The Impossibility of Moral Responsibility by Strawson

http://search.dilbert.com/comic/Free-will
Introduction and Context
There is something very important at stake in the free will vs determinism debate:  is moral responsibility possible?  Strawson's answer is "no" but this follows whether determinism is true or false! If Strawson is right, and we can't be morally responsible for our actions, why the heck do most people think that we can be morally responsible for our actions?  Because free will is an illusion.

In formulating his position, Strawson proceeds as follows:  (a)  overview of the Basic Argument and why we must accept it; (b) account of what moral responsibility means; (c)  why people think there is such a thing as moral responsibility; and (d) why, if we accept the Basic Argument (which we must), moral responsibility is impossible.

The Basic Argument:  Simplest Form
The Basic Argument goes like this:
(P1)  Nothing can be the cause of itself.
(P2)  In order to be morally responsible for one's actions, one would have to have caused one's self--at least the mental qualities and dispositions.
(C)   But since (P1), one cannot be morally responsible for one's actions.

If this seems a bit confusing, lets go through the expanded version, using an example.

Little Ami, Fries, and Mustard
When I was an elementary school student, it was a special treat to get french fries--especially for me.  I was the kid who had the lunch that all the other kids laughed at.  Usually my sandwich was something like cream cheese and alfalfa sprouts or tuna.  Instead of chocolate milk, I got plain milk.   Instead of store-bought chocolate-chip cookies I got homemade oatmeal raisin cookies.  In retrospect, I'm glad this is what I ate, but at the time it was pretty traumatic.

Anyhow, all this to say that to eat french fries at lunch was a huge deal for me.  The problem was, however, that if you got french fries at lunch, all the other kids would want some of your fries.  I'm a ten year-old kid that has to eat alfalfa sprout sandwiches and never gets fries.  I don't want to share!  At the same time, my parents raised a good boy and I was aware of the social "wrong" of being greedy and that I should share. So, what's a kid to do?

Here's what I did:  At that age, most kids hated mustard--including myself.  But in order to navigate the problem of wanting to keep all the fries to myself and conforming with social expectations I decided I would teach myself to like mustard!  Now, whenever it was a rare fortuitous fries day, I'd cover my fries in mustard and the other kids wouldn't ask me for any! (Incidentally, I did the same thing with black licorice).

Writing this, I'm starting to think I was a pretty strange kid.  And at this point you're wondering, how the f**k does this fit in with determinism and moral responsibility? Let me try to explain by taking one or two steps back:

Step 1:   Our actions are a direct consequence of who we are

Whatever we do is a consequence of a collection of our mental properties such as desires, beliefs, and psychological dispositions. This is why we can usually anticipate the action of people we know really well. We know something about their desires, beliefs, character, and dispositions. This is also how we explain why different people react differently to similar situations--they have different psychologies.

It would be really weird if our action weren't a consequence of our psychology. Our actions would seem totally random. Our actions are a direct consequence of who we are (psychologically).  This much is fairly uncontroversial.

Step 2A:  We don't cause our own psychological make-up
We don't choose our own psychological make-up.  Did you choose to like chocolate?  Did you choose to want to go to college?  Did you choose to like the people you like?  Did you choose to want the things from life that you want?  Did you choose to get sad, angry, or happy in response to various situations?  The answer, is no:  We don't choose or create our own psychological make up.

Consider little Ami.  Did he choose to like french fries so much?  Did he choose to be mindful of social conventions? The answer is no.  

Step 2B:  If we can't cause our own psychological make-up then we can't be responsible for our actions
So, here's the thing:  If we accept that our actions are a direct consequence of who we are psychologically and we don't choose or create our own psychological qualities, then how can we be morally responsible for the consequences of the actions that arise out of that psychology?  In order for us to be morally responsible for our actions we have to have created our own psychology, but this is impossible.

Step 3:   Objection:  I deny step 2A--We can cause our own psychological make-up
Suppose we deny that it's impossible to create your own psychology: you can create your own psychology and so now you are responsible for you actions.

Consider little Ami.  This seems to be what he's doing:  Ah! Ha! he says!  I know a way out of this dilemma.  I'm going to make myself like mustard!  So, there, Strawson and determinists!  Ami just created his own psychological make-up.  Basic argument falsified!

Step 4: Doh!  Stupid Infinite Regress! 
Free at last! I'm free! Free! I tell you!  Um...  Oh....  There is one small problem.  Wasn't little Ami's decision to like mustard a consequence of prior psychological properties? Isn't this the case anytime we resolve to "be" a certain way now?  Consider anytime you've resolved to change something about yourself.   You might have said something like, "I'm going to be a nicer person from now on, gosh darn it!"

