Is Electoral Exclusion Ever Legitimate?

This article endorses the premise that not all electoral exclusion is illegitimate or undemocratic. This section will briefly justify this premise by looking at the role of electoral rights.

In democratic contexts, elections stand in complex relations to other institutions and political arrangements that confer meaning to the very electoral process. Indeed, the democratic character of elections is determined by that of the broader institutional setting within which they take place (Reference BoninBonin 2017). Different democracies will have different voting procedures, but there are elements, tangential to formal electoral rights, that characterize each democratic context. I am referring to the bundle of political and civil rights that make possible the process of formation of political will and opinion, such as the right to free speech and freedom of association, the right to peaceful political contestation, the right to assemble and so forth. This corollary of rights exists to ensure people's freedom, a degree of government accountability, as well as people's opportunity to develop as political actors.^{Footnote 3} Democracies formally recognize these rights to anyone, regardless of their citizenship, residence status, or social ties to the relevant political community, in accordance with the Universal Declaration of Human Rights.^{Footnote 4}

Electoral rights are treated differently: most often they are reserved for adult citizens or conditional on one or several requirements that would-be voters have to fulfill. This is understandable as electoral rights do not exhaust the array of ways and means of people's political expression and exclusion from a country's electoral rights does not amount to not having political rights. Formal access to electoral rights in democratic countries does not guarantee greater political freedom: regardless of their ability to vote, people have the same right to express their political views or to peacefully contest a government as anyone else. Instead, having access to electoral rights entails having a greater degree of political power. This discredits a possible argument for the unconditional extension of electoral rights to migrants and other disenfranchised categories: the argument that the disenfranchised are vulnerable to fundamental violations of their political freedom and, thus, to violations of their human rights.^{Footnote 5} These rights, at least in theory, are ensured by the democratic guarantees listed above.

Lastly, it is worth mentioning that theories of global democracy, open border regimes, and the range of positions that orbit around these normative ideals tend to be against the very possibility that differential political rights could be considered democratic.^{Footnote 6} As appealing as these positions are, especially in light of timely global challenges such as climate change, global health, or forced migration, I chose to focus on standards of democratic inclusivity that can be implemented within the currently existing states, rather than in a global democracy. Far from a defense of any existing border or a denial that borders operate in unjust ways, this premise aims to acknowledge that the internal pluralism that characterizes democracies needs to be understood as coexistent with an external pluralism of polities, each with their own particular history, political identity, and democratic culture.

Critique of Functional Demos Views

The problem of the definition of the demos is the challenge of finding a democratic criterion to establish who should be part of a democratic constituency (Reference MillerMiller 2009; Reference WhelanWhelan 1983). There are, roughly, two distinct ways of approaching the boundary problem: one is to focus on whose interests are affected by political decisions; the other is to focus on the coercive power exercised by political institutions. The common rationale of these two approaches is to look at the relationship connecting people and political power and to argue that legitimacy calls for this relationship to entail a feedback loop, involving deliberation, participation, authorization, and accountability. However, they differ in that one places emphasis on evenly protecting citizens’ interests (Reference ElsterElster 2013; Reference GoodinGoodin 2007; Reference GouldGould 2004; Reference Schmitter and AxelSchmitter 1997; Reference ShapiroShapiro 2006; Reference Shapiro and CasianoShapiro and Hacker-Cordón 1999; Reference YoungYoung 2002), while the other centers on the need for a justification of the coercive power of the state (Reference AbizadehAbizadeh 2008; Reference BeckmanBeckman 2006; Reference Cohen and JonCohen 1998; Reference Habermas and WilliamHabermas 2015; Reference MillerMiller 2009; Reference Owen, Gideon, Phillip and JonathanOwen 2010). In other words, they prescribe that respectively “all affected interests” or “all coerced” should have voting rights.

