:: Article

the existentialist of hard choices

Ruth Chang interviewed by Richard Marshall.

Ruth Chang.jpg

Ruth Chang is a super-sassy sensation who confronts us with the philosophy of hard choices as she broods on the difference between incommensurate and incompatible and why the distinction counts, on how to handle hard choices, on different kinds, on why incomparability is rare but matters, on things being on a par, on causal determinism and agency, on why the hell I spend my time interviewing philosophers, on being a closet existentialist, on the philosophical relevance of dirty socks to understanding love relationships and on the recent spate of sexual harassment scandals in philosophy departments. This is a kooky deep existentialism with Ka-pow Kapazz! Woohay!

3:AM: What made you become a philosopher?

RC: I was working as a lawyer and I decided I wanted a 75% pay cut and to make my own copies.

3:AM: One of the things you’ve spent time thinking about is the difference between incommensurate and incomparable values. What’s the difference between them?

RC: The terminology here is a bit of a mess, mostly because some philosophers like to use a fancier-sounding term — ‘incommensurability’ – to mean incomparability, an idea for which we have a perfectly good ordinary term in English, namely ‘incomparability’, while others – correctly – use the fancier philosopher’s term of art to mean, well, an idea that we don’t already have an ordinary English word, for, namely incommensurability. I really wish everyone would use ‘incomparability’ when they meant incomparability and ‘incommensurability’ for some other idea, preferably, the idea of incommensurability.

Here are the two ideas of interest. One of these is that of lacking a common unit of measure, or, equivalently, not being cardinally comparable. This idea traces back to Hippasus of Metapontum, a Pythagorean, who first noticed that the length of the diagonal of a unit square (√2) could not be measured by the same unit as any of its sides (1) since, as was thought at the time, rational numbers could not be put on the same scale as irrational numbers. The ‘discovery’ that some lengths could lack a common unit of measure was a big deal because it threatened to throw into disarray the Pythagorean view that everything in the universe was measurable by ratios of integers. As the story goes, Pythagorean thugs took old Hippasus out to sea and drowned him in an attempt to keep his discovery secret. Of course, today we know about the real numbers, so the case of the length of diagonal and side of a unit square wasn’t a good example of the idea of lacking a common cardinal measure. Still, the idea of incommensurability was born. Aristotle then used the idea – and, as Alan Code confirmed to me, employed the Greek equivalent of ‘incommensurable’, asummetros – to refer to the idea that there is no common unit in terms of which the (true as opposed to expedient) value of beds and houses can be measured since a house isn’t equivalent in value to any particular number of beds. So the etymology of ‘incommensurability’ shows that we should use the term to refer to the idea of lacking a common unit of measure.

The other main idea is not being comparable. Given the ordinary language meaning of ‘incomparable’, we should use that term to mean ‘cannot be compared’, since that is in fact what the word means. There’s no need to use a fancy-sounding philosopher’s term of art to refer to the idea of not being able to be compared because we have an ordinary English word that will do the trick. So if things are incomparable, they can’t be compared; if they are incommensurable, there is no common cardinal unit that measures them and by which they can be compared.

3:AM: So if two things are incommensurable, they aren’t necessarily incomparable?

RC: Exactly. Being incommensurable is compatible with being comparable since you don’t have to have a cardinal measure of two things in order for them to be comparable – they might be ordinally comparable. There may be no unit that measures the value of writing the Great American Novel and eating a slice of pizza, but the former could be better. So incommensurability does not entail incomparability. But incomparability entails incommensurability; if two items cannot be compared, they cannot be cardinally compared.

3:AM: Okay. Now that we’ve got the ideas straight, why does any of this matter? Why are incommensurability and incomparability important?

