Blog Coordinator

« Punishing with a Compatibilist Heart | Main | A Piece on Restorative Justice and Forgiveness over at Slate »

06/28/2015

Comments

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

I, for one, *am* interested in what Eddy has to say. Here's a worry. Some people think that the folk have an incompatibilitist theory of free will, but compatibilist case by case intuitions (especially with regard to concrete cases). They are going to say at the scale taps only into people's theory and misses their case-by-case judgements .

Neil,

I didn't mean to suggest I wasn't interested in hearing what Eddy has to say! I was only noting that I already have a pretty good sense of what he'll have to say (since we have been arguing about this behind the scenes and occasionally on this blog for the past five years). But I certainly want him to add his voice to the discussion (if only to give him another chance to say more about his "theory light" approach).

That said, I want to briefly comment on the worry you raised. What you say seems to stand the worry I raised on its head. I was noting how odd it is (at least to me) that the data from the various scales has played little to no role in the debate about folk compatibilism/incompatibilism. Instead, nearly all of the attention has been focused on people's responses to cases. Yet the way you framed your worry suggests the converse is the case--namely, the parties to the debate aren't paying enough attention to the vignette data (especially the vignettes involving concrete cases).

In the post, I wasn't taking a stand on which of the two data streams is more important (or more philosophically probative), I was merely asking why people think one stream of data has received so much attention and the other so little. For my part, I think *both* types of findings have a role to play in the overall project of understanding people's beliefs, attitudes, intuitions, and theories about free will (and related concepts). That's precisely why I think it's so odd that the scale data receives so little attention.

If there were solid grounds for ignoring the scale data when it comes to the philosophical debates about free will, that's fine as far as it goes--since I think that scale data is psychologically important regardless. But folk incompatibilists surprisingly haven't appealed to this data to support their view and the folk compatibilists usually don't even bother to try to explain it away (except perhaps in passing).

Be that as it may, here is one way of trying to capture the state of play as suggested by the worry you raise: (a) most people have incompatibilist theories about free will, (b) most people have incompatibilist intuitions about abstract cases, and (c) most people have compatibilist intuitions about concrete cases. If that's right (and I have my reservations), shouldn't that be the starting point of our discussion at this stage in the debate? Instead, people have privileged data on intuitions and scenarios over data on theories and explicit statements about free will. This approach seems unmotivated to me...

Well, I'd like to hear what others have to say too, in part because I'm still away and unable to respond in detail (and need to respond to some comments on my punishment thread), but mainly because I'm really not sure what I really think about these issues--I've certainly *said* a lot to Thomas over pints and fights, but I don't always *believe* what I say, and am still trying to work out what I believe. What do others think (including you Neil)?

Eddy, I have strangely selective memory. I remember the pints well enough, but I don't remember any fights :) I do, however, remember a duel or two (but neither of us had any bullets, so they were harmless exercises in futility)!

p.s. I hope you're enjoying your retreat (or whatever it is you're intelligently doing with your summer rather than blogging).

Cognitive science provides good reasons for why cases are much more important than statements for getting at the underlying folk “conception” in a domain. It helps to look at a concrete example. Look at what is on with syntax and grammaticality judgments. According to textbook linguistics, people have an elaborate tacit theory of syntax that gives rise to their particular judgments about cases. The theory is pretty darn sophisticated and ordinary people have little idea about its contents – the theory is *tacit*.

To drive this idea home, look at this example. Everyone knows you can change:

(1a) Kim is happy

to:

(1b) Kim’s happy

But you cannot change:

(2a) Kim is happier than Tim is

to:

(2b) Kim’s happier than Tim’s

If you ask people what is the rule they are following, they will basically just make something up. The underlying rule turns out to be ridiculously complicated – basically the word following “is” can be omitted except when the “is” is on the left side of an extracted phrase. So we all have a tacit theory with a rule along these lines, and the theory gives rise to intuitions about particular cases. But if I ask people about the rule itself, they would have no idea what I am talking about.

It turns out tacit theories are fairly common in cognitive science. I have an argument in a paper in prep that a lot of philosophically-relevant intuitions arise from tacit theories. It is particularly plausible that a lot of intuitions in action-theoretic domains are tacit theory based, and free will is, in my view, an action-theoretic notion.

