Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Does moral responsibility entail libertarian free will?

  1. yes

    7 vote(s)
  2. no

    7 vote(s)
  3. don't know

    0 vote(s)
  1. May 16, 2006 #1
    Moral responsibility : When an agent performs or fails to perform a morally significant action, we sometimes think that a particular kind of response is warranted. Praise and blame are perhaps the most obvious forms this reaction might take. For example, one who encounters a car accident may be regarded as worthy of praise for having saved a child from inside the burning car, or alternatively, one may be regarded as worthy of blame for not having used one's cell phone at least to call for help. To regard such agents as worthy of one of these reactions is to ascribe moral responsibility to them on the basis of what they have done or left undone.

    Libertarian free will : The ability of an agent making a "free will" choice to have chosen otherwise (sometimes called Could Have Done Otherwise or CHDO) than what it actually did choose in a given set of circumstances, if the precise circumstances (immediately prior to the moment of choice) could be "replayed" exactly as before.

    Many defenders of the concept of libertarian free will seem to believe that such free will is a necessary pre-requisite for moral responsibility. In other words, if we deny the existence of libertarian free will, we must also deny that any agent is morally responsible for its actions.

    Many determinists, on the other hand, argue that moral responsibility can have meaning if and only if our actions are precisely determined by our detailed and prevailing motives, wishes, desires, volitions, values etc, etc, and (given that such factors will be identical if the situation is replayed under identical circumstances) the concept of CHDO is incoherent and certainly not a pre-requisite for moral responsibility.

    The question, then is :

    Is libertarian free will an essential pre-requisite for moral responsibility, and (whether you think yes or no) why?

    I'm interested to know the views of forum members on this question.


    (ps I posted this in Metaphysics and Epistemology becuse the questions it raises, regarding the requirement or otherwise for the existence of libertarian free will, are metaphysical and epistemological)
    Last edited: May 16, 2006
  2. jcsd
  3. May 19, 2006 #2
    In order for one to be held responsible for one's actions, good or bad, one must have the ability to choose his or her actions, free will.
    If I have no choice of actions or inactions, as in hard determinism then I have no responsiblity as my actions are compelled and beyound my control.
  4. May 19, 2006 #3


    User Avatar
    Staff Emeritus
    Gold Member
    Dearly Missed

    As a compatibilist I say that libertarian free will is not necessary for moral responsibility, which I only acknowledge as INNER responsibility. Other people's opinions are meaningless in this context. And all that is required for me to feel I have a moral responsibility is that it appears to me that I have freedom of choice, and that I am not aware of any really strong evidence that I don't.
  5. May 21, 2006 #4
    I think we need to distinguish between moral responsibility on the one hand and plain responsibility pure and simple on the other.

    Example : A dog can be held responsible for intentionally attacking a child, but most people would say it cannot be held morally responsible. The distinction is in the fact that though the dog controls its actions (for which we can hold it responsible in the sense that we can choose to punish it or even have it put down), it cannot be expected to have any understanding of what is morally right or wrong (hence cannot be held morally responsible).

    What do you think?

    Best Regards

  6. May 21, 2006 #5
    Interesting position. You would seem to be a determinist but at the same time you believe in the coherency of the concept of moral responsibility? (please correct me if I have misunderstood)

    Let me present a couple of arguments based on work by Norman Swartz :

    Argument #1 – There is No Moral Responsibility
    Premise 1: Every action is either caused or uncaused (i.e. a random occurrence).
    Premise 2: If an action is caused, then that action was not chosen freely and the person who performed that action is not morally responsible for what he/she has done.
    Premise 3: If an action is uncaused (i.e. is a random occurrence), then the person who performed that action is not morally responsible for what he/she has done.

    Thus: We are not morally responsible for what we do.

    Argument #2 - Causal Determinism is a Necessary Condition for Moral Responsibility
    Premise 1: Unless there are extenuating circumstances, persons are (to be) held morally responsible for their actions.
    Premise 2: Being unable reasonably to have foreseen the consequences of their actions is one such extenuating circumstance. (Recall that young children who cannot reasonably foresee the consequences of their actions are not to be held morally responsible for the consequences.)
    Premise 3: In order to be able to anticipate or foresee the likely (or even the remotely likely) consequences of one's actions, the world must not be random, i.e. the world must be fairly regular (or causally determined).

