Introduction
Power and democracy are intimately intertwined, a link most clearly visible in the literal meaning of democracy as power of the people.Footnote 1 In large-scale democracies, the right to vote takes a central place as an instrument for the exercise of power individually and collectively. By the vote, the people as a collective body holds to the power to determine, directly or indirectly, law and exercises of public power. By the vote, power is also conferred to the individual members of the electorate. Voting grants power to decide elections that determine the composition of legislative and executive bodies that in the end decide the substance of policy and legislation.
This rough picture is often reflected in claims by democratic theorists to the effect that equal voting rights are “a device for the distributing political power equally among citizens”.Footnote 2 Equal and universal suffrage grant citizens an “equal share of political power” where the “opportunities to exercise power over the state” are the same for each person.Footnote 3 Examples of the notion that voting rights are associated with political power are ubiquitous. It is present in the claim that to deny a person the vote is tantamount to denying that person political power; whenever the right to vote is abridged, the political power of that person is “diminished”.Footnote 4 It is present also among critics of democracy. According to them, there can be no moral right to vote exactly because such a right would entail “power over others”.Footnote 5
The notion that voting rights confer political power to the people as a collective body as well as to its individual members is accordingly widely shared. But is it correct? As is usually the case, the answer is conditioned by the meaning of the terms involved, in this case on the meaning of power. While the concept of power has at times been said to be “essentially contested”, in the sense that no account of its meaning is superior to another, I shall nevertheless proceed from the assumption that the root meaning of power can be spelled out without much difficulty: power is the capacity to make something happen even in the face of resistance.Footnote 6 Hence, political power is the capacity to make something happen in political life also when encountering resistance. The question whether the vote confers political power then amounts to the question whether the vote confers the capacity to make something happen with regard either to the substance of policy and law or to the composition of representative bodies or what I shall often refer to as the “electoral outcome”.
In one obvious sense, the vote does confer political power as it negates the ability of another vote to decide the outcome of elections. As Beitz notes, the equal distribution of the right to vote among the members of the demos secures equal power because each voter enjoys “equal capacity to overcome the same amount of resistance, namely the vote of another member”.Footnote 7 However, the power implied by the exercise of the vote, following Beitz’s account, does not entail the capacity to determine either the electoral outcome or the substance of policy or law but only the power to prevent another voter from holding power in either of those senses. The question asked in this paper is concerned with the powers traditionally associated with voting rights in democratic theory, namely the capacity to positively make something happen in political life, by which is here understood either the electoral outcome or the substance of policy or law. Is this image of the powers of voting plausible?
In what follows, the answer to this question is examined from the vantage point of two distinct conceptions of power. The first offers a causal explanation of the powers exercised by voting. The causal power of each voter in large-scale elections is small, perhaps even negligible. But as we are told, it is there, and can be variously expressed, for example as the “a priori voting power” that equals the probability of the individual vote being decisive to the electoral outcome.Footnote 8 A simpler way of capturing the same idea is that by the vote, each citizen is afforded an equal “probability of having it your way”.Footnote 9 So understood, the right to vote confers power by granting its holder the capacity to affect electoral outcomes causally.
The second conception explains the powers exercised by the vote in legal terms.Footnote 10 The vote is conceived of as legal power, specified as the capacity to regulate the legal position of others. Legal power so understood is exercised by the members of parliaments as they enact legislation and perform other public functions, it is employed by civil servants in the making of administrative decisions, by the members of the judiciary when they apply the law, by the police and penal officers when they enforce the law, and by citizens when they create contracts or wills, transfer property, and so on. The question though is whether citizens also exercise legal power as participants in the process that appoint and authorize the members of the legislative and executive branches or when they participate in referendum. Political rights were not mentioned by Wesley Hohfeld, who first introduced the notion of legal power, or by any of the defining contributions to the analysis of legal power since.Footnote 11 Yet, some legal scholars have cursorily remarked that voting and political participation are instances of legal power.Footnote 12 But so far only Jeremy Waldron has made a systematic attempt to explain the political powers exercised by the vote in terms of legal power.Footnote 13 I shall return to Waldron in the final section where I argue that his account is ultimately unsuccessful.
