Skip to content
 

The free cup and the extra dollar: A speculation in philosophy

The following is an essay into a topic I know next to nothing about.

As part of our endless discussion of Dilbert and Charlie Sheen, commenter Fraac linked to a blog by philosopher Edouard Machery, who tells a fascinating story:

How do we think about the intentional nature of actions? And how do people with an impaired mindreading capacity think about it?

Consider the following probes:

The Free-Cup Case

Joe was feeling quite dehydrated, so he stopped by the local smoothie shop to buy the largest sized drink available. Before ordering, the cashier told him that if he bought a Mega-Sized Smoothie he would get it in a special commemorative cup. Joe replied, ‘I don’t care about a commemorative cup, I just want the biggest smoothie you have.’ Sure enough, Joe received the Mega-Sized Smoothie in a commemorative cup. Did Joe intentionally obtain the commemorative cup?

The Extra-Dollar Case

Joe was feeling quite dehydrated, so he stopped by the local smoothie shop to buy the largest sized drink available. Before ordering, the cashier told him that the Mega-Sized Smoothies were now one dollar more than they used to be. Joe replied, ‘I don’t care if I have to pay one dollar more, I just want the biggest smoothie you have.’ Sure enough, Joe received the Mega-Sized Smoothie and paid one dollar more for it. Did Joe intentionally pay one dollar more?

You surely think that paying an extra dollar was intentional, while getting the commemorative cup was not. [Indeed, I do–AG.] So do most people (Machery, 2008).

But Tiziana Zalla and I [Machery] have found that if you had Asperger Syndrome, a mild form of autism, your judgments would be very different: You would judge that paying an extra-dollar was not intentional, just like getting the commemorative cup.

I’m not particularly interested in the Asperger’s angle (except for the linguistic oddity that most people call it Asperger’s but in the medical world it’s called Asperger; compare, for example, the headline of the linked blog to its text), but I am fascinated by the above experiment. Even after reading the description, it seems to me perfectly natural to think of the free cup as unintentional and the extra dollar as intentional. But I also agree with the implicit point that, in a deeper sense, the choice to pay the extra dollar isn’t really more intentional than the choice to take the cup. It just feels that way.

To engage in a bit of introspective reasoning (as is traditional in in the “heuristics and biases” field), I’d say the free cup just happened whereas in the second scenario Joe had to decide to pay the dollar.

But that’s not really it. The passive/active division correctly demarcates the free cup and extra dollar examples, but Machery presents other examples where both scenarios are passive, or where both scenarios are active, and you can get perceived intentionality or lack of intentionality in either case. (Just as we learned from classical decision theory and the First Law of Robotics, to not decide is itself a decision.)

Machery’s explanation (which I don’t buy)

So what is going on? In his 2008 article, “The Folk Concept of Intentional Action: Philosophical and Experimental Issues,” Machery proposes what he calls the trade-off hypothesis. I’ll state this explanation and then explain why I’m skeptical (of Machery’s theoretical explanation, not of his experiments, which I think are great).

Machery argues that (most) people see intentionality in settings with a “trade-off between a cost and a benefit.” In the free-cup example there is no tradeoff (Joe gets the cup and his extra-large smoothie) hence no feeling of intentionality, whereas in the extra-dollar case there is a tradeoff (Joe gets his drink but has to pay more) hence it does feel intentional.

The trade-off hypothesis works for this example but I don’t think it’s the key to intentionality.

Here’s why I say this. Consider the following two (hypothetical) new probes:

The No-Free-Cup Case

Joe was feeling quite dehydrated, so he stopped by the local smoothie shop to buy the largest sized drink available. Before ordering, the cashier told him that if he bought a Mega-Sized Smoothie he would no longer get it in a special commemorative cup. Joe replied, ‘I don’t care about a commemorative cup, I just want the biggest smoothie you have.’ Sure enough, Joe received the Mega-Sized Smoothie in a regular cup. Did Joe intentionally obtain the regular cup?

The One-Less-Dollar Case

Joe was feeling quite dehydrated, so he stopped by the local smoothie shop to buy the largest sized drink available. Before ordering, the cashier told him that the Mega-Sized Smoothies were now one dollar less than they used to be. Joe replied, ‘I don’t care if I have to pay one dollar more, I just want the biggest smoothie you have.’ Sure enough, Joe received the Mega-Sized Smoothie and paid one dollar less for it. Did Joe intentionally pay one dollar less?

How would you answer these? I’d say: No intentionality for the no-free-cup, and yes on intentionality for the one-less-dollar. I’m not claiming these are the correct answers (whatever that means), but I think that’s how I would answer them.

But if I’m right–in the sense that most people would answer these the way I would–then this is a counterexample to Machery’s hypothesis.

To me, the key is that the cup is incidental but the cost of the smoothie is central. Or maybe it’s that the cashier is giving Joe the cup whereas it’s Joe himself who’s paying. I’m not sure. But I think there’s something going on here.

57 Comments

  1. Paul says:

    Subjectively I'm sure I don't have Asperger's, but not only did I not see intentionality in either case, I'm having a hard time even understanding where everyone else is apparently coming from. Joe had a strong physical desire (thirst), an initial goal (get the largest drink) and was not deterred by small incidental details (a commemorative cup, a single dollar). The only connection I can see is that the cup was a bonus – if he wanted the largest drink before, there's no reason not to get it now. Adding a cost reduces the marginal value, and in some circumstance could have changed his mind. But with that view, I'd have expected switched responses to your followup question. No commemorative cup? Maybe he really wanted that cup. Versus a dollar cheaper? If he already wanted the largest, of course he'll still get it.

    Does anybody have an explanation for why they see this intentionality breakdown, because frankly I'm a little baffled.

  2. Jen says:

    Being one of the people who would say both the free-cup and one-extra-dollar scenario are unintentional, I looked around and found this answer, which explains the thought process behind my answer quite well.

  3. Phil says:

    This whole issue perplexes me. I feel like I am missing out on something, like I've just discovered that I'm colorblind or that there's a particular sound that everybody else can hear but I can't. I just can't figure out what you people — Machery, fraac, Andrew, or any of the people who have commented on this issue in previous posts — are even talking about!

