Should We Allow Prisoners To Vote?

A few days ago, I wrote a post on what I termed "quiet injustices" -- things that pretty clearly implicate questions of ethics and morality in our society, but yet rarely seem to bother us. My main example was D.C. disenfranchisement. Another is felon disenfranchisement (after their sentences have been served). Neither, I think, is in any remote way justifiable, and neither are particularly salient political issues.

But the more I think about it, the more I question whether even disenfranchising felons while they're in prison is justifiable. So, I've decided to spend some time exploring that issue.

In the field of legal philosophy, there are two primary justificatory models for why we can punish criminals. The first is "utilitarian"; we punish people because it leads to better social outcomes when we do. Deterrence, for example, is a utilitarian consideration, as is preventing the criminal from committing more crimes. The second model is "retributive"; essentially, giving criminals their just deserts. Those who commit crimes deserve to be punished (to varying degrees, depending on the crime). But the flip side of the retributive model is that you can't punish people more (or less) than they deserve, even if society would benefit (from the deterrent aspect, say). Hence, the two models are somewhat in tension with each other. Sometimes, society as a whole would be better off if a criminal was punished more (or less) severely than they "deserve". In such situations, we are forced to pick which model we prefer. Hence, I'll analyze prisoner disenfranchisement under both models, then examine several additional pragmatic considerations outside punishment theory that might advise against allowing prisoners to vote.

I. The Utilitarian Model

The utilitarian position forces us to weigh the benefits of prisoner disenfranchisements against its costs. We'll start with the costs, which should be obvious: denying someone the franchise is an extraordinarily serious breach of normal civic procedures. The harms are aggregated when it is a particular class of individuals (in this case, prisoners) who are so disenfranchised, because whatever interests they have as a class will be erased from our collective deliberations. And prisoners do have several exceedingly important interests that deserve hearing: combating prison rape and prison reform more generally, the availability of reintegration programs for non-lifers, even the simple mechanisms to prevent family dissolution during imprisonment. Moreover, prison disenfranchisement leads to the 0/5 compromise wherein mostly minority prisoners are transferred to White locales, who then gain increased voting power without additional voters. This creates a perverse incentive to deliberately jack up the incarcerated population, particularly via laws which target minority populations, and has the effect of creating unjustifiable asymmetries in political power, privileging White communities over communities of color.

The benefits of punishment in general, in the utilitarian model, are those of deterrence, and pre-empting potential future criminal activity by the imprisoned person. But outside cases where the crime itself is election-related (and in those few cases, I think disenfranchisement is justifiable), neither seems to apply. Allowing prisoners to vote would not give them any more opportunity make it any more likely that they will commit more crimes. Indeed, if anything it may make it less likely: insofar as criminal activity is positively correlated with marginalization from general social practices, integrating people into socially mainstream acts (such as civic participation) should be a bulwark against recidivism. Deterrence, too, to be inapplicable -- I have trouble imagining the prospective criminal for whom prison is not a deterrent, but loss of voting privileges is. Whatever marginal gain (if one exists at all) in deterrent force disenfranchisement offers is more than outweighed by the aforementioned costs.

II. Retributive Model

The retributive model allows for punishment as "just deserts" for criminal activity. This, obviously, is a bit fuzzy: what is "just deserts"? Obviously, we can recognize the poles: death for jaywalking is not just deserts, nor is a fine for murder. But there still remains the "how many grains of sand make a pile" problem. The line at which a punishment becomes too severe or not severe enough is blurry and contested.

However, one thing that can aid us in the present discussion is that we have, by and large, at least a metric for measuring the relative severity of punishments: prison. We can debate which crimes are the worst, and we can debate where each crime should specifically be placed on the spectrum. But we do agree that, abstractly, the worse the crime, the longer you (deserve to) spend in prison. And in the public imagination, that sentence is exhaustive of just deserts. Simply put, I think that the decision on imprisonment length alone is the reflection of what we imagine just deserts to be.

Consider a crime that we all agree should, in our status quo system, be punished by 20 years in jail. Imagine now that we agree that the prisoner should be enfranchised while he's serving his sentence. Do we now think the sentence should be 30 years instead of 20? If we do not believe that (and I don't think most people do), then we're implicitly saying that enfranchisement/disenfranchisement is not part of our internal weighing mechanism for determining just deserts. And if it doesn't meet that threshold, then under the retributive framework it is not a punishment we can legitimately impose on offenders.

