If you are a machine, do you have free will?

If the laws of nature say that everything is determined, does that mean there is no free will? It might seem to you that the answer is clearly no - if everything, including your brain, follows some set of unalterable rules, then you are not free, not in control of your actions. If this seems reasonable to you, you are most certainly not alone. There's a vigorous debate in philosophy about whether determinism is compatible with free will, with some heavyweights on both sides. 

Sam Harris, for example, is strongly of the opinion that we have no free will. He has argued for this position in numerous settings, most recently on an episode of his podcast Making Sense entitled Final Thoughts on Free Will. A high profile proponent of the opposite view is the world renowned philosopher Dan Dennett. Here's a conversation between these two thinkers on this very topic:

I want to give you one way to look at this question that I haven't exactly seen anywhere else. First, we won't be discussing here whether determinism is actually true. All we are concerned with today is the question: does determinism imply no free will? So, let's assume determinism is true. Suppose we know that your brain is basically just a complicated computer, a machine following some pre-defined set of rules, and all your thoughts, feelings, and actions are a result of this mechanistic process. Then you have no control over anything you do, right? Free will is just an illusion, isn't it?

Not so fast! Consider another situation involving computers - two computer programs playing chess. There's program W, playing white, and program B, playing black. How do they work? In simple terms, when it's W's turn, it looks at all the moves white can make, analyzes possible black's responses, white's counter-responses, etc., and evaluates the resulting positions. It then selects and makes the move that leads to more favorable positions. Of course, this is a purely mechanistic process. 

Notice, however: clearly there is a big difference between how W relates to white vs black pieces. It selects where the white pieces go, but it can't do that for the black pieces! 

But that's exactly what the word "control" is for. The compact way to describe the difference between white and black pieces in relation to W is: W controls the white but not the black pieces. This difference exists even though W is purely mechanistic.

Similarly, even if you are a complicated deterministic machine, there's still a fundamental difference between, say, your arms and everybody else's arms. Namely, you select where your arms go in a way that you can't do for other people's arms. You control your arms, but you don't control mine. 

You know this difference, you have plenty of evidence for this based on your vast experience with arms, yours and others'. So when you have a sense that you can control them, it's not an illusion as some thinkers claim.

So do we call this free will? Free will is typically described as "the ability to do otherwise". But go back to our chess-playing computers for a second and you will notice a fundamental ambiguity in the word "ability":

  1. W controls the white pieces. Another way of saying that is: it has the ability to move the white pieces but not the black ones. It is given the ability to make any (legal) move for White, and it then selects which one to actually make based on its evaluation of the situation. Let's call this type of ability "ability-1". I know, not a very creative name.
  2. W is deterministic. Someone who knows the exact situation can predict with certainty which move W will make. There is no possibility of a different move. So in that sense W has no ability to make any legal move for White. Let's call this sense of ability "ability-2".

This ambiguity is then inherited by the concept of free will, so that we have free will-1 and free will-2. So in this framework, the answer to the question "if determinism is true, do you have free will?" seems to be: yes if by "ability to do otherwise" you mean ability-1, and no if you mean ability-2. 

But wait, you might say, are you seriously suggesting that a chess program, such as W, can be said to have free will? No, because despite people's disagreements about the term there's one point of general agreement: free will only applies to a creature with "will", i.e. mental states such as intentions or desires. 

Now that we've identified the definitional ambiguity, is one definition somehow wrong and the other one right? This big but fascinating can of worms will have to wait for another article. As a quick preview, I think ability-1 corresponds more closely to what we actually sense when we say "I have a strong sense of having free will." Ability-2 is quite a weird beast, it's not immediately obvious what exactly it even means. Meanwhile, what do you think? Let me know in the comments.

27 Comments - Go to bottom

  1. Hey Dmitriy,

    I got your email and sent a rather long-winded and off-topic reply back. It's good to be back :)

    Concerning free will, I'm going to resist the temptation to write that I had no control in expressing this viewpoint, and instead point out that it doesn't seem like 1 is adequate for what most people want to call free will. If 1 is a sufficient condition for control/free will, then wouldn't I always have free will no matter what (so long as my cognitive and physical faculties were intact)?

    For instance, if I'm robbed at gun point and asked to give up all my money, according to 1 I have free will to decide to give up my money or not. I possess the ability to evaluate all my options, decide on my course of action, and then pick my course of action. I could say no and choose to pick a fight, or vice versa etc...

    But most laypeople, and most philosophers, would say that I lack the free will to decide in such situations. Not to mention that most people (presumably) and most philosophers who believe in free will, would want to say that robots lack free will (this being a property peculiar to humans and maybe some near-human animals and sophisticated AI), but they do under 1.

    It seems like 1 is a necessary condition for free will, but hardly sufficient. Ignoring the problem of compatibilism of course.

    ReplyDelete
  2. Hey Alex,

    your example with being robbed surprised me because I really don't think most philosophers who believe in free will would say you don't have free will to decide in such situations. I think that only a minority would say that, based on what I have heard and read. What makes you think otherwise?

    I think the situation you described would be characterized by most not by saying you lack the freedom to choose, but by saying that you feel one choice to be definitely better for you than the other. For example, I have the freedom to choose to bang my head against the wall many times, but I know I won't do that because I very much prefer to not be in needless pain. I don't see why the robbery case would be different.