    Joe "intended" to get a smoothie. In every case above, he got his smoothie. Whether he paid an extra dollar, whether he got a special cup, those are things he was more or less indifferent to. In the case of the cup, he was presumably almost completely indifferent: OK, it might irritate him to have to dispose of a non-recyclable cup, but it's probably not a big deal. Paying the extra dollar, well, nobody likes to pay an "extra" dollar, but a dollar is not very much money so he was presumably almost indifferent to that, too. So this question seems to me like it's asking about my definition of "intend" rather than than finding out about the deep workings of my mind.

    Almost any action has a long, long chain of consequences. Some of them are key to the decision, some of them are known but don't really come into play, and some of them aren't known at all. In my parlance, things in the first group are "intentional", things in the other groups aren't. But there aren't sharp boundaries between them, so there's not a sharp boundary on the word "intent."

    I'm obviously missing something really big, because so many people find this question interesting, puzzling, or paradoxical. I just don't get it.

  4. Mgoff says:

    I read the two cases and see how they are essentially parallel. My response was to think/feel that any intention (or lack thereof) must be the same. It may be that since the commemorative cup case came first and I responded to that with "did not intend", I felt compelled to respond to the extra dollar case in the same fashion. It's hard to say now, but I may have answered the opposite had the extra dollar case been first.

    It seems to me that I have a strong feeling that the response to the cases needs to be consistent. I'm as comfortable with the idea that Joe did intend to get the cup and spend the extra dollar as I am with my original thought that it was not intentional in either case. It doesn't sit well for me to conclude that he to intended one but not the other.

  5. ahuri says:

    I don't get it either. It seems to me that the most important bit here is Joe's thirst. As good old Victor (Hugo) wrote : necessity knows no law, or cup for the present matter.
    Maybe I have Asperger syndrome then (in addition to being colorblind).

  6. Markk says:

    Yes I am with Phil. Joe didn't care about the extra dollar or cup. He says so right in the story. Doesn't "don't care" mean have no intentionality? Of course as Phil said it is a fuzzy boundary thing, but it seems like there is some underlying Freudian thing being posited. Joe really does care about some things subconsciously even when he says he doesn't. Isn't that what you are implying with the dichotomy between the scenarios?

  7. John says:

    Getting the free cup is a side effect. Paying the extra (or less) dollar is the *cause* of the getting the smoothie. Causes are generally intentional while side effects are not.

  8. Steve says:

    Maybe it could have something to do with Joe not being able to say no to paying the extra dollar, but he could in theory say no I want a normal cup?

  9. Bill Jefferys says:

    The customer in each case made it very clear that his intent was to get the biggest smoothie. In every case he made it clear that ancillary events (kind of cup, price) were irrelevant. I do not see how any of these examples can be considered otherwise…he didn't intend the ancillary event, over which he had no control, all he wanted was the biggest smoothie.

  10. John V. says:

    I would say in response to your final scenarios that the no-free-cup was intentional but the dollar less was not intentional. I do feel that the commemorative cup was not intentional but the extra dollar was, like most people evidently.

    I guess I agree with the hypothesis that intentionality is tied to a sense of tradeoff.

    Here's another thought experiment.

    Suppose Joe buys a smoothie. He gets his smoothie, pays for it, and then the cashier hands him a commemotative cup or an extra dollar in change because of the current promotion (a bonus cup or a dollar off) of which Joe wasn't aware. Would Joe reconsider his purchase? I honestly can't imagine anyone who would.

    Now suppose Joe buys a smoothie, but as he is paying the cashier informs him that the advertized promotion (a bonus cup or a dollar off) is no longer available. Would Joe reconsider his purchase? I can easily imagine some people would.

    Or consider this hypothetical test:

    In which of the situations — yes bonus cup, no bonus cup, cost a dollar more, or cost a dollar less — would the cashier feel obliged to be sure the customer was aware of the situation before purchase?

    I think that the situations warranting ensuring full disclosure are also the ones which people feel involve intention.

  11. fraac says:

    I think you're exactly right about the flaw in the trade-off hypothesis. But I also think the Aspergers angle is useful to see what's happening in relief.

    There were also the 'help' and 'harm' cases. A CEO makes decisions to maximise profits which incidentally either help or harm the environment. Most people say harming was intentional but helping wasn't. That fits the trade-off hypothesis.

    Can you find a model that fits all of them?

  12. Bob Carpenter says:

    OK, maybe I should call a neurologist, because I'm with Phil. This is about the semantics of the term "intend", the debate of which has a long and illustrious history in philosophy, not to mention American jurisprudence (scroll down to the West legal dictionary section to get some idea of how the game of definitions goes).

    By buying groceries from Whole Foods, do I intend to put the mom and pop grocer out of business? What if I have no clue about retail economics? If I pay my taxes in the U.S., do I intend for our army to drop bombs on the rest of the world? The usual debate focuses on unintended or tangential consequences and the agent's awareness of them, especially when consequences are diffuse and have mixed preferences.

    The smoothie example at hand confuses intent with preference. In both cases, the purchaser is fully aware of what they're paying for. The purchaser might have preferred something else, but clearly knew the relevant details of what was being purchased. There aren't any far out or unintended consequences, unless that $1 or cup makes a difference elsewhere like the flapping wings of a butterfly on another continent…

  13. Frank says:

    I'm with Phil and the Asperger's folks: the guy wanted a smoothie enough that locally altering the cost-benefit analysis didn't affect his decision. I don't know why "intentionality" is a useful concept here.

    As a decision-maker, how many incidental factors can you condition your choice on? You'll practice bounded rationality — ignoring enough stuff to make your decision problem similar to others you've seen before.

  14. Anonymous says:

    I actually got hung immediately on "intentional", and decided that in neither case was the extra dollar / free cup "intentional". Joe's intention was to drink a big smoothie, period. Everything else is "incidental", and to me (I do not have Asperger's) neither "intentional" nor "unintentional" — simply beside the point. At best, I decided it was a poorly posed question since "intentional" was not defined to my satisfaction, and I did this before I got to the punchline.

