The Ethics of Chemical Castration (Part Two)

(Part One)

This is the second part in a brief series of posts looking at the ethics of chemical castration. The series is working off the recent article “Coercion, Incarceration and Chemical Castration” by Douglas et al. In this article, the authors argue that chemical castration could be permissibly used, without valid consent, if it is autonomy-enhancing in its effect.

Part one covered the background to this argument. It looked at the nature of chemical castration, the different drugs used to achieve this outcome, and the evidence for and against these drugs. It then looked at a basic coercive choice argument against the use of chemical castration. In the typical scenario, chemical castration is presented as an option to convicted sex offenders: take these drugs and you can avoid or reduce the amount of time you will spend in prison. This will be an attractive option to many. Proponents of the coercive choice argument object to this on the grounds that the choice facing the sex offender is unduly coercive. Consequently, their consent to the treatment is not morally transformative.

That gives us, roughly, the following argument:

• (1) An offender offered the choice between chemical castration and further incarceration cannot give valid consent to castration [due to coercion].

• (2) Medical interventions should not be offered in circumstances where valid consent to them is not possible.

• (3) Therefore, chemical castration should not be offered as an alternative to further incarceration.

Although one could take issue with premise (1) of this argument, Douglas et al focus their energies on premise (2). They argue that even if the consent is not valid, the presentation of the choice to undergo chemical castration may still be permissible. They do so because they think chemical castration might be autonomy-enhancing for certain individuals.

In the remainder of this post, we are going to see how this argument works. We will start with a basic defence, looking at a thought experiment the authors use to motivate their case. Then we will look at a variety of objections. As we shall see, although Douglas et al think that most objections are misplaced, there is at least one that they find mildly compelling.


1. The Autonomy-Enhancing Argument
Before we get into the meat of the argument, a few clarifying observations are in order. Douglas et al’s argument assumes that chemical castration has been proven effective in reducing recidivism rates for an identifiable class of sex offenders. As we saw the last day, the current evidence base does not always support this assumption. Furthermore, their argument assumes that the offer will always be made to convicts in a manner that reduces an ongoing prison sentence, rather than at the point of initial sentencing or as a preventive measure for certain “high risk” individuals. It’s possible that the argument they make is exportable to these scenarios, but they do not explicitly consider that possibility. Finally, they assume that it is the state, through its court system, that makes the offer and that the treatment will only be continued for as long as there are no serious side effects.

The argument itself is quite straightforward. It is based on the fact that the coercive choice objection is dependent on the value of autonomy. In other words, that the presence of coercion in a particular decision problem would not be deemed important if we didn’t think autonomy ought to be respected. Thus, autonomy is the fundamental value in this debate. If it can be shown that a particular intervention would enhance autonomy, then that might be enough to negate the problem of invalid consent.

To motivate their argument, Douglas et al imagine the following case. The case is hypothetical, but plausible given what is known about certain types of sex offender:

Jeremy’s Case: Jeremy is a convicted child rapist, currently five years into an eight year sentence. He has a history of recurrent sexual abuse and has previously been imprisoned for this. Every time he has reoffended. He despises himself, feels remorse for his victims, and experiences his sexual desires as unwanted intrusions into his mental life. He desperately wants to be free from them. He chooses to undergo chemical castration as part of deal with the state for a reduced sentence. The treatment is a success and he finds that with the drugs he is able resist his sexual urges and pursue other projects and plans.

Here’s the question: is there anything really wrong with giving Jeremy the option to undergo chemical castration? In this case, it seems like Jeremy’s sexual desires are undermining his autonomy. They are preventing him from being the kind of person he would like to be. To withhold the treatment from him, purely on the grounds that his consent to the treatment is delivered under some coercive pressure, seems unjustified. Particularly if what we care about is autonomy.

That gives us the following, simple, argument in favour of chemical castration and against the coercive choice argument:

• (4) A medical intervention can be permissibly offered if its effects are autonomy-enhancing, notwithstanding concerns about invalid consent.

• (5) The effects of chemical castration for an identifiable class of sex offenders will be autonomy-enhancing.

• (6) Therefore, chemical castration can be permissibly offered to an identifiable class of sex offenders.

Just a word or two about each of the premises before we proceed to the objections. Premise (4) is worded in a slightly odd manner. It speaks only to cases in which the treatment is “offered” to someone. But one could argue that if concerns about consent can be overridden by an autonomy-enhancing effect, then there is no reason why the forcible or compulsory imposition of medical treatment would not be permissible too. Interestingly, Douglas et al don’t seem to endorse this position — though they do make some vague gestures in its direction; they seem to think the “offer” is still important. I can see why. I would certainly feel more queasy about the situation if there were no offer. Nevertheless, keeping the offer condition may not be enough. I can imagine some people challenging the argument on the grounds that it sets up a slippery slope into the forcible use of certain medical interventions.

Premise (5) is, of course, the key claim about chemical castration. How plausible is it? One problem with the claim is that the concept of autonomy, and the conditions that serve to promote or undermine it, is highly controversial. Still, Douglas et al argue that several of the leading accounts of autonomy would support the claim that any drug that removes an overwhelming and intrusive sexual desire is autonomy-enhancing. For example, rationalist theories of autonomy hold that one is autonomous if one acts on the basis of rational beliefs and desires. The kinds of sexual desire experienced by Jeremy would be irrational. Similarly, on a Frankfurtian or Dworkinian conception of autonomy, what matters is that we act in accordance with our second order, reflective desires, not our first-order impulsive ones. In the hypothetical case, Jeremy appears to be at the mercy of his first-order desires, and so removing those would be autonomy-enhancing.