It has been noted that, when functional demos criteria are used to establish how the demos should be constituted, they encounter two problems. The first problem is the impossibility of formulating policies that take either principle seriously without endorsing a global or unbounded demos.^{Footnote 7} Along these lines, Theuns argues that functional demos principles exacerbate rather than solve the problem of the constitution of the demos: if we appoint a different demos for each politically charged decision-making procedure, whichever the functional demos standard, there will be disagreement over who adequately meets the chosen standard.^{Footnote 8} If this disagreement is to be solved democratically, the demos problem is reproduced at each instance. Further, adopting functional demos standards would result in “floating demoi”: demoi that are “radically indeterminate” in their composition, because the affected or subjected population might vary for each decision (Reference TheunsTheuns 2021: 29), with grave procedural implications. For instance, it would be impossible for individuals to know in advance who shares the same set of rights and obligations as they do, as well as which institutions could legally exercise coercive power over them. In addition, floating demoi render collective self-rule impossible, discarding one of the grounding values of democratic governance.

The second and conceptually weightier problem is that these criteria try to envision an ideally democratic state (or an ideally composed demos), while democratic governance itself is always historically grounded and non-ideal (Reference TheunsTheuns 2021). For Theuns, democratic legitimacy is realized not through conformity with an abstract ideal, but by historical reference to the illegitimate state of affairs that democratic states replace. As he puts it “the boundary problem is a category mistake—there simply is no ‘initial act of constituting the demos’; demoi are made from existing polities and are justified comparatively” (2021: 37; emphasis added). On this basis, Theuns labels democratic legitimacy a comparative^{Footnote 9} rather than an ideal value. As he concludes, legitimacy is brought about by a demand for more representative government by a bounded people and any democratic reform achieves it whenever it expands the electorate among those subjected to the relevant legal order.^{Footnote 10}

I believe that Theuns's analysis helps resolve some of the confusion that characterizes demos-related debates. It does so by making explicit that, for the sake of democratic legitimacy, it is not necessary for any demos-altering action to adhere to any purportedly perfect democratic standard. While I appreciate his arguments and I agree with his conclusions, I want to say more on the position, which I share, that democratic legitimacy should not be held to ideal standards, at least insofar as inclusion is concerned. The reason Theuns provides for this is that democratic procedures are historically grounded and problem oriented, therefore non-ideal. It is difficult to argue with the factuality of this claim, but the objection could still be raised that the empirical lack of perfect democracy, or a perfect democratizing process, should not deter us from theorizing what the perfect democracy would look like. Indeed, by analogy, virtually no existing society can be said to perfectly uphold ideal standards of justice, nor is there any shared expectation that such society will soon come into existence. Nonetheless, ideal-theoretical discussions of justice are pervasive in political philosophy. Although ideal accounts of justice face various criticisms,^{Footnote 11} they are valued for clarifying the ultimate state of affairs we should strive to achieve and have significantly influenced all post-Rawls normative political theory. Why, then, should legitimacy be treated any differently?

To address this objection, I want to suggest the narrower, but hopefully stronger claim that democratic legitimacy should be held to minimalistic (or comparative) standards at least in matters of inclusion.^{Footnote 12} This has the benefit of not excluding the possibility that other aspects of democratic legitimacy should be held to ideal standards, while allowing me to take seriously the theoretical and pragmatic drawbacks of conceiving of democratic inclusion in ideal terms.

The space of the current article is too limited to expound this issue in a satisfactory way, however there is at least one strong argument to be made against holding legitimate inclusion to standards that prove to be too vague and impractical. This has to do with the type of normative strength that the notion of legitimacy exhibits, what is sometimes called the “recognitional role of legitimacy.” This refers to legitimacy's role of rendering a political entity “a member in good standing of the state system” (Reference NaticchiaNaticchia 1999: 242). This entails, for the legitimate political entity, “the right to territorial integrity, the right to non-interference in its internal affairs, the power to make treaties, the right to make just war, and the right . . . to enforce legal rules on those within its territory” (Reference NaticchiaNaticchia 1999: 242). In sum, within the state system, legitimacy functions as a green light for international cooperation, mutual support, and ultimately international peace.^{Footnote 13} Similarly, within the state, legitimacy provides pre-emptive reasons to obey laws, pay taxes or serve on a jury when summoned.^{Footnote 14} The high stakes involved in ascribing or denying the overall legitimacy of a polity should provide a strong incentive to formulate attainable and straightforward criteria, particularly in matters of inclusion, where ideal criteria—such as functional demos views—prove approximative at best.

Upon acknowledging this recognitional role of legitimacy, and with it the need for a clear and attainable threshold, one can also appreciate its normative desirability. When the only standard to judge legitimacy in inclusion is unclear, openly unattainable, or even self-contradictory, the risk is to simply accept the status quo and provide ad hoc justifications of why it fits with a certain interpretation of the chosen standard. This results in the watering down of democratic legitimacy as an important concept and tool of critique for when the state's mechanism of democratic accountability needs reforming to retain its authority.