The legal and the causal conception of power provide radically distinct accounts of the powers conferred by the right to vote to the individual voter as well as to the collectivity of the people. It is of course conceivable that none of these conceptions of power provides a plausible account of the tenet that the vote confers power. In fact, Ronald Dworkin once argued that no interpretation of power is able to make sense of the democratic ideal of equal power.Footnote 14 Yet, Dworkin surprisingly did not consider the legal conception of power, which incidentally is the one that according to this author is offering the best prospect for a coherent account of the powers exercised by the vote. Thus, one claim defended here is that power conceptualized as legal capacity allows for a plausible reading of the claim that the people as a collective entity exercises power by voting. Another claim defended here is that legal power of the people represents the only plausible account of the powers conferred by the right to vote. That is, citizens are unable to exercise legal power as individuals by means of the vote. Moreover, neither individual citizens nor the collectivity of the people is able to exercise causal power by means of the vote. The conclusion is, in other words, that only a legal understanding of power can sustain the democratic image of “power to the people”. This is an important finding, given that democratic justifications of voting rights are premised on the vote as an instrument for power.
Democracy and power
Following Niko Kolodny democracy demands “equal opportunity to influence”, which in turn requires a fair vote in a “process that is positively sensitive to one’s choice”.Footnote 15 Indeed, a person who is able to access the polling station and to cast a vote that is subsequently counted is evidently participating in a process that “is positively sensitive” to the choice of that person. But this cannot be enough. The democratic rationale for the right to vote is not just that it grants access to a process that is sensitive to choice but that it grants access to a process that bears on politically significant outcomes. In order for that to be true, the votes cast and registered must serve some formally defined function in deciding the composition of legislative bodies or, alternatively, in deciding the substance of policy and legislation. The democratic significance of the right to vote is premised on the claim that votes bestow power over some politically significant event. The trademark of democracy is that major decisions on law and policy depend on the preferences of the people, formally expressed through the vote.Footnote 16
Based on these considerations, it is clear why democracy is regularly perceived as premised on the validity of the tenet that the right to vote confers political power. Voting rights provide the formal mechanism connecting people’s preferences and public decisions. Unless people enjoy voting rights, they would not be able to make public decisions depend on what they think. But the vote would not be essential unless it was considered important what people decide and not just what they think. After all, other methods than voting can do the job of identifying preferences. That is, the democratic understanding of the right to vote is premised on the significance of making people powerful and not just influential.Footnote 17 It is not enough that the people is influential, since it is possible for the people to be influential only because the government cares to identify what people generally prefer, want or need. In order for the people to wield power, the exercise of public power must depend on what people do and not just on what people prefer.Footnote 18 The implication is, as pointed out by David Estlund, that the process of governing in a democracy must involve “the activity of the people”.Footnote 19
Yet, the requirement that major issues of law and policy depend on the “activity of the people” can be read in two different ways. Either it refers to the powers of individual voters, or it refers to the powers of the collective of voters. The expression “power to the people” encapsulates this ambiguity by lending itself to both readings. On the one hand, it can refer to the notion that individual citizens are powerful in so far as they are able to vote. On the other hand, the phrase suggests that the people is a collective agent constituted by the electorate, the nation or the demos. This very ambiguity is reflected in Estlund’s writings, as he alternates between “the people” and “citizens”. Evidently, “the people” is a singular noun that denotes a collective agent, such as “the people of Sweden”. On the other hand, “the citizens” is a plural noun that typically refers to individual members of the people. The plural reading is endorsed by Harry Brighouse arguing that electoral outcomes should be the “non-contingent result of her [the citizen’s] participation”.Footnote 20 According to him, political power is an attribute of citizens taken separately. It is clear, however, that the democratic notion of power to the people can be read as either power to the collective body of the people or power to the individual citizen. Typically, the powers of the people and people are not distinguished.Footnote 21 The point is that even if political power serves an important role in the democratic justification of voting rights, it is not clear to whom political power should be attributed. The strategy adopted in this paper is to leave this question open. Tentatively, we assume that democracy involves the requirement that voting rights confer political power in some sense to someone. The expectation is that an exploration of legal and causal conceptions of political power should make it possible to specify to whom voting rights confer power and in what sense.
The marginalist conception of causal power
In this section, we examine to what extent one specific conception of causal power is able to accommodate the democratic expectation that the right to vote confers power to individuals and/or the people. Following this conception, the right to vote represents political power if the vote can be used to produce outcomes. Exactly how the vote can serve as the causal input explaining outcomes is subject to debate. In the following, two versions of this idea are examined: the marginalist and the threshold conception of voting power.