    Thus: Moral responsibility requires that there be causal determinism.

    There is a clear and unacceptable "tension" between these two arguments. In other words, one (or both) must be unsound.

    Note that Argument #1 seems to be saying that the concept of moral responsibility is incoherent, whereas an implicit premise is Argument #2 is that the concept of moral responsibility is coherent. One obvious solution to the tension is thus to accept Argument #1 at face value, and to say that the concept of moral responsibility is indeed incoherent (thus Argument #2 is unsound, it rests on an incoherent concept).

    From your post, it would seem that you would agree with the soundness of argument #2, but you disagree with premise 2 of argument #1. Can you perhaps explain what is wrong with this premise?

    Best Regards

    Last edited: May 21, 2006
  7. May 21, 2006 #6


    User Avatar
    Staff Emeritus
    Gold Member
    Dearly Missed

    If there were a deterministic moral responsibility, so that in principle an algorithm could tell me what I should do in every circumstance, then I would not be free. Oh I might be free to ignore the algorithm, but I would always know when I did so that I was "wrong", "non-optimal", or whatever sneer-word was current.

    As it is, I feel that there is no essence of me or us, and therefore no algorithm. But I choose to act in a way that I believe "in my heart" to be right. Call that a leap of faith or existentialism or whatever.
  8. May 21, 2006 #7


    User Avatar
    Science Advisor
    Homework Helper
    Gold Member

    Hi MF. I like the two arguments you gave. Note that they examine the issue from a specific perspective, one of assuming hard determinism or not. Like the blind men that examined the elephant that all came up with different descriptions because of their different perspectives, the arguments you gave also looks at the issue from a very specific perspective.

    The arguments you provided examine the situation assuming a specific model of physical reality, a model that compares deterministic mechanisms with indeterminate ones. I think that's more of a model than something which is 'real'.

    Someone in the judicial system for example, might have a different model and perspective, and might point out that without laws that govern society, if we didn't hold people responsible for their actions, the result would be disasterous. That model assumes people have something called free will. I wonder if free will is a phenomenon that may require emergent laws of physics that can't be reduced to determinate or indeterminate mechanisms.

    Regardless, the judicial system perspective thus concludes that people must be held accountable for their actions, that people are capable of 'deciding' what action they are going to take, and that people are capable of making those decisions. Note that this perspective is more of a 'big picture' perspective that doesn't try to determine what model we should use for microscopic physical phenomena.

    I'd have to agree with this judicial system perspective. I think the model used forces us to accept responsibility for our actions regardless of what physically may empower those decisions we make and actions we take. The model used seems to 'work'.

    I'm not sure what assumptions "libertarian free will" makes. How does the libertarian veiw model reality?
  9. May 21, 2006 #8
    Choices made by an algorithm can be "morally good", since acting in a way that others appreciate, is a smart way for you (and others) to survive. Making others responsible for their actions is also beneficial for your survival. Robots can act morally, and therefore libertarian free will is not necessary.
  10. May 21, 2006 #9


    User Avatar
    Staff Emeritus
    Gold Member
    Dearly Missed

    So if I have an algorithm that tells me "abortion doctor = murderer" and also "murderers should be executed" (and many many people in our society have algorithms of this kind), then it is good for me to gun down an abortion doctor. And SOME people, the people I like and pay attention too, will like me for doing that, and it will make them happy.

    Or if my algorithm says that Judaism is an evil disease of the blood which is infecting the true Noble Race with its poison, then... Well, SOME people will be happy.
  11. May 22, 2006 #10
    Do you believe that what is "in your heart" is uncaused (ie the source of some kind of libertarian free will), or do the feelings "in your heart" have causes just like everything else?

    There is no need to leap. Even if the world is 100% deterministic, it is impossible in principle for an agent within that world to completely and accurately predict the future of the same world. An ontically deterministic world does not mean the world is epistemically determinable.

    This supports the intuitive illusion that we are free agents.