In sum, my claim is that legitimacy in inclusion needs to be held to a set of standards that do not coincide with those appropriate for ideal justice. The reasoning is that legitimacy is a crucial requirement for democratic states and that it cannot be held to utopic standards without losing its strength and meaning. Because inclusion—for practical as well as theoretical reasons—cannot meet the standards of either functional demos view, the latter should be replaced by attainable guidelines. In light of this, the following sections will attempt to provide an actionable benchmark against which to judge electoral inclusion.

The Permanent Disenfranchisement Condition

Considering the above, I undertake a different challenge from the one that traditionally characterizes the literature on electoral rights allocation: instead of thinking about the best possible distribution of electoral rights in a well-functioning democracy, I will spell out what I take to be unviable distributions of electoral rights. This choice stems from the conviction that, while there is no universally ideal distribution of electoral rights, it is possible to think of forms of electoral exclusion that are at odds with democracy and the basic democratic principle of equal respect for people as decision-makers. These, I argue, are those which lead to involuntary permanent disenfranchisement.

For my discussion, it is important to note that, along with the default reference to functional demos views, it is common practice to cover electoral rights through discussions on citizenship. This means that scholars often speak of access to naturalization instead of access to the right to vote (Reference CarensCarens 2013; Reference De Schutter and LeaDe Schutter and Ypi 2015; Reference MillerMiller 2016; Reference WalzerWalzer 2010). This is the case because it is common practice to attach voting rights to citizenship status. Since this pairing is not a necessity, neither theoretically nor pragmatically, I distinguish between access to citizenship and to electoral rights, and focus on the latter.^{Footnote 15}

I argue that, for them to not be undemocratic, electoral regulations should not lead to any resident within the relevant territorial jurisdiction being permanently deprived of voting rights against their will. I call this the “permanent disenfranchisement condition.” The advantage of this condition as an alternative to functional demos views is threefold: it (1) clearly expresses a standard that is within reach for democratic states, (2) is not conducive to (nor does it stem from) a flattening of legitimacy as a normative standard, and (3) targets involuntary permanent disenfranchisement directly, without creating the conditions for questionable under- or over-inclusion.

In sum, this condition emphasizes residence, willingness to participate and a temporal limit on voting restrictions as means of ensuring the suitability of norms of electoral inclusion. At first glance, this suggestion may appear to be essentially a version of either one of the functional demos views, because residence could be construed as a proxy for relevant affectedness or “subjectedness.” However, no plausible account of electoral inclusion holds that all residents within a government's jurisdiction should be granted electoral rights. Minors, temporary residents, or people who took up residence one day prior to the relevant ballot can justifiably be excluded from electoral participation. Certainly, it seems implausible that a state's democratic legitimacy should hinge on its success in including such residents in the electoral process. Subjectedness and affectedness do little explanatory work in differentiating residents’ access to electoral rights.^{Footnote 16}

Rather, residence provides a good baseline for reasons that might be understood as a form of continued and relevant subjectedness/affectedness or, in my view, as signifying the individuals’ choice or necessity to share territorial contiguity, terms of cooperation, and institutional setting with one another, which translate into membership in the political community and, therefore, need for equal electoral rights.

Along with residence, the temporal prescription does the bulk of the normative work for the permanent disenfranchisement condition. The idea behind this prescription is that an appropriate amount of time is all that is needed to put anyone who is willing to participate in elections in the right conditions to do so. Aside from residence and time, there should be no hard barrier for an aspiring voter to access electoral rights: be that nationality, perceived mental fitness or any other identity-defining trait. This translates the core democratic idea of equal respect for people as decision-makers into a basic standard for electoral regulations, and it inhibits the formation of the marginalized identity of the intrinsically, unchangeably unfit to vote.