    I agree that ability-1 is insufficient for free will, I think an essential requirement for free will is having mental states, consciousness. I didn't include that in the article to keep it as short and sweet as possible. The main goal of the article was to highlight the fundamental difference between things that you can and can't control, such as your arms vs others' arms. This was meant to counter for example Sam Harris's idea that you don't control anything.

    Once it's granted that you can control certain things, the next point was to establish the ambiguity in the concept of ability and to distinguish two senses of it. Finally I claim that this ambiguity is then inherited by the concept of free will.

    But perhaps I should include a little passage saying that of course I don't think that the chess program W has free will in either sense 1 or 2, despite having ability-1. My view is roughly that:

    free will-1 = consciousness + ability-1
    free will-2 = consciousness + ability-2 = libertarian free will

    I also think it's a common but unjustified claim that most people, when they sense they have free will, sense something like free will-2. I think for the most part they sense, correctly, that they have free will-1. One reason for thinking that is that ability-2 seems like a very unclear concept at best and incoherent at worst. Given that, it seems like it would be rather hard for somebody to have a sense that they have it.

    ReplyDelete
    Replies
    1. "your example with being robbed surprised me because I really don't think most philosophers who believe in free will would say you don't have free will to decide in such situations. I think that only a minority would say that, based on what I have heard and read. What makes you think otherwise?"

      I of course can't speak for what you read, but from what I understand most philosophers would agree that free will is associated with the ability to do otherwise (although there are alternative accounts, like the sourcehood account: https://plato.stanford.edu/entries/freewill/#FreeDoOtheVsSourAcco), but would disagree that I would possess this ability while I'm being robbed. The reason for this has to do with the fact that free will is commonly associated with moral responsibility (Here is a nice resource on this relationship: https://plato.stanford.edu/entries/alternative-possibilities/).

      Basically, most philosophers think moral responsibility lies with one's possessing free will (indeed this is the very underpinning of our justice system). Because most people/philosophers think I'm not morally responsible for an immoral action (e.g. stealing) if I'm forced to do so at gunpoint, it must follow that I lacked the free will to do so. It's not an exaggeration to say that the association of moral responsibility with free will is primarily the reason we have the problem of free will to begin with. If this was not the case, then the problem would be rather easily solved (no offense).

      The real issue is that determinism is a blunt tool to solve the dilemma of moral responsibility. We want it to be the case that I'm free to act in certain situations but not others (robbery case). But determinism seems to imply that either all my actions are unfree (libertarianism), or that they are all free (compatibilism).

      That's not to say that there aren't philosophers who argue for dissociating the concept of moral responsibility from free will, but this accounts for a very small minority. In general, it's more commonly argued that the concept of free will is just logically incoherent (i.e. no such concept can satisfy all of the above conditions). The truth is that while the free will issue is controversial, many philosophers (I imagine) would find it even more problematic if we stripped free will from the moral responsibility condition. In essence, we (meaning philosophers of morality) have punted what might have been an ethical controversy into the free will domain instead.

      Delete
    2. "I also think it's a common but unjustified claim that most people, when they sense they have free will, sense something like free will-2. I think for the most part they sense, correctly, that they have free will-1. One reason for thinking that is that ability-2 seems like a very unclear concept at best and incoherent at worst. Given that, it seems like it would be rather hard for somebody to have a sense that they have it."

      Without making a comment on the merits of either ability 1 or 2, I think I disagree with the latter piece of reasoning. I don't think it's true that we can't argue that people can't believe they hold concept x, because concept x upon basic analysis seems to be logically incoherent. That seems to be an extreme version of the principle of charity wherein we don't attribute any erroneous beliefs to laypeople. I would go so far as to say that most philosophical controversies arise from a common sense belief which seems obvious at first glance, later turning out after some rudimentary conceptual analysis to be quite problematic.

      Let's take the classic problem of epistemology. Most people are convinced they possess some knowledge (i.e. I know the earth is not flat), and that knowledge requires justification, and that all rational claims which underpin our knowledge require justification of some sort. Most reasonable people, I think, are convinced that this is true, but upon further reflection we can easily see that we are going to end up with an infinite justificatory regress problem. Since we are finite beings with limited computational capacity it follows that we can't solve the regress problem, and so the very idea that we hold knowledge at all seems potentially contradictory.

      Similar problems exist in metaphysics where we encounter people's basic beliefs that objects exist (a belief even more properly basic than free will), however the concept of 'object' that most laypersons hold to seems rather logically implausible after further analysis. None of this is to say that philosophers haven't been working to reconcile the categories of knowledge or objects, or that it isn't possible to do so. I'm simply trying to point out that people hold a bunch of concepts which seem intuitively plausible to them, but that philosophers have discovered aren't (at face value) logically consistent.

      Delete
  3. Hey Alex,

    Thanks for providing some good resources on free will and moral responsibility. You are arguing that the standard conception of free will held by philosophers (and regular people) involves a stronger kind of ability than ability-1 (so more than the ability to control your arms etc.). I am arguing that the standard conception of free will inherits the ambiguity of the word "ability", with ability-1 being a legitimate variant.