It seems like we are self-creating some aspect of our psychology but is this really the case?  Doesn't there have to be an already existent psychological self that wants to change?  The desire to change exists in the prior self. Without it, no change would occur.  

Where did the self that has the desire to change come from?  Did you create that self?  Maybe you did, but then where did the psychological make-up that led to the psychological make-up that wanted to change come from?  Did you create that set of psychological properties?

Lets return to little Ami.  We might say little Ami self-created a new version of himself that likes mustard.  So there! Ami's psychology is self-created.  But again, the pre-mustard-liking Ami had a pre-existing set of psychological properties and dispositions that caused Ami to become mustard-liking-Ami--it had the pre-existing desire to want to like mustard. But Ami never chose to be pre-mustard-liking Ami.  He "just was" pre-mustard-liking Ami.

We might reply that there was a pre-pre-mustard liking Ami who chose to be pre-mustard liking Ami. But, did Ami choose to be the pre-pre-mustard-liking Ami that would eventually make possible pre-mustard-liking Ami which in turn would make possible mustard-liking Ami?  For Ami to be responsible for being pre-mustard-liking Ami, he has to have been responsible for being pre-pre-mustard-liking Ami.  At some point we have to concede that the initial psychological make up that bring about later incarnations of Ami are not self-caused.

We have, on our hands, an infinite regress problem.  The long and short of it is that every decision "you" make for being a certain way implies that there was a prior "you" that made that decision to take on a new psychological identity/property.  

Step 5:  You didn't create your prior psychology so you are not responsible for the actions caused by consequent psychological make-ups 
Did you choose to create the prior "you" that had the desire to change or be different?  If you didn't, then you can't be morally responsible for the actions that are a later consequence of that psychology.  If you did, then that decision to create a psychological make-up was a consequence of a decision made by a prior "you".   But did you intentionally create that prior "you"?  At some point we have to say "no" and since all present incarnations of your psychological dispositions are a consequence of your prior dispositions, "you" cannot be held morally responsible for the actions that emerge from those psychological dispositions.

Possible Objection:
People can change the way the are.

Reply:  True, but as has been shown, they can't do it in a way that attaches moral responsibility to their actions.

What is Do Most People Mean by Moral Responsibility and Why Do People Believe there is such a Thing?
The notion of moral responsibility that is typically implied by most people has to do with justification of punishment and reward in accordance with a principle of proportionality.   In other words, moral responsibility carries with it the idea that you should be punished (in part) in proportion with the amount of moral responsibility you bear for an action.

People believe there is moral responsibility because this is fundamentally how we experience the world. It feels like we make decisions that have moral consequences.  For example, you're going to walk into 7-11 to get you a cold pop (you smell something and you think someone is BBQ-ing).  On the way in, a homeless person asks you for a dollar.  It feels like you've got a moral choice to make.  You can give him the dollar or not.  And this is true even if you accept the Basic Argument.  Regardless, it still feels like we have freedom of choice and moral responsibility for our eventual action.  This is why we believe there is moral responsibility for actions.  It's the feels.  But it's not what is actually happening.

Overview of Freewill Vs Determinism

Introduction and Context
The debate over the degree to which we have control (if any at all) over our actions is one of the longest debated topics in philosophy and there isn't much reason to think it will end any time soon. The debate is known as the freewill-determinism debate.

Before proceeding lets quickly define some of the main positions (this in itself is a big issue in the debate):
1.  Libertarian (not political Libertarianism) freewill: In all circumstances, we are able to act otherwise (unless it entails a logical contradiction).  In other words, if I went to Roberto's and ordered the bean and cheese burrito with guacamole, I could have order something else.  Or if I was starving and stole a loaf of bread, I could have acted otherwise.  This is the strongest freewill position (not in terms of defensibility but in terms of the size of the scope of freewill).

2.  Hard determinism: The universe is physical.  The laws of physics tell us that every state is an effect of a prior state.  Freewill implies the suspension of the laws of physics because it requires that (a) an agent do something that isn't the result of prior causes and (b) the physical chain of cause and effect be broken.  (a) and (b) can't be true, therefore, all actions are determined.  

Even if someone wants to counter with "but the universe has mental properties too!" the hard determinist can rely "ok, but surely mental behaviors are also governed by causal laws.  You acted because you had a desire.  Cause and effect.  And what's more, you didn't choose to have that desire.  Did you choose to like chocolate? Or ice cream? Or philosophy 101?  No.  You just do. 

