About Rationally Speaking


Rationally Speaking is a blog maintained by Prof. Massimo Pigliucci, a philosopher at the City University of New York. The blog reflects the Enlightenment figure Marquis de Condorcet's idea of what a public intellectual (yes, we know, that's such a bad word) ought to be: someone who devotes himself to "the tracking down of prejudices in the hiding places where priests, the schools, the government, and all long-established institutions had gathered and protected them." You're welcome. Please notice that the contents of this blog can be reprinted under the standard Creative Commons license.

Monday, February 21, 2011

What do you mean when you say “rational”?

by Michael De Dora
A couple of days ago I was perusing the Wikiquote page for my favorite philosopher, Bertrand Russell, and came across the following remark: “Man is a rational animal — so at least I have been told. Throughout a long life, I have looked diligently for evidence in favor of this statement, but so far I have not had the good fortune to come across it, though I have searched in many countries spread over three continents.” I’d seen this quote before, but this time it got me thinking about rationality more in-depth than I had before. Russell, from what I gather, was lamenting the fact that most human beings believe in seemingly false claims, and commit unjustifiable and horrible acts. I am with him on that. But I have also said in public exchanges about religion that many of those beliefs and actions are what we might consider rational given the requisite assumptions. This raises a question over which I have been stewing since re-reading the quote: what does it really mean to act rationally?
I believe this question is best answered by making a distinction between two different conceptions of rationality, which I will call interior and exterior (or rational and Rational). This is the difference between acting a certain way given a starting point or context (interior or rational), and acting a certain way because it is right or true (exterior or Rational).
Interior rationality looks something like this. People have beliefs. Upon those beliefs, they base other beliefs. For a mental image, this view of beliefs wouldn’t look like a neatly stacked block of wood, but instead a jagged pile of rocks, or an intertwined mess of tree branches. The point being, beliefs are dependent on other beliefs, and together they give rise to certain behaviors, outside of what is true or not. Whether a person’s belief is true or not when he or she starts the reasoning process, it might still be rational to engage in certain acts given that belief. In short, a person can be rational within an irrational system. This is interior rationality: acting while maintaining as much coherence as possible with your beliefs and values.
For example, say I cheat on my girlfriend or wife by having sexual relations with another woman. That would be an awful thing to do. But, if I believe I am a rational and nice person, then I might seek to justify my behavior. As Carol Tavris and Elliott Aronson point out in the book Mistakes Were Made (but not by me): “Because we are not irrational and mean-spirited, any negative feelings we have about another group are justified; our dislikes are rational and well-founded.” In my example, I might wonder how a rational and nice person like me could act so irrationally and meanly. Aha! It must be someone else’s fault (i.e., mistakes were made, but not by me). So I might think things like “well, our relationship was really on the rocks anyway,” or “I think she cheated on me once, so I’m excused.” Of course these “reasons” may simply be invented: was the relationship really on the rocks? Did I ever really suspect her of cheating? Again, the point is that it all makes sense if I also believe that I am a rational and nice young man.
Yet, even as these beliefs and behaviors are rational given my assumptions about my niceness and reasonableness, they are not rational in the exterior sense. Exterior rationality does not mean acting in accordance with a given set of beliefs, but in accordance with our best assumptions as to what is right or true. Acting rationally in this sense would have me pondering whether I am really a nice and reasonable young man. If reason and evidence tell me it is not so, I’d have to adjust.
When compared to exterior rationality, interior rationality is often dismissed as necessarily irrational. However, this need not be the case. Interior rationality is prone to mistakes, but we can use exterior rationality to keep it in line. Indeed, I think both of these rationalities are worth striving for, as they mean we are acting according to what we have decided is right or true within our system of thought, and in our somewhat objective study of it. The key is to closely monitor the reasons why we are assuming we are rational. It is vital to dig through our beliefs until we get to the roots – to the basic assumptions that make us decide our behaviors are rational. The accuracy of our beliefs is a good clue into whether we’re really acting rationally. And this is critically important, because from my vantage point, while man might have impressively honed interior rationality, he has a long way to go before achieving rationality in the exterior sense proposed here.

62 comments:

  1. i think it makes more sense to come at this question from the other angle - what makes us superstitious? (ok superstitious isn't exactly synonymous with irrational, but bear with me ;)
    as part of my job i deal with scientific cameras; when an image is taken with no light falling on the CCD the image is just random noise. However, it is very easy to see patterns in the random noise - something that would indicate a fault in the camera. I believe that our intelligence is essentially spotting patterns. if our ancestors ate a red berry and became ill, then red berries were avoided in future. this 'see a pattern in a tiny sample size' phenomenon is the basis for superstition.

    ReplyDelete
  2. Coherent is already in wide-spread use to define your interior rationality. (Is it rational to favor cute dichotomies, even when they aren't needed?) And "sexual relations"? Please, I don't need to be reminded of Willy's willy any more than I currently am. Having sex, doing the nasty, banged, even the f word is preferable.

    And your take on the exterior rationality is standing on turtles all the way down. When it's impossible to derive ought from is, two inherently opposed views can still be called exterior rational. Just it's different exteriors.

    ReplyDelete
  3. Michael, your example about cheating on a wife is not about being rational but about being moral ("right" in the moral sense), and I suspect one is different from the other.
    Economists have devoted long time to debate and elaborate on rationality, since most economic theories can (and often are) construed as based on the assumption that all people behave rationally. What rationality is, whether people behave rationally, and whether economic theory actually requires agents to behave rationally, have been studied both conceptually and experimentally.
    From that long discussion, several interesting points:
    1. One may deduce or predict facts of economic life (e.g. that a lower price causes an increase in demand) from the expected behavior of a population of rational agents. But one can also predict the same fact (the same demand curve with the same decreasing slope) from other assumptions. None less than Gary Becker (1962, ref below) proved in a simplified model that agents acting randomly, or agents acting "inertially" (i.e. reacting to a price drop either with a random response or with persistence in the same behavior observed before the price drop, constrained only by a budget constraint) would generate the exact same curve.
    Beware that the demand curve is a market-size phenomenon, produced by a myriad agents, whilst the rational or irrational behavior is predicated of individuals.
    2. Rationality in economics is always defined in such a way that it is limited to the choice of means, i.e. means to achieve a certain goal dictated by unexplained (exogenous) orderings of preferences. The preferences themselves are not subject to the possibility of being rational or irrational --not at least within Economics. If you prefer a diet of 50% chicken and 50% fish, whilst I prefer 100% chicken and someone else prefers 100% vegetarian, each one may rationally choose goods at the supermarket according to such preferences and looking of course at the price tags on the articles and the amount of money in our wallets. Economics is the study of the allocation of scarce resources between different ends, and not the study of the best ends to pursue.
    3. Experimental studies such as those by Kahnemann, Tversky, Vernon Smith and others have shown that people are often not able to make a rational decision under uncertainty even if fully informed about the probability of various outcomes (not even doctors, apparently, are likely to tell you the correct probability that you have AIDS after testing positive twice). Some other authors, like Gerd Gigerenzer, have shown on the other hand that the same people make more correct decisions when the question is not framed in probability terms but in terms of actual number of cases involved. All these research programs in behavioral or experimental economics (that also cover other areas of behavior outside narrowly economic activity) insist that decisions are made under "bounded rationality", especially because acquiring all the necessary information for a correct decision may take time and resources, and thus people are apt to decide with incomplete information; even so, Gigerenzer says, they use well-tested practical heuristics that (most of the time) lead to correct decisions (as defined by some logical or mathematical framework). Some of these heuristics, other authors say (and this will not please Massimo, but so they say) have actually evolved in our long evolution as a species. For instance, they have shown babies under 2 years of age, or barely some few months old, using some of such heuristics correctly, with no prior training, and have also shown this to happen with babies of very disparate ethnic, geographic, cultural or social backgrounds. Other heuristics, instead, have to be learned, say the same authors. Some animals use similar heuristics in similar situations.

