Medical interventions such as methadone treatment for drug addicts or “chemical castration” for sex offenders have been used in several jurisdictions alongside or as an alternative to traditional punishments, such as incarceration. As our understanding of the biological basis for human behaviour develops, our criminal justice system may make increasing use of such medical techniques and may become less reliant on incarceration. This topic has great practical and current significance across jurisdictions. The Lord Justice Clerk, Lord Carloway, has recently stated that the Scottish criminal justice system should explore greater use of alternatives to prison [1]. Prison is phenomenally expensive. It costs £40,000 to keep a single inmate in prison in the UK for only one year [2]. This is more expensive than educating a boy at Eton College for the same period. In the United States, the MacArthur Foundation has made $75 million available for research and interventions relevant to reducing the prison population ( Approaches to criminal behaviour that could effectively rehabilitate offenders—such as medical interventions to reduce aggression—could spare offenders and potential victims suffering and generate economic savings. This economic benefit would be ethically desirable if the money were used to promote citizens’ welfare, e.g. if it were invested in further crime prevention measures. However, the use of medical interventions to rehabilitate offenders is likely to be opposed if they are thought to violate offenders’ rights.

Academic debate on this topic has largely focused on whether offenders can validly consent to medical interventions, given the coercive environment of the criminal justice system. Both sides in this debate share the assumption that administering medical interventions to offenders without their valid consent would be unethical [3, 4]. Recently, Thomas Douglas has mounted a formidable challenge to this “consent requirement” [5] :104]. Since support for the consent requirement is so widespread, if Douglas’s challenge succeeds, this would have major implications for thinking on this topic. Essentially, his argument rests on a comparison between prison and medical interventions. Douglas asks: if the state is entitled to impose a prison sentence on a criminal without the criminal’s consent, why is consent required for the imposition of a medical intervention?

Some theorists might question the premise that mandatory imprisonment is legitimate, given that the harsh conditions prevalent in many of today’s prisons pose serious risks to offenders’ physical and mental health (cf. Barn, this volume). To address this concern, Douglas adds the qualification that, for the purpose of his comparison, he is envisaging a type of incarceration that restricts free movement and association but otherwise poses no health risks. He calls this “minimal incarceration” [v:105]. He also stipulates for the sake of simplicity that the medical intervention being considered is the injection of a drug [v:106]. Another initial objection against Douglas’s comparison is that imprisonment and medical interventions are imposed for different reasons. The aim of prison may be retribution, whereas it is generally believed that the goal of medical interventions is rehabilitation. However, Douglas notes that many theorists would accept that rehabilitation is also a goal of imprisonment [v:106]. His argument is addressed to those theorists. So Douglas’s question, framed more precisely is: if the goal of rehabilitation justifies mandatory minimal incarceration, why does it not also justify the mandatory injection of a drug?

The most obvious way of defending the consent requirement against Douglas’s challenge appeals to the fact that incarceration merely interferes with the right to free movement, but medical interventions interfere with the right to bodily integrity.Footnote 1 This argument rests on what Douglas calls the “robustness claim”—the claim that the right to bodily integrity is more robust than the right to freedom of movement [v:110]. In other words, the right to freedom of movement loses its protective force in a wider range of circumstances than the right to bodily integrity. Douglas’s article seeks to undermine the robustness claim, by arguing that neither case-based intuitions, nor theoretical considerations support this claim.

In this article, I will attempt to raise some doubts about Douglas’s challenge to the consent requirement and the robustness claim.

Case-Based Intuitions

Our intuitive reactions to the scenario under discussion—mandatory injection of prisoners—are unlikely to take us very far. Claiming that this practice is intuitively unacceptable will lack “dialectic force” against proponents of mandatory interventions [v:111]. Douglas also casts doubt on the reliability of such intuitions—they may just reflect wariness about the unfamiliar. Although we are accustomed to the idea of employing injections to treat/prevent illnesses, we are relatively unfamiliar with the idea of using them to address criminal behaviour. Perhaps mandatory imprisonment feels more acceptable than mandatory medical interventions merely because prison is a “traditional” response to offending. Therefore, he proposes getting an indirect bearing on the issue, by examining our intuitive reactions to analogous cases that still involve comparing the robustness of rights to free movement and bodily integrity, but that do not raise the above-mentioned problems. He compares the following cases [v:112]:

Epidemic Scenario

Case 1

Imposing mandatory quarantine for three months is the only way to prevent Jill, who has a highly dangerous infectious disease, from starting an epidemic. This would infringe her right to freedom of movement.

