1. Indigenous Democracy and Political Theory

Our argument engages and advances three bodies of political theory: on Indigenous political thought, democratic theory and comparative political theory. Scholarship on Indigenous political thought, although still limited, has grown in recent years, and scholars are increasingly taking Indigenous political and legal traditions seriously “on their own terms” (Turner, Reference Turner2006; see Borrows, Reference Borrows2010; Burkhart, Reference Burkhart2019a; Mills, Reference Mills2016; Napoleon, Reference Napoleon and Tully2022). There are significant obstacles to productive engagement between Indigenous thought and mainstream political theory (Allard-Tremblay, Reference Allard-Tremblay2021; Allard-Tremblay and Coburn, Reference Allard-Tremblay and Coburn2023; Burkhart, Reference Burkhart, Marshall and York2019b; Rollo, Reference Rollo2018; Turner, Reference Turner2006, Reference Turner and Hokowhitu2020). Despite this, a number of scholars have staged direct engagements between Indigenous political thought and democratic theory, focusing inter alia on issues of territoriality (Coburn and Moore, Reference Coburn and Moore2021; Luoma, Reference Luoma2022), citizenship (Borrows, Reference Borrows2002; M. S. Williams, Reference Williams and Laycock2004), deliberation (Rollo, Reference Rollo2017; Tully, Reference Tully and Melissa2020), authority (Eisenberg, Reference Eisenberg2022) and autonomy (Allard-Tremblay, Reference Allard-Tremblay2022). We advance this scholarship through a specific exploration of democratic merits of non-electoral Indigenous practices of representation.

In this way, our article fills what remains a significant lacuna in the literature on democratic theory and representation, which, despite a growing interest in alternatives to elections, has largely failed to engage Indigenous alternatives. The conventional view in democratic theory remains that giving people equally weighted votes is necessary for fair and respectful decisions in circumstances of disagreement.Footnote ⁷ Yet, theorists increasingly highlight the potentially undemocratic features of electoral politics and debate whether randomly selecting legislators might be more democratic (Guerrero, Reference Guerrero2024; Hutton Ferris, Reference Hutton Ferris2023). Theorists of representation have also begun to explore the democratic potentials of the representative claims and the actions of various non-elected leaders (Montanaro, Reference Montanaro2017; Saward, Reference Saward2010). These and other debates can be enriched by considering the practices of customary governance discussed below and our argument that these can be understood as potentially highly democratic representative systems (Hutton Ferris, Reference Hutton Ferris2024; Rey, Reference Rey2023).

Finally, this article raises methodological issues that are salient to debates in decolonial and comparative political theory. The argument’s structure is such that we, as non-Indigenous political theorists, must evaluate the democratic merits of Indigenous political practices. This worries us because adopting an evaluative stance positions us as guardians and gatekeepers of a normative standard, “democracy,” which is derived largely from European intellectual traditions.Footnote ⁸ Even if the results come out favourably for customary Indigenous governance, it is reasonable to wonder whether holding “their” practices to “our” standards is a salutary exercise. Does it not risk distorting Indigenous practices? Is it not arrogant to presume the validity of Eurocentric categories? Would it not be better to adopt a less parochial, more reflexive approach—one that seeks to destabilize Western notions of democracy through an encounter with Indigenous thought? Challenges such as these have motivated recent movements to decolonize and deparochialize political theory, especially as reflected in the subfield of comparative political theory (Dallmayr, Reference Dallmayr2004; Euben, Reference Euben2000; Getachew and Mantena, Reference Getachew and Mantena2021; Godrej, Reference Godrej2011; Jenco, Reference Jenco2007; Sherwin, Reference Sherwin2022; M. S. Williams, Reference Williams2020; M. S. Williams and Warren, Reference Williams and Warren2014). We do not take them lightly.

We do, however, aver that the literature’s focus on destabilizing “Western” conceptions can lead it to understate the intellectual and decolonial value of evaluative approaches. An evaluative approach such as ours speaks directly to powerful institutions by shrinking the gap between unfamiliar Indigenous practices and familiar democratic ideals. Many Indigenous intellectuals have recognized this and have correspondingly defended their customary practices as consensus or intensive or participatory democracy (Kellogg, Reference Kellogg, Ackley and Stanciu2015; Napoleon, Reference Napoleon and Tully2022; Notes, Reference Notes1978; Temin, Reference Temin2021; Palmater, Reference Palmater2020; Paul, Reference Paul2022; Senate, 2010). These interventions illustrate how evaluative approaches can be valuable in Indigenous struggles. We would add that evaluative approaches can also benefit the long-term project of de-parochializing democratic theory (M. S. Williams, Reference Williams2020), even if they seem to reinforce Eurocentrism in the short run. For one thing, if an evaluative approach succeeds in motivating policy reforms to create more vibrant Indigenous democracies, this can directly benefit comparatively minded scholars by supporting new interlocutors and additional cases. Moreover, a long-term practice of theoretical comparative engagement need not accept evaluation as its only or final approach. It can instead be a prelude or complement to further dialogues, which might (for instance) decentre democracy in favour of Indigenous concepts such as Haudenosaunee Sken:nen (peace) or Anishinaabe Mino-Bimaadiziwin (living well).

