Autonomy: amorphous or just impossible?

By Charles Foster

I have just finished writing a book about dignity in bioethics. Much of it was a defence against the allegation that dignity is hopelessly amorphous; feel-good philosophical window-dressing; the name we give to whatever principle gives us the answer to a bioethical conundrum that we think is right.

This allegation usually comes from the thoroughgoing autonomists – people who think that autonomy is the only principle we need. There aren’t many of them in academic ethics, but there are lots of them in the ranks of the professional guideline drafters, (look, for instance, at the GMC’s guidelines on consenting patients) and so they have an unhealthy influence on the zeitgeist.

The allegation is ironic. The idea of autonomy is hardly less amorphous. To give it any sort of backbone you have to adopt an icy, unattractive, Millian, absolutist version of autonomy. I suspect that the widespread adoption of this account is a consequence not of a reasoned conviction that this version is correct, but of a need, rooted in cognitive dissonance, to maintain faith with the fundamentalist notions that there is a single principle in bioethics, and that that principle must keep us safe from the well-documented evils of paternalism. Autonomy-worship is primarily a reaction against paternalism. Reaction is not a good way to philosophise.

To  say that autonomy means respecting autonomous choice does not save it from all critics of its claims to be coherently normative.  One example: when X considers a potentially life-saving  treatment, whose autonomous choice is the decider? Is it X the father who wants to see his children grow up? X the husband who doesn’t want to be a burden? X the Jehovah’s Witness, fearful of hell?  X the fearful human, fearful of oblivion? Surely it is unusual to meet an X who is so well integrated as to be able to deliver a single, unequivocal answer to the question ‘What do you want?’ I’m not sure I’d like to meet such a person. Human beings (or at least the sort that you’d like to have to dinner) are defined by their contradictions and the exhilarating tensions between them.

But let’s assume that one can say that the definition ‘respecting autonomous choice’ does have a sufficiently substantive meaning to be theoretically useful. Can it in fact be honoured in the real healthcare environment?  No, says an important paper in the Journal of Medical Ethics by Agledahl, Forde and Wifstad. We have fatally failed to distinguish between the notion of autonomy in medical research (where one can have pretty pure, textbook cases of ‘informed consent’), and the same notion in the real world of medical treatment (where one can’t). We talk about a consultation with a surgeon in the same language as we talk about a consultation with the co-ordinator of a clinical trial. That language, which makes perfect sense in the RCT context, is gibberish when the subject turns to the question of the available options for managing your osteoarthritic hip.

Two examples from the paper make the point.

First : The route to a particular procedure rarely involves just one consultation between one clinician and one patient. Clinical practice is a ‘process over time and space in which several participants guide the actions that are taken…’ This, observe Agledahl et al, ‘makes it harder to define who has made the decision, identify who has the moral responsibility and understand what respecting a patient’s right to autonomous choice could mean.’

And second: patient choice is restricted by what is available in a particular healthcare system, and accordingly consultations necessarily direct patients towards the available choices. That’s not paternalism: it’s realism. ‘[C]ouldn’t I speak to [the] surgeons directly?’, asks a hypothetical young man in one of Agledahl et al’s examples. ‘Well….yes’, replies the doctor on call, ‘you could certainly call the hospital and ask for them yourself, but I’m not sure that you will be allowed to speak to them. Or else your family doctor could refer you straight to the district general hospital…[But] it will be quicker if we could examine you first and refer you based on our examination.’

The patient’s choice is being subverted. The respect given to his choice is trimmed because of the necessary organisation and financing of healthcare. One might add that even in a healthcare system unconstrained by any financial or organisational limitations, informed consent is practically impossible.  When a practitioner prescribes paracetamol, she simply cannot summarise all the complications that have been described in the literature, tailoring her assessment of the risk of that complication in the patient. The medical literature is just too vast for doctors to paint the picture for patients in anything other than a very broad, choice-truncating brush.

Agledahl et al conclude: ‘Respecting persons as autonomous persons is far more complex than eliciting choices and acknowledging informed consent….Respecting patients as autonomous persons is an important moral issue in clinical work, but it is misleading to relate this respect to the degree of choosing….[T]here is a danger that the current [autonomy-driven] discourse is creating a gap between medical practice and medical ethics, and that doctors are adapting a language from bioethics that does not properly portray their sense of moral responsibility.’