According to the marginalist view, the voter has political power if and only if she is able to decide the electoral outcome. The implied causal understanding is “marginalist” in the sense that only the final “push” that makes an event happen is considered causally effective. The voter is decisive if she is “pivotal” in the sense that her vote is sufficient to tip the balance in favor of one outcome rather than another. Apparently, the electorate as a collective agent is powerful according to this view since it always has the capacity to decide the outcome of elections.
However, individual voters have few prospects of being pivotal. Some voters might be committed to political parties or candidates that never even gains sufficient support. Yet, voting power is not defined as actual possibility of being pivotal but as the probability of being decisive when every imaginable electoral outcome is taken into account. The probability that a single voter is pivotal across all possible outcomes is her voting power.Footnote 22 Clearly, if political power is equal to voting power that is in turn equal to the probability of deciding electoral outcomes, the political power of each voter must be very small. It has been calculated that the probability of a single voter being decisive in US presidential elections is between one in ten million and one in a billion, depending on in which state he or she is entitled to vote.Footnote 23
The smallness of voting power is not itself an objection to the marginalist conception of causal power.Footnote 24 Clearly, the miniscule political power of each voter is reasonable given the scale of democracy at the level of the nation-state. The problem is rather concerned with the translation of the claim that each voter has some miniscule probability of being decisive into the claim that each voter possesses some miniscule portion of political power. Is the probability of being causally effective the same as wielding power? It does not seem congenial to the core meaning of power as the capacity to do things. I have the power to lift you up if indeed I am strong enough to do it. My power to carry you equals the potential to do it. Similarly, a nation has the power to invade and conquer another nation if it holds the military and other resources necessary to accomplish the task. A person or an agent is accordingly powerful if she has the capacity to realize some event and to overcome any barriers or resistance to it.Footnote 25
This means that power is a “success” concept. The measure of power is success in accomplishment. But it is capacity for success, not successful achievement, which counts as power. The existence of power is not conditioned on it being exercised, only on the capacity of exercising it. The capacity for success is conveyed by the claim that someone is “powerful”. Evidently, this is why power is distinct from the “flow” or “stream” of causal effects.Footnote 26 A person is powerful if she has the capacity to mobilize sufficient causal effect to generate an outcome. She may be powerful even though she does not generate any causal effect at all. Power equals ability to muster causal effect; hence, political power is the ability to be causally effective with regard to political decisions.Footnote 27
Yet, individual voters do not have access to political power in this sense. Each voter has some probability of being powerful that equals the likelihood that her vote is pivotal to the outcome. Hence, when we say that voters have political power because there is some probability of being decisive, what we really imply is that voters have some probability of having the capacity to decide the outcome. The question we should ask then is whether the probability that an agent has the capacity to do something qualifies as the attribution of power to that agent.
It would on condition that the probability of having a capacity to do something equals the capacity to do it. That is, the voter does have political power if the probability of the voter being influential represents a valid interpretation of the claim that the voter has the capacity to influence electoral outcomes.
At this point, we need distinguish between power as potentiality and possibility. Power as potentiality is power in the generic sense of capacity to overcome barriers. Yet, the possibility of power does not refer to the actuality of capacity to produce an outcome but to the existence of some trajectory of events where the capacity is actual.Footnote 28 The fact that I could buy a gun and shoot you indicates that it is possible for me to kill you, not that I presently have the power to kill you. The mere possibility that an agent can make something happen is not sufficient for the attribution of power. Now, voters do have some probability of being decisive. Is that best understood in terms of potential power or is it just possible power, which is not power at all?
It would be a mistake to discard the power of voters just because probabilities are involved. Consider for example A firing a gun at B. Though the bullet is not determined to hit B, it will do so only with some probability, it seems ludicrous to deny that A has the power to shoot B. After all, if the first bullet misses, A might go on firing until the job is done. But what if the probability of hitting B is miniscule, perhaps because A has only one try, is 400 meters away from B and the sight is bad? Does A still have the power to kill B?
The situation is similar to the question whether I have the power to win the New York Marathon. Though I am not much of a runner, there is undoubtedly some infinitely small probability that I would win the race (if, say, all other participants get sick). But does it mean that I have the power to win?