    This is related to an important issue in law, when a law states that something is only illegal if it is done "knowingly". Shades of "knowingly", "intentionally", "purposefully" permeate legal issues, where state of mind matters. If it is illegal to purchase the commemorative cup, but Joe doesn't know (or care) about that, has he "knowingly" purchased an illegal cup?

  15. Andrew,

    I think I want to say "no" on both of your examples — which would be in line with Machery's explanation, right? In neither case is Joe paying a cost in order to satisfy his goal, and that really does seem salient to me. (I might be biased: Edouard is on my dissertation committee, and I've written a couple of papers with him.)

    I wonder how you think about some of the other cases in the literature. For example, how do you see Knobe's CEO example:

    (Roughly without looking it up)

    [Harm] A CEO for a large corporation is presented with a new money-making scheme and he is told that the scheme will harm the environment. The CEO says, "I don't care about the environment, I just want to make money." The scheme is implemented, and the environment is harmed. Did the CEO harm the environment intentionally?

    [Help] A CEO for a large corporation is presented with a new money-making scheme and he is told that the scheme will help the environment. The CEO says, "I don't care about the environment, I just want to make money." The scheme is implemented, and the environment is helped. Did the CEO help the environment intentionally?

    If I'm remembering correctly, people say "yes" on the first and "no" on the second. But the CEO help case looks a lot like your second example — where Joe pays a dollar less in order to get his smoothie.

    I'll point out your post to Edouard and also to Josh Knobe — maybe they'll weigh in.

  16. Kevin Dick says:

    I'm with Bob. I would say that the whole concept of intent is fundamentally irrelevant to the outside observer.

    Maybe that's because my economics training makes me care primarily about revealed preference. All we can say is that is that the revealed preferences in these examples are consistent with the stated preference for the largest possible beverage. That getting an extra cup or paying an extra dollar did not affect his choice _at_this_margin_.

    Now, if the largest beverage were $10, $100, or $1000 more, I think we'd find that Joe did indeed care about price _at some margin_. Similarly, if we found that the drink came with 10, 100, or 1000 commemorative cups, he did indeed care about extra cups _at some margin_.

  17. lemmy caution says:

    "The customer in each case made it very clear that his intent was to get the biggest smoothie. In every case he made it clear that ancillary events (kind of cup, price) were irrelevant. I do not see how any of these examples can be considered otherwise…he didn't intend the ancillary event, over which he had no control, all he wanted was the biggest smoothie."

    You can intentionally pay an extra dollar, or you can unintentionally pay an extra dollar. It makes a difference.

    If you get one of those "free credit reports" they sign you up for a $20 a month service. Did you intend to pay the $20 a month? Yes- if you knew you were going to be charged that much and you went ahead any way. No- if you didn't read the fine print.

  18. Jerzy says:

    I'm perplexed too. My very first instinct for the extra-dollar case was that it's intentional, but after thinking about it for a second the intentionality stopped being so clear.

    Bob, at least this doesn't suddenly diagnose us with Asperger syndrome: in Machery's study, a full 1/3rd of non-Asperger respondents *didn't* think paying the extra dollar was intentional, and 1/5th of Asperger respondents *did* think it's intentional, so the question's not a perfect classifier by any means. Good to know!

    It might be good to separate out the two questions about each case: (1) did Joe *actually* act intentionally?, and (2) would most people *say* that Joe acted intentionally?
    (1) can be taken as an unanswerable question about a hypothetical situation, or as a question about the semantics and definition of the word "intend." It might be a fun topic to argue about with friends but I'm not too interested.
    (2) can be tested out experimentally, and here I agree that if you asked respondents Andrew's questions too, it could help illuminate how people tend to think about intentionality.

    LARGE trade-offs (serious costs or major benefits) obviously come into play in intentional decision-making. But in the smoothie examples, the "costs" and "benefits" are so small that Joe explicitly says he's indifferent to the cup and the dollar. To my mind, that means intentionality doesn't come into play at all…
    But if a lot of people tend to assume intentionality in the cost cases and not in the benefit cases, that would support the hypothesis that "people tend to assume intentionality if they think a trade-off is being made, even when the actor is actually indifferent to the trade-off." That would be interesting to know. On the other hand, if people tend to answer like Andrew does, the trade-off hypothesis can be debunked.
    So even though I personally don't think intentionality comes into play in any of the Joe examples, this could be a worthwhile study to run.

  19. Dubi says:

    I'm with Phil here. Let's try out this hypothetical:

    Joe was feeling quite dehydrated, so he stopped by the local smoothie shop to buy the largest sized drink available. Before ordering, the cashier tried telling him that the Mega-Sized Smoothies were now one dollar more than they used to be. But before he even began Joe cried, 'I don't care how much it costs, I just want the biggest smoothie you have!' He then paid by debit without even looking at the amount to be charged. Sure enough, Joe received the Mega-Sized Smoothie and paid one dollar more for it. Did Joe intentionally pay one dollar less?

    Clearly, he didn't. I don't understand why merely TELLING him of the price change changes his intentions, when clearly he never stopped to consider the implications. If Joe had stopped and thought it over, then decided to buy that cup anyway, we could talk about intention. But if no thought was put into it, then telling him shouldn't matter.

    Or maybe we're all just mildly autistic – that's not completely out of the question.

    I think the explanations lies in thinking about it. We ARE thinking about the added cost, so we reflect that onto Joe, and assume he thought about it too. But the story clearly indicates he did NOT think about it, and therefore did not "intend" to do anything but buy the drink.

  20. fraac says:

    Remember that AS people almost universally say the extra dollar was unintentional. They have no lack of facility with definitions and legalese, so what's going on with non-autistics must be deeper.

  21. wcw says:

    This example is awful. A tschotschke is valueless, while a dollar can be exchanged for goods and services. As a result, it boils down to, a, customer buys a smoothie and throws away a tschotschke, b, customer buys a smoothie and pays a dollar more. Did customer intend to buy a smoothie? Yes.

  22. Dane says:

    This is just one particular case of what has been called the Knobe Effect. Importantly, it doesn't seem to rely on anything inherently weird about intentionality or money (or cups, for that matter). There are a couple more cases discussed here: http://www.edge.org/3rd_culture/morality10/morali

    Knobe seems to think that the crucial factor at play is something to do with morality, but I disagree. I think these scenarios are cleverly exploiting the subtle polysemy of words like "intend" and "happy".

