Implementing comprehensive land claims agreements in Canada: Towards an analytical framework.

Alcantara, christopher

Introduction

Most federal systems were not designed to include Indigenous governments as constitutional units. In countries such as Australia, Canada, and the United States, the founders assumed that their Indigenous communities would remain subservient to the state or that they would eventually assimilate into mainstream society. Much to their surprise, Indigenous communities in these countries organized and mobilized against assimilative initiatives by seeking political and legal recognition of their rights, title, and constitutional orders through a variety of mechanisms (Papillon 2012; Scholtz 2006).

Nowhere have these efforts been more pronounced than in Canada where Indigenous groups have engaged in lobbying, litigation, negotiation and protests to achieve a range of public goods. One important tool they have used to advance their interests has been comprehensive land claims agreements, otherwise known as modern treaties. These agreements are constitutionally protected under s. 35 of the Constitution Act, 1982 and negotiated between the federal government, Indigenous communities and the provincial/territorial government in which the Indigenous groups are located. Depending on one's perspective, the agreements either transform Indigenous land tenure into property regimes that are legible to the Canadian state or they transfer title and jurisdiction from the Crown to the Indigenous groups in accordance with Canadian law. Regardless, in addition to land ownership and cash, these agreements recognize Indigenous communities as having jurisdiction over things like fish and wildlife, plants, water and ocean management, archaeological materials, environmental regulation, land use planning, and resource royalties, among other things. Some modern treaties also include self-government chapters that, under Canadian law, empower Indigenous communities to form their own governments with jurisdiction over policing, culture and recreation, housing, education, health care, and local government (Alcantara 2013). In short, these agreements are potentially transformative and are at the core of Canada's model of treaty federalism (White 2002) because they are constitutionally protected documents that provide Indigenous communities with a range of tools to pursue their interests within the confines of the Canadian state.

Given the importance of these agreements, scholars in Canada have spent considerable time studying them (Alfred 2008; Belanger 2008; Slowey 2008). Surprisingly, however, virtually nothing has been written on the dynamics of modern treaty implementation, which involves federal, provincial/territorial, and Indigenous actors working to put into practice constitutionally-protected provisions relating to the management and governance of Indigenous lands and peoples. This lacuna is surprising on two levels. For practitioners, Indigenous communities constantly complain that the federal, provincial, and territorial governments routinely fail to fully respect and implement the modern treaties (Fenge 2008; Land Claims Agreement Coalition n.d.) and yet few theoretical tools are available to make sense of these dynamics and complaints. For academics, Canadian federalism has been transformed by the emergence of new Indigenous governments and non-governmental actors (for example, land claims corporations), all of whom participate regularly in intergovernmental bargaining forums. While some of these interactions and forums have been studied (Belanger 2008; Papillon 2012; White 2002), others such as modern treaty implementation have not, so our knowledge of contemporary intergovernmental relations is partial and incomplete.

This article contributes to the literature by building an analytical framework for studying the politics of intergovernmental relations in the implementation of modern treaties in Canada. Its primary goal is to develop an analytical framework and typology to guide future theoretical and empirical analyses of this phenomenon and thus follows in the tradition of scholars who work in the area of typological theory writ large (George and Bennett 2005; Elman 2005) and those who develop typologies specific to Indigenous politics (Abele and Prince 2006). To construct the framework, the article draws upon the literature on intergovernmental relations and policy implementation to tease out key factors that may explain the dynamics of treaty implementation within and across agreements. To illustrate how this framework might be applied to specific cases of treaty implementation, this article explores four issues facing two groups in Canada: Nunatsiavut in northern Labrador and the Inuvialuit in the Northwest Territories. These groups were the focus of two Auditor General of Canada reports on treaty implementation and provide sufficient detail to demonstrate how this article's framework might be applied to the real world.

Background considerations: historical and modern treaties in Canada

According to the standard narrative among many Canadian lawyers and policymakers, the territory of Canada was acquired mainly through the signing of some historical treaties by the British and later the Canadian Crown with Indigenous peoples (see Alfred 2008 and Penikett 2006 for the counter narrative). The main goal of those agreements, at least from the perspective of the non-Indigenous officials, was to acquire title to Indigenous lands so that the Crown could use and dispose of those lands to European settlers. In exchange for those lands, Indigenous groups received much smaller parcels of land, located in usually remote areas, as well as hunting and fishing rights, money, and supplies, among other things (Miller 2009).

Although much of Canada's territory is subject to the historical treaties, large parts of Canada are not, including most of British Columbia, Northern Quebec, Newfoundland and Labrador, and parts of Yukon Territory and Northwest Territories. In 1921, the government of Canada ceased signing treaties, believing that it had sufficient authority over all of Canada's lands to no longer need to negotiate any additional historical treaties. In 1973, however, the federal government changed its mind when it initiated what it called comprehensive land claims negotiations with those groups that had never signed historical treaties with the Crown. During that year, the Supreme Court of Canada handed down a legal judgment in R. v. Calder that recognized the existence of Aboriginal title. Previous to that decision, Indigenous groups had begun a successful mobilization and lobbying campaign to convince the federal government to recognize Indigenous claims and title if it wanted to ensure that economic development proceeded relatively smoothly in northern Canada (Alcantara 2013; Miller 2009; Scholtz 2006).

As of 2016, twenty-seven Indigenous groups have signed comprehensive land claims agreements with the Crown and exercise Canadianrecognized land ownership and political/legal authority through some form of Indigenous self-government or self-governance regime. These agreements are implemented through a variety of mechanisms, such as fiscal financing agreements that set out the amount of funding provided by the Canadian governments to the Indigenous governments or land claims organizations and the list of programs and services to be provided by them; implementation plans which set out "what must be done to put the agreement into effect, who will be responsible for which implementation activity, as well as when and how these activities will be undertaken"; and implementation committees, made up of representatives of the signatories, to negotiate and oversee these implementation plans (INAC 2013, 5th para).