Case 2

Injecting Jane with a drug that renders her non-infectious (but does not treat her illness) is the only way to prevent the epidemic. This would infringe her right to bodily integrity.

According to Douglas, this scenario seems to undermine the robustness claim, as interfering with Jane’s right to bodily integrity (in Case 2) seems just as permissible as interfering with Jill’s right to freedom of movement (in Case 1). However, there are two ways in which our reactions to these cases may not be relevant to assessing the robustness claim in the context of addressing criminal behaviour.

Firstly, the example involves an emergency situation and our intuitions about what should be done in emergency situations may not be a good guide to how social institutions (such as the criminal justice system) should be regulated in ordinary circumstances. Emergencies are typically time-limited and involve such a grave threat to the existence or functioning of society that normal moral principles and rights may need to be temporarily suspended. For instance, emergencies can justify martial law, which is not normally acceptable. In contrast, (even serious) criminal behaviour is probably an inevitable feature of society under normal conditions. Furthermore, emergencies typically require the state to act quickly in order to avert an imminent danger, whereas a criminal who is awaiting punishment does not pose an immediate threat and the state has more time to deliberate. Therefore, our intuition that an individual’s right to bodily integrity should be infringed in exceptional circumstances to avert a catastrophe does not necessarily tell us what protection should be afforded to this right in the day-to-day running of the criminal justice system.

Secondly, Douglas’s way of framing the scenarios does not allow us accurately to assess the relative robustness of the right to free movement compared with the right to bodily integrity. Both of his hypothetical cases involve a choice between only two options. In case 1 the options are: a) allow the epidemic to occur or b) interfere with the right to free movement. In case 2 the options are a) allow the epidemic to occur or b) interfere with the right to bodily integrity. However, just because both the right to free movement and the right to bodily integrity can be overridden by an extremely serious countervailing consideration (such as the need to prevent an epidemic) does not show how easy it is to override the right to free movement relative to the right to bodily integrity. Consider the following analogy: knowing that Charles is much taller than both Bob and Andrew, does not allow you to compare the relative height of Bob and Andrew.

It is also worth noting that, when it comes to the topic at issue—administering medical interventions to prisoners—the state has more than two options [6]. It could impose no punishment, impose a “traditional” punishment such as prison or administer a medical intervention.Footnote 2 The state’s position is therefore more analogous to the two scenarios that I will outline below than Douglas’s epidemic scenario, which involves a stark choice between doing nothing to avert a negative outcome and infringing bodily integrity.

A type of scenario with a better chance of generating intuitions that are relevant to the robustness claim would have the following characteristics. It involves a threat that did not amount to an emergency, where three options were available: 1) doing nothing to avert a threat, 2) averting the threat by infringing the right to bodily integrity and 3) averting the threat by infringing the right to free movement. If option 3 would be preferable to option 2, this would provide support fort the robustness claim. Here is one such example:

Discipline Scenario

The parents of Anne, who is 11 years old, discover that she is engaging in highly risky behaviour. Assume that there are two forms of discipline that are likely to prevent Anne from continuing to take these risks—grounding her or administering corporal punishment—and that they will probably be equally effective. Which option is preferable?

  1. 1.

    Allow Anne to engage in highly risky behaviour without any interference.

  2. 2.

    Ground Anne (i.e. interfere with free movement).

  3. 3.

    Administer corporal punishment to Anne (i.e. interfere with bodily integrity).