So here is the picture I have in mind. People have a tacit theory of free will, much like they have a tacit theory of syntax. The theory gives rise to intuitions about cases. Much like grammaticality intuitions provide the linguist with excellent evidence for revealing our underlying tacit theory of syntax, free will intuitions provide the philosopher with excellent evidence for revealing our underlying tacit theory of free will. But if we present people with statements directly and ask essentially “Is this part of your conception of free will”? their responses are going to be unreliable. This is because the theory at issue is tacit. Just as in the case of grammar, you can’t ask people its contents directly.

Eddy, Thomas, Chandra, Neil--

In earlier posts to Gregg I wondered whether if many if not most people worldwide engage in what I termed "the false consequence argument"--namely that they believe that because they can conceive of different logically possible future courses of action, then they falsely conclude that those various courses are really accessible to them. (In my classes I term this the false inference from logical to metaphysical possibility as the latter is parsed in terms of ability.) In part this would explain why when people are faced with hypothetical scenarios of global determinism they return answers consistent with incompatibilism. But since people are also very practical when faced with specific scenarios, then it doesn't surprise me at all that they then return answers more consistent with compatibilism. Here's what I want to know: can we test the "false consequence argument" against the specific scenarios argument in some direct way that shows two different ways of thinking that people might engage in that goes beyond generalized data--namely that people sometimes think as incompatibilists and sometimes as compatibilists as specific individuals? What I want to do is narrow that claim down to specific people thinking in inconsistent ways.

Chandra,

The view you've sketched seems to work a lot better when it comes to syntax, grammar, etc. than it does to free will beliefs. Take Chomsky's famous sentence pair:

Colorless green ideas sleep furiously.
Furiously sleep ideas green colorless.

Given pairs of sentences like these--namely, one sentence that is syntactically correct but semantically nonsensical and one sentence that is neither syntactically correct nor meaningful--people will nearly universally be able to distinguish the former from the later. Moreover, as you point out, it is unlikely they will be able to identify the rules by which they arrive at these judgments--as these rules are known tacitly.

So far, so good. You run the same line on the sentences you give as well. But these kinds of judgments about syntax, etc. seem importantly different than judgments about free will (and related concepts).

For instance, one of the hallmarks of judgments about grammar is that they are stable both within and between subjects (and within and between cultures). On nativist accounts, this isn't surprising given that the underlying rules are innate (and hence deeply entrenched and stable). But this isn't what we find when it comes to judgments about free will--even judgments in response to cases (which you suggest are superior to judgments about statements). Instead, we find within and between subject disagreement both within and between cultures (depending on the details of the cases, depending on the font in which the cases are written, etc.).

This disagreement, variability, and instability suggests that the free will judgments are not driven by the same kind of mechanisms as grammaticality judgments. That's not surprising. One ought to expect our understanding of language to cut deeper than our beliefs about free will (although our beliefs about dualism and indeterminism might be a different matter).

Here is another difference between grammaticality judgments and free will judgments: Even upon reflection the folk may be unable to articulate the rules/theories that drive their grammaticality judgment. As you point, these rules/theories are sometimes ridiculously complex (even in the case of fairly basic sentences). But the same doesn't seem to hold for free will beliefs. When given the chance in class, my students regularly articulate their reasons for holding their free will beliefs and these explanations are fairly consistent.

For instance, those who naturally trend compatibilist tend to give one sort of explanation and those who naturally trend incompatibilist tend to give a different sort of explanation. And while it's certainly possible they could be making stuff up, there's no obvious way of checking to see whether this is the case in the case of free will judgments--which is yet another difference between these judgments and grammaticality judgments. In the case of the latter, we can identify the underlying phrase structure rules.

So, when people make judgments that comport with these rules but then offer explanations that don't appeal to them, we can see that they are making stuff up, etc. But what is the analog to phrase structure rules in the free will case?

Of course, you might still insist that despite the differences between free will judgments and grammaticality judgments, our theories about free will are nevertheless tacit. But then you'd need a different argument for why we ought to privilege judgments about cases over judgments about statements in the case of free will. You can't just draw an analogy between our free will judgments and our grammaticality judgments to support the view.

It's also worth pointing out that it's unclear to me whether the example you use involves cases rather than statements. I was using case to mean vignettes. And it doesn't seem like the "cases" used to probe grammaticality judgments are vignettes. Instead, people are presented with statements/sentences and asked simple questions like: Does this sentence make sense?

All that being said, I want to ask you a different question. Given that you don't think scales gives us access to tacit theories, to what do you think they provide us access? Presumably, you think they shed light on psychological constructs. So, let me phrase my initial question differently: Why think that data on tacit theories about free will are more important than data about the psychological constructs that presumably undergird these theories. After all, it would be odd if these psychological constructs played no part in the tacit theories.