Each signatory has its own mechanisms for overseeing treaty implementation. At the federal level, the cabinet in 2015 passed a directive for how federal departments are to manage their comprehensive land claims obligations across the country. The directive also called for the creation of a Deputy Ministers' Oversight Committee, chaired by the deputy minister of the Department of Indigenous and Northern Affairs Canada, to coordinate the federal government's approach to modern treaty implementation. The Oversight Committee's activities are supported by the Modern Treaty Implementation Office, located in Indigenous and Northern Affairs Canada (INAC 2015). Provincial and territorial governments tend to manage their implementation responsibilities through their Indigenous Affairs departments. In Newfoundland and Labrador, for instance, it is the Labrador and Aboriginal Affairs Office, located in the Premier's Office, which oversees treaty implementation. Indigenous groups rely on their land claims organizations, such as the Inuvialuit Regional Corporation (IRC) in the Northwest Territories, or a part of their government, such as Nunatsaivut Affairs in Nunatsiavut, to oversee treaty implementation. These bodies report directly to the political leadership of their communities, such as the executive branch in Nunatsiavut or the IRC's board ot directors in the Inuvialuit Settlement Region.

In this article, the federal, provincial, and Indigenous signatories are treated as unitary actors when they interact with each other at the implementation table. This approach is consistent with the assumptions of rational choice and game theory (Kam 2000) but also other empirical studies which find that, despite internal diversity, it is still analytically useful and accurate to treat these actors in a unitary fashion when discussing how they interact with each other during treaty negotiations (Alcantara 2013). At the federal level, for instance, the Privy Council exercises general oversight; the Treasury Board is sensitive to fiscal issues; Justice is interested in the constitutional and legal implications; and Natural Resources Canada is sensitive to how treaties might affect their responsibilities. Similar divisions exist at the provincial and Indigenous levels. When federal, provincial, and Indigenous officials meet to negotiate settlements or implementation, such divisions are less apparent at the table, although they are certainly important for crafting mandates and negotiating positions before and after the parties arrive at the negotiating table.

Analyzing treaty implementation: a framework and typology

Modern treaty implementation is a unique political phenomenon. At its core, it involves domestic government actors from at least three levels working together to figure out how to implement a constitutionally-protected agreement relating to the management and governance of Indigenous lands and peoples. On the one hand, modern treaty implementation is similar to other instances of intergovernmental cooperation, although in this case the intergovernmental interactions are guided by a unique constitutionally-protected document: a comprehensive land claims agreement. On the other hand, this phenomenon is also similar to governments, departments, and street level bureaucrats implementing public policies in accordance with new legislation or regulations. The key difference is that modern treaty implementation involves three levels of government rather than governments and stakeholders. I draw upon both literatures to construct a framework.

Political scientists have long been interested in uncovering the conditions that facilitate collective action and cooperation across a variety of spatial and temporal settings. Scholars have found that the production of public goods depends heavily on the motivations and capabilities of the relevant actors (for example, governments, organizations, groups, and individuals) and the institutional environment in which they operate. Actors are driven by a set of endogenous or exogenous preferences and they pursue particular strategies in hopes of maximizing those preferences, depending on the incentives generated by the institutional rules that govern the particular public good to be distributed or decision to be made (Knill and Lenschow 2001; Munck 1994; North 1990). Richard Simeon's (2006) classic book on federal-provincial diplomacy, for instance, examines how eight actor- and institution-oriented variables interact to produce a range of outcomes relating to pensions, taxation, and constitutional change. Inwood, Johns, and O'Reilly (2011) take a similar approach in their study of intergovernmental policy capacity across four policy fields. Similarly, Alcantara (2013) and Alcantara and Nelles (2016) focus on the interaction between actor preferences and institutional incentives for explaining variation in modern treaty settlements and Indigenous-local intergovernmental partnerships in Canada.

Policy-implementation scholars take a similar approach but focus on different levels or scales of analysis. Some scholars subscribe to a top-down approach, where the analytical focus is on macro-level institutions, policies, and actors. Implementation occurs as a result of central-level forces exerting their will on local populations (Pressman and Wildavsky 1973; Ferman 1990). Other scholars argue that policy implementation occurs through a bottom-up approach, where micro-level institutions, policies, and actors exert considerable agency in dealing with macro-level forces (Mazmanian and Sabatier 1983; Berman 1980). Both approaches recognize that actors and institutions, albeit at different levels, matter for determining whether and how a policy is implemented across multiple jurisdictions (deLeon and deLeon 2002; Mailand 1995; Paudel 2009). In the early 1990s, public policy scholars turned away from these approaches and began to use game theory and principal-agent theory to study policy implementation (Milward and Provan 1998; Scholz 1991). In these studies, though, intergovernmental actors and relations are rarely taken into account.

This article draws on these literatures to examine how Indigenous and non-Indigenous government actors interact when tasked with implementing comprehensive land claims agreements in Canada. At their core, the dynamics of modern treaty implementation can be understood by focusing on two characteristics: 1) actor congruence/incongruence; and 2) the coherence/incoherence of the treaty provisions. Actor congruence/incongruence refers to the extent to which the federal, provincial/territorial, and Indigenous government signatories agree or disagree on the goals and means of a particular treaty provision. Here, the relevant questions are: do the three governments agree on what the main goal of the provision is and how that provision is to be implemented? Policy or provision coherence/incoherence refers to the extent to which the relevant treaty provision is clear or ambiguous in terms of its goals and means. Here, the crucial question is: is the provision vague or specific in terms of its goals and means? The logic here is that sometimes legislation can simply state that citizens should be engaged in the policymaking process or it can describe in great detail how and why those citizens should be engaged (Matland 1995). These two characteristics are similar to the approach of some policy implementation scholars who focus on the interaction between policy (or actor) conflict and policy ambiguity (Matland 1995; deLeon and deLeon 2002). It also resonates with the focus of the political science literature on actors interacting within the rules prescribed by the institutional environment (Alcantara 2013; Simeon 2006).