2. Indigenous Government in Canada

The landscape of Indigenous governance in Canada is complex. Prior to colonization, Indigenous peoples in Canada organized themselves in a range of clans, nations, confederacies and coalitions, adopting diverse leadership structures and modes of government. Some of these communities were exceptionally egalitarian, even to the point of lacking permanent leadership roles. Others were hierarchical, with layers of status separating the highest leaders from their subordinates. To our knowledge, none relied on the periodic elections of leaders or representatives. These pre-colonial institutions and the cultures they nourished continue to inform Indigenous peoples’ identities, worldviews, laws and political traditions, even as they have been assaulted and undermined by generations of settler colonialism.

Today, the Canadian constitution recognizes three categories of Aboriginal peoples: Indians, Inuit and Métis (Canada, 1982). The term “Indian” generally denotes individuals who are registered as Indians under the Indian Act and thus have Indian Status, although the term can also apply to “non-status Indians”—persons who are descended from individuals with Indian status who were stripped of their status by various provisions of the Indian Act. Under the Indian Act, there are more than 600 “Indian Bands.” Indigenous peoples have long objected to the term “Indian,” and in modern usage, the term “Indian Band” has been replaced by “First Nation,” and “Indian” by “First Nation person.” But this usage can be misleading if it is taken to imply that the Indian Band/First Nation is politically continuous with a pre-colonial Indigenous polity. In fact, most First Nations are administrative units that were created by the Indian Department’sFootnote ⁹ policies of divide and conquer during the nineteenth and early twentieth centuries, and their boundaries may or may not correspond to those of any pre-existing Indigenous polity.

This essay focuses primarily on the experiences of First Nations peoples, and some of our arguments will not apply to the Métis or Inuit. More specifically, we focus on the Anishinaabe, Wet’suwet’en and Haudenosaunee peoples. These peoples have very different economic, philosophical and governance traditions and experiences of colonization. By discussing all three, we seek to avoid pan-Indigenous generalization while nevertheless showing that our arguments are reasonably robust (Kirshner and Spinner-Halev, Reference Kirshner and Spinner-Halev2024).

Imperial and later settler colonial governments have persistently sought to weaken and subordinate traditional political orders to strengthen their own claims to territorial sovereignty. Since the late nineteenth century, this has involved incorporating independent Indigenous nations as part of the settler state through the Indian Act (of 1876), which divided Indigenous polities into Indian Bands that were located on reserves and placed under the authority of the Indian Department and the colonial executive. A further barrage of assimilationist policies in the late nineteenth century then sought to completely destroy Indigenous cultures and political orders and refashion Indigenous people into “ordinary” British subjects, including by introducing mandatory residential schooling and banning customary governance practices such as the potlatch.

Elective leadership emerged as a part of this overall process of sovereignty-assertion. Since 1869, Canada has claimed the authority to impose elections on Indigenous communities and has done so on groups, including the Haudenosaunee, who resisted the imposition of the Indian Act, in an attempt to break the power of traditional leaders (Titley, Reference Titley1992: 110–28).Footnote ¹⁰ A 1899 amendment to the Indian Act made elections compulsory for all communities east of Ontario (Walls, Reference Walls2010).Footnote ¹¹ Elections were then introduced across much of the West after the Second World War in an attempt to shore up the legitimacy of Band Councils, devolve more administration to them and, in this way, cut costs for the Canadian state (Hawthorn and Canada: Indian Affairs Branch, 1966).Footnote ¹²

The customary leadership structures that survived these changes now face new kinds of challenges, which are connected to electoral bias. Since the 1990s the settler state has been subject to judicial mandates to consult with Indigenous communities. Electoral bias has coloured the representatives they have seen as legitimate and sought to engage with, which has strengthened the status and practical importance of elected Band Councils.

An observer might suppose that Canada’s policies of “self-government”—which have been rhetorically central to the federal agenda since the 1970s (Abele and Prince, Reference Abele and Prince2006; Nichols, Reference Nichols2019; Weaver, Reference Weaver1981)—would have expanded the scope for customary governance, but this is mostly not the case. Instead, self-government has been pursued through Indian Bands and Band Councils, reinforcing the marginalization of pre-colonial polities and governance structures.Footnote ¹³ In most cases, policies that the federal government describes as “self-government” involve extending the authority of Indian Act Bands (First Nations). The most impactful instruments of self-government are Self Government Agreements. These agreements create sui generis institutions based on multilateral negotiations with Indigenous, federal and provincial/territorial governments. As a negotiated mechanism, the agreements theoretically allow for tremendous innovation. In practice, however, their scope is dictated by federal and provincial/territorial negotiating mandates, which permit greater local autonomy but require significant institutional continuity. Existing agreements dissolve one or more Indian BandsFootnote ¹⁴ and replace them with new institutions that inherit all Band assets and exercise jurisdiction described by the agreement rather than the Indian Act.

The close connection between self-government policies and Indian Bands has generated significant criticism from Indigenous advocates, particularly those associated with the resurgence movement. Resurgence scholars see Canada inappropriately committed to propping up colonially imposed institutions (G. R. Alfred, Reference Alfred2005; Coulthard, Reference Coulthard2014; Diabo, Reference Diabo2017; Simpson, Reference Simpson2011). Advocates of resurgence instead favour the revivification of Indigenous governance traditions and Indigenous law (Coulthard, Reference Coulthard2014; Simpson, Reference Simpson2017; Borrows, Reference Borrows2002; Borrows and Tully, Reference Borrows, Tully and Asch2018; Stark et al., Reference Stark, Craft and Hōkūlani2023). Many resurgence initiatives, including the Unist’ot’en Camp at the heart of the Wet’suwet’en/Coastal Gaslink conflict, are primarily focused on rebuilding and revivifying Indigenous legal, political, cultural and economic traditions to create the foundation for subsequent engagement with the state and settler society (Coulthard, Reference Coulthard2014; Elliott, Reference Elliott2018; McCreary, Reference McCreary2024). In this sense, resurgence is continuous with the unbroken history of Indigenous peoples fighting to preserve their distinctive ways of life (Simpson, Reference Simpson2017).