Arguably, grounding is the best option, corporal punishment the next best and doing nothing the worst option. A plausible explanation for this intuition is that interfering with free movement is preferable to interfering with bodily integrity. It might be objected that there are alternative explanations for the intuitive response to this case. For instance, it might be thought that corporal punishment would inflict more distress than grounding. However, that explanation is not convincing. Anne might well find grounding more distressing than corporal punishment, e.g. if being grounded meant she would miss her friend’s birthday party. Another competing explanation for our intuitive preference for grounding is that people might reject the stipulation that corporal punishment is effective. Yet, even if this is one consideration that counts against corporal punishment, the idea that infringing bodily integrity is more disrespectful than interfering with free movement is arguably a more important consideration [7]. In support of this point one might consider the fact that corporal punishment of adult criminals, e.g. flogging, was rejected not merely on the basis that it is ineffective. (It is not clear that our current practices—e.g. prison, are much more effective in reducing reoffending. Nor is it clear, to return to the previous objection, that flogging would necessarily be more distressing than a long prison term.) A more compelling reason for rejecting corporal punishment is that interfering with bodily integrity is more disrespectful than interfering with freedom of movement and therefore that the right to bodily integrity is more robust than the right to free movement.

The abandonment of corporal punishment of adults (and the growing opposition to the corporal punishment of children) is particularly relevant to the question at issue—the acceptability of the mandatory injection of prisoners. Douglas claims that opposition to such injections may be based on unreliable intuitions that are simply the result of our unfamiliarity with this method of responding to criminal behavior. However, society is familiar with the more general issue of responding to misconduct of adults and children by interfering with their bodily integrity through corporal punishment, so our intuitions about that matter are more likely to be reliable. Intuitions about corporal punishment seem to lend support to the idea that the right to bodily integrity is more robust than the right to freedom of movement. Therefore, these intuitions provide some indirect support for the claim that acknowledging the state’s right to imprison offenders on a mandatory basis (i.e. to infringe free movement) does not entail that the state also has the right to administer injections to prisoners on a mandatory basis (i.e. to interfere with bodily integrity).

The discipline scenario is not the only thought experiment that lends support to the robustness claim. Consider the following scenario:

Driving Scenario

Joe has a medical condition that causes him unpredictably to lose consciousness while driving. He lives in an isolated cottage in a remote area and relies on his car to meet friends and participate in activities he enjoys (although he does have access to medical and emergency services and can get food delivered etc. without his car). Imagine that the authorities only have the following three options. Which option is preferable?

  1. 1.

    Allow Joe to put other road users at risk, without any interference.

  2. 2.

    Compel Joe to stop driving.

  3. 3.

    Forcibly inject Joe with a drug that does not treat his underlying illness, but does prevent him from losing consciousness for the duration of his car journeys.

It seems that, all other things being equal, it would be preferable for the authorities to compel Joe to stop driving than for them to forcibly inject him. A plausible explanation for this intuition is that interfering with free movement is preferable to interfering with bodily integrity. It is unlikely that intuitive opposition to option 3 is merely the result of unfamiliarity with the practice of injecting people in such circumstances. We are, after all, familiar with the general issue of using injections to alleviate medical symptoms. Alternatively, it might be objected that society’s attitude towards compulsory injections is not as clear-cut as the above scenario suggests. There are some professions, e.g. health workers, where injections are required as part of the job [v:115]. However, such examples are quite rare and consent to the injections is, in fact, given at the stage when the health worker agrees to the terms of the employment contract. Also, unlike the issue of injecting criminals, the party offering the injection has only two feasible options: allow the health worker to put patients at risk or make vaccination a requirement of the job. Whereas, in the criminal justice context, the state has at least three options: imposing a traditional punishment, a medical intervention or no punishment at all. Furthermore, the health worker still has a choice: she could choose another job. However, Douglas’s argument suggests that prisoners should be deprived of any choice in the matter—mandatory injections would mean that the offender could not even opt to stay in prison as an alternative. Young children may be given injections without their consent [v:115]. However, this is disanalogous to the criminal justice situation for at least two reasons. Firstly, children are given injections for their own benefit, whereas the main purpose of injecting offenders is to benefit others.Footnote 3 Secondly, children may lack the cognitive abilities and life experience to be able to judge for themselves what is in their own best interests, whereas offenders are autonomous adults who possess the relevant abilities and life experience.

