Theorizing the resilience of the Indian Act.

Morden, Michael

Introduction

The Indian Act, 1876--cradle-to-grave legislation overlooking every instant of life for status First Nations in Canada--has been the subject of a disproportionately small amount of direct scholarship. There is a strong, self-evident empirical case to be made for studying its origins, form and function. There is also a good theoretical case. The Indian Act defies most of the expectations of the field, born as it is from racist and authoritarian norms that are outwardly repudiated in the modern context. This alone is insufficient reason to exclude the Indian Act from the Canadian model of difference management. Its form, function, and consequences over a century and a half offer lessons for the comparative project.

There are several research puzzles connected with the Indian Act, but none is more compelling than the question of its extraordinary resilience. The broad, substantive outline of the Indian Act has persisted for seven generations (Borrows 2008) in the face of a complete legitimacy deficit. By extending autocratic control over First Nations and displacing Indigenous institutions and norms of governance, it provoked anger from its earliest enactment. This anger has never abated, but has rather intensified at particular historical junctures since. By ultimately failing its intended purpose of assimilation, and maintaining a kind of negative discrimination that undermines the integrity of the Canadian political community, it has also been a source of shame and indignation for non-Natives and state actors, at least since the mid-20th century. At present, antipathy toward the Indian Act is shared across the political spectrum and national divide. Despite this, it survives to this day having only undergone partial reform. In fact, serious conflict tends to collect around attempts to change the Act.

Fortunately, a rich literature exists precisely for the purposes of explaining institutional resilience. Several possible explanations emerge from comparative politics, including the three (no longer very new) "new institutionalisms": sociological institutionalism (SI), rational choice institutionalism (RI), historical institutionalism (HI). I will argue below that SI is a poor fit, but both RI and HI offer some clues for explaining the survival of the Indian Act. RI helps to explain the constitutional conservatism of Canadian and First Nations political leaders, who oppose changes to a legal regime that limits the field of play and confers authority upon them. HI places emphasis on the feedback effects of past decisions, which prima facie fits the case of the Indian Act nicely. But both can also be used to construct an argument anticipating that fundamental institutional change that has never come.

A newer, fourth school--discursive institutionalism (DI)--helps to fill the gaps that are left. DI appropriately places norms, ideas and discourse at the centre of the story. DI reveals a paradox: normative repudiation of the Indian Act and actors associated with it actually has the effect of inhibiting efforts at reform within "normal politics." The Indian Act is a case of "legitimacy trap": it limits the field of play by according or denying political actors standing in the policy discussion, but political leaders associated with it lack legitimacy in their constituencies because of that association. The legitimacy trap is cyclical and self-reinforcing; failed reforms and reforms perceived to be illegitimate deepen the legitimacy deficit of the Indian Act and actors associated with it, making future reforms ever more difficult. Norms are themselves path dependent, as illustrated in the collective memory attached to the White Paper episode and other acts of betrayal and aggressive assimilation. These historical narratives regenerate a deep reservoir of distrust that becomes activated in any state-led attempts to reform the Act. This historical narrative feeds back on political behaviour and disrupts change.

In short, there are rational, historical, and normative reasons for the persistence of the Indian Act. Each will be explored below.

The Indian Act

When the Canadian model of difference management is discussed, particularly in the comparative context, it is often with one significant omission. This is the 1876 Indian Act, a work of cradle-to-grave legislation that seeks to govern every aspect of the lives of qualifying First Nations in Canada. The Indian Act does not define the institutional relationship between First Nations and Canada in its entirety. It has been complemented over the decades by multiple other elements--both complementary and contradictory elements, such as land claims processes, Supreme Court jurisprudence, provincial-First Nations relations, and legislation outside the Indian Act. But in structuring the basic political relationship between First Nations and the Canadian state, and establishing the form and limits of band governance, it is quasi-constitutional. In every sense, it is an institution designed to manage a profound, identity-based social cleavage. It has done so since the middle of the 19th century.

The Indian Act as we know it was introduced in 1876, but this was only an amalgamation of already-existing laws related to "Indian Affairs." Formal colonial institutions of Indian policy were established in British North America in the 1760s, but at this time the Indian Department served a primarily diplomatic function (Allen 1996; Calloway 1987; Miller 2000a; Willig 2008). This began to change, albeit unevenly, at the beginning of the 19th century (Allen 1996: 175; Calloway 1987: 3). Indian Affairs was transferred from military to civilian authority, and it underwent a dramatic shift from a diplomatic apparatus designed chiefly to maintain good relations to a bureaucratic machine dedicated to bringing First Nations within the governance architecture of orderly "high modernism" (Allen 1996: 178; Scott 1998).

The Gradual Civilization Act was introduced in 1857, which formalized the new goal of assimilating First Nations people into sedentary, Christian, agricultural communities (Milloy 1991: 59). This was followed in 1869 by the Gradual Enfranchisement Act, which enabled the creation of underpowered municipality-like band council governments on reserves. Those councils were offered a limited range of peripheral jurisdictional responsibilities, but they worked through Indian Agents who operate near-complete control. Canada also accorded itself the power to disallow any by-laws passed by the band councils, and to depose any elected councillors who were deemed nuisances. These acts were combined in the Indian Act, 1876. It incorporated a huge range of separate functions, including defining Indian status and the entitlements and legal limitations that accompanied it, establishing land management regimes on the reserve, managing the sale of natural resources, structuring the administration of Indian monies, and defining band council power and electoral systems. This law was often amended over the next several decades, increasing the control exercised by the Government of Canada's Department of Indian Affairs (DIA) over First Nations people. Amendments were passed prohibiting ousted band councillors from seeking immediate re-election (1884), outlawing religious ceremonies and dances including the potlatch and tamanawas (1885), allowing non-Native municipalities to expropriate reserve land for public works (1911), and prohibiting anyone from hiring a lawyer to work on behalf of an Indian band (1927).

