Look below the fold for the answer.
There have been 107 responses to yesterday's survey, but since I'm too cheap to "upgrade" my Survey Monkey account, I get to see stats on only the first 100. So don't bother voting any more.
At any rate, there was a significant majority of respondents who voted "No." Specifically, 62%, with 38% responding "Yes." I suppose I could have added a "Not sure" option, or a "degrees of confidence" meter, but I didn't.
From the few comments that came in as well on Flickers of Freedom, perhaps some important details from the prompt were missing. I threw in the "drunk" comment only as a (very minor) joke, not anything that should be thought to undermine Max's reliability at replacing intentions (which he can do easily and always successfully and by whatever means you're happy with); it was only a way to explain his being mistaken over the identification of the relevant intention.
Here was the point: manipulation cases work on our intuitions typically by having some external manipulator replace one intention with a very different sort of intention. I was wondering if people thought, then, that it was the manipulation itself that undermined attributability, rather than the content of that intention. What was especially interesting about my case, I thought, was that Max was guided by precisely the same sort of considerations by which I was guided, namely, the reasons there were in favor of punching (although this might not have come out clearly enough in the prompt). If what guided me in the formation of some intention also guided Max in the intention he implants in me, and those intentions have identical content, then why wouldn't the subsequent intention be attributable to me? Apparently, though, many people think otherwise: the manipulation alone is sufficient to undermine attributability. I guess I am a bit surprised by this result, although again, it may also be a product of insufficient clarity in the prompt.
Thanks to all who participated.
Hi Dave,
I wish that you had included your survey question, since whenever I click on the survey link now, I just get a message thanking me for taking the survey. But I’ll go on my faulty memory.
In any case, you write: “what guided me in the formation of [my original and subsequently replaced] intention also guided Max in the intention he implants in me, and those intentions have identical content.” You also say, “Max was guided by precisely the same sort of considerations by which I was guided, namely, the reasons there were in favor of punching.”
I think of reasons as facts that count in favor of some response. But do you think that Max and I are guided by the facts? It seems to me that I was guided both by what I perceived the facts to be (which can, of course, differ from what the facts are) and by my judgments about whether those apparent facts count in favor of my punching. Max, by contrast, was guided both by what he perceived the facts to be and by his judgments about whether those apparent facts count in favor of his replacing my intention. Indeed, don’t you stipulate that the only reason Max manipulates me is that he falsely believes that I’m forming an intention that I’m not in fact intending. If that’s right, you’re mistaken in your claim that what guided me in the formation of my original intention (viz., my beliefs) was the same as what guided Max in his decision to replace my intention with one of his own creation. For if that right, then Max is guided by something that I’m not guided by — namely, the false belief about what intention I’m forming.
So it seems to me that they weren’t guided by the same thing. Moreover, it seems that what’s being guided is different in each case. In my case, what’s being guided is my formation of either an intention to punch or an intention to walk away. By contrast, what’s being guided in Max’s case is the formation of either an intention to manipulate or not to manipulate. What guides me in my intention formation is what I take to be considerations in favor of punching and walking away. And what guides Max in his intention formation is what he believes my intentions are and what he takes to be considerations in favor or against replacing my intentions.
By the way, I voted that “no.” It seems to me that because the implanted intention is not responsive to my judgments about reasons to punch or not but is instead responsive to Max’s overt acts of manipulation which in turn are responsive to his (not my) judgments about reasons to manipulate or not, the implanted intention cannot be attributed to me. Also, I think that the prompt was clear, and that I just disagree with you on whether it is correct to say that Max and I were guided by the same things.
Sorry, here’s the prompt:
Suppose we are in a bar and I intend to punch you for insulting my sister. My intention is formed in light of my consideration of the good reasons there are in favor of punching you. Suppose Max, the powerful demon/neurologist able to manipulate my intentions, is monitoring me, but he takes himself to be my guardian angel, ready to replace my intentions with ones based on what he views as good reasons. He is also a little drunk, so he mistakes my intention to punch for an intention to hug. “But this is wrong,” he thinks, “as the insult calls for a punch, not a hug.” He thus eliminates my actual intention to punch (what he thinks is an intention to hug) and replaces it with an exactly similar intention to punch. I then act on it.
Is the implanted intention attributable to me?
While Max’s decision is indeed guided by his judgments about reasons (which are different from mine), the intention he implants is determined in the same way my original intention was formed, namely, by judgments about what intention I ought to have in light of the reasons there are. So it still seems to me that the content of the intention is exactly similar (to punch), and the considerations going into the formation of that intention are exactly similar.