    (character limit; to be continued).

    ReplyDelete
  4. (continued)


    4. When you make a decision, and I know nothing about you, I may make the judgment that you are irrational; e.g. if I see you purchasing a more expensive good when a cheaper equivalent is available. After learning some other things about you I may correct my initial judgment (perhaps you prefer the more expensive brand, for whatever reason). If even this test fails, I may learn also that your choice was guided by some other principle (e.g. the cheaper good is at another shop four blocks away, and you don't want to spend time going there, because the value of your time for you is larger than the savings you will make). In the end, if I "internalize" more and more constraints operating on your choice, I may end up considering that practically ANY behavior you adopt may be "rational" all things considered. (In a similar way, your guilt in a crime may be excused by acting in legitimate defense, by not having being able to perceive what you were doing, by temporary insanity, by having being beaten or bullied or sexually abused as a child, or by whatever other excuse the courts may happen to accept in your case; with sufficient "extenuating circumstances" almost any crime might go unpunished). Internalization of more and more conditions (other than your own preferences and your budget constraint) risks making "rationality" moot.
    5. Being rational in your mind may not necessarily lead to your acting rationally. "I know what is good but I do what is bad" (approximate quote from one St Paul epistle). You may rationally know that smoking is bad for your health, but you keep smoking: that behavior probably does not mean you are not a rational being.
    Hope these random thoughts help pushing the discussion forwards.

    ReplyDelete
  5. Omitted reference:
    Becker, Gary S. (1962). Irrational behavior and economic theory. The Journal of Political Economy, 70(1):1-13.

    ReplyDelete
  6. Michael,

    Your post strikes me as less than clear. First, tet us take the following. You write:

    "Exterior rationality does not mean acting in accordance with a given set of beliefs, but in accordance with our best assumptions as to what is right or true."

    However, acting in accordance with 'what is right or true' entails acting in accordance with a set of beliefs, i.e., acting in accordance with what one believes to be be right or true.

    If, given his evidence, A believes p is true to a high degree of probability, then even if p in the end turns out to be false, A is still rational to act in accordance with p (p may be a research theory and his actions may be the allocation of otherwise scarce research resources). Indeed, I suspect you would agree that A would be irrational not to so act, in which one need not act in accordance with what 'is' true in order to be rational; rather, one need only act in accordance with the best available evidence.

    Furthermore, you seem to conflate irrational actions and irrational beliefs: one can have rational beliefs but act irrationally and one can have irrational beliefs and act rationally with respect to those (irrational) beliefs. On this you are not clear.

    ReplyDelete
  7. What I mean when I say "rational" is not what I mean when I say "rationalize."

    ReplyDelete
  8. A continuation of my comments before has not appeared (though an omitted reference, posted after the continuation, did). So I briefly summarised the missing continuation.

    4. An apparently irrational behavior may be seen to be rational once more information is available or is considered. One person looking for an article may prefer paying $10 at supermarket A, instead of paying $8 at supermarket B. Irrational, perhaps, unless the person does not know about B, and actually finding the lower price would cost time (with uncertain results), or (even already knowing about the cheaper price) it may be further away and thus the cost of going there (time + transportation) may be higher than the savings. This is quite trivial. But even if such incorporation of further information fails to render the decision rational, one may always imagine some other reason why the actual behavior was rational: the person may prefer brand A instead of B, for unknown reasons. Or she may have a "top price to pay" so she buys in the first shop offering the article at a price below the (previously decided) top price to pay. Reckoning that looking for additional information is always costly, settling for a not optimal choice may be rational after all.

    Internationalization of circumstances: Along this path of turning something into rational by way of "internalizing" more external conditions previously left aside may ultimately turn every imaginable behavior into a rational one (just like admitting more and more extenuating circumstances may end up rendering all crimes unpunishable). Thus rationality should be defined within the bounds of available information, given preferences, given uncertainty, given costs of seeking additional information, and expected benefits from additional information. This concept has led economists to the idea of "bounded rationality", amply discussed in the literature during at least four/five decades.
    Karl Popper came out with a similar idea in his attempt to formulate a "situated logic" for the social sciences instead of (or complementary to) the ruthless falsificacionism applied in the natural sciences. This situated or situational logic starts by asserting that intentional actions MUST be rational, leaving to the social scientist the task of finding the reasons behind the actions that make them rational (to the actor) at the time of acting. This "rationality principle" has been generally frowned upon by epistemologists of the social sciences, but in fact can be seen as an heuristic for such disciplines: start assuming the behavior is rational; find how and why it is; if you fail, then look for other explanations (insanity, drunkenness, knee-jerk reflexes and the like). The rule is similar to an Occam-razor-type of strategy used in other disciplines: a doctor faced with a patient's complaint may look first for the most straightforward explanation of the symptoms; if that fails, look for more unusual medical explanations, and if that also fails look for psycho-somatic causes or adopt the (Dr House-like) "the patient is plain lying" explanations. Start with the simplest explanation, and keep it as simple as possible (but not more), as Einstein used to advise.

    ReplyDelete
  9. Paraconsistent,

    Please give an example of an irrational action based on rational beliefs and vice verse. I'm not necessarily skeptical; just curious.

    ReplyDelete
  10. Paul,

    I did not say actions based on rational beliefs may be identified as irrational; however, I did state the converse. I mentioned that one can act irrationally even though one's belief sets are perfectly rational in order to divorce Michael's conflation between acting rationally and holding rational beliefs. An example of one whose actions were irrational yet whose belief sets were rational would be one who, say, believes a food is unhealthy, wishes to maintain a healthful diet, yet succumbs to his impulses and eats it anyway.

    An example of one who acts rationally with respect to irrational beliefs may be one who, say, believes in homeopathy and refrains from taking unnecessary homeopathic remedies because (according to his belief) doing so would result in an undesired outcome.

    ReplyDelete
  11. Oops! In my comment at 4.11 PM, the penultimate paragraph starts with "Internationalization" where I meant to write "Internalization". Sorry.