A joint committee of the Senate and House of Commons examined Indian Affairs in 1946-48, and proposed some incremental liberalizing revisions (Leslie 2002). Amongst other changes, a centralized Indian Register was created, carrying forward and intensifying pre-existing gender discriminatory provisions governing Indian Status, designed to limit access to status and drive assimilation. Another review process was undertaken from 1963-67 on a much larger scale, under the auspices of the Hawthorn Commission. Hawthorn recommended more incremental change based around the notion of Indians as "citizens plus," but the report was never adopted as policy.

The most dramatic act of reform ever attempted came in 1969, when the government of Prime Minister Trudeau released its "Statement by the Government of Canada on Indian Policy"--which, green cover notwithstanding, became commonly known as the White Paper (Tennant 1990:150). The White Paper proposed winding down the reserve system, devolving some Indian Affairs powers to the provinces, and generally ending any kind of special legal status for First Nations. It broke radically with past practice, and stimulated massive contentious mobilization across the country. The White Paper is correctly credited with introducing a new epoch of Indigenous-settler relations, one often characterized by direct action outside of state institutions. The policy itself was quickly retired in the face of opposition the government had never anticipated.

Some other targeted changes were made to the Indian Act after 1969, including BU1-C31 in 1985, which addressed some gender discriminatory provisions governing Indian status by reinstating women who had lost status through marriage to non-Status husbands (further reinstatement amendments were made in 2010). In 2002, there was another failed attempt at reform on a large scale. Bill C-7, the First Nations Governance Act, aimed to broaden the law-making authority of band councils, and would invest band councils with legal personhood, but also imposed harsher mechanisms of review and accountability. The Assembly of First Nations (AFN) announced its "complete and unequivocal rejection" to the finished product (Cassidy 2003: 48), the Bill died on the order paper, and it was not resurrected.

The Indian Act of today is not the Indian Act of 1876, to be sure. It is important to note that where grand efforts at reform have universally failed, limited change--including the devolution of administrative responsibilities to band councils, opt-in sectoral legislative agreements adopted by some First Nations, administrative changes to transfer payments and other critical policy instruments--has occurred. Moreover, the Indian Act's effects have been altered and--in some senses--diminished by the complex of sometimes contradictory institutions and policies that have grown up around it to structure the First Nations-settler relationship in parallel. Nonetheless, its resilience must be considered extraordinary, particularly in light of its profoundly colonial substance. Recently described as a "Victorian horror" (Swain 2010: 205), the Act notoriously gives chiefs too much power vis-a-vis their communities, and too little power vis-a-vis the Canadian government (Imai 2007). Rules governing Indian status have had a profound effect on the social, material, political, and cultural make-up of First Nations societies (see Dickson-Gilmore 1999; Palmater 2011). For these reasons and more, it has been a central source of First Nations anger for its entire long life. Moreover, the Indian Act has failed the assimilated tasks imagined for it originally by the non-Native fathers of Indian Affairs, preserving both positive and negative varieties of legal discrimination. Despite all this, the Indian Act remains in force today. Since voices from across the political spectrum now call for its dismantling (and this is nothing new), it is important to revisit the history of the Indian Act and understand its staying power.

Theorizing institutional change and continuity

There is an obvious and rich comparative theoretical literature to turn to in search of help explaining the resilience of the Indian Act. The so-called "new institutionalisms" are precisely engineered for making sense of institutional continuity. Indeed, the commonly cited failure of these approaches is that they do a poor job of explaining change. Several possible explanations emerge from the comparative literature on institutional resilience, and the three traditional "new" institutionalisms: sociological institutionalism, rational choice institutionalism, and historical institutionalism. Each will be addressed here briefly--with acknowledgements that this is now well-worn terrain--along with discussion of the ways that they both fit and are dissonant with the experience of the Indian Act.

Sociological Institutionalism (SI) is the first school addressed here because it offers the least with respect to explaining the resilience of the Indian Act. SI adopts broad understanding of institutions, as "culturally-specific practices, akin to the myths and ceremonies devised by many societies, and assimilated into organizations, not necessarily to enhance their formal means-ends efficiency, but as a result of the kind of processes associated with the transmission of cultural practices" (Hall and Taylor 1996: 946-947). Institutions inform behaviour not just by structuring incentives, but by reifying cultural norms and values, and defining individual roles, scripts, and expectations. When people act the roles prescribed to them, they reinforce the institutions that generate those scripts, and continuity prevails.

It is difficult to reconcile this vision of institutions and institutional dynamics with the resilience of the Indian Act, for the simple reason that the Indian Act has never operated according to a "logic of social appropriateness" (Hall and Taylor 1996: 949). From its earliest enactment, it was seen by First Nations as an inappropriate and authoritarian imposition. While the political response was muted, and some First Nations political organizations of the time, such as the largely Anishinabek Grand General Indian Council, even welcomed it (Shields 2001: 57), this only reflected a lack clarity about the new legal regime, and about the degree to which it was intended to violate the treaties-based national model that had presided previously. Small-scale local protest met and often rendered impossible the implementation of elements of the Act (see, for example, Cole and Chaikin 1990; Loo 1992). While Canadian policy-makers worked hard to operate within the footprint of the old treaties model in order to escape major conflict, the cumulative effect of DIA domination in First Nations communities had set the Indian Act firmly outside of norms of appropriateness in Indigenous society by the late 19th century.