Given that you were a member of the team that constructed The Free Will Inventory, I am especially keen to hear what you have to say on this front. Is it that you think that the scale data is psychologically probative but not philosophically probative or is it that you think that the scale data is simply less relevant or probative than the vignette data (or something else)?

Thomas, You raise a lot of really excellent points. I can’t address them all because I am not supposed to be working (I am at a beach house in Kiawah Island all week, probably a couple of miles away from you). But here are a couple of quick points.

I am using tacit theories of syntax and grammaticality intuitions to illustrate one of the most important property of tacit theories: opacity. We can apply the principles that constitute the theory reliably to concrete cases, but we can’t articulate the principles themselves. The grammar case makes the idea vivid. I am not implying all tacit theories resemble the grammar case. There are lots of ways in which syntax/grammar domain differs from the free will domain, which you correctly point out. I am fine with that.

You rightly point out one big difference is that in the free will domain, the principles will likely be more ordinary -- they won’t appeal to ridiculously abstract ideas like Chomsky’s notion of c-command, etc. That said, free will is linked with having certain and abilities and powers, and formal approaches to characterizing the semantic structure for statements involving these -- such as the Lewis-Kratzer semantics -- are indeed pretty abstract and sophisticated.

I have no idea what we measure with the Free Will scale. The scale is itself fine in the psychologist’s sense -- it has great psychometric properties. But, I don’t think you can make much progress on the kinds of questions that interest philosophers using a scale. That doesn’t mean Xphi has no role. XPhi can be very helpful in uncovering tacit theories. The XPhi methods that are helpful, however, are different. They aren’t scales.

1. I'm really enjoying the back-and-forth between Thomas and Chandra.

2. Chandra raises an interesting analogy with grammar, which I had not considered before. I share Thomas's reservations, but I also acknowledge the force of Chandra's argument.

3. I lean skeptical about free will, and here is my default view on surveys. I'm worried about this situation: people have a thoughtful definition of free will, which they define in abstract conditions. This definition makes sense for various philosophical reasons. When people try to apply it to concrete cases, however, they *fudge* the application because their answers are results-oriented. They are fudging the application to get the results they want.

Think of it this way: the survey taker wants to eat his cake and have it too. The survey taker wants the benefits of the libertarian definition: freedom from the past, radical self-creation and self-direction. But the survey taker also wants to say that free will actually exists, confirming our prejudices and status quo system of moral responsibility and criminal justice. The survey taker thinks at first that he can have both, but realizes (upon careful application) that you can't, because of Strawson's Basic Argument and similar arguments. So the survey taker fudges the application.

So my view is a competing hypothesis to Chandra's theory of tacit concepts. Which one is correct? I'm not sure I can prove which one is correct now. But I want the debate to be framed in a way that acknowledges the potential for my theory to be true, pending empirical falsification.

4. Another point: similar to point 3, I continue to suspect that, in many contexts, definitions *precede* applications. So people are more likely to get definitions accurate than to get applications accurate. For example, if you ask people what the number 5 is, they are more likely to get that correct than if you ask them what 5*5*5 is. This is because in the first question, they only have to define 5 correctly. In the second question, they have to get 5, *, and the overall computation correctly. By simple probability, A is more likely than A&B&C. This seems to me to be generally true and a kind of default view on definitions versus applications. But we still have to consider Chandra's grammar analogy as an important counter-example.

5. The last thing I'll say is that: Chandra's theory of tacit concepts seems to result in some really weird consequences. If we assume that people's application of free will is correct (in some sense) and reveals a correct-but-tacit concept of free will, well what the heck is that tacit concept going to look like? I think it's going to be a freaky kludge, because we'll discover that a 1,000 different factors end up significantly affecting how people make attributions of free will. There is a giant swath of psychology studies showing how things that *should* be irrelevant nevertheless have significant impacts on our behavior (see, as one example: the situationist literature on automatic reactions, or the Knobe effect). I'm not sure that Chandra actually wants to follow his tacit-concept argument to where it will ultimately lead.

Hi Thomas-

I don't know how illustrative my own experience is, but I've never discussed inventory work, despite some affection for empirical work on these issues. This isn't out of dislike for inventory-based work but (I think) mainly a byproduct of the case-based work getting published and kicked around first. Work takes time to be propagated and digested, and the case-based stuff got there first. (Right? Or maybe I'm just missing something—was there a comparable amount of reasonable inventory data available early? I remember thinking the inventories created without philosophers being problematic in some important ways so I just never seriously engaged with that. I gather that the newer, philosopher-involved inventories have solved those problems, but this is still a relatively recent phenomenon.)