    You use an example of being coerced at gunpoint. But this quote from the Principle of Alternative Possibilities article you linked to talks about such a case and I think supports my position:

    "Keyne is confronted by a “your money or your life” mugger, and hands over his wallet. According to PAP, whether Keyne is responsible for doing so turns, in part, on whether he could have refused. So could he have? Considering the example in one sort of context, we will say Yes: his arm was under his control and he was fully aware of what noncompliance involved and how to achieve it. But in another sort of context we might say No: Keyne was rational and valued his life far more than the money in his wallet, as would any reasonable person such circumstances; fix this cognitive background, and the mugger’s demand left him no other option. So in handing over his wallet, has Keyne met PAP’s requirement on responsibility or not? There seems to be no clear answer to this question."

    This seems to parallel my account of the ambiguity of "ability".

    ReplyDelete
    Replies
    1. I can respond to some other points you made but wanted to see first what your take on the quote is.

      Briefly, I fully accept the connection between free will and moral responsibility.

      Delete
    2. Hey Dmitriy,

      "But perhaps I should include a little passage saying that of course I don't think that the chess program W has free will in either sense 1 or 2, despite having ability-1."

      This might be helpful, as I personally made the mistake of thinking otherwise.

      "Thanks for providing some good resources on free will and moral responsibility. You are arguing that the standard conception of free will held by philosophers (and regular people) involves a stronger kind of ability than ability-1 (so more than the ability to control your arms etc.). I am arguing that the standard conception of free will inherits the ambiguity of the word "ability", with ability-1 being a legitimate variant."

      My pleasure. I must say that I find your stance rather interesting. Do you mean to say that you can't think of a single example wherein one can possess ability 1 to commit an action, and yet not be held morally responsible for that action? That surely seems extreme. I don't know about you, but if I entrusted you with a precious item of mine and you were robbed at gunpoint and forced to give it up for the price of your life; I wouldn't hold you liable for giving away my item. Nor would I charge you/demand some recompense. Even though the action of giving away my valuable belonging, in a normal context, seems wrong.

      But the above is the logical outcome of accepting the consequence of ability 1 + consciousness= free will (which entails your having free will in the robbery case) and "I fully accept the connection between free will and moral responsibility."

      It's possible to argue of course that one can be morally responsible for an action but not morally blameworthy. So maybe you're morally responsible for giving up my item, but I shouldn't blame you for doing so. But this seems to heavily dilute the meaning of the phrase "moral responsibility" outside the bounds of semantic convention. In any case, even if you really think you're morally responsible/blameworthy (Yikes! Remind me not to carry any valuables of yours 😊) for the above action, surely it's reasonable to believe that I should be able to come up with at least one example which defeats the free will/ability 1 connection?

      As for the example given of Keyne; I agree that it is definitely an ambiguous situation. But I disagree that most philosophers would interpret the ambiguity in the way you suggest (as implying that ability 1 combined with moral responsibility is a legitimate alternative). I think the ambiguity is meant to show that there is a reasonable argument to be made between dissociating free will and moral responsibility in the first place (in other words, this would count as an argument against PAP). That's not to say that it can't logically follow that this ambiguity can work in favor of ability 1 being a legitimate alternative, but that leads to the undesirable consequence I described above.

      "I can respond to some other points you made but wanted to see first what your take on the quote is"

      About your other argument; I should mention that I wrote:
      "I don't think it's true that we can't argue that people can't believe they hold concept x, because concept x upon basic analysis seems to be logically incoherent"

      Whereas you used the word 'sense' as opposed to belief. I didn't phrase my objection too well and I meant to actually capture this use of 'sense' in my counterexamples, which is why I used the dual examples of knowledge and objects. That's because we have a direct sense that we possess knowledge, and a direct sense of what an object is (i.e. I don't need to think about the fact that objects exist, I can directly perceive their existence). Sorry if I implied that you were arguing that we can't assume that laypersons don't have erroneous beliefs.

      I hope that makes it more clear.

      Delete
    3. We should assume that laypersons don't have erroneous beliefs*

      Delete
    4. I should also mention that I’m not an authority on most philosophers. I’m just speaking from my anecdotal experience + my understanding of the texts.

      Delete
  4. Hey Alex,

    >But the above [robbery victim is morally responsible] is the logical outcome of accepting the consequence of ability 1 + consciousness= free will (which entails your having free will in the robbery case) and "I fully accept the connection between free will and moral responsibility."

    By "connection" I don't mean something like equivalence. These concepts are deeply connected, but the logical outcome you were speaking about needs more than that. It would need something like: "I accept that if someone exercised free will they are morally responsible for what they did."

    But I don't accept that, I think there are many counterexamples. What if someone freely chose to do something but was unaware that this was wrong, maybe due to diminished capacity of some kind? If we can't possibly expect them to understand the wrongness of their (freely chosen) actions then holding them morally responsible seems dubious.

    I don't think it's a radical view that more than free will is needed for moral responsibility. But with regard to your robbery example, we can even grant that whatever that "more" is, it's there. For example, when I surrender your item to the robber, I fully understand how bad it is for you to lose that item. Granting that, yes, I think as long as I made a choice and exercised ability-1 to move my arm and give your item to the robber, it would be reasonable to hold me morally responsible - but crucially that doesn't mean I should be blamed!

    You already foresaw this possibility:

    >It's possible to argue of course that one can be morally responsible for an action but not morally blameworthy. So maybe you're morally responsible for giving up my item, but I shouldn't blame you for doing so.

    But then you say:

    >But this seems to heavily dilute the meaning of the phrase "moral responsibility" outside the bounds of semantic convention.