Hi Dave,
What is a consideration? Is it a fact or is it a belief? If it is a fact, then what is it for a fact to “go into the formation of an intention”? If it is a belief, do you think that beliefs held by different subjects are relevantly similar so long as they have the same propositional content? If so, why think that this is what’s relevant when judging attributibility?
Given that it was a survey mining for conceptual intuitions about attributability, I don’t have to answer your badgering questions, Doug.
Seriously, I suppose the relevant considerations here are facts. Put it this way, though. All I want is to intend to act on the best reasons (I don’t know how to put this in a more precise way, I’m afraid). All Max wants is for me to intend to act on the best reasons. In this case the best reasons counsel punching. Both Max and I clearly and distinctly perceive these to be the best reasons. I form the intention to punch based on my judgment of what my intention to act on the best reasons would consist in. Max mistakenly thinks I have formed a different intention, so he replaces the one I formed with an exactly similar intention to punch, one that he bases on his judgment of what my intention to act on the best reasons would consist in.
Put this way, do you still think there’s a difference other than the bare fact of manipulation? And if not, isn’t what happened sufficient for attributability of the intention to me?
I think that some attitude can rightly be attributed to me only if its formation was guided by (and/or was responsive to) my judgments about reasons for and against having that attitude. And, for it to be guided by (and/or responsive to) my judgments about reasons for and against having that attitude, it would have to be that whether or not I form that attitude depends on what my judgments are. So I don’t see how my intentions could be guided by the facts as opposed to my judgments. More importantly, though, it seems to me that, as you’ve described the case, the intention that I’m given by Max was sensitive not to my judgments but to Max’s judgments. Or maybe I’m misunderstanding the case.
Is it true that Max would have given me the intention to kiss you if he had judged that your insult called for a kiss from me and that he would have given me this intention so long as he so judged and, thus, irrespective of whether or not I had judged that your insult calls for a kiss from me? That’s how I’d been understanding the prompt. And if this understanding is right, then I don’t see how you can rightly claim that “what guided [emphasis added] me in the formation of [my original and subsequently replaced] intention also guided Max in the intention he implants in me.” What guided Max was his judgments about whether a punch was called for, whereas what guided me in the formation of my original intention was my judgment that a punch was called. Max and I may both have judged that I ought to punch you and for the same perceived reasons. And consequently Max decided to manipulate me so as to give me a new intention to punch you, but that doesn’t mean that this new intention is sensitive my judgments about reasons in the way that my old intention was sensitive to my judgments about reasons.
So there still seems to me to be a difference other than the bare fact of manipulation, for Max’s manipulations are sensitive to his judgments, not to my judgments.
Consider a different case: Max is caused by my judgments to give me an intention to do whatever I judge there to be decisive reason to do. Thus, even if he thinks that it would be best for me to kiss you, I form the intention to do whatever I think it would be best for me to do. In this case, what intention I form is sensitive to my judgments in a way that it isn’t in your case. Right? Or am I misunderstanding the case.
Hello,
I would like to ask if the demon is powerful enough to be able to share my perspective. If I form the intention to punch you on the basis of some reasons, I belive those reasons will include the belief/fact that she is MY sister. The demon can perhaps exactly duplicate my intention, but I do not think he can base the duplicate on the same reasons, as my sister is not HIS sister. His reasons might at most include something like “if she were MY sister, this would call for a punch”. Putting aside the issue whether he can really know what it is like for me to be in the relationship to my sister, I think the two reasons are different. I wonder whether the thought experiment could be rediscribed to provide exactly the same reasons on both sides. Anyway, I believe that manipulation is itself sufficient for non-attributabilty, so I voted NO.
Thanks, Doug and Radim, this is turning out to be extremely helpful after all. Would the following change in the scenario, then (drawn from Doug’s amendment), cause you to change your vote?
Suppose Max is an ensuring manipulator: he is caused by my judgments to give me an intention to do whatever I judge there to be decisive reason to do, in cases where there’s a screw-up on my part. So when I’m weak-willed, say, Max replaces my weak intention with the one based on what I’ve already judged I have a decisive reason to intend. In the current case, though, he’s wrong: he thinks I’ve formed an intention to hug, when I’ve actually formed an intention to punch. He then replaces my intention to punch with an exactly similar intention to punch. Is it now attributable to me?
It sounds like you’re still describing a case where what causes Max to manipulate my intentions is not just my judgments but Max’s judgments. After all, he manipulates me, in the case you’ve just described, because he has mistakenly judged that I’ve formed an intention to hug. Doesn’t this mean that my intention formations will be guided by his possibly false beliefs? It seems so. Thus, I still don’t think that my new intention is attributable to me given that it is guided by Max’s false judgments.