RC: Well, my own view is that incommensurability isn’t all that important but incomparability is. To see why, we need to distinguish the relata, or the things that either incommensurable, on the one hand, or incomparable, on the other. The items that are typically said to be either incommensurable or incomparable are either abstract values, like justice and beauty and cruelty, or bearers of values such as policies, paintings, and actions. The main upshot of incommensurable values is that attempts to model values by standard utility functions, which presuppose cardinal measure of whatever the function is a function over, won’t work. And if bearers of values – most significantly alternatives for choice – are incommensurable, then attempts to model the rationality of choice by such functions – for example by cost-benefit analysis – are similarly wrong-headed because the values of incommensurable alternatives do not admit of cardinal measure. There are other implications, but that’s the main thing at stake if either values or their bearers are incommensurable. Since many people (but not economists and decision theorists!) already reject such modeling, incommensurability isn’t really all that big a deal among most philosophers these days.

As for incomparability, if values are incomparable, again, not a whole lot follows. We have to reject standard decision-theoretic models of values, which, as I’ve said, a lot of people already do, and we have to accept that abstract values are incomparable. But this latter fact doesn’t itself entail much – it doesn’t entail that there are irreducibly plural values for instance since incomparable values – say, justice and beauty – may nevertheless be constituents of a single value – say, impartial goodness of outcomes. Nor does it entail that all bearers of those incomparable values are themselves incomparable since two abstract values can be incomparable without every instantiation of them (as manifested by bearers of value) being incomparable. That is, justice and beauty might be incomparable even though an insignificant beauty can be worse than a massive justice. So the incomparability of values, like the incommensurability of values or of their bearers, isn’t such a big deal.

But if alternatives for choice are incomparable, then all hell breaks loose – at least if we accept certain natural assumptions. We naturally think that we’re justified in choosing one alternative over the others so long there is some truth about their relative value, and being incomparable denies that there is any such truth. So if alternatives for choice are incomparable, practical reason breaks down; we haven’t got any basis for being justified in choosing one alternative over the others. And if there’s a lot of incomparability among our alternatives for choice, it looks like most of the choices we make aren’t guided by practical reason.

Put dramatically, maybe the Enlightenment conception of ourselves as rational agents who go around discovering and responding to reasons is a sham – and maybe the existentialists were right after all. You think you’re using reason when you decide to go vegetarian or start an exercise regime or buy this toaster rather than that one, but all you can do is existentially ‘plump’ for this rather than that since incomparability is everywhere. What’s a stake if alternatives for choice are incomparable is, then, our very understanding of ourselves as human agents.

3:AM: But you don’t think there is a lot of incomparability – if any, right? Can you give us a sense of why, on the one hand, you think incomparability of alternatives is a big deal, and on the other, you think there isn’t a lot of it?

RC: Fifteen years ago, when I was a grad student, I pored through the literature and collected what I thought were (and still think are) the seven main arguments for the existence of incomparability among goods or alternatives for choice. After raising objections against each argument and pronouncing each pretty much dead, I expected a rainshower of corrections, criticisms, and abuse from incomparabilists but surprisingly got very little pushback. I’m still waiting for incomparabilists to come up with a good argument for incomparability.

But of course this doesn’t mean we shouldn’t believe that there’s no incomparability — I’m the first to believe that we should believe things for which there are no good arguments. Incomparability isn’t one of those things, though, because I think there’s a diagnosis for why we think we should believe it that has nothing to do with whether there is any.

The problem is that we need some way to understand ‘hard choices’, roughly, choices between alternatives where one alternative is better in some of the relevant respects, the other is better in other relevant respects, and there seems to be no overall truth about whether one is better or whether they are equally good. Stock cases include whether to become a philosopher or a lawyer and whether to tell your friend that her husband is cheating on her or to stay silent because it’s none of your business and they will work out their issues or not on their own. In many hard cases, you can be practically certain that neither alternative is better overall and nor are they equally good, so the hardness isn’t simply a matter of uncertainty about the relevant facts. And since, so I’ve argued, the hardness doesn’t seem to be in our language but rather in our values, we need to explain the hardness in terms of a normative fact about the alternatives. The fact that they cannot be compared seems pretty naturally to fit the bill. That’s why the choice between them is hard. And since hard choices seem to be pretty common, we need a fair bit of incomparability to account for them.

