Discussion of game theory inevitably brings up the question of whether game theory relies on an assumption of rational behavior, and, if so, whether this is a weakness or a strength. Rather than respond, I thought I’d dig up this old post from my long-abandoned (but still planned-to-be-revived-one-day) Word for Wednesday series.
Shorter JQ: the word ‘rational’ has no meaning that cannot better be conveyed by some alternative term. Avoid it.
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
- reasonable as opposed to emotional
- calculating as opposed to intuitive
- self-interested as opposed to altruistic
- materialistic as opposed to non-materialistic
- 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.
John Quiggin 
Wasnt economic rationalism coined for political purposes? The Labor folks couldnt call it by its normal name, “economic liberalism” because that was the name of their political opponents.
Ludwig Von Mises (from the Austrian school of economic thought) put a lot of energy into this topic. In essence his view was that if we are respectful of people we need to accept that the choices they make are rational within the framework of subjective values and perceptions that they hold. Nobody can make a more rational choice for them.
This position seems to accept that a person may value their emotions above their intellect, be intuitive rather than calculating, altruistic rather than self interested, non-materialistic or even illogical. None of these take away from the fact that people are better positioned than anybody or anything else to optimise their own choices.
I regard myself as a libertarian which means I think people should be free to make their own choices in life so long as those choices do not involve unprovoked violence against others. I am often regarded as an being an “economic rationalist”. The critics of economic rationalism are typically dismissive of it because they believe that people are frequently not rational. This to me shows a certain prejudice about what is rational. The presume that some system other than freedom can lead to better decision making because they presuppose that some choices are objectively better than others. However our lives are subjective not objective and nobody can live your life for you. I believe in economic rationalism not because I think people are calculating and unemotive or even that they should be but simply because I believe in the virtues of live and let live.
Probably more useful than a discussion about rational choices versus irrational choices is a discussion about functional behaviour and disfunctional behaviour.
Functional behaviour is behaviour that generally leads us to results we desire. Disfunctional behaviour leads us to results we don’t desire. So if you really want to be friends with somebody and you always insult them they you are probably behaving in a manner that is disfunctional.
I have a friend that smokes casually. He supports bans on smoking and hight taxes on cigarettes because he sees his own behaviour as being disfunctional. He feels that my liberal attitudes would damage him because I would set free his disfunctional tendancies.
My own view is that all behaviour is essentially driven by one of two independent psychological forces:-
a) Our beliefs. eg what is good, what is bad?
b) Our neural assocations. eg what is pleasurable, what is painful?
Disfunction arises due to mixed signals in the brain. If we want to be loved but we find pain in opening up to people then we may not achieve our aim of being loved. If we want to be rich but we think it is evil to be rich then we will stay poor. What we believe and what we associate with pain or pleasure dictates how we act, but only in so far as this mental states are adequately aligned.
Learned helplessness is a particularly harsh form of disfunction.
All successful advertising seeks to influence our beliefs (will those IR reforms be good or bad) or our emotions (if I drive a Toyota then “oh what a feeling”).
Personally I “believe” that it is disfuctional to trust that governments can make our life choices for us more effectively than we can ourselves. And because of my belief I act accordingly.
This seems entirely reasonable, but it doesn’t say much. Rational behaviour requires a rational choice of utility function, as well as rational reasoning about the optionas once the function has been chosen.
Maybe the problem with the word rational is that it is only used to describe the second behaviour, whereas the utility function is taken as given. For example, “economic rationalism” _assumes_ a utility function of maximizing GDP or wealth (or something like that – I am not an economist), and then rational or irrational behaviour is with respect to that function. That annoys everyone who doesn’t believe in the utility function.
However, I believe there are rational ways to arrive at a utility function, but it requires a fair bit of introspection. For example, in my case, when discussing life decisions with my wife, I try to get to the root of what really makes us happy or unhappy – that’s building the utility function if you like – and then look for strategies to optimize our joint happiness, ie optimize the utility.
That’s how I see some therapy – as a mechanism for rationally discovering your own utility function.
“None of these take away from the fact that people are better positioned than anybody or anything else to optimize their own choices.”
If this were true, we should support a wide array of behaviors that are now sanctioned up to thing like like free markets for things like addictive drugs, guns, and small tactical nuclear weapons.
Society has always been run on the basis that people need a certain amount of protecting from themselves. I’m not allowed to become a junkie. If I want to kill my neighbor, the state will attempt to stop me. The usual argument used to support this is based on competing rights, but a truly libertarian approach to this would be “do the crime do the time”. This is a line I have actually heard US lawyers use. It sounded strange.
While we might imagine ourselves to be rational beings, I tend to see rationality as the tip of the iceberg. It’s a view based in psychology and biology. What lies beneath surface, partially visible, is a mish-mash of competing desires, conflicting impulses, rationalizations, subtle and not so subtle addictions. Lower brain functions, if you like.