    ReplyDelete
  12. Has this distinction anything to do with the idea that Bayesian rationality simply assumes that your subjective beliefs satisfy the axioms of probability ('interior' rationality?), but that this does not entail that your beliefs are an accurate representation of objective frequencies ('exterior' rationality?)? I would like to know what do you think about the connections.
    .
    On the other hand, I'm curious about another possibility: that 'rational' is not taken an all-or-nothing term, but as an essentially comparative notion. Some actions can be 'more' rational than others (and less so than others), and even we might say that some people are 'more' rational than others (and less so than others). I wonder how the mathematical theories of action might represent this comparative notion.

    ReplyDelete
  13. So was I taught that "man is a rational animal" and I have arrived at the conclusion of how wrong this statement is. There are different ways of arriving at such conclusion, not only observation of human behaviour.
    Neurologically, the impulses to act in a given manner, are not generated in the so-called rational part of the brain and neither are the choices of premises for our thinkink. Those choices seem more or less originated by our feeling and unconscious part of the brain and may come from either genetics or culture - perhaps a very minor part may be influenced by feedback in very particular cases.
    All in all, I vote for "man is an emotional animal"

    ReplyDelete
  14. I'm trying to understand your distinction between interior and exterior rationality. Seems more like a difference between rationality and rationalization to me. The case of the cheating husband seems like a case of retrospective rationality; it seems less a case of him being rational, but him trying to rationalize his own behavior.

    Perhaps one can argue that the husband was being rational if he had evaluated all possible gains and losses from committing the adultery vis-a-vis not committing the act. If he has more to gain from cheating than not cheating, it would be rational for him to cheat no?

    ReplyDelete
  15. And to top it off, the continuation of my original comment DID appear after all, although very belatedly. I hope my points are clear after all this. Commentaria non sunt multiplicanda, would Mr Ockham say.

    ReplyDelete
  16. Weiye (February 22, 2011 2:00 PM):

    you have a valid point when you introduce the "difference between rationality and rationalization". However, that difference is not the one that appears in the case of the cheating spouse. That case, as I pointed out before, is one of conflation between being rational and being moral. Your own second paragraph correctly points out how a rational cheating husband may proceed (evaluating all gains and losses, taking his decision, and also proceeding with the necessary prudence to avoid detection, I suppose. That the action is moral or immoral has nothing to do with the issue of its being rational or non-rational.
    A more significant example is how a rational Auschwitz commander would proceed about his orders to exterminate Jews and other undesirables from the Third Reich. Assuming he prefers to obey the orders, or worse, that he concurs with the need of such extermination, his rational procedure would be (as it actually was in the documented examples of many lagers) to use careful and detailed organisation, economizing on time, personnel, poison and other scarce resources, use the captive labor force as long as feasible before sending them too to the gas chambers, and altogether proceeding with the utmost efficiency and promptness in the execution of his tasks. Such tasks look horrendous to us, and maybe also to him, but that is not a relevant question for an inquire about rationality, be it about rationality in action, in the sense of using adequate means to achieve given means; or rationality in thought, in the sense of drawing logical conclusions from a given premiss.

    Now about "rationalization": there is a psychoanalytical meaning for this word, bet let us leave that aside. In more common usage, it is an ex post (rational) explanation for something that was not so explained ex ante. This is not always a bad practice. In epistemology it is quite usual to speak of a theoretical model as something that "rationalizes" a given observation, an use very frequent in economics. For instance, an empirical research on consumer behavior may state that "an Constant Elasticity of Substitution (CES) demand function rationalizes observed consumer behavior relative to the choice between desktok and laptop computers". This means that IF a consumer's preference orderings (or those of an aggregate of consumers) are such that they can be represented by a CES function, then the behavior to be expected is precisely the observed one, plus minus statistical error. Similar judgments are also made in natural science: "Such and such evolutionarily sustainable paths rationalize observable fossil record and observed behavior for this species of mammal", in the sense that IF the animals followed those paths by natural selection, then the observed fossil and zoological record would be expected to arise. Thus the analysts SHOW TO BE RATIONAL something otherwise irrational or difficult to understand.

    ReplyDelete
  17. Erratum near the end of the second paragraph of my comment at 2.25PM:

    Where it is written:
    "in the sense of using adequate means to achieve given means"

    it was meant:
    " in the sense of using adequate means to achieve given ends".

    Sorry again. Writing too fast lately.

    ReplyDelete
  18. What I mean when I say "rational" is not what I mean when I say "rationalize."

    Rationalize means lying to yourself and others with an explanation of events and purposes with a pretense of as close to a logical and therefor rational process as possible.

    ReplyDelete
  19. This account seems confused.

    Interior rationality as you have defined it would permit me to say in defence of any action X that I held the belief that "I should perform action X", hence my action.

    You defined exterior rationality just as "doing what's right and true". Well I guess most people can agree with you there, but the interesting question is what distinguishes more effective approaches to that end (e.g. critical analysis, careful reflect, science) from less effective ones (superstition, dogma, religion). You have explained nothing.

    ReplyDelete
  20. This is an interesting take on rationality, but I think you'd be better off with the term "internal consistency" for your "interior rationality."

    I think when we say "rational" we mean something like "not in some way mistaken." The cheating person in your example is mistaken factually and morally, and all the internal consistency in the universe changes that not a whit.

    As an aside, Yudkowsky has a good fable on why consistency ain't nearly enough: suppose you're an anti-inductionist who subscribes to the gambler's fallacy; hence you think that if a coin has come up heads 16 times, that means tails is "due anytime now."

    We might be tempted to say "but this gambling strategy never works for you - look, you're poor!"

    To which you reply: "That just proves my luck is due to change!"

    ReplyDelete
  21. Ian Pollock opines: 'I think when we say "rational" we mean something like "not in some way mistaken."'

    Probably yes in some sense of the word "mistaken", probably not in others. But overall the judgment is sloppy. Rationality in thought (the dominion in which you can be or not be "mistaken") means applying certain rules to derive conclusions from given antecedents. If the premiss is false, of course a rational consequence of it may (will) be false too. From "Greek philosophers are immortal" one could derive "Socrates is immortal", and be mistaken though perfectly rational.
    The same is valid for rationality in action, for which "mistaken" (in other comments, not yours) tends to be replaced by "ethically wrong". You may belong to a cannibal society, and thus your plan to obtain the brain or leg of a fellow human being for your dinner may make perfect sense. Moreover, it would not be morally wrong for you within your particular culture, to it may seem disgustingly immoral to some foreign devil like me.
    Anyway, "rationality" as commonly used nowadays is not applicable to "morally right" actions, but to actions logically designed to attain a goal while complying as well with other constraints you may have established or been forced to accept (not exceeding your resources, not breaking any laws, choosing the most economical procedure, and so on). If you have a goal to attain (such as getting some human organ for dinner), being rational (as regards action) means that you would go after that goal in a logical, efficient and orderly manner.