    "Intent" can refer to one's ultimate aim, such as Joe wanting the largest drink irrespective of anything else. However, "intent" can also refer to any voluntary action knowingly undertaken by a person, such as Joe consenting to paying a different amount of money, or accepting or forgoing a commemorative cup. The only thing that changes between the various scenarios is which definition of "intent" the reader chooses to employ when interpreting the question. This isn't an arbitrary decision–a reader uses context to guess at which meaning is more likely to be the one the positer is using.

    In the free cup case, the reader assumes the salient question is whether or not Joe actually wanted the promotional cup (was this his goal?). Promotional cups, being marketing ploys, are created to attract customers and/or encourage people to buy larger size drinks. Our special knowledge of Joe's situation informs us that it was not the promotion that caused Joe buy a large drink… buying a large drink was in fact his goal all along.

    In the extra dollar case, it would be rather inane to ask whether Joe's goal was to spend an extra dollar, and a reader accordingly interprets the question to mean "did joe voluntarily and knowingly pay an extra dollar?"

    In the case of the CEO and the environment (see link), the choice of definition of 'intent' hinges on the questions of blameworthiness or praiseworthiness, and the definitions switch accordingly. Assigning blame is contingent upon whether the CEO voluntarily and knowingly caused harm to the environment. Deserving praise is contingent upon whether the CEO's ultimate goal was to help the environment.

    Similarly, in the case of Maria, "happy" can refer to an emotional state or it can refer to wholesomeness (as in, 'a happy home').

  23. Arthur B. says:

    A +1 here, this is really perplexing me. I don't even know what you mean by "intend", and it seems like the whole thing is a discussion of the semantic of the word.

    If you'd ask me in real life, "Did you intend to get the cup / pay an extra dollar?", my real life answer would have been "Huuh, what do you mean?"

    Now that I've read this, I might be tempted to come up with a definition of "intend" that matches some meaningful concept. Here's my try:

    Intended consequences are the predictable, desirable (to me, ex ante) consequences of my actions.

    Getting an extra dollar is not intended, because it is negative.
    Getting the commemorative cup is intended, because it is positive.

    However, when I learn that the commemorative cup is tied to the mega-smoothie, I also realize that the price of the cup has been marked into the price of the smoothie and my expectation of the price jumps by a dollar. Paying that extra dollar is unintended.

  24. Epanechnikov says:

    This is the way I see the problem:

    Before going to the local smoothie shop Joe made a decision based on his subjective expectation on the product features (and subsequently the level of utility/enjoyment he would gain by consuming the smoothie). His decision is materialized in a purpose which leads to an action. It is now probably rational to assume that the purpose is distinguished by its levels or degrees of intensity – Li. The higher the expected enjoyment the higher the intensity of purpose.

    When Joe reaches the shop his expectations are updated. If the new utility expectation differs from the old one then Joe's intensity of purpose will change (from L1 to L2=L1+ΔL). Some new purpose (ΔL) and a new intention will have been created. This will lead to a decision updating.

    This rationale is consistent with the way most of us see the problem presented here. The level of purpose remains unchanged in the Free-Cup Case while it changes* (which means that some new intention is created) in the One-More/Less-Dollar Case. I believe that the sentence "I don't care if I have to pay one dollar more" does not imply indifference but decision. It just means that his expected utility from consuming the smoothie is much higher than the smoothie's new price. Would he have wanted the smoothie if its price were now 50 or 100 dollars more than it used to be? Can't we rationally believe that the price of a product does always change our degree of willingness to consume it?

  25. Bill Jefferys says:

    His stated loss or utility doesn't care about the ancillary events. So I don't agree.

  26. Epanechnikov says:

    We however know that the surface form of a sentence is not the only reality.

    Consider the following example:

    Thief: "Your money or your life"
    Victim: "I don't care about the money. Take them all. I just want my life"

    Does that mean that the victim's welfare has not been decreased after giving his money to the thief just because he said so?

  27. fraac says:

    Can crack this case wide open if you have psychological profiles of enough non-autistics who say it was unintentional. The beauty of this stuff is it's only philosophy until you take it into the lab, then it becomes legit. :p

    Machery seems to have realised the question "Why do Aspergers people see it differently?" is key to unlocking the mystery.

    I suggest objective definitions don't matter (and don't exist anyway!) because autistics and non-autistics will have the same arguments. Context is being read differently. Why? These are smart, rational people on both sides.

    My guess is autistics are reading the context, including the metacontextual question which defines a universe where both answers make sense, and saying "How is it possible to unintentionally pay extra as the question implies? If there is a level between what you do now and your bigger plans. Okay, I see." In another context, with another person asking the question, the answer would be different.

    Most non-autistics seem to be addressing the question but not the questioner, as if their answers are already tethered to a universe. 'Theory of Mind' would presumably be this universe.

    Tbh I think a non-autistic's chance of working this out are comparable to an ant understanding a picnic, but the science cannot lie. Get more results. Need input!

  28. Phil says:

    This is just killing me. I can see that you all think there is something here that is not just non-obvious, it's fascinating and surprising…and I don't see it at all. Can somebody who finds this interesting please try to explain why this isn't just an argument about the definition of "intent"?

    By the way, lemmy, you said you were quoting me but that quote was from Bill Jeffreys.

  29. Phil,

    I'm not sure I can satisfy your request, but here is a feeble attempt. The first part of the problem is about the folk theory of intentional action. At the outset, Knobe, Machery, and others assume that there is something like an ordinary notion or concept of intentional action. There are then two questions: (1) under what conditions will people generally say that an action was intentional or that an outcome was brought about intentionally, and (2) what psychological or neurological mechanisms/processes are activated when people make intentional action judgments?

    One direct way in which the study involving Asperger subjects is interesting is that it tells us something about (2) — or at least, it might do so.