These two characteristics--actor congruence (incongruence) and provision coherence (incoherence)--interact with each other to produce four types or styles of treaty implementation: 1) administrative implementation; 2) experimental implementation and policy learning; 3) compromised implementation, and; 4) no action or zero sum implementation. (1) Each of these types can be ranked in terms of the amount of conflict that is likely to occur between the treaty signatories, the likely outcomes to be produced, and the key factors (2) necessary for implementation to occur. The four styles and their features are summarized in Tables 1 and 2.

In the first implementation style of strong actor congruence and provision coherence, we anticipate little conflict: the treaty provisions are clear in terms of the policy goals and the means to achieve them. The government actors agree with each other about the policy goals and the means (for example, technologies and instruments) required to accomplish those goals, so no conflict is expected when implementing the provisions. Here the likely scenario is administrative implementation in which the bureaucrats at the relevant level carry out the requirements of the agreement, much like traditional public administration. The key enabling factor for successful implementation (rather than the status quo of no implementation) is the availability of resources (3) for the responsible actors to deliver the policies or provisions (Matland 1995:162).

In the second type of strong actor congruence and weak provision coherence, we anticipate weak conflict to occur between the actors, compared to the previous scenario. Although the treaty provision is ambiguous or vague in terms of policy goals and how to achieve them, the relevant actors agree on what the provisions should achieve and how they should be achieved. This scenario should result in what Matland (1995) calls experimental administration and policy learning. Local actors with delegated responsibility for putting a provision into practice will at first experiment with a particular implementation strategy before modifying it as they receive feedback from its application. Alternatively, if the provision is applicable to a broad range of policy or political contexts, we will likely see local actors across these contexts diverge initially from each other in how they implement the provision. As time progresses and local actors receive feedback from initial efforts, we will likely see policy learning and perhaps policy convergence (Meseguer 2005), with modifications inspired by internal and external feedback.

In the third style of weak actor congruence and strong policy coherence, we anticipate strong conflict among the signatories. While the treaty provision is clear about goals and how they are to be advanced, one set of actors may disagree with the goals of the policy or the means used to advance those goals. In such situations, the likely outcome is compromise. The key factor shaping implementation is power, stemming from the internal characteristics of the actors and the nature of the institutional environment (Alcantara 2013; Simeon 2006). One of the actors must force the others to agree with them--either through coercion or bargaining in the form of side payments or log-rolling--otherwise, implementation is unlikely to occur. It could also involve the parties agreeing to an abeyance or a legal resolution through the courts or a dispute resolution body. An implementation solution is more likely to occur in this situation compared to the one described below because the provision is coherent rather than incoherent; the range of solutions is narrower than when the provision is ambiguous (and thus the range of solutions is wider).

In the final type, the treaty provision is ambiguous in terms of its goals and the means to achieve them and the actors disagree strongly with each other about these issues. The result is maximum conflict, relative to the other scenarios, likely generating zero-sum dynamics and outcomes or, more likely, no implementation at all. Again, the key factor here is power, and resolution will depend on one or more of the actors engaging in coercion or bargaining. Resolution may also be pursued through judicial bodies, although the lack of coherence in the provision may result in the courts pushing the issue back to the actors for negotiation. Indeed, Canadian courts have made it clear that negotiation rather than litigation, and especially for these kinds of issues, is the preferred route (Alcantara 2013; Haida Nation v. British Columbia 2004).

This typology provides a platform for analyzing the dynamics of how modern treaties are implemented (or not) in Canada. It is designed to be applicable to single-n case studies and comparative analysis of multiple cases. However, the four implementation styles described above are "ideal types," meaning that real phenomena may blend or overlap across cells. The next section illustrates how the typology can help illuminate the dynamics of modern treaty implementation in Canada. It draws upon examples found mainly in two Auditor General reports addressing the implementation of the Inuvialuit and the Labrador Inuit Final Agreements.

Implementation styles: illustrations from Nunatsiavut and the Inuvialuit Settlement Region

Administrative implementation

A good example of the administrative implementation style, with its emphasis on strong actor congruence and strong provision coherence, is the way in which the federal government provides fiscal financing to the Nunatsiavut Government. Section 18.1.1 of the Labrador Inuit Land Claims Agreement (2004: 296) requires the signatories to the treaty to negotiate a fiscal financing agreement indicating which programs and services the Nunatsiavut Government is to provide on behalf of the Crown. This agreement, to be re-negotiated at least every five years, also lays out the funds the Nunatsiavut Government will receive to deliver these programs and services. From 2005 to 2014, the federal government allocated over $300 million to the Nunatsiavut Government under fiscal financing agreements (Auditor General of Canada 2015: 17; see also Implementation Committee of the LILCA 2014: 6; Implementation Committee of the LILCA 2013: 6).

Two issues related to fiscal financing illustrate this implementation style. Under the terms of fiscal financing agreements since 2005, the Nunatsiavut government was given the task of delivering the federal government's "Non-Insured Health Benefits Program for First Nations and Inuit" and the "construction, renovation, and repairs to housing" (Auditor General of Canada 2015: 17). During negotiations to renew one fiscal financing agreement, Nunatsiavut officials told federal counterparts that they "could not provide non-insured health benefits comparable to those that First Nations and Inuit receive under the program elsewhere in Canada" (Auditor General of Canada 2015: 17). The federal government, in response, provided additional funding to the Nunatsiavut government because it delivers a non-insured health benefits program to First Nations and Inuit communities elsewhere in Canada. The existence of this program meant it could allocate money from that program to the Nunatsiavut Government (Auditor General of Canada 2015:17-18).

