Public defence: Ainar Miyata

Humans are neither purely nor perfectly rational creatures. Sometimes we are influenced by things that we ought not (rationally) be influenced by, such as whether a piece of information is framed in terms of loss or in terms of gain, or whether an item costs 8$ or 7.99$. Other times our reason is bypassed completely, such as when we are affected by music. 

Such influence, whether it happens intentionally or not, matters. It matters for what people end up believing, wanting, feeling, and doing. It affects everything from matters of love—how you choose your partner, say—to matters of life and death like in the doctor-patient relationship.

This thesis is an examination of the moral status of such non-rational influence, viewed primarily through the lens of the latter example (though the example of love shows up, too— Paper 3 for those who are interested).

Nudging

When one uses non-rational influence intentionally, it is sometimes called nudging. It has been suggested that, because patients don’t always choose wisely, doctors ought to nudge them toward better choices. As I argue in the first paper that makes up this thesis, this is probably not a good idea in practice. 

Although commonly raised arguments against such nudging fail—nudging isn’t incompatible with respecting autonomy or with informed consent, as is sometimes claimed—a doctor nudging their patient toward what the doctor thinks the patient ought to do is not likely to make things go better for the patient. This is because it is seldom the case that there is only one medically sensible thing to do, and among the options that are on the table, doctors from different hospitals and different countries, would disagree on what is best. 

The only thing nudging in such situations would serve is to enhance the influence of the personal opinion of one doctor at the expense of the opinion of the person whose preferences matter the most in such a situation, i.e., the patient.

Principles

This doesn’t mean all nudging is bad, however. Quite to the contrary: although nudging toward specific treatments is usually a bad idea, nudging the patient toward a better state of understanding, such as by presenting statistics in a way that is easier to parse is usually a good idea. Such nudging is a great way of following what I call the principle of epistemic beneficence, according to which doctors ought to put their patients in the best possible epistemic position to make the best choice in their situation.

Further, as long as doctors make their patients’ understanding worse, i.e., as long as doctors abide by what I call the principle of epistemic non-maleficence, doctors ought to nudge their patients in those relatively rare situations where there is only one medically sensible course of action, for example, if the patient is hesitant about doing some exercise or washing their hands.

Bumping

When non-rational influence is unintentional, I call it bumping rather than nudging. I believe that the moral significance of bumping has been overlooked for two reasons: 1) the unintentional is equated with the random, and 2) unintentional action is assumed to be inappropriate for moral appraisal, like praise or blame. In the second paper in the thesis, I argue that both 1) and 2) are false. 

Even though bumping is unintentional, it can be systematic, and it is likely that doctors systematically bump their patients. Moreover, doctors who do so can be morally responsible if they harm their patients in this way.

Theories of influence

Non-rational influence is a big topic, not the least because influence itself is a big, philosophically under-explored topic, despite its centrality to human interaction. 

Throughout this thesis, and especially in the introductory synthesis, I have attempted to answer some fundamental questions both about influence simpliciter and non-rational influence. This has resulted in an exploration of theories of rationality and reasons, as well as a relatively extensive typology of influence.

The type of influence known as manipulation is an important part of the typology both because of its relevance to nudging and its connection to moral matters in general. 

In the third and final paper of the thesis, I develop a new theory of manipulation, which I call the attitude-sensitive account of manipulation. I identify two different and significant senses of the word “manipulation” that can be captured with their respective attitude-sensitive accounts. In one of these senses, nudging turns out to always be manipulative, but in the other sense it is not necessarily so, and that is the most morally relevant sense.