Having pleaded for the accidents of history, I suppose I'm sympathetic to some of what Chandra and Neil say about the comparative appeal of cases to inventory measures. But, I wonder if there isn't a disciplinary aspect to this as well. I suspect my own sympathies are partly a product of philosophical training that was in no small measure grounded in thought experiments and arguments based on cases. Indeed, the most familiar way to test the plausibility of various claims (including, perhaps, those in the inventory) was usually to try them out in concrete cases. So, given this kind of training in philosophy, it seems to me unsurprising that many philosophers would gravitate toward case-based studies. That's not an argument against the utility or informativeness of inventory-based studies, of course—just a gesture at part of an explanation for the limited uptake on inventory studies (thus far).

Okay, even if my inquiry about my "false consequence argument" test of abstraction against specific scenarios din't exactly light this thread on fire. . .

Can anyone tell me if there have been statistical analyses of responses comparing the results of more abstract questions against those of particular instance scenarios to extract how many individuals, on some sort of average against some sort of group, espouse contradictory views about FW and responsibility on that particular basis? What bothers me is that stats about one or the other kinds of questions may not themselves yield insight about how many individuals out of some ensemble contradict themselves. Contradictory views as some sort of stat contrast is not as informative as how many individual people actually invoke inconsistent views, at least to me.

I'm assuming because of my ignorance of the literature that someone has done this kind of analysis, at least with respect to some demographic groups.

Alan,

Sorry, I spent up all my energy and time yesterday responding to Chandra (in this post) and Tamler (in the other post). Here are a few quick replies to your question for this morning! First, most of the studies that have been run on folk intuitions have relied on a between subject design rather than a within subject design. That's because it's the easier way to cleanly isolate the variable being manipulated. So, for instance, one group gets an abstract case and another group gets a concrete case. Here we find differences between the cases. You certainly could run these cases with a within subject design (with a randomized ordering of the cases). If anything, I would expect to find that the way participants respond to the first case anchors their responses and then influences how they answer the second case. In order to preserve consistency/avoid inconsistency, they will tend to respond to the both cases similarly. If this prediction's right, what we would find would be fewer differences between case types (e.g., abstract and concrete). But this consistency would be dependent on the order in which the participant happened to receive the cases. That's partly why most of the work has used between subject designs. Of course, it would be interesting to rerun some of these studies with a within subject design. But as things presently stand, I am not sure, off hand, how many of these sorts of studies have been run thus far.

Now you also mentioned (in your earlier comment) the issue of conceivability. There has been some very recent work that attempts to address this issue (albeit not exactly in the way you mention). Shaun Nichols, David Rose, and Wes Buckwalter have a paper (under review) that criticizes the work done by Eddy Nahmias on "perfect neuroprediction." Nichols et al. suggest that people have an intuitive indeterministic metaphysics that intrudes on their intuitions about the cases in a way that makes it hard for them to properly understand the deterministic features of the scenarios. So, while their responses appear to support compatibilism, this is only because they failed to adequately grasp the implications of determinism. If you email David Rose (a grad student at Rutgers) I am sure he would be happy to share the paper. I think they have hit upon a very important issue--one that has bothered me ever since Eddy and I ran our first studies back in 2004. Indeed, I am presently in the process of re-running our original studies to test for a complimentary hypothesis--namely, that deterministic scenarios are met with so-called "imaginative resistance" such that people either don't properly imagine the deterministic scenarios as described (for reasons having to do with motivated cognition). My prediction is that by coupling the same old scenarios with questions better suited for detecting and identifying intrusion effects and imaginative resistance, we will find far fewer properly compatibilist responses even with the cases usually used to support folk compatibilism. I am almost done uploading the studies and questions (along with several scales which we didn't have available back in 2004). I should start collecting data within the next week or so. Once I have something to share, I will post something here at Flickers.

OK, that's it for now. I hope that helps.

Thanks so much Thomas, that was very helpful. I may well be so bold as to write Rose or Shaun.

My thesis about the confusion of logical and metaphysical possibility arose from my experience with my 101 students over many, many years (remember we study the FW problem for the entire semester). Specifically, familiar logical possibilities (drawn from past experience--going left or right out a door on different occasion, e.g.) that are strongly epistemically available are thus held to be metaphysically accessible for future use--an inference from "was" epistemically to present/future "can" with at least dual-(if not more)ability. But that inference isn't warranted, even if psychologically it is hard to abandon.

