Home > Regular Features > Word for Wednesday: Rational (no definition offered)

Word for Wednesday: Rational (no definition offered)

June 11th, 2003

The term ‘rational’ and its variants (rationality, rationalism) are used in a lot of contexts in economic debate, both positively and negatively, but nearly always sloppily or dishonestly. A specimen I’ve seen on more occasions than I can count is the line (usually presented with a sense of witty originality) ‘if you are opposed to economic rationalism, you must be in favor of economic irrationalism’.

In keeping with the idea of this regular feature, I thought about providing a definition that would clarify the issues surrounding this word and the reasons it causes so much confusion. In reflecting on the problem, however, I’ve come to the conclusion that the word ‘rational’ has no meaning that cannot better be conveyed by some alternative term and that the best advice is probably to avoid it altogether.

The basic problems surround the kind of use that is standard in economics and related discipline, in which ‘rational’ choices are those that maximise the value of some objective function. A lot of energy has been dissipated on disputes over whether this is a normatively compelling or descriptively accurate, or whether some alternative such as ‘satisficing’ would do better.

Rather than taking sides in this dispute, I will offer the following purely mathematical claim. Given any data on any observed set of problems involving the selection of one or more choices from a set of alternatives, the observed choices can be represented as the maximisation of an appropriately specified function. To give an easy example, satisficing can be represented (rationalised) as optimising, taking calculation costs into account, or alternatively as a combination of set-valued maximisation with a selection rule based on the order in which alternatives are presented.

If this claim is accepted, it’s evident that the definition of rational choices as those that maximise an objective function is empty, since all choices satisfy this criterion.
Most uses of the term rational and the opposed ‘irrational’ involve some confused mix of the following connotations

  1. reasonable as opposed to emotional
  2. calculating as opposed to intuitive
  3. self-interested as opposed to altruistic
  4. materialistic as opposed to non-materialistic
  5. logically consistent as opposed to inconsistent

Of these points, the last requires some further explanation. Various consistency properties have been proposed as requirements for rationality. The one that is most obviously reasonable, though not invariably compelling, is transitivity. If I prefer A to B, and B to C, I should prefer A to C.

The problem with debates involving the word ‘rationality’ is that people tend to shift from one meaning to another, sometimes deliberately and sometimes without realising what they are doing. Given the entrenched nature of all five uses listed above, and the tangled relationships between them, it’s impossible to specify a ‘right’ meaning. The best option is probably to avoid the word altogether, and to use the specific terms I’ve suggested as appropriate to its various components.

Categories: Regular Features Tags:
  1. June 11th, 2003 at 18:26 | #1


    A specimen I’ve seen on more occasions than I can count is the line (usually presented with a sense of witty originality) ‘if you are opposed to economic rationalism, you must be in favor of economic irrationalism’.

    Reminds me of a certain Sydney philosopher who asserted that if you didn’t accept his notion of rationality you were eo ipso an irrationalist.

  2. cs
    June 11th, 2003 at 19:37 | #2

    Interesting. Because of the alleged antecedents of a certain Sydney philosopher, I’ve been dipping back into Hobbes over coffee, and he seems to have used ‘reason’ consistently with 2. & 5, while denying the possibility of “right reason constituted by nature” (which, he concluded, means that disputes must go to agreed judges). The most relevant passage goes:

    …and when men that think themselves wiser than all others, clamour and demand right reason for judge; yet seek no more, but that things should be determined, by no other men’s reason but their own, it is as intolerable in a society of men, as it is in play after a trump is turned, to use the trump on every occasion … revealing their want of right reason, by the claim they lay to it.

  3. June 11th, 2003 at 21:51 | #3

    definition 6

    rational simply means an action which does not result in an outcome that conflicts with higher goals.

    irrational means there is a conflict. for example, i want a job at a bank, it would be irrational to wear jeans to the interview.

    irrespective of whether i know what the best strategy is. a friend can inform me my wearing jeans is irrational, and it makes sense.

    (i spent some significant time in a course on rationality definitions and most came up empty, not conflicting with higher goals is the most meaningful term. and i think its a fair synonym for 1 (reasonable versus emotional)

  4. James Farrell
    June 12th, 2003 at 00:42 | #4

    As far as the concept is concerned, I think definitions 1-6 clear things up nicely. But we are left with the problem of how to use the term in public policy debate.