One consequence of the history surveyed above is that, although vast majority of First Nations communities in Canada use periodic elections to select their leaders, there is considerable variation in how elections are institutionalized (Table 1). According to Indigenous Services Canada, in 2023, 136 First Nations followed election procedures outlined in the Indian Act. These procedures are particularly constraining, as they mandate biennial elections under close federal supervision. Indigenous criticisms of these arrangements (Senate, 2010) led to the First Nations Election Act, 2014 (J. Canada, 2024), which established an alternative regime with four-year terms and an independent oversight process. In total, 80 First Nations have opted into this system. Another 366 First Nations are classified as following custom codes, of which there are two varieties: 175 First Nations once adhered to Indian Act elections and have used the Conversion to Community Election System Policy (I. S. Canada, 2015), introduced in 1988, to draft their own election code, subject to approval by the federal government; the other 191 First Nations (almost all in western Canada) have never formally adhered to Indian Act election rules.Footnote ¹⁵ No centralized data on the latter group exist, and more research is needed. Generally speaking, these communities appear to operate according to electoral procedures that have either evolved organically from quasi-electoral methods introduced by Indian Agents or were established by the Band in the twentieth century without recourse to Indian Act procedures. A small minority of “always custom” First Nations do not rely on periodic leadership elections.Footnote ¹⁶ Finally, 43 First Nations have negotiated 22 “Self Government Agreements” that have created a variety of governance structures (Alcantara and Whitfield, Reference Alcantara and Whitfield2010). No Self Government negotiation has yet produced a non-electoral regime,Footnote ¹⁷ although some establish non-electoral advisory bodies,Footnote ¹⁸ and others structure elections according to traditional clan divisions.Footnote ¹⁹ The Self Government that has gone the furthest in this direction is the Ka:’yu:’k’t’h’/Che:k’tles7et’h’ government, which has established a legislative council in which four seats out of nine seats are directly appointed by the Ha’wiih Advisory Council, which itself operates under Maa’nulth law (Plummer, Reference Plummer2025).

Table 1. Indigenous Leadership Selection in 2023Footnote ²⁰

The relationship between traditional and electoral practices varies widely across First Nations communities, as one might expect given the range of historical trajectories. Some communities, such as the Mississauga at Credit River, shifted their customary approach to an electoral one as early as the 1830s, well before Canada required it (Smith, Reference Smith2013; Walters, Reference Walters1998). For other communities, including the Wet’suwet’en, the two institutions exist in parallel and have long histories of cooperation and conflict. Other communities have attempted to fuse the institutions by electing traditional leaders or by giving the Band Council a mandate to lead the transition to customary self-government, as Kahanawá:ke did in 1982 (G. R. Alfred, Reference Alfred1995: 136). In many cases, state-mandated electoral institutions have been adapted to reflect Indigenous law, institutions and cultures. For example, the Anishinabek community of Mishkeegogamang has creatively re-fashioned its elections to suit Anishinaabe principles (Sieciechowicz, Reference Sieciechowicz, Barber, Leach and Lem2012). The leadup to elections involves iterative consultations between different families to establish the key issues facing the community and the “front-runners” to take office. On voting day, the outcome has been predetermined by this deliberative process; the drama lies in tacitly coordinating the votes to ensure that the victory margins are slim, in accordance with strongly egalitarian norms that insist the leader is not “above” the people.

3. Collective Self-Government and Shared Agency

Democracy is an exercise in collective self-rule and, as such, requires governance institutions that both realize and express our shared agency. Elections can sometimes help realize shared agency by making governance more inclusive (see, for example, Schattschneider, Reference Schattschneider1960), and they can sometimes express shared agency by sending to citizens a signal that they are expected and able to participate in rule together (Booth Chapman, Reference Booth Chapman2022). But some Indigenous people do not see electoral institutions as an appropriate vehicle for their shared agency and participation in rule—and this is reasonable, considering the colonial legacies they embody. When many Indigenous citizens reasonably refuse to participate in elections, customary governments may serve as more appropriate vehicles for collective self-rule.

Philosophers have sometimes been wary of strong holistic conceptions of shared (or collective, or group) agency, which may seem to imply the existence of macrosubjects with intentions and cognitions that are irreducible to those of individuals. Yet, we do sometimes think of ourselves as acting together, and very weak or individualistic accounts of group agency that see it merely as a matter of mutual co-ordination cannot distinguish acting together (for example, walking together) from acting alongside (for example, walking in parallel without bumping into each other). Over the last few decades, Michael Bratman has developed a “planning conception” of shared agency that steers a middle course between overly strong and overly weak accounts (Bratman, Reference Bratman2013). To oversimplify somewhat, Bratman’s account suggests that shared agency occurs when individuals jointly intend to act together and adjust their plans accordingly in response to each other (see Bratman, Reference Bratman2013, chap. 4). Emilee Chapman shows that this conception can help us talk about democratic self-rule without making implausible claims about macrosocial subjects with intentions and mental states (Booth Chapman, Reference Booth Chapman2022: 199–205). Democratic self-rule occurs when each member of a group intends that the group rules themselves collectively and when they adjust their plans reciprocally on the basis of the shared intention. Self-rule therefore depends on the existence of appropriate “participatory intentions” (Kutz, Reference Kutz2000) across most—ideally all—of the community: a widespread desire to take part in practices of self-rule.