    Regarding your orderings of preference, the only requisite that is ordinarily regarded as necessary for other consequences of rational action to arise is that the preference ordering is internally consistent. E.g., if A is preferred to B, and B to C, then A must be preferred to C. But even if you happen to have a non-consistent ordering of preferences, you may still be rational in the way you act when taking your decisions; for instance, you may decide to buy A over B if both cost the same, and B over C. And of course, you must prefer A to C if both cost the same. If you don't, your preferences are confusing, but have still not acted either rationally or irrationally in this matter. Once you have made your mind upon buying, say, A, then to be rational you should study the marketplace, choose the best shop, avoid unnecessary costs (such as fancy packing) unless required (e.g. for a gift), not get fooled or distracted by others calling upon you to change your mind and bur perhaps Z, and so on.

    Returning to Ian Pollock: the cheating spouse may be "morally mistaken" (in your view, not his or hers), but it he proceeds carefully he will still be a rational philanderer. About him being "factually mistaken" I cannot see how: perhaps the dalliance was disappointing in the end? But at the time of the decision it was factually right: "I like this lady", he may have thought, "therefore I'll go after her, taking care my wife doesn't get word of it". Perfectly rational, and factually accurate, I'd say.

    ReplyDelete
  22. Hector,
    Apparently, you can be rational and wrong at the same time, which is OK as long as you weren't purposely wrong, and especially if you were confident of your logic. Except that it's not a given that: if A is preferred to B, and B to C, then A must be preferred to C.
    If costs are not the problem, then the reasons for the preference could upset that order. A may be preferred to B for an entirely different reason then B to C, but for other reasons already in the mix, the elimination of B may cause C to be preferred to A. Since reasons can be expected to come in combinations of a different order then the options. Logic is not as mathematically correct as mathematics are logically correct.

    If you knew that already but didn't see the relevance, than your entire lecture may have been an excercise in rationalization.

    ReplyDelete
  23. @Hector:
    I said he is mistaken factually, because Michael had him inventing excuses about the marriage being "on the rocks" which were (ex hypothesi) not true.

    When I say "rationality," to be very explicit, I mean:
    The art of correctly applying logic, probability and reason to (1) obtain an accurate predictive map of the world, and (2) further one's terminal values.

    This is a very "big" definition, and for me it does essentially subsume ethics as a special case.

    >...the cheating spouse may be "morally mistaken" (in your view, not his or hers), but it he proceeds carefully he will still be a rational philanderer... "I like this lady", he may have thought, "therefore I'll go after her, taking care my wife doesn't get word of it". Perfectly rational, and factually accurate, I'd say.

    Erm... no. Let's break it down here. Some considerations (assuming he has made a monogamous commitment, which was not stated but implied):

    (1) Based on my own experience with the crooked timber of humanity, he is likely to regret the decision later, whether or not he is caught. I have a hard time believing that you can ever regret a rational decision ("regret" in the sense of considering your decision to be mistaken).

    (2) Making a decision to cheat on a spouse with whom one has taken on a consensual obligation is analogous to defecting on the Prisoner's Dilemma, in that it counterfactually would give other people, and in particular your spouse, good reason to cheat on you when they can get away with it. If you think that your spouse would be wrong to cheat on you even if you never found out, symmetrical reasoning applies to you as well. This is a subtle but powerful notion brought out more fully in Good and Real.

    (3) Cheating on a romantic partner can usually be seen as a surrender to akrasia (which also does great justice to our psychological state when we are tempted to perform such acts). Akrasia is best explained via the wholly irrational phenomenon of hyperbolic discounting in our utility functions (nobody would explicitly choose ice-cream over health, but an ice-cream now still somehow feels worth it). If akrasia is irrational (it is), then to the extent that the cheating reflects akrasia, it is also irrational.

    I anticipate an objection to the effect that I am assuming too much about this hypothetical man; I reply that I am only assuming he is a typical, non-sociopathic human with a utility function that includes things other than his immediate gratification, and concerns the welfare of people other than him. I don't think I've lost much generality.

    ReplyDelete
  24. Reason, rationality and their cousin science are simply methods. Given a desire and a set of environmental constraints on the achievement of that desire what then is the best method of achieving it? Therein lies rationality. If the goal is to have sex with as many women as possible (to maximize the chances of genetic transmission or pleasure etc.) then cheating on your girlfriend/wife is a perfectly rational thing to do. Of course doing that in a culture which views such behavior as wrong/sinful whilst trying to think of yourself as a 'good' person is a little more problematic. In that case you might need additional mental rationalizations such as those you mentioned to resolve that inherent conflict.

    Irrationality would come when doing something out of sync with your intended goal. For instance - doing extensive price researching would be irrational if your goal is to buy something in the shortest time period possible and vice versa. Neither extensive price researching nor buying things quickly could be considered irrational absent the context of the goal.

    Additionally I think people stray from the realm of reason when their desires (and the actions they take to achieve them) are directed towards things which no evidence supports the possibility of - things like entering into an afterlife when you die. Therefore while wanting to visit another star would be wildly optimistic it would not be irrational whereas wanting to sit at the right hand of a deity would be since there is no compelling proof of their existence.

    ReplyDelete
  25. @Massimo:
    This sounds, at least vaguely, analogous to the distinction between the coherence (internal rationality) and correspondence (external rationality) theories of truth.

    And, I'm about as likely to use "rational" for such internal rationality in an example of the nature you cite as I am to use "truth" for analogous examples supporting a coherence theory of truth.

    So, it's an interesting discussion, but, not the best use of the English language.

    ===
    @Hector:
    You offer some interesting thoughts on strength/validity of reasoning vs. truth-value of propositions in logical argumentation which also are somewhat analogous.

    ReplyDelete
  26. I'm a bit confused about your distinction. Our "best beliefs" are not always true. Truth is too much to ask for. We can only try our best. And coherence is always important for rationality. The idea of rationality is to have "justified beliefs." How exactly are (internally) rational beliefs justified differently from (externally) Rational beliefs? Both sets of beliefs would prefer to be true and based on evidence, wouldn't they?

    ReplyDelete
  27. Gadfly, this isn't my article, it's Michael's...

    ReplyDelete
  28. Baron P, the condition of transitivity (A>B and B>C imply A>C) is normally required of orderings of preferences so that these orderings (coupled with rational choice) lead to the kind of optimal states predicted in neoclassical economic theory. But the property of transitivity is not a requirement for the behavior to be rational, it is just a requirement for the (aggregate) outcome to have some specific properties (e.g. Pareto optimality). Of course people may have inconsistent preferences, and moreover, as you suggest, the introduction or elimination of one alternative may alter the ordering of the others. But that is all about preferences, which are considered exogenous and given. Choice is about choosing among the various options (A, B, C, ...) subject to certain constraints (chiefly the budget constraint mandating that you cannot devote to your preferred alternative more resources than you have).
    On the other hand, the whole theory of "rational choice" in economics has been under fire from several quarters, and the whole theory of neoclassical preferences have been in practice been abandoned since Paul Samuelson launched his concept of "revealed preferences" in 1937, whereby all theorizing about preferences as such is abandoned and all economic analysis proceeds on the basis of choices actually made.
    But all this refers, again, to preferences, not to the rationality of choosing one or another thing UNDER YOUR GIVEN PREFERENCES. And, of course, it is never the case that "costs are not a problem". There is no such thing as a free lunch.