The distinction here is between outcomes that are mere probabilities and outcomes that can be established by the agent in spite of probabilities. Whence a single shot is fired, the outcome is not determined. If more shots can be fired, the agent is able to determine the outcome. In the latter case, but not in the first, the agent does have the power to kill the other. By analogy, I do not have the power to win a lottery if given only one single try. If I can draw as many tickets I like, I do have the power to win the lottery. Power is not chance. Power is the ability to overcome chance.Footnote 29
The implication of these observations is that voters do not have the power to determine elections although there is some small probability that they will be decisive. A low probability signals that the voter is rarely ever pivotal and that she rarely ever has the capacity to decide the outcome.
But perhaps these remarks are based on an ungenerous reading of the probabilistic account of voting power. What advocates of this view propose might be better explained by the claim that there is some probability that voters are influential and therefore powerful—not that political power equals the probability of being influential. Voters have some probability of actual power, as it were. The claim that the vote confers the probability of actual power avoids the troubles associated with saying that voters have actual power because they have a probability of being causally decisive. The implication is of course that voters do not always possess political power. A voter has political power if and only if her vote is pivotal to the outcome and accordingly has no political power if her vote is not pivotal. From the fact that there is some probability of each voter being pivotal, we could not deduce that each voter has political power.
This move remains true to the facts but comes at the price of undermining the democratic claim that political power is an attribute of the voter. If only pivotal voters do have political power, then most voters normally do not have political power. Hence, if we think that there is some probability of some single voter being pivotal and that political power is present only among voters who are pivotal, it follows that there is some probability of some single voter holding political power. Yet, since real-world elections are scarcely decided by singular pivotal voters, the implication is that, normally, no single member of the electorate holds political power.Footnote 30
The threshold conception of causal power
Let us consider a different conception of causal power. According to the threshold view, any member of the set sufficient for an event to take place is causally effective and hence powerful. In the context of elections, the implication is that political power is an attribute not solely of the voter whose vote is pivotal but of every vote required for the winning alternative to win. The proper analogy here is to lift a heavy object where the combined strength of, say, six people is necessary. The causal power involved in lifting the object is an attribute of all the six people whose strengths where required. It would be plain wrong to submit that only the sixth person (who was “pivotal”) exercised causal power.Footnote 31 This “threshold” conception of causal power offers a distinct and more “inclusive” understanding of causality than the marginalist conception discussed above.
The threshold conception avoids the seemingly paradoxical result that no single voter holds political power unless some voter is pivotal. Following the threshold view, political power is present among every voter required for the candidate to win since each of them contributes to the outcome. When a vote is won by more votes than required (as is usually the case) the political power of each voter equals the probability of being part of the necessary set of votes. If the winning candidate required just 40 percent of the votes received, each voter who did vote for the winner had a 0.4 probability of being powerful. But this seems to bring us back to original problem, noted above, of translating statistical chances into power. From the fact that there is some probability that each voter was part of the set that together caused the outcome, it does not follow that any voter had the power to make the outcome happen. All that follows is that there is some probability that each voter had the power to do so. Thus, the problem of attributing political power to individual voters based on probabilistic estimates of causal power remains.Footnote 32
Is either version of the causal conception of power able to accommodate the claim that democracy confers power to the people as a collective entity? The problem is that either account would only recognize that some sub-set of the electorate is causally effective. And it would be to commit the fallacy of composition to conclude that the electorate as group is decisive from the fact that each voter has some probability of being decisive. The collective body of the people does not have the power to decide elections just because some of its members do. The saying that in a democracy, the collective body of the people has the power to determine electoral outcomes, does not square with the observation that in democratic elections only some sub-group of the electorate exercises causal power.
Furthermore, if power is just to have a causal impact, the people is exercising power even if outcomes are determined by other agents (lobbyists, for example). By contrast, the democratic ideal of popular power is the supposition “that not only should the people have an influence on government, they should have an influence that tends in a certain direction”.Footnote 33 Following Pettit, we should be more concerned with the extent to which the people are in “control” than with the extent to which they have the ability to exert causal power.
However, Pettit’s point is concerned with influence over policy, not with influence over electoral outcomes. For the latter question, the primary concern is the relative causal power of the votes cast by the electorate in relation to other causal trajectories. Pettit’s observation reminds us that the causal power of the electorate is part of a grander flow of causal effects that determine elections. The causal understanding of power does seem to preclude that the power of the people has some special status. The people should not only have causal power, the people should be the sole source of power. But if power is just short for causal power, the powers of the people are just one in many. Though the people has the capacity to intervene in the chain of causes that determine elections, so do many other variables: the social structure, political ideology, the design of the electoral system, voters’ perception of economic reality, are among the causes of electoral outcomes.Footnote 34 As Barry once noted, the purely causal analysis of power must grant that the disposition of voters is conditioned by arbitrary factors such as the weather, success in the World Cup, and so on.Footnote 35 The point is that the decision made by the members of the electorate represents just one (intermediate) causal input among several others to the full causal story of elections.