If we inspect hard choices carefully, however, we’ll see that there are two distinct sorts of hard case we need to separate: those that are hard because there is a complete breakdown of practical reason and so we must respond to them outside of the scope of our rational agency, and those that are hard simply as a substantive normative matter, perhaps in something like the way that some math problems are substantively – mathematically – hard. Maybe moral dilemmas, if they exist, are like the former. Maybe Sophie, in choosing between which of Jan and Eva to save from the Nazi gas chambers, was forced to do something that was not guided by reasons. When she chose to save Jan, she was not acting within the scope of her rational agency because practical reason had broken down – she had to existentially plump rather than rationally choose.

Hard choices of the second variety allow responses within the scope of rational agency — you remain a rational agent as you agonize over what to do just as you remain a mathematician as you painstakingly work out the next line of the proof. In these substantively hard cases, it makes sense to continue to deliberate, agonize, ask your mother for advice, and so on. When we choose in such cases, we are exercising our rational agency, not simply plumping like Sartrean existential agents.

When a hard choice falls outside the scope of practical reason, the right thing to say, I think, is that the alternatives are incomparable. When a hard choice is substantively hard, the right thing to say, I think, is that the alternatives are comparable, but related by some relation beyond ‘better than’, ‘worse than’, and ‘equally good’ – I dub this relation ‘on a par’.

Here’s an example. Suppose you’re choosing between spending Spring Break on a beach in Florida or camping with pals in Wyoming. You’re practically certain that, with respect to whatever matters in the choice between them – say how costly, fun, and relaxing each holiday is –, neither is better than the other. If we improve one of the vacations – say, the price of the Florida vacation is reduced by $100 – does it thereby follow that the Florida vacation is better? Probably not. If the alternatives are comparable – another argument suggests that they are – then they must be related by some fourth comparative relation beyond ‘better than’, ‘worse than’ or ‘equally good’ – they are on a par. Roughly speaking, they are in the same league or neighborhood of value, even though their values are very different. I think a lot of hard choices are like this and that the cases that might be thought to be cases of incomparability are really cases of parity.

3:AM: Right. You’re known for arguing that even if one thing isn’t better or worse than another thing and they’re not equally good, they can still be comparable – you say they are ‘on a par’. Why should we think there is parity and what is it?

RC: Well, if someone asked me to say what it is for things to be better or equally good, I’d try to describe what those relations involve by describing features of the evaluative differences they denote. If A is better than B, then the evaluative difference between them favors A. If A and B are equally good, then there is a zero evaluative difference between them. If A and B are on a par, there is a non-zero evaluative difference between them, but that difference doesn’t favor one over the other. One reason it’s hard to wrap our minds around the idea of parity – or non-zero, non-favoring evaluative differences – is that we’re so used to understanding value on the model of the reals. Once you assume that value behaves like mass or length, you’re stuck with the view that one value has got to be more, less or equal to another since mass and length can be measured by real numbers, and real numbers must stand in one of those three relations. One of the upshots of entertaining the possibility of parity is that we begin to question at a really fundamental level understanding value in the same way we understand most nonevaluative properties in the world.

As for arguments for parity, that’s too big a question to tackle here, but I can sketch a couple of big ideas that might help. We’ve already seen one argument: we need one concept – incomparability – to describe what’s going on when practical reason breaks down and all we can do is existentially plump for an alternative, and another concept – parity – to describe what’s going on when practical reason is humming along fine but there’s a problem in the way the values or reasons relate that don’t determine what we should do.

Here’s a little exercise you can perform at home to help you get on the road to parity. Suppose you’re contemplating a choice between two rather different things – maybe a career in law and one in sky diving — about which you’re pretty sure that neither is better than the other. Now ask yourself whether a small improvement in one of them – say a $100 increase in salary – would necessarily make it better than the other. If not, then they aren’t equally good. Now ask yourself whether they are comparable. One way to help yourself think about this is to consider successive detractions along a spectrum from one of the items and successive improvements along a spectrum in the other item until it’s pretty clear that the detracted one is worse than the massively improved one. Now ask yourself whether it makes sense to think that somewhere along the spectrum we shift from comparability to incomparability. As we go down one spectrum and up another, we maintain comparability, but then somewhere in the middle we have incomparability and then moving in opposite directions along the spectra we again have comparability. The differences between each successive neighbor on the spectrum are by hypothesis small differences in whatever matters to the choice between them. I think it’s pretty implausible to think that the items could be arrayed so that the items at the top of one spectrum are comparable with the items on the bottom of another, and yet the middle items of each spectrum are incomparable. There are complexities I’m ignoring over here, but the basic idea is that it’s hard to believe that all such cases are ones where we’ve got incomparability instead of comparability. And if we’ve got comparability, we’ve got a comparative relation that isn’t ‘better than’, ‘worse than’, or ‘equally good’: we’ve got parity.