    ReplyDelete
  29. Ian Pollock,
    1. Your definition of rationality covers too much, in my humble opinion, in comparison with current scientific views of the issue. You define it as "The art of correctly applying logic, probability and reason to (1) obtain an accurate predictive map of the world, and (2) further one's terminal values".

    I agree with the use of logic, probability and reason, but I am not sure about the rest. First, what does "to" mean in "to (1) obtain....."? It means that is your PURPOSE, or that is your OUTCOME? You may act rationally with those purposes, yet base your reasoning on faulty data and therefore fail to obtain the outcome. Even if the premiss is good, the world may turn out otherwise. Expected outcomes are not guaranteed by rational decisions. People may mistakes, even when applying logic, rationality and probability. The predictive map of the world could turn out to be all wrong, even if you anticipated it through entirely rational processes of thought. The future is inherently uncertain.
    Besides, let me comment that your definition is entirely about thought. Rationality is also about decision and action.
    2. Being rational today, with the information you have today, including your probability assessment that you may end up being wrong, does not imply that you could not be wrong in the end and regret your decision (Note besides that "regretting" is a feeling, nothing to do with rationality: you may discover you were wrong but nonetheless feel no regret since you feel you acted to the best of your knowledge, or simply because you are not the regretting type). Rational doctors decide to operate or not on a patient, rational businessmen decide to invest or not on a given asset, rational travelers decide to take or not to take that plane or train, and all may later find reasons to "regret" their decisions, although they were all completely rational (as is your action of crossing a highway if you reckon the risk of being run over by a car is negligible in that occasion; but you may be crushed by a truck anyway).

    3. Of course, "rationalizing" ex post just a weak way of inventing excuses, alleging non existent facts such as (in your example) that the marriage was going sour, is akin to the sense of rationalization in psychoanalysis. The false excuses may be invented consciously or inconsciously (the latter being the psyc. sense). But anyway, this kind of rationalization serves other purposes quite alien to the rationality of your actions. It may serve you to ethically explain or otherwise justify your conduct to your friends, spouse, or others (thus rationally trying to present to them a better image of yourself) or may serve you subjectively to appease your own feelings of guilt (probably a rational way of appeasing them; an alternative could be trying to get over that guilt entirely). But that has nothing to do with the fact that you decided to engage in (and actually conducted) your extramarital affair rationally or irrationally. Millions of people do such things every day, many without any regret or guilt, and most do it in a rational way, i.e. taking account of their subjective preferences and using the appropriate means to maximize the benefits and minimize the costs.

    ReplyDelete
  30. Says Gadfly: "@Hector:You offer some interesting thoughts on strength/validity of reasoning vs. truth-value of propositions in logical argumentation which also are somewhat analogous."

    In fact. In armchair reasoning, rationality is about the application of logic to certain premisses in order to reach conclusions. It is a formal process of thought, irrespective of the truth value of the premisses. Lewis Carroll wrote a whole treatise of Logic based on absurd premisses, just to show this point.
    As regards decisions (i.e. actions), more or less the same applies. You take certain data as given, draw some conclusions, and act accordingly. You may be disastrously wrong in taking that course of action, of course, because you never (by definition) know all the necessary information, part of which refers to the unknown future (and other part, although belongs to the past, is too costly to acquire in the time allowed, e.g. all the possible moves in a game of chess). (By the way, both players in a game of chess are generally quite rational, but nonetheless one of them always fails to win). But a possibly negative outcome does not detract from the rationality of your choice.

    ReplyDelete
  31. Thameron says: " I think people stray from the realm of reason when their desires (and the actions they take to achieve them) are directed towards things which no evidence supports the possibility of - things like entering into an afterlife when you die."

    I disagree. Believing or not in a deity is not an issue for being rational in your behavior towards that supposed deity. Just as advancing towards a hill in the belief that it is full of enemies (when in fact no enemy is there) does not detract from your conducting your troops rationally: good deployment and spreading of troops, prior bombarding of the hill, ordering soldiers to crawl and not present an easy target, and so on. In the end, your troubles were useless because no enemy was there, just as all your Rosary mantras and Sunday Masses may turn out to be useless if there is no afterlife or God. But (as Pascal famously thought) it may be rational to believe in God, just in case (not my opinion, by the way, but as you see this is all a matter of preference, not a matter of rationality in reasoning, decision or action based on your preferences or beliefs).

    ReplyDelete
  32. The Pascal quip, by the way, gives away the silliness of attaching rationality (or irrationality) to belief in something not observable (such as God of the afterlife). "Believing just in case" would be rational in an utilitarian calculating way, but that is not really "believing". It is only "acting as if one believes", e.g. going to church or reciting prayers. The act of believing is not something you can do based on a calculation: either you actually believe, or you don't. If you pray and go to church Pascal-wise, "just in case", then you are not a believer.

    ReplyDelete
  33. Hector,
    May I remind you that in your syllogistic example of how preference and choice will flow, rather than "A>B and B>C imply A>C," you had it that "A must be preferred to C." Which is perhaps how classical deduction would have it, except in your attempt to justify its use, you seem to have inductively switched systems on us. Just saying.

    ReplyDelete
  34. @Hector:
    >Being rational today, with the information you have today, including your probability assessment that you may end up being wrong, does not imply that you could not be wrong in the end and regret your decision...

    Actually, it does. Suppose I drive a friend home, and on the way somebody runs a red light and crashes into my car from the side, killing my friend. I would wish I hadn't driven through that particular intersection at that particular time, but I would not regret the decision (i.e., I would not consider it a wrong decision).

    This holds even if there is some way I could have found out the car was coming; e.g., maybe listening to police radio would have told me that staying away from that street was a good idea. The fact remains that without the benefit of hindsight, I was perfectly correct not to listen to police radio, since it really just isn't worth the slight increase in safety it entails.

    >Besides, let me comment that your definition is entirely about thought. Rationality is also about decision and action.

    I said "applying reason... to further one's terminal values." That is action-oriented.

    >But that has nothing to do with the fact that you decided to engage in (and actually conducted) your extramarital affair rationally or irrationally. Millions of people do such things every day, many without any regret or guilt, and most do it in a rational way, i.e. taking account of their subjective preferences and using the appropriate means to maximize the benefits and minimize the costs.

    Let me clarify: I think that a feeling of guilt or regret is evidence that you have violated your own subjective preferences, which are in the general case really really complicated (expected in an evolved creature). You can keep repeating the spherical-cow theory that people only value immediate gratification, but it is really implausible. In general, our utility functions include others to some extent, and also include preferences about matters of principle like deception.

    ReplyDelete
  35. Baron P, sorry for the ambiguity, but by "A>B" (as is clear from the context) what is meant is not a quantitative difference, as between two numbers, but a preference relationship. If you do not like the > sign you may substitute one that is clearer, such as "pref", as in "A pref B". The kind of relationship implied is preference. Thus, may be move ahead once that ambiguity is removed?