Many Indigenous citizens reasonably refuse to form the relevant kind of participatory intentions in relation to elective Band Councils. Consider, for instance, the Haudenosaunee community at Six Nations, which is the most populous First Nation in Canada, or the Algonquins of Barriere Lake. At Six Nations, only a small fraction of the population votes in Band Council elections (Senate, 2010: 18), and most boycott them in ongoing protest against the coercive imposition of electoral government by Canada in 1924 (Titley, Reference Titley1992: 110–34). At Barriere Lake, the Canadian government called the first-ever election in the community in 2014, which was intended to replace the customary council, and only five people participated in the process, which was widely seen as illegitimate by the community (Pasternak, Reference Pasternak2017: 190–213).Footnote ²¹ In cases such as these, popular customary governments may better realize and express a community’s shared agency.

Even outside of the most dramatic examples of violent imposition, many Indigenous people object to Band Council elections because they rightly link them to unjust colonial policies. Most Band Councils were imposed on First Nations between the 1890s and 1960s as part of an assimilationist project and were designed to undermine, rather than facilitate, Indigenous self-rule in at least four ways. First, elections were intended to aid in displacing leaders who opposed government policies. When Band Council elections were first introduced into Canadian law, legislators were motivated by a desire to bypass traditional leaders who opposed efforts to parcel out reserves as private property (Milloy, Reference Milloy1992: 150–51). Second, Canadian officials hoped that participation Band Council elections would “educate” Indigenous peoples to prepare them for assimilation into settler institutions (Nichols, Reference Nichols2019: 249). Third, officials hoped that election would erode social support for customary governments and cultural practices by creating attractive career opportunities for ambitious community members (see, for example, Sproat, Reference Sproat1879). Fourth, male-franchise elections (until 1951) disempowered of Indigenous women from traditional leadership roles, for example, as Haudenosaunee Clan Mothers or on Anishinaabe Women’s councils.

Indigenous critics also see evidence of colonialism in the poor design and institutional weakness of Band Councils. Perhaps the most common criticism of Indian Act elections, both historically and in the present, is that biannual elections in small communities produce factionalism, divide communities and prevent leaders from gaining adequate experience representing their people (Senate, 2010; Walls, Reference Walls2010). Add to this the structural subordination of elected leaders to the Indian Department (which historically could and often did remove leaders it opposed) (Satzewich and Mahood, Reference Satzewich and Mahood1994) and their extremely limited formal powers (Abele, Reference Abele2007; Milloy, Reference Milloy2008), and it is easy to see why the institution fails to gain profound allegiance.

Given the association of electoral institutions with colonial violence, assimilation and patriarchy, it is easy to see why customary Indigenous governments might be better able to realize and express Indigenous peoples shared agency and shared participation in self-rule. Traditional structures often have deep cultural legitimacy; their representative claims rest on their “deeper roots” in the community and the ties of tradition (Saward, Reference Saward2009: 10). Support for traditional Indigenous governments also expresses a widespread desire to reject the genocidal program of assimilation and instead participate in the collective survival of Indigenous nations.Footnote ²² These factors make customary governments attractive vehicles for democratic collective agency, even before we consider their specific democratic mechanisms.Footnote ²³

A skeptic may argue that, as a matter of fact, elections are now normally endorsed by Indigenous communities. To our knowledge, this question has never been systematically investigated by pollsters, although we do not doubt it is true in many cases. One piece of evidence in support of this claim are the high levels of voter turnout in some Band Council elections. But people can have many reasons for voting that do not reflect an endorsement of the system itself; background conditions of domination render participation an unclear signal of consent by constraining the range of viable alternatives (Luoma, Reference Luoma2022: 36). By creating pathways to establishing non-electoral arrangements, our proposal will allow more accurate assessment of the extent of Indigenous popular support for electoral systems.

Another line of skepticism toward our argument might focus on showing that the refusal to embrace elections is unreasonable because the injustices associated with historical imposition of elections have now passed. After all, many just practices (such as state redistribution to the poor) owe their existence to historical violence (state-building). What matters is that Band Council elections today are plausibly just.

Many Indigenous people would reject the claim that the injustice of election imposition is confined to the past, pointing, for example, to their continuing role in dividing communities and legitimating land dispossession. But even if we imagine these issues are somehow addressed and the injustices really are mostly historical, the objection remains unpersuasive for reasons that Annie Stilz articulates. She compares the situation of colonized nations with that of a woman married to a formerly abusive husband. Even if he is now causing her no harm or injustice, it would be reasonable for her to refuse to participate in the marriage (Stilz, Reference Stilz2015). The thought experiment is, we think, apt and reveals that Indigenous people do not reasonably owe just elections their allegiance given the history of violence and injustice we have documented.

A final objection to our argument might hold that elections in Indigenous communities are worth keeping despite these problems because of their role creating democratic responsiveness. Elections are an important mechanism through which democracies ensure the accountability of leaders to the people. In the absence of elections, the worry goes, leaders will not have sufficient incentives to remain responsive to the will of the community and perhaps especially its weaker members.