However, if you describe a case both where Max will push button A if and only if, and because, I judge that there is decisive reason to punch (and thus irrespective of any of Max’s judgments) and where a neuro-chip in my brain will activate causing me to form the intention to punch if and only if, and because, Max pushes button A, then I would hold that even if my intention were formed as a result of the neuro-chip activating that intention is still attributable to me, for that formation was still guided solely by my judgments.
Hi Dave,
If you just want to test whether bare manipulation is a factor in determining attributability, why muddy the waters by including an agent who manipulates depending of his or her possibly false beliefs? Why not just have a case in which an agent has an implanted computer chip that manipulates her only if she is weak willed. So the implanted computer chip is capable of both monitoring and manipulating her brain and is such that it causes her to form the intention to do x if and only if, and because, she judges that she has decisive reason to do x but is not going to form the intention to do x on her own (the computer can tell whether the necessary precursors to such an intention formation are present). People who has a problem with weakness of the will have these chip surgically implanted in their brain. I think that we would attribute the intentions that are formed as a result of these chips to the agents that have them implanted.
One reason to preserve Max is that I want an exactly similar intention implanted to the one I would have had otherwise. I think in the weakness of will case, some might balk at saying the non-weak-willed-implanted intention is mine, given that it’s different from the intention I would otherwise have had. My main point in constructing this example has been that I think what’s been doing work on people’s intuitions in the manipulation cases is the differing content of the intention, not the bare fact of manipulation itself.
Just change the example so that the computer always intervenes and stipulate that the agent would have formed the intention to do x even if there had been no intervention.
I was just going to point out a similar scenario to Doug´s. Suppose the causal connection from your reason assessment ability to your intention forming ability is somehow dysfunctional (could be weak will or brain damage, whatever). Suppose you only use the demon to cause the intentions that you would normally have caused yourself if you had the ability. You assess your reasons, run the reasons through the demon, just like your blood through an artificial kidney machine, and he causes you to form the intention on your reasons. I still do not think I would accept the intention is attributable to you, exactly because the causal chain goes outside your body and because it is not you who causes the intention.
Parfit dealt with a similar issue in his definition of memory in Reasons and Persons. There the issue was whether a memory that you lose after a brain damage and regain after a hypnotist implants it to you (on the basis of a detailed desription by someone whom you told about it before the loss) is your memory. His decision was that it was not, the reason being that this was not the “right kind of causal relation”. Perhaps Dave´s issue here is sufficiently similar. In both cases the causal chain goes outside your mind, in neither case the “author” of the mental state is you, thus, in neither case the mental state is attributable to you.
Just another quick comment. Perhaps the difference between the demon and Doug´s chip is that the demon is a free agent who could choose a different course of action, not a mechanical tool, a blind extension of your own causal capacities. I am less inclined to accept non-attributability in the chip case than in the demon causal loop case.
The chip case will get advocates of the extended mind thesis on the cheap, which I don’t want. But otherwise stipulating that Max operates in that way may get what I want, while also preserving the usual understanding of what the manipulator is.
Radim: do you not think, then, that the blood that comes back to you post-artificial kidney is yours? You seem to have a pretty restrictive notion of authorship.
Parfit’s comments on issues like this go back and forth. Sometimes he thinks the right kind of cause matters (where this includes a narrow set of causes), whereas other times he thinks that the right kind of cause can be any cause.
David:
I am not very happy about the kidney machine example now. It was an illustration of the causal loop, rather than an exact analogy for the intention case. However, it is not the blood, but the cleanness of the blood that is at issue. I have not caused the state of my blood being clean. I cannot be accountable for it. Likewise, I have not caused the state of me intending to punch the bad guy.
Let me try to put it another way. The scenario at which we arrived is a case in which an external agent interprets your reasons for action as a set of instruction for the formation of an intention, which he then creates in your brain. Its like you outsource the intention forming process, because you cannot do that yourself, or perhaps, you “have your intentions formed” for you. If this is the correct description of the case, the question then is:
If
– you provide the instructions
– you cannot carry them out yourself
– an external agent carries them out for you
– he carries them out exactly as you would have done, if you could,
– Is the output attributable to you?
Unfortunately, my intuitions are not very consistent here. First, take the negative case in which the agent misinterprets the instructions and forms and intention in you to kill the bad guy. That intention, in my view, is not attributable to you. If that is so, why should the correct intention be attributable to you? Why should attributability depend on such contingent issue as the demon´s ability to interpret what you would have intended yourself? Wouldn´t it be better to say that intentions caused by other agents, however similar to intentions that I would form myself, if I could, are never attributable to me?