I consider permanent disenfranchisement to be undemocratic if and only if it is involuntary. Should a person decide to never exercise their electoral rights or never display an interest in acquiring electoral rights in the first place, this would fall outside of the scope of the direct responsibility of the state. Similarly, should a person decide not to meet whichever attainable requirement the government establishes for electoral access, this would also not be the direct responsibility of the state. I speak of direct responsibility because, of course, political disengagement tends to not exist in a vacuum, instead it often expresses widespread discontent with political outcomes and distrust of democratic institutions, which do have the responsibility to react to low turnouts. In fact, when abstentionism becomes prevalent, issues of democratic legitimacy may also arise, but this is a different problem than the one at stake, which is the state's potential to actively exclude specific individuals from electoral participation against their will.

To an extent, the criterion I suggest might seem uncontroversial. Of course, the criterion entails that voting rights cannot be denied on the basis of most identity-defining (or socially salient) characteristics, such as gender, religious affiliation, race, sexual orientation, and so forth. But as intuitive as the criterion might seem, there are two categories of permanently disenfranchised individuals whose condition is widely accepted, if not prescribed, by the relevant literature: I am referring to denizens and people who are not considered mentally fit enough to vote.^{Footnote 17}

The case of denizens is interesting because, given that the electoral rights of migrants are typically discussed through access to naturalization (see above), most scholars argue for the necessity to allow migrants to take up citizenship after some time. What this tends to disregard is that immigrants might have strong and valid reasons not to naturalize in the host country. For instance, it might be the case that to naturalize, the migrant would have to give up their former citizenship, causing them to forfeit hereditary titles, the right to own land in the country of origin, or simply entailing a serious rift in the person's identity. In such cases, naturalization is not a viable and free choice. This means that states that mandate citizenship as a condition for electoral rights do not meet the permanent disenfranchisement condition if they do not feature different provisions for those whose naturalization would cost the loss of a native citizenship.^{Footnote 18}

Resistance to the mentally ill or disabled accessing electoral rights is usually motivated by their purported inability to make well-informed political choices due to their mental conditions. Historically, the very same argument has been used to exclude on the basis of census, race, and gender; but regardless of its problematic record, this position is so widely held that it grants serious consideration. Methods to assess whether individuals are mentally fit vary widely in different countries and many of their rationales are sometimes questionable, but for the sake of the argument, let us just assume they correlate to some degree with individuals’ actual political competence. In this case, it is useful to reflect on whether individuals’ actual political competence should impact their right to partake in elections.

Ottonelli has shown how expecting equal electoral access to be responsive to people's actual competence as political decision-makers is implausible and leads to a conflation of justice and democratic legitimacy with problematic normative consequences (Reference OttonelliOttonelli 2012). She begins by positing that equal respect for people as competent decision-makers is substantially distinctive of democratic rule, unlike mere respect for people as individuals with interests and human dignity, which democratic and undemocratic political organization alike may endorse.

She also notes that people's actual competence as decision-makers cannot, in seriousness, be considered equal across the electorate. As she puts it:

To the problem of the implausibility of this assumption, she argues, we should add that it stands in the way of redressing the mentioned differentiating factors, by positing the improvement of people's political competence as broadly unnecessary. Her suggestion is to disentangle the idea of equal respect from any actual expectations about individuals’ equal competence, thus to understand the principle of equal respect as one of “respectful treatment, rather than a requirement of sincere responsiveness to competence” (Reference OttonelliOttonelli 2012: 211). The reason to treat people as if they were equally politically competent is to be found, for Ottonelli, in the performative function of rejecting “hierarchies of status entrenched in political institutions,” even if this comes at the cost of suboptimal political outcomes in terms of efficiency or substantive justice (2012: 212).

Ottonelli does not use her arguments to advocate for (or against) mental illnesses and disabilities to be irrelevant for people's access to electoral rights, but those cases do constitute a very fitting example of the stigmatization and institutional hierarchization that her reading of the principle of equal respect aims to defeat.

If one is persuaded by Ottonelli's argument, one could still claim that opting for a minimal standard of political competence does not entail conceiving of all voters as actually equally competent, but merely as competent enough. For one thing, Ottonelli is clear that citizens are to be treated as equally competent “no matter what their actual skills or competence level,” which seems to rule out the notion of “competent enough” as implementable in electoral inclusion (2012: 202). However, more self-referentially, I note that the condition of permanent disenfranchisement is formulated in a such a way that it keeps electoral participation among those for whom it is at least somewhat meaningful, by allowing voluntary permanent disenfranchisement. When the mental illness or disability is in fact so severe that political participation becomes senseless for the individual, this would be conveyed by their disinterest in participating. In any other case, whenever the individual expresses a will to participate in the public sphere of the relevant political community, any hard institutional barrier standing in their way constitutes a democratic failure which can hardly be justified through reference to the individuals’ alleged insufficient competence.

