Philosophy, Film, Politics, Etc.

Saturday, March 22, 2014

Cratylus and Kripke

I've just turned my attention to Plato's Cratylus for the first time.  Nothing of significance brought it to my attention, but when I learned it is considered his only sustained treatment of language, and that it deals with the question of whether meaning is natural or conventional, I thought I should become familiar with it.  I have only looked at the very beginning, but already I wish I had more time to explore it and its relation to 20th-century philosophy of language.

At the beginning of the dialogue, Hermogenes affirms a view of names which is relativistic:  the designation of a name is entirely dependent on conventional usage.  The rightness or wrongness of word meaning is relative to a linguistic community.  If I use the word "horse" to refer to humans, and "human" to refer to horses, then I am right as far as my own usage goes, though I am not in line with the majority.

Socrates leads Hermogenes to find a problem with this view with a curious argument.  He gets Hermogenes to admit three points:

(1) Propositions are either true or false.
(2) The most basic element of a proposition is a name.
(3)  If a proposition is true or false, then its parts are true or false.

From this it follows that names are either true or false.  So there must be something wrong with Hermogenes' conventionalism.

It would seem that (3) is a point of weakness in Socrates' argument, but I want to suggest otherwise.  We might be tempted to say that only propositions as a whole can be true or false.  In one sense, this is correct.  Propositions predicate properties, and in this way they can be true or false in ways which names cannot.  That is, unless we take names to also predicate properties to their designators--however, if we did that, then we would arrive at an infinite regress, because every name would express a proposition which would be composed of yet more names, ad infinitum.  So there must be a sense in which propositions do more than individual names; and there is a sense in which they alone, only as a whole, can be true or false by virtue of that function.

I'm afraid it is too easy to use this point to reject Socrates' argument.  I have not read further into the dialogue than this, so I am not sure what Plato makes of it.  But just taking the argument as it stands, I think we can find something much more interesting to consider about it.  For there is another sense in which a proposition can be true or false:  namely, a proposition can be true or false in so far as it is used correctly.  This is what Kripke calls truth in a metalinguistic sense (see Kripke's Wittgenstein on Rules and Private Language).  If we abandon the notion of metalinguistic truth, Kripke suggests, we undermine our very conception of meaning.  (Actually, this is not necessarily Kripke's argument; it is an argument Kripke formulates as inspired by Wittgenstein, and so it may be better to call it "Kripkenstein's argument.")

In short, Socrates' argument may be this:  If we adopt conventionalism with respect to names, we are claiming that metalinguistic truth is relative, and so the objective truth of all of our utterances is undermined.

I don't know when I'll have time to read the rest of Plato's dialogue or to investigate the literature on it.  I haven't even read as much Kripke as I should.  For now I can only appreciate the intriguing connection between Plato's thought and 20th-century philosophy of language.

Thursday, March 20, 2014

Dennett's response to stubborn scientists

It is perhaps well known that there is more than one way of interpreting the phrase "free will" and that many scientists are very stubborn about how they are willing to interpret it.  Dan Dennett has recently (in the last several months) acknowledged that there are benefits to giving up the term, though also that there are serious costs and risks.  This has led to some speculation by Gregg Caruso at Flickers of Freedom about whether or not Dennett has shifted his position wrt compatibilism.  One commenter, Randall Harp, draws attention to this lecture from November 2013, in which Dennett makes and expands upon the same point (Dennett starts speaking around to 21-minute mark):

This is entirely in line with Dennett's well-known views.  In fact, it's worth watching as a concise introduction to Dennett's thoughts on free will and moral responsibility.  He has not shifted his position and he still uses the phrase "free will" as he always has.  As I posted at Flickers of Freedom,

Dennett is clearly not endorsing abandoning the term "free will." He is acknowleding that it is not necessarily a good idea to insist on using it (because many scientists are very stubborn about how they are willing to interpret the term), but he is not giving it up. He's rather saying that the people who insist on not using it are missing the point. He has not switched sides or made any substantive concessions.

As I also posted in a Facebook thread started by Rick Repetti, Dennett is
just being a bit wily. If anything, he thinks that merely dropping the word "free will" from our vocabulary would be very misleading and dangerous. He's just saying that people who are fussing over the term "free will" are missing what is really at issue here. He thinks they're just as wrong as ever about the substantive issues--about whether or not determinism is compatible with moral responsibility and a robust sense of rational agency. He suggests that maybe, when talking to scientists, it is better to address these issues in other terms, and not insist on using the term "free will." But he is also saying that even if we stop using the term free will, as people like Sam Harris suggest, we still can't say the sorts of things Sam Harris wants to say about rational agency and moral responsibility. And he still seems to think that the phrase "free will" is very useful.

