Blog Coordinator

« Very Bad Wizards Ep. 25: Burning Armchairs with Josh Knobe | Main | Congratulations to Kadri Vihvelin »



Feed You can follow this conversation by subscribing to the comment feed for this post.


I think this post is exactly right, and your ideas are a crucial contribution to the debate. You say:

"All I want to accomplish here is to illustrate that measures that limit the influence of present luck drive up the influence of constitutive luck, and vice versa."

You can see this in the Basic Argument or any other "tracing"/"ultimacy"/"regression" argument. Essentially, the arguments begin with the present and say "luck or no luck?" If no luck, then just move backward. As you move backward, you discover that past luck fills in the gaps. Eventually, we go so far back that we start to hit "constitutive luck" or luck (good or bad) in being the kinds-of-persons we are.

To the extent that manipulation arguments pump good intuitions (as opposed to bypassing or scapegoating intuitions), these arguments are simply making vivid the reality of constitutive luck.

Of course, constitutive luck is a tricky concept. It's not exactly clear what it is, when it applies, or where it's boundaries are. It doesn't make much sense to say that I'm lucky not to be a car, or that a dog is lucky (or unlucky) to not be a human. Why not? And why does it make more sense to say that I'm lucky to be born rational enough, and sweet tempered enough, not to become a mass murderer? You've done great work to put the first stakes in the foundation on constitutive luck, but there is still a lot of work to do.

Great post Neil--really got me thinking.

I was going to use a sports analogy by cloning Usain Bolt, but then realized a better example is available. Drag racing.

We construct any number of drag racers with the exact same specs, right down to the point of invoking any more philosophical problems about the arbitrary degree of measurement, and thus produce two race cars that, in theory, can do a quarter-mile in a near-identical time. The constitutive luck of the vehicles has been deliberately evened out as equally competitive. We take two drivers of proven even-enough competency, thus trying in another way to eliminate another source of constitutive luck. They race. I doubt they will photo-finish as co-winners. Present luck--small fluctuations of response, slight differences of track texture, etc. will result in a winner, even if only detectable by the photo-finish.

Take two dragsters from very different teams constitutively-speaking. One is well-funded, lots of tech, several replacement engines of the highest quality, etc., and the other with more limited resources, not the highest horsepower, fewer replacement parts and lower tech. Now seat approximately equally skilled drivers. Thought present luck might make a difference--the better car suffers a blowout from an unseen bit of metal on the track--on average the better funded team wins (so I'd say).

The equal drivers in both scenarios serve as stand-ins for agents in real life who on the whole have equal abilities (as assessed in the context of relevant circumstances), but who in the first case are subject mostly to present luck, and who in the second case are subject mainly to constitutive luck.

While present luck can make a difference in the second scenario in a few individual cases, over the long haul constitutive luck wins (by something like Bayesian assumptions). In the first case present luck is everything, even as differentiating small agential variations of skill.

My intuition (gag) is that this favors a role of constitutive luck as more important than present luck if all other factors (assessed as agential in driver competence) are evened out.

Of course this is all artificial, and assumes conditions that do not reflect conditions that real agents (drivers) face. A thought experiment with at best limited use.

I'm too depressed about last night to give any sort of substantive comment except to say that I think sports is an underused analogy for our field. There's so much that maps on nicely (as your post illustrates). Why do we not have a problem praising people with obvious physical advantages? That's another example of constitutive luck snowballing. They get better coaches and more opportunities in high school, college, and on. Why do we feel somewhat (but not entirely) comfortable judging results (consequentialist luck)? And why does God hate Boston right now? Why do we have to lose every postseason game in every sport in the most excruciating way?


Does being a dogmatic compatibilist mean that if someone physically took your finger and forced you to pull the trigger of a gun to hit some bystander, you would still be morally responsible for shooting the bystander in a deterministic universe? Is this what all hardline compatibilists have to commit to? I can't see how anyone would argue for this sort of position. Certainly no folk intuitions would support this. (This question probably would have been better suited for the last thread.)

You have some nerve complaining, Tamler. In recent years the Almighty has not only seen fit to remove the Curse to the tune of 2 WS wins, but you guys have also been blessed with a Stanley Cup, an NBA title, and several Super Bowl wins.