Another example Douglas offers in defence of his position is the forcible injection of mentally disordered individuals who pose a danger to themselves or others, but nevertheless have sufficient mental capacities to be competent to refuse or grant consent [v:113].Footnote 4 He points out that in many jurisdictions the legal requirements that must be satisfied before such a person may be forcibly injected are no more onerous than the preconditions for interfering with the person’s freedom of movement [8]. Admittedly, this is a particularly challenging case for the defender of the robustness claim. My earlier criticisms of the epidemic scenario (the availability of only two options and the exceptional nature of the scenario) do not apply to Douglas’s mental disorder example. The state generally does have more than two options when dealing with dangerous mentally ill individuals: inaction, compulsory drug treatment, or compulsory detention. Furthermore, responding to severe mental illness is a problem that society must face under normal conditions—it is not an exceptional, emergency scenario. However, the mental disorder example may face difficulties of its own. Two possible strategies could be adopted to defend the robustness claim against this example. The first strategy is to deny that our current practices regarding the mentally ill are justified. The second strategy is to defend our current practices, but to distinguish them from the compulsory injection of sane offenders.

A proponent of the first strategy might point out that society has traditionally taken an excessively paternalistic attitude towards people with mental disorders. It has been slow to recognize that mental disorder need not preclude the ability to make autonomous decisions about one’s own treatment. Excessive paternalism towards people with mental disorders could explain why society has, on the one hand, been willing to disregard a competent mentally ill person’s preference for physical detention over drug treatment; while, on the other hand, society has generally been unwilling to countenance depriving sane, autonomous offenders of this choice. However, those who value the right to bodily integrity should be troubled by the fact that a genuinely competent person’s preference for physical detention over drug treatment can be ignored simply because this person suffers from episodes of mental illness. Furthermore, disregard for the individual’s preference is particularly troubling if a reasonable case can be made for the preference. If indeed it is unjustifiably paternalistic forcibly to inject competent people who suffer from mental illnesses, despite their preference for detention, then our misguided acceptance of this practice cannot be used to justify forcible injection of sane offenders.

Turning to the second strategy, our current practice of forcibly treating certain severely mentally ill individuals, regardless of their apparent competence to refuse consent, might be justified on epistemic grounds. In a situation where we must weigh the individual’s best interests (and also potentially the interests of others) against the value of respecting the individual’s choices, and where there is uncertainty about whether that choice is really autonomous, the balance might tip in favour of promoting best interests. If drug treatment is a) indisputably in the individual’s best interests and b) we know that the individual suffers from a severe mental illness, we might doubt whether that individual is genuinely competent to refuse treatment (even during periods of apparent lucidity). It might be thought that the very nature of the situation just described necessarily gives rise to at least some doubt about the individual’s competence to refuse treatment. If this is correct, and if this degree of doubt about competence would tilt the balance in favour of the best interests considerations, this could explain why the law disregards the refusal of consent in this situation. The more confident we can be that an individual is competent to refuse treatment, the weaker are the paternalistic arguments for imposing mandatory treatment on that individual. This strategy would allow the case of mentally ill patients to be distinguished from the case of sane offenders on the basis that there is less reason to doubt that the latter are competent to refuse treatment.

These strategies are mutually supportive. The most plausible criticism of the first strategy is probably the idea that a mentally ill person who is a danger to herself or others should not be permitted to opt for detention over treatment if treatment is indisputably in that person’s best interests. However, the more one is convinced that refusing treatment in a given situation would be wholly unreasonable, the more plausible the second strategy becomes, i.e. the more grounds there are for doubting that the individual genuinely is competent.