    I don't think so at all. Here's a quote from that same article on PAP:

    To be morally responsible is to be the proper object of the “reactive attitudes,” such as respect, praise, forgiveness, blame, indignation, and the like.

    So I think it would be reasonable to say that my normally bad act of giving your item away is forgivable because I had morally sufficient reasons to do it. In other words I did make a choice, and it was the right choice morally speaking.

    ReplyDelete
    Replies
    1. To summarize, I don't present any theory of moral responsibility in the article. But I think something like this is a reasonable view:

      free will-1 = ability-1 + consciousness
      moral responsibility = free will-1 + other bits like understanding wrongness

      Do you think this view is out of line with modern philosophy and/or has deep problems?

      Delete
  5. In response to your most recent comments:

    "By "connection" I don't mean something like equivalence."

    I see. I mistakenly interpreted your statement to meant something like "I fully accept the connection (that was being discussed thus far) between free will and moral responsibility".

    "But I don't accept that, I think there are many counterexamples. What if someone freely chose to do something but was unaware that this was wrong, maybe due to diminished capacity of some kind? If we can't possibly expect them to understand the wrongness of their (freely chosen) actions then holding them morally responsible seems dubious.

    I don't think it's a radical view that more than free will is needed for moral responsibility."

    Right. But I think most in the field would try to argue that such examples constitute instances of diminished free will. So that diminished intellectual capacity contributes to one having diminished/non-existent free will in the first place, as opposed to such examples demonstrating that one has free will but still requires an extraneous condition to establish moral responsibility (that's not to say that there aren't any extra conditions however).

    The reason for this can be attributed to ordinary language assessment. When we ask people why they shouldn't be held morally responsible for their unfortunate actions in the robbed-at-gunpoint case, they will say things like "I didn't have a choice; I had to give up item x".

    How should we parse such language? According to your interpretation, we have to interpret such talk as being figurative. Ordinary persons don't really mean they lack free will (or if they do, then they are wrong) but something else. There's nothing really wrong with this interpretation, but philosophers (especially in modern times) are shy to engage in such revision of ordinary language unless necessary/as a last resort. But it all depends on what school of thought you belong to.

    I would say that attempting to reconcile the concept of free will with statements of ordinary language is the reason that it is often thought that such instances constitute examples where free will is absent. Hence this "something more" (at least in such cases) is attributed to the concept of free will, which is what makes the entire notion so troublesome.

    ReplyDelete
    Replies
    1. About dissociating moral responsibility from blameworthiness:

      I wrote "It's possible to argue of course that one can be morally responsible for an action but not morally blameworthy." when I meant to say "for [such] an action but not..."

      In other words, it logically follows that if you are morally responsible for a bad/wrong action, then you should be held blameworthy. This is the standard view of ethics that most philosophers in that field subscribe to, as far as I am aware. The quote you mentioned was speaking of the relationship between moral responsibility (for any action) and the assignment of 'reactive attitudes' in a much broader sense (not just about bad/wrong actions).

      Thus the question at hand is whether your action of giving up the item was morally wrong. This depends on how we choose to interpret the word 'action'. One interpretation, and this is the conventional interpretation, is that actions are morally wrong/good in and of themselves. In this sense, stealing is always wrong, but there are some instances when stealing can be condoned (for instance, if I'm stealing from Hitler). Most philosophers of morality (again from my anecdotal experience) would say that in such a case you shouldn't be held morally responsible (and hence aren't blameworthy) but that nevertheless you committed a wrong action.

      A second interpretation, which I suspect you may subscribe to, is that any attribution of moral properties like goodness/badness to actions is contextual. One can't decide whether an action like stealing is good or bad, unless one has looked at the full context (in this context stealing from Hitler is good, or at least not bad). Without getting into the merits/demerits of either, I want to briefly say that the second interpretation creates a problem of how exactly we are supposed to contextually interpret the moral consequences of an action.

      Presumably, such interpretation works by a mechanism of summing up the moral consequences of moral particulars, and then deciding the overall moral status of the action. In other words (stealing=somewhat bad) + (hurting Hitler's finances= really good) = (stealing from Hitler = pretty good). So the wrongness of stealing (it encourages lawlessness in the society etc..) is more than outweighed by the good. But notice that we still have to attribute some absolute moral status to a moral particular. Even if you didn't agree with the above, it still seems like you would have to agree that there exist moral particulars of some kind which carry absolute moral status, otherwise such contextual analysis just doesn't seem possible.

      If this is so, then any disagreement between adherents of the first and second interpretations is really just reducible to whether one thinks the actions being discussed (e.g. giving up a valuable item) constitute moral particulars or not. I think most philosophers would want to say that even if the sum composite action of stealing from Hitler was okay/good, you still had to commit a wrong action at some point (stealing) in undertaking the composite action. Thus, there will continue to exist a question of whether we should always be held morally responsible for committing wrong actions, and most would say no.


      To summarize:

      If you believe you can be held morally responsible for action x, but don't think you are morally blameworthy for undertaking x, then either x and all component parts of x are not bad actions (but this runs counter to the above analysis), or you just don't believe moral responsibility (for committing bad actions) is tied to being blamed (for committing bad actions). But this definitely runs counter to the contemporary position, and the quote you mentioned.

      So, the way most philosophers get around this is by denying any fixed connection between moral responsibility and committing a wrong action. Unfortunately, this has implications for the free will debate if you subscribe to the PAP-moral responsibility relationship.