Nunatsiavut officials also raised concerns about their ability to build and repair housing for its members under the terms of the fiscal financing agreement. This time, however, the federal government balked at providing additional resources because "there was no federal housing program for Inuit south of the 60th parallel. Consequently, no federal funding could be identified to address the housing needs of Nunatsiavut residents" (Auditor General of Canada 2015: 18). Some money was available through the Canada Mortgage and Housing Corporation (CMHC) but those resources went to provincial governments rather than specific Indigenous communities. To access that money, the Nunatsiavut Government had to apply and compete against other applicants in the province for the money.

These two disputes are illustrative of the administrative implementation style. The treaty provisions and policies governing fiscal financing were clear and coherent, with the treaty requiring the signatories to negotiate a fiscal financing agreement that outlined the sources of funding and the slate of programs to be delivered by the Nunatsiavut Government. The parties successfully negotiated these agreements within the timeframes established in Chapter 18 of the Labrador Inuit Land Claims Agreement (2004: 296) and tended to do so with little difficulty or conflict (Implementation Committee of the LILCA 2014: 6). The narrative told by the Auditor General report indicates that the actors involved, such as the Nunatsiavut Government, the federal Department of Aboriginal Affairs and Northern Canada and Health Canada, easily agreed upon the goals and means used to implement those commitments.

Resources were the key factor at the centre of both issues. For noninsured health benefits, Nunatsiavut officials informed the federal government that it lacked the necessary funding to provide equivalent coverage for its members and so the federal government provided the necessary funding from a similar program it had in place for other Indigenous groups in Canada. For housing, Nunatsiavut officials informed the federal government about its inability to fund adequate housing construction and repair. Although the federal government agreed that this was a legitimate problem, and that it would work with the Nunatsiavut Government and the provincial government to "examine potential solutions to address the issue of housing for the Labrador Inuit south of 60," it could not provide additional resources because it had no equivalent program for Inuit housing south of 60 outside of the land claims region (Auditor General of Canada 2015: 19). Consequently, an implementation solution had not been reached at the time of the Auditor General's investigation. Shortly thereafter, and likely due to the coherence of the treaty provision and the congruence among the relevant implementation actors, the federal government did identify additional resources for Inuit housing south of 60 in its 2016 Budget thus allowing administrative implementation to occur. In short, strong actor congruence combined with strong provision coherence produced a low-conflict, traditional public administration, implementation style, where policy implementation depended mainly on the availability of resources.

Experimental implementation and policy learning

A good example of the experimental implementation style, with its emphasis on strong actor congruence and weak provision coherence, is the dispute related to the Labrador Inuit Park Impacts and Benefit Agreement (IBA) and the management of archaeological materials and sites in the Torngat Mountains National Park. Section 9.2.2 of the Labrador Inuit Land Claims Agreement requires that the Government of Canada and the Nunatsiavut Government negotiate a Park IBA prior to the creation of any national park on lands subject to the modern treaty (Labrador Inuit Land Claims Agreement 2004: 129). The parties negotiated this IBA in 2005 and did so prior to creating the Torngat Mountains National Park in the northern part of Labrador under chapter 9 of the agreement (Labrador Inuit Land Claims Agreement 2004: 129). This IBA, which is not publicly available, formalizes the relationship, obligations, and responsibilities between the Nunatsiavut Government and Parks Canada. Among other things, it outlines the kinds of training and economic opportunities for Nunatsiavut beneficiaries and businesses, and calls on the signatories to "develop and implement a memorandum of understanding on the presentation, management, and safekeeping of archaeological materials found in the park" (Auditor General of Canada 2015:15).

In short, Parks Canada and the Nunatsiavut officials engaged in experimental implementation and policy learning, with Parks Canada changing its policy implementation approach (for example, from unilateralism to joint management) in response to varying circumstances, experiences, and feedback from the Nunatsiavut Government.

Although this IBA has been in effect since 2005, the memorandum of understanding has yet to be negotiated and ratified. Several disputes have arisen relating to the management of archaeological materials and sites located within the National Park. After the ratification of the land claims agreement, Parks Canada "decided, without consulting the Nunatsiavut Government, to move archaeological materials from the park (stored in Dartmouth, Nova Scotia), to Gatineau, Quebec" (Auditor General of Canada 2015: 16). Nunatsiavut officials immediately expressed concerns about the transfer and the materials were eventually returned to the park. In 2013, Nunatsiavut Government officials raised concerns with Parks Canada about visitors damaging an archaeological site in the park. Parks Canada responded by mapping and assessing the site, developing a plan to restrict access to the site and requiring that a Parks Canada staff member accompany any and all visitors to the site. Although Nunatsiavut officials appreciated these efforts, they were still concerned that Parks Canada had no formal process in place to map, assess, and protect other sites in the National Park. As a result of that and the previous dispute, Parks Canada and the Nunatsiavut Government began negotiating a comprehensive memorandum of understanding shortly thereafter to address the management and protection of all archaeological materials and sites in the park (Auditor General of Canada 2015:16-17).

These disputes display the key features of the experimental implementation and policy learning style. The policies and provisions governing archaeological materials in the National Park were vague and simply called on the parties to negotiate a memorandum of understanding to address this issue. In short, there were no clear rules about how archaeological materials and sites were to be managed. Both actors--Parks Canada and the Nunatsiavut Government--shared the same goal of protecting Inuit archaeological sites and agreed to work together to develop a formal process and plan for doing so. Recently, the parties agreed to create a seven-member all-Inuit Torngat Mountains National Park Cooperative Management Board to provide advice to Parks Canada for managing the park. Parks Canada and the Nunatsiavut Government have finally drafted a memorandum of understanding relating to archaeological management in the park and are currently reviewing it in preparation for discussing and signing it later this year (Auditor General of Canada 2015: 15-17). In short, Parks Canada and the Nunatsiavut officials engaged in experimental implementation and policy learning, with Parks Canada changing its policy implementation approach (for example, from unilateralism to joint management) in response to varying circumstances, experiences, and feedback from the Nunatsiavut Government.