    But here is another thing that the study might be doing. Many philosophers adopt a meta-theory* (or a theory of evidence) according to which what people ordinarily say about a term or concept X provides evidence for the correct account of X. In fact, lots of philosophical discussions take the same basic form. For example, philosophers will ask, "What is knowledge?" and then they will start pushing around example cases, theories, and counter-examples in the attempt to figure out what knowledge really is.

    Such discussions presuppose both that there is some correct account and that there is going to be widespread agreement among ordinary people with respect to their judgments about particular cases. (I tend to be skeptical that there are any answers independent of what we might do with concepts like knowledge or intentional action, which is part of what makes me a Pragmatist.)

    For a few years now, Machery and others have been carrying out experiments showing that there is no widespread agreement on many philosophically interesting cases (ranging from the semantics of reference to moral judgments). More importantly, there is often disagreement across theoretically salient sub-populations, e.g. American subjects will say something different than Japanese subjects (e.g. about reference), men will say something different than women (e.g. about knowledge), high SES people will say something different than low SES people (e.g. about conventional moral norms), Asperger subjects will say something different than non-Asperger subjects (e.g. about intentional action), etc. If you don't have a commitment to there being a correct account of such things as knowledge or intentional action, then no problem. But if you do think there is a correct account — and many philosophers seem to — then you need to say why one group or another has privileged intuitions. What privileges Western, educated, rich, philosophy professors, for example? Answering "nothing" appears to present a serious challenge to one common strain of philosophical methodology.

    Does that help?

    * I have a couple of recent blog posts discussing two broad kinds of meta-theory with respect to causation that might be helpful, if you're interested.

  30. J.J. Hayes says:

    I do not think the question has been answered as to why this is not purely definitional. But lest we get bogged down in definitions about definitions it does seem that to say that there is a "correct account of intentional action" seems to beg the definitional question. It seems as if it falls into essentialist thinking, rather than merely labeling what is going on for ease of discussion. If we substitute "x" for what is being done with regard to the cup and "y" with respect to the dollar, and then ask if "x is the same as y" is true the whole process is different and the word "intention" might never be introduced and our previous ideas about intention might not enter our thinking at all. The very introduction of the word "intentional" implies an essence of intentionality which skews the argument.

  31. ceolaf says:

    I do not think that this is merely a discussion about definitions. Rather, I think it is about constructs. (Similar, but not quite the same, eh?)

    Livengood's most recent comment points to part of the issue. Some — he ascribes it to philosphers — believe that there are deeper constructs in our heads to which we make reference with words, terms or phrases. This is not merely about definitions, because definitions are words used to explain other words. But this idea (that I referring to as "constructs") is about notions of ideas that are in our heads without necessarily being exactly tied to a word or set of words. Let's say that "romantic love" is a construct that does not rely on words or a set of words to exist in our head. Let's say that "fairness" is another.

    This is a discussion about "intention" or "intentionality." What is that? Not "how would you define that?" Rather, what is the deeper construct? (Yes, definitions try to get closer to the construct, but the defintion is another tool to represent the construct, not the construct itself.)

    To me, an interesting thing is that the same construct can have different meaning to different groups. I looked at this example — even before reading the whole paper — and thought that different groups will see it differently. Lawyers have a very precise meaning for intentionality that most of them have internalized. I think that psychologists wrestle with this idea a fair amount and therefore probably have a different idea that lay people. I tend to believe that philosphers have this tendency to take apart ideas, so intentionality — like everything else — has a different feeling for them than for lay people.

    (I'm the son of a lawyer and pshrink, and have been told my philosphers that I think like a philosopher. I know that I like to go through the process that Livengood described.)

    So, the interesting quesiton here, to me, is NOT whether something was inentional or not. I don't think there is just one answer to that. Rather, I am fascinated by which groups think it one way or the other, and why. I am most fascinated by how this construct has different meaning for different groups. Is it that the only thing they have in common is the word? Is there overlap in meaning? What and why is/are the difference(s)?

    And then, in the context of this particular paper, there is another question. Given the different meanings for different groups of the construct that we refer to as intetionality, what goes on with those on austism spectrum (i.e. the most current term, I believe). Is there is difference in the meaning of the construct, or is it that rather a problem with mindreading? Unfortunately, because the authors don't acknowledge that the construct might have different meanings for different groups, they assume that the problem for those on the autism spectrum is entirely about mindreading.

    (And for those who did not read the article, 95% of people in the sample say that the cup is intentional, and 45% of them say the dollar is intentional. I am not clear how to extrapolate those 126 undegraduates to the general population, however.)

  32. Phil says:

    Jonathan L and coelaf, thank you for that attempt. I think this may be something I just can't understand.

    Joe's situation seems so clear to me, I just don't see what there is to discuss, or what seems paradoxical. Joe wanted the biggest smoothie, and didn't care if he got a cup or if he paid an extra dollar. Whether we say he "intended" to spend the extra dollar, that just feels like an argument about a word: if you consider something to be a very slightly undesirable consequence of an action, and you decide to take that action anyway, then do we say you "intended" for that consequence to happen, or not?

    Does the following story give you the same feeling of cognitive weirdness?

    I decide to get a smoothie. As I approach the counter, I see Joe walking toward the counter too. I hurry my pace so I get there first, not wanting to have to wait while the smoothier (?) makes Joe's smoothie. I definitely "intended" to get there ahead of Joe; and I definitely "intended" to reduce my wait time by doing so. But did I "intend" to make Joe wait? I was given a choice between (Phil waits, Joe doesn't), and (Joe waits, Phil doesn't), and I "intentionally" chose the latter, so clearly I must have "intended" to make Joe wait. And yet, if you were to ask me, I would say, No, I did not intend to make Joe wait, but I did intend to get in front of him!

    Is that the same sort of thing you people are talking about? If it is, then I guess I understand the discussion — how can you say you didn't intend to do something, if you did it knowingly — but I still don't see it as a paradox or puzzle. In my example above, I intend to reduce my own wait time, but I am essentially indifferent to Joe's. If a second smoothier were to show up, so that they can make Joe's smoothie at the same time mine is being made, it's not like I'd say "darn, I intended to make Joe wait for his smoothie."

    Eh, I guess I still just don't get it. Thanks for trying to explain it to me, though!