Papi made himself into a great ballplayer just as much as Pedroia did. I don't have a problem praising either one, as both made the most out of their talent. Why should the fact that the former was more gifted to begin with detract from his achievements? I suppose degree of effort will also be chalked up to some sort of innate advantage, begging the following question:

Luck as compared to what, Neil? This discussion is beginning to remind me of debates over altruism: every single example of a term being rejected until the contrast implicit in its meaning is completely nullified. Even LFW and thus its exercise appears here as a matter of good fortune, since it is either bestowed upon a creature or not. But now I have lost all sense of what it means to catch a break.

//Why do we have to lose every postseason game in every sport in the most excruciating way?//

While I'm sure last night's loss was excrutiating for you, you don't have to look too far back to find another post-season loss that was more excrutiating. But it was inflicted /by/ your team, not /on/ it. You might actually have found /that/ game exhilerating -- much as I found last night's game exhilerating. Evidence for sports relativism?

Anyway, I'd be interested in seeing a defence of the claim that it's wrong to actually award the Stanley Cup to any particular team, even if we have to do so anyway (due to the terms of its donation and the decisions of its Trustees -- the laws of its universe). It's all down to present and constitutive luck, innit?

I don't have any problem praising athletes with obvious physical advantages, musicians with innate talent, or humorists who were born into a funny family and absorbed that talent themselves. Constitutive luck? Sure, but constitutive, not luck, is the key word there. Good stuff is good. I'm grateful to anyone who reliably brings it. And maybe even sometimes unreliably.

Brent, keep up. Recall that I am a skeptic about moral responsibility.

Robert, do you really think this Papi person (of whom I know nothing) worked or works harder than other players of this benighted game? There are thousands and thousands of kids out there working just as hard or harder, I guarantee it. I talked about effort in the book. What I said there transfers badly to this context, but in the work context it is a complete myth that the lawyer deserves her salary because she works harder. Lawyers work really hard: so do Bangladeshi sweat shop works (comparable hours). The difference is the lawyer gets respect from a job that is intrinsically satisfying as well as extra compensation. If that's not due to luck, then the word has no meaning.

I think the way to press the emptiness criticism is by working through the account I offer. Here are the definitions (non chancy luck = constitutive luck, roughly):

Chancy Luck:
An event or state of affairs occurring in the actual world is chancy lucky for an agent if (i) that event or state of affairs is significant for that agent; (ii) the agent lacks direct control over that event or state of affairs and (iii) that event or state of affairs fails to occur in many nearby worlds; the proportion of nearby worlds that is large enough for the event to be chancy lucky is inverse to the significance of the event to the agent.

Non-chancy Luck:
An event or state of affairs occurring in the actual world which affects an agent’s psychological traits or dispositions is non-chancy lucky for an agent if (i) that event or state of affairs is significant for that agent; (ii) the agent lacks direct control over that event or state of affairs; (iii) events or states of affairs of that kind vary across the relevant reference group, and (iv) in a large enough proportion of cases that event or state of affairs fails to occur or be instantiated in the reference group in the way in which it occurred or was instantiated in the actual case.


I thought you were skeptical about moral responsibility for the reason of luck, but not for the reason of determinism. Haven't you said that you are compatibilist in that regard? Apologies if I got that wrong. But IF you are a hardline compatibilist (I do realize you are MR skeptical for other reasons), then I think you still have to answer for the type of case I presented.

Brent, you have me right. But not even the most dogmatic compatibilist is going to worry about your unless there is some reason to think it is analogous to determinism. Compatibilism 101: deterministic causation is not analogous to coercion or constraint. That's the very central plank of the compatibilist case. So I doubt anyone is going to be moved by the case.

I am also irked when professionals boast of their work ethic as a means of justifying their high incomes. That said, I also believe that there is self-cultivation going on in each person and that in many cases it separates winners from losers. Why would that be so hard to believe in a case of sustained excellence such a Sir Alex and Man U?

Your writings abound with putative examples of both CL and NCL. What I need to drop the emptiness charge is a good, old-fashioned case of desert: someone exercising self-control, doing something meritorious all by himself, unaided by God, Man, or Nature so as to deserve a measure of praise. But if you cite an instance of such a thing, say a Campbellian 'effort of the will,' we'll have to say that not everything we are tempted to praise is a matter of CL or NCL.