3.  Freewill Skepticism is closely related to hard determinism except it adds a dilemma for the freewill supporter.  At the quantum level of physics, events aren't determined but random.  So, either events are causally determined (by the environment and the physical make-up of the agent) or they are random.  If it is the former, they are not free.  If it is the latter, they are not free because randomness isn't freewill either.   Therefore, since all events are either determined or random, there is no freewill.

Positions 1-3 subscribe to incompatibilism which is the idea that freewill and causal determinism are incompatible.  You can only have one or the other, but not both. 

In contrast to incompatibilism there is also compatibilism, the idea that freewill and causal determinism are not mutually exclusive.  The classical defense of this position redefines freedom as a decision being free from compulsion or coercion.  So, suppose someone puts a gun to your head and says "write an essay about freewill and determinism," the classical compatibilist would say your decision to write the paper isn't free.  So, now "free actions" applies to those actions where there was no coercion or compulsion even though they are causally determined. 

The more modern compatibilist position is that there is a distinction between acting freely and having freedom of will.  I might be suggested that freedom of will is a matter of acting on our desires.  But Frankfurt replies that this isn't enough.  Freedom of will requires that we have some control over what we desire and that those desires motivate us to action.  This is as opposed to us being some entity that is pre-programmed with desires but has no say in what those desires are.  Note that this position still agrees that causal determinism is true.

Why the Crap Does the Freewill Determinism Debate Matter?
The main reason people think it matters is because the notion of moral responsibility implies that people have some control over whether or not they do something.  Do we hold our computers morally responsible for their actions?  Except for that one time I had a paper due in 15 minutes and the printer decided at that moment not to work, the answer is usually no.  Most of us think that justice, in part, requires that a punishment have some sort of proportional relationship to moral responsibility, but if my actions were all causally determined, it seems to follow that I'm not morally responsible for them. 

Related to the question of proportionality, we also think people deserve praise, reward, or condemnation based on their actions.  Again, if there is no freewill, it doesn't seem to make sense to praise them.  But maybe we're causally determined to praise/punish them anyway!  And maybe we are causally determined to puzzle over this problem...This can get very confusing very fast.

What Can We Know About Morality from the Concept of Freedom?

Is Morality Self-Evident?

Preamble

     Up until now we have just taken it on faith that Kant is right in that the supreme principle of morality is the Categorical Imperative (CI) in its various incarnations (i.e., formulation of universalization, of natural law, and of humanity as an end).  Now Kant begins his argument for why the CI is the only fundamental moral law.  For Kant it is not enough to give evidence for the CI, he wants to give a logical proof.  In philo-speak, Kant wants an a priori proof of the CI rather than a synthetic proof; that is to say he wants to prove the CI without any appeal to the experiential world; he wants to show that we can come to know the CI simply through rational/logical reflection of concepts.

     An obvious question is, why is Kant so hung up on avoiding appeal to experience?  The general answer is that if we appeal to external reasons to support moral thinking then if those reasons change, so will our moral laws.  For example, suppose after cooking you dinner your friend asks you how the meal was.  In fact, it was terrible but you know telling him this would break his precious heart, so you lie; that is, you lie for the reason that you don't want to hurt your friend's feelings.  

     A few weeks later the same friend announces they will be a contestant on a cooking show and would like your feedback on a dish they will enter in the competition.  He serves you the same dish they served you the previous week; this time you tell your friend that the dish isn't so good for the reason that you want to help them win the competition.  This also seems like a perfectly normal thing to do but the problem is we are left no clear guide to determining the "correct" moral action in situations where people ask for your opinion.  

     Every time our reasons to act change, so do our actions; and reasons are contingent upon our (often) ephemeral desires.  How can morality be so fickle and still have any worth and meaning?  And besides in both situations there are multiple, sometimes contradictory reasons according to which we could have acted.  Maybe in the first dinner I could have told him the truth based on the reason that I think my friend wants to improve his cooking and can't do so without honest criticism.  Of course one could reply that the rule according to which we are acting is "do whatever is going to produce the most happiness for the most people"; and this might be true, but this maxim comes with a boat load of its own problems and discussing them will take us far afield from the task at hand, so I leave it for now.  The essential point is that, for Kant, if we appeal to circumstantial reasons for choosing our behaviour, then what is moral is at the mercy of our circumstances. 