  33. anon11 says:

    "If you don't have a commitment to there being a correct account of such things as knowledge or intentional action, then no problem."

    Exactly. Intentional action and knowledge are just shorthand models for a diffuse set of emergent neurological phenomena. Have western philosophers really wasted this much time and energy because they mistook their own ill-defined model for a fundamental truth that needed to be discovered?

  34. fraac says:

    "I do not think the question has been answered as to why this is not purely definitional."

    Because of the control group with a different neurology. It's perceptional, the crucial part happens before conscious thought. All talk of definitions is red herrings as it fails to distinguish from the control group.

    Phil: smoothista.

  35. Phil, you keep saying that you don't get it. You're right, in that you answer is a legitimate view. But you leave the question as to why most other people disagree with you. Are they stupid? Do they use different definitions?

    What makes the problem interesting is that it appears to show that people have sharply different ways of thinking about the intentions of others. There is also disagreement about just what those differences are, as Andrew disagrees with Machery.

  36. Phil says:

    Roger, I don't know why most other people disagree with me. That's what I find so frustrating! Do you remember those "magic eye" figures (or something like that) that were popular about ten years ago? If you aligned your eyes right, you'd see a stereo image: a figure of something would pop out of the picture. Sometimes you'd see people looking at a bizarre colorful image, and everyone would be saying "oooh, that's really neat!" Well, I have a friend who couldn't see those; no matter how he tried to align his eyes, or follow the tricks that other people suggested, he never once saw any 3D shapes in one of these things. He got very frustrated, but he could just never see the illusion. Well, that's what I feel like: I cannot see a cognitive illusion here. I would like to see it; I get it that everyone else sees it and finds it interesting; but I simply don't see it at all. I don't know why.

    fraac: ;)

  37. Phil,

    I'm starting to be perplexed about what exactly you think you aren't getting. Is it that you don't understand that two different groups — with some known mental/neurological differences — say different things about the cases? Is it that you don't understand why the difference in judgments is interesting? Is it that you can't see why anyone would answer differently than you answer the questions?

    In short, what is it precisely that you are not seeing? (Or does it require seeing in order to say what you are missing?)

  38. Phil says:

    Jonathan, I understand that different people answer the question differently, and I understand that there is a systematic difference between how people with and without Asperger's answer the question.

    What I am missing is (1) why people without Asperger's say that, for the original story, most people say that Joe intended to spend the dollar but did not intend to get the cup, and (2) why people think the difference in answers tells us something especially interesting or unexpected.

    Note that I am one of the large minority of "normal" people (I do not have Asperger's) who think the answer for the cup is the same as for the dollar. However, unlike most people I am nearly indifferent between saying that they were both intentional, or both unintentional. It seems to me that this depends entirely on how you define "intentional." By my personal definition, I'd say that an effect is "intended" if you realize it is likely to happen, it plays a significant role in your decision-making, and you bring it about anyway. So my inclination is to say that neither getting the cup nor paying the dollar was intentional: as the story is told, neither played a significant role in the decision. However, if someone said "that's not what is meant by 'intentional', intentional just means that it's something you thought about, it doesn't have to be something you consider important," I'd say Oh, OK, well in that case both the getting the cup and spending the dollar were intentional. I still don't see a puzzle or paradox.

    For what it's worth, I would also say that, in the Machery paper, the hypothetical company that either helps or harms the environment does not "intend" to do either. I would be outraged if they don't care that they're harming the environment, but that doesn't mean I would see it as intentional.

    The two paragraphs above seem very natural to me, to the point that it's hard for me to see how someone could disagree. And yet, almost everybody disagrees! And it's clear that not only do they disagree with me, they get a funny feeling of cognitive dissonance when they think about it. I'm missing that completely.

    Well, not completely. In fact, upon reflection, it would be much easier for me to argue that the hypothetical company is "intentionally" harming the environment than that they are intentionally helping it. So I do see a bit of asymmetry that is worth thinking about. Maybe I just have the feeling to a much, much lesser degree than most people. But is that because of my worldview, or because of what I think "intended" means?

  39. fraac says:

    I really think the non-autistics saying intentional are projecting themselves into the scene and ignoring the (meta-)context. Maybe the non-autistics saying unintentional are projecting themselves into the scene and referring to their own history in a different way. Very tempted to grab some students and do some street psychology on this.

  40. Phil,

    Maybe it would be helpful to review some of the context for the study. Philosophers (and lawyers) have an interest in providing an account of intentional action. But they disagree about what the proper account is. One place where contemporary philosophical theories of intentional action disagree is on whether side-effects of an intentional action are brought about intentionally. As I said earlier, if you don't think it makes sense to talk about a proper or correct or true account of intentional action, then you have already gotten off the bus. Maybe getting off the bus there is the best thing to do, but let's ride a little longer just to see the scenery.

    Josh Knobe — a philosopher working on intentional action, among other things — suggested breaking the deadlock by looking at what ordinary people think about some cases. Knobe suggested that it matters whether the side-effect is thought to be morally good or morally bad. As he showed, in the CEO case, people are much more likely to say that the side-effect was intentional in the harm case than in the help case. Here is a nice video summarizing that initial result.

    Machery's studies — initially on plain old ordinary college kids — were designed to undermine the claim that ordinary judgments of intentionality for side effects depended on a moral judgment. Instead, he argued that it was about paying a cost. If an intentional action has a side-effect that costs the agent, then the side-effect was brought about intentionally.

    Backing up, one might wonder why we want an account of intentional action at all. For one thing, it seems to matter for judgments about moral and legal responsibility. If the CEO brought about harm to the environment intentionally, then the CEO is responsible for that harm in a way that the CEO is not responsible if the harm was not brought about intentionally. At least, a lot of people have that intuition.

    We might resolve the issue, in a sense, by simply stipulating a definition of intentional action and intentionality for side-effects, but many people think that would be unsatisfying. I am sympathetic to simply stipulating a definition, but I can also see that there are constraints. For example, I would be unhappy with a definition that stipulated that actions and their side-effects will be intentional iff the agent is wearing a chicken suit while performing the action in question. Less extravagantly, I don't want a definition that counts every side-effect of an intentional action as having been brought about intentionally.