Friday, March 7, 2014

Ahab, Dennett and Rational Agency

"Vengeance on a dumb brute!" cried Starbuck, "that simply smote thee from blindest instinct! Madness! To be enraged with a dumb thing, Captain Ahab, seems blasphemous." 
"Hark ye yet again- the little lower layer. All visible objects, man, are but as pasteboard masks. But in each event- in the living act, the undoubted deed- there, some unknown but still reasoning thing puts forth the mouldings of its features from behind the unreasoning mask. If man will strike, strike though the mask! How can the prisoner reach outside except by thrusting through the wall? To me, the white whale is that wall, shoved near to me. Sometimes I think there's naught beyond. But 'tis enough. He tasks me; he heaps me; I see in him outrageous strength, with an inscrutable malice sinewing it. That inscrutable thing is chiefly what I hate; and be the white whale agent, or be the white whale principal, I will wreak that hate upon him. Talk not to me of blasphemy, man; I'd strike the sun if it insulted me.
                                                                                      --Moby Dick, Chapter 36, by Herman Melville

Ahab does not distinguish between whale, sun and man.  All are pasteboard masks, all giving the appearance of order or truth, but hiding some as yet unknown power or force.  Does it make sense to punish a whale or strike the sun if we feel it has done us wrong?  What about punishing a person?  At what point can we say that an entity--whale, sun, or human being--deserves to feel the force of our wrath?

Ahab hates Moby Dick because he sees malice in the whale.  Is the whale the source, the ultimate cause of this malice?  In other words, is Moby Dick the "principal"?   Or is the white whale merely an agent of another's will?  It does not matter, says Ahab.  The whale is all he has to fight against.  If there is some other power beyond the whale, then he will strike at it by striking the whale itself.

I don't want to get into whether or not Ahab is right to hate Moby Dick, or whether his fight against Moby Dick is rational.  What interests me is the distinction between principal and agent, and whether or not we should follow Ahab's lead in rejecting it.

Let's approach this in philosophical terms.  When we are talking about sources of intentional action--minds capable of acting on reasons--we are taking what Daniel C. Dennett calls the intentional stance.  Taking the intentional stance is treating an entity as if it had beliefs and desires, and the capacity to act rationally on them.  So, if Ahab were to regard Moby Dick as the principal--the reasoner behind the malice--then he would be treating Moby Dick as a rational agent with beliefs and desires.  Starbuck says this is blasphemous.  We should not treat whales as if they were rational agents.  Ahab says that he would treat the sun the same way.  In other words, he would treat the sun as an intentional agent, too.

But wait.  Ahab does not exactly say that he is treating Moby Dick as a rational agent.  He says it doesn't matter.  Ahab is equally fine regarding the whale as an agent of another's will.  In other words, he is fine treating the whale as if it were designed in order to serve some purpose.  So the whale might be understood as the intentional creation of another's will.  To treat the whale this way, we would be taking what Dennett calls the design stance.  This is just the intentional stance applied to artifacts.  To take the design stance, you imagine an object as the product of some intentional design. You imagine how the designer intended the object to function--you imagine what beliefs the designer had and what desires the object was meant to satisfy.  In doing so, you can predict how the object will behave.  So, Ahab can treat the whale as a "blind brute," an unthinking robot which is merely carrying out a program designed by some other intelligent being.

Ahab says it doesn't matter.  Either way, we are imagining the actions of the whale as the result of intention--as the result of beliefs and desires.  But it does seem to matter, doesn't it?  Don't we want to know if the whale is an intelligent creature?

Let's look at another example:  a thermostat.  Dennett says that we can treat thermostats with the intentional stance.  We can imagine them as having a rudimentary set of beliefs and desires, and this allows us to predict how they will respond to changes in room temperature.  But, of course, we know that thermostats are designed.  So we can (and should, I would say) take the design stance towards them.  They do not have beliefs and desires of their own, but are artifacts which carry out the intentions of their designers.

Dennett says that there is no real difference here.  He rejects the distinction between "original intentionality" and "derived intentionality."  In other words, there is no difference between treating an object as having beliefs and desires of its own and treating it as if it is carrying out the intentions of some other agent.  It would seem Dennett agrees with Ahab, then.  It does not matter if Moby Dick is principal or merely agent.

