There have been endless discussions among libertarians "even the anarchist variety" about the meaning of voting. Somehow, it seems, we cannot agree on what it means to vote: Is it self-defence to support the "lesser evil" or is it not? And even if we believe it is not, many libertarians cannot help casting a ballot on Election Day; they cannot resist the urge to spend their time, money, and energy to support a candidate that seems to be closest to their own in terms of values and ideals — even if that candidate is not very close at all.
Before we get into discussing the meaning of ballot-casting, we need to set some things straight. Many make a number of mistakes in trying to prove that they are doing nothing wrong by voting; we need to avoid making the same mistakes. For instance, the choice of the "lesser evil" is a choice between evils. This simple truth follows directly from the constitution and meaning of "lesser evil." Also, it assumes a very limited number of choices — choosing the lesser of two evils means there are only two choices. Of course, I would choose shooting myself in the foot before doing the same in my head at any time — but this conclusion holds only if I had no choice whatsoever but to pick only one of the two.
Picking the "lesser of two evils," it must be concluded, is a universal truth of action. In terms of opportunity cost, any actor would choose that action which maximises the anticipated outcome. This is true even if all alternatives are in the negative, for, as Mises pointed out, "[t]he incentive that impels a man to act is always some uneasiness" and the purpose of taking action is to remove that uneasiness. In this sense, does it not make sense to vote?
The answer is no, and the reason for this is quite simple: No election day provides a setup with only such a limited set of choices. The choice to be made, in America, is not between Republicans and Democrats, but between the aforementioned parties and available third party candidates — and not voting at all. (The latter, of course, implies positive value, since your time and energy can be used productively.) Even in countries where voting is compulsory, such as in Australia, there are ways of not supporting any of the parties through either casting a blank ballot, voting for oneself, or even doing one's best in making the ballot invalid.
Now that we have rid ourselves of some of the annoying misconceptions of the act of voting, let us move on to evaluate the meaning of the action. We have now established that it simply doesn't make sense to shoot oneself in the foot (unless one has a masochist bent) if the available options are a shot in the head and not pulling the trigger at all. Considering a normal actor and Mises' insight, Election Day should not see many libertarians casting their ballots if they see it as an evil act. This brings us to the analysis of the act itself, and whether it is indeed an act of evil.
The action itself may seem quite harmless to most people; after all, it usually involves either putting a piece of paper in an envelope (something we do very often in our daily lives when endorsing the USPS), punching holes in a punch card, or pushing keys or a touch screen on some computerised voting machine. But the carrying out of the action itself is not a sufficient analysis. Let us yet again use Mises' definition of action as "purposeful action." Writes Mises:
Acting man is eager to substitute a more satisfactory state of affairs for a less satisfactory. His mind imagines conditions which suit him better, and his action aims at bringing about this desired state. The incentive that impels a man to act is always some uneasiness. A man perfectly content with the state of his affairs would have no incentive to change things. He would have neither wishes nor desires; he would be perfectly happy. (Human Action, p. 13)
Following Mises, the act of voting, rather than simply the mechanical carrying out of casting the ballot, necessarily includes the actor's purpose and perceived or anticipated value of making this choice. In other words, the voter is "eager" to vote because he believes voting will make him better off than any perceivable alternative action; the actor deems voting more productive in terms of his personal utility than any other possible action.
Let us therefore take a look at the reason for voting as commonly stated by voting libertarians: The choosing of the lesser evil over the greater evil. We have already established that the choice to vote is not a choice between the two dominating parties (or, even, any parties) but also between voting and not voting. From this follows that the "lesser evil" argument necessarily implies that the individual voter should anticipate being better off (in terms of psychic income) from voting for the specific candidate than acting in any other possible way.
"Lesser evil" should therefore be an unsatisfactory term to describe this particular action, since the choice hardly ever is one between only destructive ("evil") options. It is inconceivable that voting libertarians really believe that all other possible courses of action on Election Day are destructive — or, at least, less productive than the casting of a ballot.
But to understand the meaning of voting, we need to also analyse the possible incentives for engaging in this type of action, i.e. what the actor should perceive as possible outcomes and how voting may bring about a "more satisfactory state of affairs." We therefore need to have a look at the rationale for or meaning of the act of voting.
Stating that voting serves primarily one purpose — to get one's preferred candidate elected for office and, as a result, influence the direction of public policy — should be quite banal. But it is not — it is often considered a very provocative statement. The reason it is provocative is that many voting libertarians would claim their intention is not to influence public policy in a certain direction, but that their choice of candidate is primarily negative: They vote against an alternative, and even greater, evil.
But voting against one or many alternatives in a general election only makes sense under very specific circumstances: We need to yet again adopt the very limited setup of the "lesser evil" argument discussed above, and we also need to assume that the candidate or party of choice can be counted on to work consistently for policies directly the opposite of those we consider "evil." The candidate or party simply cannot agree with the "greater evil" ditto on any issue that we consider important, but has to act as a perfect counterweight to our most feared evil policies.
Under these particular conditions, voting could indeed be considered an action taken in self-defence or against a greater evil. (We here disregard the statistical fact that any individual vote has less than marginal impact on the outcome of any election.)