Quite right. We need urgently to disown the monolithic fundamentalism of pop-ethics, and embrace a truly liberal pluralism that listens respectfully to the voices of many principles. Proper pluralism isn’t incoherence. I’ll be suggesting that the harmonious joinder of all those voices is the sound of dignity.

  • Facebook
  • Twitter
  • Reddit

6 Responses to Autonomy: amorphous or just impossible?

  • Marco Antonio Oliveira de Azevedo says:

    Very good points. I agree that there is a difference between consent in clinical trials and consent in clinical practice. Nevertheless, it doesn’t mean that in clinical practice autonomy is less important. The difference is not between a context where autonomy applies absolutely and a context where the principle applies only prima facie, or not absolutely. Even if pluralism is true, in both contexts *autonomy* means that patients have, in some very general sense, a right to refuse: in clinical trials, a right to refuse the invitation to ingress in a study; in the context of a patient-physician relationship, a right to refuse any recommended treatment. In both cases we do not have a full autonomous (kantian) choice – or even a deliberative aristotelian choice. But in both cases, even if we have reasons to accept those invitations and medical prescriptions, they do not imply any kind of obligation to obey. They are not, obviously, *imperative* commands (actually, I think they are and cannot be commands in any sense). But, certainly, the relationships are very different. In the clinical trial case, we do not have any professional consultation in advance. We can say, if you please, that the two situations are phenomenologically distinct. The point is that in clinical practice, physicians are considered by their patients as expert authorities in the context of a consultant relation. Researches are not consultant relationships (that is, the subjects of a trial are not *patients*).

    • Charles Foster says:

      Marco. Many thanks. I think we agree. The crucial interests that are represented by what you call 'autonomy' are just as important to me as they are to you.

  • Peter Wicks says:

    At first glance autonomy seems to be a more concrete concept than dignity, so I’ll be interested to see how you define it and spin it out into the “harmonious joinder” we all want. Any chance of a sneak preview of the basic ideas?

  • Charles Foster says:

    Peter. Many thanks. The ideas will be published later this year by Hart in a book called 'Dignity in Bioethics and Law'. I give the thesis its first public outing at the Oxford-Mount Sinai Bioethics Colloquium in NYC tomorrow morning. In short, I adopt an account of dignity based on human flourishing. One doesn't necessarily need to do anything as a human (or have the capacity to do anything) in order to flourish in the sense in which I define the word. It is one's human being that matters. That doesn't at all mean that I adopt the traditional Catholic idea of dignity as status: I don't. In deciding what one should do or not do in bioethics (or anything else for that matter), one should strive to maximise the amount of human flourishing, taking into account the dignity/flourishing interests of all the stakeholders, including society, weighting the interests appropriately. It doesn't necessarily follow that the patient's interests will be weighted more heavily than those of anyone else.

    • Peter Wicks says:

      Looks interesting, thanks! I agree that the interest of the patient is not primordial – as you allude to in your post it makes prioritisation and sound management of resources impossible – so we should go for a more directly utilitarian approach (i.e. weighing the different interests involved). I'm interested in the association you make between the words "dignity" and "flourishing" – this indeed looks quite promising from my (still relatively naïve!) perspective.

  • Matt Sharp says:

    Charles, I was wondering what your thoughts are on autonomy with regards to prisoners. It seems obvious that autonomy and patient choice is already restricted within the prison system. By it's very nature, incarceration is meant to reduce at least freedom of *movement*, so patients (presumably?) can't choose to attend any GP or hospital they like.

    Should this extend to a reduction in freedom for a prisoner to choose whether to be treated at all? So, consider if a prisoner is either (a) diagnosed with what would be a terminal disease without treatment, or (b) attempts suicide by means of a hunger strike. In these cases, assume the prisoner decides they do not wish to be treated or force-fed. Also assume the prisoner has committed a crime with obvious victims. Is there a duty for the prison service and medical practitioners to keep the patient alive, for the sake of the victims of the crime? Should justice overrule autonomy?

    As an example, when Harold Shipman hung himself, the granddaughter of one of his victims stated "This seems like an easy way out for him. He never showed any remorse or any guilt and that door is now closed to us."

    Thanks for your time; I appreciate if you're unable to give a thorough answer. If there any papers on this area that you'd recommend, that would be most helpful.