Of course, one might take the opposite stance on my hypothetical, and say that both factors (enfranchisement and sentence length) are relevant considerations for measuring just deserts. At that point, the question shifts to whether disenfranchisement is an inherently unjust punishment. For example, if in the previous hypothetical we replaced the variable of "enfranchisement" with the variable "tortured daily", I might agree that we should adjust the sentence length downward to make up for the increased severity of the punishment. But prior to that I'd hold that torture is not a just punishment -- inherently it can never be considered part of just deserts, because it represents a extreme violation of human dignity. The question is whether voter disenfranchisement occupies a like category. I think that there is a solid argument that it does. Voting is an absolutely bedrock, fundamental part of being a person in a democratic citizenry. To be barred from voting is to be rendered a non-person in our political community. Whatever else prisoners are, I think they remain persons and citizens. They are not "men without countries", and they should not be treated as such.

This, I recognize, is more of a judgment call. But particularly since I believe we can affect just deserts via imprisonment alone, without disenfranchisement, I do not believe we have good grounds to add disenfranchisement as a punishment options outside of very special circumstances (where the felon's crime was election related).

III. Pragmatic Considerations

Having dealt with the question at a level of theory, the final issue is whether there are any pragmatic bars that would advise against allowing prisoners to vote. I think a few present themselves, but they do not strike me as sufficient to outweigh the arguments laid out above.

The first potential problem is both the most intuitively obvious (to me, anyway) as well as the most easily dispatched. That is that prisoners, already having demonstrated themselves to have anti-social tendencies, will vote for policies that sanction their sociopathic activities (i.e., all the robbery inmates will get together and vote to legalize robbery). This strikes me as an unfounded fear, in part because I don't actually think most robbery convicts actually believe that robbery should be legal, but primarily because there aren't enough prisoners to accomplish such an end. Prison enfranchisement is important because it allows prisoners to form political alliances with like-minded interest groups who also are concerned with their problems. But while there are voices on the outside who care about the problem of prison rape, there are very few, I'd imagine, which believe we should legalize robbery. The only way this could be circumvented is if prisoners actually became a democratic majority, in which case I'd argue we have bigger problems on our hands. Particularly since I think prisoners should still be considered "residents", for voting purposes, of their hometowns, and not where the prison is located, it strikes me as unlikely that prisoners will ever gain a working democratic majority.

The second potential problem is that of independent agency in a prison environment. Prison is a institution of power and control, both among prisoners, and between the guards and the prisoners. But voting demands the free exercise of one's own autonomous faculties. This creates tension. But I think the tension can be overcome. The biggest risk for intimidation comes where the prisoners are all voting in the same jurisdiction. But where the prisoners are spread out of many jurisdictions (because they hail from different hometowns), the potential for adverse influence dissipates. It doesn't go away entirely (statewide and national elections will see significant overlap in interests), but then again, the potential for voter intimidation always exists anyway. So long as ballots are secret, the risk strikes me as minimal.

The third potential problem is how to organize an election in prison. Prisoners should be given access to news, so they can make informed decisions (I do not think utter isolation from current events is justified by either utilitarian or retributive considerations). And since I envision the prisoners as being registered in their home jurisdictions, rather than residents of whether the prison is located, the election itself would be standard absentee ballot fare.

IV. Choice of Punishment: A Middle Ground

A middle ground option, alluded to early, that exists within the retributive framework would be to give prisoners an option of taking a reduced sentence that includes disenfranchisement, or a lengthened sentence that allows one to vote. But that would entail figuring out an exchange rate between enfranchisement and time-served (how many years of a sentence is the right to vote worth?). Pragmatically, it is simpler to use prison length on its own as a metric of just deserts. Moreover, insofar as a just democratic polity incorporates (to the greatest degree possible) the voices of the entirety of the citizenry, we should promulgate policies that encourage people to vote, and be particularly wary of discouraging voting amongst particular classes or groups of the citizenry. But since I suspect few prisoners would choose voting rights over a reduced sentence, adding voting rights as a punishment consideration would cut against the benefits of inclusion that I've articulated.

V. Conclusion

So in conclusion, I believe that it is difficult to justify disenfranchising prisoners, under normal circumstances, under either a utilitarian or a retributive model of punishment. Furthermore, the pragmatic arguments against prison voting are similarly lacking in merit. Meanwhile, allowing prisoners to vote would benefit society by making it more democratically legitimate, by including pertinent voices on deliberations of great social import, and even by potentially reducing recidivism rates.