes that 'mechanisms for resource management and collective action are borrowed and constructed from existing institutions, styles of thinking and sanctioned social relationships' (Cleaver 2002, 16).In the social sciences, the concept of bricolage emerges from Lévi-Strauss' (1962) usage of the term in reference to an unexpected movement, or to novel, inventive ways of using available materials.Intellectual bricolage is a similar process by which people (individuals, societies) "take to pieces and reconstruct sets of events (…) and use them as so many indestructible pieces for structural patterns" (22).
Drawing on Lévi Strauss' conception of bricolage, Mary Douglas (1973Douglas ( , 1987) applies it to institutional analysis.In her view, the socially sanctioned thought formulae are used to construct patterns of behaviour, or institutions.Institutions thus formed offer easier classification ('do the thinking') of events and criteria for ascertaining legitimacy of actions.More importantly, these thought and behaviour patterns manifest 'institutional leakage' by metaphorically connecting with new situations, and allowing "meaning to leak from one context to another along the formal similarities that they show" (Douglas 1973, 13).
Still, in Douglas' conception, the institutions (styles of thinking, patterns of social interaction) are most often applied involuntarily, even unconsciously, and are very hard to change.Cleaver (2002Cleaver ( , 2012)), on the other hand, has convincingly demonstrated that such social patterns are subject to the agency of individuals and they are constantly (if subtly) shaped and reshaped in the process of everyday usage of the institutions.
It is this emphasis on agency and its related concepts of power and authority, that gives, in our view, the concept of institutional bricolage more explanatory power than institutional bargaining and institutional layering.
The institutional approaches outlined in the two previous sections, useful as they are to our analysis, fall short on explaining why some of the new regulations imposed on the Sámi reindeer herders by the local and central bureaucracies (e.g.slaughtering quotas, the use of undefined 'commons') have apparently been followed, at least by some herders.Those who do follow them can be said to benefit from the subsidies, but it is not obvious that the only motivation is financial (e.g.Ulvevadet 2000;Marin 2006).This apparent compliance tells us little about the effect of layering of regulations.It may be an indication that the new layers of rules have managed to infiltrate Sami herders' patterns of thinking and practicing reindeer herding.Yet, we propose that this is a strategic behaviour of bricolage on the part of some of the herders drawing on two sets of 'accepted behaviours' (often in opposition to each other): One couched in the conception of the 'commons' introduced by the State administration, the other embedded in the Sámi culture and practice of herding.
There are two mechanisms at work in this process of bricolage. 6On the one hand, when herders employ the formal state-derived conception of 'commons' actively and explicitly in their daily interactions, they reinforce its legitimacy as a valid way of thinking.It is therefore a legitimizing practice with constitutive effects in social practice (cf.Sikor and Lund 2009).On the other, this reference to pastures as 'commons' may reflect a shortcut way of accessing resources, rather than a genuine recognition of the validity of this conception (although, involuntarily, it may have this effect).
We propose that the partial acceptance of the tern 'commons' reflects what Cleaver (2012) refers to as "naturalisation by analogy" (48).The new arrangements derived from bricolage, gain legitimacy if they manage to relate to the "broader views of (…) the world and the proper place of people within such systems" (ibid).The State-backed idea of a county-encompassing 'commons' with very broad physical and social borders has become 'naturalised' as an accepted style of thinking among some of the reindeer herders because it partly resonates with the Sámi conception of sharing of resources, as illustrated earlier in the article.
Still, as we have shown above, there are fundamental differences between commons thus defined and what reindeer herders consider proper behaviour on the ranges, the rights and obligations related to resources, and the sources of legitimacy of these.Thus, for the herders the idea of the obligation to share is combined with the right to access and use resources by people in need, but also with the concept of 'first rights' (Sámi: vuostasriekti) that well-defined entities (siidas) have (Hågvar 1989, 143).We see herein an "active sponsorship" typical of institutional change through layering (Streeck and Thelen 2005, 24): the idea of sharing is emphasized while the idea of first rights is down-played.This results in the subtle naturalization of the 'sharing' principle into the formal new regulations, while removing the complex circumstantial calculations connected to sharing.
In addition to the mechanism of naturalisation, institutional bricolage relies also on authoritative processes.Thus, certain elements from the repertoire of 'proper principles' are emphasized by actors with power and interests to derive institutional arrangements favouring them.In Finnmark, declaring the springautumn pastures as commons, has in fact allowed herders that spend summer on pastures bordering the spring-autumn pastures to use resources in ways that have been detrimental to other herders (out of season grazing), and in opposition to earlier practices (Riseth 2000).
The process of granting authority to certain ways of doing things is thus connected to the "unequal exercise of power and the capture of its benefits" (Cleaver 2012, 51).Nevertheless, such processes of authority formation can be challenged during the process of bricolage.This can be achieved in two ways: through public negotiation (of proper behaviours) or through 'daily practical enacting' (ibid).