Compromised implementation

A good example of the "compromise" implementation style, with its emphasis on weak actor congruence and strong provision coherence, is the dispute over shrimp fishing between the Government of Canada and the Nunatsiavut Government (Implementation Committee of the LILCA 2014: 7). The Nunatsiavut government claims that the Department of Fisheries and Oceans (DFO) has not lived up to ss. 13.12.7 and 13.12.9 of the modern treaty given how the department has gone about managing shrimp fishing licenses and quotas in the region. These sections, especially when compared to other modern treaty provisions, are highly coherent and clear. Section 13.12.7 of the Labrador Inuit Land Claims Agreement reads: if the federal Minister of Fisheries and Oceans "decides to issue more Commercial Fishing Licenses to fish for shrimp in Waters Adjacent to the Zone than the number available for issuance in the year of the Agreement, the Minister shall offer access to the Nunatsiavut Government through an additional Commercial Fishing License issued to the Nunatsiavut Government or by some other means to 11 percent of the quantity to be Harvested under those licenses." Furthermore, section 13.12.9 states that if "the system for allocating commercial opportunities in relation to a species or stock of Fish ... changes from the system existing on the Effective Date (of the agreement), the Minister shall offer to the Nunatsiavut Government participation under the new system that is at least as favourable as that set out" under s. 13.12.7 (Labrador Inuit Land Claims Agreement 2004: 214). In short, these are highly detailed and specific treaty provisions.

Although DFO has not issued any new shrimp fishing licenses since 2005, it increased the amount of shrimp that can be fished by commercial fishers in 2008, 2012, and 2013. In 2012, DFO allocated an additional 300 metric tonnes to the Nunatsiavut government "which represented 6.5 percent of the total increase in allowable catch" (Auditor General of Canada 2015: 11). This increase, according to the department, was a special allocation and not related to ss. 13.12.7 and 13.12.9.

Two other issues complicate this dispute. In 2013, DFO changed the boundaries that it uses to manage its shrimp fishing in northern Labrador. The Nunatsiavut government maintains this constituted a change to the system referred to in s. 13.12.9. Also, in 2014, "in response to conservation concerns, the Minister lowered the amount of northern shrimp that the fishing industry was allowed to catch" which included a reduction for the share allocated to the Nunatsiavut Government (Auditor General of Canada 2015: 13). Nunatsiavut officials maintain they should not have been subject to a reduction since their share of the quota was substantially less than the 11% referred to in s. 13.12.7 and that reducing their share "was inconsistent with the constitutional protection provided by the agreement" (Auditor General of Canada 2015:13).

The solution to the impasse will turn on power relations and the ability of the actors to coerce (for example, threatening to seek redress from the dispute resolution body or from Canadian courts) or bargain with each other.

The Auditor General's report recommended that the two parties engage in negotiations to clarify their obligations and responsibilities as they relate to ss. 13.12.7 and 13.12.9. Should negotiations fail, they are to consider bringing the matter to the dispute resolution board, which has the power to deal with such issues as laid out in Chapter 21 of the Labrador Inuit Land Claims Agreement. In response to these recommendations, DFO agreed to do what the Auditor General suggested but with the proviso that "the Minister's discretion under the Fisheries Act is not fettered. Timelines for implementing this recommendation will be discussed with the Nunatsiavut Government as part of our ongoing discussion on fisheries-related issues" (Auditor General of Canada 2015:14).

This dispute has all of the characteristics of the compromised implementation style. The ss. 13.12.7 and 13.12.9 provisions are fairly coherent and clear, focusing on the obligations of the signatories should there be an increase in the number of commercial fishing licenses or a change in "the system for allocating commercial opportunities in relation to a species or stock of Fish." The crux of the dispute is a disagreement between DFO and the Nunatsiavut government on the "means" of the provisions. Specifically, the parties disagree on whether increases and decreases in fishing allocations and changes to fishing boundaries invoke the two provisions in chapter 13 of the modern treaty. The solution to the impasse will turn on power relations and the ability of the actors to coerce (for example, threatening to seek redress from the dispute resolution body or from Canadian courts) or bargain with each other. Both parties have indicated that, although discussion is likely to occur, they are willing to flex their different power resources to defend their positions. As a result, strong conflict is likely to continue as evidenced by Nunatsiavut officials referencing the constitutional nature of their modern treaty and department officials mentioning the importance of ministerial discretion under the Fisheries Act.

No action or zero sum implementation

A good example of the "no action or zero sum" implementation style, with its emphasis on weak actor congruence and weak provision coherence, is the dispute over the economic measures chapter of the Inuvialuit Final Agreement. Section 16.2 states that "Canada and the Inuvialuit agree that the economic measures set out in this section should relate to and support achievement of the following objectives: (a) full Inuvialuit participation in the northern Canadian economy; and (b) Inuvialuit integration into Canadian society through development of an adequate level of economic self-reliance and a solid economic base" (Inuvialuit Final Agreement 2005: 54). Section 16.3 required that the parties conduct a joint review in 2000 and five years thereafter of the extent to which the objectives in s. 16.2 have been achieved.