Should we be satisfied with this approach?  Well, what of a mercury thermometer?  Like a thermostat, it tells the temperature.  And like the thermostat, it was designed.  When I take up the design stance towards the mercury thermometer, I am taking up the intentional stance towards its designer.  So, following Ahab and Dennett, I should not think there is a significant difference here.  I may as well treat the mercury thermometer as a rational agent in its own right.  Ahab and Dennett lead us to conclude that any designed object can be regarded as a rational agent with beliefs and desires of its own.

Ahab goes one step further.  He says the sun, too, can be regarded in the same way.  And indeed, why not?  Well, we might object that the sun was not designed.  We certainly have no evidence that it was designed. But that is not the deciding factor.  According to Dennett, the intentional stance is justified just in case it is useful as a predictive strategy.  If we can make successful predictions about the sun by taking up the intentional stance towards it, then we are justified in doing so.

It's natural to use the language of intentionality to talk about natural processes.  We usually refer to this as personification and metaphor.  However, Dennett's view is that, when it comes to using the intentional stance, then it is not a metaphor.  If we are making successful predictions about the sun's behavior, we are dealing with a real intentional system--as real as any can be.

It goes without saying that people have, throughout history, regarded the sun, the moon and countless other natural objects as intentional agents.  They made predictions on this basis--some more successful than others--but they did the best they could.  Aristotle (and others) regarded all entities in teleological terms, and made reasonably successful predictions accordingly.  We therefore can say that it is justifiable to regard even the sun as a rational agent, or as exhibiting the intentions of some other agency.  Remember, Ahab and Dennett say there is no significant difference.

I, for one, am not satisfied by this approach.  There are obvious advantages to distinguishing between rational agents and mere artifacts.  We should not regard a clock as a rational agent merely because it carries out the intentions of its designer.  We should not regard any complex computer as a rational agent merely because it carries out the intentions of its designer.  It is not enough to say that the intentional stance is useful and successful.  To say that an object is a rational agent, we should have some means of determining whether or not the intentional stance is justifiable as opposed to the design stance. That is, we should think we are dealing with rational agency if and only if we are dealing with apparently rational behaviour which cannot be explained as the result of another agent's intentions.

We can explain the behaviour of the sun without appealing to rational agency of any sort.  Therefore, we should not regard the sun as an intentional system.  We can thoroughly explain the behaviour of a thermostat by appeal to the designer's intentions, and so we should not regard a thermostat as a rational agent.  The same goes for more advanced systems, like chess-playing computers.  Human beings, however, appear to be rational agents and we have no explanation for our behaviour as the result of any other agent's intentions.  Therefore, we are justified in regarding ourselves as rational agents.

In short, Ahab is wrong.  And so is Dennett.

P.S. I realize I have not given necessary or sufficient conditions for rational behaviour here.  Perhaps a sufficient condition is something like what Dennett suggests:  That we are justified in regarding an entity as a rational agent if doing so leads to successful predictions.  I am just adding the caveat that we must not be able to explain the behaviour as the result of another agent's design.  I would also add that whether or not an entity really is a rational agent is not merely a matter of whether the intentional stance can be used to make predictions about the system, but a matter of whether or not the intentional stance is necessary to make the best possible predictions about the system in question.  So, people may be justified in using the intentional stance to talk about the sun, but there is a real sense in which they are wrong.  We do not need the intentional stance to make the best possible predictions about the sun.  I believe we do need the intentional stance to make the best predictions possible about human beings.  This conclusion is based on the fact that we do, in fact, apply the intentional stance.  Perhaps we do so by design--but even then, we are still taking up a predictive strategy, which means we still have beliefs and desires. So even if we are designed, we are rational agents.  (There's no reason to assume that rational agents cannot be designed.)  The fact that we do take up the intentional stance proves that we are rational agents.  As such, we must be regarded as rational agents or we will not be fully understood.  We might suppose that some alien creatures could make predictions about our bodily movements without regarding us as rational agents. However, all this shows is that we--our identity as rational agents--is not defined by our bodily movements.  We may supervene on our bodily movements, but we are not identifiable with them.  So those aliens will not be describing human action.  They will only be describing bodily movements.  And since human action is real, their descriptions will be incomplete.  At least, that's the position I'm leaning towards now.