One might respond by setting the historically less powerful party aside and arguing that the Indian Act has fit comfortably within and actively reified settler cultural norms. This was true when a particular model of forceful assimilation, Christianization, and "civilization" were broadly embraced across the political spectrum as the ideal end of "Indian policy" in Canada. But these society-wide norms were retired beginning in the post-war era. At least since the 1960s, the Indian Act has violated non-Native notions of appropriateness in both form and function. Growing internal discomfort with the Victorian aims embodied in the Indian Act was widely expressed at mid-century. The DIA was popularly derided as an anachronism already in the 1960s, referred to in the press as the "dinosaur rattling his bones" (Weaver 1981: 72). The Hawthorn Report of 1966 suggested that the defining objectives of Indian Administration had already been overtaken by "obsolescence" (Hawthorn 1966: 400). Nowhere is this normative repudiation quite so clear than in Prime Minister Trudeau's White Paper of 1969, which reflected a commitment to "colour-blind" procedural liberalism. The White Paper reflected a broader, deeply significant normative shift took place in the last half of the 20th century, which is too-often missed in the study of settler colonialism because it has not necessarily altered the desired ends of Indian policy. This brand of liberalism, while still assimilatory in its application to First Nations, is nonetheless irreconcilable with the positive and negative discriminatory features of the Indian Act--precisely because the Indian Act has failed to assimilate First Nations and instead maintains a separate status for First Nations. In short, the Indian Act operates outside of cultural norms and conventions in both Indigenous and non-Indigenous societies, and has done so for some time. For this reason, SI is not well suited to explaining its persistence.

In contrast to the culture orientation of SI, Rational Choice Institutionalism (RI) identifies a calculus explanation for institutional survival. Institutions establish the rules of the game, and structure the strategic interactions between actors that determine political behaviour (Hall and Taylor 1996). A "structure-induced equilibrium" prevails (Keman 1999: 253; Shepsle 1989), where political order is maintained in an institutionally-defined arena that circumscribes the behavioural options open to actors. The Indian Act, under this framework, reduces the transaction costs of First Nations-Canadian bargaining and guarantees a degree of certainty. It is important, in adopting this approach, to consider what actors are responsible for maintaining the institution, and how they are acted upon by the institutional incentive structure.

RI emphasizes the incentive structure that is both created by and can reinforce institutions. With respect to the Indian Act, RI identifies the rational calculus that dissuades both First Nations and non-Indigenous leaders from pursuing wholesale reform. It should of course be understood that the fact of dysfunction and inefficiency--prima facie irrationality--is not in itself reason to reject RI as a possible fit (Miller 2000b). With respect to non-Indigenous elites, the Indian Act clearly serves a range of instrumental ends, despite the fact that it has not done what it was designed to do--assimilate First Nations people. The reserve and band council system contain the Indigenous challenge to Canadian sovereignty. Theoretically, they funnel political activism through institutions designed to mitigate that challenge. This is an ideal venue for non-Indigenous political actors because here the Indigenous challenge is re-constructed as an administrative or public policy issue, rather than as an inter-group dispute dragging Canadian sovereignty claims themselves into question. As Keman (1999: 253-54) explains, in equilibrium, institutions provide rules that structure conflicts, and provide a consensual basis for problem-defining and problem-solving.

A particularly key function the Indian Act serves is to limit the field of play, by conferring standing in the policy discussion only to those First Nations leaders the Canadian state wishes to deal with. Select First Nations political actors are given stakes in a system conceived in Ottawa, and this can be perceived as strongly serving Canadian elite interests. Therefore, while the Indian Act system is derided as costly, one could argue that it is embraced for the certainty it provides.

Some First Nations political leaders also, arguably, have an incentive to maintain the Indian Act status quo, despite the fact that they desire far more autonomy than it provides. Most elective chiefs and band councils draw the governing authority they possess from the Indian Act. This can breed conservatism, and fear of loss-of-position in whatever reforms are to follow. Shin Imai (2007) has argued that two ostensibly contradictory claims about the Indian Act are both true--that it gives both too much and too little power to chiefs (see also Alcantara, Spicer and Leone 2012). "Too much power" is a misnomer of sorts--radical centralization persists in Indian administration--but it is at least the case that politically, chiefs and council operate independent of some expected community-accountability mechanisms in many communities (ibid.). At the same time, real power remains with the Department of Aboriginal Affairs.

RI analysis reminds us to be conscious of "nested games" (Tsebelis 1990); in this case, First Nations leaders must strategize both to achieve their interests vis-a-vis the Canadian state, and also vis-a-vis other people in their communities. While the Indian Act appears to be sub-optimal in the inter-group sphere, it is possible that it preserves the position of chiefs and councils sufficiently in the intra-group sphere to justify its survival. This presentation of band politics often appears in the normative literature. "The majority of band chiefs don't care about community accountability and questions of integrity," writes Taiaiake Alfred, "because the colonial gravy train keeps dropping loads of cash into their coffers. As a result, they continue to play their designated and essential role in the colonial system" (2005: 44). This point is often overstated, as the real world of First Nations politics is more complex. The world of First Nations politics is deeply heterogeneous, and many First Nations leaders deeply embedded within the Indian Act system have been champions of change. Nevertheless, this political dynamic is alive and must be acknowledged.

Here lies a partial explanation for the resilience of the Indian Act. The regime represents a suboptimal outcome for both state actors (for whom it is an expensive violation of liberal norms), and those First Nations leaders integrated into the band council system (for whom it is an inhibitive and insulting violation of treaty and sovereignty norms). But it accords a degree of certainty to both. Certainty is an important commodity, particularly because no consensus exists with respect to alternatives to the Indian Act. Political leaders are therefore likely to fear what emerges from a more open field of play. In this sense, the Indian Act has at times been perversely effective at mediating the relationship. RI clearly has some theoretical value and applicability in making sense of the Indian Act--but it is not wholly satisfying either.

In fact, an RI argument can be made that the Indian Act is and has been susceptible to major change. Typically, in RI models change occurs via exogenous shocks that alter the parameters framing the institutional equilibrium, or through small-scale endogenous change that slowly undermines that equilibrium. Both sources of change can be identified in the Indian Act.