This is of course why other activities, most notably the capacity to influence public opinion, sometimes grant you more political power than voting (Brennan, 2012, 15). In so far as political power is understood as causal power, there is no basis for crediting the electorate with a privileged place in the explanation of elections. Given that political power is conceptualized as the ability to influence an outcome and the fact that electoral outcomes are influenced by a wealth of distinct factors, the exclusive attribution of political power to the members of the electorate remains unattainable.Footnote 36
Democracy and legal power
Similar to causal power, legal power is the capacity to produce outcomes. The distinctive feature of legal power is that the activities performed and the outcomes they generate are both understood in legal terms. An agent holds legal power if and only if she holds the legal capacity to produce a legally recognized outcome. If the agent holds an additional claim-right to the legal capacity, she holds a right to legal power.Footnote 37 The requirement that the capacity and the outcome are defined in legal terms is articulated by Carl Wellman’s definition of legal power as the “legal capacity to perform some act that has specific legal consequences and that is normally performed in order to bring about some such consequences.”Footnote 38
By implication, capacities by agents and outcomes generated by them that are not recognized by the law do not amount to legal power. For example, a person with the capacity to stop the traffic on the highway is evidently in possession of the physical and causal power to make this happen. Yet, unless she is a police officer or otherwise vested with the appropriate legal authority to undertake such acts, she is nevertheless without legal power to do so. Causal power is not sufficient for legal power.Footnote 39 There are also other logical possibilities where either the capacity or the outcome is not legally recognized. Consider for instance an agent with the legal capacity to make a decision but where the outcome of the decision is not legally recognized. This would be the case if the decision is subsequently invalidated by some legal authority. As it turned out, the agent did not possess the legal power to make that decision. Or, consider the perhaps less intuitive case of an agent without legal capacity to make a decision that nevertheless does make a decision and where the outcome of the decision is subsequently recognized as valid. An example would be unilateral renouncement of membership in an organization, where no legally defined procedure to that purpose exists, but where other members accept the validity of the decision. The agent apparently had no legal power to make the decision but nonetheless managed to produce a legally valid outcome.
In order for the right to vote to confer legal power to the people, it must be true that the people has a legally recognized capacity to vote and that it is able to produce a legally recognized outcome by the exercise of the vote. The first condition is easily satisfied. Most adult citizens are in most countries granted legal voting rights. But while this condition is not so hard to satisfy, we know that enjoyment of legal voting rights is insufficient for the vote to be effective as a means for political power. Incumbents may manipulate elections, suppress competition, censor the media, with the result that voters are deprived of any effective means to produce changes in the legal positions of people in office.
If legal voting rights are to be legal powers, a second condition must be fulfilled, namely that they provide the capacity to produce legally recognized outcomes. Exactly what this involves is far from clear, however. Following Wellman’s account of legal power, the vote must be able to produce some legal outcome. The right to vote satisfies this condition whenever it entails the capacity to cast ballots that are counted and summarized in some legally defined way. Clearly, the legal registration of your vote, possibly including the public recognition of your vote in the tally of total votes counted does represent a legally defined outcome. This is obvious from the fact that casting your ballot in the wrong place or at the wrong time would result in the failure to achieve legal recognition of your vote. By the right to vote, each voter is consequently afforded legal power over public officials tasked with the computation of the electoral result. This usage appears consistent with Wesley Hohfeld’s original definition of legal power as “control” over changes in “legal relations”.Footnote 40
Although this understanding appears consistent with previous definitions of legal power, there is a thicker conception of legal power that is also more attractive to a democratic account of the right to vote. The more demanding view is that legal powers are “second-order rights”. Legal powers confer the capacity to redefine, abolish or introduce permissions, prohibitions and requirements. In short, legal power exists only where there is legal capacity to regulate legal rights.Footnote 41 Following this more exacting notion of legal power, votes cast in elections must affect the legal rights of others in order for voting rights to count as legal power. But do electoral outcomes affect others’ legal rights?