Let's try to be a little more open-minded when it comes to sports talk. I am an American enamored of futbol and also enjoy watching Australian Rules Football. I don't understand cricket, but I'm sure with a tutorial from you and Brian Weatherson I could easily get into that too. Tamler and I would gladly do the same thing for you when it comes to baseball.

I bet it wouldn't work, Robert. There are people who love sport, any sport, and people who are bored by it. I don't belong to either. I have lived in this country for 30 years without ever managing to care about Aussie Rules. I live in Melbourne, the most sports mad city in the world (per capita) and aussie rules is a religion. Interestingly, it crosses all classes and gender: I think that female interest is almost as high as male. But I don't care. I enjoy tennis and cricket, but I can take them or leave them. And then there's soccer. It is just has a power for me that is unique. I don't think that there is a good argument from "soccer is special for me" to "soccer is uniquely valuable", though.

Of course there is self-cultivation. And I hope that my skepticism (what John called me Neilism) doesn't entail that some people don't do a better job of it than others. I agree with Paul: good stuff is good. But I want to distinguish creditworthy from good.

“Compatibilism 101: deterministic causation is not analogous to coercion of constraint.”

But don't some compatibilists take the "bite the bullet" strategy to handle manipulation arguments? Conceding that there is no real difference between mere determinism and blatant manipulation, they’d still rather say that the agent is at least somewhat responsible in all 4 cases than none of them at all.

I guess this is the type of compatibilist I was referring to as “dogmatic”. Perhaps I’m confusing the hard line reply compatibilists with the term “dogmatic”. But I still think that any compatibilist who takes the hardline "bite the bullet" approach to manipulation arguments would indeed have to hold the agent in my example morally responsible.

"You have some nerve complaining, Tamler. In recent years the Almighty has not only seen fit to remove the Curse to the tune of 2 WS wins, but you guys have also been blessed with a Stanley Cup, an NBA title, and several Super Bowl wins."

Robert, you're a 100% right. That was bullshit, I had no business complaining given the success we've had. I apologize. It was like Romney and the 1% thing. The only thing I can say in my defense is that we've lost huge games in horrible kicked in the groin ways (2008 Giants superbowl, 2011 Giants superbowl, 2010 Lakers, 2013 Ravens) to teams that I hate. But that's not even a good excuse because I don't hate the Blackhawks, I respect them, and they deserved the series. So I just take responsibility, I deserve blame for that comment.

I have a confession to make. Dan Dennis is a pseudonym I employ when contributing to philosophy blogs. My real name is Andy Murray and I am just taking a break from Wimbledon to contribute to this discussion as I thought my sporting insight might be relevant.

I want to tell you the story of my identical twin brother Sandy. We are physically identical and have had an identical upbringing, imbibing similar attitudes and ideas from our parents, and sharing friends, toys, clothes, everything.

One day when we were 13 and came home from tennis practice he as usual followed his desires and habits so went upstairs to play computer games and watch pornography. I was about to join him but instead asked myself ‘Why?’ Why do that, why not do something else? I started reflecting trying to find answers. Through reflection and reading I gradually deepened my understanding of ethics, gained new ideas, attitudes, criteria etc. As a result I started acting differently to Sandy. I rejected merely following base desires and habits, and I instead sought to achieve something worthwhile with my life. I acquired a work ethic, I sought to treat people with respect and set a good example, and so our ways parted. Now I am a rich and successful tennis player and he, well, sadly he ekes out a living as a cognitive psychologist.

The parting of our ways all goes back to when I asked, ‘Why?’ You may ask me why I did that. Well, I just stood back and recognised I faced a choice, and took responsibility for that choice. You may ask, why did I do that and Sandy did not. I am afraid I cannot answer that question. You may say that there must have been some difference in our upbringings prior to that point which is responsible for my making that choice: but I would say that this is bare assertion. Can you prove that to be the case?

So we are left with your bare assertion that my asking why and Sandy’s not asking why is due to differences in our upbringing; and my bare assertion that I took responsibility for my choices and sought to deepen my understanding of ethics etc, and from this the wealth, respect and glory followed. Is there any reason to prefer your assertion?