First, there have been exogenous changes that have conceivably altered the parameters of the Indian Act regime and produced disequilibrium. If the Indian Act worked to limit participation, produce certainty and impose a shared, limited problem definition, this has been largely undermined by external developments. Rulings of the Supreme Court and lower courts, in particular, have effected parametric change in Indian Affairs. Rulings first recognized then incrementally expanded the meaning and application of Aboriginal rights, opened status membership to growing numbers of First Nations people, and have even begun to move toward some recognition of traditional authorities as rights-holders. Successive rulings have also reworked the problem definition by recognizing in incrementally more expansive ways treaty and Aboriginal rights. Court decisions have introduced uncertainty, including over the possible future addition of other recognized Aboriginal groups under the purview of federal Aboriginal Affairs. These exogenous impacts could be regarded as parametric shifts, the kinds of changes anticipated in RI to produce disequilibrium and provoke institutional change.

Endogenous change is also identifiable. Greif and Laitin propose an RI model for endogenous change that fits the experience of Indian Administration in Canada in important ways. They highlight how institutional processes can "affect[] aspects of the situation apart from behavior in the transaction under consideration" (2004: 634) by endogenously shifting "quasi-parameters," which can in time destabilize an equilibrium. "Self-undermining" institutions are institutions which, over time, produce marginal shifts in quasi-parameters such that institutionally-prescribed behaviour eventually ceases to be self-enforcing. By limiting the field of play, the Indian Act regime has pushed excluded First Nations people outside of institutions and toward competing loci of authority--including traditional orders of government in some communities, traditionalist organizations like warrior societies, and ad hoc movements for contentious collective action. In the short-run, co-opting a sub-segment of First Nations political leaders benefitted the Indian Act regime. But over a longer term, this had the effect of strengthening networks of First Nations operating outside the regime. This has increased the level of uncertainty faced by both parties and has particularly increased the costs of participation in the Indian Act system for First Nations leaders. To cite a recent example, in late 2012 and early 2013, a loose network of Indigenous and ally activists stimulated country-wide mobilization in opposition to the Indian Act regime, under the auspices of the "Idle No More" movement. This movement had a dramatic effect on band and national First Nations politics. In its immediate aftermath, First Nations leaders most closely identified with the Indian Act system faced political challenges, and outbidding behaviour was discernible.

In short, RI produces equivocal predictions about the Indian Act. It is possible to recognize a kind of Indian Act equilibrium that can explain relative stasis. But this has become weak and self-undermining over time, in a way that makes major change possible.

The final conventional approach, historical institutionalism (HI), emphasizes path dependency and the reinforcing feedback mechanisms that preserve institutions over time. Path dependent processes can mean the survival of institutions that preserve power asymmetries, that no longer serve their initially intended purposes, and that have outlived their normative or rational usefulness. There are variants of the argument, but a commonly cited model identifies path dependency where "a contingent historical event triggers a subsequent sequence that follows a relatively deterministic pattern" (Mahoney 2000: 535). A path dependent sequence is durable after a critical juncture; again, change is expected to stem from exogenous shocks to the system (though this traditional expectation is challenged by recent developments in the field, which will be addressed below). Because of the weight accorded to history and sequence, the Indian Act outcome may not appear surprising when viewed through an HI lens, despite its lack of legitimacy.

One can identify examples of policy decisions taken early in the Indian Act era that foreclosed or made vastly difficult the realization of new policy directions, such as the offloading of meaningful governance responsibility to band councils. Path dependent centre-periphery governance dynamics have been observed by HI scholars in other contexts; extreme centralization has been found to degrade regional and local governance capacity, making devolution impossible (see Levy 1999). Following this pattern, systematic efforts on the part of the agents of the Indian Act to deplete local First Nations governance resources stimulated a state of dependency over time, which now interferes with the neoliberal push for decentralization and devolution. The Indian Act badly eroded institutional governance memory in First Nations communities--for example, by displacing established governments like the hereditary Haudenosaunee Confederacy. The regime also specifically sought to stem the development of a politicized civil society--for example, by responding punitively to early activists such as organizers in the League of Indians. The radical centralization inherent in the Indian Act system has an enduring impact on First Nations communities, and has made it difficult for Canadian elites to shed those governance responsibilities they no longer care to operate.

Another (mildly paradoxical) lesson that recent exercises in HI contribute is that endogenous and incremental changes can occur to institutions in lieu of formal change, and that this can contribute to formal continuity by defusing societal pressure. Hacker, Pierson and Thelen (2013) identify two related processes: institutional "drift"--"when institutions ... are deliberately held in place while their context shifts, changing their effects" (p. 1), and "conversion"--"when political actors are able to redirect institutions or policies to new ends--that is, use them for purposes beyond their original intent" (p. 2). Both are applicable in our context. Actors inside of it have long recognized the potential to subvert the functioning of the Indian Act to serve First Nations interests. For example, an historical study of two Indian agencies at Georgian Bay in the early 20th century finds that even at the apex of Indian Affairs power, Anishinabek constituents used self-interested Indian agents mindful of the costs of social assistance to lobby for exemptions to game and fish regulatory regimes, committing the resources of the Indian Act toward articulating an early Aboriginal rights discourse (Brownlie 2003). Martin Papillon (2008) describes more recent and significant efforts at converting the institutions of the Indian Act to serve new ends, by describing how in the 1980s, through administrative agreements and executive orders, significant powers related to education, policing, health, social assistance, and economic development were conferred on band councils (p. 301). In effect, the infrastructure created in the Indian Act to exercise control over Indigenous communities has been rededicated to allow for a kind of "self-administration" (ibid.) quite unanticipated by the fathers of the Indian Act.

HI, like RI, adds compelling theoretical structure to the Indian Act story. But it still fails to complete the explanation. One difficulty remaining for HI is in accounting for the White Paper episode, a clear breakdown that led to a critical juncture in Indian administration. While the status quo ultimately prevailed, it is difficult to read the White Paper as anything other than a moment of heightened contingency, from which something radically different could have emerged. There is more to explaining continuity and the Indian Act than just rational calculus and history. There is also a constellation of ideas, norms, and attitudes that has perversely contributed to the survival of the institution.