We never really conquer this stuff, we just (hopefully) reach some kind of happy working relationship with it, enjoying it’s thrills and pleasures but keeping it’s destructive tendencies in check. There’s always risks. In families, communities and societies – even in libertarian society meetings – we continuously monitor ourselves and others and act to curb excesses.
For what its worth I think that people who want to do drugs or buy guns should be allowed to (and the market in these things should in general be mostly free). So long as people don’t initiate unprovoked violence towards others they should be allowed to buy and sell almost anything. Laws that supposedly protect people from themselves are generally a menace. These laws just make criminals out of ordinary people.
I am happy for small tactical nuclear weapons to be prohibited so long as the government is also prohibited from having them. And obviously I don’t think people should be free to buy and sell other people.
If we are going to abolish words which have more than one meaning, or are capable of being used dishonestly, then we have to abolish more than “rational”. We also have to abolish “love” and perhaps “honestly” itself.
Certainly it is always important to ask “rational for whom?” and “rational when?”, and perhaps “rational how?” and “rational at what cost, and to whom?”. It is often necesary to explore premises and beliefs in order to understand what “rational” means in a particular context, and to understand that what is rational for me may not be rational for you.
These clarifications are the outcomes of debate. To achieve them is largely why debate, particularly policy debate, is important. I wonder whether an attempt to abolish “rational” implies a desire to also abolish debate. I don’t for a moment believe that any process of substitution of other words for “rational” is going to enable some kind of mystical one-off argumentative coup which would render debate superfluous, though many people (particularly in the public relations game) would like to believe it possible. Because there are many different sorts of people in the world, all with their own ideas about what is right and what is good for them and for others, any such coup will inevitably turn out to be a coup de theatre – which is quite good enough for the aforesaid PR crew, by the way!
As I have commented before, people who are worried about rationality and perhaps the word itself would do themselves a favour by reading JRSaul’s “Voltaire’s Bastards”. Infuriating as the book sometimes is (discursive, sometimes repetitive and unnecessarily long), it is worth the time, particularly for economists.
Do I take it, Terje, that you think people should have personal nuclear weapons if the government has them? That is, every American is entitled to have as many nuclear weapons as he or she likes?
Mike,
I think the US government should not have nuclear weapons.
Regards,
Terje.
Once again a libertarian troll steers a discussion (of the word rational) to his preferred looney discourse.
Steering is a simple and deniable tactic: simply lard your argument with enough unnecessary mentions of ridiculous libertarian hobbyhorses, and somebody will pick up on it.
I’ve been sick of that sort of libertarian evangelism for decades now. If you are too, you may want to check my Critiques Of Libertarianism web site.
A few years ago, I wrote a small article called Skepticism Of Rationality. Here’s my favorite bit:
Usually, “rationality” is a Humpty-Dumpty word, that means whatever the user finds is convenient to club his oponents with. Either by defining rationality to specifically disclude an opponent’s methods (as objectivists do) or by making rationality so diffuse that all the users methods are included (much the same way new-agers use the term “energy”.)
Pop quiz time: Which of these would you call rationality? Deduction, induction, abduction, science, experimentation, heuristics, tradition, faith, imitation, guesswork, dreaming, tripping, learned reflex, unlearned reflex, tropism, metabolism? The list can go on very extensively.
Read the article for my opinion. 🙂
Pious of you, Terje, to say that “I think the US government should not have nuclear weapons” but that wasn’t under discussion. Not at all. The topic was freedom.
I asked whether, given that the govt does have them, you think every single person has a right to them (as you had implied). You have chosen to avoid answering. I guess you have figured out that answering it is going to put you in an impossible position, logically.
Could it be your whole “freedom” philosophy is actually illogical, ie irrational?
test
testing
“None of these take away from the fact that people are better positioned than anybody or anything else to optimise their own choices.”
Not so, I think. If I buy new clothes (rarely so I’m never going to get much practice) I’d rather trust my wife’s final judgement than my own. I only see myself in odd angles in the mirror, whereas she sees me directly from all angles quite often. She’d have a much better idea of what I look like.
Questions for John Quiggin:
1. Why is the word “should” in the sentence “If I prefer A to B, and B to C, I should prefer A to C”? The definition of transitivity (used in game theory, general equilibrium theory, decision theory) does not include a “should”.
2. Statements:
a. “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.�
b. “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.�
Questions:
c. What do you mean by saying “if this claim is accepted� (Would it be helpful to present the exact theorem).
d. Why does (b) follow from (a)?
Questions for Terje Peterson:
1. Which part of the anatomy is being used by a person about whom it is said that he or she “may value their emotions above their intellect”? (I tried to make such a choice by means of a thought experiment – I can’t do it.)