In Finnmark, as we have shown above, practical enacting is paramount to what Paine (2009) calls 'politics through reindeer'.In this sense, practices, as markers of meanings and authority¸ re-negotiate institutional change.For Sámi reindeer herders such practices and their importance are well-documented (Solem 1970;Tønnesen 1979).Paine (2009) refers to such practices as illustrating people's belonging to territories or neighbourhoods, and also that such belonging entailed a 'pragmatic understanding of usufruct' (57) that was constantly reshaped to a certain degree.This suggests the 'culturally creative act' that de Certeau (1984,123) associates with 'marking out boundaries'.To him, the process entails both the construction of spatial stories about a piece of land (how it has been used, what are its 'proper' boundaries etc.), spatial meta-stories (genealogies of places, what are the proper ways of marking the boundaries) as well as acts.Interestingly he maintains that the result of marking out boundaries is itself a process of bricolage and defines these everyday practices of bricolage as 'ways of operating' that are 'victories of the weak ' (1984, xix).
Such acts (or enactments) are nevertheless not necessarily overt resistance to institutions that are perceived as unfair.Instead, they are more subtle authoritative processes by which less powerful actors bypass rules or try to reshape them by their enactments.Following Reed-Danahay (1993), we propose that this type of resistance (which she terms 'débruillardise') is a type of bricolage in that it implies both accommodation and resistance, and is a way of outwitting the dominant ideologies, of disentangling oneself from rules that are perceived as illegitimate or in contradiction with institutional apparatus perceived as legitimate.Such creative forms of 'make do', rely on contriving ways "to subvert or outwit dominant cultural forms and meanings" (ibid, 228) from within the system, without openly confronting it.In this sense, even if reindeer herders make use of the subsidy system put in place by the State, they may actually be engaged in a process of bricolage based on subtle resistance.
By the same token, the fact that some (even if few) herders actively argue for, or enact, the discourse of 'commons' does not necessarily mean they support the dominant ideology of control and management as the solution against an accepted 'tragedy'.They simply make do in the current system of multiple layers of authority and regulations in order to achieve their goals -and accept the social stigma or general negative outcomes as a (temporary) outcome of this.On the one hand, by expanding their pasture area with reference to the official (and legal) idea of "commons", such acts will often be considered as trespassing and might sever their relations to neighbouring siidas and bring about sanctions upon the trespassers and/or their herd.On the other hand, there is a traditional acceptance of the fact that "lucky" reindeer owners expanding their herd will need larger pastures.However, if such an expansion is to be accepted within the pastoral society, it should not take place at the expense of the original users.In other words, such acts have to be negotiated and mediated lest they end up as encroachments.The current enactment of the official version of "commons" opens up for accusations of encroachment, which sometimes end up in the legal system and the media as presentations of "anarchy" and "lawlessness" among the herders.This in turn legitimizes more State interventions aimed at designing perfect institutions for preventing the 'obvious' tragedy.

Conclusion
There is an accepted public perception that large tracts of Finnmark are a common property regime that involves 'common' rights to pasture.This oversimplified misinterpretation has naturalized Hardin's assertion about "the tragedy of the commons" and justified a major institutional reform of reindeer policy measures during the last decades.We have in this article problematized the use of the term 'common' in relation to natural resources in Finnmark by pointing out distinctions between rights to land and rights to exploit certain resources under particular circumstances.We show that by assuming the existence of common rights to territories that belong to undefined groups, the formal institutions developed by the State are at best confusing.
Moreover, the customary institutional arrangements pertaining to reindeer herding in Finnmark hardly fits into the narrow category of 'commons'.Flexibility in the membership of the individual 'commons' (siida), the absence of the right to exclude (difficult circumstances can in principle provide access to areas used by other siidas), rights to specific activities and resources for non-members -all this suggests a far more complex situation.Thus, a focus on grazing rights alone, explains very little of the reality of Sámi reindeer herders simply because the 'commons' refers only to a limited part of the Sami institutional complex used for the exploitation of resources in Finnmark.
Our analysis illustrates how the mismatch between the formal and informal institutions and their underlying logics and styles of thinking has resulted in a complex institutional dynamic affecting land and resource tenure.We identify three major processes that have shaped this dynamic: institutional bargaining, layering, and bricolage.Our analysis of institutional bargaining shows how different institutions (the Norwegian Parliament, the Sámi Parliament, and the International Labour Organisation) have negotiated the emergence of the most recent legislation regulating land tenure.Rather than removing the current informal regulation that Sámi herders enact in their daily practices, this new legislation (The Finnmark Land Act of 2005) as well as numerous others, slowly and gradually infiltrates the current complex of legitimate resource management through the active sponsorship of some principles (e.g.obligation to share resources).This is achieved by layering new rules on top of old ones and the legitimating enactments of these new rules by some herders.Still, we propose that rather than indicating that herders support and legitimise the new regulations, such enactments represent 'make do' processes within the existing situation.7These pragmatic reactions characteristic of institutional bricolage, have nevertheless the unintended effect of granting more authority and legitimacy to the new, formal institutions.
We propose that this historically informed, integrative approach of processes of institutional change is better able to explain the complex social processes taking place in Finnmark.Our analysis suggests that in order to understand whose 'styles of thinking' about resource governance get precedence and are formalised in dominant institutions, we need such cross-scalar analyses.Finally, we argue that a more adequate form of resource governance in Finnmark must operate with concepts and categories that reflect Sámi herders' understanding and customary practice of legitimate and adequate resource-use.Simply attempting to redefine resources as 'common' could create further social, economic, and environmental problems.