This argument implies a trade-off between two factors: participatory intention and responsiveness. We accept that, in some cases, the loss in responsiveness will outweigh the participatory gain. In other cases, however, the reverse will be true. Moreover, some cases may not involve any drop in responsiveness, because customary Indigenous governments have their own mechanisms for inducing this quality.

4. Responsiveness through Exit

In this section and the next, we describe some mechanisms used by customary governments to promote responsiveness. We focus our account on aspects of the Wet’suwet’en bahlat (feasts or potlaches), the Anishinaabe doodem system and the Haudenosaunee Council. Each is continuous with pre-colonial traditions, has been targeted by colonial violence and operates to some degree today. The purpose of our reconstructions is not to provide a comprehensive or authoritative overview of these institutions, but to sketch an underlying logic of responsiveness through voice or exit as a basis for comparison with elections. This approach has the merit of highlighting and making legible the significance of these Indigenous practices to democratic theory; it has the drawback of presenting an incomplete and one-dimensional picture of each institution. Readers should consult our sources for more detailed accounts. When available, we have relied on accounts authored by members of the Indigenous community in question; in other cases, we have preferred sources by authors with long histories of public collaboration with the communities.

Any system of representation involves the distribution of opportunities for influence unequally between leaders and ordinary citizens. The challenge for democrats is to find ways of structuring systems of representation that nonetheless approximate the ideal of political equality by stopping unacceptably large, durable, or harmful kinds of gaps opening between the two groups (Hutton Ferris, Reference Hutton Ferris2024; Mansbridge, Reference Mansbridge, Castiglione and Pollak2019). This requires finding ways to make leaders responsive to people.

Elections can mitigate the inequality inherent in representation by creating responsiveness to citizens practices of “exit” from political groupings (Hirschman, Reference Hirschman1970; Warren, Reference Warren2011; and for “exit” by Indigenous citizens, see Rollo, Reference Rollo2017). Mark Warren identifies two kinds of exit relevant to democratic theory: enabled exit, which secures non-domination and which exists when individuals have the right to exit a relationship or grouping, and institutionalized exit, which serves responsiveness and exists when individual exit decisions are institutionally linked to leaders in a way that evokes responsiveness (Warren, Reference Warren2011). Elections are opportunities for institutionalized exit because they give citizens structural opportunities to sanction leaders by withdrawing their electoral support for them. Leaders, in response, are encouraged to perform anticipatory representation, that is, to try to act in ways they guess their constituents will approve of to avoid electoral sanction (Mansbridge, Reference Mansbridge2003).

Some Indigenous customary practices require leaders to secure the active support from members in ways that are analogous to, but perhaps more robust than, elections. For example, a Wet’suwet’en person seeking a chiefly name must lead their House in hosting a feast (also called bahlat or potlatch) one year after the death of the previous name-holder. By necessity and custom, feasts are expensive events that require considerable material support from others; chiefs cannot “afford a name” without the support of “their mother’s and father’s respective clans, of the spouses of their father’s clan … and of their own spouse’s clan” (Mills, Reference Mills1994: 117–18). House-members therefore have the option of “exiting” the coalition of support for the chief by lending material support to an alternative candidate. Aspirants are correspondingly incentivized to engage in anticipatory representation, building a broad base of support. This enabled and institutionalized exit, as we argue below, is also connected to voice because this capacity to grant or withdraw support is linked to dense practices of co-operation associated with resource gathering and feasting that create opportunities for various kinds of discursive influence over would-be chiefs.

If the bahlat system in some ways resembles an electoral campaign, other Indigenous practices go beyond this—so much so, in fact, that we believe they represent a third sort of exit relevant to democratic theory: empowered exit. This occurs when the right to exit is protected (as in enabled exit) and institutions connect individual exit decisions with representatives in a way that evokes responsiveness (as in institutionalized exit) but exit is also strongly facilitated by social and political institutions that increase its feasibility. When individuals are empowered to exit in this way, it creates greater pressure for responsiveness for leaders, relative to situations when people are merely permitted to exit. It is also likely to better promote non-domination and constrain leaders’ coercive ability.Footnote ²⁴

Historical Anishinaabe societies (and many of their neighbouring societies) had strong mechanisms of enabled exit. These mechanisms ensured what David Graber and David Wengrow (Reference Graeber and Wengrow2022: 132) describe as the “freedom to abandon one’s community, knowing one will be welcomed in faraway lands.” One such mechanism was the Anishinaabe doodem (clan) system. Each Anishinaabe had a doodem, associated with a particular animal, which they inherited from their father (Treuer, Reference Treuer2010: 18).Footnote ²⁵ Anishinaabe who shared a doodem were regarded as close kin and owed strong duties of mutual aid and hospitality (Miller, Reference Miller2010: 39). These doodem networks cut across community and even linguistic lines, ensuring that Anishinabek individuals could expect welcome and support far from their immediate kinship networks. These networks are further extended through spousal relations since marriage between members of the same doodem is forbidden, and hospitality is also owed to children or spouses of doodem-kin (Bohaker, Reference Bohaker2021: 78–85). These networks were always accessible to Anishinaabe, not just enabling but empowering exit.

