Determinism and Disneyland

Okay, I think I am convinced that moral responsibility doesn't require the ability to do otherwise.  I find the Frankfurt-type counterexamples compelling in this regard.  And perhaps even independently of the counterexamples, I think it makes more sense to say that moral responsibility must depend on whether or not you endorse what you are doing, or like what you are doing, or identify with your action, etc. than on whether or not you could have done otherwise.

That said, however, I'll also say that I'm not convinced that moral responsibility is compatible with determinism.  In fact, I think the two are incompatible.  Determinism, I think, is less like a Frankfurt-style counterexample and more like Disneyland.  Let me explain.

You know that Disneyland ride where you get into a car that's fixed to a track and then you ride around on the car?  Well, every time I go there I always have to fight to get the seat with the steering wheel. (We all want to be the driver, don't we?)  And occasionally I succeed, and I get to pretend like I'm driving the car.  Of course, I'm not actually driving the car, and I realize this. 

But now suppose that I don't know that the car is on a track, and in fact I think that I am controlling the car.  I turn the steering wheel to the right when I come to a turn, and (what do you know?) the car goes to the right.  I have no idea that I didn't have any effect on the direction that the car turned. It seems to me that this is what determinism would be like, if it were true.  In this case, I think I'm not morally responsible for the fact that the car went to the right.  Of course, I may be morally responsible for something -- like turning the wheel, or something like that.  But not for the fact that the car went to the right.

Compare this to a case where I am actually driving a real car that's not on a track and I turn right, but unbeknownst to me there was a counterfactual intervener that would have manipulated my brain and made me turn right had I shown an inclination to turn left.  As it happens, I turn right on my own and the counterfactual intervener need not intervene.  In this case, I think I am morally responsible for the fact that the car went to the right.

It may be that what's worrying me is sourcehood.  And I'll admit - it's the fact that determinism is an actual-sequence mechanism that bothers me, whereas a counterfactual intervener is just a alternative-sequence mechanism.  So, the question for you source incompatibilists out there -- why is determinism more like Disneyland?  What about its being an actual-sequence mechanism makes the difference?  And for you compatibilists out there -- why isn't determinism like Disneyland?

Comments

[Andrei A. Buckareff]

Is it really accurate to say that the way our actions would be determined if determinism is true is how you characterized it? You have conceded that something like an internalist theory of personal autonomy (e.g., the views of Bratman, Frankfurt, Velleman, and Watson) gets things right for purposes of moral responsibility; but the sort of autonomous agency we have on such a view is allegedly undermined by determinism because determinism is, in your words, 'an actual sequence mechanism'. Now, on internalist theories of personal autonomy, the actual sequence is all that matters for the purposes of ascertaining whether or not some exercise of agency was autonomous. The agent must identify with her object level motivational states, or something like that must occur (we'll ignore the details), in order for the action to be autonomous.

On the foregoing rough sketch of autonomous agency, what makes your action autonomous is that it comes from you and is controlled by you (I'm ignoring any details). If determinism is true, however, the control does not come from outside of you. And this is what is wrong with the Disneyland carride analogy. You are not like the driver of the car on that ride who thinks she is in control, but is not. As a matter of fact you are in control, only, since you are part of the natural order, your activity is like any other event in the rest of nature and is the consequence of deterministic causal processes and subject to the laws of nature. Keep in mind that the mechanisms which are efficacious in the etiology of your action are all within you. They're not like the tracks of that horrible ride at Disneyland.

At this point I have to shift a burden back on to you. What good will indeterminism do for you if you go with the sort of theory of personal autonomy you believe allows you to reject PAP but you remain an incompatibilist? Is the phenomenology of agency any different if determinism or indeterminism is true? If not, does this count for anything? I'm just curious.

(P.S. I've intentionally avoided using the term 'free'. I do not believe that 'free' and 'autonomous' denote the same concept. I am not sure this will make much of a difference for your question. I suspect it won't.)

[Eddy Nahmias]

The disneyland image can be found, I believe, somewhere in Carl Ginet's work. I think it plays a powerful role in arousing incompatibilist intuitions. And I think it is very misleading (largely for reasons Andrei suggests).

The image conflates determinism with consciousness epiphenomenalism and hence with a vague sort of fatalism). Consider the counterfactuals. If you were to chose to turn the steering wheel the other direction on the ride it would have no causal influence on where you go. But if you were to choose to do otherwise, it would (barring bizarre conditions) have a causal influence on what you do. Your actions depend on your choices in a way the ride does not. (And it may also be that what past or laws are actual also depends on your choices, but that's a more contentious matter.)