Here we discover the value of a fourth institutionalism, which has attracted increasing attention in recent years. Vivien Schmidt terms this approach "discursive institutionalism" (DI) (2008; 2010). Schmidt acknowledges that some will see DI as only a nuance added to the existing approaches, and also that it incorporates a range of theoretically and methodologically divergent works. But it is sufficiently distinctive to merit discussing outside the existing typology. The defining feature of DI is that it "take[s] account of the substantive content of ideas and the interactive processes by which ideas are conveyed and exchanged through discourse" (2010: 3). DI is less structural and deterministic than the other approaches, acknowledging a role for "sentient" agents to strategically communicate new ideas, or reframe old ideas (2010: 4). This activity can result in institutional dynamism. Change occurs continuously and endogenously, through communicative action. Schmidt argues that DI can correct a primary failure of the traditional new institutionalisms--their struggle to explain change in the absence of exogenous shock. This may be a particular contribution of DI, but not the only one. It will be suggested below that DI can also help to explain the resilience of the Indian Act in a way that the traditional institutionalisms cannot.

The potential for DI to explain continuity as well as change is so far mostly unrealized (Hope and Raudla 2012). But as Hope and Raudla point out, there is a place for discourse in discussions of continuity: "Where action is in actors interest, where institutions are open to change, and where cultural norms are permissive to action, but there remains policy stasis, discourse may be the causal factor which explains this stasis" (2012: 402-403). DI is applied in a more qualified way here; discourse has operated as a complement to historical structures where they have become weak and vulnerable to collapse--as the mortar that has filled cracks in the structural masonry, and preserved the basic outline of the Indian Act. It has done so in some perverse and paradoxical ways.

DI is, in its present iteration, limited in at least two ways. The first, discussed above, is that it has been primarily directed at explaining institutional change rather than institutional continuity. But a persuasive model for institutional dynamics should be able to explain either outcome with approximately the same success. The second is that in application, an unnecessarily limited concept of discourse has been adopted. Discourse is typically taken to refer to policy ideas and accompanying communicative processes. If this has been broadly true in actual empirical work, Schmidt suggests that her model, in theory, can take in considerably more, including "different levels of ideas (policy, programmatic, and philosophical ...) and different types of ideas (cognitive and normative), but also different forms of ideas--narratives, myths, frames, collective memories, stories ... and more" (2008: 309). The latter, broader category of ideas is the one most relevant in our context. In particular, one "type" of idea--legitimacy norms--matters, and it is strengthened by "forms" of ideas such as myths and collective memory. This ideational content in discourse plays a role in undermining efforts at reform where structures might otherwise allow them, and agents might otherwise pursue them. Importantly, and paradoxically, this process preserves the institutions of the Indian Act while simultaneously fostering normative repudiation of it.

The deep legitimacy deficits associated with the Indian Act that persist in both First Nations and settler societies perversely play some role in preserving the regime, by denying the political actors embedded in it the credibility to pursue change, and fuelling distrust of change initiatives. Legitimacy--"a psychological property of an authority, institution, or social arrangement that leads those connected to it to believe that it is appropriate, proper, and just" (Tyler 2006: 375)--is an old concept that has appeared inconsistently in the field, and which is not now sufficiently developed in the new institutionalisms. Gilley argues persuasively that "taking ideas and ideals seriously means taking legitimacy seriously as the critical link between what institutions do and how they are changed" (2008: 278). There is reasonably strong empirical evidence to support the notion that legitimacy shapes the way people behave in relation to institutions. For example, in exploring why people obey the law, Tom Tyler has famously found that the legitimacy of laws and the institutions supporting them is more significant than fear of punishment (Tyler 2006). It should be anticipated that legitimacy also plays a role in stimulating or inhibiting institutional change. But the nature of that role is complex. For scholars working in the SI tradition, institutions are assumed to be articulations of what is legitimate or "appropriate"; where institutions lack legitimacy, they must be vulnerable to collapse. Both HI and RI present arguments for how institutions may survive legitimacy deficits, but the Indian Act story underscores a different, stronger finding: that institutions can sustain themselves not despite, but because of legitimacy deficits.

When legitimacy is introduced to studies of institutional change, it is often with the assumption of a simple unidirectional relationship: the more legitimate an institution is, the more resistant it is to change. The Indian Act story suggests that legitimacy works in more complex ways. This point relates back to one central function served by the Indian Act that has been discussed above: limiting the political field of play, including imposing limitations on entry and standing in the policy discussion. The Indian Act defines the legal-constitutional political classes, for both First Nations peoples and Canada. While challenges from outside formal Canadian institutions are strong (and in some communities, traditional institutions have retained sufficient robustness to challenge the presence of the Indian Act regime on a local level), most high level First Nations political actors are made party to the Indian Act. The AFN--the ostensibly independent peak advocacy organization for First Nations--follows a template established in the Indian Act by according membership and voting rights to band council chiefs, for example. Of course, Canada's representatives on the Aboriginal Affairs file--those who are responsible for advancing reform proposals on behalf of the state--also draw their authority from the regime. The Indian Act is a gatekeeper. But association with the Indian Act profoundly undermines the legitimacy of the leaders it empowers to make decisions, preventing those leaders from successfully advancing reform. This results in what may be described as a "legitimacy trap." It is a cyclical phenomenon, in which failure to reform illegitimate institutions deepens distrust of and between leaders, increasing the political challenge of selling reform in the future. Where culture, interests and history are permissive to change, then, action can be constrained by the normative overlay of institutions.