    ReplyDelete
  36. Ian Pollock, your last remark I think misses the point altogether: " I think that a feeling of guilt or regret is evidence that you have violated your own subjective preferences, which are in the general case really really complicated (expected in an evolved creature). You can keep repeating the spherical-cow theory that people only value immediate gratification, but it is really implausible. In general, our utility functions include others to some extent, and also include preferences about matters of principle like deception."

    1. A feeling of regret or guilt may or may not be evidence that I violated my preferences, which are of course "complicated" (not because I am an "evolved creature": all creatures are, from the lowly viruses and bacteria and including all the rest).
    2. I never, but never said (or intended to say) anything that could remotely implied that "people only value immediate gratification". Our utility functions (which are not "things", but mathematical representations of our observed behaviors) usually take into account not only non-immediate gratification but also the gratification of others (e.g. our own children) and may also include matters such as avoiding deception (which in some instances may be preferred, or not, over other alternatives involving deception, since all alternatives may touch on different aspects of our preferences, and normally we express preferences over "baskets" of things to be decided upon as wholes, such an extramaterital dalliance, with all their strings attached). No problem with that. Preferences are GIVEN, they do not matter as regards the rationality of your behaviour based on them. You prefer, e.g. to have a clandestine lover, to not having it, under the given circumstances and given your preferences and values. Now, given that, you may conduct your affair in a rational or irrational manner.

    Since this example is so entangled with moral values and emotions, and the possibility of committing all the follies of love, it is not really a good example for the matter at hand, I think. It is too much prone to confusion and error. If you wish to pursue the discussion, it would be better to limit the examples to more manageable cases, such as (for instance) decisions not involving morality or sexual folly, or those involving morality but not folly. Once the simplest cases are resolved we may add other complications gradually.

    ReplyDelete
  37. Ian, one more thing. Suppose you decide doing A, and later you regret it because you find out that by doing A you have somewhat "violated your values" (e.g. betraying your wife, which you are against doing, even if at the moment you valuated that at a lower "price" than indulging your sexual desires for a lover, i.e. you expressed an order of preference in practice). Perhaps the sexual gratifications wanes with time, and only the regret about your wife remains. That is, I suppose, how people are. But this does not mean that (1) your decision ex ante to engage a lover was not based on your ex ante judgment about the balance of costs and benefits, nor (b) that your conduct of the dalliance was not made with due diligence and rational care about things such as keeping your wife from discovering it, or not spending way too much in your lover instead of your family, or choosing your casual partner according to your sexual preferences. It does not detract, either, from the rationality of acting, afterwards, according to the order of preferences arising after the fact (the sex was not so great after all, and you still love your wife, and ....).
    The problem is that all this babble about inner preferences may be good for armchair philosophizing, but it is probably no use for a practical person (say, an economist of marital behavior such as Gary Becker, or a civil judge adjudicating on a demand of divorce for adultery), just as your regrets at having shot a customer when robbing a bank will not impress the judge at your criminal trial. For this reason, economists have ceased long ago to discuss the matter of subjective weighing of utilities and disutilities, and use instead the notion of revealed preference: at any time t, your actions A(t) are an expression of your preferences at that time, P(t), period. Whether you acted rationally in the pursuit of those preferences is a completely different matter.

    ReplyDelete
  38. Hector, my quarrel was not with the > sign, it was with the way you used "must" in the first example and after my questioning of the certainty required, you seem to have changed the imperative to the probable and replaced the "must" with "imply." Syllogistic inference is not the same as suggestive inference, which was my quarrel with that example to begin with.

    ReplyDelete
  39. Baron P., I used "must" in the sense of what is usually considered to be a consistent ordering of preferences. The two requisites economic theory mentions for such consistent preferences are: (1) non reflexivity (if A is preferred to B, then B is not preferred to A) and transitivity (if A pref B, and B pref C, then A pref C). A number of theorems in economic theory require such properties. If people's preferences do not have such properties, then such theorems are false. This does not say that people "must" (or "have the obligation of") having such preferences, nor even that people actually "do" have them.

    ReplyDelete
  40. >This does not say that people "must" (or "have the obligation of") having such preferences, nor even that people actually "do" have them.<

    No, and of course I should have expected that to be the point, or you would not have had to go to such great lengths to so persuasively convince us of it.

    ReplyDelete
  41. So, Baron, the semantic matter is settled, I hope. I clarified what I meant, and you apparently understood what I meant. That an action is rational does not have to do with fulfilling moral obligations or with implementing righteous values, but only with the application of adequate means to achieve an exogenously given end. Likewise, that a process of thought is rational means that the derivation of conclusions agrees with the rules of logic, regardless of the truth or worthiness of the premisses (or the moral/political/personal/social implications of the conclusions, to add a further point).

    ReplyDelete
  42. Hector, I believe I understand what you meant to mean, which doesn't mean I'm compelled by your logic to agree with it. You've bent those rules a bit to achieve your purpose, and I expect your answer then would be that if so, you bent them rationally.

    My stated differences there as well might bear repeating:
    >Logic is not as mathematically correct as mathematics are logically correct.
    Syllogistic inference is not the same as suggestive inference.<
    There's more, but I expect again these were irrelevancies to you. Differences without the distinction of being rules, perhaps. Or easily classified as unstated and thus irrelevant premises. Thus nothing there to really break.
    But to assert that a conclusion which hadn't breached the bounded rules of logic was therefor rationally arrived at, regardless of the worthiness of the premises involved, is bizarre.
    I'll leave off here with a reference I'm sure you'll recognize:
    Bertrand Russell, then 92, was said to have asserted that the rational had its limitations: "Beware of rational argument," he said; "you need only one false premise in order to prove anything you please by logic."

    ReplyDelete
  43. Baron, I entirely subscribe Bertrand Russell's words. If a premiss is false you will draw false conclusions. Which of course has nothing to do with our argument.
    That is precisely the basis for asserting that "to assert that a conclusion which hadn't breached the bounded rules of logic was therefore rationally arrived at, regardless of the worthiness of the premises involved". That is not certainly bizarre. It is plainly correct. And please look: it is so in whatever sense of the vague word "worthiness" you were thinking. But especially if you meant it to mean that the premises were true. In fact, the only requirement that the premises need for the conclusions to be true, is to be true themselves. They do not need to be "worthy" in any other sense (e.g. morally righteous, or aesthetically beautiful, or economically valuable, or whatever).
    Any textbook of logic will tell you that given some premises, rational reasoning will give you conclusions logically derivable from those premises. Whether the conclusions are actually true would depend on the truth of the premises, of course, but even if the premises are false, this does not mean that the reasoning was faulty. Example:
    Premise 1: Martians are blue-haires.
    Premise 2: You, Baron P., are a Martian.
    Rationally deduced logical conclusion: Therefore you are blue-haired.
    If new information arrives to my desk convincingly telling me that the color of Martians' hair is not blue (or that Martians do not exist, or the color of their hair has not been ascertained as yet), or that you are not actually Martian, of course I shall proceed to change my premises. Once the premises are changed, of course the rationally deducible conclusion will be different. That, in turn, would not mean that the new premises or conclusion are true: it would only mean that IF they are true, THEN the conclusion would also be true. Until new information arrives. And so on.