In 2001, the signatories (for example, the Inuvialuit Regional Corporation, the Government of Canada, and the Government of the Northwest Territories) completed the review and "found that the economy of the Region had not improved since the signing of the Agreement, that the economic objectives had not been met, and that the Inuvialuit were falling behind their Northern neighbours" (Auditor General of Canada 2007 17-18). After seventeen months of silence from the federal Department oi Indian and Northern Affairs Canada (INAC), the Inuvialuit Regional Corporation proposed 26 recommendations to address the findings oi the 2001 joint economic measures review report. INAC agreed to implement four of the recommendations and to consider implementing eighl others. According to the Auditor General of Canada (2007: 18), the federal government failed to implement any of the Inuvialuit's recommendations nor did it conduct a subsequent economic review since the firsl one in 2001.

Successful implementation of any sort will likely require the Inuvialuit Regional Corporation or the Government of Canada to exercise power through coercive means or through bargaining.

Here implementation stalled as a result of weak provision coherence and weak actor congruence. Section 16.2 is highly ambiguous and vague, calling for the full participation of the Inuvialuit in the northern Canadian economy and their integration through "economic self-reliance and a solid economic base." These provisions are similar to other highly ambiguous policies identified in the policy implementation literature; Mailand (1995: 169) and others (deLeon and deLeon 2002: 486) point to the U.S. War on Poverty legislation that created the Community Action Agencies, as an example of a highly ambiguous policy. The policy goal of these agencies was to "facilitate local citizenry empowerment. What this meant was unclear to virtually all participants" (Matland 1995: 169; deLeon and deLeon 2002: 486).

Similar dynamics are evident in the economic measures chapter in the Inuvialuit Final Agreement. The vagueness of s. 16.2 was accompanied by disagreement between the Government of Canada and the Inuvialuit Regional Corporation about what this section meant "foi their respective roles and responsibilities. It [the 2001 review] noted that the Inuvialuit believe these measures were intended to create a pro-active, collaborative partnership, while Canada focuses on specific, clearly identified obligations" (Auditor General of Canada 2007 18). Actor incongruence centred on actor disagreement about how tc go about putting the obligations under s. 16.2 into practice. The result has been no action. Successful implementation of any sort will likely require the Inuvialuit Regional Corporation or the Governmenl of Canada to exercise power through coercive means or through bargaining.

Discussion and conclusion

Given the lack of academic literature on this topic, this article has sought to make a scholarly and practical contribution by constructing a typology of implementation styles that can be applied to a wide range of modern treaties in Canada and perhaps land claims treaties in other countries. It suggests that the level of actor congruence/incongruence and the level of treaty provision coherence/incoherence can have a powerful effect on how and why treaties are implemented (or not), and interact to produce different implementation styles with distinctive levels of conflict, dynamics, and likely outcomes.

It suggests that the level of actor congruence/incongruence and the level of treaty provision coherence/incoherence can have a powerful effect on how and why treaties are implemented (or not).

The next step is for researchers to further develop this framework by elaborating its theoretical foundations and through empirical application. For instance, what kinds of theoretical expectations drive whether actor congruence or incongruence is likely to occur in any given situation? Here the tools of rational choice and political economy might prove useful because together they stress the importance of economic self-interest or interdependent self-interest for why different actors behave the way that they do. Researchers interested in understanding the forces underpinning actor congruence and incongruence may benefit by drawing on a large body of research that has found that modern treaties are very much driven by economic considerations, especially among federal, provincial, and territorial actors, but also Indigenous ones as well (Alcantara 2013; MacDonald 2011; Penikett 2006; Slowey 2008).

Further work is needed to investigate whether the complexity inherent within each of the signatories has an effect on treaty implementation. This article assumed that each signatory could be treated as a unitary actor, but each typically contains a diversity of organizations and interests. To what extent does the internal diversity of the federal, provincial/territorial, and Indigenous actors affect actor congruence or incongruence?

Similar theoretical work is needed to understand the dynamics and effects of treaty provision coherence. The level of coherence could be a function of the legacies of treaty negotiations. Provisions that have weak coherence may have been the most difficult issues to conclude during negotiations, or those were the provisions negotiators decided to purposely leave vague because they could not agree at the time (thus putting the issue into abeyance to achieve a settlement), or the lack of coherence of a certain provision may have been due to the lack of foresight. To shed light on these dynamics, scholars might fruitfully draw upon the historical institutionalism literature with its emphasis on path dependency, sequencing, and historical legacies (Thelen 1999).

Historical institutionalism might also be helpful for systematically analyzing the effect of time and other contextual factors on treaty implementation (Pierson 2004). The framework set out in this article parsimoniously captures the key issues and dynamics occurring at the treaty implementation table at a given point in time, but does not capture the effects of political, economic, and social conditions over time. Future studies might draw upon work summarized by Lindquist and Wanna (2015: 221-3) which suggests that policy outputs and target-group compliance vary over time as a result of government and staff turnover, changes in public and constituency support, and gradual and sudden changes in socioeconomic conditions, among other factors (see also Mazmanian and Sabatier 1983). Figure 1 identifies four implementation trajectories similar to the Matland typology and the ones presented in this article but more sensitive to contextual and temporal factors (Lindquist and Wanna 2015: 222).

The first trajectory in Figure 1 is effective implementation, akin to administrative implementation (strong provision coherence and actor congruence), in that the actors can effectively and with minimal conflict, implement the policy despite changes to government staff and leaders over time, the latter of which may be mitigated by continued public and constituency support for the initial implementation. The second trajectory is cumulative incrementalism, which is similar to experimental implementation and policy learning (weak coherence and strong congruence): the initial set of actors are modest in their initial efforts to implement the policy but the implementation becomes more successful over time as policy learning occurs and public and constituency support increase. The rejuvenation trajectory - where implementation success relies heavily on the will and interests of leaders and staff as well as changing socioeconomic conditions - is similar to compromised implementation (strong coherence and weak congruence). Finally, the gradual erosion trajectory is similar to zero-sum implementation (weak coherence and congruence) in that implementation is initially problematic and continues to decline due to lack of agreement among street level actors, unsupportive executive and legislative leaders, and changing socio-economic conditions, among other factors (see also Mazmanian and Sabatier 1983).