2. Why do the names von Mises (and van Hayek) pop up about 30 years after Mount and Reiter have made precise van Hayek’s writings on the role of prices to coordinate decentralised decisions and about 20 years after it has been shown that the equilibria (solutions) of theoretical models of ‘competitive private ownership economies’ with incomplete markets are generically inefficient? These results do not mean that therefore a central planner must make all decisions for everybody. In the first instance, it means that repeating phrases such as ‘freedom of choice is good’ and ‘government is bad’ is not helpful because these models assume that people have ‘freedom of choice’ and people are assumed to be capable of making such choices (ie rationality is defined such that it excludes the behaviour of the character in the comedy Open All Hours, who comes to the shop but can’t make up her mind on what she wants) and there is no government. The interesting question is: Are markets incomplete? This is an empirical question. The answer is YES (at least for the foreseeable future). I am using the term ‘models of non-dictatorial resource allocation’ for the class of models which assume ‘freedom of choice’ and ‘rationality’ in the sense indicated above. There are many more examples one could give where the solutions to models of non-dictatorial resource allocation systems do not correspond to the solutions the ‘economic rationalists’ want to believe in. One of the problem areas arises from the legal artefact of a ‘corporation’. Neither von Mises nor van Hayek is able to tell what they think in the year 2005. I can’t see a purpose in speculating about what they might say. Quoting from von Mises and from van Hayek’s writings does not help either because they did not ask the questions which have been of interest for at least 20 years and which remain of interest. Hence my question.
3. No matter what criticisms can be advanced against the writings of von Mises and van Hayek, it is my personal opinion that neither of them deserves to be used as a springboard for ‘moving forward to the past’. But this has been happening. Attempts have been made for quite a while to reintroduce the economic aspects of 18th and 19th libertarianism by means of ‘talking a lot’ about von Mises and van Hayek. This amounts to asking people to ‘dumb down’ in the sense of discarding about 100 years of work done by quite impressive brains and to try to persuade people in general to be blind to the obvious. Would Terje Peterson concede that people reject economic rationalism because they know what is best for them?
Mike H is correct in that it seems that my reference to Libertarianism has somewhat hijacked the thread. My apology. Lets try and get it back on track.
To Mike P, I concede that people should not own personal nuclear weapons, especially suicidal bombers. The government should have a general monopoly on violence. This does not mean that arguments about freedom are all flawed and/or illogical/irrational.
Ernestein:-
1. my guess is that it is our emotions that decide whether to value our emotions more or less than our intellect. Of course both are a product of the mind so the origin is the brain.
2. The names of von Mises and van Hayek never went away so I think the premise of your question is flawed. Thatcher used Hayek as the inspiration for many of her reforms at that was a few decades back now.
3. People reject economic rationalism for a multitude of reasons. Given my inclination towards freedom I expect many people to make choices and conclusions differently to me. I would guess it is usually because they don’t like the real or perceived result of economic rationalism. Although I am sure many people reject it because they see it as untruthful or some such thing.
My understanding of the term “Economic Rationalism” is that it does not originate from the notion that Economics should be rational (an meaningless axiom). It originates from notions about individual economic agents being rational in their approach to making choices. Rationalism
http://en.wikipedia.org/wiki/Homo_economicus
http://en.wikipedia.org/wiki/Economic_rationalism
I too have struggled with use of the term ‘rational’ (and ‘bounded rationality’ – don’t know what you think about this one John) while studying whether there are any consuemr policy implications thrown up by the behavioural economics literature – particularly time inconsistency of preferences. I liked the following quote from “Dieting as an Exercise in Behavioural Economicsc“, by Herman and Polivy, ch. 16 in “Time and Decision’ – Economic and Psychological Perspectives on Intertemporal Choice” (2003) by Loewenstein, Read and Baumeister (eds):
“Laboratory research has focused on factors that interfere with dieters’ resolve to diet. Somehow, it seems to be more interesting to study dieters’ failures than their successes, but what is the ultimate goal of such studies? For the most part, both the public and granting agencies assume that our goal is to learn enough about the factors that make diets fail so that we can construct better diets, or better dieters. By the same token, much of the literature on intertemporal choice appears to be concerned with why people act imprudently. The assumption is that the choice made in the cool light of reason (usually some sort of abstinence) is the correct choice, and that the choice made in the heat of the moment (usually some form of indulgence) is the erroneous choice. As far as dieting is concerned, however, we are not convinced that dieting is a particularly worthwhile goal; we are happy to point out the difficulties and dangers of dieting … Our research on dieting is not intended to make dieting more effective; more often we are interested in highlighting the difficulties of dieting, to suggest that these very difficulties are “nature’s wayâ€? of indicating that dieting may not be worth pursuing. Certainly in evolutionary terms, abstaining from available, high-calorie food has little or no survival value. More generally, we are interested in the costs of behaving ‘prudently’. Is it possible that the imprudence – the focus of the intertemporal choice literature – is something we are working too hard to eradicate? From our engaged perspective, if dieting is to serve as a perfect example of anything, perhaps it is a perfect example of how people set out to achieve the impossible, and in failing, reveals that people are not designed to achieve perfection, and that perfection – perfect restraint at least – in fact is not desirable, as our own experience tries to tell usâ€? .