    Best Regards

  12. May 22, 2006 #11
    I should point out that I borrowed these from Norman Swartz.

    Agreed. Any argument must assume certain premises, both implicit and explicit.

    I disagree. We do not need to assume "free will" (in the libertarian sense of free will) to understand why laws are necessary and efficacious from society's perspective. We need only assume that human agents can make choices, based on their rational wishes/desires/intentions, which determine their actions. Neither the choices, nor the rational wishes/desires/intentions which underlie them, need necessarily be based on any kind of libertarian free will. The entire judicial system and its objectives can be defended quite rationally from a purely deterministic perspective.

    Agreed. But none of this requires libertarian free will to be effective. It works under a purely deterministic model.

    Agreed it works. But none of this requires libertarian free will to be effective. It works under a purely deterministic model.

    A libertarian would presumably claim that to be morally responsible for our actions we must effectively be autonomous agents, the sources of our actions must be neither deterministic nor indeterministic but "something else" (this something else being mystical free will which nobody can explain). The concept seems incoherent to me.

    Best Regards

  13. May 22, 2006 #12
    Lars Laborious is arguing that an algorithm could make moral judgements, but he is not claiming that such moral judgements would always conform to everyone's idea of what is "right". The judgements made by the algorithm would depend on the underlying value systems (the raw data) that the algorithm is working on.

    Humans work in just the same way. We all have different "raw data" that makes up our personal value systems, which is why there is no agreement between humans on whether abortion is morally right or wrong. If humans cannot unambiguously determine what is morally right or wrong, why should we expect an algorithm to be able to do so?

    None of this, by the way, shows that our decision making is not deterministic.

    Best regards

  14. May 22, 2006 #13


    User Avatar
    Staff Emeritus
    Gold Member
    Dearly Missed

    I can't visualize a situation in which every individual forms a unique but deterministic moral code and an overall moral code emerges. Rawls' imaginary contract perhaps? But some individuals would not ever feel bound by such a contract. Calling them sociopaths doesn't make them go away, and in a deterministic case they are just as justified as anyone else.
  15. May 22, 2006 #14


    User Avatar
    Science Advisor
    Homework Helper
    Gold Member

    Hi MF.
    You've provided us with Norman's Argument 1. Using premise 1 and 2, he's concluded "we are not morally responsible for what we do" if the world is ontically deterministic. I believe the conclusion stems from the concept that we could not have done otherwise. Our actions throughout life are ontically deterministic, even if the world is not epistemically determinable.

    Similarly, I can understand his conclusion using premise 1 and 3 in which the world is NOT deterministic, but contains random elements. However, in this case, the conclusion reached stems from the fact there is no ability for some person to control their actions. I'm not sure this is reasonable and will attempt to address this in a moment.

    About Argument 2, Norman says:
    Ref: http://www.sfu.ca/philosophy/freewill3.htm

    He then goes on to suggest that Argument 1 is faulty saying:
    His argument for this is summed up on the next page of his notes where he talks about what he means by "causally determined".
    Ref: http://www.sfu.ca/philosophy/freewill4.htm

    I can't be sure what he's really suggesting from all this. I understand his Argument 2 better, but I don't think he provides anything that refutes his Argument 1 as he claims. Regardless, the arguments are interesting.

    I believe the following regards what you are calling "Libertarian free will". Regarding CHDO, along with using the model that the world is either ontically deterministic or just "fairly regular", I'm thinking there can be only the following conclusions:

    1) Ontically deterministic: If we make the assumption that the world is ontically deterministic, (ie: if we "model" the world as having only determinite mechanisms), then I come to the conclusion that the world history line is unchangable. Nothing we do is going to change the future. We do what we do because there are mechanisms which create what happens which we have no control over, regardless of any subjective feelings to the contrary. From the second we are born to the second we die, there can't be any variation to anything that happens regardless of what a judicial system does. If you are to die by electric chair, then assuming the world is ontically deterministic, there is no way you are going to avoid that fate.
    2) Fairly Regular: If we make the assumption that the world is only fairly regular but NOT ontically deterministic, (ie: if we "model" the world as having a small number of random mechanisms), then perhaps in this case we could make the case that there is some benefit to there being positive and negative outcomes to any given situation. In this case, there seems to be a way of changing the person's 'behavior' in the future to reduce the probability of certain unwanted actions. In this case, if the unwanted behavior resulted in a negative outcome (ie: the jury found the person guilty and sentanced them to a $5000 fine) then perhaps there is a chance that a person's negative behavior could be modified to become less likely in the future and their positive behaviors could become more likely in the future. Their behaviors would be modified in a fairly deterministic way depending on the positive or negative result of a random occurance. So in this case, the random mechanism doesn't benefit the immediate decision, but it may result in determinate mechanisms in the future changing the likelihood of that particular random occurance from happening again. This is analogous to evolution in which a random mutation may or may not benefit the individual but more importantly, the random mutation changes the likelihood of that particular occurance happening again in the future either for better or worse.

    I don't see any way of modifying a person's behavior for Case 1, where we model the world as being ontically deterministic. Do you see any different conclusions to the two cases I've presented? When you say, "The entire judicial system and its objectives can be defended quite rationally from a purely deterministic perspective." are you making the same assumptions I am for an "ontically deterministic" world? If so, if you agree there is no change one can make to the future, that the future is just as determined as the past, then I'd like to understand how you come to that conclusion.

    On a separate note, I think one might be able to argue that the world can't be properly modeled using only deterministic and/or random causal actions. I'd suggest there is another option to these two mechanisms which regards strongly emergent phenomena, thus I think one can suggest that the model is faulty to begin with and argument 1 provided by Norman is faulty because premise 1 is faulty. In this case, I think one has to clarify the question: Does moral responsibility entail libertarian free will? The clarification is that the world can't be modeled as being subject to only deterministic and/or random causal actions, that the world also depends on strongly emergent phenomena which can act to change the outcome of a random event. The world is dependant on a phenomena which can not be reduced to specific events or causal actions.
  16. May 23, 2006 #15
    What happens if different individuals have different moral values, for example on the abortion question? The only way to arrive at a single overall moral code is then by some form of compromise. But there is no guarantee that such compromise could always be reached, which ultimately would mean the overall moral code is determined either by majority vote, or by dictatorship.

    In a deterministic case yes everyone is "justified", because in the face of strict determinism (I believe) moral responsibility is an incoherent concept.

    One needs to ask fundamental questions like "what is the purpose of a (secular) moral code in the first place", and "what is the purpose of secular law and the enforcement of secular law?". The answers to these questions I think are very revealing.

    Best Regards

  17. May 23, 2006 #16



    I’ve studied Norman’s work and corresponded with him in some detail.
    Norman believes (and I have no problem with accepting this idea) that the “laws of nature” are descriptivist rather than prescriptivist. But he then goes on to argue that though determinism may be true, determinism rests on the assumption of “laws of nature”, and human free will is somehow able to intervene to “determine” (some of) the laws of nature at the moment of choice. This is why he argues that premise 2 of Argument 1 is false, ie he is arguing that (based on his notion of free will) that an action can be both caused and yet at the same time chosen freely, because in his explanation it is our free will choice which actually determines (to some extent) the laws of nature.

    Unfortunately, this to me just seems like so much handwaving. What he fails to do, to my mind, is to provide any rational or coherent explanation of “how” this mechanism (that free will is somehow a rational and yet “uncaused cause”) is supposed to work.

    Yes, but be careful. What does it mean to say something like “change the future”, since the future has not happened yet?
    In a deterministic world the future is determined by what we do, thus to simply say that “nothing we do is going to change the future” can be misleading. If I do A the future will be one way, if I do B the future will be another way. Therefore what I do determines the future.

    Agreed. But this is not the same as saying “nothing we do is going to change the future”.
    It also, by the way, is not a reason for saying that secular punishment for crimes committed is “wrong”, even if it always was determined that those crimes would be committed (we can get to that later).

    But exactly the same argument applies if the world is 100% deterministic. In a 100% deterministic case, if the unwanted behavior resulted in a negative outcome (ie: the jury found the person guilty and sentanced them to a $5000 fine) then perhaps there is a chance that a person's negative behavior could be modified (deterministically) to become less (epistemically) likely in the future and their positive behaviors could become more (epistemically) likely in the future.