A third strategy could also be adopted that involves distinguishing mentally ill people from sane offenders on the basis that treatment of the former is intended to restore their rationality, [9, 10] whereas sane offenders are already rational and treating them is intended to improve their moral motivation. The right to bodily integrity implies that individuals’ autonomous refusal of treatment should be respected. However, if a mentally ill person’s refusal of treatment would mean that in the future she will become non-rational and non-autonomous, then arguably it shows greater respect for her autonomy to impose treatment. Footnote 5 In contrast, the same argument could not be made for imposing mandatory moral enhancement on offenders unless one adopts the controversial meta-ethical position that immorality is a kind of irrationality. Indeed, many crimes could actually be considered rational in the sense of maximizing the preferences of the criminal. Also, mentally ill people could be distinguished from sane offenders on the basis that interference with the liberty of the former is justified for paternalistic reasons (i.e. to safeguard the interests of the person treated), whereas punishment might have a different justification (e.g. to promote the interests of society). It should also be emphasized that paternalistic arguments are rarely considered capable of justifying the mandatory treatment of competent individuals and could only do so in exceptional circumstances and for limited periods of time.

To summarise: this section critiqued two examples advanced by Douglas—the epidemic scenario and the compulsory treatment of the mentally ill. These cases purport to undermine the robustness claim—the idea that the right to bodily integrity is more robust than freedom of movement. This, in turn, supposedly undermines the “consent requirement”—our current practice of requiring consent for the administration of medical treatment (despite our willingness to impose mandatory incarceration on offenders). I have argued that these examples do not necessarily undermine the robustness claim or the consent requirement. I have offered two examples of my own—the discipline scenario and the driving scenario that seem to generate intuitions that support the robustness claim and that provide indirect support for the consent requirement.

Theoretical Considerations

As well as discussing case-based intuitions, Douglas responds to some more theoretical arguments that might be advanced in support of the claim that the right to bodily integrity is more robust than the right to free movement. He anticipates that defenders of the robustness claim will point out that certain rights do seem more robust than the right to free movement. For instance, rights against killing, torture and public humiliation seem particularly robust. Douglas therefore considers whether the considerations that underpin these seemingly robust rights also underpin the right to bodily integrity. He discusses two considerations - harm and threats to agency.


I do not claim that the special robustness of the bodily integrity rests primarily on the idea that infringements of this right (e.g. through compulsory injections) are especially harmful. This is an empirical question, which we do not have enough data to answer. However, I will attempt to cast doubt on Douglas’s arguments that compulsory injection of prisoners is unlikely to cause especially serious harm.

Douglas claims it is doubtful that compulsory injections of prisoners would be experienced in seriously negative ways, since healthcare workers and children do not find their experience of compulsory (or strongly encouraged) vaccination an extremely negative experience [v:115]. However, there are reasons for thinking that prisoners might experience these injections differently from children or healthcare workers. As, noted above the injection of healthcare workers is not truly compulsory (in the way that injection of offenders would be under Douglas’s proposal) as healthcare workers could always choose another job.Footnote 6 The injection of young children can be compulsory, but it is plausible that adults might react differently to compulsory injections. Adults have grown to expect that they will have control over their own bodies and often find it highly distressing to be treated like children. In addition, vaccination of children and healthcare workers is intended to prevent physical illnesses. The thought that the injection is for one’s own physical welfare might reduce the distress of the procedure. In contrast, the aim of injecting offenders is to affect their future decisions and moral motivation, mainly in order to benefit others. The offender is likely to doubt whether the injection will promote his own welfare. Furthermore, the thought of being forcibly injected with a chemical that alters one’s motivations in a way that bypasses ones rational faculties could be particularly distressing [11].Footnote 7

A closer analogy is the example, discussed above, involving the compulsory injection with a mind-altering drug of a mentally ill person who is a danger to herself or others. However, this example tends to support the proposition that such injections are particularly distressing and that we should not deprive the individual of the option of detention. Empirical research suggests that, “most patients who had experienced some form of behavioural control clearly preferred the use of seclusion or physical restraint to being forced to take medication. Moreover patients’ opinions were established when they were in remission from the mental disturbance that had led to their confinement so their views were not clouded by psychotic symptoms or emotional arousal” [12] :146, citing [13].