Hmm. Language is tricky, isn’t it?
I think there is a world of difference between economic rationalists and rational economists.
Terje,
Thank you for your reply. I’d like to say a few words on 2 items.
1. You write your “guess is that it is our emotions that decide whether to value our emotions more or less than our intellect. Of course both are a product of the mind so the origin is the brain.�
My guess is that you may be speaking for those people who say that they ‘value’ their emotions more than their intellect.
The concept of rationality in game theory (and general equilibrium theory and choice theory) uses the logic of mathematics. This excludes making a ‘choice between emotions and intellect’ (assuming the distinction between ‘emotions’ and ‘intellect’ is meaningful in this context to begin with) because a (methodological) choice has already been made.
Incidentally, game theory has been quite successfully applied by biologists. The behaviour of fish can be represented quite well by game theory (dynamic). Assuming the only objective of fish is to survive, some ‘little fish’ seem to be bright enough to know that swimming close together is good for them. The ‘shark’ seems to have figured out that swimming alone works. (Lucky biologists, they don’t have to deal with ‘sharks’ who pretend they are ‘little fish’.)
2. Your suggestion that the premise of my question is flawed does not require a reply, particularly given your answer that “Thatcher used Hayek as the inspiration for many of her reforms.�
1. Do you think that we can not make rational choices without mathematical skills? Is it not possible that evolution (or creation) has endowed us with the ability to do mathematical calculations intuitively. I don’t consciously do the maths when I calculate my tragetory in crossing a busy road. In fact if I did attempt to do the calculation consciously instead of intuitively then I would probably get killed.
In the economics of household budgeting I imagine that most people also do things intuitively. I do the groceries with a sence for value and an idea about thrift but I don’t run the calculator over the shopping list.
The fact that fish can do mathematics before deciding their behaviour suggests that its not all about intellect.
It is possible to be an “economic rationalist” and accept that people don’t do advanced calculus before buying goods or selling their labour. The mechanics of the process (ie the algorithm chosen to make choices) is not so critical. And contrary to what critics of “economic rationalism” might suggest I see no contradiction between economic rationalism and goofing off in the back yard on a sunny afternoon.
2. Please educate me. My emotion wants my intellect to be slightly more taxed at this moment.
what an interesting discussion. ProfQ wrote:
“The term ‘rational’ and its variants .. are used in a lot of contexts … 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’.
The first time i heard it put this witty way, it issued from Paul Keating when Treasurer to an interviewer on the ABC. Presumably he wasn’t the only one to say it like that but that’s the first official instance i recall here.
Since eco rationalism’s meaning is being discussed.. …i think Derick’s comments say it all, and, yes indeedy DerickC, that old Englidge Languish is mighty tricky so….
MikeH must be soundly applauded for his neologism (?) “disclude’! WOW!! haven’t seen such a good one for a while – why not forget the handy ‘exclude’ and make up your own similar word – that’s certainly one approach to conveying an argument. Its also critical at this point to note that my entire argument here hinges upon the word “derationalisationeering” I hope you dont mind.
I notice that the media does this a lot at the moment tho, presumably because our so-called journos increasingly copy, cut and paste stories holus bolus from other countries’ media outlets where people there actually do use words like “comedic” instead of comic and “gifted” instead of give, gave or given.
I guess our media also has to edit our stories to remove really obvious non-Australian words like ‘lorry’ or the fabulous ‘momentaneously’ and replace them with words we recognise to make it look like they actually write the stuff they put their name on.
We the great unwashed out here also borrow in a rather engaging manner I think because I occasionally hear people on the telly say someone should “bunker down” instead of the American phrase “hunker down”, which might say a lot about our state of mind OR just the fact we don’t know what a hunker is or why they must go down 🙂 🙂
On the other hand we seem to have somehow lost our own ‘half-arsed’ or ‘kick arse’ and now I hear “ass”. Maybe it apears in the New York Times these days instead? I prefer a half-burro myself or to get out there and kick donkey.
These days personally speaking, I can cope with ‘orientATE’ and I am trying really hard in this vibrant intellectual company to deal with the spread of “academIA” rather than the readily available “academe” but I really would prefer if the Neologists out there concentrated upon the “NEO” bit, ie the ‘new’ aspect of this label that we borrowed from the frogs.
Anyway blah blah whinge whinge whinge, on to the more important discussion of rationalism…excuse the interruption!