3AM: Can we go back to what you said about practical reason? Is this the right taxonomy of possibilities you have in mind? If incomparable, then we can’t choose as rational agents because the choice involves a breakdown of practical reason; if equally good, then, I suppose, we flip a coin; if on a par, then…what? What are we supposed to do? Flip a coin? How does parity differ from equality?

RC: ‘On a par’ and ‘equally good’ are different relations because they have different formal properties. ‘Equally good’ is reflexive – a is as equally good as a — and transitive — if a is as equally as good as b which is as equally good as c, then a is as equally as good as c. ‘On a par’ is irreflexive — a isn’t on a par with itself – a is as equally as good as itself — and nontransitive — if a is on a par with b which is on a par with c, it doesn’t follow that a is on a par with c. But they are both ways in which items can be compared.

You’ve put your finger on the other main importance of parity – besides its inviting us to give up our deeply held, implicit conception of value as akin to mass or length with respect to measurability. The most important difference between ‘on a par’ and ‘equally good’ shows up in what we should do, practically speaking, when faced with such alternatives. If alternatives are equally good with respect to what matters in the choice between them, it’s always permissible to flip a coin between them. Not so when things are on a par.

The key thing about parity is that it opens up a new way of understanding rational agency that is a substitute for the usual Enlightenment conception according to which we are essentially creatures who discover and respond to reasons. On that view, our agency is essentially passive – our reasons are ones given to us and not made by us. Our freedom as rational agents consists in the discovery of and appropriate response to reasons given to us and not created by us. Parity allows us to see that our agency may have a role in determining what reasons we have in the first place. So we might be free in a deeper sense – we are free to create reasons for ourselves under certain conditions.

3:AM: Boo-rah! Say more! Is the thought that we aren’t subject to causal determination but can break the causal chain with an act of agency that creates reasons?

RC: I don’t think we can do anything to escape causal determinism – that is, if causal determinism is true.

The idea instead is that when our reasons are on a par, we have the normative power to create new, ‘will-based’ reasons in favor of one alternative as opposed to another. Take a toy example. You can have the banana split or the chocolate mousse for dessert. They are on a par with respect to deliciousness, which is what matters in the choice between them. You have the normative power to put your agency behind – to ‘will’ – the chocolateyness of the chocolate mousse to be a reason for you to have it, thereby, perhaps, giving yourself most all-things-considered reasons to choose the chocolate mousse. Your act of agency is what makes it the case that you now have most reason to choose the mousse. This is an active view of rational agency because instead of sitting back and discovering what reasons we have, we can create reasons – when our non-will-based reasons – what I call our ‘given’ reasons – are on a par.

It’s in this way, I suggest, that we forge our own identities as, say, chocoholics or people who love extreme sports or care about the environment or work to alleviate poverty or any number of things that help define each of us as distinctive rational agents with particular concerns and projects. This is, I think the most interesting way in which we are– as philosophers like to say – the ‘authors of our lives’.

One way to get an intuitive handle on this alternative view of agency is by considering the way you spend your Saturday afternoons. Say you spend yours interviewing philosophers. Could it be true that you have most reason to spend your Saturdays this way, rather than, say, going for walks, learning the piano, or working in a soup kitchen? Probably not. Could it be true that you have sufficient reason to interview philosophers as well as many other things, and you just arbitrarily plump for interviewing philosophers, where this plumping isn’t an exercise of rational agency but the agential equivalent of flipping a coin? Our choices of how to spend our free time don’t always feel that deeply random. What we do instead, on the view I believe parity makes possible, is put ourselves behind one activity rather than another – we identify with it, we commit to it – for the time being perhaps – we take it on as something we’ll do. When we put our agency behind something, it feels like we have most reason to do what we’re doing. And that’s because we have conferred normativity on that activity. Putting your agency behind spending your Saturdays interviewing philosophers is how you make yourself into the distinctive rational agent that you are – someone curious about things philosophical.