Threats to Agency

Douglas distinguishes between two ways in which interfering with bodily integrity might be thought to threaten agency. Because the mind is “dependent on and influenced by the body” interfering with the body might pose a “causal” threat to agency by impairing the capacities required for agency [v:116]. Alternatively, interfering with B’s bodily integrity might pose a “communicative” threat to agency by expressing “the proposition that B is not an agent, or at least, that A does not care whether B is an agent or not” [v:116]. According to Douglas, it is plausible that torture, killing and public humiliation threaten agency to a greater extent than incarceration (i.e. interfering with free movement). This could explain why the rights that protect against such torture, killing and public humiliation are more robust than the right to free movement. However, he questions whether forcibly injecting prisoners would threaten agency more than imprisonment would.

I agree with Douglas that injecting offenders with a drug designed to rehabilitate them need not necessarily pose a causal threat to agency. It might happen to undermine agency, depending on how the drug operated [14]. If, for example, the drug instilled in offenders a fear of reoffending that was so powerful that they were unable to choose to reoffend then this might restrict their agency. However, if it merely altered the offender’s preferences regarding reoffending so that they more closely resembled the preferences of a typical non-offender, then it seems the offender would not be less of a free agent than the non-offender [15].Footnote 8 Nevertheless, the offender’s belief that the forcible injection of a mind-altering drug would pose a causal threat to agency might result in great distress. This supports the harm-based objection against such forcible injections, discussed previously. It might be argued that officials could relieve this distress by explaining to offenders that the injection would not undermine any of the capacities required for free decision-making. However, in reality, if one were faced with the prospect of being forced to undergo a medical procedure that would alter one’s own thought-processes, even rational arguments about the benign nature of such interventions are still unlikely to be very reassuring.

Perhaps the most significant problem with mandatory injection of prisoners is that it poses a communicative threat to agency. Pinning someone down and forcibly injecting her with a mind-altering drug is likely to send out a more disrespectful message about that individual than incarceration would. Violating bodily (and mental) integrity sends out a disrespectful message, because it invades a particularly intimate sphere. The individual’s body and mind are constitutive of the person and invading the mind and body therefore amounts to a fundamental attack on the person, in a way that interfering with free movement does not (Shaw, 2016, Against the mandatory use of neurointerventions in criminal sentencing. in Birks, D. & Douglas, T. (eds), Treatment for crime: Philosophical essays on neurointerventions in criminal justice. Oxford University Press, unpublished).Footnote 9

The claim that interfering with bodily integrity is particularly disrespectful is supported by the fact that society has historically moved away from physical punishment of adults (and, as noted above, of children), preferring constraints on free movement such as imprisonment (or in the case of children grounding, or detention). Admittedly, as Douglas claims, killing, torture and public humiliation are even greater threats to agency than the kind of forcible injection under discussion. A closer analogy with forcible injection might be non-public flogging of moderate intensity that does not inflict serious physical injuries. If such flogging is less acceptable than even long-term incarceration (as our legal system assumes) this suggests a similar basis for preferring incarceration to forcibly administered injections. It might be objected that the problem with the kind of flogging just described is the distress it inflicts rather than the disrespect it communicates. However, it is far from clear that such flogging would necessarily inflict more distress than long-term incarceration, even under good prison conditions. Arguably, the forced injection with a mind-altering drug would even involve an additional kind of disrespect that is not present in the flogging example [16]. The injection sends out the message that the offender’s moral motivation is so deficient that it needs to be directly re-engineered. In contrast, flogging engages the offender’s agency as it is—either by providing him with a prudential reason to refrain from reoffending, if it is administered as a deterrent, or by responding in a retributive way to his free choice to do wrong. It is also worth noting that many considered the forced psychiatric treatment of dissidents in the former Soviet Union to be particularly sinister—wrong in a distinctive way—when compared with incarceration. Compulsory treatment was not sinister purely because of the distress it inflicted, but because it showed utter contempt for the dissident’s status as an agent.