Dear Andy,

Don't you have more pressing issues to worry about right now? The Wimbledon draw is wide open this year. Since you asked, I will tell you why I don't think you deserve credit for becoming the player you are, even at the risk of depressing you and causing yet another title to escape.

I am not committed to thinking that the difference between you and Sandy is due to your upbringing. It might be, but I can't tell (it might be upbringing, it might be genes - identical twins are not entirely identical - or it might be something else. I think there is some explanation for why you made the choices you did, though. The rights thought struck you, with the right force, at the right time (why: well any number of reasons. Brain processes are stochastic; they might even be indeterministic. In addition, there are all sorts of chance stimuli in the environment which might serve to prime you ... the effect could be due to a word on an advertising hoarding. You and Sandy could have been reading the same book at the same time, but when you happened to glance up the word 'good' had a priming effect on you alone, due to a chance micro saccade). I claim that if the difference isn't due to upbringing, or genes (or inherited epigenetics) it was due to some such chance event. There is some explanation of why you chose the way you did, and that explanation can be factored out into present lucky events and constitutively lucky states and nothing else.

There is, however, a possibility I am wrong that there is an explanation of why you made the choice you did. Maybe you made the choice for no reason at all! That's possible, but - for mostly familiar reasons - I don't think that helps establish your responsibility. Either the familiar luck argument against event causal libertarianism shows that you are not responsible for your choice, or my (new) argument against agent-causal libertarianism, according to which the agent-causal power can't be exercised for reasons, shows you are not responsible for your choice. I am pretty sure, by the way, that this new argument can be extended to cover non-causal accounts.

But don't be too depressed. You might not deserve your skill, but you and I can still take pleasure in its exercise.

Doesn't the tone of Andy's story support my previous argument that MR enthusiasts are really just trying sure up praise for not turning out horrible? It sure seems like Andy is begging for us to say something like: you sure did deserve to be an all-star tennis player and make millions of dollars.

On a side note, Andy's story also seems to imply that successful people are usually the virtuous ones. If Andy did have a twin brother who wasn't a famous millionaire, I’d put my money on him being a "better person".

Hi Neil

Thanks for your concern. After winning this evening I feel I deserve to relax with a bit of light blogging.

If I thought I was not responsible for losing then I might feel under less pressure, so might have more chance of winning – though on the other hand I might just try less hard…

So it seems like you don’t know what explains my asking ‘Why?’ and Sandy’s not asking taking ‘Why?’ (I don’t blame you for this, you don’t have enough information about my life – or indeed anyone’s – to give confident explanations for why a certain person asks ‘Why?’, though you list a few possible explanations). Rather you simply *assume* there must be something that explains my asking ‘Why?’ In other words you assume that my asking ‘Why?’ must be explainable in terms of something else. So you assume that there must be something more fundamental than my asking ‘Why?’ which explains my asking ‘Why?’ (In other words, you assume that my act of asking ‘Why? can be explained in terms of something more fundamental). Furthermore, it seems like you have in mind an explanation in terms of a causal sequence of physical events, with chance events playing a role in that causal sequence of physical events.

Would be I be wrong to think all the work is done by the aforementioned assumptions? So once those assumptions are made the rest of your argument goes through. What though is your argument for accepting those assumptions?

May I not reject those assumptions? May I not think that my asking ‘Why?’, my taking responsibility for my decision-making is fundamental. Not explainable in other terms. It is a rich complex, sui generis notion which is not subject to analysis in other terms. You either grasp it or you don’t. I asked ‘Why?’ thereby taking responsibility for my decision-making but Sandy did not – there’s an end on’t.

Furthermore, the very nature of this asking ‘Why?’ this taking responsibility for my decisions (I might add, the very concept of freedom) is such that it cannot be part of a causal sequence of physical events. If I am caused by certain physical events to ask ‘Why?’ then in asking ‘Why?’ I am not taking responsibility (am not free) but am merely a cog in a mechanism. Thus in claiming I am responsible for asking ‘Why?’ I am precisely denying that my asking ‘Why?’ is merely part of a causal sequence of physical events. It just ain’t that sort of thing (indeed it ain’t a thing at all). How can you prove that it is?