This freedom of mobility contributed to democratic leadership in Anishinaabe societies. Ogimaag (civil chiefs, usually hereditary) could not rely on coercive authority, instead relying on persuasion for influence (Miller, Reference Miller2010: 36; Simpson, Reference Simpson2011: 206; Treuer, Reference Treuer2010: 22). Non-coercion was a social norm that manifested throughout Anishinaabe culture, including in parenting practices (Miller, Reference Miller2010: 36; Simpson, Reference Simpson2011: 232–36). But it was also enforced by the opportunity of exit. The definition of ogimaag, according to Elder Basil Johnston, is a “he (or she) who counts a number of followers” (Johnston and Johnson, Reference Johnston and Johnson1995: 23). Behaving inappropriately, for example by acting contrary to the interests of the community, would result in a loss of influence and followers, without which one would cease to be an ogimaag (hereditary entitlement notwithstanding). In practice, ogimaag depended heavily on the three village councils, civil councils (usually composed of male family heads), warriors councils and women’s councils, to make decisions and were careful to avoid speaking on behalf of any group who had not given their specific consent on the issue at hand (Miller, Reference Miller2010: 110). Moreover, civil authority was not the only kind of authority recognized, and some issues belonged in the domain of religious authorities (Midewiwin), military authorities, or women (Treuer, Reference Treuer2010: 14). Specific doodem also had particular expertise, often connected to the characteristics of their animals. The result, as Leanne Betasamosake Simpson emphasizes, was a system of leadership that was “diffuse, shared, and emergent arising out of need” (Simpson, Reference Simpson2011: 233). These fluid structures accommodated seasonal variations in the size of Anishinaabe groups, who gathered in the summer and dispersed in the winter. It also contributed to malleable forms of diplomacy, in which larger confederacies could form without infringing the autonomy of local communities, who could withdraw their support at any time (Kugel, Reference Kugel1998: 5–6; Miller, Reference Miller2010: 5; White, Reference White2010).

Over time, the Anishinabek’s freedom as mobilityFootnote ²⁶ has been severely weakened by settler colonialism, which has concentrated them on reserves, disrupted doodem networks and imbricated Anishinaabe people within a capitalist economy that creates significant financial and material barriers to relocation (Bohaker, Reference Bohaker2021: 170–98). This has, in turn, weakened the ability of Anishinaabe dissenters to exit local communities and correspondingly reduced the need for anticipatory representation by political leaders. Modern Anishinaabe Indian Bands generally define membership according to rules regarding Indian status and residency, rather then doodem affiliation.

Nevertheless, the tradition of kin-based community mobility is a promising site for democratic innovation and renewal. Doodem identities remain important for many Anishinabek, and there are contemporary efforts to revitalize traditional Anishinaabe understandings of belonging (Bohaker, Reference Bohaker2021: 200; Lee, Reference Lee2015; Mills, Reference Mills2020). To date, these efforts have been primarily motivated by the need to address problems of membership generated by the Indian Status regime.Footnote ²⁷ It seems to us, however, that an additional benefit of restoring doodem-based membership could be to empower political exit and thereby encourage democratic responsiveness in Anishinaabe communities. The recently signed Anishinaabe Nation Agreement, which creates a federation of First Nations (former Indian Bands) who control their own membership codes, could be a fertile site for experimentation in this direction (Canada, 2022).

While it is beyond our scope to describe the path these experiments should take, a few speculative remarks can illustrate the feasibility of the proposal. Recall that, historically, empowered exit was supported by strong reciprocal duties of hospitality, which allowed individuals to leave their community with assurances that their basic needs would be met and that they would be politically welcomed. A modern revival of this tradition might place particular emphasis on securing the material foundations of mobility for all Anishinabek. For example, geographically dispersed doodem-based organizations could directly subsize the costs of moving for members. More expansively, doodem networks could control and allocate certain resources, such as housing, childcare and even employment opportunities, lowering the logistical and financial obstacles to mobility. These reforms would be most effective if they were accompanied by a broader set of changes that improved Anishinaabe peoples’ economic circumstances and also, crucially, increased their access to their traditional territories beyond reserves.Footnote ²⁸ These material functions would complement the traditional function of welcoming doodem-kin into the community and ensuring their social belonging. Such a network would empower political exit from local Anishinaabe communities by reducing the material costs of mobility, potentially enhancing responsiveness.

Indigenous exit-based responsiveness mechanisms such as the Anishinaabe system of kin-based empowered mobility may be just as powerful at inducing responsiveness from leaders as democratic elections.Footnote ²⁹ The influence of these mechanisms on leaders has not been studied empirically, and so we can only make informed guesses about the relative influence of elections versus kin-based institutions of empowered mobility. There is no a priori reason to expect either practice of exit to be better at generating democratic responsiveness. Empirically, elections have some widespread and uncontentious limitations in this respect, such as when citizens in the safe seats created in many electoral systems cannot realistically affect outcomes. There is also a substantial amount of debate amongst political scientists about just how good elections are at inducing exit-based responsiveness (see, for example, Achen and Bartels, Reference Achen and Bartels2016). One growing body of evidence appears to suggest that elected governments change course when the wealthy want them to but ignore the opinions of everyone else (Gilens, Reference Gilens2012; Lupu and Warner, Reference Lupu and Warner2022). While the jury is still out on the interpretation of these findings (Elkjær and Iversen, Reference Elkjær and Iversen2020), the debate shows that we cannot simply assume that elections are any good at generating democratic responsiveness to citizens exit. The mobility-based mechanisms customary governments may be just as powerful, if not more so.

5. Responsiveness through Voice

Democratic systems of representation need to involve responsiveness to citizen voice. As Hirschman observes, exit is not always practically feasible or desirable in either politics or economics. Consider, for instance, the case of a commuter who must use a particular railway line, a blue-collar worker in a two-party system who knows that only one party ever favours pro-labour policies or a citizen who cannot (except at great risk or cost) migrate away from a polity whose government is about to make a decision that is fundamentally unjust and threatens their basic interests. In such cases it becomes crucial for people to use their voice, for instance, by writing in protest to the railway company or to a legislator.