Consider, first, the claim that votes affect the legal rights of the people subject to state institutions. It is undeniable that elections provide important input to the political process that eventually determine the legal rights of anyone in the jurisdiction. For example, social and economic rights are likely to change, depending on whether left-leaning or right-leaning political parties are successful in parliamentary elections. But are these political consequences sufficient to conclude that people regulate the rights of others by means of the vote?
In fact, we should distinguish between the legal power to regulate the rights of another and the legal power to determine an outcome that affects the legal rights of another. For example, A and B may enter a contract that confers A with the legal power to decide the hourly wages of B. As a consequence of the legal power of A, the ability of B to buy a new car is contingent on the decision of A. Yet, this is not equal to the claim that A has the legal power to decide whether B shall buy a car or not. By analogy, though citizens have the legal power to decide electoral outcomes, which are in turn consequential for the legal rights of the subjects of the relevant authority, citizens are not thereby conferred the legal power to decide the legal rights of these subjects.
Consider, second, the ability of voters to decide the legal rights of the people elected. This seems more promising. The legal right to hold office depends on the decisions made by the voters. The right to hold public office and the duties to leave public office are contingent on the electoral outcome. Here, then, is a clear sense in which voters are able to make decisions that regulate the rights of others. And yet, the conclusion is not entirely satisfactory. There is a difference between being in control of the legal rights of others and being in control of to whom the rights already defined by others should be apportioned. In fact, the electorate is not regulating the legal rights of people elected to office; they only decide who should enjoy the legal rights that others have already decided that people elected to office should enjoy. The powers held by the electorate are analogous to the legal powers held by nomination committees; they are not making decisions about rights, they are making decisions about people. In a representative democracy, the enfranchised portion of the people equals the nomination committee.
The question then is whether the power to appoint someone into office amounts to legal power. There are two major types of legal power, the power to change laws and the power to change legal relations in accordance with law. Jeremy Bentham termed the first the “Powers of imperation”; they comprise the legal capacity to impose, revise or abolish the legal duties of others by the making, modifying or revoking laws. Clearly, as we have just seen, the powers of imperation pertain only to the members of legislative bodies and the judiciary and not to the members of the electorate.
The second type of legal power Bentham called the “powers of aggregation”.Footnote 42 As a private person, I have the legal capacity to change the legal position of others and myself, by the making of contracts, wills, and so on. There is of course the language of “social contract” that seems to suggest that the legal powers of making contracts is pertinent to the relationship between citizens and their representatives as well. In actuality, the relationship is contractual merely in a metaphorical sense. Citizens do not have the ability to create legally binding rights or duties for the people elected by them. The right to vote does not equal the power of aggregation either.
However, as observed by Bentham, the powers of imperation possessed by lawmakers are incomplete in an important sense.Footnote 43 Bentham notes that law making involves the creation of legal duties only for general classes of persons. The task of legislating is the task of deciding the legal rights and duties of “property owners”, “judges”, or “citizens”. Law making does not entail decisions about the legal rights and duties of particular individuals. The legal rights and duties of particular individuals depend on some additional decision that distributes the status of property owner, judge or citizen among them. Thus, in order to translate the rights and duties of general classes of persons to rights and duties of specific individuals, someone must “settle the membership of those general classes”. Bentham explained the latter by saying that the powers of imperation must be “broken into shares”. The “shares” refers to the requirement that legal power is separated into one agency with the power to regulate the rights and duties of member of classes and another agency with the power to regulate membership in these classes.
Hart and Bentham used the distinction between imperation and its shares to explain the division of legal powers between the legislative and administrative branches of government. None of them contemplated the possibility that legal power is exercised by citizens too. Yet, it seems very clear that Bentham’s distinction is useful for explaining the legal powers of the electorate. We have already established that the electorate cannot exercise the legal powers of “imperation” in the sense defined by Bentham. Only the members of legislative bodies do. Nonetheless, it is consistent with Bentham’s terminology to conclude that the electorate holds a “share” in the powers of imperation. The members of the electorate do so by virtue of deciding membership in the class of legislators and thereby completing the legal powers of legislators. In a representative democracy, there can be no legislators without a legally valid electoral result deciding the composition of legislative bodies. In that very specific sense, the legal right to vote confers legal power to the electorate.
Now, the conclusion that the electorate is in possession of legal power represents advancement compared to the causal conception of power. Remember, the causal view could not explain the democratic intuition that the people is in control of electoral outcomes. Given the multiplicity of causal factors that influence the electoral process, the causal power of the people is reduced to one among several determinants of electoral outcomes. Yet, in a democracy, we do not wish to say that the people is merely one among several causes of electoral outcomes. Rather, we wish to say that only the people has the exclusive power to decide elections.