Here is my definitive list in order of value.

1. Behavioral data in real life interactions.

2. Intuitions about real life cases, well described.

3. Intuitions about hypoethical but very realistic cases, well described.

4. Behavioral data in experimental settings. (eg. third party punishment in public goods games)

Drop-off

5. Intuitions about concrete realistic cases, where features of the universe are specified BUT all terms like 'morally responsible' or 'blameworthy' are clearly described in settings where participants can ask questions and get further clarfication.

Long long drop-off

5. Intuitions about hypothetical non-realistic cases.

6. Intuitions about general principles in deterministic universes, however they're described.

7. The free will and determinism scales. (Fuck those scales.)

I hope, Thomas, that you will also test for another kind of 'imaginative resistance'--the difficulty some people have in recognizing that determinism does *not* entail bypassing. Now, some might say that this difficulty is itself evidence for intuitive libertarianism, since it suggests that people are so wedded to the idea that decision-making requires indeterminism (or non-physicalism) that they interpret determinism (or physicalism) to conflict with, or undermine, the causal impact of mental states during decision-making. I think that's the wrong interpretation of people's intuitions (or theories), though we need to do more work to figure out who's right. Instead, I think people are apt to read a description of a sufficient cause in the distant past (or at a low level, such as neural activity) to leave no room for an essential causal role at the more proximal (or higher) level, such as mental states and conscious deliberations—and more apt to do so if the descriptions prime such bypassing. But determinism (or physicalism) does not have that consequence. And folk who recognize that it doesn't tend to be compatibilists about FW/MR and determinism (or physicalism).

The bypassing mistake is exacerbated, since we lack a theory to explain how conscious mental states relate to physical (e.g., neural) states, and hence we don't have an intuitive way to understand how our mental lives make crucial causal differences while being part of the lawlike physical world. Similarly, we couldn't understand how the earth flies around the sun though we feel unmoving until Galileo's theory of inertia made sense of it. A physicalist theory of conscious mental activity will, I think, play a crucial role in weakening bypassing intuitions.

Regarding the discussion above, I'm not sure how to interpret people's responses to many of our Part 2 statements. As you know my current take on the existing results is that somewhere around 20-30% of people (at least in U.S.) have libertarian dualist intuitions, likely internalized from a religious upbringing that explicitly teaches such a theory about the mind/soul and free will. Most others are what I call 'theory-lite' meaning that, on the one hand, they think free will requires that we have (as we experience) genuine alternatives in making choices, that (most of) our choices are caused by our mental activity, and that we are, in some sense, the causes sources of our behavior, but on the other hand, they don't have a theory of, or developed ideas about, how these features of free will work or are implemented. Hence, they don't read scenarios that describe determinism or physicalism as undermining these features of free will, since such possibilities do not obviously conflict with alternatives, choice, or sourcehood (as compatibilists illuminate!). People might see a conflict when given arguments like the CA or Manipulation Arguments, but at that point we're in the thick of the philosophical debates and compatibilists will offer responses to those arguments. But pre-argument (pre-theoretically), I submit that most folk do not have libertarian or incompatibilist intuitions (or theories).

Then the question is why do so many people agree to the statements that seem to express libertarian theories or beliefs. I'm not sure. Beyond the ones who I think are committed to those theories, I speculate that many others are reading them in a theory-lite way or perhaps not interpreting them to entail incompatibilism (I know I risk raising the same worry that one might raise about people not internalizing the deterministic/physicalist theories in scenario studies). For instance, since people think free will does involve having alternatives and being essential causes of their actions, they might agree to statements that begin:
"Free will is the ability to make different choices..." or
"To have free will is to be able to cause things to happen in the world..."
And not then interpret the rest of the statements as committing them to more metaphysically loaded theories. The statement about responsibility requiring responsibility for all prior decisions is more perplexing to me. I think I would disagree with it, but I also accept tracing conditions, so find it intuitive, at least if ‘all’ is replaced with ‘some’.

I also incline towards Chandra's response regarding the relevance of responding to theories vs. statements, though I wish psychologists would use FWI rather than FAD if they are going to try to measure beliefs about free will (and determinism and dualism).

@Kip, Not all factors that influence an intuitive judgment process is part of the tacit theory associated with that judgment process. Theorists distinguish the theory that underwrites competence from all the factors that influence performance. Memory limits, intoxication, drowsiness, font size -- these all affect performance. They aren't part of the one's syntactic competence. The competence/performance distinction isn't always easy to draw. But it gets drawn routinely by cognitive scientists, so there is no special reason to be skeptical about it (I bring up this last point because some Xphi folks are surprisingly strong skeptics that the distinction can be drawn).