Elections may facilitate responsiveness to voice. The threat of losing office to unsatisfied constituents may help encourage representatives to listen to citizens voices and perform a kind of deliberative responsiveness to them (Hutton Ferris, Reference Hutton Ferris2025), including in public meetings, private communications and constituency services. A representative system with electorally vulnerable governments at the centre may also encourage more diffuse kinds of responsiveness to voice, as empowered representatives in formal institutions serve as conduits for discourses emerging and circulating in civil society (Habermas, Reference Habermas1996).

One alternative voice-based responsiveness mechanism is the Wet’suwet’en practice of communal co-operation, competition and resource gathering associated with community feasts. According to Michael Luoma (Reference Luoma2022: 34), “[a]ll Wet’suwet’en members are able to participate meaningfully in public deliberation and are integral to the continued power of chiefs and the success of clan-based political decisions.” He attributes this inclusiveness partly to Wet’suwet’en political culture, which has “robust norms of horizontal contestation, accountability and open public deliberation” and partly to the small size of the Wet’suwet’en community, which facilitates regular practices of face-to-face interaction that mean that “Wet’suwet’en citizens may be able to participate more effectively in meaningful public deliberation and political contestation than citizens of large and anonymous societies practicing representative democracy with fixed election cycles” (34). Small size alone, however, is not sufficient to secure inclusion; it is buttressed by the dense patterns of interaction, mutual support and reciprocity associated with the bahlat/potlatch system. Frequent potlatching requires community members to constantly loan each other money and provide other forms of material support, and this both requires and sustains discourse that keeps chiefs and would-be chiefs informed about and attentive to the opinions, desires and understandings of community members. While it is mostly chiefs who speak in the feasts themselves (Mills, Reference Mills1994: 113), the bahlat sits at the centre of and activates dense and more inclusive practices of consultation and debate. In this way, Wet’suwet’en governance encourages the kind of iterative back and forth talk necessary for democratic representation.Footnote ³⁰

Another kind of responsiveness to voice occurs through the customary governance of the Haudenosaunee Confederacy. The Haudenosaunee Confederacy was established by a heroic Peacemaker, who joined five warring nations (the Mohawk, Oneida, Onondaga, Cayuga and Seneca) together under the constitution of the Kayaneroko:wa (the Great Law of Peace).Footnote ³¹ The regime established by the Kayaneroko:wa is both deliberative and inclusive. It is structured by a series of nested and parallel councils, at the apex of which is the Confederacy Council, which is composed of 50 Rotiyanehson (usually translated “Chiefs,” singular: Royaner) representing the various clans of the five nations, each of whom is selected and can be removed by a Yakoyaner (Clan Mother).

The smallest and least formal units in the Haudenosaunee system of council democracy are the men’s (or “warriors”) and women’s council for each clan (Hill, Reference Hill2017: 61–62). These can be convened whenever the group identifies a need, including to discuss the appointment of a Yakoyaner or Royaner, and their decisions or recommendations are sometimes raised by the clan’s Royaner for discussion in the Confederacy Council. Within the Confederacy Council itself, deliberation is cumulative process that moves from smaller to larger groupings. It begins with small groups of two to four Rotiyanehson within a nation, usually those who share a clan (for example, the Bear Clan of the Mohawk Nation). These small groups must reach consensus on the issue before engaging with Rotiyanehson from other clans within their nation, who will then deliberate until national consensus is reached (K. P. Williams, Reference Williams2018: 281–82). At this point, the deliberation proceeds to the inter-nation level of the Confederacy. Responsibility for introducing topics at the Confederacy level falls to the two Elder Brother nations, the Mohawks and Seneca. First, these Elder Brothers with deliberate with one another. Once they reach consensus, deliberations expand “across the fire” to include the Younger Brother Nations, which originally were the Oneida and Cayuga and now include the Tuscarora as well. The Younger Brothers will consider the matter in their own clan and nation groupings before giving their assent or passing the matter back across the fire for additional discussion. When the Younger Brothers and Elder Brothers are aligned, the matter passes to the Onondaga, who serve as Fire Keepers and are responsible for convening the assembly and ensuring that proper procedure is followed. The Onondaga either give their assent or send suggestions back for further discussion. All Confederacy Council deliberations are observed by Yakoyaner, who will ensure that their Royaner rightly represents his clan on pain of removal. Individual Royaner frequently consult with their Yakoyaner, and informal conversations between Rotiyanehson outside the longhouse plays an important role in achieving consensus. As with the systems of council democracy advocated by Hannah Arendt and some nineteenth century socialists, this system functions to draw the voice of every individual clan member up into the highest echelons of decision-making in a distinctively deliberative way (Arendt, Reference Arendt1963; and on council democracy, see Muldoon, Reference Muldoon2018, Reference Muldoon2020).