This is exactly what we can say if power is understood in legal terms. Given that the law stipulates that elections should be determined on the basis of a popular vote, and that the members of the people are legally entitled to vote, it follows that the legal powers exercised by the people alone decide the electoral outcome. The legal powers afforded to the collective body of the people, by the right to vote, equals popular “control” of the electoral outcome. It does not matter, from this point of view, that the causal determinants of the elections are multi-faceted since causal influence does not equal legal power. Even if other factors play a causal role in explaining who wins the election, only the actions undertaken by the people are legally effective in deciding the outcome. In this very direct sense, the notion of legal power is better suited to explain one basic democratic intuition.Footnote 44
The point is analogous to the relationship between the judge in a court of law and the judgement reached by the court. In order to explain the judgment, a variety of causal factors is pertinent and it would be quite misleading to submit that the opinion of the judge alone explains the judgment. Evidently, the evidence presented before the court, the strategy and skill of the attorneys and prosecutors, etc., are causally pertinent. However, once we ask who has the legal power to settle the case, it is obvious that only the judge holds the relevant legal power. The judge has exclusive legal power to decide judgments in a court of law, just like the people has exclusive legal power to decide elections in a democracy.
Individual and joint legal power
As stated initially, legal power is the ability to regulate the legal positions of others. The question whether the vote confers legal power accordingly translates into the question whether a single voter is able to regulate the legal position of others. In reply, Waldron remarks that the legal power of the voter is held jointly with other voters. A jointly held legal power is one where the description of the power “makes essential reference to the actions of others”.Footnote 45 Waldron explains this claim by analogy to a jointly held check account where the account-holders have the capacity to sign for as well as against withdrawal from the account. The implication is that one person’s signature in favor of withdrawal from the account is effective only depending on what other account-holders decide. Sometimes the signature will produce the desired legal effect, sometimes it will not. The situation is similar to voting, according to Waldron. Sometimes, your vote is effective because a sufficient number of votes are cast for the same candidate or party. But sometimes your vote is just wasted. The point emphasized by Waldron is that your signature (or vote) is legally effective whether it contributes to the desired result or not. Thus, even if A is unable to withdraw money from the account by his signature, the outcome is “exactly the consequences to which the rule about the signatures […] pertains”.Footnote 46 The point then is that if legal power is a jointly held capacity, then legal power is exercised whenever an agent performs an action that changes the legal position of others if enough others perform it too. Waldron concludes by noting that a vote is “all that has to be done and can be done, for his part, in the eyes of the law, to bring about that legal consequence”.Footnote 47
Evidently, Waldron’s analysis is premised on a particular understanding of legal power. As defined in this paper, legal power is the capacity to regulate the legal rights of others. By contrast, Waldron endorses the view that a person can hold legal power despite the absence of immediate legal effect from her actions provided that such effects will follow if enough others perform the relevant actions. A sufficient condition for legal power, following his account, is that an agent could be a member of a set that is sufficient to transform the legal rights of others. This is why Waldron is satisfied with the claim that A holds legal power even when her signature for withdrawal from the joint bank account is always defeated by the joint account-holders. The same should then be true for voters in national elections. Even if A’s vote is always “wasted”, because A consistently votes for the loosing candidate or party, it may still be true that A has legal power. The reason why is that it remains possible for A’s vote to be legally effective, depending on how others vote.
Waldron’s definition of legal power is the ability “to perform an action which (if enough others also perform it) alters the assignment of rights and duties in the political community”.Footnote 48 The words within the parenthesis are crucial as they seemingly entails the rejection of the view that legal power equals legal capacity to regulate the legal rights of others. Rather, Waldron’s understanding of legal power appears consistent with the claim that the exercise of legal power may or may not alter the legal rights of others and that the extent to which it does is ultimately a matter of coincidence. This means that the conceptual apparatus employed by Waldron deviates from the one employed in this paper and, possibly from established terminology.
In reply, Waldron might say that quibbles over terminology are of little substantive interests and he might just accept the claim that the attribution of legal power to an individual does not entail the capacity of that person to regulate the legal position of another. But in biting this bullet, the analysis of the right to vote in terms of legal power is distorted.