On the other hand, thinking about some very different cases, we often attribute outcomes to people in cases in which they do not cause the outcomes directly, but have them caused instead. For instance, architects provide builders with detailed instructions for the construciton of buildings. They never cause the buildings to exist themselves. Still, we tend to attribute the buildings to the architects, not the construction companies. Reportedly, some of Da Vinci´s paintings were painted by his pupils according to his instructions. Still, we attribute them to Da Vinci.
So maybe the inconsistency in my intuitions is in tune with Parfit´s oscillation between the right cause and any cause.
“Why should attributability depend on such contingent issue as the demon´s ability to interpret what you would have intended yourself?”
Because it’s the consistency of the content of the intention with my overall motivational character that matters w/r/t attributability, I’m thinking.
There is perhaps another issue that might be relevant and has not been pointed out and that is the autorization of the manipulator. If you have a chip implanted in your brain to do your intending for you, I guess it is a form o aurhorization of the process. But, as you write in your first scenario, “the demon takes himself to be my guardian angel”. Is this something I have authorized? If not, I think even if the content of the intention is consistent with my overall motivational character, I could defend myself that the intention is not attributable to me.
This raises the question, Radim, as to what authorization consists in. It can’t be the case that only those psychic elements I’ve consciously endorsed are mine, say. Despite all the focus in the literature on identification being “active,” I think there are plenty of cases in which I simply “find myself” with various attitudes that I recognize to be mine, independently of any explicit, active authorization on my part. So I think authorization can be passive: as long as there’s the right structural relationship to my cares and commitments, that’s good enough (and that should explain why I think the manipulator in my case doesn’t undermine attributability).
Hello David,
maybe we should clarify what is meant by attributability, then. I do not know what plans you have with that concept in your theory, in what definitions you are going to use it, etc. But do you think that attributability entails responsibility? If that is so, I would still insist that the above intention cannot be attributable to me. Let me explain:
Maybe it is my English, but it seems to me what you are describing is authorship, which is different from authorization as I used the concept. No doubt I can be the author of a lot of mental states that I am not consciously aware of. But if someone is going to use those mental states plus some objective facts (by these two sets I mean the reasons for punching the bad guy), transform them WITHOUT MY PERMISSION, and implant the output in my brain in the form of an intention, then even if the intention is exactly like the one I would have formed myself, if I could, I wouldn´t accept responsibility for it. If attributability entails responsibility, I couldn´t accept attributability, either. Maybe an illustration to support my point. Suppose I am the president and I often like to make speeches to the public. Now suppose that there is a person who works in my office and has access to all the relevant data that I have access to, so he knows exactly what I am going to say in my next speech. Now suppose further that this person has absolutely no authorization from me to be my spokesperson. Still, he goes out and makes the speech to the public, saying exactly the same things that I would have said, if I made the speech myself. I think it would be legitimate for me to refuse responsibility for any consequences this might have, because this person did not have my permission. Again, if attributability entails responsibility, the speech (or the content of the speech) cannot be attributable to me.
I guess there are two things that I mind about the manipulator case: he does not have my permission and he is a free agent. This makes the chip case a very different story – the chip is an automaton and assuming I freely underwent the operation, I sort of gave permission for the machine to form my intentions.
Radim: Part of what I’m doing is trolling for conceptual intuitions on “attributability.” I’m thus trying not to beg any questions about its nature here. But since asked, I think the relation between attributability and responsibility is very complicated, in ways I won’t go into.
For now, let me just say that you’re putting an awful lot of weight on the requirement of explicit permission for any attitude to speak for me (to be attributable to me). This strikes me as too strong. I seem to have all sorts of attitudes that arise without my permission that I nevertheless consider mine. Perhaps you are thinking about the expression of attitudes, but that’s something else entirely. Or perhaps you’re thinking of the phenomenon of taking responsibility for an attitude, but again, that’s something different.
David, when you say “I seem to have all sorts of attitudes that arise without my permission that I nevertheless consider mine”, I do not have a problem with that, because such attitudes have the normal cause. It is me who causes (consciously or unconsciously)these attitudes. Where I require authorization is all the cases in which we do not have the normal cause and the cause of my attitudes involves another free agent (who may not even be a reliable cause). Also, I do not require an explicit authorization, and I do not require the authorizaton of each intention that the agent causes in me. Rather, what I want is the authorisation of the whole channel.
So, if I am to mine my intuitions about what I believe attributability to be, then it has to involve things that either I produce (consciously or not) or are produced by channels that I at least endorse. If what I get is just an exact duplicate of an intention that I would normally form myself, arising from the same reasons, but produced by an unauthorized free agent who is not me, I “intuit” that it is too weak for attributability.
But maybe it is just that I cannot divorce attributability from responsibility or that I just don´t get the concept of attributability at all. Anyway, it has been interesting stuff for thought. Thank you.