Of course, what you choose may be causally determined, etc., bringing back (source) incompatibilist worries. But they should not be worries driven by the idea (or image) that our conscious deliberations and decisions are not really having any causal influence on our lives. Now, there may be other reasons to worry about consciousness epiphenomenalism, but they are worries that are entirely distinct from determinism and that are consistent with indeterminism. I'll post a paper about this point one of these days, but my view is that the free will debate should be about the question of mental causation and potential threats from certain reductionistic views and that the issue of determinism grabs us primarily because it suggests these threats (and perhaps others it does not actually entail). Happy holidays!

[Clyde Dunton]

I would also agree with Andrei in that it seems Neal is attempting to describe some sort of scenario that I would not identify with determinism as I concieve it. To be determined is to have a real steering wheel in your hand but the turn you are about to make has been in the cards forever (ok maybe not forever but a long darn time). Sourcehood is also very problematic for me to. I have found that Fischer, Ravizza, Frankfurt, and Wolf (re: praise deserving moral responsibility) all fail to address this concern to my satisfaction. Also I can't help but notice that many incompatibilists rely on the 'feel' of things or other factors that ultimately do not bear on the issue of grounding responsibility in/on the agent. If an agent has the sensation of owning an act that may be enough for some, but I feel that genuine characteristics of moral responsibility must include some type of source/control/origin requirement that cannot be fulfilled simply by making appeals to the qualitative nature of agency. Even if the agent sees themself as an agent and has an "actish" feel these are decidedly not sufficient characteristics of moral responsibility. If I am shooting a three ball pool combo and am able to make the second ball in the combo feel like an agent and have an "actish" qualitative feel, that feel does nothing to ground responcibility. While it may be cute that the second ball asserts its responsibility and demands praise for sinking the third ball it still seems that it has nothing more than causal responsibility insofar as it would be impossible for the same event to happen as it did without the second ball present. Is qualitative sensation alone sufficient for generating characteristics of genuine moral responsibility? I can't help but feel that this is where many compatibilists base their intuitions...

[John Fischer]

Yes, as Eddy Nahmias pointed out, the Disneyland ride image comes from Carl Ginet; I discuss it in The Metaphysics of Free Will. Put briefly and no doubt unsatisfactorily, Neal, I believe that not all causal sequences are created equal. The car goes the way it goes in the ride via a certain sort of causal sequence, one which is not the agent's own, moderately reasons-responsive process. On the other hand, causal determination in itself does not entail that the sequence in question is relevantly similar; causal determinaton is, Ravizza and I have argued, compatible with guidance control, i.e., compatible with the pertinent action's issuing from the agent's own, suitably reasons-responsive process.

Of course, Disneyland IS the "happiest place on earth," and I doubt that we semicompatibilists have much to offer in this respect....

[Clyde Dunton]

"causal determinaton is... compatible with guidance control, i.e., compatible with the pertinent action's issuing from the agent's own, suitably reasons-responsive process." i am wondering, to interject sloppily, what thing/stuff is doing the guiding and the controlling in the causally determined scenario? Is it the causal momentum of the universe (that seems to me unresponcive)? Also i'm wondering what type of ownership the agent has over the R-R process? The history requirement seems to help but doesn't the history have a history of "its own"? Does this type of ownership only occur when the causation has trickled up/down to the agental level? Does the agent's history's history own anything? If determinism is so it sure does some guiding and controlling... of the agent's R-R process, or so it would seem.

[John Fischer]

Good questions. Especially given the exigencies of the holidays, the best I can do is refer you to Responsibilty and Control: A Theory of Moral Responsibility (Cambridge U Press), Mark Ravizza and me. There is a precis in the last chapter. Also, there is a precis in the PPR book symposium.

And, by the way, these books make a WONDERFUL Christmas present; since they are green, they can go under the tree beautifully, even unwrapped.

[Conscious Robot]

Is this open for anyone to have a go?

‘Knight takes bishop’ has a series of potential outcomes that a chess computer must analyse. It must ‘imagine’ the outcome of each potential response to ‘knight takes bishop’. If it were a conscious human, it would ‘see’ the knight in its new position, and spot the threat from the rook. It would know that losing the knight wasn’t a ‘good outcome’ and would look for an alternate move. Maybe this would feel like having a choice, but it would simply be following its programmer’s instructions.

I have a choice of routes on the way to work – this morning, before I ‘chose’ whether to steer left or right, I thought about the routes ahead, imagining the possible traffic, the scenery and the radio reception. I then chose left in anticipation of the beauty of the frost on the overhanging trees. It felt like my own choice. But I didn’t chose that the frost on the overhanging branches would be ‘pleasurable’, any more than the chess computer chose that it wanted to achieve checkmate.