The Concept of Freedom as an Explanation for Autonomy of the Will

     Lets get a couple of definitions out of the way:  The first is 'will'; by will Kant means the ability to cause yourself to act (provided you are a rational being!); for example when you get up in the morning you are acting on your will to wake up.  Freedom is the property of the will that allows you to act independently of external causes.  If we were merely subject to external causes we'd be no different than a ball of tumble weed getting blown around.  Within philosophy the notion of free will is by no means a settled matter; but Kant assumes it nonetheless because it is required for morality.  For example, if we didn't have freewill (our actions were nothing but the total effects of external causes) how could be be culpable for our actions?  It is an entirely reasonable position that morality only makes sense if we have freewill; but the degree to which we have it, if at all, is still an open debate.  The essential point is that in our actions we are (to varying degrees) free from the influence of external causes--through our will and the fact that our will is freewill,.  The technical term for this type of freedom is negative freedom--that is, freedom from external causes.

     Kant is not satisfied with this feeble notion of freedom and argues for a more robust positive freedom as well.  Positive freedom is not just freedom from external causes but the ability to be the cause our own actions (i.e., act on our will).  But although our freedom (USA! USA! USA!) somehow grants us exemption status to external causes (i.e., natural laws which apply to objects) it does not mean our freedom is "lawless".  Along with causation (i.e., we can cause our own actions) comes the notion of laws according to which causation must conform.  For if our freedom of action were lawless our actions would be random, and that is to be no more free than to be subject to external causes; rather our freedom must be a force that conforms with "a special kind" of immutable laws.  

     I'm a little skeptical of this claim.  It seems pretty ad hoc; I mean, isn't it convenient that there are special kinds of immutable laws to which our will ought(?) to conform?  How does Kant know? Is he Jesus?  And doesn't it seem strange that, although we are (as far as we can tell) comprised of only matter, yet we are somehow exempt from the laws of nature (causal chains) when it comes to determining our own action?  These are some problems but are tangential to the issue so lets move on...

     So, given that the special causal power that rational beings have (freewill) conforms to some special kinds of laws Kant defines freedom of will as autonomy.  "What else can freedom of will be but autonomy--that is, the property which will has of being law to itself?" What he means here is that freedom is the ability to act on the laws that you make.  This idea can be a bit tricky to wrap your head around but it is a powerful one: every time you act, you act according to some principle; so every action is, in a sense, an instance of you writing a law.  

     For example, The other day I inadvertently dropped my phone.  A person who saw me drop it ran after me to give it to me.  In doing this action they wrote a law of the will which might be: "if someone inadvertently drops their phone, return it to them".  As I mentioned before, this is an idea I really like, and sometime I like to go through my day thinking (before I act) that what I do will become a law--try it! It's fun!

     Next Kant takes, as he will later admit, a logical jump; he goes from the plausibly defensible "what ever action I do becomes a law" to "I should only act in a way such that whatever I do will be a good universal law".   And the joy does not stop there; if we do make this step we can see that "a free will and a will under moral laws are one and the same."  I think what he means is that a will that is free will always act in such a way that its actions make universalizable law and a (rational) will will always follow the "special laws" governing the will (which magically turn out to line up exactly with the CI--i.e., the universalizable laws).  It all sounds a bit too good to be true...

     Remember, the whole passage started with the intent of trying to extract, a priori, purely from concepts the principle of morality.  Kant says if we analyze the concept of freedom of the will (aka. autonomy) we can, without reference to the empirical matters, deduce the principle of morality (CI); that is to say, the notion of the CI in contained in the notion of freewill.  But not so fast.  First of all, this whole thing only works if we suppose freedom of the will.  And second, even if we suppose freewill and from freewill deduce the notion of morality, this does nothing to guarantee we can discover the content of morality--i.e., "a good will is one whose maxim can always have as its content itself considered as a universal law"--there is no reason why the content of morality might not be different; what if if it is, "act only in such a way that will maximize happiness for the greatest number of people?"  Simply analyzing the concept of freewill doesn't give us any indication either way...

     Recap: We started by saying that rational beings have free will in a negative sense (I can cause myself to act), then we said we also have positive free will (I can do the things I freely chose to do and I am not subject to external causes on how I act) and since positive freewill causes things to happen, it must be subject to some "special" causal laws (i.e., other than the laws of objects).  Because I have the ability to chose what I want (negative freedom) there is such a thing a morality; morality arises out of the fact that I can chose between "good" and "bad" things/courses of actions; i.e., the concept of morality can be known from rational reflection of the concept of negative freedom.  Regarding what can be known from the concept of positive freedom, we can know that there are (moral) laws that govern the will (because anything with causal power must act according to laws); however, even Kant admits that we cannot know what these laws are purely by examining the concept of positive freedom; we can only know that they exist, thus we need to look elsewhere for an a priori grounding of the CI...