This temporal perspective is helpful for theorizing how a variety of contextual factors might take shape, given the initial conditions captured by the implementation styles described in Tables 1 and 2. By factoring in time into the treaty implementation trajectories, implementation (as it relates to conformity with the statute or policy) can be flat as what occurs in the effective implementation style, sloped upwards as what occurs in the cumulative incrementalism style, or can vary significantly as what occurs in the other two styles.

While the policy implementation trajectories outlined by Lindquist and Wanna (2015) emphasize factors sensitive to centre-local relations, they can be modified to be sensitive to contextual factors relevant to treaty implementation, such as changing norms and notions regarding what is appropriate in terms of policy responses and solutions. Whereas residential schools, for instance, were once seen as appropriate, today's social and political landscape would preclude any government actor from imposing such a policy. This perspective should encourage treaty-implementation scholars to incorporate a wider range of contextual and temporal factors, particularly those relevant to treaty implementation. An interesting question concerns what the relevant time horizons should be: are we considering five, ten, twenty or more years?

Alternatively, scholars might develop competing theoretical expectations and frameworks and empirically test them against the one proposed here. For example, treaty implementation may depend solely on the costs attached to, or the intrinsic importance of, each treaty provision. Or perhaps a different framework might address the implementation process more holistically and capture the dynamics of modern treaty implementation, rather than divide specific issues into four ideal types. I contend that the framework presented in this article incorporates all of these factors but in different degrees and not in explicit ways. In all four implementation styles, available resources are a necessary condition but also a sufficient condition in the administration implementation style. Similarly, the intrinsic importance of treaty provisions are somewhat incorporated in the actor congruence/incongruence characteristic in that the actors' views about the goals and means of a treaty provision will likely depend on the intrinsic importance of the provision to the actors in question. The issue of analyzing treaty implementation holistically is somewhat captured by the notion of log-rolling and bargaining, although perhaps the legacies of past interactions on other issues need to be better addressed. Adding a historical-institutional element to the framework, as suggested above, may help us better capture these dynamics.

Despite these potential limitations, this article has taken an important first step on a topic virtually ignored in the literature. More theoretically informed empirical studies are needed. Do the implementation styles hold up against what has transpired in modern treaty regions across Canada? Are treaty implementation dynamics in Australia or New Zealand subject to the same characteristics that seem to affect actors in Canada? This article has contributed a matrix of implementation styles and used it to examine several implementation problems uncovered by the Auditor General of Canada. The hope is that Canadian and comparative researchers will draw upon this work to develop a more robust theoretical and empirical literature on this important topic.

Notes

(1) This framework was inspired by the policy implementation work of Matland (1995), who used policy conflict and policy ambiguity instead of actor congruence/incongruence and provision coherence/incoherence. Matland's matrix was designed to analyze how policy implementation occurs within national and sub-national political systems, rather than at the intersection of intergovernmental relations between conventional and unique levels of government (for example, federal, provincial, territorial, and Indigenous governments). The framework in this article modifies his categories and typology in light of these realities.

(2) The identification of a single key factor for each scenario does not preclude other factors from emerging to influence outcomes. The influence of other possible factors is discussed below.

(3) Sufficient resources are a necessary condition for all of the scenarios. In this case, however, the availability of resources happens to also be sufficient for implementation to occur.

References

Abele, Frances and Michael Prince. 2006. "Four pathways to Aboriginal self- government in Canada." American Review of Canadian Studies 36 (4): 568-95.

Alcantara, Christopher. 2013. Negotiating the Deal: Comprehensive Land Claims Agreements in Canada. Toronto: University of Toronto Press.

Alcantara, Christopher and Jen Nelles. 2016. A Quiet Evolution: The Emergence of Indigenous Local Intergovernmental Partnerships in Canada. Toronto: University of Toronto Press.

Alfred, Taiaiake. 2008. Peace, Power, Righteousness. Toronto: Oxford University Press.

Auditor General of Canada. 2007. "Chapter 3: Inuvialuit final agreement." In Report of the Auditor General of Canada to the House of Commons. Ottawa: The Auditor General of Canada, October.

--. 2015. "Report 3: implementing the Labrador Inuit land claims agreement." In Reports of the Auditor General of Canada. Ottawa: The Auditor General of Canada, Fall.

Belanger, Yale. Ed. 2008. Aboriginal Self-Government in Canada. Saskatoon: Purich Publishing.

Berman, Paul. 1980. "Thinking about programmed and adaptive implementation." In Why Policies Succeed or Fail, edited by Helen Ingram and Dean Mann. Beverley Hills, CA: Sage, pp. 205-230.

deLeon, Peter and Linda deLeon. 2002. "What ever happened to policy implementation?" journal of Public Administration Research and Theory 12 (4): 467-92.

Elman, Colin. 2005. "Explanatory typologies in qualitative studies of international politics." International Organization 59 (2): 293-326.

Fenge, Terry. 2008. "Implementing comprehensive land claims agreements." Policy Options July-August: 81-85.

Ferman, Barbara. 1990. "When failure is success." In Implementation and the Policy Process, edited by Dennis Palumbo and Donald Calista. Conn: Greenwood.

George, Alexander and Andrew Bennett. 2005. Case Studies and Theory Development in the Social Sciences. Cambridge: MIT Press, pp. 39-50.

Haida Nation v. British Columbia. 2004. 3 Supreme Court Reporter 511.

Implementation Committee of the LILCA. 2013. 2011-2012 Labrador Inuit Land Claims Agreement Annual Report. Ottawa: Public Works and Government Services Canada.

--. 2014. 2012-2013 Labrador Inuit Land Claims Agreement Annual Report. Ottawa: Public Works and Government Services Canada.