The chess computer was created by a human that programmed it to achieve one aim – checkmate. It doesn’t ‘know’ that it’s playing chess, it just knows which outcome is preferred. I was created by an automatic process of natural selection to achieve one aim – to maximise the survival chances of my genes. I don’t need to know consciously that this is my aim, I can leave that assessment to my non-conscious. My conscious mind has been programmed to fulfil desires and to maximise its experience of well-being: my non-conscious sets the desires, and informs me whether or not frosty branches will make me feel good. My non-conscious was programmed by natural selection, and my conscious mind merely follows the instructions. I am a conscious robot.

If I don’t yet know why frost on overhanging branches improves the survival chances of my genes, that doesn’t disprove that my choices - like that of the chess computer - are the result of automatic movements of electrons through silicon or carbon-based matrices.

If I were really free, I’d choose to ‘feel good’ all day, whichever route I took to work. This choice would of course be no more free than my current existence, but I’d enjoy it a lot more.

[Carl Ginet]

Yes, in my paper "Might We Have No Choice?" (published in Lehrer ed., Freedom and Determinism, 1966) I used the example of an amusement park ride with a steering wheel to illustrate the possibility of having an illusion that one is choosing which of several alternatives will take place (p. 102). I also noted that the example does not provide a perfect analogy to our situation if determinism (or rather a certain deterministic hypothesis I labelled "H") is true, because in the example "the ineffective or illusory choice of path is independent of the path actually taken..." I added an elaboration of the example to make it a better analogy by supposing that the path the car takes and the rider's impressions of seeming to choose its path are both controlled by some person other than the rider in such a way as to make them coincide. This makes the seeming choice of path and the actual path interdependent but (contrary to what I suggested in that paper) it still doesn't make it a perfect analogy to what our situation would be if determinism were true. For if determinism were true it would be the case, not only that the choice of path and the path taken have a common cause independent of the chooser, but also that this independent cause causes the path taken through causing the prior choice of path, which in turn causes the path taken.

[Michael Aune]

It is obvious enough that much of the frustration arising from the 'determinism v. moral responsibility' problem is the assumption, whether stated or not, that moral responsibility is inextricably linked to some concept of free will. Whether explicitly stated or not, this premise always seems to be lurking somewhere. For those who do not feel that the two are connected, the following may be of little consequence.
I ask, however, why is it necessary to be free in order to be morally accountable for one's actions? I understand the intuitive pull that the free will/moral responsibility entanglement has, but I question as to whether it is enough.
To state my point more clearly: For the sake of argument, assume that the determinism hypothesis is true. Others have commented that determinism removes moral responsibilty from an agent based on the fact that the agent's actions were determined by outside forces. Essentially, that we are all on a ride over which we have no control, as in the Disneyland example stated above. This example seems, however, to make a distinction between the agent and the natural laws. The agent is not something above and beyond the determined world; s/he is a manifestation of those laws, a manifestation as real as the feelings of moral responsibility most every person experiences. The natural order has given rise to the feelings of moral responsibility we attribute to agents, and I see no reason why there is any problem with this. Even if we are mere automatons, mechanically going about our lives, a part of that mechanism - a very real part, I argue - is the application of moral responsibility through our persons. In this light moral responsibilty stems directly from the most basic laws of physics, as fundamental a basis for anything as any strict materialist could hope to find.
I admit that this anaylsis (as rough as it is) will probably not satisfy many, but I believe that it is important not to elevate agents and the emotions, thoughts, etc. above the natural order in which they are found. Once one attempts to do this, the 'agent' and 'nature' seem destined to come into conflict.

[Joe Campbell]

Michael,

I wonder what it is that you think 'free will' means. If you think that "moral responsibilty stems directly from the most basic laws of physics," then why can't some kind of freedom or control stem directly from the laws of physics?

In answer to your original argument, I think that it is clear that some folks are not morally responsible for their actions. If we consider those who (a) have the the required cognitive capacities and (b) knowingly and intentionally perform some action yet (c) do not seem to be morally responsible for their actions, it seems that they lack some kind of freedom or control, which I and others call 'free will.' This is a kind of generalization argument.

My guess, though, is that you have some particular kind of freedom or control in mind when you use the term 'free will,' something that is by definition incompatible with determinism.