    Thus, invoking indeterminism does not change the situation.

    The analogy to evolution by natural selection is misleading. Mutations could be 100% ontically deterministic, there is no a priori reason why they need be indeterministic. In which case the analogy is false.

    Sure we can. In a 100% deterministic scenario the person’s behaviour is determined by antecedent conditions. If those antecedent conditions include the fact that the person has been fined already for breaking the law, it is reasonable to assume that this could be a contributing determining factor which causes him to avoid breaking the law in future. There is no need to invoke indeterminism (which doesn’t actually change the situation in any useful way anyway), or unexplained metaphysical free will (which is an attempt at hocus pocus).

    My view on law and punishment is as follows :
    Why do we have secular law in the first place, what is the purpose of secular law, and what is the purpose of enforcing secular law?

    If we assume the premise of determinism, then any agent P which commits a morally wrong act A is causally determined to commit that act. If A occurs then, given determinism, P could not have failed to commit (was powerless to prevent itself from committing) A, hence A could not have not occurred. In other words, A was nomologically necessary. I think you will agree with this.

    Apportioning “blame” or seeking “retribution” for the occurrence of A is therefore (as an end in itself) meaningless, and simply apportioning blame or seeking retribution is (I suggest) not the ultimate purpose of either secular law or secular punishment.

    The purpose of secular law is, quite simply, to try to ensure that A does not occur in the first place. The presence of laws, and the enforcement of those laws via punishment, acts as part of the antecedent conditions which determine whether A either occurs or does not occur. If, despite the prevailing law and punishment, A still occurs then this points to a failure of the law or punishment, and not necessarily to a failure of the agent (the agent, after all, is acting deterministically – how can it be held responsible?). A perfectly functioning and "ideal" society is one in which the laws and punishments act as part of the deterministic antecedent conditions to ensure that no morally wrong acts occur (ie to prevent morally wrong acts occurring in the first place). The ultimate purpose of secular law in relation to moral responsibility is therefore preventive – it’s purpose is to prevent morally unacceptable acts from occurring in the first place

    Contrast this with divine law and divine judgement. In the case of divine law (eg the so-called day of judgement), the purpose IS to judge, it is to apportion blame and it is to seek retribution. In the case of divine judgement, divine law by definition cannot be wrong, therefore if an agent is found wanting under divine law then it must be the case that the agent is responsible. Theism therefore must assume free will, and even in a perfectly functioning society with free will, the laws and punishments may not necessarily prevent wrong acts from occurring (because the behaviour of the agents in that society is not determined). The ultimate purpose of divine law and punishment is therefore NOT simply preventative (ie its ultimate purpose, unlike secular law, is not solely to prevent moral wrongs from occurring), instead part of its ultimate purpose is indeed to apportion blame and to punish those free will agents who commit morally wrong acts.

    I think that I have shown above that it can be.

    If you think the deterministic explanation fails to accurately model the world I would be interested to know exactly how you think it fails, because I cannot see how it does. What does it fail to explain?

    I do not deny that there are “emergent phemonena”. But I do not believe anyone can come up with any rational explanation of how such phenomena can give rise to free will “ab initio”. To suppose that free will exists seems to me to indulge in hand-waving, I have not seen anyone able to make a rational attempt at explaining how free will can possibly arise.

    Sounds metaphysical to me. Are you suggesting that free will exists, but the mechanism which gives rise to free will cannot be rationally understood or exlained?

    But why posit the existence of such metaphysical free will anyway? What is it exactly about the world that remains unexplained if we assume determinism is true and free will an illusion?

    Best Regards

  18. May 27, 2006 #17


    User Avatar
    Science Advisor
    Homework Helper
    Gold Member

    Hi MF,
    If all events in time are fixed, then there is no 'decision' being made in the sense that a decision allows the selection of more than one possibility. Determinism reduces a 'decision' to nothing more than another causal action. Which raises the question of why anyone should have the subjective experience of making a decision if they are nothing but a marionette dancing to the tugs of strings that they know nothing about. The concept of making a decision is meaningless.