Kip raises a crucial issue with his point 4 - and then gets the answer exactly backwards! Definitions typically aren't highly accessible to consciousness. I think cognitive science research has made this terrifically clear. But if you don't believe me, try asking the man on the Clapham Omnibus for a definition of "chair". A sizable majority of the time, you'll get something that excludes beanbag chairs, Art Deco chairs, or various other clear cases of chairs; or includes beds, swingsets, and other things people sit on. Time and again, the human ability to recognize instances far outstrips our ability to define the category. To use Kahneman's terms, System 1 can handle many tasks that System 2 need never ponder.

Math is exceptional - sometimes. I swear I often do math intuitively, even while having forgotten the relevant equation.

Tom,
I think people's responses to FWI items *are* evidence - highly defeasible evidence. Parodying item (8) of Part II,

(8') To be human, one must also have human parents and grandparents and so on, that led up to the present generation.

Or more concisely,
(8'') To be human, one must also have human parents.

I bet you could get an overwhelming majority of the folk to agree to (8''), even among those who subscribe to evolution. That is, until you point out the contradiction. Plausible sounding principles are cheap.

I stole the (8'') analogy from Dennett. It's a thing of beauty - Stephen Colbert, eat your heart out - so I just had to steal it. I deserve everything I steal. ("You deserve everything you steal" is a line from a great children's book, http://www.amazon.com/The-Curse-Cobweb-Queen-Adventure/dp/0679838783 )

Eddy, you say that "The bypassing mistake is exacerbated, since we lack a theory to explain how conscious mental states relate to physical (e.g., neural) states, and hence we don't have an intuitive way to understand how our mental lives make crucial causal differences while being part of the lawlike physical world."

On the hypothesis that the majority of folk aren't dualists, then they should (or could easily be led to) understand that the causal differences our mental lives make can't be anything over and above the causal differences our brain processes make. Non-dualists already know that mental states don't add to what the brain already accomplishes in behavior control. So I don't see how having a final physicalist theory of conscious mental activity is necessary for weakening bypassing intuitions (although it likely will help cure people of any substance dualism or libertarianism they might harbor).

Hi Tom, as a non-dualist, I reject the claim that "the causal differences our mental lives make can't be anything over and above the causal differences our brain processes make," at least on some ways of understanding that claim. I am more inclined to accept: "The causal differences our mental processes make are made in virtue of their being token-identical with physical states (typically mostly or entirely extremely complex neural processes), in accord with the laws of nature." But notice that both of our formulations are pretty complicated, not very folksy. In any case, the fact that most people do *not* believe that (conscious) mental states (CMS) are merely brain states does *not* mean that most people believe that CMS are (or must be) *non-physical* states (whatever the hell such states are supposed to be!). I think most people are theory-lite about the mind-body relation (even many who talk in terms of 'souls'), though that might mean they have some sort of functionalist theory of mind, at least implicitly.

In any case, because we do not yet have a theory that makes sense of *how* CMS are related to physical states, when people are told that "the brain does it all," it's easy to interpret that to mean that CMS do nothing (or to be a bit baffled in the way a pre-Galilean would be by the Copernican claim that we're flying around the sun). Similarly, it's easy to interpret the (bewildering) claim that the millions-light-year-wide light cone of events in the distant past are (or the Big Bang is) a sufficient cause for (or a difference-making cause for!) my current decision as meaning that my CMS don't make a causal difference to my current decision.

Like you, I predict the physicalist theory of CMS we eventually end up with will indeed cure people of any lingering dualist or libertarian intuitions they might harbor (our current promissory theories, unfilled out as they are, already cure a lot of younger people, in my experience), but such a theory will also offer a good account of how CMS make causal differences and hence will also weaken their bypassing intuitions *and* their lingering incompatibilist intuitions. Recall that there's no reason to be an incompatibilist if one doesn't have some libertarian intuitions about what free will and MR require...

Thanks Eddy. If and when the folk's theories of the mind-brain relation start to align with neuroscience, they will be less likely to imagine that CMS give us an unconditional ability to do otherwise. They will then opt either for compatibilism or FW/MR skepticism. If it's the former, I hope they end up with your retribution-lite views on punishment.

The comments to this entry are closed.

Books about Agency


3QD Prize 2014: Marcus Arvan