I acknowledge that a convincing case for the removal of any restriction on the basis of mental fitness may require more argumentation. For instance, it might have to address practical concerns such as the possibility of manipulation of voters with cognitive impairments. The point stands, I believe, that such restrictions express problematic, undemocratic attitudes toward an already stigmatized segment of the population. Indeed, it is worth reiterating that

This should place a heavier burden of proof on the curtailment of these electoral rights rather than on their expansion. Practical concerns may arise in response to any expansion of the electorate, but permanent disenfranchisement hardly seems like a justifiable solution, as it inevitably fails to treat political agents with equal respect.

Contextual Considerations

Democratic polities came about in different social, economic, and cultural contexts, in a variety of ways,^{Footnote 19} and their electoral regulations are one of the possible expressions of collective political identity, as they have the potential to emphasize this or that desideratum of optimal political participation.^{Footnote 20} My current proposal is that, past the minimal threshold of permanent disenfranchisement, any curtailment of electoral rights should be justifiable through the specificities of the relevant democratic culture. I call this the democratic ethos proviso.^{Footnote 21} This is a softer normative recommendation: I argue that democratic polities have good reasons to comply with the democratic ethos proviso, but its breach does not compromise a government's democratic legitimacy. This proviso points to a preferred mode of justification for context-specific electoral regulations, but it refrains from labeling any one type of requirement as universally permissible or forbidden. Provided the permanent disenfranchisement condition is met, the democratic ethos proviso encourages an endogenous approach to voting requirements. In this sense, it is an attempt at complementing my universalizable claims about permanent disenfranchisement with contextual considerations.

The main argument I bring in support of this proviso is straightforward: there is value in individual democratic cultures and the molding of political interactions is an appropriate site for these differences to be expressed. Reference FerraraFerrara (2015) argues that the presence of a democratic ethos that characterizes and evolves with the relevant polity is a necessary guarantee for procedural criteria of democratic legitimacy not to incur in what he terms “trivializing emulation.” For Ferrara, the presence and consistent evolution of a stable democratic culture warrants that requirements of formal democracy are not merely performed, but that they do in fact contribute to a legitimate democratic process. In his view, the democratic ethos does not have a comprehensive, invariable character that is autochthonous to any given political context and that should be exported to democratize other polities. Rather, it exists in multiple versions, each coextensive of the relevant polity's sociocultural underlayer. By way of example, he illustrates this pluralism of democratic ethics through the lens of religious diversity. Bracketing the many points of convergence of religiously differentiated democratic contexts,^{Footnote 22} he highlights two main points of divergence: the tendency to prioritize rights over duties and the role played by political conflict, seen either as a sign of a healthy democratic life or as an undesirable rift within the democratic body. Thus combining these variables, he distinguishes four possible kinds of democratic ethos, leading to liberal democracy, strong republicanism, consociationalist democracy and consensualist democracy, each with full potential for democratic legitimacy. This is one of the typologies of democratic ethics that is possible to draw from the interplay of sociological, cultural, and political factors.^{Footnote 23}

Ferrara does not apply his advocacy of democratic pluralism to the specific case of electoral regulations, but his approach sheds light on how the institutional configuration of different polities that share a democratic orientation is not (nor should it be) accidental, but rather it stems from their differentiated understanding and attachment to democratic values. In my view, electoral regulations are no different. Electoral rights are emblematic means of democratic participation and “the fundamental institution of representative democracy is the electoral system” (Reference KatebKateb 1981). That is, elections are a necessary means to the practice of legitimate constitution of authority, and they embody and establish (mutual) relations of power (Reference Ceva and ValeriaCeva and Ottonelli 2022). The way in which they do so as well as the readings of democracy that they are premised upon, cannot be completely extrapolated by the collectivity that exercises them. In this sense, there can be no ideal picture of electoral inclusion for all polities to strive toward.