Sunday, March 2, 2014

The Intentional Stance

One of Dennett's most well-known contributions to philosophy is the idea of the intentional stance.  This is supposed to be a template for understanding minds and rational agency.  However, Dennett's view of the intentional stance has shifted since he first began its formulation in the mid-1970s.  His earliest work on the topic indicates the following thesis:  An object is an intentional system just in case it is advantageous to regard it as such.  Here he is, in 1976:

An Intentional system is a system whose behavior can be (at least sometimes) explained and predicted by relying on ascriptions to the system of beliefs and desires (and other Intentionally characterized features--what I will call Intentions here, meaning to include hopes, fears, intentions, perceptions, expectations, etc.). There may in every case be other ways of predicting and explaining the behavior of an Intentional system--for instance, mechanistic or physical ways--but the Intentional stance may be the handiest or most effective or in any case a successful stance to adopt, which suffices for the object to be an Intentional system. So defined, Intentional systems are obviously not all persons.
Dennett goes on to explain that even plants can be talked about as if they were intentional systems, which means that they are, lo and behold, "very low-grade Intentional systems."

How does Dennett differentiate the grades here?  Presumably it has to do with how many orders of belief attribution the object in question can muster.  However, once we treat an object as an intentional system, there's no reason to stop at any particular level.  As FC Young (1979) points out, we always have the option of multiplying orders of intentionality.  We can always reframe our explanations in terms of highly complicated teleological terms.  The fact that we can do it does not mean we should.  So why shouldn't we?

Dennett's response to this problem has earned him a fair bit of criticism.  His response is to try to have it both ways.  On the one hand, he does not clearly give up his original position.  To be an intentional system just is the ability to be regarded as one through a successful application of the intentional stance.  On the other hand, he acknowledges objective constraints which determine whether or not we are justified in applying the intentional stance to an object.  The problem is, he cannot have it both ways.

Dennett (1981) presents a situation where he thinks the intentional stance is obviously unjustified:  his lecturn.  Clearly such an inanimate object cannot be usefully regarded with the intentional stance.  His argument is that the addition of teleological elements does not add to the predictive power we already have.  This presupposes that we are not the sort of people to regard all objects as having teleological functions.  It also means that there are constraints on what can or cannot justify the application of the intentional stance.  Sometimes it gives us added value, sometimes it does not.  If it does not, then it should be cut (via Ockam's Razor, basically).  But in that case, being an intentional system is not merely being thought of as one (in the context of making successful predictions).  It is being thought of as one when doing so is necessary to sustain our ability to make predictions.

The problem for Dennett is profound.  If there are conditions on what should or should not be considered an intentional system, then our theory of intentionality should help us understand why.  We want a theory of intentionality that gives us the criteria, or at least points us in the right direction.  The direction Dennett gives us is not so useful.  His position amounts to this:  The objective criteria for determining whether or not an object is an intentional system can be found by determining whether or not our ability to make predictions about that object requires that we regard it as an intentional system.  What we want to know, of course, is how to determine if it is necessary to regard a system as an intentional system.  Dennett's approach seems to have us chasing our tail.

It is often argued that Dennett's position is a useful way of observing the similarities between human beings and other animals, and even computers.  Part of Dennett's position is that there is no "magic moment" when systems become so complex that they become real rational agents, as opposed to the sort that are defined merely by taking up the intentional stance towards them.  So, for example, it is said that even chess-playing computers are rational agents, because the intentional stance is highly useful (even necessary, it is argued) for predicting their outputs.

Indeed, if you are playing chess against a computer, you can imagine that the computer has belief and desires about the game.  Dennett says your best strategy is to imagine that you are playing against a rational agent, a system which has true beliefs about the game and which wants to win.  I am not sure.  It may, in fact, be more productive to think of the computer as having been designed to simulate a chess-player without reproducing the beliefs and desires that a person would have while playing chess.  There is no demonstrable need to imagine that the computer is a rational agent at all.  We should therefore recognize that the computer simulates the behavioral outputs of a rational chess-player without having any beliefs and desires of its own.

How do I know the computer does not have any beliefs and desires of its own?  Dennett provides us with the answer here:  Because I do not gain any predictive value by supposing it has beliefs and desires of its own.  I know it is a programmed simulation.  I do not need to suppose it actually wants to beat me.

Dennett challenges us to find a situation in which we simply must regard a system as having beliefs and desires of its own.  But that is not a hard challenge to meet at all.  In your next conversation with a person--any person--talk to them as if they do not have beliefs and desires of their own.  In a real-life situation with other walking, talking human beings, actually deny that they have beliefs and desires of their own.  How far will that get you in explaining and understanding their behaviour?

But okay, that sort of challenge might be complicated, and it might be difficult to interpret the results.  So here's a much simpler example:  Yourself.

Dennett makes this very simple for us.  He acknowledges that we all treat various objects as if they have beliefs and desires.  We all take up the intentional stance.  This is a given.  It is also given that only rational agents can employ the intentional stance.  It follows that we all must be rational agents.  We could not employ the intentional stance if we were not.