In addition to all this, undergoing chemical castration would be autonomy-enhancing in a much more mundane sense. Being imprisoned is a constraint on autonomy (justified though it may be), and so anything that removes that constraint might be said to be autonomy-enhancing (provided it doesn’t have other, autonomy-reducing, consequences). That would appear to be true in Jeremy’s case.

So the conclusion seems to hold: chemical castration may be permissibly offered to an offender, even if valid consent is a problem.


2. Objections and Replies
Douglas et al look at a variety of objections to this simple argument in their paper. I’ll focus on the four that I think are the most significant.

The first objection appeals to the priority (perhaps even the lexical priority) of maintaining the current level of autonomy over-and-above promoting long-term autonomy at the expense of current autonomy. Similar arguments abound elsewhere in moral and political philosophy. For example, negative utilitarians hold that our primary goal should be to minimise suffering, not to promote well-being. Thus, if forced to choose between maintaining the current level of well-being and reducing it in order to achieve some long-term gain, we should choose the former not the latter. The idea in the chemical castration case is that we should not pursue any policies that undermine or negate autonomy in the short-term, even if there are long-term benefits. This might be encapsulated in the oft-heard slogan that we should “respect autonomy”.

There are a few problems with this objection. It might have some force if the autonomy gains were going to be minimal (e.g. treatment for some minor autonomy-undermining condition). But in the kinds of case envisaged by Douglas et al, the compulsive desires are likely to have a severely debilitative effect. Think of the short-term discomfort of a vaccination versus the long-term benefits of immunity from the disease and you might get the picture. Much more importantly, however, is the fact that the use of chemical castration may not actually deplete autonomy in the short term at all. The baseline level of autonomy for the prisoner is low. The offer does not decrease it any further. If anything, availing of the offer immediately increases the level of autonomy.

The second objection holds that not all increases in the number of options facing a particular agent are autonomy-enhancing. If true, this would be a problem for Douglas et al’s argument as they are essentially arguing that the sex offender’s autonomy is enhanced (immediately) in the case under consideration because the number of options they face increases from one (stay in prison) to two (stay in prison, or be released).

But is it true? The objection rests on the belief that sometimes presenting an option to an agent can be autonomy-depleting if the option in question appeals to some irrational desire. For example, a heroin addict who has the option to take a shot of heroin is arguably less autonomous than a heroin addict with no access to the drug. Could the same not be said of the imprisoned sex offender? Could they not have an irrational desire to be released from prison, or to please the public by undergoing chemical castration? Maybe, but it’s difficult to see why this would have to be the case, unless we assume that no one in their right mind could ever volunteer for chemical castration. Surely in some cases the desire to avoid imprisonment and to be rid of compulsive sexual urges is not irrational? In those cases, the choice could be legitimately presented to the offender.

The third objection holds that for some people the sexual urges will not be autonomy-undermining and hence their removal will not be autonomy-enhancing. Consider once more the Frankfurtian account of autonomy, according to which autonomy involves consistency with second-order reflective desires. If there is a sex offender who endorses his sexual desires, in a second-order and reflective manner, then they will be untouched by Douglas et al’s argument.

But then so what? The argument was never that all sex offenders could and should be presented with the option of chemical castration, only that an identifiable class could be. Evidence also suggests that sex offenders that do not suffer from recognised paraphilias are less responsive to the drugs. For this class of sex offenders, the option should not be presented. This doesn’t defeat the autonomy-enhancing argument.

The fourth, and final objection, is that in some cases the existing conditions of incarceration are so awful that adding the option of voluntary chemical castration for a reduction in sentence is not a moral improvement. Douglas et al’s argument assumes that the continuing incarceration is itself a justifiable restriction of autonomy. In other words, it assumes that the autonomy baseline in the scenario is itself morally justified. But if this assumption is wrong — if the incarceration is actually unjustifiable — adding the option of chemical castration may simply make a bad situation even worse.

Consider the following analogy:

Kidnap: Alan kidnaps Bob’s son. He then offers to release the child if Bob pays a ransom.

Clearly, the release of the child would be autonomy-enhancing in this scenario. But, just as clearly, this doesn’t mean that Alan is justified in presenting that option to Bob. It might be that incarceration is like this. As Douglas et al put it, in at least some jurisdictions prisoners are subjected to all manner of ill-treatments, from rape and physical assault, to serious deprivation and forced labour. This would probably make the incarceration unjustifiable, and consequently the option of chemical castration improper.

Douglas et al accept the force of this criticism, and argue that if this is the case we should aim to reform the system of incarceration, not offer piecemeal reforms to it (such as offering individual prisoners the choice of chemical castration).


3. Conclusion
In summary, the current evidence on chemical castration is mixed. It may reduce recidivism rates among certain sex offenders, but further studies are required. But if we assume that it is effective, and that the side effects of long-term usage can be minimised, then we may have reason to think it ought to be presented as an option to a certain class of sex offenders. For not only would the medication have a social benefit in terms of reduced rates of recidivism, it would also have a personal benefit for the sex offender in terms of autonomy enhancement.

This is a provocative argument, and Douglas et al do a reasonably good job of defending it in their article. Still, even they concede that there may be cases in which its use is morally impermissible. This would be particularly true whenever the conditions of incarceration are themselves unjustifiable.