    Similarly, there is no A or B. The future will not be one way or another in a deterministic world. The world will only ever be one way.

    I have to disagree with that, the two cases are different. In the case of a 100% deterministic world, there is no reason to suggest there is a 'me, you, we, I' because this infers the relatively ancient belief that there is some kind of 'soul' or responsible agent that the body is a container for.

    If we claim the actions of a physical body is determinate, there is NO chance that a person's future actions can be changed, any more than the actions of the judicial system can have more than one possible future.

    My impression is you're suggesting that future events can be influenced somehow by applying a legal system with laws and rules. You wrote "occurs or does not occur", which is incorrect. In a deterministic universe, these laws and rules have no influence over future events, they are simply the points in time leading up to a determined event. We can't change any future morally wrong act in a deterministic world. Those morally wrong acts, whatever they may be, will occur just as the acts of the judicial system will occur.

    In the case of a deterministic world, there is no possibility to change future events. We are not preventing, nor reducing the chance of any event. You can claim we can't possibly know the future, but that doesn't change the fact that if we assume determinism, the future is fixed and no change can be made. Determinism (to me) clearly means there are no agents of any kind which have any influence over future events. To suggest there is a 'we, me, I, them' is to make the error of falling back on the idea there is some kind of responsible agent as suggested by religion or even before such concepts were created.


    I believe our opinions also differ considerably about what a strongly emergent phenomena is. Please correct me if I'm wrong, but I believe you are of the opinion that strongly emergent phenomena describes more than molecular interactions and conscious phenomena. As far as I know, science only recognizes some molecular interactions as being strongly emergent. I don't claim to understand solid state physics, but from what I've read, strongly emergent phenomena at the molecular level may be called for.

    I don't believe consciousness is generally considered a strongly emergent phenomena, it is only weakly emergent, meaning it is reducible to the interactions of it's constituent parts (ie: neurons). To suggest it is strongly emergent (to me) indicates one can not reduce the phenomena to the interactions of it's constituent parts, but that is ill defined. There is no theory at present to determine if and how something can be reduced to constituent parts in order to determine if it truly is strongly emergent or not.
  19. May 28, 2006 #18
    Do you mean “Why should anyone have the subjective experience of acting freely?”
    Simply because we are not aware (imho) of the precise deterministic reasons (the causes behind) our decisions. The detailed reasons are hidden from us (what Metzinger calls the “hidden darkness” of our consciousness), hence we naively believe that we are somehow “free agents”.

    But think about it carefully – if you believe our source of will is not deterministic, then what is it? Random? Stochastic? What exactly?

    Not at all. Let us say that you do either A or B. If you do A then the future happens one way, if you do B then the future happens another way. The future depends on what you do, and as a causal agent you choose to do either A or B. I am sure you will not say that the concept of “doing A or B” is meaningless (since the future depends on whether you do A or B). What is meaningless (imho) is the notion that we act with (libertarian) free will when we choose to do either A or B.

    That is correct. The future “will only ever be one way” in just the same way that the past “only ever was one way”. But we do not know which way the future will be in advance, which gives us the illusion that we act with free will. Another way of saying this is that the world is ontically deterministic but epistemically indeterminable.

    Incorrect. There is indeed an “I” and a “you” etc, in the sense of the “deterministic agent causing the action”. Where we are misled (imho) is in thinking that this agent is an “uncaused cause”, acting with (libertarian) free will. There is no “free will I” or “free will you”, but there is a deterministic agent which we call “I” and “you”.

    Think carefully about what you have just written. “Changed” from what? A person’s future actions have not happened yet, how can it make sense to talk of “changing” actions which have not happened yet? The present determines the future (imho), therefore the presence of laws and the enforcement of those laws can have a real deterministic effect on the future actions of agents.

    And you don’t? Seriously? Do you instead think that the presence of laws and the enforcement of those laws is not an influential factor in encouraging people to be law-abiding citizens? (Think carefully)

    The whole purpose of having secular law, and of enforcing that law, is precisely to encourage people to behave in accordance with certain standards. How can it possibly do this unless there is some “causal relationship” between the “law” and “people following the law”?