You might say, "Sure, I know I am a rational agent, but how do I know you are?"

It is a reasonable deduction, given that we are of the same species and function in more or less similar ways.  I certainly appear to be employing the intentional stance, and I certainly seem to function just like you.  Evolutionary theory gives us a basis for understanding these commonalities.  So it is reasonable for you to assume that other human beings are just as much rational agents as you are.

The question is, why should we think that other objects have the same rational agency we do?  Dennett's argument appears to be that it is useful to act as if they do.  While that is true, this usefulness has limits and can often lead us astray.  Dennett should recognize that there is a difference between making good use of a metaphor and mistaking that metaphor for a literal truth.  We need some way of distinguishing between the metaphorical use of the intentional stance and the literal one.  Dennett does not offer one, and so his project is unsuccessful.

Dennett acknowledges that there is no doubting we are rational agents.  Yet, he says that we might, in theory, be able to predict all of our behaviour without taking the intentional stance at all.  We might do it with only the physical stance.  This is one of his most controversial ideas:  If we learned enough about our physical bodies, we would not need to regard ourselves as rational agents at all!  Yet, we would still be rational agents.  So, our physical account would be leaving out true facts about us.  Recall that, to avoid regarding his lecturn as a rational agent, Dennett must acknowledge physical facts which determine whether or not he is a rational agent.  So his complete physical account would seem to leave out the physical facts which determine that he is a rational agent.  This is impossible.

Perhaps Dennett could say the physical description only leaves out those facts under the description that they determine he is a rational agent.  However, this is highly problematic.  It is not clear what other sort of description would be available, since being a rational agent should be intelligible under the physical stance--if it were not, the fact that Dennett's lecturn is not intentional could not be decided by appeal to physical facts.  I might be missing some subtleties here, so I cannot discount this option altogether, but it does seem problematic.

Another option is to just claim that rational agency is intelligible as such via the physical stance.  However, in that case, the difference between the physical and intentional stances collapses.  This could be taken to mean that all physical systems exhibit intentionality, or it could mean that intentionality is an irreducible aspect of physical systems.

Here are Dennett's options, as I see them:  (1) Appeal to the physical stance in order to ground rational agency (in which case the intentional stance collapses into the physical stance).  In this case, intentionality is an irreducible property of physical systems, which means either all physical systems are intentional or there is a property dualism between mind and body.  (2) Regard rational agency as both real and irreducible to the level of physical cause and effect.  This option looks more like substance dualism.  (3) Claim that intentional states can be intelligible via the physical stance, just under a different description.  This might be Dennett's best bet, but I think it is very problematic, if Dennett wants to claim that there are physical facts which we can appeal to in order to determine whether or not an object is a rational agent.  (I don't think he wants to claim this, but I'm not sure he has a better choice.)

Of course, Dennett might choose a fourth option, to abandon realism altogether, saying that, by regarding ourselves only with the physical or design stances, we would stop being rational agents.  This would make rational agency completely subjective, so that being regarded as a rational agent is a necessary condition for rational agency. Thus, we would be making predictions about our behavior, wanting to predict outcomes and having beliefs about how to do that--we would exhibit all the qualities of rational agency without being rational agents. I suppose we would say things like, "I am not a rational agent; I only believed I was so I could make better predictions about my behavior.  Now I know better!"  That's rather like saying, "I used to think, but now I know better!"  No, that would be absurd.  I'm not even sure I should count this as an option.

I should note that it will not do to try to reduce the intentional stance to the design stance.  The design and intentional stances are not really distinct stances at all.  To say that object x was designed is to say that it serves some function.  In other words, it has a purpose or intention behind it.  Designed objects are intentional objects; when we make predictions according to design, we are making predictions about how they were intended to behave.  We aren't supposing that the objects are the source of their intentionality, but we are still interpreting them as (extended) parts of an intentional system.  The design stance is the intentional stance applied to artifacts of intentional agents.  This can be useful even if there was no intention behind the design (as, for example, in the case of evolved traits), of course; but we should recognize a difference between merely being a useful metaphor and being literally true.  We should be cautious about putting too much stock in the design stance for the same reason we want stronger criteria before we start believing that chess-playing computers are rational agents.

P.S. I don't have free access to it, but judging by the abstract, Paul Yu and Gary Fuller (1986) seem to make an argument which is in some ways parallel to my own.

[Editorial Note:  On March 4, 2014, I significantly altered a portion towards the end of this post, correcting my presentation of Dennett's position and significantly improving my argument.]