A second reason to comply with this proviso concerns the potential for problematic conceptions of the demos, which may be publicly reinforced when the demos is defined not by reference to relevant constitutional, civic, or democratic contexts, but by ethnocultural identification instead.^{Footnote 24} The issue with defining the demos through an ethnocultural lens is that it introduces an underlying hierarchical logic that clashes with the democratic commitment to political equality. Indeed, an ethnoculturally defined demos conflicts with the first condition I outlined, which forbids permanent disenfranchisement. This should be quite clear: ethnocultural affiliations broadly understood are integral parts of people's identity. If inclusion is premised on ethnocultural assimilation, the state places an unreasonable demand on prospective members of the electorate. As mentioned above, this proviso does not establish substantial limits to inclusion criteria; rather, it promotes a specific way of justifying them. As such, this proviso is not tasked with defining ethnocultural criteria of inclusion, such as a formal requirement of full cultural assimilation, because these fall under the scope of the permanent disenfranchisement condition. Instead, it discourages ethnocultural justifications of electoral inclusion criteria. To provide an example of how this proviso may work, it can be used to argue that language proficiency as a condition for electoral access is appropriate if justifiable through the distinctively strong deliberative tradition of the relevant polity but not primarily through reference to objectives of cultural assimilation.^{Footnote 25} This amounts to the difference between conveying the perfectly acceptable message that “you need to learn our language, so that you can self-govern alongside us,” versus the problematic “you need to become like us, in order to self-govern alongside us.”

Closing Remarks

Clearly my proposal leaves room for states to establish a range of time during which newcomers are not entitled to electoral participation by virtue of the fact that residence over time is considered to correlate with better ability to vote on an informed basis. The desired knowledge is not limited to the hard facts of the relevant institutional and political arrangements, but it extends to the domain of social and cultural life that are properly grasped only through prolonged and varied interaction with other members of a political community.

My article suspends judgment on whether prior residence is a good way of ensuring valid electoral participation, rather it allows this requirement because it does not express a judgment on the individual's intrinsic ability or worthiness to participate in elections.^{Footnote 26} This is the case because the temporarily excluded individuals are viewed as simply requiring prolonged exposure to the relevant democratic culture to become contextually politically informed.^{Footnote 27} Certainly, we can easily imagine a politically well-informed immigrant, whose knowledge of institutions and policies is superior to those of the average citizen and who might be significantly more motivated than the average citizen to engage in politics.^{Footnote 28} For this person, it might be unfair to have to wait for years to be able to cast a vote, while their politically uninterested neighbor can vote and run for elections. However, it is a prerogative of a democratic community to rely on a variable that to some degree correlates to a sufficient familiarity with the relevant political setting, as well as to personal identification and attachment to that same political context. This is analogous to what age-related limits on voting access do. They fail to be a perfect measure of whether someone is appropriately politically informed and motivated. However, age requirements posit that everyone needs the same amount time to become politically active in a meaningful way. However arbitrary or unfair the choice of any specific age will be for some individuals, it is justifiable for democratic communities to resort to such an imperfect standard, without necessarily running into a problem of democratic legitimacy. The waiting period does not contradict the democratic principle of equal respect: it does not label the person as intrinsically and unchangeably unfit to become a political agent.

This article has endeavored to provide a pragmatic approach to addressing electoral inclusion, starting from the conceptualization of permanent disenfranchisement as a widespread democratic failure. It has called into question classical approaches to the boundary problem, and it has tentatively offered a condition to assess electoral regulations. Building on Theuns's contribution to the debate, I have argued that functional demos views erase the difference in normative weight between the notions of legitimacy and justice, effectively obscuring the main theoretical device available to address legitimacy problems in electoral policies. As a more suitable mark of legitimacy, I have proposed the permanent disenfranchisement condition, which prohibits governments from permanently disenfranchising any resident expressing a desire to participate in elections. In addition to providing unambiguous and actionable guidelines for the allocation of electoral rights, this condition targets directly one of the most pressing problems of internal legitimacy in contemporary democracies: that is, the permanent electoral exclusion of relatively vast segments of their populations.

A subordinated democratic ethos proviso, which only applies if the first condition is respected, stipulates that voting prerequisites should be justifiable through reference to the relevant democratic culture. The aim is to secure a minimal but normatively strong standard for legitimacy in democratic inclusion while framing electoral inclusion in a context-sensitive manner, which translates the general belief that there is no such thing as an ideally constituted electorate to aim for, regardless of the political setting in which voting rights are to be exercised.