    And, of course, you may still have blue hair even if you aren't a Martian, and/or even if Martians do not have blue hair. That is a completely different matter: The conclusion may be true even if the premises are false.

    Elementary, my dear Baron.

    ReplyDelete
  44. "Logic is not as mathematically correct as mathematics are logically correct". In fact, logic and mathematics are hard to tell apart, especially after the developments in both in the first 2-3 decades of the 20th century. Both are equally "correct", both mathematically and logically (if anyone can tell the difference) and moreover both are (Godel says) always incomplete in a special way.
    Besides wordplay I do not see any relevant point there, Baron.

    ReplyDelete
  45. "Believing or not in a deity is not an issue for being rational in your behavior towards that supposed deity."

    I think that believing in something that you have never seen, heard, touched, tasted or smelled; something which no instrument or experiment can detect is inherently irrational as is any similarly imperceptible goal based on that premise - receiving 40 virgins/raisins for all eternity in the afterlife or what have you. The desire for such a belief may be rational (or at least explicable) given human brain wiring and cultural indoctrination, but the belief itself is not.

    Your example of the enemies on the hill does not hold. Enemies are just other human beings (perfectly perceptible) with or without weapons (also perceptible) who want to kill you. Nothing fantastic there. A god on the other hand is generally an utterly imperceptible being assumed to have a certain set of personality characteristics and abilities. These two assumptions are not really comparable.

    ReplyDelete
  46. @Michael (Since, as Massimo pointed out and I now realize):
    This sounds, at least vaguely, analogous to the distinction between the coherence (internal rationality) and correspondence (external rationality) theories of truth.

    And, I'm about as likely to use "rational" for such internal rationality in an example of the nature you cite as I am to use "truth" for analogous examples supporting a coherence theory of truth.

    So, it's an interesting discussion, but, not the best use of the English language.

    ===

    @Hector: True that we can reach all sorts of bad/wrong conclusions while rationally reasoning (apologies for what has a redundant sound to it) while acting on incorrect data. But, that's why, of course, in logic, a valid argument must have both valid reasoning and true warrants. This gets back to why I don't like Michael's use of "interior rationality" in the first place. And so ...

    @All
    @Michael
    What this really comes off to me is a problem in philosophy of mind, theory of mind, and related issues such as Dennett's "intentional stance" as well as a bit of Lewis Carroll's Humpty Dumpty.

    To use a phrase such as "interior rationality" leaves open the connotation that different people can define rationality differently and that we aren't in a position to object when they do so.

    Or, to riff on Dennett again (not that I actually agree with him a whole lot), Michael's "interior rationality" risks coming off a whole lot like his "folk psychology."

    For people who don't apply some degree of skeptical self-examination, I don't think "interior rationality" is as rational as you might hope it is, Michael.

    ReplyDelete
  47. Thameron, for your information I also think detities do not exist. But we are not talking about that here. People believing in deities may base their belief on a number of thought processes (e.g. St Thomas Aquinas 'proofs') and all sorts of empirical data (the wonders of Nature requiring a designer), or simply (and most likely for the majority) on inner feelings without much influence from reason or data. They may also believe in other less supernatural things (dragons in the seas, enemies on a hill, prophesies by some reputed prophet, and so on). Casting judgment on the 'rationality' of such beliefs is not a very promising venue, I surmise.
    However, no matter what, once you accept that some people believe certain things, or prefer certain things over others, you can start thinking, on firmer ground, about what the rational behavior (or derived beliefs) of such people should be. Praying to the (supposedly existing) deity seems quite a rational thing to do, like bombarding the hill just in case enemies are there is too. As Pascal put it, you can decide you better start praying just in case God exists.

    In any case, GIVEN the premises, rationality concerns the derivation of conclusions, not the truth of the premises. Even if turn your attention on the rationality of the premises themselves (are they grounded on evidence? Are they internally consistent? etc), you still need OTHER premises (axioms, assumptions) to proceed in the task of evaluating those premises (in this case, premises about deities). Then those assumptions themselves would need to be justified, and then you are in infinite regress, ultimately resting on unproven axioms. (And then Godel steps in to inform you that even with your axioms you cannot prove everything, including some propositions that are known to be true, but that's another story).
    I started my participation in this thread with examples from economics, where inner states of mind or feeling or preference are taken as given, restricting the analysis of rationality to evaluations and decisions made on the basis of those subjective states. I still think that is the better way to proceed. About inner preferences, you can only demand a degree of coherence (such as non reflexivity and transitivity) but little more.

    ReplyDelete
  48. Hector, If you use premises that you don't believe are in some fashion accurate, and call the process logical, you may be rational, but likely for the purposes of accomplishing deceit. And if I'm not mistaken, the historical purpose of the logical and rational has been predictive, as even a liar has a predictive, albeit ultimately irrational, purpose. And ask yourself what all predictions are meant to have in common.

    ReplyDelete
  49. I say "Logic is not as mathematically correct as mathematics are logically correct" and Hector says "Both are equally "correct", both mathematically and logically."
    "Besides wordplay I do not see any relevant point there, Baron."
    And I say again that the correctness of logic depends upon the correctness of the logician's premises, and the correctness of mathematical results is dependent on the correctness of those furnished by the logician become mathematician.
    In either case the premises must be accurate to accomplish either the logician or the mathematician's predictive purposes. The difference is that if the premises are wrong the operation fails to accurately predict, and the fault lies with the logician-cum-mathematician rather than the mathematics. You've tried to separate logic from its premises to make it the equivalent of mathematics. But there is no logic possible without premises. And ultimately no mathematics.

    ReplyDelete
  50. Baron's words: ", If you use premises that you don't believe are in some fashion accurate, and call the process logical, you may be rational, but likely for the purposes of accomplishing deceit."

    Of course, but that is not the matter. First, the premises are not supposed to be known in advance to be false. You take the premises, whatever they are, and draw conclusions. If the premises turn out to be false, so the conclusions could not be affirmed on such grounds. Second, even if you (as an observer) know that the premises of some other person are false, you can still predict his or her behaviour (e.g. you may predict that this person will buy Bach's flowers or homeopathic preparations in order to cure an illness, since you know he/she believes in the curative powers of such products). The reasoning will be logically correct, and the conclusion will be true (i.e. the prediction will come to fruition) even if the person is mistaken in her/his belief about such "alternative medicine" remedies. By the same token you may start an advertising campaign on some false premises (i.e. suggesting, albeit not actually saying, that purchasing a certain deodorant will get you laid more often) and have success in increasing your deodorant sales (i.e. getting some people to act, purchasing the product on that false promise of sexual success). You do not actually know WHICH persons will fall for it, but you may confidently expect (based on prior tests or experience) that a sufficient number will, thus making the ads economically worthwhile. All this time you would be acting rationally in the pursuit of your goal of selling deodorant, and the poor sods that buy the stuff will be also acting rationally insofar as they gullibly believe the ads' message. They could be later disabused, of course, but that's another story.