You may claim that either I am caused or my action is random. But may I not reject this claim, and point out that this claim begs the question, relying as it does on those assumptions which I dispute. My whole claim is that my asking ‘Why?’, my taking responsibility for my decisions, is neither caused nor random. It is precisely something else, not capturable in those simplistic terms. It is mysterious and wonderful – but that’s fine…

Hi Brent

I am simply concerned about truth and ethics – I have all the adulation I need thanks.

I don’t think my life story does imply anything about the relative virtue of successful and unsuccessful people. What I wrote rather implies that those who ask ‘Why?’ and take responsibility for their lives are likely to be better (in the sense of more ethical) people than those who do not. Whether that translates into their being more or less likely to be rich and successful, I don’t know.

Hi D(an)dy,

You are quite right about my assumptions. I think it is a desperate move to appeal to the inexplicable, so I am not especially worried that I can't block your reply. But while I can't block it, I can defeat: that's the point of my argument against agent-causal libertarianism. To me, the agent-causal response to the luck objection sounds as desperate as yours. That reply is: the exercise of the agent-causal power is essentially an exercise of control. But my strategy in the book is to grant it, and show that it can't be exercised for reasons. I think the same move will work here. Of course, you may be unmoved by the argument (I have had three different kinds of responses to that argument from well-informed people: there's nothing new here except window dressing; that response is the best thing in the book (not meant as a putdown, I think!); that argument is an embarrassment. Perhaps you will fall into the third class of respondents).

Hi Neil

So are you saying your assumptions are inexplicable? Does not every system of thought contain something inexplicable, primitive? So in your system of thought your assumptions are inexplicable, primitive; in mine asking why, learning, taking responsibility, ‘freewill’ are inexplicable, primitive (so not explicable in terms of ‘exercise of control’).

DD: I don't think my assumptions are primitive. They are supported by their enormous explanatory success. The assumptions that events have causes and the assumption that mental processes are best understood by postulating simpler mechanisms are amply justified by their track record.

'But my strategy in the book is to grant it, and show that it can't be exercised for reasons.'

Sure you can exercise your ACP for reasons: they are the 'material cause' of your choice, what it is made out of. You find yourself in a particular state of mind, consisting of various beliefs and desires, and on the basis of it, will to do one thing rather than another. Afterward you will cite those B & D to justify your choice, they will be for yourself and others the reasons why you made it, although not in an efficient causal sense.

Robert, you are attempting to refute an argument without having seen it! Here's the quick version. Everyone agrees that our reasons play some kind of role in helping us to select options. On the agent-causal story as developed by Clarke and O'Connor, our reasons array options before us, and we then exercise our agent-causal power to select from those reasons. I argue that the force of reasons is fully accounted for in arraying those options before us, so that if the agent-causal power is itself said to be exercised for reasons, we are double counting some of our reasons. It is easiest to work through Clarke and O'Connor, who have given us worked out accounts of agent causation, but I claim that there is no good alternatives to holding that (a) our reasons constrain the options that are choiceworthy for us and (b) all our reasons are brought to bear in arraying choiceworthy options; grant (a) and (b) and my argument goes through. Fail to grant (a) and (b) and you run into equally serious problems elsewhere.

A lot depends on what you want from an account of libertarian free will, though. If all you want is sourcehood, then you can have choice for reasons - it is true that however you choose, you choose an option for which you had reasons - and you can have the satisfaction of a Kane-like UR condition. If all you want is sourcehood, then you can have those things on an event-causal theory too. There is no problem from which event-causal theories suffer that agent-causal theories solve.

What's always puzzled me is that reasons and causes in any form of LFW have no good explanatory inter-relationship to support that view. Reasons must come from somewhere. Either that source is caused or not. If caused then their source doesn't have L ultimacy, if not then they appear to be subject to luck as sourced. A process of picking among such reasons is caused or not. The same options hold. There is no place to locate ultimacy at all, except by question-begging posit at some point.

There is little in the FW/MR controversy I am committed to, but I see nothing acceptably explanatory about agent responsibility in LFW (in its own terms), though of course I do see how an attempted defense of it serves larger world-view motivation.

The comments to this entry are closed.

Books about Agency

3QD Prize 2014: Marcus Arvan