      I hope all of this made sense.

      Delete
  6. > But I think most in the field would try to argue that such examples constitute instances of diminished free will.

    I would be surprised by that. Suppose God made me forget that people can feel pain. Suppose then I squeezed my little nephew's hand really tight to show my affection, not realizing that this could hurt him. I think the most natural account is that I had made a free choice in squeezing his hand but am not blameworthy because the knowledge of human pain was removed from me. I would guess that most in the field wouldn't argue that I lacked free will there. But if they did I would argue with them:)

    I think you can be "ignorantly free", i.e. have normal free will but lack some specific knowledge that some particular action will cause undue harm, and therefore not be blameworthy. This seems a very natural position to me.

    If that's right then we have a case that something else is needed besides free will for moral responsibility, so that would be in support of

    moral responsibility = free will-1 + other bits like understanding wrongness.
    ----------------------------------------------------------------------------------------------

    >When we ask people why they shouldn't be held morally responsible [specifically blameworthy] for their unfortunate actions in the robbed-at-gunpoint case, they will say things like "I didn't have a choice; I had to give up item x".

    >How should we parse such language? According to your interpretation, we have to interpret such talk as being figurative. Ordinary persons don't really mean they lack free will (or if they do, then they are wrong) but something else.

    I don't think when an ordinary person says "I had to..." the literal meaning is I lacked free will, so that part isn't figurative at all. The first part, "I didn't have a choice" is slightly figurative I think, that's a more or less idiomatic expression that just means the choice was clear, A was a much better option than B.

    >If you believe you can be held morally responsible for action x, but don't think you are morally blameworthy for undertaking x, then either x and all component parts of x are not bad actions (but this runs counter to the above analysis), or you just don't believe moral responsibility (for committing bad actions) is tied to being blamed (for committing bad actions). But this definitely runs counter to the contemporary position, and the quote you mentioned.

    I was trying to think of the easiest way to address this without getting too much into the weeds to the point that nobody except us can follow what's going on. This might work:

    A dentist causes a child pain because he performs an important dental procedure. He has free will but is blameless. And that's despite the fact that one of the components is bad, namely the pain. He is blameless because he had morally sufficient reasons to cause this pain in order to save the child's teeth. Do you think this account runs counter to the contemporary position?

    If yes, in what way?

    If not, then my account of the robbery case is pretty much the same - I had morally sufficient reasons to give your item away.

    ReplyDelete
    Replies
    1. I think you can be "ignorantly free", i.e. have normal free will but lack some specific knowledge that some particular action will cause undue harm, and therefore not be blameworthy. This seems a very natural position to me.

      If that's right then we have a case that something else is needed besides free will for moral responsibility, so that would be in support of"

      Right. Like I mentioned, I definitely agree that there are additional components to moral responsibility than free will, and that's the generally understood view anyway. The point is that there exist many cases, like the robbed at gunpoint case, or cases involving people with impaired mental/intellectual faculties, where it is thought that free will is impaired/not present.

      This goes back to ordinary language, usually people will say that "x should not be allowed to make important choice y" because x has impaired mental faculties and is incapable of deciding for themselves. The latter being the key part. Again, such ordinary language is definitely open to interpretation. But it is the general position that we should construe such talk to be about free will; that free will isn't just ability/control to do x but something more.

      Denying this seems to lead to some extreme consequences. For one, it would mean that almost every time people say "I was free to do this" or "I couldn't do this" that they are speaking somewhat figuratively. In other words, how many times do you hear people speaking of freedom or the ability to choose in the manner which you suggest (as just meaning ability 1 + consciousness)? Not very often I will say.

      The point is that we use the phrase 'free will' in our language as being something one can possess on/off depending on circumstance. But if your interpretation is correct then I basically always possess free will; so long as I remain alive/my cognitive faculties were reasonably intact.

      "A dentist causes a child pain because he performs an important dental procedure. He has free will but is blameless. And that's despite the fact that one of the components is bad, namely the pain. He is blameless because he had morally sufficient reasons to cause this pain in order to save the child's teeth. Do you think this account runs counter to the contemporary position?"

      I think everyone can agree that the dentist is blameless. The question is why? Briefly, here are the possible reasons:
      1)The action in question (causing children pain) is not wrong
      2)The dentist is not morally responsible for the above action (and hence is not to be blamed)
      3) The dentist is morally responsible but not blameworthy

      I actually think in this case, most philosophers would say that either 1 or 2 is true. In the robbery case, it's definitely 2. In the dentist case, 1 is left open because (don't cause children pain) doesn't necessarily seem like a prima facie wrong action. There are all kinds of reasons we might want to condone pain, like the above, or for discipline reasons. Whereas stealing doesn't seem like something we want to generally condone, some amount of pain in children (within limits obviously) seems generally condonable.

      Anyways, I don't want to get too much into the weeds here; the point of the above analysis which you quoted was to show that the reason for the lack of blameworthiness in such cases is not attributed to there existing a disconnect between moral responsibility and blameworthiness.

      Delete
    2. To elaborate more about the stuff on ordinary language. The point there is that usually we deem that people are speaking figuratively about x whenever x is typically interpreted to mean y, and it seems like in this case they are not actually saying y. However, the words "freedom" and "choice" and the phrase "I could not do otherwise" aren't frequently employed in the manner you suggest (as being about ability 1).