The legitimacy trap appears in discourse. Canadian and First Nations political leaders are tarnished by their association with the Indian Act, including in the context of pursuing change. At the occasion of an attempted major reform to the Indian Act--the First Nations Governance Act, introduced by the Chretien government in 2002--the Indian Affairs minister Robert Nault was taunted by protesters, who said "We are through with Indian Agents ... You are fired" (Canadian Press, 15 March 2003). The minister was made to wear the legitimacy deficit of the institution, becoming identified with Indian agents, historic and symbolic keepers of the colonial order (see also, Bear 2001). He subsequently lost his standing in the policy community, along with his ability to direct change. The same epithet--"Indian Agent"--was commonly applied to Shawn Atleo, National Chief of the Assembly of First Nations, by his political opponents because of a perceived conciliatory orientation towards Ottawa in talks over changes to the Act and especially, to First Nations education programming (see, for example, Ball 2014). Opponents derided his willingness to participate in the law-making of "colonial governments," in light of the existence of Aboriginal rights and title that "transcend, or are more fundamental ... than anything that can be presumed upon us by way of the Indian Act or any other federal or provincial legislation that is written on paper"--in the words of Saulteaux Anishinabek chief Derek Nepinak (Galloway 2013). In short, the legitimacy deficit associated with the Indian Act is so profound as to sap the political capital of anyone working within it, including in the interest of changing it. This has the perverse effect of preserving the Act in its broad strokes.

The legitimacy of actors is affected by association with the Indian Act regime. The legitimacy of reform action itself is also impacted by a heavy normative burden, as narrative and collective memory play a powerful role in fuelling distrust of reform processes. The myth is another under-studied normative dimension in DI scholarship. Gerard Bouchard (2013) has recently developed the concept of national myths by suggesting four broad characteristics they generally share: a) they include some hybrid of fact and fiction, though the balance between the two is struck differently in different national myths; b) they are dual, in the sense of telling specific stories about specific places and events, but also replicating universal narrative forms; c) they entail "a kind of sacredness ... that confers upon their contents a self-constraining power"--that is, they are more unassailable than simple policy ideas; and d) that myths are energy-producing, and sufficient for stimulating or inhibiting political mobilization (pp. 2-3).

Myths are discussed in sociological institutionalism, but they are understood differently. In SI, myths are themselves--in a sense--institutions. In contrast, DI concerns itself with myths about institutions--the stories we tell that shape our understanding of those institutions and alternatives to them. Narrative can undermine institutions and generate a belief in the need for change. Narrative can also reinforce institutions, and it can do so in several different ways. The Indian Act occupies a central place in

Indigenous national mythologies, which accurately reflect the pre-eminent role that the Indian Act has played in the establishment of settler colonial control over Indigenous societies. The Indian Act regime is the set-piece for innumerable narratives of calamitous abuse, from contrived famines (Daschuk 2013), to forced settlement, residential schools, and the general degradation of the treaty relationship. But narrative also conveys the dangers associated specifically with reforming or abandoning the Indian Act. The White Paper episode offers one illustrative example of the causal importance of narrative, living on in collective memory and fuelling skepticism of reform processes.

It has been suggested above that the White Paper should be viewed as a critical juncture, during which contingency was heightened and the possibility of radical change to the Indian Act regime was real. Politics killed the White Paper, and quickly. The federal government was wholly unprepared for the contentious response by Indigenous peoples, as was non-Native society generally. By June of 1970, Trudeau was apologizing for "being very naive" (Weaver 1981: 185), and the policy was formally retracted in 1971. But the ultimate, firm re-establishment of the status quo was by no means a foregone conclusion in the historical moment. However, the disastrous White Paper episode, living alongside a vast cultural repertoire of stories of betrayal, stimulated a new and powerful narrative counselling distrust of state-led efforts at reforming the Indian Act.

Though the policy was retracted, the condition of intergroup trust was not restored to the (already low) level that it had existed prior to 1969. Many Indigenous actors continued to express a belief that a White Paper-type abolition of the treaty order was on the horizon. In 1975, Harold Cardinal reported that "Many of our elders feel that, within the foreseeable future, we will witness the demise of the Indian reserve system ... The question is ... how can we salvage the basic tenets of our nationhood?" (1977: 220). Low-grade fear that a White Paper-style obliteration of the treaty order was imminent was to become a stable feature of Indigenous narrative. Periodic attempts to revive individual elements of the White Paper were met with unmovable opposition.

The White Paper narrative has remained in force, as a resonant "chosen trauma" (Ross 2001: 166) that evokes other betrayals by Canada, and is cited frequently to bolster opposition to the Canadian government and collaborators. This is evident in the discourse surrounding some recent mobilization. In 2012, for example, a Crown-First Nations Gathering between the AFN and the government of Canada was derided by hardliner opponents of the AFN leadership as "like the 1969 White Paper assimilation plan using modern words" (Palmater 2012). A proposed First Nations Property Ownership Act, which would create an opt-in framework for bands to convert some reserve land to fee-simple ownership, was dubbed by a leading Indigenous commentator "the White Paper Lite" (Vowel 2012). An omnibus bill that included some changes to the Indian Act was called "the White Paper with a twist" (Palmater 2013), with protesters insisting: "If you look at the White Paper ... that's exactly what the Conservative government wants to do" (Winter 2012). The original text of the White Paper was available at the official website of Idle No More--a movement that brought unprecedented levels of Indigenous protest activity across the country in late 2012 and early 2013--with the comment that it "is the fight we are currently invested in" (Idle No More 2013). At a 2014 policy conference, the Aboriginal wing of the Liberal Party of Canada promoted a resolution, ultimately accepted by the membership, which pledged the party to officially apologize for having ever introduced the aborted policy, 45 years previous. The power of this particular historical narrative on contemporary politics is clear.

The effect of the White Paper has been to deepen distrust of the Canadian government, but also to contribute new discursive content to the notion of Indian Act reform or abolition. That process is now twinned in First Nations narratives with the abolition of treaty relations and legal recognition of First Nations difference. The symbolic content of Indian Act reform inhibits the ability of leaders to pursue their own self-interest and abandon the institutions. It also helps to explain what RI does not, which is that opposition to changes to the Indian Act often involves "mass-first" mobilization (Stroschein 2011)--mobilization amongst non-elite First Nations with no personal or instrumental stakes in the current system, creating pressure on First Nations leaders. This was true of the Idle No More movement, for example.