Consensus decision-making is an important manifestation of the Great Law’s emphasis of Sken:nen (Peace) and is promoted through a strong emphasis on respectful deliberation. Each council begins with a Kanonhweratonhsera (literally, words that come before all else), a thanksgiving address that expresses gratitude for each aspect of creation. Each section of the address concludes with the affirmation “and so our minds are one,” to which all participants respond with their assent (K. P. Williams, Reference Williams2018: 281), initiating a procedure focused on unity. Deliberations themselves are guided by a strong sense of decorum. Participants are expected to be clear-minded, not influenced by their passions, and to speak from the impersonal perspective of a representative or delegate of a wider community. Speakers generally begin by complementing the previous speaker and apologizing in advance for anything they say that might be understood as “sharp” or injurious. When speaker finishes, there is a long pause during which they can reflect and ensure they have not omitted anything important; when the next speaker replies, they begin by restating what was originally said to ensure mutual understanding and allow misunderstandings to be corrected. When consensus is elusive, strategies for resolving disagreements include concentrating on principles rather than details (which can be delegated), breaking matters into parts and saving the most difficult for later or agreeing to defer to local community decisions instead of deciding collectively where necessary (K. P. Williams, Reference Williams2018: 281–94).

Customary governments such as those described above might secure deliberative democratic inclusion better than possible electoral counterparts. Even though democratic theorists are likely to concede that elections in large complex societies help encourage diffuse responsiveness across systems of representation to the voices of ordinary people, they are rarely very optimistic about the quality of this responsiveness. They are more likely to emphasize instead the way elections encourage representatives to adopt a strategic orientation to constituents (Disch, Reference Disch2011) and to lament the way this distorts processes of communicative influence from civil society to the state (Habermas, Reference Habermas1996). There is currently a lot of debate about whether responsiveness processes are so broken that elections for top legislative office should be replaced by processes of random selection (Guerrero, Reference Guerrero2024; Landemore, Reference Landemore2020). This alternative is sometimes defended with the help of an argument from partisan rigidity (see Hutton Ferris, Reference Hutton Ferris2023) that claims, like some Haudenosaunee authors (such as K. P. Williams, Reference Williams2018: 23–24), that elections breed destructive forms of division and that non-elected leaders can engage in unusually fair-minded and co-operative truth-seeking. While responsiveness to voice in electoral systems often looks questionable, unreliable, partial, or diffuse, some Indigenous customary practices present attractive alternative models.Footnote ³²

6. The Need for Reform

Our defense of the democratic merits of customary governments rests on three arguments. Most obviously, customary governance may be a better vehicle for self-determination in places where people reasonably fail to affirm Band Councils elections as institutions adequately reflective of their group’s shared agency (see also Luoma, Reference Luoma2022). Indigenous citizens may prefer to engage with non-leaders even if they are less responsive than elected counterparts. But, second, Indigenous responsiveness mechanisms based on mobility might be just as valuable for securing responsiveness as elections. Specifically, some Indigenous societies institutionalize exit opportunities that are not only formally available to citizens (enabled exit) and induce responsiveness from leaders (institutionalized exit) but are strongly facilitated by kinship structures that crosscut local political units and make mobility a real and often attractive option for citizens (a situation we have called empowered exit). Finally, many communities are also governed as genuinely deliberative democracies that secure high-quality deliberative responsiveness to the voice of ordinary people.

Despite this, it is presently impractical for Indigenous communities to have non-electoral governance practices recognized by the Canadian state. The precise nature of the administrative and legal obstacles varies depending on circumstances,Footnote ³³ but any First Nation pursuing this path would face possible judicial review for compliance with both Canadian administrative law and constitutional law.Footnote ³⁴ Previous Supreme Court jurisprudence has found a Charter right for off-reserve band members to vote (Corbiere v. Canada 1999); it is difficult to anticipate how this principle—and more generally the prevailing judicial understanding of political equality rights—would be reconciled with a non-electoral regime that violated one-person, one-vote. Thus, any First Nations seeking to adopt a non-electoral regime would be relying on novel constitutional arguments.Footnote ³⁵ Even if it were to succeed, the expense and uncertainty creates a significant barrier to innovation.

To address these obstacles, Canada should loosen its constraints on Indigenous government experimentation. Specifically, we recommend five policy changes. None require Canada to alter its preference for democratic leadership within Indigenous communities; they only require that Canada adopt a less parochial conception of democracy. They are that (1) the conversion to community election system policy should be replaced by a conversion to Indigenous democratic system policy; (2) self-government negotiators should encourage communities to devise innovative democratic solutions and include language in self-government agreements to facilitate this; (3) Parliament should encourage courts to take a broad view of how Indigenous political systems can protect Indigenous individuals’ democratic rights; (4) Canada should create a clear process for communities with parallel customary and electoral leadership structures to enter an internal process of institutional reconciliation, without a predetermined (that is, electoral) outcome; and (5) Canada should establish a freestanding Indigenous law tribunal with jurisdiction over issues of political representation within Indigenous communities.

The final proposal is perhaps the most important. As with any government, Indigenous customary governments can break down or become captured by powerful individuals or factions within the community. The tribunal would be to provide Indigenous citizens with a recourse in such cases without needing to rely on Canadian law. Such a tribunal would be a boon to the ongoing and important but still limited efforts to create additional space for Indigenous law within Canada’s legal system (Borrows, Reference Borrows2010; Grammond, Reference Grammond2022).

These reforms would be pro-democratic in the sense that they would give Indigenous communities greater control over (some aspects of) their own constitutional regime. They would also have the salutary effect of stimulating democratic experimentation and innovation. These innovations, in turn, will help to advance global understandings of democracy, illuminating new paths and exposing new questions. However, these paths remain closed so long as the colonial state remains wedded to policies that discourage non-electoral models. A first step, therefore, is to break the bounds of electoral bias and open the field for Indigenous democratic innovation.