    My colleague Tom Valentine is happy to be called an economic rationalist. When he gave his inaugural professorial lecture some years ago, however, he said he preferred the term ‘rational economics’ to ‘economic rationalism’. This he contrasted with what he called the ‘entitlement consensus’, and quoted instances of various public intellectuals proposing policies that took no account of costs. His favourite example was an unnamed woman who said seriously, ‘I think everything should be subsidised…’ Of course I disagree Valentine on a lot of things, but we all know there are people like her, and it would be handy from time to time to have a phrase for what they lack.

    Unfortunately this opportunity has been missed, and the term has become inseparable from anti-interventionist ideology. It seems to have entered the Australian discourse around the mid eighties, and I have a have a feeling that it may have been John Stone who popularised it, though I may be wrong. Used by the likes of Stone it is a put-down for anyone who won’t accept their logic. Meanwhile on the left it is a pejorative, connoting a callous obsession with efficiency or a naive faith in markets, or both. A Marxist colleague mainly uses it to characterise any sort of analysis that precludes altruism as a significant motivation for behaviour.

    If ‘economic rationalism’ is going to be, like ‘political correctness’, just a device for posturing and name-calling, we might as well do without it. But I think it can still be usefully and dispassionately employed to denote a general doctrine. In an Economics Society lecture in 1993 John Nevile defined it as the belief that the market is the only legitimate allocator of resources. This is neither too vague nor too constraining and it’s something we need a phrase for. Libertarian is too ambiguous and not restricted to the economic sphere, while laissez-faire has too much nineteenth century baggage. So although it’s annoying that subscribers to this doctrine have been allowed to appropriate the word rational, I think we’ll have to live with it.

  5. June 12th, 2003 at 00:42 | #5

    Isn’t that same constructing of terms at the bottom of “social democracy” too? That is, isn’t it a mere association of words without any NECESSARY and INHERENT corresponding connection of underlying concepts? In either case, if the listener is willing to accept the basis of discussion sight unseen, the inferences follow from the premises almost tautologously. It is up to the listener not to accept that a particular thing is (necessarily, etc.) both rational and in the economic sphere, just as it is up to the listener not to accept that a particular thing is both social (with an implied “-ist”) and democratic.

    If you don’t keep things rigorous you risk falling foul of “no cat has eight tails, every cat has one more tail than no cats, therefore every cat has nine tails”.

  6. June 12th, 2003 at 01:16 | #6

    There isn’t any special problem with the word rational unless it is used as a buzzword for something which is always good. By all of your five criteria Genghis Khan was not only rational, but probably more rational than any leader contemporary to him. (This is not a throwaway line, I can argue it at lenghth). We may disagree about his goals, but people who look at his methods find them rational.

    I have always thought of rational as calculation of ends/means, cost/benefit, input/output, all expressible as ratios. “Rational organization” is centralized organization organized around some standardized ends/means decision. Alternatives are traditional organization (including custom, religious organization and law) and negotiated settlements or compromises (which tend to be messy and not rational).

    In arguments where “rational” is used as a buzzword, I usually see shifting definitions opportunistically deployed, with “rational” meaning some specific form of organization at one point in the argument, but at other key points simply means “organization which is not stupid and insane”.

    When a rational alternative is proposed to a political compromise or traditional legal or customary arrangement, the rationaist always seems smarter as long as he controls the terms of the argument. But rationality always comes in specific forms, and if the rationaist has assigned his ends and means badly the rational form is not superior.

  7. Shai
    June 12th, 2003 at 03:54 | #7

    One day, on the news, a reporter was interviewing a professor of statistics on the topic of an impending powerball lottery. “Buying lottery tickets is irrational”, he said. “The expected payout is much, much lower than the cost of a ticket. You’re much more likely to die on the drive over here than have the winning numbers.” The reporter then asked him if he had bought a ticket. “Yes, but if you don’t buy a ticket you can’t win” he replied, looking somewhat embarrassed.