    ReplyDelete
  51. That the prediction is actually fulfilled is not a necessary requirement for the prediction to be rational. Every time a stockbroker sells a share, another stockbroker buys the same share at the same agreed price, both acting on rational analysis that predicts the likelihood of a gain as a result of said transaction. One of them (or both) may ultimately be proven wrong, but nonetheless their actions can both be regarded as rational.
    Please introduce time, and thus uncertainty about the future, in your arguments in this regard.

    ReplyDelete
  52. Hector, the point was and is that logic has a predictive purpose. Which is to attain a rational result. Both premises and conclusions are predictive, one meant to perfect the other. Try mathematically to prove that you can effectively use one without the other and see what happens. Predictively speaking that is.

    And since you asked, I'll point out what should be obvious, that time and uncertainty about the future have made the use of predictive logic a necessity.

    ReplyDelete
  53. " you may be rational, but likely for the purposes of accomplishing deceit". Indeed. Precisely my point. You can rationally do good, or rationally accomplish evil. Morality of goals has nothing to do with rationality in the use of means.

    ReplyDelete
  54. Baron, precisely because the future is uncertain (and information limited, even about the past), one can legitimately assume a premise, in the belief that it is true, only to find out later that it was false. In the meantime, one was able to draw logical conclusions from such premise, in a completely rational manner.
    Of course, if I were Hegel, I would think that Reason, an entity existing outside me in the realm of the Spirit, cannot be satisfied because Reason, being such, would be aware that my premise was false. Unfortunately, I do not believe in such entities as the Spirit or Reason, astute as they might be. In the real world, we have only us, imperfect human beings applying logical procedures to available information.

    ReplyDelete
  55. I'd say that morality of goals has everything to do with rationality in the use of means, since it goes to the process by which you hope to accomplish your purposes. Hopes that in turn depend on the accuracy of your presumptive premises.

    Hector, we can go round and round on this all day, and predictably get nowhere. You can't construct a thesis that the rational process can achieve a purpose without a premise when that thesis is in itself an argument based on that premise.

    ReplyDelete
  56. As ypu say, Baron P, we both seem incapable of seeing each other's point. I would let other readers judge and (as least for my part) end the discussion here.

    ReplyDelete
  57. "Reason is the greatest enemy faith has worship 830 1045."

    All this fuss over Michael's argument and not one of you noticed the cult of Pythagoras has taken over a First Baptist Church.

    ReplyDelete
  58. Hector, I'll end it here as well, and let that last little unworthy snipe of yours go unanswered.

    ReplyDelete
  59. Hey all,

    Sorry for the late response. I took a couple weeks off from the blogosphere to focus on some professional duties. I'm just getting back into the swing of things now. Thank you all for the responses.

    However, I would like to answer a common misconception about my article. Many here seem to think that interior rationality is merely the human process of rationalizing. I was actually proposing two different systems of rationality. One is internal (within a system based on the prior assumptions) and one is external (a sort of bird's eye view of the system).

    In the internal system of rationality, it makes sense to act (or predict one will act) in a certain way given baser assumptions. Again, this is not about post-hoc rationalization. The external system of rationality is a more critical system of questioning all of our assumptions to see if we are really acting rationally. Perhaps it is the difference between what we think is true and what is actually true, but it need not be so, for interior rationality is not automatically irrational.

    Hope this helps.

    ReplyDelete
  60. Michael, it helps. Some questions remain, nonetheless.
    1. "Internal" rationality, you say, is based on "assumptions". I think one should distinguish between "assumptions about the real world" and "preferences among alternative choices". The former can be wrong or right (e.g. one may wrongly believe that a rain-making dance around a tribal fire may cause rainfall). The latter are not so questionable: I say potahto, you say potayto, you like beef and I don't, you prefer fishing and I prefer hiking, and so on. Everyone has tastes and preferences and priorities, and usually everyone else cannot tell whether they are "rational". Rationality is usually defined in terms of using adequate means to achieve an end, the end being "practical" ends (such as obtaining food) or "values" (such as "doing good"), hence Max Weber distinction between rational action oriented to goals and rational actions oriented by values. But either goals or values are not themselves subject to "rationality" characterization (except in the limited sense that the most common models of "rational choice" or "rational action" in Economics require that preferences be non-reflexive and transitive: this is not a requirement to make the choice more rational, but for the models to work in a consistent manner, i.e. producing consistent responses to a given stimulus).Conventional economic theory also uses to require that preferences are relatively stable, and treats them as given, but again this is just a simplification for short-term models, since everyone agrees that preferences may change over time, given sufficient time, and entire branches of industry are devoted to change your preferences, e.g. through advertising).

    ReplyDelete
  61. Michael, I repost my comment in a better form that (by mistake) I posted at your more recent post ("Refute or promote").

    Your clarification helps. Some questions remain, nonetheless.

    1. "Internal" rationality, you say, is based on "assumptions". I think one should distinguish between "assumptions about the real world" and "preferences among alternative choices". The former can be wrong or right (e.g. one may wrongly believe that a rain-making dance around a tribal fire may cause rainfall). The latter are not so questionable: I say potahto, you say potayto, you like beef and I don't, you prefer fishing and I prefer hiking, I love this particular girl and you don't, and so on. Everyone has tastes and preferences and priorities, and usually everyone else cannot tell whether they are "rational", and I'm not sure that it matters at all. If you actually believe that your rain dancing causes rainfall, then dancing is the (internally) rational thing to do if your tribe faces drought, although some foreign witches may tell you that it is useless, and call this belief of theirs "externally rational" and "objectively true". Foreign witches are weird, aren't they?

    2. Rationality is usually defined in terms of using adequate means to achieve certain ends, the ends being "practical" goals (such as obtaining food) or "values" (such as "doing good" or "proclaiming your faith"), hence Max Weber distinction between rational action oriented to goals and rational actions oriented by values. But either goals or values are not themselves subject to "rationality" characterization (except in the limited sense that the most common models of "rational choice" or "rational action", especially in Economics, require that preferences be non-reflexive and transitive: this is not a requirement to make the choice more rational, but for the models to work in a consistent manner, i.e. producing consistent responses to a given stimulus, so that an external observer may say that they "make sense" in terms of your beliefs and preferences.
    Conventional economic theory also uses to require that preferences are relatively stable, and treats them as given, but again this is just a simplification for short-term models, since everyone agrees that preferences may change over time, given sufficient time, and entire branches of industry are devoted to change your preferences, e.g. through advertising).

    ReplyDelete
  62. Prof. Pigliucci,

    I came across your article on interior and exterior reality. I was wondering if you could direct me to literature that uses this concept in respect to the choices agents make. Preferably texts which offer an introduction (or a gentle application) of choice theory or game theory.

    I'd be grateful for any advice.

    ReplyDelete

Note: Only a member of this blog may post a comment.