3:AM: Okay, so you’re saying that when our ‘given’ reasons are on a par – the reasons we discover and respond to – then I can create reasons for myself by putting my agency behind an activity, like interviewing philosophers. Presumably, my willing things to be reasons isn’t itself something that is guided by reason but a pure act of agency. Doesn’t this make you a closet existentialist?

RC: In a way, yes. I think that both Kant and the existentialists were right about some very deep matters. Kant — on some interpretations — was right in thinking that practical normativity could have its source in the will. Sartre was right in thinking that choices can be a matter of agential fiat. They both were too ambitious, however, and took their ideas to be the linchpin of human agency. They were each partly right. Some reasons – the will-based ones – have their source in the will but others – the ones that are on a par when will-based reasons can kick in – are ‘given’ to us just as the Enlightenment view says. And only some choices are a matter of agential fiat – for example, ones where we our ‘given’ reasons are on a par. Crucially, the existentialists eschewed any possibility of normativity before the act of agential fiat. When we make ourselves into chocoholics or do-gooders or philosophical explorers, we do so in an already-existing normative landscape. Or so I think. So that’s another way my view differs from existentialism.

3:AM: You also think that this agential fiat is what explains what’s distinctive about love relationships, don’t you? Can you explain what you have in mind here?

RC: Exactly what this agential fiat is is a big question but it shouldn’t be thought of as something necessarily conscious or deliberate requiring effort and a furrowed brow. I’ve suggested that it’s akin to stipulating the meaning of a word; you can just start using ‘gnarly’ a certain way. That is indeed how slang gets going. Putting your agency behind something is something you do, but you might do it unconsciously and certainly not deliberately.

Here’s a story that illustrates what I have in mind. Suppose you’re invited to have dinner at the house of a casual acquaintance, and you see that she has left various pairs of her dirty socks lying on the living room floor. Your acquaintance might be a bit of a slob. Her socks don’t strike you as providing you with a reason to pick them up. They’re her socks, after all, and it’s her home, which she can keep however she likes. Contrast this with the case in which you at home with your beloved. She has left her dirty socks strewn on the living room floor. When you see the socks, they might strike you as providing you with a reason to pick them up. You might desist (she’s got to learn to pick up after herself) but the phenomenology of the socks is different in the two cases. What explains the difference in phenomenology? I suggest that the phenomenology is explained by something you do. By putting your agency behind her and her interests, you have given yourself special reasons of a committed relationship that you otherwise wouldn’t have. Putting your agency behind things makes you see the world differently, including people. Your beloved, thought to be a cranky ass by everyone else, is your lovable honeycake.

3:AM: Now turning to something completely different. What is your take on the spate of sexual harassment scandals that have rocked the profession recently? What, in your opinion, can be done to help make the profession less susceptible to bad behavior by male senior philosophers pursing relationships with female philosophy students?

RC: Well, I can’t believe that philosophers are worse than other academics, or people in general, although the distinctive, intense one-on-one discussions that mark training in philosophy probably makes our profession especially vulnerable to sexual harassment, inappropriate behavior, misread signals and the like.

The two most prevalent ‘unsympathetic’ reactions to all the press about sexual harassment or sexually inappropriate behavior I’ve had – all from senior male philosophers, some of some fame – are both of apiece with what we do as philosophers and therefore not altogether surprising. But I find them pretty dispiriting.

The first is that we all have to remain neutral, that we can’t express even conditional moral disapprobation or sympathy for a party until we ourselves have the proof in hand and we can make our own judgment about the matter. Allied with this reaction is the intellectual reflex to think of all the counterarguments to any allegation or counter-interpretations to data with which we are presented. We are trained to be this way – to see the world in terms of arguments for and against a proposition, and to withhold judgment until all the arguments and data are in. This reaction is usually cloaked under cries of ‘due process!’.