    Saw that exchange on CNN last year, and thought it was cute.

  8. June 12th, 2003 at 06:46 | #8

    As regards c8to’s definition 6 above, that meaning is quite well covered by the term “practical”, I see no reason to use “rational” in this context.

  9. Tim Dymond
    June 12th, 2003 at 11:41 | #9

    Number 3 ‘self-interested as opposed to altruistic’ gets you into a means/ends dilemma. Sharing my wealth may appear ‘altruistic’, and keeping it ‘self-interested’ – however if my relative wealth to other people’s poverty breeds resentment in and isolation by those other people, might it not be in my ‘self-interest’ to engage in a little redistribution in order to provide a longer term safer environment in which to enjoy my remaining wealth?

  10. June 12th, 2003 at 12:25 | #10

    practical means implementable.

    i may want to take over the world. it might be impractical. acquiring nuclear weapons as a first step, while impractical, would be rational.

    rational has meaning within this framework:

    1) you have a hierachy of goals.
    2) an action is rational if it doesnt conflict with higher goals, (and perhaps helps you achieve one of them, but that goes without saying)

    another example. i want to have sex. if a higher goal is to not have children, it would be rational to wear a condom. if i want to have children, it would be irrational.

    (this is irrespective of my understanding of conception. lets assume there is a guy who doesnt understand the link between sex and conception, but doesnt want kids. we could still point out to him the link, and then assert that its rational to use contraceptives)

    (apologies to sensitive readers for the lewdness, but im getting away from quiggin’s violence in philosophy)

  11. June 12th, 2003 at 15:30 | #11

    A specimen I’ve seen on more occasions than I can count is the line (usually presented with a sense of witty originality) ‘if you are opposed to economic rationalism, you must be in favor of economic irrationalism’.

    Funny, I’ve often heard (usually presented with a sense of witty originality) “economic rationalism isn’t” or something similar.

    I’ve never really considered what ‘rational’ means outside an economic context – but with regards to ‘economic rationalism’ I have concluded that it is an insult and nothing more. It is like ‘asshole’ – which is meant simply to insult, and not to literally imply that a particular human has turned into an anus. Used in it’s context as an insult, ‘economic rationalism’ is defined as believing in bad things… and the user of the words generally goes on to prove that it is bad – which (given the strawman definition given) essentially is an excercise of proving that bad things are bad.

    And this is how Clive Hamilton and Lindy Edwards earn their money…

  12. Me No No
    June 13th, 2003 at 15:27 | #12

    Like J Farrell, I have often thought that “political correctness” and “economic rationalism” are conceptually similar, that is, pretty empty and meaningless terms of abuse that carry a kind of implied superciliousness and superiority on behalf of the name-caller, and caricature the callee.

    They’re used by different sorts of people, of course, although you could say that Hansonites would employ both terms. In fact they might say that “economic rationalism” is “politically correct”.

    I think it was Michael Pusey who coined ER.

  13. June 14th, 2003 at 00:27 | #13

    The British military historian John Keegan described Adam Smith and his collegues as ‘the economic rationalists of the 18th century’ in a book written in the early 1990s so the phrase definately took hold in the UK.

  14. Andy
    June 18th, 2003 at 20:56 | #14

    it’s evident that the definition of rational choices as those that maximise an objective function is empty, since all choices satisfy this criterion

    That’s not correct.

    Choices which suggest non-transitive preferences don’t. This is not logically inconsistent either – for example, when the choices are based on majority votes.

  15. John
    June 18th, 2003 at 21:13 | #15

    Andy, in the non-transitive case, the choice isn’t to pick an optimal element from a large set, but to make a sequence of restricted choice sets. For example, in a voting situation, the agenda typically produces a sequences of pairwise choices. So you set up the domain of the maximisation function so that all preferences are dependent on the restricted choice set, and this can be represented as maximisation of an appopriately defined function. Regret theory in choice under uncertainty is an example (actually, it’s a minimisation, but this difference is trivial).

  16. Andy
    June 19th, 2003 at 01:11 | #16


    So, in other words, it’s still not true that “all choices satisfy this criterion”. Sometimes you have to restrict the available choices to a subset to get it to work.

Comments are closed.