If legal power is divorced from the legal capacity to regulate the legal position of others, it turns out that we are entitled to attribute legal power to individual citizens by virtue of the legal right to vote even in political systems where the exercise of that right is impotent. Too see this, consider a political system where the government regularly intervenes in elections to ensure that it remains in power. As long as the interventions are recognized by the law (such as the government is legally empowered to change the result of elections), it seems perfectly consistent with Waldron’s conceptual apparatus to submit that citizens exercise legal power by the vote. The reason why is that it remains true that the each citizen is potentially able to regulate the legal rights of others by the vote if “enough others” perform the relevant legal action. Following the reasoning of Waldron, it is still possible that the votes cast by the members of the electorate are decisive. The possibility would materialize if, for example, the government realizes that it does not have to intervene in the election in order to secure a favorable outcome. The conclusion is very different if legal power is defined as the legal capacity to produce legally recognized outcomes. In that case, the fact that the government retains the legal right to intervene and decide the election is undermining the legal power of the electorate to do exactly that.
The observation that voting rights will be legal powers even in cases where the citizens are deprived of the right to decisively determine the outcome of elections means that Waldron’s conception of legal power is unable to distinguish between the powers entailed by the vote in democratic and authoritarian systems respectively. Based on Waldron’s concept of legal power, we could no longer account for the claim of democratic theorists that voting rights are essential for the empowerment of the people.
Earlier in this paper, we noticed that power is ordinarily defined as the capacity to produce outcomes and to overcome resistance in the process. In a democracy, the legal power of the electorate is inherent in its capacity to determine electoral outcomes unobstructed by legal challenges. By Waldron’s understanding of legal power, the electorate is not necessarily able to quell resistance by authoritarian incumbents. Hence, Waldron’s usage of legal power departs from the ordinary meaning of power. The price paid for Waldron’s analysis of voting as the exercise of legal power is in other words that power is taken out of the vote.
Concluding remarks
Democratic theorists argue that the right to vote confers power; by the right to vote the citizen holds a piece of political power and, together, the members of the electorate exercise power in elections. The position defended in this paper is that this view is largely mistaken. In so far as power is understood in causal terms, we are not entitled to conclude that voting rights confer power. In particular, we argue that the theory of “voting power”, the tenet that the powers conferred by the vote equal the probability of being decisive or “pivotal” in elections, does not warrant the conclusion that voters hold political power. Given the meaning of power as the capacity to produce outcomes, the mere probability that an agent produces an outcome does not equal the claim that the agent holds power. Rather than saying that voters have a small probability of being powerful by the exercise of the vote, we should conclude that voters normally have no power to influence the outcome of elections.
The notion that the collective body of voters has the capacity to determine electoral outcomes does not fare much better. The problem with any causal conception of power is that it is unable to recognize the people as the unique owner of political power. The powers exercised by the members of the electorate are just one among several causes of electoral outcomes and hence does not have the capacity to determine the result unilaterally.
The vision of voting rights as means for political power is sustainable only on the basis of a legal understanding of power. In fact, it can be sustained only in so far as we are concerned with the legal powers of the electorate as a collective body. Legal voting rights do confer legal power to the electorate by virtue of the fact that the vote provides the legal capacity to appoint public officials. However, the individual voter does not have the capacity to determine the legal position of others and is accordingly unable to exercise legal power by the vote.
The contrast between legal and causal conceptions of power illuminates the wider question about the relationship between individual citizens and the state in a democracy. When the right to vote is understood as the opportunity to wield causal power, the relationship between the voter and the state depends on the likelihood of the vote being an effective means for political change. To the extent that citizens are unable to exert causal power, the democratic significance of voting dissipates and they are reduced to subjects of public power. By contrast, when voting rights are conceived of as legal powers, citizens are members of a legal category whose legal relationship to the state is unmoved by the causal effectiveness of voting. The Swedish political scientist Herbert Tingsten once captured the gist of this notion by the observation that the electorate is, in a legal sense, an “organ of the state”.Footnote 49 This now largely forgotten view is also expressed in Norberto Bobbio’s claim that from the vantage point of a legal analysis of the state, there “cannot be any power other than legal power”.Footnote 50 If legal norms can only be changed by legal means, and if the people have the capacity to participate in the creation of legal norms, the rights of the people to participate in that process must also be conceived of as legal powers. The legal conception of power in other words elucidates not just the place of power in democratic theory but also the relationship between citizens and the state in legal analysis.