It should be quite obvious that such situations do not emerge in the real world, but can only be created in perfectly controlled experiments devoid of large-scale human action. It is simply not the case that either of the dominant parties in U.S. politics offers a credible alternative to the other party's recognised evil actions. Since voting is an action taken under uncertainty of future conditions, the individual voter cannot rely solely on party rhetoric but needs to assess the party's or candidate's real action during the length of the term. (He would also need to assess the probable outcome of the election and the likely influence of his vote.)
For most libertarians, this should leave only third party candidates as possible alternatives for voting "against" a certain policy or policies. As the American political system is set up, voting for a third party candidate adds a new set of problems to consider when making the decision. Since third parties are unlikely to have any real impact on the outcome of elections and even less on public policy, such votes should be expected to have no or almost no value at all to the voter. Hence, we can but conclude that it is not possible to vote to offset evil — one cannot vote against a party, candidate, or policy. The democratic system of voting is positive only; votes are blank checks offered in support of the winning candidate.
These practical considerations set aside, we still need to investigate the meaning of voting — what does it mean to cast a ballot? What are the possible and expected outcomes of voting?
Most political systems in the western world are plagued by the influence of special interests on public policy. However, in contrast to popular rhetoric, politicians have a lot to gain from allowing special interests to influence public policy. Not only are they offered vast sums of money in payment for such influence, but they draw on special interest lobbyists as sources of information and arguments to use in public discourse in support of preferred policies. It is therefore difficult to figure out a candidate's true belief (if any) and whether he or she consistently will work for certain issues in a certain manner.
Save this uncertainty of the applicability of candidates' and parties' programs, a vote is necessarily cast in support (rather than rejection) of that candidate or party. It may be the case that the voter under very specific circumstances (see above) wished for the vote to counteract certain policies, but even such counteraction cannot be anticipated without acting to support the candidate with opposing views in one or many policies. Only through having the "counter candidate" elected can "evil" be counterbalanced. It follows that any (boundedly) rational voter must realise and take into account the fact that a vote is primarily offered, and will be counted, as an act of support.
In a democratic system requiring unanimity, i.e. a "negative" democracy, any vote would be granted veto power. In such a system, any vote could be used to counteract evil and set things right — and voting would consequently be both in support of and, in effect but not in principle, be used to reject parties, candidates, and policies. It is safe to say such a system would not be likely ever to be able to increase the realm of politics for as long as the unanimity requirement is intact. But this is certainly not the case in our so-called "liberal democracy" systems in the western world. Not only is nobody granted veto over political power, but our government organisations are consistently growing in size and assumed authority.
It follows that a vote must be offered only in support a political agenda and therefore that it neither can, nor can be assumed to, counteract unwanted political interests. Since politicians cannot be held accountable for their actions when elected for office, other than at the end of the term (in the next election), we must conclude that a vote is in many respects equal to handing a blank check to a candidate or party seeking power. But it is not a check drawn only on your personal checking account, but a check drawn on everybody's account.
A vote is not, and is not intended as, a contract between the voter and the candidate, but a claim made on political power under the assumption that the elected candidate will gain influence over government. The non-contractual nature of political power here also implies that it is fundamentally collective, not individual, which means the act of voting is a positive claim on power over society (and everyone in it). It is necessarily an act of violence, however indirect and — perhaps — unanticipated.
The latter may seem unintuitive or even unwarranted, but it really is not. Political influence implies the power to direct and command society — government assumes the right to design and/or redesign the organisation and structures of society as well as influence the actions of individuals through offering either rewards or penalties. There are no true freedoms from this reign of government except for the ones granted by government itself, since it neither allows for unregulated secession nor exceptions. This organisation enjoys, according to the common definition of state or government, a monopoly of violence and uses this privileged position to form a kind of society desired by those controlling government.
It is necessary to explore the possible meaning of voting (in political elections) in its true context, i.e. in the context of state society. Since a vote is an act carried out to influence the election of representatives of government and thereby its policies, it necessarily denotes a claim on political power. It follows that the vote is not a defensive but offensive act, and it furthermore necessarily follows that it is an offensive or positive claim on power over society. In other words, the vote cannot be interpreted in terms of its limited mechanics or its real influence — it has to be interpreted in terms of the purpose of this human action: The aspiration to directly influence the rule of government and therefore, indirectly, the forceful subjection of any individual to the power of government.
The act of voting, therefore, cannot be considered a market action since it implies the use of (indirect) force. Rather, voting falls under Oppenheimer’s definition of "political means" in the same sense that doing business with (or, more accurately, asking someone else to, on your behalf, contract with) a contract killer does: It is the solicitation of violent services offering as only payment an anonymous statement of support for an unenforceable advertised political program.
Whether a vote in this sense is indeed a support for the system as a whole is another, but related, matter that we need not analyse. We have already established the meaning and significance of voting; the analysis of voting as an act of support for the system per se, which may now appear to the reader as a quite probable conclusion, can only make the act itself appear as even worse. But as fun as such an analysis may be, it is a purely philosophical matter and should at this point be considered quite irrelevant.