Douglas claims that an action only poses a communicative threat to agency if it reflects contemptuous attitudes held by the person who performs that action. However, this is an overly subjective interpretation of the meaning of actions. Actions have a social meaning that is to some extent independent of the personal motives of the individual who performs the action. To return to the flogging analogy, an official responsible for administering floggings might not deny that offenders are moral agents. He might personally view flogging as a way of communicating moral censure to a responsible agent who freely chose to do wrong and who (the official believes) will understand the message of censure being communicated. He need not, personally, be contemptuous of prisoners. Perhaps he thinks that they faced strong temptations or pressures that he himself might have succumbed to had he been in their position and, if that had been the case, he would have been willing to be flogged himself. Yet this official’s personal views would not prevent flogging from communicating a disrespectful message about the offender. Footnote 10

Social meanings can of course change. If such a change occurred with regard to the forcible injection of offenders, one might think that it was acceptable at that point to engage in this practice. However, it is difficult to imagine that “re-education” alone would effect such change. This change would, as noted above, represent a reversal of the historical trend away from using state-sanctioned interference with bodily integrity as a way of enforcing social/legal conformity. The fact that non-consensual interference with bodily integrity typically communicates disrespect is unlikely to be an arbitrary convention. Gardner and Shute give the example of a woman who is raped while unconscious and suffers no adverse physical effects and never discovers what has happened to her [17]. What makes this rape still seriously wrong is the fact that it treats the victim with profound disrespect. This point is not confined to rape. Imagine a similar scenario where skilful surgeons perform a procedure on an anaesthetised patient despite knowing the patient would not consent to the particular procedure that had been carried out. This still seems disrespectful even if it posed no risk of negative physical consequences and the victim never discovers what happened. The disrespect involved in these cases does not seem to be the result of an arbitrary convention. The law could not change the meaning of these actions in the way that it could change road traffic rules, which are merely conventional. The disrespect involved in these cases is much more extreme than similar interferences with freedom of movement. Imagine locking someone in a room while they are asleep, without causing physical harm and without the “victim” ever discovering that this had been done. It might be objected that the reason why the examples involving non-consensual interference with bodily integrity communicate more disrespect than the example of the locked room is that the former is a greater violation of the individuals’ preferences – greater because the preferences are stronger. However, this objection would be question begging, because the greater intensity of the preferences arguably reflects the more serious view we take towards violating bodily integrity, compared with infringing free movement. Alternatively advocates of mandatory injections may propose implementing this practice first as a means of changing the attitudes towards it. If this approach were taken then administering forcible injections during a period when attitudes were still opposed or transitioning would risk violating the rights of individuals treated during this period. Furthermore, it is doubtful whether changing attitudes to the mandatory injection of offenders would be desirable. It is difficult to imagine altering such attitudes in a compartmentalised fashion. Eroding the significance of the right to bodily integrity in the context of injecting offenders carries a high risk of eroding the significance accorded to that right in other contexts as well, which might have adverse unintended consequences.

To summarise: In this section, I have discussed two considerations that could provide a basis for a) the claim that the right to bodily integrity is more robust than the right to free movement and hence a basis for b) the practice of requiring that offenders give consent before medical interventions are permissible, while still imposing incarceration on a mandatory basis. These considerations are that interfering with bodily integrity (through medical interventions) can cause more harm than interfering with free movement (through incarceration) and that the former can also pose a greater communicative threat to agency than the latter—i.e. communicates a message of disrespect for agency. I have argued that there is more reason for taking the harm consideration seriously than Douglas acknowledges. However, the more significant consideration, in my view, is the communicative threat to agency that I have argued would arise from mandatory medical treatment of offenders.


Douglas has highlighted an issue of great theoretical interest and practical significance. There is widespread academic acceptance of the proposition that mandatory incarceration of offenders is acceptable, while mandatory medical rehabilitation is unacceptable. However, the moral basis for this position has not been thoroughly examined. If the current position were not well founded then that would have important implications for the criminal justice system. Douglas has mounted a very thought-provoking challenge to the current consensus. In this article, I have attempted to raise some doubts about this challenge. I do not claim to have provided a complete defence of the current stance regarding mandatory medical treatment versus mandatory incarceration. It is to be hoped that more work will be done in this area in the future [18].