By focusing attention on discursive dimensions of institutional change, including legitimacy norms and narrative, we are able to identify how the Indian Act remains so deeply rooted in the political bedrock. Perversely, it has done so by violating cultural norms so completely, and functioning so irrationally as to have entirely deprived the institutional processes of First Nations-Canadian relations of legitimacy. Participating in incremental change is seen as complicity, and this precludes the possibility of "contained institutional change"--which has been found to preserve and enhance institutional legitimacy in the face of social change (Gilley 2008). Instead, political mobilization is pushed outside of institutions and into contentious action. In some communities, it has also meant a return to (or reinvention of) traditional institutions that are unrecognized in the Canadian legal order.

Conclusions

Non-Indigenous columnists and commentators routinely lament the sabotage of purported "moderate" First Nations leaders, and accuse the opponents of incremental change to the Indian Act of making "perfect the enemy of the good" (see, for example, Yakabuski 2014). But unwillingness to participate in changes to the Indian Act reflects, paradoxically, the legitimacy deficit it has cultivated. The experience of the Indian Act illustrates the complicated relationship that ideas, norms and discourse can have to institutions. Much of the DI literature simply points to cases where ideas supporting policy change lead to policy change under cultural, rational and historical circumstances that appear to make change unlikely. That model has begun to be applied in the opposite direction, where ideas opposing change have prevented it where structures and elite game-play would have permitted it (Hope and Raudla 2012). I have argued that the negative norms and historical associations attached to the Indian Act are causally important, but that they actually contribute to its resilience. The argument made here can be extended to other institutions that serve the constitutional function of defining the policy-making and political playing field, and conferring standing on political leaders. In a context like Canada, where actual collapse of all or part of the state is exceptionally unlikely, the legitimacy trap can lock in a pattern of painfully resilient dysfunction.

The legitimacy trap is an important contemporary engine for the ongoing failure that is intrinsic in federal control of First Nations communities. This extends beyond solely the functioning of the Indian Act to other organs of government that have grown up around it in the post-White Paper era. Indian Act band councils and the Ministry of Aboriginal Affairs remain the principle actors in policy lobbying and land claims negotiations. As a result, conflict increasingly occurs outside of institutions and so-called normal politics, and organs such as the Assembly of First Nations are likely to continue to struggle to establish firm legitimacy in Indigenous communities. Policy change will continue to be driven in large by decisions of the Supreme Court--decisions like the recent Tsilqot'in ruling, which alter the parameters of the institutional relationship. Policy-makers will also continue to be dogged by the simple policy challenge that no clear alternative to the Indian Act has been consistently articulated. But this problem is itself a symptom of the broader dysfunction, and will not be overcome before the trust and legitimacy deficits are themselves overcome to some degree. In the absence of willingness on the part of Canada to consider wholesale or "uncontained" institutional change, these patterns are likely to endure.

References

Alcantara, Christopher, Zack Spicer and Rob Leone. 2012. "Institutional design and the accountability paradox: A case study of three Aboriginal accountability regimes in Canada." Canadian Public Administration 55 (1): 69-90.

Alfred, Taiaiake. 2005. Wasa'se: Indigenous Pathways of Action, and Freedom. Toronto: University of Toronto Press.

Allen, Robert. 1996. His Majesty's Indian Allies: British Indian Policy in the Defence of Canada 1774-1815. Toronto: Dundum.

Ball, David. 2014. "The writing was simply on the wall for years." Windspeaker 32 (3): 10.

Bear, Jeff. 2001. "Kill the sacred cow." Windspeaker 19 (2): 6.

Borrows, John. 2008. Seven Generations, Seven Teachings: Ending the Indian Act. Vancouver: National Centre for First Nations Governance. Available at http://fngovemance.org/ resources_docs/7_Generations_7_Teachings.pdf (Accessed 12 May 2014).

Bouchard, Gerard. 2013. "The big nation with small dreams: Quebec national myths". In National Myths: Constructed Pasts, Contested Presents, edited by Gerard Bouchard. New York: Routledge.

Brownlie, Robin. 2003. A Fatherly Eye: Indian Agents Government Power, and Aboriginal Resistance in Ontario, 1918-1939. Toronto: University of Toronto Press.

Calloway, Colin G. 1987. Crown and Calumet: British-Indian Relations, 1783-1815. Norman: University of Oklahoma Press.

Canadian Press. 2003. "Marchers fill Kenora, Ontario Street to protest the First Nations Governance Act." Canadian Press Newswire. 15 March 2003.

Cardinal, Harold. 1977. The Rebirth of Canada's Indians. Edmonton: Hurtig Publishers Ltd.

Cassidy, Frank. 2003. "The First Nations Governance Act: A legacy of loss." Policy Options April: 46-50.

Cole, Douglas and Ira Chaikin. 1990. An Iron Hand upon the People: The Law Against the Potlatch on the Northwest Coast. Vancouver: Douglas & McIntyre.

Daschuk, James. 2013. Clearing the Plains: Disease, Politics of Starvation, and the Loss of Aboriginal Life. Regina: University of Regina Press.

Dickson-Gilmore, E.J. 1999. "Iati-Onkwehonwe: Blood quantum, membership and the politics of exclusion in Kahnawake." Citizenship Studies 3 (1): 27-43.

Galloway, Gloria. 2013. "Emerging First Nations leader says indigenous rights 'transcend' any piece of legislation." The Globe and Mail. 26 July 2013. Available at http://www.theglobeandmail.com/news/politics/ emerging-first-nations-leader-says-indigenous-rightstranscend-any-piece-of-legislation/ article13469885/(Accessed January 14, 2016).

Gilley, Bruce. 2008. "Legitimacy and institutional change: The case of China." Comparative Political Studies 41 (3): 259-284.