      So how can you say that it is reasonable to believe that people are speaking figuratively about x as not really meaning y, when in the vast majority of instances that x is used, it's being proclaimed to be about y? That seems to be rather backwards from how we understand figurative cases of ordinary language (as usually just being about minority cases).

      Now you can argue for exceptions to this rule, like how in modern times the word "literally' is employed to mean anything but. But it must be understood that the usage of 'literal' changed through time. The modern definition is incorporated from historical practice, where the word certainly was never employed to mean "figurative".

      Further, the word if it used enough to mean something else, may one day be understood to actually mean "figurative". However, in our case, I don't think that people have historically used expressions like "I could not have done otherwise" to mean something analogous to ability 1 most of the time.

      Delete
  7. There are a lot of threads now, so let me try to summarize them to make sure we don't get dizzy. Let me know if I don't represent your arguments accurately. What do you think of maybe turning this into a collaborative article on the question of which definition of free will is "right"? I was going to write about that in any case.

    Main thesis you are questioning: free will-1 is within the range of legitimate meanings of the term "free will".

    Your argument against this position is

    >The point is that we use the phrase 'free will' in our language as being something one can possess on/off depending on circumstance. But if your interpretation is correct then I basically always possess free will; so long as I remain alive/my cognitive faculties were reasonably intact.

    To support that, you have two lines of argument:

    - The robbery case is one example where you think in our language we don't have free will but we do have free will-1.
    - Another example you give is when people say "I didn't have a choice, I had to".

    1. The robbery case, where I am forced to give up your item at gunpoint.

    We both grant that I'm blameless, and that I have free will-1. You are arguing that this is a counterexample to my main thesis, i.e. I have no free will, in any legitimate sense, when I give up your item.

    Your argument for that is that here having free will is in serious tension with being blameless, at least according to the contemporary majority position on moral responsibility. Specifically, you claim that for this case free will would entail moral responsibility, and moral responsibility would entail blameworthiness.

    I am questioning these entailments with the dentist counterexample. Your response to it seems to indicate that the main (or at least a common) account of it is that the dentist isn't morally responsible.

    My response then: since he presumably has free will, that seems to break the first of the two entailments. So how do you demonstrate it for the robbery case in a way that wouldn't apply to the dentist?

    2. Ordinary expressions like "I had to...", "I didn't have a choice".

    I'll try to formulate a compact form of the argument you are making:

    A. When people say things like that, they think they lacked free will, according to their understanding of "free will". Examples of such people cover all legitimate understandings of the term "free will".
    B. If they think they lacked it then they actually did lack it. But they had free will-1.
    C. Therefore whichever legitimate meaning of "free will" you pick, it's possible to lack free will while having free will-1.
    D. Conclusion: free will-1 is not among legitimate meanings of "free will".

    I think A and B questionable. I think I'll just grant B to simplify the discussion.

    ReplyDelete
    Replies
    1. Your argument for A.

      Here's a representative passage:
      "Denying this seems to lead to some extreme consequences. For one, it would mean that almost every time people say "I was free to do this" or "I couldn't do this" that they are speaking somewhat figuratively. In other words, how many times do you hear people speaking of freedom or the ability to choose in the manner which you suggest (as just meaning ability 1 + consciousness)? Not very often I will say."

      And here's my representative passage:
      "I don't think when an ordinary person says "I had to..." the literal meaning is I lacked free will, so that part isn't figurative at all. The first part, "I didn't have a choice" is slightly figurative I think, that's a more or less idiomatic expression that just means the choice was clear, A was a much better option than B."

      And one from you:
      "Usually we deem that people are speaking figuratively about x whenever x is typically interpreted to mean y, and it seems like in this case they are not actually saying y. However, the words "freedom" and "choice" and the phrase "I could not do otherwise" aren't frequently employed in the manner you suggest (as being about ability 1)."

      My response to that:
      I'm dubious that figurative uses of x must happen less often than literal uses of x. But let's leave aside the meaning of the word "figurative", I'll use a different word. I think expressions like "I had no choice" are essentially idioms, i.e. most of the time they are not used to mean what the naive, literal parsing of its syntax would suggest. That's not rare in English, there are countless such expressions ("I am starving", "I'll be there in a minute", "I am absolutely thrilled", "I would die to be able to..."). When somebody says that they'll be there in a minute, they usually mean in a few minutes or soon. When a surgeon says "I didn't have a choice, I had to amputate" she doesn't mean that there were no choices for her to select from. She means that she evaluated other choices to be clearly worse than the unfortunate choice of amputation, and then intentionally selected the best option and executed it.

      The expression "I couldn't" is used sometimes idiomatically and sometimes literally. "I couldn't reach the top shelf" is literal, it's talking about ability-1. "I couldn't come to your party because I had to perform an amputation" is idiomat in the same sense as "I had no choice". So I don't see great examples establishing A, i.e. things people say in situations where they had free will-1 to show that they didn't have free will.

      Delete
    2. About the dentist case, I would say that I personally at least, don't see inflicting pain as a wrong (within limits of course). I would add the caveat that involuntarily inflicting pain is wrong. So a a person inflicting pain on a consenting masochist who wants to experience it, or a doctor doing this to a willing patient isn't doing a wrong thing by my eyes. Also I don't see even involuntary pain as always wrong. For example, if someone lightly pinches me to get my attention, I don't think that they committed a wrong.

