Strawson on free will and interpersonal relationships

In his classic paper “Freedom and Resentment,” P. F. Strawson addresses the question of what difference the widespread acceptance of determinism would make to our everyday ways of dealing with each other.  He judges that our commonsense conception of human behavior is too deeply rooted in our nature to be dislodged even in such a scenario.  As in Strawson’s other work, he urges attention to details of ordinary experience that are ineliminable but often overlooked by revisionist systems of metaphysics.

Strawson distinguishes between what he calls the participant reactive attitudes that are the norm in interpersonal relationships, and the objective attitude that can in exceptional cases supplant it.  Let’s begin with the first set of attitudes, which can be understood by reference to the different ways we can harm or benefit each other, and the different ways we react to these different harms and benefits.

Consider two occasions, one in which someone deliberately pushes you, and the other in which someone pushes into you by accident (as a result of tripping, say).  The physical harm caused to you might be the same in either case.  But the first person’s action will generate in you a feeling of resentment that the latter person’s will not.  Or consider a case where you are stranded and a stranger gives you money for cab fare, and also a case where you are stranded and find some money that someone has lost and with which you can pay for a cab.  The benefit to you will be the same in either case, but the first will produce in you a feeling of gratitude that the second will not.

Actions that produce in you a feeling like resentment do so because you take it that what moves the other person to perform them are factors like malevolence, rudeness, indifference, contempt, or an intention to insult.  Actions that produce in you a feeling like gratitude do so because you take it that what moves the other person to perform them are factors like kindness, goodwill, affection, and esteem. 

Consider also forgiveness.  Strawson notes that this presupposes that the action being forgiven was indeed appropriately met with resentment, but that the guilty party acknowledges this and repudiates the action.  The forgiving person accepts this repudiation on the part of the offender and, in turn, cancels the resentment.

Resentment, gratitude, and forgiveness are “participant reactive attitudes” insofar as they involve reacting to other human beings as fellow participants in a common interpersonal form of life.  We see each other, not as impersonal forces, but as rational agents having various beliefs, desires, and other propositional attitudes (as philosophers call them), as acting on the basis of these and thus subject to moral and rational appraisal, and as, accordingly, to be held responsible for the things we choose to do on their basis.

Within this general framework, various specific kinds of relationships exist – between family members, friends, lovers, colleagues, those who have some interest in common (such as a political cause or hobby), those we deal with temporarily in chance encounters (such as a grocery store employee or a clerk at the post office), and so on.  The nature of these relationships determines the specific expectations we have of people, the ways they might offend or please us, etc.  But they are all variations on the same basic theme of regarding others as fellow persons or rational agents, to whom it is appropriate to take the various participant reactive attitudes.

Now, focusing on resentment in particular, Strawson says that there are, within this commonsense framework for dealing with others, two sorts of cases where resentment at another person’s action can be removed or at least mitigated.  First, there are the cases where, after more carefully considering the matter, we make judgments of the kind expressed in sentences like “He didn’t mean to,” “He didn’t know,” “He couldn’t help it,” “He had no alternative,” and so on.  We thereby absolve the person of responsibility.  However, Strawson emphasizes, this is not because we judge that the participant reactive attitudes (resentment, gratitude, forgiveness, etc.) are not after all applicable to the person.  On the contrary, we believe that they are in general applicable to him, that he is a fellow member of the community of rational agents.  Rather, we simply judge that in the particular case at hand, the attitude of resentment is not after all appropriate.  For though we initially supposed that the person’s action was the result of ill will, we later come to judge that it was not.

Second, though, there are cases where we drop the attitude of resentment because we judge that the other person is deranged, or acting under compulsion, or that “He wasn’t himself,” or that he is just a small child.  In other words, we attribute the person’s behavior to psychological abnormality or immaturity, and for that reason judge that he does not bear the sort of responsibility required in order for an attitude of resentment to be reasonable. 

Now in this sort of case, Strawson says, we do not treat the person in question as a normal fellow member of the community of rational agents, to whom it is appropriate to take the participant reactive attitudes.  Here is where the “objective attitude” comes in.  It involves not dealing with another in the ordinary interpersonal way, but instead as something to be managed, treated, cured, handled, trained, controlled, or the like.

Think of the way we regard wind or rain that damages the house, or a stray neighborhood cat that leaves its droppings on the lawn.  We might get angry at these things, but we don’t literally resent them, because we don’t conceive of them in the personal terms that alone could make such an attitude intelligible.  Nor, for the same reason, do we literally feel gratitude toward wind and rain that pass without leaving damage behind, or a cat that does its business somewhere other than the lawn.  We manage or control a situation involving wind, rain, or a cat, rather than dealing with them the way we would a fellow member of the community of rational agents.