Greif, Avner and David D. Laitin. 2004. "A theory of endogenous institutional change." American Political Science Review 98 (4): 633-652.

Hacker, Jacob S., Kathleen Thelen and Paul Pierson. 2013. "Drift and Conversion: Hidden Faces of Institutional Change." Paper presented for APSA 2013 Annual Meeting, 29 August-September 1. Chicago, USA.

Hall, Peter A. and Rosemary C. R. Taylor. 1996. "Political science and the three new institutionalisms." Political Studies 44 (5): 936-957.

Hawthorn, E.B., ed. 1966. A Survey of the Contemporary Indians of Canada, V.1. Ottawa: Indian Affairs and Northern Development.

Hope, Mat and Ringa Raudla. 2012. "Discursive institutionalism and policy stasis in simple and compound polities: The cases of Estonian fiscal policy and United States climate change policy." Policy Studies 33 (5): 399-448.

Idle No More. 2013. "The white and red papers." Idle No More. Available at http://idlenomore.ca/ index.php/articles/resources/selected-readings/item/98-the-white-and-red-papers (Accessed March 8 2013).

Imai, Shin. 2007. The Structure of the Indian Act: Accountability in Governance. West Vancouver: The National Centre for First Nations Governance.

Keman, Hans. 1999. "Political stability in divided societies: A rational-institutional explanation." Australian Journal of Political Science 34 (2) (1999): 249-268.

Leslie, John F. 2002. "The Indian Act: An historical perspective." Canadian Parliamentary Review 25 (2): 23-27.

Levy, Jonah D. 1999. Tocqueville's Revenge: State, Society, and Economy in Contemporary France. New Haven: Harvard University Press.

Loo, Tina. 1992. "Dan Cranmer's potlatch: Law as coercion, symbol, and rhetoric in British Columbia, 1884-1951". Canadian Historical Review 73 (2): 125-165.

Mahoney, James. 2000. "Path dependence in historical sociology." Theory and Society 29 (4): 507-548.

Miller, James Rodger. 2000a. Skyscrapers Hide the Heavens: A History of Indian-White Relations in Canada. Toronto: University of Toronto Press.

Miller, Gary. 2000b. "Rational choice and dysfunctional institutions." Governance 13 (4): 535-547.

Milloy, John. 1991. "The early Indian Acts: Developmental strategy and constitutional change." In Sweet Promises: A Reader on Indian-White Relations in Canada, edited by J.R. Miller. Toronto: University of Toronto Press.

Palmater, Pamela D. 2011. Beyond Blood: Rethinking Indigenous Identity. Saskatoon: Purich Publishing.

--. 2012. "AFN Election 2012: Stopping the Assimilation of First Nations in its Tracks." Rabble.ca. 19 March 2012. Available at http://rabble.ca/blogs/bloggers/pamela-palmater/2012/ 03/afn-election-2012-stopping-assimilation-first-nations-its-tra (Accessed March 8, 2013).

--. 2013. "Idle No More: What do we want and where are we headed?" Rabble.ca. 4 January 2013. Available at: http://rabble.ca/blogs/bloggers/pamela-palmater/2013/01/whatidle-no-more-movement-really (Accessed March 8, 2013).

Papillon, Martin. 2008. "Is the secret to have a good dentist? Canadian contributions to the study of federalism in divided societies." In The Comparative Turn in Canadian Political Science, edited by Linda A White, Richard Simeon, Robert Vipond and Jennifer Wallner. Vancouver: UBC Press.

Ross, Marc Howard. 2001. "Psychocultural interpretations and dramas: Identity dynamics in ethnic conflict." Political Psychology 22 (1): 157-178.

Schmidt, Vivien A. 2008. "Discursive institutionalism: The explanatory power of ideas and discourse." Annual Review of Political Science 11 (1): 303-326.

--. 2010. "Taking ideas and discourse seriously: Explaining change through discursive institutionalism as the fourth 'new Institutionalism.'" European Political Science Review 2 (1): 1-25.

Scott, James C. 1998. Seeing Like a State: How Certain Schemes to Improve the Human Condition Have Failed. New Haven: Yale University Press.

Shepsle, Kenneth A. 1989. "Studying institutions: Some lessons from the rational choice approach." Journal of Theoretical Politics 1 (2): 131-147.

Shields, Norman. 2001. Anishinabek Political Alliance in the Post-Confederation Period: The Grand General Indian Council of Ontario, 1870-1936. Queen's University. Unpublished MA Thesis.

Stroschein, Sherrill. 2011. "Microdynamics of bilateral ethnic mobilization." Ethnopolitics 10 (1): 1-34.

Swain, Harry. 2010. Oka: A Political Crisis and its Legacy. Toronto: Douglas and McIntyre.

Tennant, Paul. 1990. Aboriginal People and Politics: The Indian Land Question in British Columbia, 1849-1989. Vancouver: University of British Columbia Press.

Tsebelis, George. 1990. Nested Games: Rational Choice in Comparative Politics. California: University of California Press.

Tyler, Tom R. 2006. Why People Obey the Law. Princeton: Princeton University Press.

Vowel, Chelsea. 2012. "White paper, what paper?" apihtawikosisan. 22 August 2012. Available at http://apihtawikosisan.com/2012/08/22/white-paper-what-paper/(Accessed March 8, 2013).

Weaver, Sally M. 1981. Making Canadian Indian Policy: The Hidden Agenda, 1968-1970. Toronto: University of Toronto Press.

Willig, Timothy D. 2008. Restoring the Chain of Friendship: British Policy and the Indians of the Great Lakes, 1783-1815. Lincoln: University of Nebraska Press.

Yakabuski, Konrad. 2014. "The lost generation." The Globe and Mail. 10 May 2014. Available at http://www.theglobeandmail.com/news/national/canadas-lost-generation/article18586165/ (Accessed 10 May 2014).

Michael Morden is a postdoctoral research fellow at The University of Western Ontario.