      Certainly we can take the dentist example to the extreme by concocting a scenario wherein a doctor has to perform procedure x to save the life of a child, but unfortunately does not have any access to anesthesia. The procedure will be incredibly painful, and the child does not want to undergo it. Nevertheless, for the good of the child, the parents and doctor decide to undertake the procedure.

      I would say that in this case, the general position here would be that the action might have been wrong, but that the doctor and parents are blameless (i.e. not morally responsible). This goes back to free will being a necessary condition for moral responsibility, but not a sufficient condition.

      That said, I wouldn't read too much into any of this. Just because most philosophers think x has to be defined a certain way, doesn't mean that we should do so. If your intuitions align differently, by all means feel free to do otherwise.

      Delete
  8. "Your argument for that is that here having free will is in serious tension with being blameless... I am questioning these entailments with the dentist counterexample... So how do you demonstrate it for the robbery case in a way that wouldn't apply to the dentist"

    Firstly, I'd like to point out that I'm not arguing for anything. I'm simply putting forward the general understanding of free will, as philosophers have conceptualized it. Like I mentioned, it's definitely true that there is more to moral responsibility/moral blameworthiness than just possessing free will. Free will is a necessary condition for moral responsibly, but not a sufficient condition. This was aptly demonstrated by your dentist case and my stealing from Hitler case.

    The reason for our lacking free will in the robbery case is supposed to be because of the argument from ordinary language (i.e. most people would say they lack free will in such a situation). Generally, when conducting conceptual analysis, philosophers start from ordinary language statements, and then try to reconcile potentially contradictory aspects of this concept. It is not important that the concept be something we actually possess, or even logically coherent; for it may be that it is actually neither. So premise B doesn't have to work.

    ReplyDelete
    Replies
    1. " 'I'm dubious that figurative uses of x must happen less often than literal uses of x. But let's leave aside the meaning of the word "figurative", I'll use a different word. I think expressions like "I had no choice" are essentially idioms"

      I meant in cases where x is a word (e.g. 'Couldn't' or 'Freedom'). "Dying" is also most often not used figuratively, at least historically speaking.

      Additionally, it should be pointed out that it is quite clear that someone in the robbery case has the physical ability to make many choices. We all agree that this is true. So of course if one asked an ordinary person whether they had the physical ability to not give away their item, it would be generally understood that the answer is yes. The real question is, is that the general understood meaning of human ability or freedom to act?

      The easiest way to decide if people really mean what they say when they speak phrases like "I'm dying to see this" is to simply ask them. It's pretty clear that most people would say that they aren't literally dying, or literally starving, or literally about to arrive in a minute etc... when pressed. However, this isn't the case when it comes to our freedom of will, and our human ability to do otherwise.

      Another way to see this is in the courtroom, or in legalese. Generally, legal speak is quite literal and doesn't leave much to the figurative imagination. It would be quite strange to learn that the law which speaks of our freedom to do x (like the freedom to bear arms), is not being literal but is actually speaking of something else, especially because in this domain the parsing of words is so important.

      And free will of course goes hand in hand with our freedoms; if you don't have the freedom to do anything, then you don't have free will (I hope this isn't a controversial statement, it's basically in the dictionary and the general understanding that free will is a synonym for autonomy etc...).

      In the courtroom this is even more obvious, it is commonly asked whether the defendant had the freedom of will to undertake action x, and it is commonly understand that they do not by reasons of insanity or coercion. Even though insane people and coerced people possess ability 1. Again (going back to legalese), freedom is generally understood as the ability to do x without some "undue burden" and not simply as the physical capacity to do x.

      If you undertook polling, most people think that they lack the freedom of speech if they can't say x in public because of governmental repression, or criticism etc... Most don't understand their freedoms to be so literal, that they would interpret freedom of speech as just being the ability to physically move their mouthparts and reproduce discernable sound (even if you pressed them to be literal).

      Similarly, in the robbery case; I'm pretty sure that most people would say that they really did, in a literal sense, lack the freedom to choose otherwise (regardless if they could physically choose). But even if I'm wrong, it's obviously not going to be anywhere close to a ~0% percentage of people who would say so (as would be the case involving the polling of people who are asked if they were being literal about their dying, assuming the respondents were not pranking us).

      So freedom, and consequently our free will, is typically associated as being much more meaningful than ability 1. Another reason for this, besides ordinary language analysis, is that we want free will and human autonomy to not be something so trivial that any animal could possess it. But ability 1 (even ability 1 + consciousness) seems like it would apply to my cat.

      Anyways, the point of all this is to get into why philosophers have defined free will in the way they have. Obviously you are free to use the expression in any way you desire, as long as we understand each other.

      Delete
  9. In essence this entire debate is over a disagreement of magnitude. It is clear, I think, that free will is associated with freedoms. The freedom to evaluate a course of action, the freedom to physically implement a course of action, the freedom of speech, the freedom to not be exposed to harm, or forced to do something undesirable etc...

    The question then becomes whether all/most of these freedoms are part of free will (as most philosophers contend) or whether only some of them are (like ability 1). I have given some arguments for why I think the general understanding is substantially more expansive than you allow, but I'm not wedded to any position that free will has to be x or can't be y. I think such controversies are mostly semantic in nature and not really important. But that's my personal opinion.