[Kip Werking]

I think one of philosophy's better kept secrets is the fact that the term "free will" is horribly ambiguous and (if Richard Double is right) does not refer to anything in particular. I remain fascinated with the problem because I think there is a disturbing *idea*, regardless of the semantics, which many people fail to appreciate (even if most sophisticated compatibilists in academia appreciate it): the problem is that all of our decisions can be traced back to accidents of birth and this seems to eclipse any *ultimate* control we have over our lives. This is the sense in which I use the term and I think Robert Kane refers to this sense in defining free will as the "ability to be the originator and sustainer of one's own ends and purposes." But I acknowledge, like Kane (and Fischer, who regards free will as an umbrella term), that there are other senses of free will. One reason for favoring this disturbing understanding of "free will" is the lack of alternative definitions that retain some controversy. If free will is just, as Hobbes said, the simple power to do what you want (regardless of how you acquired these wants), it is difficult to see why anyone would ever deny that we had this power.

If we focus upon this disturbing notion of free will, related to the tracing problem, I think we will realize that determinism makes the problem especially *conspicuous* but that indeterminism cannot help (as everyone except the libertarians seems to agree).

Furthermore, the term "morally responsible" is just as ambiguous (I think) as "free will" is. Nevertheless, there is a disturbing notion of "morally responsible" which correlates with the disturbing notion of free will. This notion is related to the problem whereby all of our actions can be traced back to accidents of birth. As with the term "free will", I think this understanding of "morally responsible" should govern in the philosophical context of the free will problem, and that free will and moral responsibility should roughly track each other in this sense.

Finally, as Galen Strawson said in his contribution to Kane's handbook, there is much more to be said about free will "and this is just the beginning." "But it is the beginning." I don't intend for this understand of the terms "free will" and "moral responsibility" to be final, although there is an obvious sense in which they seem to be, so much as I expect them to be *proper*. Using any other definition is, I think, a red herring. There is more work to be done in, for example, incorporating cognitive science, evolutionary psychology, and meta-ethics into this framing of the free will problem: what gave rise to this problem and, in the face of it, how should we revise our social and ethical systems? There is other work to be done in other areas too.

But, I humbly suggest, this is the proper way to understand the free will problem.

[Michael Aune]

Joe,

I must admit that I am using an ambiguous notion of free will, but I feel that there is little other choice. I also chose to use an incompatibilist definition since it seemed more pertinent to the discussion at hand. Those who already think there is no fundamental opposition between free actions and determinism have little reason to enter this discussion!
As Kip Werking commented above, there seems to be little consensus as to what exactly 'free will' constitutes. Most definitions implicitly or explicitly involve notions of moral responsibility, as well as the incompatibility or compatibility of in/determinism with it. There are numerous reasons why defining it as the ability to do otherwise is unsatisfactory, so I hesitate to use that within a definition. The same hesitation also arises whenever concepts of sourcehood or 'freedom' from the natural order is used. In short, while I have an idea as to what 'free will' should not be, I am currently unable to produce a strict definition. Forgive my ambiguity.
In any case, to respond to your first point, I see a rather large distinction between the concepts of 'moral responsibility' and 'systems of control' or 'freedom'. Feelings of moral responsibilty, which I should have clarified as the only reality I postulate for moral responsibility, can rather easily be seen as stemming from the natural order, provided one equates the brain with the mind. Systems of control, however seem to me to be distinctly different, owing to the fact that their arisal would involve a fundamental break with everything else we encounter in nature. No longer is it simply the collision and interplay of trillions of particles; now it is, somehow, the 'control' of those trillions of particles by a system of other particles which, while arising from that same mess, is somehow above it. While perhaps not logically impossible, it does seem implausible. To date I have not seen an argument which describes how or why such a system could arise. Of course, I am certainly not an expert on modern physics, and so I must defer to those who are more familiar with that topic.
Last, I hope to avoid any generalizations on my part. Obviously enough there are those that are outside of our moral consideration, at least where blame or praise is to be had. Those lacking the cognitive ability to distinguish 'right' from 'wrong', who nonetheless purposefully and intentionally perform actions, are morally unresponsible. I do not believe that this has anything to do, however, with a lack of free will on their part. If one views it as a lack of the required 'moral sense', that ability to recognize the 'moral' aspect of certain actions, that seems to be all that is needed. No mention of 'free will', 'freedom' or 'control' is necessary. If you are morally blind, it is much harder to hold you morally accountable. I find it very compelling that an analysis of this kind has no need to bring in theories or definitions of 'free will'. All it need be is a 'mechanical' reaction to a given impetus - a reaction that does not cause the attribution of moral responsibility to arise in the minds of others. This can fairly simply explain the reason why some are not held responisble in purely (again, for the sake of argument) deterministic, physical terms.
I hope that helps clarify my previous post in some way. For a first time philosophy blog experience, I certainly chose one hell of a topic to discuss!