    Finally, as I and others have already said, there is a separate issue: everyone being studied is a speaker of English and presumably has a working definition/concept of intentionality. Yet, Asperger and non-Asperger subjects have a systematic disagreement about whether the side-effects are intentional in Knobe and Machery cases. Why is that? Is it that they evaluate the moral valences differently? Is it connected to differences in so-called mind-reading abilities? (Mind-reading here just means something like an ability to ascribe mental states to others.) Is one or the other group applying the concept correctly — i.e. saying true things about intentional action — while the other is not?

    Anyway, I don't think I can do much better than that at making the puzzles here interesting or comprehensible.

  41. fraac says:

    Great summary, bonus points for the Halmosian conditional.

  42. anon11 says:

    When lawyers and philosophers claim to have obtained laws of nature, it's usually an intellectual fraud. Physical scientists do this for a living and still screw it up, so don't try it at home unless you're ready to be wrong.

    The convenient, low-dimensional measurement for neurological state is a self-report. First of all, it's a fairly crude measure that could reflect all kinds of things. Now think about how many confounders there are between people diagnosed with asperger and the control group. Anyone who thinks that this case control study design will produce a "correct" (whatever that means), natural, definition of intentionality has got to be delusional.

    So what if it's unsatisfying that philosophers and lawyers are left with is a definitional argument? We're stuck with the universe we have, and in this universe many abstractions of humankind will not turn out not to map to a simple, (legitimately) publishable truth of nature.

  43. Phil says:

    Jonathan, thank you for that careful explanation.

    I have just realized something…but let me give a digression first. I have a friend who once went to a doctor about some issue or other, and the doctor noted that my friend has weak hands. He suggested some hand exercises. My friend was in his thirties at the time, but had never realized that his hands were weaker than anyone else. In retrospect there were lots of clues, like the fact that he felt unsafe when he operated his bike brakes from the brake hoods (the way 80% of cyclists do it 80% of the time), and various other things that he noticed but had somehow never put together. He knew that he experienced these things differently from other people, but hadn't recognized that there was this simple fact that explained everything.

    I am having something of the same experience. I have just realized that I have a long history of drawing a distinction between something happening "intentionally" and having it happen as an unintended side-effect, including some cases almost exactly like the CEO with the (anti)-environmental project. I dislike distortion and propaganda, even when it's for a cause that I support, so I've often been the guy disagreeing with friends who say things like "the Tea Partiers want the children of illegal immigrants to die due to lack of health care"; I'll say "it's not true that they want the kids to die; it's just that they don't much care if they do." Sorry for the inflammatory example, it's the first real-world example I remember, but there have been plenty of others.

    Anyway, I now realize that my perception of "intent" has _always_ been different from most people's. It's like I've been using the word wrong for all these years…and not just the word, the concept (as in the above example, which does not use the word 'intent'). As I mentioned in an earlier context, I've always felt that something is intentional if it plays a substantial role in your decision-making and you decide to make it happen, but obviously that's not the way most people think about this issue! How is it I never noticed this before? I guess my thinking is close enough to other people's that I could get away with thinking about it 'my way' for all these years (decades!).

    Now all I need to do is to figure out what people _do_ mean and I hope I will be well on the way to experiencing this interesting puzzle myself. Jonathan, I think you have given me plenty to work with. I'll think about it.

  44. fraac says:

    Maybe you're autistic.

  45. Andrew Gelman says:

    Phil:

    That story explains why I'm so impressed that Scott Adams can open jars with his bare hands (to connect to an earlier thread).

  46. Phil says:

    fraac: No, I don't think so. Nobody else has ever suggested that possibility. Andrew knows me quite well, perhaps he can weigh in.

    Honestly, though, this is eye-opening. What do other people mean when they say something is "intentional"? What we might call the standard or "normal" view of it seems bizarre to me. Surely we would all agree that intent is in the mind of the actor, not the observer, but…well, surely we would not all agree to that, since that view is clearly contradicted by the examples we're discussing!

    Hoo, boy. I feel like I have to go back to second grade and re-learn this.

    Andrew: I have a friend who once held a "Festivus" celebration — it's a made-up holiday from Seinfeld — and the ritual "test of strength" was to try to open a jar that nobody in the household had been able to open.

  47. Andrew Gelman says:

    Phil:

    You are unusual in your interactions with others, but no more than I am. You have no loyalty but are very aware of the effects of your actions on others. I am the opposite. I don't think either of us could be characterized as even close to autistic.

  48. fraac says:

    "The convenient, low-dimensional measurement for neurological state is a self-report. First of all, it's a fairly crude measure that could reflect all kinds of things. Now think about how many confounders there are between people diagnosed with asperger and the control group. Anyone who thinks that this case control study design will produce a "correct" (whatever that means), natural, definition of intentionality has got to be delusional."

    I agree a lot of this won't be immediately amenable to the scientific method but I have at my disposal some of the most insightful natural psychologists on the planet. (One read Mein Kampf and said Hitler definitely had Aspergers. Many truths are inaccessible to social science). We can work out a model and then let the pros spend years harvesting results that fit it. My curiosity needs to be sated, I'm not much bothered about the official story. I'm not much interested in intentionality either; it seems epiphenomenal at a level more interesting to lawyers than… I was going to say behaviourists but I think I just mean people.

  49. luosha says:

    here's an alternative that accounts for your intuitions in both cases. but it requires that we don't take you as seriously when you talk about not caring about money as when you talk about not caring about the cup.

    whether you claim you care or not, it seems like the dollar more or dollar less will have some effect on your well-being. whereas having the commemorative cup will not. if you knowingly do something that will affect your wellbeing, we count it as intentional. if you knowingly do something that does not affect your wellbeing (in our judgement), then we don't count it as intentional.

  50. Edouard Machery says:

    Many thanks to Andrew and to all who have commented on his post. This is an extremely interesting and illuminating exchange. While I won't be able to answer to all the points made above, I want to address some of them at some length.

    Let me start with a few preliminaries.

    1. I highly recommend reading the posts by Jonathan Livengood, which describe the background of the studies I have been running accurately and clearly.