    ReplyDelete
  10. Hey Alex,

    > I'd like to point out that I'm not arguing for anything. I'm simply putting forward the general understanding of free will, as philosophers have conceptualized it.

    Aren't you arguing that you are putting it forward correctly? :) Because my position is that free will-1 is actually in line with the general understanding of free will. More specifically, in my summary above, I wrote:

    Main thesis you are questioning: free will-1 is within the range of legitimate meanings of the term "free will".

    Your argument against this position is

    >The point is that we use the phrase 'free will' in our language as being something one can possess on/off depending on circumstance. But if your interpretation is correct then I basically always possess free will; so long as I remain alive/my cognitive faculties were reasonably intact.

    Let me know if we're on the same page about this being the topic. Importantly, by "legitimate" I mean that in our language there is some fuzzy boundary between what any word, say "chair", can mean and what it can't. For example, in this sense it's illegitimate to apply the word "chair" to a cat. I can decide to call a cat as a chair if I want, but I would be redefining the word "chair". On the other hand, if I take a chair and remove three or four of its legs, I'm guessing some people would still call that a chair and some wouldn't. So that may be a grey area, but I wouldn't be guilty of redefining the word chair if I applied it to that object. In other words, I say that this object is "within the range of legitimate meanings" of the word "chair", whereas a cat isn't.

    Does that clarify what my main thesis is and do you agree that that's what we are talking about?

    ReplyDelete
    Replies
    1. Hey,

      “Aren't you arguing that you are putting it forward correctly? :) Because my position is that free will-1 is actually in line with the general understanding of free will.”

      There are, I think, two separate things here. One is what philosophers generally conceptualize as free will. The other is what free will is generally construed to be (by the population at large).

      So what I’m saying is not meant to be against this: “Because my position is that free will-1 is actually in line with the general understanding of free will.”

      As it’s possible of course that the majority of philosophers have gotten it wrong. Naturally, I could also be wrong about what the general philosophical understanding of the subject is, but I don’t think so, based on my reading of the subject and my anecdotal conversations with people (more experienced than I) in the field. However, keep in mind that this topic is not my area of expertise.

      Having said all this, I do have one last quibble to make. Concerning:
      “ free will-1 is within the range of legitimate meanings of the term "free will".”

      I don’t quite think that’s the best way to put it. Specifically, I think it pretty clear that free will 1 is one of the legitimate expressions of free will. Meaning that people’s talk about their free will can sometimes be captured by ability 1.

      That is because the “philosophical definition of free will” (if you’ll permit a bit of hyperbole) or PFW, includes ability 1, whereas the latter is exclusive of the former.

      I would say then that the real argument here concerns whether the additional burden of evidence required to establish PFW is met or not (i.e. whether the above arguments work).

      So it’s like you’re arguing that whales are mammals, and the other people are saying that they are aquatic mammals (my apologies for the analogy working against you).

      Delete
  11. As an addendum, I want to clarify this matter which I never fully addressed.

    “> But I think most in the field would try to argue that such examples (of diminished physical/intellectual capacity) constitute instances of diminished free will.>

    I would be surprised by that. Suppose God made me forget that people can feel pain“

    It is generally understood that cases wherein people lack moral responsibility due to diminished intellectual faculties (E.g. mentally handicapped person or brain tumor patient) are cases attributable to lack of free will.

    Knowledge isn’t a mental faculty and so your example about ignorance of pain wouldn’t apply. Indeed it would be rather absurd if ignorance could be interpreted as diminishing ones free will (since we all lack knowledge of something). However it’s definitely true that ignorance can eliminate moral culpability in certain circumstances.

    ReplyDelete
    Replies
    1. I think I misinterpreted what you meant by diminished intellectual capacity; which contributed to the confusion there.

      Delete
  12. Hey Alex,

    >There are, I think, two separate things here. One is what philosophers generally conceptualize as free will. The other is what free will is generally construed to be (by the population at large).

    100% agreed.

    > So what I’m saying is not meant to be against this: “Because my position is that free will-1 is actually in line with the general understanding of free will.”

    Ok, this is interesting, because I thought you were arguing using what ordinary people mean when they say "I had no choice. So I thought you were in fact saying that free will-1 is not in line with what ordinary people mean by free will.

    >As it’s possible of course that the majority of philosophers have gotten it wrong.

    In some sense I do think that is what happened. But I am saying that there is enough range of - non-fringe - views in modern philosophy that identifying free will-1 with free will is not out of line with it.

    So I am claiming that my identifying free will with free will-1 is (a) not out of line with the ordinary language understanding of free will, nor (b) with modern philosophy. It's (a) that I am expressing with “ free will-1 is within the range of legitimate meanings of the term "free will".

    Does that clarify what it is that I'm claiming? Was I incorrectly interpreting you as arguing against both (a) and (b) and you are only arguing against (b)? I haven't addressed some of your recent points yet, because I want to make sure we both understand what each of us is claiming.

    >So it’s like you’re arguing that whales are mammals, and the other people are saying that they are aquatic mammals.

    This might be a sign that there's still a misunderstanding regarding my main thesis. I am not saying that fw1 can be taken as a necessary condition of fw. I am saying it can legitimately be taken as a necessary and sufficient condition. "Legitimately", as I discussed earlier, means "without redefining the term free will, without calling a cat a dog"

    ReplyDelete

Post a Comment