    Why is it incorrect? It must be the case that event A “either occurs or does not occur”, or do you think otherwise?

    Incorrect. In a deterministic universe, each event in the present or past (eg the enactment of a law, the enforcement of a law) may be an antecedent and therefore causal event which influences events in the future.

    There you go again. It makes no sense to talk of “changing the future”, because the future hasn’t happened yet. But the events of the present can and do have influence on the events of the future, The future “is what it is” only because the present “is what it is”. And secular law is part of the present.

    Yes, in a deterministic world, if event A happens then event A was always going to happen. But it is wrong to conclude from this that “the future is what it is regardless of what we do today”, because the future depends precisely on what we do today. Our actions today determine the future.

    Of course we are. The present determines the future. What we do today determines what will happen in the future. How can you say that this means “we are not preventing nor reducing the chance of any event”?

    Incorrect. It means there are no “free agents” who can act to cause events “through their own free will”; but deterministic agents have causal powers, and their actions determine future events. It is instead the entire concept of libertarian free will which is incoherent.

    You are with respect confusing “agent” here with “free will agent”. An agent by definition can be 100% deterministic, and can refer to itself as “I”, but you seem to think that all agents by definition must be “free will agents”, I’m not sure why.

    Can anyone show how strongly emergent phenomena could account for libertarian free will? I don’t think so (which makes the appeal to emergent phenomena as an explanation for free will ineffectual). Any event A must be either deterministic (determined by antecedent events), or indeterministic (random), or stochastic (probabilistic). There is no logical or rational alternative to one, or a combination of, these. Would you like to explain how any combination of these might give rise to something we call libertarian free will?

    The mechanisms of emergent phenomena can be rationally and coherently explained, at least in principle, in a manner which is consistent with determinism. Would you like to propose one phenomenon that is generally agreed to be an emergent phenomenon but cannot be rationally explained in this way?

    The problems with the notion of libertarian free will is (a) that it explains nothing that we observe about the world that cannot be explained by determinism and (b) that there is no way even in principle that the mechanism of free will can be explained on a rational or coherent basis.

    Best Regards

    Last edited: May 28, 2006
  20. Jun 1, 2006 #19


    User Avatar
    Staff Emeritus
    Gold Member

    I've voted no, simply because I hold that the only requirement a society needs to hold its member morally accountable for their actions is a fairly well agreed upon set of moral norms as well a fairly well agreed upon set of circumstances under which a person is to be held accountable. There may or may not be any metaethical justification for this practice, but I don't think there needs to be. A purely pragmatic justification is enough. The members of a society have the right to create a culture by which their society will continue to cohesively exist, and holding members of that society morally accountable for deviating from the norms of the society is one way of doing this. This does get us into very tricky waters of what exactly constitutes an acceptable norm and under what circumstances a member of a given society is justified in not accepting an unjust norm, but I don't think these need to be addressed for the purposes of this discussion.

    The above argument is question-begging, though. Premises 2 and 3 both contain hypothetical conditionals concluding that one is not morally responsible for one's actions, and uses these to conclude that one is not morally responsible for one's actions. A further argument needs to be made for why we should accept these hypothetical conditionals. Why should an action being either caused or uncaused be reason to preclude its actor from moral responsibility? That seems to be the very question we are attempting to answer, and one cannot argue for a negative answer by assuming that answer from the outset.
  21. Jun 1, 2006 #20
    Hi loseyourname

    Thank you for a very well written post.

    Agreed. The premises make implicit assumptions about the definition of moral responsibility.
    Premise (2) is making an implicit assumption that moral responsibility entails free will.
    Premise (3) is also making an implicit assumption that moral responsibility entails actions are not random.

    I believe the assumption in Premise (3) is valid, but the question as to whether moral responsibility entails free will is the very question of this thread. I also believe it does not.

    If we assume that moral responsibility does not entail free will, then Premise (2) is false, and the Argument is unsound.

    Best Regards

Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?

Similar Discussions: Does moral responsibility entail libertarian free will?
  1. Morals (Replies: 45)

  2. Logical entailment (Replies: 8)