INAC. 2013. Final Agreements and Related Implementation Matters. Available at https://www. aadnc-aandc.gc.ca/eng/1100100030583/1100100030584

--. 2015. Cabinet Directive on the Federal Approach to Treaty Implementation. Available at https://www.aadnc-aandc.gc.ca/eng/1436450503766/1436450578774

Inuvialuit Final Agreement. 2005. Ottawa: Minister of AANDC.

Inwood, Gregory J., Carolyn Johns, and Patricia L. O'Reilly. 2011. Intergovernmental Policy Capacity in Canada. Montreal-Kingston: MQUP.

Kam, Christopher. 2000. "Not just parliamentary 'Cowboys and Indians'" Governance 13: 365-92.

Knill, Christoph and Andrea Lenschow. 2001. "Seek and ye shall find! Linking different perspectives on institutional change." Comparative Political Studies 34: 187-215.

Labrador Inuit Land Claims Agreement. 2004. Ottawa: Minister of DIAND.

Land Claims Agreement Coalition, n.d. A New Land Claims Implementation Policy. Ottawa: LCAC. Available at http://www.landclaimscoalition.ca/assets/LCAC_Discussion_Paper_ 0404162.pdf

Lindquist, Evert and John Wanna. 2015. "Is implementation only about policy execution?" In Nerv Accountabilities, New Challenges, edited by J. Wanna, E. Lindquist, and P. Marshall. Canberra: ANU Press, pp. 209-42.

MacDonald, Fiona. 2011. "Indigenous peoples and neoliberal 'privatization' in Canada." Canadian journal of Political Science 44 (2): 257-73.

Matland, Richard E. 1995. "Synthesizing the implementation literature." Journal of Public Administration Research and Theory 5 (2): 145-74.

Mazmanian, Daniel and Paul Sabatier. 1983. Implementation and Public Policy. Lanham: University Press of America.

Meseguer, Covadonga. 2005. "Policy learning, policy diffusion, and the making of a new order." The Annals of the American Academy of Political and Social Science 598 (1): 67-82.

Miller, Jim R. 2009. Compact, Contract, Covenant. Toronto: University of Toronto Press.

Milward, H. Brinton and Keith G. Provan. 1998. "Principles for controlling agents." Journal of Public Administration Research and Theory 8 (2): 203-22.

Munck, Gerardo. 1994. "Democratic transitions in comparative perspective." Comparative Politics 26 (3): 355-75.

North, Douglass. 1990. Institutions, Institutional Change and Economic Performance. Cambridge University Press.

Papillon, Martin. 2012. "Adapting federalism." Pitblius: Journal of Federalism 42 (2): 289-312.

Paudel, Narendra Raj. 2009. "A critical account of policy implementation theories." Nepalese Journal of Public Policy and Governance 25 (2): 36-54.

Penikett, Tony. 2006. Reconciliation. Vancouver: Douglas & McArthur.

Pierson, Paul. 2004. Politics in Time. Princeton: Princeton University Press.

Pressman, Jeffrey and Aaron Wildavsky. 1973. Implementation. Berkeley: University of California Press.

Scholtz, Chrisa. 2006. Negotiating Claims. New York: Routledge.

Scholz, John T. 1991. "Cooperative regulatory enforcement and the politics of administrative effectiveness." American Political Science Review 85 (1): 115-36.

Simeon, Richard. 2006. Federal-Provincial Diplomacy (second edition). Toronto: University of Toronto Press.

Slowey, Gabrielle. 2008. Navigating Neoliberalism. Vancouver: UBC Press.

Thelen, Kathleen. 1999. "Historical institutionalism in comparative politics." Annual Review of Political Science 2: 369-404.

White, Graham. 2002. "Treaty federalism in Northern Canada." Publius: Journal of Federalism 32 (3): 89-114.

Christopher Alcantara is Associate Professor, Department of Political Science, The University of Western Ontario, London, Ontario. The Traditions & Transition SSHRC Partnership Grant, administered out of Memorial University, financially supported this article. The author thanks Stephanie Irlbacher-Fox, who several years ago invited the author to submit a title and abstract for a SSHRC Connection Grant application on modern treaty implementation. Her invitation nudged the author to start thinking about this important topic. Thanks as well to Graham White, Evert Lindquist, and the three anonymous reviewers of this journal for their helpful advice on an earlier version of the manuscript. Finally, the author expresses his gratitude to Bob Young, who has not only been an incredible mentor during the first year at Western, but also suggested that the author look at the policy implementation literature for this article.

Caption: Figure 1. The Effects of Time and Context on Policy Implementation Table 1. Matrix of Implementation Styles Strong Provision Weak Provision Coherence Coherence Strong Actor Administrative Experimental Implementador Congruence Implementation and Policy Learning Weak Actor Compromised No Action or Zero Sum Congruence Implementation Implementation Table 2. Implementation Styles and Their Features Treaty Implementation Conflict Level Likely Outcome Characteristics Strong Actor Congruence No Conflict Administrative Strong Provision Implementation Coherence Strong Actor Congruence Weak Conflict Experimental Weak Provision Coherence Implementation and Policy Learning Weak Actor Congruence Strong Conflict Compromised Strong Actor Congruence Implementation Weak Actor Congruence Maximum No Action or Weak Provision Coherence Conflict Zero Sum Implementation Treaty Implementation Key Factor for Example Characteristics Implementation to Occur Strong Actor Congruence Resources Fiscal Financing of Nunatsia-vut Housing (and drawing down power) Strong Provision Labrador Inuit Park IBA Coherence and archaeological protection Strong Actor Congruence Local Context NG Shrimp Fishing Weak Provision Coherence Inuvialuit Economic Measures Provision Weak Actor Congruence Power Strong Actor Congruence (Bargaining and Coercion) Weak Actor Congruence Power (Coercion Weak Provision Coherence and Bargaining)