    2. As several commentators (Bob Carpenter, Phil, anonymous, etc.) note, the studies hang on what "intentional" means among lay speakers. Indeed, as explained by Jonathan, the very point of the studies reported here is to examine what meaning or what meanings ordinary speakers of English (you, me… well, you: I am French) associate with the word "intentional". We present people with vignettes and we asking lay people to judge whether an outcome is intentionally brought about; we then try to interpret lay people's answers.

    There are many reasons to be interested in this question. First, imo, it is intrinsically interesting. Second, the ordinary concept of an intentional action – viz. the meaning ordinarily associated with "intentional" – is important to folk morality (see Jonathan's point about responsibility above) and to the law. By examining this concept, we can understand better what the law or what folk morality requires for an action to count as intentional. Stipulating here won't do. Third, many philosophers think that the lay understanding of intentional action (including our intuitions about whether some particular actions in some vignettes count as intentional) constrain the philosophical theories about what makes an action really intentional. (There are many ways of conceiving of the relations between philosophical theories and lay intuitions, but this is not the place to dicuss this here.)

    3. Please keep in mind that the two vignettes are not meant to test for Asperger syndrome (see the posts by Andrew, Jerzy, Bob Carpenter, Phil, and others). Not all people with AS respond alike (the same is true of "neurotypical" individuals).

    At the same time, the differences between people with Asperger and individuals matched for SES, IQ, age. etc. is very large and striking. We are in the process of replicating these results.

    4. As a few commentators rightly note (e.g., Mgoff), it makes a difference when the two vignettes are presented together (in a within subjects design) and when, each subject seeing a single vignette, the answers of two groups are compared (in a between subjects design). The effect is stronger in the latter case than in the former case, even though, as Andrew's judgments illustrate, the effect is also present when one sees the two vignettes.

    5. It is important to keep in mind that the question is about what makes an action *intentional* – not about what *is intended* or about what the agent *wants*.

    What seems interesting is that while the agent in a sense does not want to pay one extra dollar his action is still intentional. I think that the pattern of judgments found in lay people reveals a fairly subtle sense of what makes an action intentional.

    These points having been made, I can turn to Andrew's counterexample.

    6. Andrew's pair of cases is very ingenious. I may run them, and get some data about them.

    Before discussing them, I'd like to know whether in the second case the agent should say "'I don't care if I have to pay one dollar LESS" instead of "'I don't care if I have to pay one dollar MORE" which sounds a bit weird given the context.

    In any case, my intuitions are not entirely clear, and I may share Jonathan's intuitions: No to each case. This would be consistent with the trade-off hypothesis (so my intuitions may be biased!): there is no trade off in either case, so the outcome that is concomitant to the agent's goal (getting an ordinary cup or paying a dollar less) is not intentional.

    But suppose that Andrew's intuitions are common among lay people. Then, here is what I should say.

    The no in the ordinary cup case is consistent with the trade off hypothesis: no trade off so the concomitant outcome (getting the ordinary cup) is not intentionally brought about.

    However, the yes in the one dollar less case is more problematic: There is no trade off, so the trade off hypothesis entails that the concomitant outcome (paying one dollar less) should not viewed as intentional. This is a genuine counter example indeed.

    Imo, this shows that the presence of a trade off is not *necessary* for a concomitant outcome to be judged intentional. It may still be the case that it is *sufficient* for a concomitant outcome to be judged intentional that a trade off be involved. This would entail a revision of the trade off hypothesis, but only a circumscribed one.

    I hope this makes sense!

    Thanks again for all these comments!

    Edouard

  51. K? O'Rourke says:

    Andrew, Phil: One of the important things I learned for Don Rubin was not to take different as less helpful (or more wrong).

    You have to endure your half empties with the celebration of your half "Phils" (not meant in any way to be negative – I just can't help it).

    Anyway, bit surpised their was not a Bayes analysis i.e. prior of autisim * likelihood of not getting intentionality given autism, etc.

    By the way, Wittenstein, lamented his lack of getting intentionality when realizing the failure of his tractatus – but then I am running into my half empty.

    K?

  52. K? O'Rourke says:

    Arghh – for – from, must be my screen

    K?

  53. fraac says:

    Ed, I think you may be able to separate the intentional/uninentional extra dollar non-autistics with a question like "How confident are you generally?", though it's a longshot. I'm going to take the behaviourist approach and see if we can spot anything.

  54. Murat Sincan says:

    I don't necessarily agree with your two hypothetical cases in the end. It is not the trade-off between any two things that qualifies a decision as intentional but the value associated with that trade off that makes or even quantifies the intentionality of the decision. Since Joe doesn't care if he gets or doesn't get the commemorative cup it doesn't matter to him so the decision is unintentional but a dollar always matters no matter you gain or loose it so in both the original case and your second case it is perceived as an intentional decision. Thats how I understand this concept.

  55. fraac says:

    Here is a good analysis by a non-autistic: http://adamcadre.ac/calendar/12696.html

    The line the stands out, that I believe shows a method universal among non-autistics (except maybe bpds and/or psychopaths) is this:

    "I just put myself in Joe's place and felt how much intentionality each case required me to summon."

    He put himself in Joe's place and ignored Joe's own testimony. This is exactly what I think is going on most of the time in most conversations taking place on Earth.

  56. stormy says:

    This is a completely bogus experiment and proves nothing….why? because its completely subjective. There is not majority answer that is right, period.

    I would add BOTH experiments are both NOT intentional and Both ARE intentional at the same time. It depends on your perspective and its relative.

    In addition, if a speaker says he "he does not care", its not complicated! Its means HE DOES NOT CARE, and so nothing can ever be intentional period, except the one thing he cares about. Its also possible, he could not care and care at the same time in all four cases…He wanted his drink at the size he wanted, but then because he has a secondary motive, of saving money and collecting cups to save even more money, it helped him anyway. He got home and switched….he wanted BOTH a drink, but also the second motive…..so maybe he lied! hehe…..human nature is alot more complicated than most psychologist pretend it to be. Its why they STILL dont understand aspergers, autism and are constantly redefining it. A bunch of morons who have yet to define their field of study, if you ask me.

  57. Andrew Gelman says:

    Hey, no need to shout!