When we take what Strawson calls the “objective attitude” to another human being, we do something similar.  We judge that the person has fallen from normal interpersonal capacity (in the case of severe psychological abnormality) or that he hasn’t yet risen to it (in the case of immaturity).  And thus we see his behavior as something to manage or control rather than to feel resentment, gratitude, forgiveness, etc. toward.

Now, suppose belief in determinism became widespread in society, rather than something that only intellectuals thought much about.  It might seem that the result would be the disappearance of the participant reactive attitudes.  For those attitudes presuppose responsibility, and determinism might seem to undermine the view that people are responsible.  But Strawson thinks that while it may be possible in theory to abandon the participant reactive attitudes, in practice it could never happen. 

For one thing, he notes, accepting determinism would not entail abandoning attitudes like resentment for the usual reasons we cancel our resentment against another person.  Again, there are two such reasons.  First, we might judge that the person was after all acting out of goodwill rather than ill will.  But accepting determinism would hardly lead us to conclude that all people are really after all always acting out of goodwill!  It would not entail that anytime anyone does anything, it’s because he didn’t know any better, didn’t mean it, etc.  Second, we might judge that a person was either psychologically abnormal or immature and thus not responsible.  But accepting determinism would hardly lead us to conclude that all people are after all psychologically abnormal or immature!  So, again, determinism would not entail that we should extend our usual reasons for denying that someone is responsible for an action to all people and all actions. 

Strawson’s point here seems to be that it is the context of our ordinary participant reactive attitudes that gives sense or intelligibility to our decision not to hold someone responsible.  And that context makes the denial of responsibility the exception rather than the rule.  What we cannot plausibly do is extend this denial of responsibility so that it becomes the rule rather than the exception.

Strawson also suggests that to abandon the participant reactive attitudes would entail a level of “human isolation” that we are simply incapable of bearing.  Imagine trying consistently to adopt the objective attitude across the board to all human beings and all their actions – seriously regarding them as if they were like the wind, rain, or neighborhood cat, something to be managed or controlled rather than sincerely reasoned with, grateful or resentful toward, etc.  It would be like trying to think of them on the model of an AI device (Alexa or Siri) or an ATM machine rather than a real human being.  Moreover, to be consistent you’d need to take this attitude toward yourself as well.  This is psychologically impossible and would lead to a mental breakdown if seriously attempted.  The participant reactive attitude simply goes too deep in human nature, no less than eating, sleeping, perceiving, and walking do.  It is, Strawson says, not something we can simply choose either to accept or reject.

To expand on Strawson’s point, we might note that the participant reactive attitude goes so deep that it is evident even in the behavior of those who deny the reality of free will, and indeed even in the very act of defending this denial and drawing implications from it.  For example, those who claim (quite mistakenly) that neuroscience or some other scientific finding shows free will to be illusory try to convince people who disagree, sometimes accuse them of intellectual dishonesty if they resist, may express contempt for them and their purported ignorance, etc.  None of this is consistent with the denial of free will.  If people don’t have free will, then while you may speak to them as if you are reasoning with them (as you might with a dog or an AI like Alexa), you cannot really do so.  Nor can you blame them for remaining unconvinced by your arguments if they are simply unfree in everything they think and do, including in their persistence in believing that they are free.

Those who claim that free will is illusory also often take this to entail that we should abandon the notion of responsibility and, consequently, the institution of punishment.  But as Strawson’s analysis implies, to give up the participant reactive attitudes would entail abandoning far more than just that.  We would also have to give up attitudes like resentment, gratitude, and forgiveness, along with praise and blame of any kind, any attempt to reason with others, and so on.  Again, to be consistent we would have to adopt the “objective attitude” wholesale, treating others and ourselves as things to be managed or controlled, just as we would the wind and rain, a dog, an ATM machine, or the like.  There is nothing special about punishment, specifically, that makes it intelligible why we would give that up but keep most of the other aspects of the participant reactive attitude.  It all goes, or none of it does.  And as Strawson says, it simply isn’t possible for it all to go.

Exactly what are the metaphysical implications of Strawson’s analysis?  It could be taken in different directions.  For example, one could (though I certainly would not) interpret it in a compatibilist way, holding that we are free as long as we do what we want to do without external compulsion of the ordinary kinds, but where this does not rule out our wants being themselves determined.  One could take it in a Kantian direction, holding that it shows that we must of necessity think of ourselves as if we are free, but where this falls short of a metaphysical demonstration that we really are free.  Or one could take it to provide the ingredients for such a demonstration, an argument to the effect that denying free will is ultimately incoherent. 

I am partial to the latter approach, though demonstrating the reality of free will is a topic for another time.  (In fact it is one of the many topics addressed in the book on the soul that I am working on and hope at last to finish by the end of the summer.)