Now I’m a lawyer, and I love due process probably more than most philosophers, but I really think that these attitudes and reactions are misplaced. The real world is one in which none of us is ever going to get all the evidence and data needed to make the kind of well-informed, dispassionate judgment about a case, and — crucially – people who tend to be on the receiving end of harm require the profession’s support rather than silence. So I think each of us has to make our own judgment, on the basis of whatever data we can be reasonably expected to get, including our understanding of how things typically roll in the world, and take a moral stand on cases of alleged sexual misconduct in the profession. It’s not that we have to blog about it or call up the victim, whomever we might believe him/her to be, but even casual remarks to colleagues in a department go a long way toward establishing a departmental culture or professional community where, eventually, people in that community have the sense: ‘We are a place that cares about the harm sexual harassment does to junior people in the profession and will take that junior person seriously’ or, to take just one possible alternative, ‘We are a place that cares more about the possible injustice done to an alleged perpetrator of sexual harassment and will stand behind such a person until he is proven guilty’. Context really matters here. Given that sexual harassment is a very real and serious problem in the world, I would much rather be in the former culture than the latter. Of course, we can be wrong about any particular judgments we make. But that wouldn’t be the end of the world since most of us aren’t saddled with decision-making authority over the relevant parties. It’s time to stop pretending that we are university disciplinary committees and quit creating a passive ‘due process’ professional culture. Failure to build a welcoming, safe, and caring culture for the profession – one that reflects the realities of how the world usually rolls and thus errs on the side of supporting the alleged victim of sexual harassment – is, in my view, crucial to the health of the profession.

The second reaction I’ve had from senior male philosophers is that women undergraduates have all the power: with one little complaint they can ruin the career and reputation of a senior male philosopher who is guilty of nothing more than expressing romantic interest in her, and so the wisest thing a male philosopher can do is to steer clear of mentoring or working with any female student. The idea here is that if a senior male philosopher hasn’t got the judgment to know when his behavior is inappropriate, unwelcome, and the like, he’ll just punish all female students by taking his ball home and refusing to play with the girls. Louise Antony, in a NYT Opinionator piece, does a good job of skewering this view. I would add just one small point. At institutions where dating students is not prohibited, any senior person – graduate student or faculty member – obviously has the burden to make sure he’s not unwittingly exploiting the power imbalance between himself and a female student when he begins to pursue a romantic relationship with her. One way to check whether he’s doing that is for him to ask of every communication he has with the student, ‘Is this something I wouldn’t mind having videotaped/copied and shown to the university’s committee on sexual harassment?’ That might help pierce some self-deception.

I suppose that a silver lining in all the bad press philosophy has been receiving is that, like STEM researchers, philosophers might now be motivated to use the tools of their trade to figure out ways to make sexual harassment just unequivocally not okay within the profession. After the Larry Summers blowup, people in the STEM fields rolled up their sleeves and starting tackling the possibly related problem of the underrepresentation of women in those fields in the way they knew how – by collecting data, doing studies, and offering hypotheses based on rigorous analysis of that data. We philosophers are handicapped by not having the skills or resources — they have NSF funding — to collect data or to run proper studies about sexual harassment ourselves, but we can always try to do the same thing — to effect a shift in the culture of the profession — from our armchairs. I think that is already happening. But some people don’t like it and are being dragged along, kicking and screaming. So it’s a slow and painful process.

3:AM: Just so we don’t end on the note of pain, for the readers here at 3:AM, are there five books that you could recommend that would take us further into your philosophical world?

RC: Aristotle, Nichomachean Ethics
Sartre, Existentialism is a Humanism
Kant, 1st two Critiques
Parfit, On What Matters
Korsgaard, The Sources of Normativity
Scheffler, Boundaries and Allegiances

I guess that’s seven, but who’s counting.

Richard Marshall is still biding his time.

First published in 3:AM Magazine: Friday, April 4th, 2014.